THE MAKING, SHAPING 
 
 AND 
 
 TREATING OF STEEL 
 
 BY 
 J. M. CAMP 
 
 AND 
 
 C. B. FRANCIS 
 
 SECOND EDITION 
 
 PUBLISHED BY 
 
 THE CARNEGIE STEEL COMPANY 
 
 PITTSBURGH, PA. 
 

 
 Copyright 1920 by 
 
 CARNEGIE STEEL COMPANY 
 
 Pittsburgh, Pa. 
 
TO 
 
 HOMER D. WILLIAMS, 
 
 President of the Carnegie Steel Company 
 
 AND 
 
 HIS ASSISTANTS 
 
 For Their Valued Suggestions and Words of Appreciation 
 THIS BOOK IS DEDICATED 
 
 425274 
 
PREFACE TO SECOND EDITION 
 
 This book has been written especially for the nontechnical employees 
 of the Carnegie Steel Company, and others, who, seeking self instruction, 
 may desire to secure in the shortest time possible a general knowledge of 
 the metallurgy of iron and steel. 
 
 The book is the outcome of several years experience in attempting to 
 teach the metallurgy of steel to our salesmen and other nontechnical 
 employees. From the first, the method pursued in this work has been that 
 of taking the students, under proper guidance, into the mills, where they 
 obtain, first hand and individually, such information as they desire and 
 are able to collect, and of supplementing the knowledge gained from these 
 visits with talks and explanations delivered in a classroom where conditions 
 are more favorable for this kind 01 instruction than they are in the mills. 
 These talks, in a condensed form, were put in writing, and a copy given 
 to each of the students. As the demand for these lectures increased, it was 
 decided that, for the sake of convenience, they should be printed; and 
 accordingly they were re vised and are here assembled in the present volume. 
 
 In order to increase the value of the book as a reference book, we 
 have aimed to condense the information and to avoid every unnecessary 
 word, both by omitting all matter from the text not absolutely essential 
 and by the free use of tables, drawings and diagrams, which we permit to 
 tell their own story. However, knowing our readers will be men imbued 
 with a desire to learn, we have not avoided discussing many scientific aspects 
 of our subjects. But here we have tried to make it easy even for the 
 general reader. We have aimed to use language as simple as possible, 
 consistent with clearness, and to treat our subjects in such a way that, 
 aside from what a limited education supplies, no prerequisites will be 
 required. We start with the elementary subjects of Physics and 
 Chemistry, the logical prerequisites, and build our metallurgy upon that 
 foundation. The book will, therefore, prove of most value to those 
 connected with the steel business, and not technically educated, who are 
 really anxious to learn more about the wonderful industry in which they 
 are engaged. For such as these, we have aimed to make this book at 
 least a stepping stone to higher and better things. 
 
With regard to the subject matter of the book, we claim but little in 
 the way of originality. We have no new theories to advance and no new 
 discoveries to reveal. Our aim throughout has been to describe conditions 
 and things as they are and to explain the causes for their being as they are, 
 rather than to tell how they might be or how they ought to be. To 
 accomplish our first purpose, we have been compelled to rely mainly upon 
 our personal observation and experiences, but in explaining the causes of 
 things we have freely consulted all those whose published opinions relating 
 to each particular subject have been available and of value. We are, 
 therefore, indebted to many for aid, and to all these we wish to express 
 our thanks. Wherever we have drawn upon these sources of information 
 we have aimed to give the credit to the authors by mention in the text 
 or by foot note references. As guides to collateral reading these 
 references will have an additional value to our readers. We also desire to 
 thank the superintendents and the many heads of departments of our various 
 plants for the courtesies they have shown us and for the many bits of val- 
 uable information which they were ever ready to give. 
 
TABLE OF CONTENTS 
 
 PART I. 
 MAKING OF STEEL 
 
 CHAPTER I Some Principles of Physics and Chemistry 
 
 SECTION I. INTRODUCTION 
 
 1. Iron, the Master Metal 1 
 
 2. Metallurgy defined 2 
 
 3. Matter 2 
 
 a. Fundamental Laws and States of Matter .... 2 
 
 b. Molecules 2 
 
 c. Sciences of Matter 3 
 
 SECTION II. PHYSICAL PROPERTIES OF MATTER. 
 
 1. Classes of Properties 3 
 
 a. General Properties: 3 
 
 i. Inertia 3 
 
 ii. Extension 4 
 
 iii. Mass 4 
 
 iv. Density and Specific Gravity 4 
 
 v. Porosity 4 
 
 vi. Impenetrability 4 
 
 b. Special Properties 4 
 
 i. Cohesion and Adhesion 4 
 
 ii. Elasticity 4 
 
 iii. Plasticity 5 
 
 iv. Ductility 5 
 
 v. Malleability 5 
 
 vi. Hardness 5 
 
 vii. Crystallization 5 
 
 viii. Diffusion . . . -. 5 
 
 ix. Effusion 5 
 
 x. Absorption ." 5 
 
 SECTION III. ENERGY, HEAT AND TEMPERATURE AND THE 
 ETHER: 
 
 1. Energy Law of Conservation 5 
 
 a. Kinds of Energy Kinetic and Potential . . 6 
 
 2. Heat and Temperature 6 
 
 a. Effects of Heat Law of gas expansion and 
 
 kinetic theory 6 
 
 b. Temperature Scales 6 
 
 c. Measurement of Heat 7 
 
 3. The Ether.. 7 
 
TABLE OF CONTENTS 
 
 SECTION IV. CHANGES IN MATTER: 
 
 1. Physical and Chemical Changes 8 
 
 2. Mechanical Mixtures and Chemical Compounds: 
 
 a. Solutions and Alloys 8 
 
 b. Acids, bases, salts, and non-electrolytes 8 
 
 3. The Chemical Elements 9 
 
 a. Classification of the elements 9 
 
 b. Chemical Symbols 9 
 
 c. Fundamental Laws of Chemical Changes. ... 9 
 Definite and Multiple Proportions 9 
 
 SECTION V. THE ATOMIC AND ELECTRON THEORIES: 
 
 1. Atoms 10 
 
 a. Atomic weights 10 
 
 b. Valence 10 
 
 c. Table of Elements with symbols, etc 11 
 
 2. Electrons. 13 
 
 SECTION VI. CHEMICAL FORMULA AND REACTIONS: 
 
 1. Formula of Compounds 13 
 
 2. Formula of Molecules of Elements 13 
 
 3. Chemical Equations 13 
 
 a. Balancing reactions 14 
 
 b. Radicals 14 
 
 4. Ions and Electrolysis 14 
 
 5. Dry and Wet Chemistry 15 
 
 a. Acids, bases and "Salts of Dry Chemistry . . 15 
 
 b. Table of anhydrides 16 
 
 6. Kinds of Reactions 16 
 
 7. Laws of Chemical Reactions 17 
 
 SECTION VII. CHEMICAL NOMENCLATURE: 
 
 1. The General Principle of Nomenclature. . . 18 
 
 2. Terminology of Binary Compounds 18 
 
 3. Terminology of Ternary Compounds 18 
 
 4. Terminology of Acids 18 
 
 5. Terminology of Bases 19 
 
 6. Terminology of Salts ....... 19 
 
 SECTION VIII. CHEMICAL CALCULATIONS: 
 
 1. Kinds of Problems 19 
 
 2. Problems involving weight only 19 
 
 3. Problems involving volume only 20 
 
 4. Problems involving weight and volume 21 
 
TABLE OF CONTENTS ix 
 
 SECTION IX. DESCRIPTION OF ELEMENTS IMPORTANT IN 
 
 IRON AND STEEL MAKING: 
 /^~* 
 
 1. Occurrence, Preparation, Properties and Some 
 
 Compounds of Oxygen 22 
 
 2. Occurrence, Preparation, Properties and Some 
 
 Compounds of Hydrogen 22 
 
 3. Occurrence, Preparation, Properties and Some 
 
 Compounds of Sulphur 23 
 
 4. Occurrence, Preparation, Properties and Some 
 
 Compounds of Carbon 23 
 
 5. Occurrence, Preparation, Properties and Some 
 
 Compounds of Silicon 24 
 
 6. Occurrence, Preparation, Properties and Some 
 
 Compounds of Nitrogen 25 
 
 7. Occurrence, Preparation, Properties and Some 
 
 Compounds of Phosphorous 25 
 
 8. Occurrence, Preparation, Properties and Some 
 
 Compounds of Calcium and Magnesium 25 
 
 9. Occurrence, Preparation, Properties and Some 
 
 Compounds of Aluminum 26 
 
 10. Occurrence, Preparation, Properties and Some 
 
 Compounds of Chromium 26 
 
 11. Occurrence, Preparation, Properties and Some 
 
 Compounds of Manganese 27 
 
 12. Occurrence, Preparation, Properties and Some 
 
 Compounds of Iron 27 
 
 CHAPTER II. Refractories. 
 
 SECTION I. NATURE OF REFRACTORIES: 
 
 1. Importance 28 
 
 2. Requirements of Refractories 28 
 
 3. Classes of Refractories 28 
 
 SECTION II. ACID REFRACTORIES: 
 
 1. Chemical Composition 29 
 
 2. Silica Bricks 29 
 
 3. Clay 29 
 
 a. The Impurities in Clays 29 
 
 b. The Process of Making Fire Clay Brick. ... 30 
 
TABLE OF CONTENTS 
 
 SECTION III. BASIC REFRACTORIES: 
 
 1. Magnesia 30 
 
 2. Lime 30 
 
 3. Dolomite ...... 31 
 
 4. Bauxite 31 
 
 SECTION IV. NEUTRAL REFRACTORIES: 
 
 1. The Ideal Furnace Lining .?... 31 
 
 2. Graphite 31 
 
 3. Chromite 31 
 
 4. Protection for Refractories 31 
 
 5. Table Chemical Analyses of Refractories 32 
 
 SECTION V. TESTING REFRACTORIES: 
 
 1. Trial Tests and Laboratory Tests. . 33 
 
 2. Fusion Temperature 33 
 
 3. Resistance to Compression 33 
 
 4. Expansion and Contraction 34 
 
 5. Slagging Test 34 
 
 6. Density 34 
 
 7. The Impact Test 35 
 
 8. The Abrasion Test 35 
 
 9. Spalling Test 35 
 
 H AFTER HI. Iron Ores. 
 
 SECTION I. ORES AND THE IRON BEARING MINERALS: 
 
 1. Minerals and Ores 36 
 
 2. The Iron Bearing Minerals 36 
 
 a. Magnetite Group 37 
 
 b. Hematite Group 37 
 
 c. Limonite or Brown Ore Group 37 
 
 d. The Carbonate Group 37 
 
 3. The Mineralogical Make-up of Iron Ores 37 
 
 SECTION II. VALUATION OF ORES: 
 
 1. Factors in the Valuation of Ores 38 
 
 a. The Impurities that Are Never Reduced in 
 
 the Blast Furnace 39 
 
 b. The Impurities that May Be Partially 
 
 Reduced 39 
 
 c. The Impurities Always Reduced 40 
 
 d. Water ... 40 
 
 e. Accessibility 41 
 
TABLE OF CONTENTS 
 
 SECTION III. THE BIRMINGHAM DISTRICT: 
 
 1. Location and General Geology 42 
 
 2. Method of Mining 42 
 
 SECTION IV. THE LAKE SUPERIOR DISTRICT: 
 
 1. Importance, Location and General Geology 42 
 
 a. The Marquette Range 43 
 
 b. The Menominee Range 43 
 
 c. The Gogebic Range 43 
 
 d. The Vermilion Range 46 
 
 c. The Missabe Range 46 
 
 f. The Cuyuna Range 47 
 
 SECTION V. MINING THE LAKE ORES: 
 
 1. Prospecting and Exploration 47 
 
 a. Prospecting 47 
 
 b. Drjll Exploration 47 
 
 2. Methods of Mining 50 
 
 a. Open Pit Mining 50 
 
 i. Siteamshovel Mining 50 
 
 ii. Milling 52 
 
 iji. Scramming 53 
 
 iv. Advantages of Open Pit Mining 53 
 
 b. Underground Mining Slicing 53 
 
 i. Advantages of the Slicing System of 
 
 Mining 55 
 
 ii. Depth of Mine Shafts 55 
 
 3. Grading the Ores 55 
 
 4. Transporting the Ores 56 
 
 5. Mining and Grading in Winter 57 
 
 CHAPTER IV. Fuels. 
 
 SECTION I. SOME P RE-REQUISITES TO THE STUDY OF FUELS: 
 
 1. Introductory 58 
 
 2. Sensible and Specific Heat 58 
 
 3. Latent Heat and Change of State 59 
 
 a. Laws of Fusion 59 
 
 b. Laws of Evaporation 59 
 
 c. Laws of Ebullition 59 
 
 4. Transmission of Heat 59 
 
 5. Fuels and Combustion 60 
 
 6. Fuels and Chemical Energy ; 60 
 
 7. Measurement of Calorific Power t 60 
 
 8. The Calorific Power of Some Common Elements . 61 
 
 9. Calculating Calorific Power , 61 
 
 10. Practical Heat Tests. . 62 
 
Xll 
 
 TABLE OF CONTENTS 
 
 SECTION I. Continued. 
 
 11. Laboratory Heat Tests 62 
 
 12. Calorific Intensity 63 
 
 13. Methods of Conserving Heat 63 
 
 14. Pyrometers 64 
 
 a. Specific Heat, or Water, Pyrometer 64 
 
 b. Electric Resistance Pyrometers 64 
 
 c. Thermo-Electric Pyrometers : 64 
 
 d. Radiation Pyrometers 65 
 
 e. Optical Pyrometers 65 
 
 SECTION II. CLASSIFICATION OF FUELS: 
 
 1. Table Classification of Fuels 66 
 
 2. Plan of Study 67 
 
 SECTION III. INCIDENTAL AND LIQUID FUELS: 
 
 1. Incidental Fuels 67 
 
 2. Tar 67 
 
 3. Petroleum 68 
 
 a. Composition of Petroleum 68 
 
 b. Hydrocarbons Generalized, Empirical and 
 
 Structural Formulas 68 
 
 c. Table The Different Homologous Series of 
 
 Hydrocarbons 69 
 
 d. Fuel Oil and Other Products of Petroleum. 69 
 
 SECTION IV. GASEOUS FUELS: 
 
 1. Advantages of Gaseous Fuels 70 
 
 2. Table Hydrocarbons in Natural Gas and Petroleum 70 
 
 3. Natural Gas 71 
 
 4. Artificial Gases 71 
 
 a. Table Composition of Gaseous Fuels 71 
 
 b. Principle of the Gas Producer 71 
 
 c . Factors Affecting the Efficiency of the Producer 73 
 
 d. The Hughes Producer as an Example of 
 
 Mechanically Poked Producer 73 
 
 e. Conditions and Reactions 74 
 
 f. Operation of the Hughes Producer 75 
 
 SECTION V. THE SOLID NATURAL FUELS: 
 
 1. Analysis of Solid Natural Fuels: 75 
 
 a. Table Analysis of a Solid Fuel, Coal, by 
 
 the Three Different Methods 76 
 
 2. Wood 76 
 
 3. Peat , 77 
 
 4. Lignite and Brown Coal 77 
 
TABLE OF CONTENTS xiii 
 
 SECTION V. Continued. 
 
 5. Table Approximate Analyses of the Different 
 
 Solid Fuels 78 
 
 6. Diagram Depicting Geologic Periods in which 
 
 Gas, Oil, and the Valuable Minerals are Found. . 79 
 
 7. Coal 80 
 
 a. Bituminous Coal '. . 80 
 
 b. Ash in Coal 80 
 
 SECTION VI. PREPARED SOLID FUELS: 
 
 1. Powdered Coal 81 
 
 a. Requirements for Use of Powdered Coal 81 
 
 b. Advantages of Powdered Coal 82 
 
 c. The Sharon Powdered Coal Plant 82 
 
 i. Description of Pulverizing Plant 82 
 
 d. Clairton and Homestead Powdered Coal Plants 83 
 
 2. Coke 85 
 
 a. Methods of Manufacturing Coke 85 
 
 SECTION VII. THE BEEHIVE PROCESS FOR THE MANUFACTURE 
 OF COKE: 
 
 1. The Continental No. 1 Plant of the H. C. Frick 
 
 Coke Company 86 
 
 a. The Mine 86 
 
 b. The Coking Plant 86 
 
 i. Construction and Arrangement of Ovens 86 
 
 ii. Waste Heat System 87 
 
 iii. Charging the Ovens 87 
 
 iv. The Coking Process 88 
 
 v. Watering and Drawing the Coke 89 
 
 vi. Longitudinal Ovens 90 
 
 SECTION VIII. THE BY-PRODUCT PROCESS FOR MANUFAC- 
 TURING COKE: 
 
 1. General Features of the Process 90 
 
 2. Advantages of the By-Product Process 91 
 
 3. The Plant of the Clairton By-Product Coke 
 
 Company 91 
 
 a. Construction of the Ovens 93 
 
 b. Heating the Ovens 95 
 
 c. Drying and Heating New Ovens 96 
 
 d. Operation of the Ovens 96 
 
xiv TABLE OF CONTENTS 
 
 SECTION IX. THE BY-PRODUCT PLANT: 
 
 1. The Volatile Matter of Coal 98 
 
 2. Gas Mains and Coolers 98 
 
 3. Separation of the Tar and Ammonia Liquor .... 99 
 
 4. Compressors and Tar Extractors 99 
 
 5. Recovery of Ammonia , . . . 100 
 
 6. Debenzolating the Gas 101 
 
 SECTION X. THE BENZOL PLANT: 
 
 1. Light Oil 101 
 
 2. Composition of Light Oil 102 
 
 3. Construction and Principles of the Still 102 
 
 4. Operation of the Crude Still 102 
 
 5. Washing the Products of the Crude Stills 103 
 
 6. The Pure Stills 103 
 
 SECTION XI. SOME PROPERTIES AND USES OF THE RAW BY- 
 PRODUCTS FROM THE COKE WORKS: 
 
 1. Characteristics of Benzol, Toluol and Naphtha 105 
 
 a. Some Members of the Benzene Series, and 
 
 their Physical Properties 106 
 
 2. Commercial Benzol 107 
 
 a. Uses of Commercial Benzol 107 
 
 b. Motor Benzol. 107 
 
 3. Properties and Uses of Pure Benzol, or Benzene. . . . 109 
 
 a. Table Diagram Showing Some of the 
 
 Products Derived from Benzene, Their 
 
 Formulas and Their Uses 108 
 
 b. Table Reactions Showing How Phenol, 
 
 Picric Acid and Resorcinol may be 
 
 Derived from Benzene 110 
 
 c. Table Reactions Showing How Aniline and 
 
 Benzidine are Derived from Benzene 109 
 
 4. Usesof Toluene Ill 
 
 a. Table Some Products Derived from Toluene, 
 
 Their Formulas and Uses 112 
 
 b. Commercial Toluol and Solvent Naptha Ill 
 
 5. Uses of Naphthalene 113 
 
 a. Table Showing Some Products Derived 
 
 from Naphthalene 114 
 
 6. Tar 114 
 
 a. Diagram Illustrative of the Refining of Tar . 115 
 
 7. Ammonia 118 
 
 a. Ammonium Sulphate 116 
 
 b. Use of Ammonium Sulphate as a Fertilizer. . 116 
 
TABLE OF CONTENTS 
 
 CHAPTER V. Fluxes and Slags. 
 SECTION I. FLUXES: 
 
 1. Smelting and the Functions of a Flux 117 
 
 2. The Selection of the Proper Flux for a Given Process 117 
 
 3. Acid Fluxes 118 
 
 a. Alumina 118 
 
 4. Basic Fluxes 118 
 
 a. Available Base 118 
 
 b. Limestone 119 
 
 c. Supply of Limestone 119 
 
 d. Action of Limestone in Furnaces 119 
 
 5. Neutral Fluxes 120 
 
 SECTION II. SLAGS: 
 
 1. Slag 120 
 
 2. Functions of Slags 120 
 
 3. Importance of Slags , 120 
 
 4. Chemical Composition of Slags 121 
 
 5. Relation of Acids to Bases in Blast Furnace Slags. . 121 
 
 6. Ratio of Acids to Bases in Open Hearth Slags 122 
 
 7. Acid to Base in Acid Furnaces 122 
 
 8. Electric Steel Furnace Slags 122 
 
 9. Acids Formed by Silicon . . 122 
 
 10. So-called Acid and Basic Slags 123 
 
 11. Classification of Slags.... 123 
 
 12. Uses of Slags 124 
 
 CHAPTER VI. The Manufacture of Pig Iron. 
 
 SECTION I. SOME INTERESTING HISTORICAL FACTS: 
 
 1. Early History of Iron 125 
 
 2. Old American Furnaces 126 
 
 3. The Importance of Iron 126 
 
 SECTION It. COMPOSITION AND CONSTITUTION OF PIG IRON: 
 
 1 1. Constitution of Pig Iron 126 
 
 2. Chemical Elements in Pig Iron 127 
 
 a. Carbon 127 
 
 b. Silicon 127 
 
 c. Manganese 128 
 
 d. Sulphur 128 
 
 e. Phosphorus 129 
 
 3. Grading Pig Iron 129 
 
TABLE OF CONTENTS 
 
 SECTION III. A BRIEF OUTLINE OF THE PROCESS AND EQUIP- 
 MENT FOR THE MANUFACTURE OF PIG IRON: 
 
 1. Trend of Modern Improvements 130 
 
 2. Essentials of the Process 130 
 
 3. Essential Equipment 130 
 
 SECTION IV. CONSTRUCTION OF THE BLAST FURNACE PROPER: 
 
 1. The Gross Features of the Furnace Proper 131 
 
 2. The Foundation 131 
 
 3. The Hearth or Crucible 132 
 
 4. The Bottom 132 
 
 5. Tapping Hole 132 
 
 6. Cinder Notches 134 
 
 7. Tuyeres 134 
 
 8. Tuyere Connections 134 
 
 9. Boshes 135 
 
 10. Mantle 136 
 
 11. Shaft, or Stack, and In-Walls 136 
 
 a. Thick Wall Type 136 
 
 i. The furnace Lines and Bosh Angles 136 
 
 b. Intermediate, or Semi-Thin, Wall Type 137 
 
 c. Thin Walled Type 137 
 
 d. Furnace Linings 137 
 
 12. Water Trough 139 
 
 13. Tops 139 
 
 a. Stock Distributor 140 
 
 b. Hoisting Appliances 140 
 
 c. Top Openings 140 
 
 d. General Consideration for Top Construction. . 141 
 
 14. Runners 141 
 
 SECTION V. BLAST FURNACE ACCESSORIES: 
 
 1. The Stoves 142 
 
 a. Stove Burners and Valves 143 
 
 b. Other Stove Openings 143 
 
 c. Stove Linings 145 
 
 2. Dust Catcher and Gas Mains 146 
 
 3. Arrangement of Furnaces and Cleaning Plant at 
 
 Duquesne 146 
 
 a. Primary Division 148 
 
 i. Methods of Scrubbing the Gas 148 
 
 ii. The Fans 149 
 
 iii. Water Separator 149 
 
 b. The Secondary Division 149 
 
 SECTION VI. EQUIPMENT FOR HANDLING RAW AND FINISHED 
 
 MATERIAL- 
 1. The Boiler House, Power Plant, Pumping Station, 
 
 Blowing Engines, etc 150 
 
TABLE OF CONTENTS xvii 
 
 SECTION VI. Continued. 
 
 2. Dry Blast 150 
 
 3. Cold and Hot Blast Mains 150 
 
 4. Appliances for Handling Ores, Coke and Stone 150 
 
 5. Stock House Equipment 151 
 
 6. Disposal Equipment for the Iron 152 
 
 7. Equipment for Slag Disposal 152 
 
 SECTION VII. OPERATING THE FURNACE: 
 
 1. Blowing In 152 
 
 2. Drying 152 
 
 3. Filling 153 
 
 4. Lighting 153 
 
 5. Heating the Bottom 154 
 
 6. The Heating of the Stoves 154 
 
 7. Tapping 154 
 
 8. Care of Runners 155 
 
 9. Sampling the Iron 155 
 
 10. Tapping Slag 156 
 
 11. Changing Stoves 156 
 
 12. Charging the Furnace 156 
 
 13. Some Irregularities of Furnace Operation 157 
 
 a. Slips 157 
 
 b. Scaffolding 158 
 
 c. Chimneying and Hot Spots 158 
 
 d. Loss of Tuyeres and Chilled Hearth 158 
 
 14. Uncertainties and Variables in Furnace Control 158 
 
 15. Banking 159 
 
 16. Blowing Out. 159 
 
 SECTION VIII. THE BLAST FURNACE BURDEN: 
 
 1 . Burdening the Furnace 159 
 
 a. Outline of a Method for Solving a Burdening 
 
 Problem 161 ' 
 
 b. The Burden Sheet 161 
 
 2. Table 27. Analysis of Raw Materials Used in the 
 
 Blast Furnace 160 
 
 SECTION IX. CHEMISTRY OF THE PROCESS: 
 
 1. Methods of Investigating the Reactions of the Blast 
 
 Furnace 163 
 
 2. The Functions of Oxygen and Carbon 163 
 
 3. Behavior of Nitrogen in the Furnace 165 
 
 4. Action of Phosphorus in the Furnace 165 
 
 5. Disposition of Sulphur in the Furnace 165 
 
 6. Behavior of Silicon 165 
 
 7. Action of Calcium and Magnesium . . . . : 166 
 
 8. Action of Aluminum. . 166 
 
TABLE OF CONTENTS 
 
 SECTION IX. Continued. 
 
 9. Action of Less Abundant Elements 166 
 
 10. The Reactions Within the Furnace 167 
 
 11. Tracing the Materials Through the Furnace 170 
 
 12. Conditions Affecting the Amount of Silicon and 
 
 Sulphur in the Metal 171 
 
 CHAPTER VII. The Bessemer Process of Manufacturing Steel. 
 
 SECTION I. THE CLASSIFICATION OF FERROUS PRODUCTS: 
 
 1. Introductory 172 
 
 2. Pig Iron and Cast Iron , 172 
 
 3. Malleable Cast Iron 172 
 
 4. Wrought Iron 173 
 
 5. Steel 173 
 
 6. Methods of Making Steel 174 
 
 7. General Principles of the Methods of Purifying 
 
 Pig Iron 175 
 
 SECTION II. PRINCIPLES AND HISTORY OF THE BESSEMER 
 PROCESS: 
 
 1. Principles of the Process 175 
 
 2. Some Incidents Connected with the Early History 
 
 of the Process ' 176 
 
 3. Importance of Manganese 176 
 
 4. Thomas and Gilchrist Process 177 
 
 5. Other Improvements 177 
 
 6. Plan of Study 177 
 
 SECTION III. EQUIPMENT AND ARRANGEMENT OF THE EDGAR 
 THOMSON PLANT: 
 
 1. The Converter House 177 
 
 2. The Larger Accessories 179 
 
 a. The Cupolas 179 
 
 b. Charging the Cupola 180 
 
 c. The Blast 180 
 
 d. The Mixers, 181 
 
 i. Importance of the Mixer 181 
 
 e. The Stripper 181 
 
 f. The Casting Equipment 182 
 
 i. The Ingot Moulds 182 
 
 SECTION IV. CONVERTER CONSTRUCTION AND REPAIRS: 
 
 1. General Features Pertaining to Converters 183 
 
 2. Parts of Converter 183 
 
 a. Lining of the Converter 184 
 
 b. The Bottom 185 
 
 i. Relining the Bottom 185 
 
TABLE OF CONTENTS 
 
 SECTION V. THE CONVERTER IN OPERATION PURIFYING 
 THE METAL: 
 
 1. Charging the Vessel 187 
 
 2. The Blow 187 
 
 3. Controlling the Blow 188 
 
 4. The End of the Blow \ 189 
 
 SECTION VI. FINISHING OPERATIONS CONVERTING THE 
 
 PURIFIED METAL INTO STEEL: 
 
 1. Deoxidation and Recarburization 190 
 
 2. Loss of Recarburizer and Deoxidizer 191 
 
 3. Examples of Recarburizing 191 
 
 4. Ladle Reaction 191 
 
 5. Teeming 192 
 
 6. Sampling the Steel for Chemical Analyses 192 
 
 SECTION VII. CHEMISTRY OF THE PROCESS: 
 
 1. The Order of Elimination of the Elements 193 
 
 2. The Laws and Conditions Governing the Reactions 
 
 in the Converter 193 
 
 3. Reactions of the First Period 194 
 
 4. Reactions of the Second Period 195 
 
 5. Chemistry of Recarburizing and Deoxidizing 196 
 
 CHAPTER VIII. The Basic Open Hearth Process. 
 
 SECTION I. SOME GENERAL FEATURES OF THE SIEMENS 
 
 PROCESS: 
 
 1. Early History of the Process 198 
 
 2. Principles of Siemens Pig and Ore Process 199 
 
 3. Advantages of the Process 199 
 
 4. Mechanical Changes and Improvements in Siemen's 
 
 Process 200 
 
 5. Metallurgical Improvements 200 
 
 6. The Process for the Pittsburgh District 201 
 
 SECTION II. EQUIPMENT FOR A MODERN BASIC OPEN 
 
 HEARTH PLANT: 
 
 1. The Modern Plant 202 
 
 2. Calcining Plant 202 
 
 3. Fuels 203 
 
 4. Fuel Consumption 203 
 
 5. Hot Metal Mixer 204 
 
 6. Spiegel Cupolas 204 
 
 7. The Steel Ladles 204 
 
 8. The Stripper 205 
 
 9. Moulds 205 
 
 10. The Charging Machine 206 
 
 11. Charging Boxes 207 
 
 12. Stock Yard 207 
 
 13. Arrangement of the Plant 207 
 
TABLE OF CONTENTS 
 
 SECTION III. CHIEF FEATURES OF BASIC OPEN HEARTH 
 CONSTRUCTION: 
 
 1. Parts of the Open Hearth Furnace and Their 
 
 Arrangement 209 
 
 2. The Furnace Proper 209 
 
 a. The Hearth 210 
 
 b. The Walls 210 
 
 c. The Roof 211 
 
 d. The Bulk Heads 211 
 
 3. The Ports 211 
 
 4. The Up-and-Down-Takes 211 
 
 a. Arrangement of Up-and-Down Takes for 
 Natural Gas, Coke Oven Gas, Powdered 
 
 Coal and Tar 212 
 
 5. Slag Pockets 212 
 
 6. Regenerators for Producer Gas 212 
 
 7. Regenerators for Natural and Coke Oven Gases. . . . 216 
 
 8. Regenerators for Powdered Coal 217 
 
 9. Flues and Valves 217 
 
 10. The Stack 217 
 
 SECTION IV. OPERATION OF A BASIC OPEN HEARTH- 
 PURIFYING THE METAL: 
 
 1. Furnace Attendants and Their Duties 218 
 
 2. Preparation. of the Furnace for its First Charge. . . . 218 
 
 3. Charging 219 
 
 a. The Order of Charging Raw Materials 220 
 
 4. Melting Down the Charge 220 
 
 5. The Addition of the Hot Metal 221 
 
 6. The Purification Periods 221 
 
 a. The Ore Boil 222 
 
 i. The Run off 222 
 
 b. The Lime Boil 223 
 
 c. The Working Period. ,...' 223 
 
 i. Methods of Working the Heat 223 
 
 ii. Testing for Carbon 224 
 
 iii. Control of Carbon and Temperature . . . 224 
 
 iv. Judging the Temperature of the Bath. . . 225 
 
 7. Tapping 225 
 
 SECTION V. FINISHING THE HEAT MAKING STEEL FROM 
 THE PURIFIED METAL: 
 
 1. Methods of Finishing the Steel 226 
 
 2. Some Features that Make the Finishing of the Steel 
 
 Difficult 227 
 
 3. Teeming 228 
 
 4. Sampling 229 
 
TABLE OF CONTENTS xxi 
 
 SECTION VI. KEEPING THE FURNACE IN REPAIR: 
 
 1. Preparation of the Furnace for the Next Charge 229 
 
 2. Furnace Troubles 230 
 
 3. Repair Materials 231 
 
 a. Dolomite 231 
 
 b. Magnesite 231 
 
 c. Chrome Ore 231 . 
 
 SECTION VII. CHEMISTRY OF THE BASIC PROCESS: 
 
 1. Some of the Principles and Conditions Involved.,. 232 
 
 2. Properties of Iron and Its Oxides 232 
 
 a. The Importance of Ferrous Oxide, FeO, in the 
 Part Played by the Oxides of Iron in the 
 
 Process 233 
 
 3. Properties of Silicon and Its Oxide, Silica 234 
 
 4. Properties of Manganese and Its Oxides 235 
 
 5. Sulphur and Its Oxides 236 
 
 a. Sulphur from the Fuel 237 
 
 6. Phosphorus and Its Oxides 237 
 
 7. Carbon and Its Oxides 238 
 
 a. The Action of the Limestone 239 
 
 b. Effect of Carbon Elimination on Slag 
 
 Composition 239 
 
 8. The Order of Elimination 239 
 
 a. Factors Opposing this Order of Elimination 240 
 
 9. Resume 241 
 
 CHAPTER IX. Manufacture of Steel in Electric Furnaces. 
 
 SECTION I. INTRODUCTORY: 
 
 1. The Plan of Study 243 
 
 2. Force, Work, Energy and Potential 243 
 
 3. Power 240 
 
 4. Transmission of Energy 244 
 
 5. Electromotive Force (E. M. F.) 246 
 
 SECTION II. THE DEVELOPMENT OP ELECTROMOTIVE FORCES 
 
 OR "GENERATION OF CURRENT:" 
 
 1. Methods for Setting Up Electric Currents 246 
 
 2. Magnetism , 246 
 
 a. Magnets and Magnetic Substances 247 
 
 b. Magnetic Fields and Electric Currents 248 
 
 3. Electromagnetic Induction 249 
 
 a. Laws of Electromagnetic Induction 250 
 
 b. The Dynamo 250 
 
 SECTION III. KINDS OF CURRENT: 
 
 1. Alternating Current 251 
 
 a. Graphic Representation of Alternating Current 252 
 
TABLE OF CONTENTS 
 
 SECTION III. Continued. 
 
 2. Direct Currents 253 
 
 3. Polyphase Currents 253 
 
 4. The Two Schemes of Wiring for Three Phase Current 254 
 
 SECTION IV. TRANSMISSION OF THE CURRENT: 
 
 1. Ohm's Law 256 
 
 2. Resistance of Conductors 256 
 
 a. Effect of Temperature on Conductors 257 
 
 b. Resistance in Series and Parallel 258 
 
 c. Currents Through Divided Circuits 259 
 
 3. * Self-induction, Impedance, Power Factor 259 
 
 4. Heat Developed in Conductors 259 
 
 5. The Stationary Transformer, 260 
 
 a. Kinds of Stationary Transformers 260 
 
 SECTION V. THE UTILIZATION OF THE CURRENT IN ELECTRIC 
 FURNACES: 
 
 1. Effects Produced by Electric Current 261 
 
 a. Chemical Action Produced by the Electric 
 
 Current 261 
 
 i. Electrical Units of Measurements 262 
 
 b. The Magnetic Influence of the Current 262 
 
 2. Heating the Bath 262 
 
 a. Heating by Direct Resistance 263 
 
 b. Indirect Resistance Heating 264 
 
 c. Arc Heating 264 
 
 d. Methods of Applying the Arc in Arc Furnaces. 265 
 
 i. The Stassano Furnace 265 
 
 ii. Girod Furnaces 266 
 
 iii. The Principle of the Heroult Furnace .. 266 
 
 3. Some General Conclusions 267 
 
 SECTION VI. GENERAL FEATURES PERTAINING TO THE 
 METALLURGY OF STEEL MADE BY ELECTRO- 
 THERMAL PROCESSES: 
 
 1. Advantages of Electric Heating. 267 
 
 2. Refining Procedure 267 
 
 a. The Oxidizing Period 268 
 
 b. The Reducing Period ' 269 
 
 i. Oxygen 269 
 
 ii. Removal of Sulphur 269 
 
 c. The Finishing Period 270 
 
 3. Some Comparisons 271 
 
 4. Fluxing Materials 271 
 
 5. General Manufacturing Practice 271 
 
TABLE OF CONTENTS 
 
 SECTION VII. THE DUQUESNE PLANT FEATURES PERTAIN- 
 ING TO ITS CONSTRUCTION: 
 
 1 . Equipment 275 
 
 2. Construction of the Furnace Shell 275 
 
 3. The Furnace Lining 275 
 
 4. The Roof 277 
 
 5. Controlling the Electrodes 277 
 
 a. The Electrode Holders 277 
 
 6. The Electrodes 279 
 
 7. Furnace Openings 279 
 
 SECTION VIII. OPERATION OF THE FURNACE: 
 
 1. Practice at Duquesne Plant 279 
 
 a. Charging 280 
 
 b. Deoxidizing 280 
 
 c. Finishing the Heats 281 
 
 d. Tapping and Teeming 281 
 
 e. Scrap Heats 281 
 
 SECTION IX. THE CHEMISTRY OF THE PROCESS: 
 
 1. Deoxidation of the Bath 286 
 
 2. Desulphurizing the Metal 286 
 
 3. Difficult Specifications 288 
 
 SECTION X. PROPERTIES AND USES OF ELECTRIC STEEL: 
 
 1. Properties of Electric Steel 289 
 
 2. Illinois Steel Company's Tests on Rails 290 
 
 3. Uses of Electric Steel 291 
 
 4. Summary 291 
 
 CHAPTER X. The Duplex and Triplex Processes. 
 
 SECTION I. GENERAL FEATURES OF THE DUPLEX PROCESS: 
 
 1. What the Duplex Process Is 293 
 
 2. Advantages and Disadvantages of the Process .... 293 
 
 3. Methods of Duplexing 294 
 
 4. The Talbot Furnace 294 
 
 SECTION II. OPERATION OF THE PROCESS: 
 
 1. An Example of the Duplexing Process 295 
 
 2. Preparing the Furnace for Charging 295 
 
 3. Charging Molten Metal from the Converters for 
 
 the First Heat 295 
 
 4. Tapping and Recarburizing the First Heat 296 
 
 5. Preparing the Furnace for the Second Heat 296 
 
 6. Closing Down the Furnace for the Week End 297 
 
 7. Slag 297 
 
 SECTION III. COMBINATION PROCESSES IN THE SOUTH: 
 
 1. The Duplex Process in the South 297 
 
 2. The Southern Triplexing Process 298 
 
TABLE OF CONTENTS 
 
 PART II. 
 THE SHAPING OF STEEL 
 
 CHAPTER I. The Mechanical Properties of Steel. 
 
 SECTION I. GENERAL REMARKS PERTAINING TO THE TESTING 
 OF STEEL: 
 
 1. The Factors that Affect the Mechanical Properties 
 
 of Steel 299 
 
 2. The Two Objects in the Testing of Steel 299 
 
 3. Relative Importance of Physical and Chemical Testing 300 
 
 4. Nature of Physical Testing 300 
 
 SECTION II. THE TESTING OF STRUCTURAL AND OTHER SOFT 
 STEELS: 
 
 1. The Pulling Test 301 
 
 a. Procuring the Test Pieces 301 
 
 b. Preparation of the Test Piece 302 
 
 c. Pulling the Test 303 
 
 i. Graphic Representation of Tests 304 
 
 ii. Reasons for the Points of Yield and 
 
 Maximum Stress 304 
 
 d. Examination of Test After Pulling 305 
 
 e. Calculating the Results of the Test 306 
 
 2. The Modulus of Elasticity, or Young's Modulus ... 307 
 
 3. Relative Importance of the Mechanical Properties as 
 
 Determined by the Pulling Test 307 
 
 4. Bending Tests. 308 
 
 SECTION III. THE TESTING OF THE HIGHER CARBON AND 
 HEAT TREATED STEELS: 
 
 1. Kinds of Tests Applied to the Higher Carbon and 
 
 Heat-treated Steels 308 
 
 2. The Tensile Test 308 
 
 3. The Impact Test 309 
 
 4. Hardness Tests .... 309 
 
 a. Shore Scleroscope 310 
 
 b. Brinell Hardness 310 
 
 i. Relation of Brinell Number to Tensile 
 
 Strength 311 
 
 CHAPTER II. The Mechanical Treatment of Steel. 
 
 SECTION I. METHODS AND EFFECTS OF MECHANICALLY 
 WORKING STEEL: 
 
 1. Methods of Shaping Steel 312 
 
 2. Benefits of Mechanical Working 312 
 
 3. Hot and Cold Working 313 
 
TABLE OF CONTENTS 
 
 SECTION II. 
 
 SUMMARY OF THE HISTORY AND PRINCIPLES OF 
 WORKING STEEL: 
 
 1. The Three Methods for Mechanically Working Steel. 316 
 
 a. Hammer Forging 316 
 
 i. Principles and Effects of Hammering .... 317 
 
 b. The Forging Press 317 
 
 i. The Effect of Pressing 317 
 
 ii. Advantages of the Press '. 318 
 
 c. Rolling 318 
 
 i. Principle and Effect of Rolling 319 
 
 ii. Rolling Compared with Hammering and 
 
 Pressing 320 
 
 iii. Rolling and Pressing Ingots 321 
 
 CHAPTER III. Essentials of Rolling Mill Construction and 
 
 Operation. 
 
 SECTION I. THE ROLLS THEIR PREPARATION AND ARRANGE- 
 MENT: 
 
 1. Parts and Equipment of the Simplest Type of 
 
 Rolling Mill 322 
 
 2. The Rolls and Their Parts 322 
 
 a. The Manufacture of Rolls 323 
 
 b. The Sand Roll 323 
 
 i. The Materials Used in Sand Cast Rolls.. 324 
 
 c. Chilled Rolls 324 
 
 i. Difficulties in Making Chilled Rolls 326 
 
 d. Steel Rolls 326 
 
 e. Other Rolls 327 
 
 f. The Size of Rolls 327 
 
 g. Roll Design 327 
 
 i. Methods of Procedure in Designing Rolls 328 
 
 ii. Difficulties in Designing Rolls 328 
 
 h. Turning the Rolls 329 
 
 i. Dressing the Rolls 329 
 
 3. Types of Mills 330 
 
 SECTION II. PARTS OF THE MILL ESSENTIAL TO THE OPER- 
 ATION OF THE ROLLS: 
 
 1. The Chocks 331 
 
 a. The Arrangement of the Chocks 331 
 
 b. The Function of the Chocks 332 
 
 2. The Housings 332 
 
 a. The Adjusting Equipment 333 
 
TABLE OF CONTENTS 
 
 SECTION II.- Continued. 
 
 3. The Pinions v 333 
 
 4. The Connections . 334 
 
 5. Guides and Guards 334 
 
 6. Additional Equipment 335 
 
 SECTION III. SOME GENERAL FEATURES PERTAINING TO 
 OPERATION OF THE ROLLING MILL: 
 
 1. The Mill Force 336 
 
 a. Duties of the Roller 336 
 
 2. Fins 336 
 
 3. The Different Passes and Stands 337 
 
 4. Factors Affecting the Rolling Operation 337 
 
 5. Effects of Temperature 337 
 
 6. Effect of Chemical Composition 338 
 
 7. The Effect of Speed 339 
 
 8. Draught 339 
 
 9. The Effect of Diameter of Rolls 340 
 
 CHAPTER IV. Preparation of the Steel for Rolling. 
 
 SECTION I. INGOTS AND THEIR DEFECTS: 
 
 1. Preparation of Ingots 342 
 
 2. Ingot Defects 342 
 
 3. The Nature of the Cooling of an Ingot 343 
 
 a. Pipes 343 
 
 i. Methods of Reducing Waste due to 
 
 the Pipe 346 
 
 b. Blow Holes 346 
 
 c. Crystallization 348 
 
 d. Segregation 348 
 
 e. Checking and Scabs 349 
 
 f . Slag Inclusions 349 
 
 4. Size and Shape of Ingots 352 
 
 SECTION II. THE CONSTRUCTION OF THE SOAKING PIT: 
 
 1. General Features of the Pit 352 
 
 2. Arrangement of the Pits 353 
 
 3. Equipment for Handling Ingots 353 
 
 4. Construction of the Pits 353 
 
 a. The Air Regenerators 355 
 
 b. The Pit Covers ,.. 356 
 
 c. Fuel and Air Valves, etc 357 
 
 d. Stack-Flues and Stack 357 
 
 e. The Course of the Gases Through the Pits 358 
 
 5. Eight Ingot Pits 358 
 
 6. Making up the Bottom of the Pit 358 
 
TABLE OF CONTENTS xxvii 
 
 SECTION III. SOAKING THE INGOTS FOR ROLLING: 
 
 1. Charging the Ingots 359 
 
 2. Heating the Ingots 360 
 
 a. Week-end Charges 360 
 
 b. Soaking Hot and Cold Ingots 360 
 
 c. Soaking Hot Spring Steel 361 
 
 d. Soaking Low Carbon Hot Steel 361 
 
 e. Soaking Medium Steels 361 
 
 f. Soaking Screw Stock . . . 361 
 
 g. Soaking Alloy Steels 362 
 
 3. Drawing the Ingots 362 
 
 4. Heat Balance of Pits 362 
 
 5. Disposition of Ingot Products 363 
 
 CHAPTER V. The Rolling of Blooms and Slabs. 
 SECTION I. INTRODUCTORY: 
 
 1. Outline of the Plan of Study 364 
 
 2. Blooms, Slabs and Billets 365 
 
 SECTION II. SOME GENERAL' FEATURES PERTAINING TO 
 BLOOMING MILLS: 
 
 1. Size of Blooming Mills 366 
 
 2. Types of Bloomers, Their Advantages and Dis- 
 
 advantages 366 
 
 3. Drive for Reversing Mills 368 
 
 SECTION III. AN EXAMPLE OF REVERSING MILLS THE 40" 
 MILL AT DUQUESNE: 
 
 1. The Engine 368 
 
 2. Driving Connections 368 
 
 3. Pinions and Pinion Housings 369 
 
 4. Spindles and Coupling Boxes 369 
 
 5. Roll Housings 370 
 
 6. Rolls 371 
 
 7. Roll Bearings 371 
 
 8. Hydraulic Shears 372 
 
 9. Steam Shears 372 
 
 10. Manipulator 373 
 
 11. Design of the Rolls 373 
 
 12. Operation of Rolling 375 
 
 SECTION IV. EXAMPLE OF A THREE-HIGH BLOOMING MILL: 
 
 1. Plan of Study 377 
 
 2. The 40" Three-high Mill at Edgar Thomson 377 
 
 a. The Engine and Connections 378 
 
xxviii TABLE OF CONTENTS 
 
 SECTION IV. Continued. 
 
 b. The Pinions and Spindles 378 
 
 c. The Roll Housings 378 
 
 d. The Rolls 378 
 
 e. Lifting Tables 380 
 
 3. Roll Design for Three-high Bloomers 381 
 
 a. An Example of Roll Design for Three-high 
 
 Blooming Mill 381 
 
 SECTION V. THE ROLLING OF SLABS: 
 
 1. The Rolling of the Slab 385 
 
 2. The 32" Mill at Homestead as an Example of a 
 
 Slabbing Mill 385 
 
 a. The Horizontal Mill 385 
 
 b. The Vertical Mill 386 
 
 3. Precautions to be Observed in Rolling Slabs 387 
 
 4. Removal of Scale 388 
 
 5. Shearing Slabs at 32" Mill 389 
 
 CHAPTER VI. The Rolling of Billets and Other Semi- 
 Finished Products. 
 
 SECTION I. THE THREE-HIGH BILLET MILL: 
 
 1. General Features of Rolling Billets 391 
 
 2. Example of Three-High Billet Mill The 28" Mill 
 
 at Duquesne 391 
 
 a. Engine 391 
 
 b. Drive 392 
 
 c. Pinions and Their Housings 392 
 
 d. Housings and Roll Bearings 393 
 
 e. Rolls 393 
 
 f. Guide Cages 395 
 
 g. Tables 395 
 
 SECTION II. THE CONTINUOUS BILLET MILL: 
 
 1. General Features of the Continuous Mill 397 
 
 2. Advantages and Disadvantages of Continuous Mills . . 397 
 
 3. Example of Continuous Billet Mill 398 
 
 a. Drive 398 
 
 b. Pinions and Housings 399 
 
 c. Rolls and Housings 399 
 
 i. Adjustment of Rolls 399 
 
 d. Arrangement of Roll Stands and Guides 400 
 
 e. The Rolls 400 
 
 f . Cropping Shears 401 
 
 g. Flying Shears 404 
 
 h. Hot Beds.. 404 
 
TABLE OF CONTENTS xxix 
 
 SECTION III. ROLLING OF SHEET BARS AND SKELP: 
 
 1 . Difficulties and Methods of Rolling Semi-Finished Flats 406 
 
 2. The Tongue and Groove Pass 406 
 
 3. Sheet Bar 408 
 
 4. The 21" Mill at Duquesne .-,.'.:.. 408 
 
 a. The Layout for the Mill 409 
 
 b. Arrangement of the Roll Tables 409 
 
 c. Hot Saws and Shears 410 
 
 d. Drive 410 
 
 e. Pinions and Housings 411 
 
 f . Rolls and Roll Housings 412 
 
 SECTION IV. SOME GENERAL PRECAUTIONS TO BE OBSERVED 
 IN ROLLING SEMI-FINISHED PRODUCTS: 
 
 1. Reasons for Studying Defects 413 
 
 2. Rough Surface Due to Scale ". 413 
 
 3. Cobbling 414 
 
 4. Laps 414 
 
 5.- Collar Marks 414 
 
 6. Guide Marks 415 
 
 7. Ragging Marks 415 
 
 8. Off Size 415 
 
 9. Unequal Draughts 415 
 
 10. Seams 415 
 
 11. Slivers < 415 
 
 12. Scabs 415 
 
 13. Shearing Defects 415 
 
 14. Splits or Cracks in Billets and Blooms 417 
 
 15. Inspection 417 
 
 CHAPTER VII. The Rolling of the Heavier Finished Products 
 Plates. 
 
 SECTION I. PREPARATION OF THE STEEL FOR ROLLING 
 FINISHED PRODUCTS: 
 
 1. Reheating 418 
 
 2. Types of Reheating Furnaces 419 
 
 a. The Regenerative Reheating Furnace 419 
 
 b. The Recuperative, or "Continuous" Furnace. . . 421 
 
 3. The Advantages of Continuous Reheating Furnaces. . . 421 
 
 SECTION II. THE ROLLING OF SHEARED PLATES: 
 
 1. Methods of Rolling Plates 423 
 
 2. The 140" Mill at Homestead as an Example of a 
 
 Sheared Plate Mill 423 
 
 a. The Drive and Connections 424 
 
xxx TABLE OF CONTENTS 
 
 SECTION II. Continued. 
 
 3. Difficulties in Rolling Sheared Plates 425 
 
 4. The Rolling Process 426 
 
 5. Cooling and Straightening , 427 
 
 6. Laying-out and Stamping 427 
 
 7. Test Pieces 429 
 
 8. Shearing 429 
 
 a. Shearing Tolerances , . . 430 
 
 9. Size Inspection 430 
 
 10. Weighers .... 430 
 
 11. Checkers 430 
 
 12. Slip Maker 431 
 
 13. Recorder 431 
 
 SECTION III. UNIVERSAL MILL PLATES: 
 
 1 . The 48" Mill at Homestead as an Example of Universal 
 
 Plate Mills 431 
 
 2. The Operation of Rolling 432 
 
 3. Straightening, Marking and Shearing U. M. Plate . ." . . 433 
 
 4. Advantages of Universal Mill Plates 433 
 
 5. Physical Properties of Plates 433 
 
 6. Inspection of Plates 434 
 
 CHAPTER VIII. The Rolling of Large Sections. 
 SECTION I. RAILROAD RAILS: 
 
 1. Development of Rail Manufacture 436 
 
 2. Methods of Rolling Rails 437 
 
 3. How to Study Roll Design 438 
 
 4. Precautions to be Observed in Designing the Rolls. . . 438 
 
 5. Stages of Reduction 439 
 
 6. The Different Steps in the Design of the Rolls 439 
 
 a. The Section 439 
 
 b. The Cold Templet 439 
 
 c. The Hot Templet 441 
 
 d. The Pass Templet 441 
 
 e. Preparation for the Rolling 443 
 
 7. The Diagonal Method 447 
 
 8. The Mills 447 
 
 9. Rolling Heavy Rails 448 
 
 10 Unavoidable Variations 448 
 
 11. The Various Steps in Shaping of Rails 449 
 
 12. Cutting 450 
 
 13. Recording 451 
 
 14. Finishing and Inspection 451 
 
 15. Light Rails 452 
 
TABLE OF CONTENTS xxxi 
 
 SECTION II. THE SHAPING OF RAIL JOINTS: 
 
 1. Rolling Rail Joints 454 
 
 2. Methods of Finishing Rail Joints 458 
 
 a. Cold Worked Bars 459 
 
 b. Cold Worked and Annealed Bars 459 
 
 c. Hot Worked Bars 460 
 
 d. Hot Worked and Oil Quenched 461 
 
 3. The Edgar Thomson Splice Bar Shop 458 
 
 SECTION III. STRUCTURAL AND OTHER SHAPES: 
 
 1. Plan of Study 461 
 
 2. Angles, Methods of Rolling 462 
 
 a. The Three Methods Compared 462 
 
 3. The Channel 464 
 
 4. Beams, Ties, and Piling 464 
 
 5. Zees and Tees 466 
 
 6. Finishing Sections 466 
 
 7. Rounds 467 
 
 8. Cutting and Straightening Rounds 468 
 
 9. Flats 468 
 
 10. Hexagons 469 
 
 11. Deformed Bars 469 
 
 CHAPTER IX. The Rolling of Strip and Merchant Mill Products 
 
 SECTION I. STRIP, OR HOOP, MILLS AND THEIR PRODUCTS: 
 
 1. Meaning of the Word Hoop 470 
 
 2. Hoop as a Rolling Specialty 470 
 
 3. The Carnegie Hoop Mills 470 
 
 4. Methods of Rolling Hoop 471 
 
 5. Precautions Required in Rolling Hoop. 471 
 
 6. Finishing Hoop 473 
 
 SECTION II. MERCHANT MILLS: 
 
 1. What the Merchant Mill Is 474 
 
 2. Kinds of Merchant Mills 474 
 
 a. The Guide Mill 475 
 
 b. The Belgian and Looping Mills 477 
 
 c. The Semi-continuous or Combination Mill. . . . 477 
 
 d. The Cross Country Mill 479 
 
 3. Future Development 479 
 
 SECTION III. DESIGNING ROLLS AND MAKING UP SCHEDULES 
 FOR MERCHANT MILLS: 
 
 1. Roll Designing for Merchant Mills 482 
 
 2. Economic Features of Roll Designing 482 
 
 3. The Order in the Office 484 
 
 4. The Order at the Mill Size of Billet or Bloom . . 486 
 
TABLE OF CONTENTS 
 
 SECTION IV. ROLLING PRACTICE IN MERCHANT MILLS: 
 
 1. The Roller His Importance 487 
 
 2. Precautions in Rolling 487 
 
 3. Rolling Defects 488 
 
 4. Two Different Finishes 489 
 
 a. Common Finish 489 
 
 b. Special Finish 489 
 
 I 
 
 SECTION V. SHEARING AND BUNDLING MERCHANT MILL 
 PRODUCTS: 
 
 1. The Methods of Shearing and Bundling 490 
 
 2. Duties of the Shear Foreman 490 
 
 3. Bundling Export Material 491 
 
 4. Special Bundling 491 
 
 5. Handling the Material in the Warehouse 491 
 
 6. Straightening 492 
 
 7. Invoicing 492 
 
 SECTION VI. INSPECTION DEPARTMENT OF A MERCHANT MILL 
 PLANT: 
 
 1. The Inspection Department 493 
 
 2. Function of the Inspection Department 493 
 
 3. Surface Defects 494 
 
 a. Buckles and Kinks 494 
 
 b. Fins 494 
 
 c. Underfills 494 
 
 d. Slivers 494 
 
 e. Laps 494 
 
 f . Seams 494 
 
 g. Burned Steel 495 
 
 h. Roll Marks 495 
 
 i. Finish 495 
 
 j. Pipe 495 
 
 4. Testing for Defects 495 
 
 5. Other Duties of Inspectors 495 
 
 6. Manner of Gauging Different Sections 496 
 
 CHAPTER X. Circular Shapes. 
 
 SECTION I. SOME GENERAL FEATURES PERTAINING TO THE 
 ROLLING OF CIRCULAR SHAPES: 
 
 1. The Rolling of Circular Shapes 497 
 
 2. Preparing the Blanks 497 
 
TABLE OF CONTENTS 
 
 SECTION II. THE CARNEGIE SCHOEN METHOD FOR MANU- 
 FACTURING STEEL WHEELS: 
 
 1. The Carnegie Schoen Method 498 
 
 a. Forging the Blanks: 
 
 i. First Method of Forging 499 
 
 ii. Second Method of Forging . . . . 500 
 
 b. Rolling the Forged Blank: 
 
 i. The Rolling Mill 500 
 
 ii. The Rolling Process 503 
 
 iii. Effect of the Rolling 504 
 
 c. Punching Web Holes and Coning 504 
 
 d. Inspection of Carnegie Schoen Wheels 504 
 
 e. Heat Treating Car Wheels 506 
 
 2. The Forging of Circular Shapes 507 
 
 CHAPTER XI. Forging of Axles, Shafts and Other Round 
 Shapes. 
 
 SECTION I. HOWARD AXLE WORKS AS AN EXAMPLE OF A 
 FORGING SHOP: 
 
 1. The Plant and its Equipment 508 
 
 2. Precautions to be Observed in the Manufacture of Axles 508 
 
 3. Inspection of the Blooms 509 
 
 4. Heating the Blooms 509 
 
 5. The Rolling and Forging Operation 509 
 
 a. Advantages of the Method 510 
 
 SECTION II. FINISHING PROCESSES FOR FORCINGS: 
 
 1. Straightening 511 
 
 2. Cutting-off and Centering 511 
 
 3. Rough Turning 512 
 
 4. Hollow Boring 512 
 
 a. Piping and Segregation 512 
 
 b. Strength and Weight 512 
 
 c. Hollow Boring and Heat Treating 513 
 
 5. The Heat Treating Plant 513 
 
 a. The Furnaces 513 
 
 b. The Quenching Tanks : 514 
 
 c. The Testing Equipment '. 515 
 
 d. Advantages of Heat Treating Axles 515 
 
xxxiv TABLE OF CONTENTS 
 
 PART III. 
 
 THE CONSTITUTION, HEAT TREATMENT 
 AND COMPOSITION OF STEEL. 
 
 1. INTRODUCTORY 516 
 
 CHAPTER I. The Constitution and Structure of Plain Steel. 
 
 SECTION I. STEEL AS AN ALLOY OF IRON AND CARBON: 
 
 1. The Constituents of Steel 518 
 
 a. Ferrite 518 
 
 b. Cementite 518 
 
 c. Pearlite 520 
 
 2. Manner of Freezing of Solutions and Alloys 520 
 
 a. An Example of the First Class of Solutions . . . 520 
 
 b. Example of the Second Class of Solutions Salt 
 
 and Water 521 
 
 c. Lead and Tin Solutions as Another Example of 
 
 the Second Kind of Freezing 523 
 
 d. The Iron-Carbon Eutectic 524 
 
 i. Formation of Pearlite and the Eutectoid. . 525 
 
 3. Structural Composition of Slowly Cooled Steel 526 
 
 a. Effect of the Constituents Formed Upon the 
 
 Physical Properties 526 
 
 SECTION II. THERMAL CRITICAL POINTS OF STEEL: 
 
 1 . Nature of Critical Points or Ranges of Steel 527 
 
 2. Thermal Critical Point for Eutectoid Steel 527 
 
 3. Thermal Critical Points for Pure Iron 528 
 
 4. Thermal Critical Points of Low Carbon Steel 528 
 
 5. Thermal Critical Points of Medium Carbon Steel. . . 528 
 
 6. The Carbon-Iron Diagram for Steels and Methods 
 
 of Notation 529 
 
 7. The Position of the Critical Ranges 530 
 
 8. Changes at the Thermal Critical Points 530 
 
 a. Changes at AS 530 
 
 b. Changes at Ag 531 
 
 i. The Magnetic Properties 531 
 
 c. Changes at AS, 2 531 
 
 d. Changes at AI 53 1 
 
 9. Causes of the Thermal Critical Points in Steel... 532 
 
TABLE OF CONTENTS 
 
 SECTION III. THE CRYSTALLINE STRUCTURE OF STEEL: 
 
 1 . Crystals and Grains 533 
 
 2. Crystallization of Steel: 533 
 
 a. Crystallization of Eutectoid Steels 534 
 
 b. Crystallization of Hypo-Eutectoid Steel 534 
 
 c. Crystallization of Hyper-Eutectoid Steels 535 
 
 d. The Effect of Work on Grain Size 535 
 
 e. Crystalline Changes on Heating Steel: 535 
 
 i. Crystalline Refinement on Heating 536 
 
 3. Practical Importance of Grain Structure 537 
 
 4. Summary of Chapter I 538 
 
 CHAPTER II. Heat Treating Theory and Practice. 
 
 INTRODUCTION 539 
 
 SECTION I. ANNEALING: 
 
 1. The Annealing Operation 539 
 
 2. Purpose of Annealing 539 
 
 3. True Annealing and "Process" or "Works" Annealing 540 
 
 4. Heating for True Annealing: 540 
 
 a. Importance of Time in Heating for Annealing. . 542 
 
 5. Cooling 542 
 
 a. Effect of Cooling on the Net- Work 543 
 
 b. The Effect of Cooling Upon Pearlite 543 
 
 c. Other Factors 545 
 
 d. Methods of Cooling 545 
 
 e. Combination Methods of Cooling 545 
 
 6. Double Annealing 546 
 
 7. Box Annealing 546 
 
 8. Annealing Hyper-Eutectoid Steels 547 
 
 9. Normalizing and Spheroidizing 547 
 
 SECTION II. HARDENING: 
 
 1. The Hardening Operation 548 
 
 2. Heating for Hardening 548 
 
 3. Cooling for Hardening 549 
 
 a. Cooling or Quenching Media 551 
 
 b. Combination Methods of Quenching 551 
 
 c. Manner of Quenching 553 
 
 4. Progressive Hardening 553 
 
 5. Hardening Eutectoid Steels 553 
 
 6. Hardening Hyper-Eutectoid Steels 553 
 
 7. Hardening Hypo-Eutectoid Steels 554 
 
TABLE OF CONTENTS 
 
 SECTION III. THE TEMPERING OF HARDENED STEEL: 
 
 1. The Tempering Process 554 
 
 2. Nature and Theory of Tempering 555 
 
 3. Methods of Determining Tempering Temperatures. 556 
 
 4. Influence of Time in Tempering 557 
 
 5. Physical Properties Affected by Tempering 557 
 
 6. Tempering the Steels of Different Structural 
 
 Composition 557 
 
 a. Tempering Austenitic Steels 558 
 
 b. Tempering Martensitic Steels 558 
 
 c. Tempering Troostitic Steels 558 
 
 d. Sorbite.. 559 
 
 SECTION IV. THE TOUGHENING OF STEEL: 
 
 1. Toughening Defined 559 
 
 2. Benefits of Toughening 559 
 
 3. Quenching for Toughening 559 
 
 4. Change of Constituents Due to Toughening Fig. 118. . 560 
 
 5. Physical Properties of Toughened Steel Tab. 62 561 
 
 SECTION V. CASE HARDENING: 
 
 1. The Process of Carburizing Iron 562 
 
 2. Application of Case Hardening 562 
 
 3. The Two Periods of the Case Hardening Process .... 562 
 
 4. Kinds of Steel Suitable for Case Hardening 563 
 
 5. Case Hardening Properties of the Elements: 
 
 a. Carbon 563 
 
 b. Manganese 563 
 
 c. Silicon 563 
 
 d. Phosphorus and Sulphur 563 
 
 e. Nickel 563 
 
 f . Vanadium 563 
 
 g. Chromium 564 
 
 6. The Carburizing Agent 564 
 
 7. Carburizing Materials: 564 
 
 a. Packing and the Action of Charcoal Carburizer 564 
 
 b. Carburizing Mixtures and Compounds 565 
 
 8. Heating the Carburizing Pack 566 
 
 a. Controlling the Temperature 567 
 
 9. Removal of the Articles from the Boxes After 
 
 Carburizing 567 
 
 10. Heat Treatment of Case Hardened Articles 568 
 
 11. Superficial Hardening 568 
 
TABLE OF CONTENTS xxxvii 
 
 CHAPTER III. Constituent Elements of Commercial Carbon 
 Steel and Their Influence Upon Its 
 Mechanical Properties. 
 
 INTRODUCTORY 569 
 
 1. Properties of Iron 569 
 
 2. Effect of Carbon 570 
 
 3. Influence of Manganese 570 
 
 a. Influence of Manganese in Heat Treatment . . . 571 
 
 b. Influence of Manganese on Sulphur 572 
 
 4. Influence of Sulphur 572 
 
 a. Why Manganese Neutralizes Effect of Sulphur . 572 
 
 b. Uses for Sulphur in Steel 572 
 
 5. Influence of Phosphorus 573 
 
 a. The Two Evils of Phosphorus 574 
 
 6. Influence of Silicon 574 
 
 7. The Influence of Oxygen 575 
 
 8. Combined Effect of the Elements on Tensile Strength 
 
 of Steel 575 
 
 9. The Influence of Copper 576 
 
 10. Influence of Tin 577 
 
 11. Influence of Arsenic 577 
 
 CHAPTER IV. Alloy Steels. 
 
 SECTION I. INTRODUCTORY. 
 
 1. Definitions 578 
 
 2. Carnegie Types and Grades 579 
 
 SECTION II. NICKEL STEEL: 
 
 1. Manufacture of Simple Nickel Steel 580 
 
 2. The Different Nickel Steels and Their General 
 
 Characteristics 581 
 
 3. Reasons for These Peculiarities of the Nickel Steels . . 582 
 
 4. Structural Changes Due to Nickel 584 
 
 a. Pearlitic-Nickel Steels 584 
 
 b. Martensitic-Nickel Steels 584 
 
 c. Austenitic-Nickel Steels 584 
 
 5. The Constitutional Theory of Ternary Steels 584 
 
 6. Heat Treating Pearlitic Nickel Steels 585 
 
 SECTION III. CHROME STEEL: 
 
 1. The Manufacture of Simple Chromium Steels 587 
 
 2. Influence of Chromium 587 
 
 a. The Microscopic Constituents of the Chrome 
 
 Steels 588 
 
 3. Uses of the Simple Chrome Steels : 589 
 
 4. Heat Treatment of Chrome Steel. . 589 
 
xxxviii TABLE OF CONTENTS 
 
 SECTION IV. CHROME NICKEL STEELS: 
 
 1. Influence of Chromium and Nickel When Combined. 590 
 
 2. Types of Chrome-Nickel Steel 590 
 
 3. Mayari Steel 591 
 
 4. Uses of Chrome-Nickel Steels 591 
 
 5. Heat Treatment of Chrome-Nickel Steels 592 
 
 6. Physical Properties of Chrome-Nickel Steels 593 
 
 SECTION V. VANADIUM STEELS: 
 
 1. Simple Vanadium Steels 595 
 
 2. Influence of Vanadium .- 595 
 
 SECTION VI. CHROME-VANADIUM STEELS: 
 
 1. Effect of Combining Chromium and Vanadium 596 
 
 2. Properties and Uses of Chrome-Vanadium Steels 596 
 
PART 1. 
 
 THE MAKING OF STEEL. 
 
 CHAPTER 1. 
 
 SOME FUNDAMENTAL PRINCIPLES OF PHYSICS AND 
 CHEMISTRY. 
 
 SECTION 1. 
 
 INTRODUCTION. 
 
 1. Iron, the Master Metal: In beginning this very brief study 
 of the metallurgy of the most important metal of a metallic age, it is difficult 
 to refrain from pointing out a few of the qualities that have made iron the 
 master metal, although its importance really needs no comment here. A 
 little reflection shows it to be as vital to modern civilization as air and water 
 are to life; and it has become so common that, like air and water, its true 
 importance is lost sight of by most people, who look upon its abundance as 
 a matter of course and value it accordingly . No other one metal has contributed 
 so much to the welfare and comfort of man. There is scarcely an article 
 used in our daily lives that has not been produced from iron or by means 
 of it. Consider bread as an example. Plows made of iron turn the soil, 
 harrows of iron level it, and drills of iron sow the seed; machines of iron harvest 
 the wheat and thrash it; rolls of iron crush the grain to separate the flour; 
 engines of iron bring the flour to our homes, where it is made into dough in 
 iron pans and baked in an iron stove; finally the bread is sliced from the loaf 
 with an iron knife, and served to us at a table made with iron tools. It has no 
 exact substitute in nature, and without it most of our modern conveniences 
 would have been impossible of development. The railroads, the automobile, 
 and the watch are three of the many notable examples of such conveniences 
 No other metal is capable of giving the great range in physical properties, 
 that makes iron available for an almost unlimited number of purposes. Thus, 
 from our towering skyscrapers, our massive bridges and our immense ships, 
 where, as great beams, cables and plates, it supports loads almost greater 
 than the mind can conceive, we can trace it even to our parlors, where, as 
 invisible hairpins, it supports milady's tresses and, as the strings of her piano, 
 sends forth at her magic touch sweet sounds of melody. One property which 
 it possesses in a far greater degree. than any of the other metals is that of 
 magnetism. This property is so pronounced in iron and so slight in other 
 metals that, from a practical viewpoint, iron and one of its compounds may 
 be considered as the only magnetic substances. Hence, our modern magnetic 
 
2 MFTA-LLURGY, MATTER 
 
 and electrical' Appliances "are <kpendent upon this one metal; and we find 
 it forming the essential parts of the dynamo, the electric motor, the telegraph, 
 the telephone, the wireless telegraph, the compass, and a large number of other 
 instruments of less importance. And so we might continue at great length 
 upon this one topic of the importance of iron, but our time is too short to 
 permit our giving much of it to a theme which the reader may develop for 
 himself. Hastening on, then, to more important matters, we find the first 
 question that confronts us is, What is Metallurgy? 
 
 Metallurgy: In general, Metallurgy is defined as the science which 
 deals with the preparation of the metals and their adaptation to the uses 
 for which they are intended. It is an advanced and specialized science, 
 hence a difficult one. Even a slight understanding of the subject requires 
 a previous knowledge of the fundamental sciences of Physics and Chemistry. 
 For those who may not have had the necessary preparation in these pre- 
 requisites, this study is. becomingly introduced by a brief consideration of 
 some of the more important principles of these two sciences. To present 
 these principles in as concise and simple a manner as possible is the object 
 of this chapter. 
 
 Matter : Through the various senses of sight, touch and hearing, the 
 human intellect becomes aware of the existence of things which, collectively, 
 are called matter. Limited portions of space that contain matter are 
 termed bodies, and the different kinds of matter are, in general, spoken 
 of as substances. Matter is a fundamental thing and cannot be accurately 
 defined. It is described by its properties, which will be discussed later. 
 
 The Fundamental Law and the States of Matter: Certain facts 
 about matter, however, are plainly evident. It occupies space, and can 
 be neither increased nor decreased in amount. These last two facts are 
 commonly known as the Law of the Conservation of Matter. It exists 
 in any one of three states; solids, which have definite masses, sizes and 
 shapes; liquids, which have definite masses and sizes but not form; and gases, 
 which possess definite masses only. A common example is water, which 
 at ordinary temperatures exists in all three states; namely, ice, water and 
 vapor. Liquids and gases together are called fluids on account of their 
 flowing properties, and in many instances they are subject to the same laws. 
 They are distinguished from each other by their relative compressibility. 
 Liquids are but slightly compressible, while gases are highly compressible. 
 The volume of a gas varies inversely as the pressure applied to it. For 
 example, if a certain mass of gas has a volume of 10 cu. ft. under a pressure of 
 100 Ibs., the same mass of gas will occupy but 5 cu. ft. at 200 Ibs. pressure. 
 
 Molecules: Furthermore, while the conception may seem difficult to 
 establish as a fact, there are strong reasons for believing that the relatively 
 large bodies, in which form matter makes itself evident to the human 
 are composed of minute particles, called molecules. This belief is 
 
MOLECULES 
 
 founded upon many facts, and can be arrived at by some such process of 
 reasoning as follows: Mental conception concerning the constitution of 
 matter may be based on either one of two hypotheses; namely, that matter 
 is infinitely divisible or that it is made up of small particles. According 
 to the first hypothesis, a body of matter could be divided indefinitely, if 
 means were available to make such a process possible, without changing 
 any of its characteristics, except its size; that is, a piece of chalk, for 
 example, would remain chalk even as the particles resulting from the infinite 
 division approached zero in size and weight. This retention of original 
 characteristics would imply that each individual kind of matter, the chalk 
 in the present instance, is an elementary substance. But this conclusion 
 is contrary to the facts, for it is a matter of common knowledge that many 
 substances like iron ore, limestone, sugar, etc., are composed of substances 
 quite different from the original. Thus, through the application of heat 
 alone, limestone and sugar are decomposed, the former into quick lime 
 and carbonic acid gas, and the latter into carbon, or charcoal, and water. 
 Only the second hypothesis remains, and it agrees with these facts, for ft 
 assumes that the larger masses of limestone and sugar, so evident to our 
 senses, are made up of small particles, each of which is composed of 
 portions of these simpler things into which the limestone and sugar are 
 decomposed by the heat. These ultimate particles of the different 
 substances are called molecules, which are, therefore, defined as the 
 smallest particles of a substance that retain the characteristics of that 
 substance. 
 
 Sciences of Matter: Two great classes of matter are evident; animate 
 or living matter, such as the living bodies of plants and animals, and 
 inanimate matter, such as glass, water, air, etc. Animate matter is treated 
 of by the sciences of Biology, Zoology and Botany; inanimate, by Physics 
 and Chemistry, formerly included under the one head of Natural Philosophy. 
 Thus, Metallurgy may be looked upon as a highly specialized branch of 
 Natural Philosophy. All these sciences are so closely related that a 
 knowledge of all is essential to a complete understanding of any one. 
 
 SECTION II. 
 
 SOME PHYSICAL PROPERTIES OF MATTER. 
 
 Properties : That a better understanding of matter may be obtained 
 from a study of its properties has already been indicated. By properties 
 is meant those characteristics by which the different kinds of matter are 
 distinguished and by which it may be described and defined. They are 
 of two classes, namely, general and special. General properties are 
 common to all matter, while special properties are peculiar to certain kinds 
 of matter only. The general properties are as follows: 
 
 Inertia: This property causes matter to resist any attempt to change 
 its state of rest or motion. 
 
PHYSICAL PROPERTIES 
 
 Extension is that property by virtue of which matter occupies space. 
 There are two systems of measuring extension, the English and the metric. 
 In the English system, the linear unit is the yard, while the volumetric 
 units, as established by custom, are the gallon, the bushel, and the cubic 
 yard. Corresponding units in the metric system are the meter=l. 09361 
 yards=39. 37 inches; the kilometer=.62137 mile; the liter=.26417 gallon= 
 1.0567 quarts, liquid, or .908 quart, dry; and the cubic meter=1.308 cubic 
 yards. 
 
 Mass refers to the amount of matter. It is measured in grams, which 
 is the mass of one cubic centimeter of pure water at the temperature of its 
 greatest density, 4 centigrade. Commercially, the unit is the kilogram, 
 equal to 1000 grams. From a scientific standpoint there is no exact English 
 equivalent, because weight involves the force of gravity, which may vary, 
 whilst mass is constant. However, the pound has been standardized so 
 that one kilogram=2. 20462 pounds. 
 
 Density is the weight or mass of a unit volume of matter. It is usually 
 expressed in grams per cubic centimeter. 
 
 Specific Gravity is the number of times a body is heavier than an 
 equal volume of some substance used as a standard. For liquids and solids 
 this standard is water; for gases it is air or hydrogen. In the metric system 
 density and specific gravity are numerically the same, since the weight of 
 one cubic centimeter of water is one gram. 
 
 Porosity: All matter is porous. The molecules, it is thought, are 
 separated, even in the densest materials, by spaces larger than the mole- 
 cules themselves. 
 
 Impenetrability: Two bodies of matter cannot occupy the same space 
 at the same time, and to this property of matter the term impenetrability 
 is applied. 
 
 Special Properties: The chief special properties of matter, some of 
 which are of supreme importance in the manufacture of steel, are as follows: 
 
 Cohesion and Adhesion : According to the law of gravitation, every 
 particle of matter in the physical universe attracts every other particle 
 with a force whose direction is that of a line joining the two particles and 
 whose magnitude varies directly as the product of the two masses, and 
 inversely as the square of the distance between them. Applied to molecules, 
 this attraction is known as cohesion and adhesion; the former is the 
 attraction of molecules of the same kind for each other, the latter, the 
 attraction of unlike molecules. The clinging of a drop of water to the end 
 of a glass rod exemplifies both of these forces. 
 
 Elasticity is the power of matter to assume its original shape after 
 having been distorted. The property of cohesion causes all bodies to resist 
 change in form, but only solids have elasticity of form. When a solid body 
 
PHYSICAL PROPERTIES 
 
 is deformed, the resistance it offers is called the stress; the deformation 
 which produces this stress is called the strain. Hooke's law states that, 
 up to the elastic limit, the strain is proportional to the stress. In practice, 
 the stress is measured in terms of a force or forces applied externally 
 to the body being tested. There are four methods of calling forth the 
 elasticity of bodies: namely, by pressure, by stretching, by bending and 
 by twisting. Stretching and bending are the methods most commonly 
 employed in testing the elasticity of steel. 
 
 Plasticity is the opposite of elasticity. A plastic body once distorted 
 will not regain its original shape. 
 
 Ductility is sometimes defined as the property by virtue of which 
 matter may be drawn into fine wires. As the term is employed in the 
 testing of steel, ductility is the distortion or strain a body undergoes in 
 being ruptured. 
 
 Malleability: Some kinds of matter, metals in particular, can be 
 hammered or rolled into thin sheets. This property is called malleability. 
 
 Hardness is the ability to withstand abrasion, or resist penetration. 
 
 Crystallization : Some substances in changing from a liquid to a solid 
 form, separate not as a continuous compact mass but as bodies having a 
 definite shape and color, called crystals. That crystals may form, it is 
 necessary that the molecules be free to arrange themselves in a definite 
 order. This condition is secured when a substance is in solution or in a 
 molten state. In steel manufacture this property is of great importance. 
 
 Diffusion is most characteristic of liquids and gases. It is the property 
 that causes two fluids in contact to intermingle. Liquids diffuse slowly, 
 but gases much more rapidly. 
 
 Effusion is the term applied to that property of gases which causes 
 them to pass through porous solids. The rates of effusion of different gases 
 is inversely proportioned to the square roots of their relative weights. 
 
 Absorption: Many porous bodies, like coke, charcoal, platinum 
 sponge, etc., are capable of absorbing large quantities of gases. Thus, 
 one cubic centimeter of charcoal is capable of absorbing from thirty to 
 thirty-five times its own volume of carbon dioxide. Gases are condensed 
 on the surface of all solids, and porous bodies offer a large surface for con- 
 densation. 
 
 SECTION III. 
 
 ENERGY, HEAT AND TEMPERATURE, AND THE ETHER. 
 
 Energy : Physics and Chemistry, however, have to do with more than 
 matter. The senses also reveal the presence of a second factor in nature, 
 called energy. Like matter, energy is a fundamental that cannot be 
 satisfactorily defined. It is not a thing. It is that which gives a body 
 the ability to move against a resistance; that is, the ability to do work. 
 
ENERGY AND HEAT 
 
 Thus, a body may possess energy and still neither move nor do any work. 
 Like matter, energy is conserved. It may be changed from one form to 
 another or be transferred from one point to another, but the total energy 
 of the Universe remains constant. This fact is known as the Law of Con- 
 servation of Energy. 
 
 Kinds of Energy: There are two kinds of energy; namely, potential 
 or stored up energy, sometimes called energy of position, and kinetic energy, 
 or the energy possessed by a body by virtue of its motion. Thus, a weight 
 on the top of a building possesses potential energy with respect to the 
 ground by virtue of its position; if it is permitted to fall, its energy then 
 becomes kinetic. Energy is measured in terms of the work which it is 
 capable of doing. 
 
 Heat and Temperature: One form of energy is heat. Heat must 
 not be confused with temperature. The latter measures one of the effects 
 of the former. The difference may be illustrated thus: Let it be supposed 
 that two portions of natural gas, each of a cubic foot, are burned completely, 
 so that the heat liberated is entirely absorbed by two bodies of water 
 initially at the same temperature, the volumes of which are a quart and 
 a gallon, respectively. It is evident from common experience that the 
 temperature of the smaller portion of water will be raised the higher, though 
 the quantity of heat imparted to each is precisely the same. 
 
 Effects of Heat : Heat produces marked effects on matter. All matter 
 expands by the application of heat alone, though there are many apparent 
 exceptions. The volume of a gas varies directly as the absolute temper= 
 ature, other conditions remaining constant. Change of state may be 
 caused by heat. Thus, glass or iron, dense solids at ordinary temperatures, 
 readily assume the fluid state on being heated above their fusion point. 
 According to the kinetic theory of heat, the molecules of matter always 
 have a certain amount of independent motion, and the effect of adding 
 heat is to increase the energy of this motion, the molecules being thereby 
 forced farther and farther apart. This forcing apart of the molecules 
 accounts for the change of state as well as the expansion of bodies on being 
 heated. 
 
 Temperature is determined by measuring the expansion it produces in 
 a volume of mercury enclosed in a small glass tube, called a thermometer. 
 The length of the tube is marked off into small divisions, which constitutes 
 the scale of the thermometer. There are four thermometer scales in 
 common use; the Centigrade, Fahrenheit, Reaumur and Absolute. The 
 difference among them consists of the number of divisions between the 
 freezing point and the boiling point of water, and the numbers applied to 
 these divisions. 
 
 The Centigrade is the thermometer employed in all scientific work. 
 In it the freezing point is marked zero and the boiling point 100. The only 
 
ENERGY AND HEAT 
 
 difference between this scale and the Absolute is that, in the latter,^ the 
 zero point is 273 below the Centigrade zero. 
 
 In the Fahrenheit thermometer, the space between the freezing and the 
 boiling points is divided into 180 equal parts, and zero is 32 of these parts 
 below the freezing point of water. The boiling point, therefore, is 212. 
 
 In the Reaumur scale, the 
 freezing point is marked zero 
 and the boiling point 80. For 
 high temperatures, instruments 
 called pyrometers are used. 
 
 These relations of the 
 various scales are shown in the 
 accompanying diagram: From 
 this diagram the following for- 
 mulas are readily developed: 
 
 Temp. A=Temp. C+273. 
 Temp. C=Temp. A 273. 
 Temp. F=Temp. %C+32. 
 Temp. C=Temp. (F 32)%. 
 
 Boiling Point 
 
 Cioo c 
 
 F 212 
 
 A373 fop" 
 
 of Water 
 
 Degrees 
 Intervening; , 
 
 1 
 
 1 
 
 B 
 
 | 
 
 
 ft 
 
 
 
 freezing Point , 
 
 
 , 
 
 
 32 > 
 
 
 273 , 
 
 
 
 
 of Water 
 
 
 
 O 5 
 
 FIQ. 1. Diagram showing relations of the 
 various thermometer scales. 
 
 
 Measurement of Heat: Heat is measured in calories. A calorie (cal.) 
 is the heat required to raise the temperature of one gram of water one 
 degree centigrade. In practice the large calorie (Cal.) is employed. It 
 is the amount of heat required to raise one kilogram of water through 
 one degree centigrade. The corresponding unit in the English system is 
 the B. t. u. (British thermal unit), which is the heat required to raise 
 the temperature of one pound of water one degree Fahrenheit. These units 
 may be converted from one to the other by use of the following factors: 
 
 1 Calorie=3.968 B. t. u. or 
 1 B. t. u.= .252 Cal. 
 
 The Ether: A third factor composing the Universe is the Ether. 
 Little is known about it except that it fills all space, permeates all matter 
 and transmits light, heat, and electric waves. Its properties are very 
 difficult to analyze, because the senses are not directly affected by it. Its 
 presence was first suspected through the study of the transmission of light. 
 By comparatively simple experiments, it was shown as early as 1802 that 
 light is transmitted by a wave motion, and since light is transmitted through 
 a vacuum, something other than matter must act as the medium. The 
 same conclusion is arrived at by a study of heat radiation. The develop- 
 ment of wireless telegraphy was based on this supposition, and its success 
 is further evidence of the existence of the Ether. 
 
CHANGES IN MATTER 
 
 SECTION IV. 
 
 CHANGES IN MATTER. 
 
 Physical and Chemical Changes: Matter is constantly undergoing 
 changes. A close observer soon discerns that these changes are of two 
 kinds, namely, one in which the nature and composition of the matter 
 undergoing the change remains the same, called a physical change, and 
 another in which the nature and composition are affected, called achemical 
 change. The bending of a stick, the freezing of water, the fusion of steel 
 are examples of the former, while the burning of coal is a common example 
 of the second. In many physical changes and in all chemical changes heat 
 is involved, being either absorbed or liberated. Chemical changes that 
 liberate heat furnish a source of energy chemical energy. 
 
 The Make=up of Material Bodies: In considering the make-up of 
 the various bodies of matter, it is necessary to distinguish between mere 
 mixtures and more closely combined substances. A mechanical mixture 
 is a mixture of two or more substances, which is not homogeneous and the 
 components of which can be separated by mechanical means. Such mix- 
 tures are made up of molecules of different kinds. A chemical compound 
 is homogeneous throughout its mass, and its components cannot be separat- 
 ed by mechanical means, that is, its molecules are all of the same kind. The 
 components of any given chemical compound are always in the same pro- 
 portions for that compound. This fact distinguishes compounds from 
 alloys and solutions which, though they are practically homogeneous and 
 sometimes are practically impossible of separation by mechanical means, 
 are never constant in composition. An alloy is a solid solution of one 
 metal in another. 
 
 Kinds of Chemical Compounds: A close study of a great number 
 of chemical compounds will show that all substances fall into four classes; 
 namely, acids, bases, salts and non-electrolytes. 
 
 Acids are characterized by the fact that they all have a sour taste 
 when in water solution and change the color of certain chemicals, called 
 indicators. One of the most common of these indicators is litmus, of 
 which there are two colors, a blue and a red. Acids change the color of 
 blue litumus to red. Vinegar is chiefly a dilute solution of acetic acid. 
 
 Bases have the power of neutralizing acids, and may be looked upon 
 as their opposites. Examples are quick lime, lye, etc. Bases change the 
 color of red or neutral litmus to blue. 
 
 A Salt is the product formed when an acid is neutralized by a base. 
 Common table salt, made by neutralizing hydrochloric acid with sodium 
 carbonate, is an example. As a rule acids, bases, and salts are electrolytes, 
 that is, their water solutions will conduct the electric current. 
 
 Non Electrolytes : There are some compounds that do not resemble 
 either acids or bases, nor can they be classed as salts. They are char- 
 acterized by the fact that their water solutions will not conduct the electric 
 current, so are termed non-electrolytes. Benzene, methane and distilled 
 water are examples. 
 
CHANGES IN MATTER 
 
 Chemical Elements: Notwithstanding the fact that chemical com- 
 pounds are homogeneous and cannot be separated by mechanical means, 
 they are readily divided into simpler substances by chemical processes, 
 These simpler substances are called elements, and each and every chemical 
 compound is composed of two or more chemical elements. While the 
 number of chemical compounds is almost unlimited, there are compara- 
 tively few elements. In 1918, the discovery of 84 elements had been 
 reported. The total number lies between 92 and 97. Of these, twelve 
 compose about 99 per cent, of the earth's crust. It has been estimated 
 that the solid crust of the earth is made up approximately as follows: 
 
 Oxygen 49.85% Calcium 3. 18% Hydrogen 97% 
 
 Silicon 26.03% Sodium 2.33% Titanium 41% 
 
 Aluminum 7.28% Potassium 2.33% Chlorine 20% 
 
 Iron 4.12% Magnesium 2.11% Carbon 19% 
 
 TOTAL 99.00% 
 
 Classification of Chemical Elements: A study of the elements 
 reveals the fact that there are two great classes; namely, those that combine 
 with oxygen and hydrogen to form bases, and those that combine with oxygen 
 and hydrogen, or hydrogen alone, to form acids. The former are sometimes 
 called metals and the latter non=metals, or metalloids. The line of 
 division is not a sharp one. Some elements form both acids and bases, 
 but the tendency is more pronounced in the one direction than in the other. 
 Furthermore, like plants or animals, these two divisions may be sub-divided 
 into families or groups, the members of which possess similar properties. 
 These divisions and groups are shown in a subjoined table. 
 
 Symbols: For convenience and brevity, each element is represented 
 by a symbol. These symbols are composed of the first letter, capitalized, 
 of the English or Latin names of the elements, combined, where necessary 
 as a distinguishing mark, with some succeeding letter. Thus, C=carbon, 
 Ca=calcium, Cd=cadmium, F=fluorine, Fe=ferrum (iron), etc. 
 
 Fundamental Laws of Chemical Changes: Now, where the elements 
 combine to form a compound they always do so in definite proportions 
 by weight. Thus, fifty-six parts by weight of iron will combine with 
 sixteen parts by weight of oxygen, or fourteen parts by weight of nitrogen 
 with sixteen parts by weight of oxygen. This fact is known as the Law of 
 Definite Proportions, and the definite weights are called combining weights. 
 Further investigation along this line shows that some pairs of elements form 
 more than one compound and that the combining weights of the elements 
 in these different compounds are simple multiples of each other. Concisely 
 stated, the law is this: Whenever two elements unite to form more than 
 one compound, if we consider a fixed weight of the one, the weights of 
 the other which combine with it are integral multiples of one 
 another. This fact is known as the Law of Multiple Proportions, or 
 Dalton's second law. The following compounds formed by the two 
 
10 ATOMS 
 
 elements nitrogen and oxygen are well known examples : 
 
 COMPOUND PARTS BY WEIGHT PARTS BY WEIGHT 
 
 OF NITROGEN OF OXYGEN 
 
 Nitrous Oxide 28 16 
 
 Nitric Oxide 28 32 
 
 Nitrogen Trioxide 28 48 
 
 Nitrogen Peroxide 28 64 
 
 Nitrogen Pentoxide 28 80 
 
 SECTION V. 
 
 THE ATOMIC AND ELECTRON THEORIES. 
 
 Atoms: Upon the facts just stated in the preceding section, the 
 English chemist Dalton founded a very important hypothesis, now known 
 as the atomic theory. In order to explain the laws stated above, reasoning 
 led to the following assumptions: 
 
 1st: The molecules of matter are themselves made up of small particles. 
 
 2d: These particles possess the power of attracting other particles or 
 otherwise attaching themselves to them. 
 
 3d: These particles do not subdivide in taking part in chemical changes. 
 
 These particles are called atoms. All the atoms of the same element 
 have the same mass or weight, the same form, and the same combining 
 power, while atoms of different elements differ in one or more of these 
 respects. 
 
 Atomic Weights: The atom is so small that it is useless to hope that 
 its mass or weight will ever be determined absolutely. However, the 
 weight of one atom of an element must be proportional to the combining 
 weight of that element. Since the combining mass or weight of hydrogen 
 is the least of all the otjier elements, it is assumed that its atom is the 
 lightest. Therefore, the atomic weight of hydrogen was made one by 
 Dalton and the atomic weight of the other elements multiples of it. However, 
 since hydrogen forms with other elements comparatively few compounds 
 that can be used for atomic weight determinations and oxygen more than 
 any other element, it was decided later to make the latter element the 
 standard. Accordingly, the atomic weight of oxygen is made 16, and the 
 atomic weights of other elements are compared with it as a standard,thus 
 making hydrogen 1.008. This system' of comparative weights is known as 
 the international table of atomic weights. 
 
 Valence : Concerning the attractive power of the atoms of the various 
 elements, it may be pointed out that the law of multiple proportions indicates 
 that the atom of an element may combine with one or more atoms of another 
 element in forming compounds with it. Here, again, hydrogen is used as 
 a standard, for since its combining weight is the least of all the other 
 elements, it is assumed that the holding power of its atoms must also 
 be the least. Therefore, the valency of an atom is properly denned 
 as the number of hydrogen atoms it is capable of combining with 
 or replacing. The valencies of the atoms of the elements are by no 
 means fixed quantities, but vary in some cases from one to seven. 
 
ELEMENTS 
 
 11 
 
 The following table is intended to furnish a complete list of the elements, 
 with their symbols and atomic weights, and to show, also, the classification 
 and valencies of the more common ones. Very rare elements are placed 
 in a separate list. Elements that ordinarily are both acid and basic 
 are marked with an *, and those important in the manufacture of steel 
 are printed in Italics. 
 
 Table I The Chemical Elements. 
 
 Showing Physical Constants of the More Common Ones. 
 
 
 
 Group 
 
 Name 
 
 I 
 
 
 
 1920 
 
 Atomic 
 Weight 
 
 Valence 
 
 Specific 
 Gravity 
 
 Melting 
 Point 
 
 Boiling 
 Point 
 
 0=16 
 
 Water 
 
 Air 
 
 C 
 
 C 
 
 BASE FORMING 
 
 Potassium 
 Calcium. . . 
 
 Magnesium 
 Silver 
 Aluminum 
 
 Lead 
 Chromium 
 
 Manganese 
 Iron 
 
 Palladium. 
 Gold 
 
 Lithium 
 Potassium 
 Sodium 
 
 Li 
 
 K 
 
 Na 
 
 Ca 
 Ba 
 Gl 
 
 Sr 
 
 Mg 
 Zn 
 Cd 
 
 Ag 
 Cu 
 Hg 
 
 Al 
 Ga 
 Tl 
 Sc 
 
 Pb 
 
 Sn 
 
 Cr 
 
 Mo 
 
 W 
 
 Mn 
 
 Fe 
 Co 
 
 Ni 
 
 Pd 
 Ru 
 Rh 
 
 Pt 
 Au 
 
 6.94 
 39.1 
 23.0 
 
 40.07 
 137.37 
 9.1 
 87.63 
 
 24.32 
 65.37 
 112.40 
 
 107.88 
 63.57 
 200.60 
 
 27.1 
 70.1 
 204.0 
 44.1 
 
 207.2 
 118.7 
 
 52.0 
 96.0 
 184.0 
 
 54.93 
 
 55.84 
 58.97 
 58.68 
 
 106.7 
 101.7 
 102.9 
 
 195.2 
 197.2 
 
 I 
 
 / 
 / 
 
 // 
 II 
 II 
 II 
 
 // 
 II 
 II 
 
 I 
 I-II 
 I-II 
 
 III 
 
 .59 
 
 .87 
 .97 
 
 1.58 
 3.8 
 1.85 
 2.5 
 
 1.7 
 7.1 
 8.6 
 
 10.5 
 8.9 
 13.60 
 
 2.58 
 5.95 
 11.85 
 
 11.3 
 77.3 
 
 6.9 
 8.8 
 18.8 
 
 7.4 
 
 7.78 
 8.7 
 8.7 
 
 11.8 
 8.6 
 12.1 
 
 21.5 
 19.3 
 
 
 186 
 62.5 
 97.6 
 
 810 
 850 
 960 
 900 
 
 650 
 419 
 321 
 
 961 
 1083 
 38.8 
 
 657 
 30 
 302 
 1200 
 
 327 
 232 
 
 1505 
 2500 
 1700 
 
 1225 
 
 1520 
 1610 
 
 1450 
 
 1550 
 1950 
 1940 
 
 1753 
 1062 
 
 1400 
 
 667 
 742 
 
 Calcium 
 
 Barium 
 Glucinum 
 
 950 
 
 Strontium 
 
 
 Magnesium 
 Zinc . . . 
 
 1120 
 918 
 
 778 
 
 1955 
 2100 
 357 
 
 1800 
 
 Cadmium 
 Silver 
 
 Copper 
 
 Mercury 
 * Aluminum. 
 
 Gallium (rare) . . . 
 Thallium " ... 
 Scandium " ... 
 
 *Lead 
 *Tin 
 
 Chromium . 
 
 
 II-IV 
 II-IV 
 
 II-III-VI 
 II-IV 
 
 II-III 
 II 
 II 
 
 IV 
 
 I-III 
 
 1525 
 1525 
 
 2200 
 
 * Molybdenum . . . 
 
 *Tungsten 
 * Manganese 
 
 Iron 
 Cobalt 
 
 1900 
 2450 
 
 Nickel 
 
 2530 
 
 *Palladium 
 Ruthenium (rare) 
 Rhodium (rare) . . 
 
 *Platinum 
 
 
 *Gold 
 
12 
 
 ELEMENTS 
 
 Table I The Chemical Elements Continued. 
 
 
 
 Group 
 
 Name 
 
 I 
 
 1920 
 Atomic 
 Weight 
 
 Valence 
 
 Specific 
 Gravity 
 
 Melting 
 Point 
 
 Boiling 
 Point 
 
 0=16 
 
 Water 
 
 Air 
 
 C 
 
 C 
 
 
 Chlorine . . 
 
 Fluorine 
 
 F 
 
 19.0 
 
 j 
 
 
 1.26 
 
 223 
 
 187 
 
 
 
 Chlorine.. 
 
 Cl 
 
 35.46 
 
 I 
 
 
 2.49 
 
 102 
 
 37.6 
 
 
 
 Bromine 
 
 Br 
 
 79.92 
 
 I 
 
 3.1 
 
 
 7.3 
 
 59 
 
 
 
 Iodine 
 
 I 
 
 126.92 
 
 I 
 
 4.9 
 
 
 114 
 
 184 
 
 
 Sulphur. . . 
 
 Sulphur 
 
 5 ' 
 
 32.06 
 
 II-IV-VI 
 
 2.0 
 
 
 116.5 
 
 444.6 
 
 
 
 Selenium (rare) . . 
 
 Se 
 
 79.20 
 
 
 4.3 
 
 
 218.5 
 
 690 
 
 
 
 Tellurium " 
 
 Te 
 
 127.50 
 
 
 6.2 
 
 
 451 
 
 1390 
 
 p 
 
 
 
 
 
 
 
 
 
 
 ^H 
 
 Carbon . . . 
 
 Carbon 
 
 C 
 
 12.0 
 
 IV 
 
 1.7-3.5 
 
 
 Infusible 
 
 3500 
 
 
 
 
 Silicon 
 
 Si 
 
 28.3 
 
 IV 
 
 2.49 
 
 
 1420 
 
 3500 
 
 ^-4 
 
 
 Boron. . 
 
 B 
 
 10.9 
 
 III 
 
 2.40 
 
 
 2250 
 
 3500 
 
 C 
 
 
 Titanium 
 
 Ti 
 
 48.10 
 
 IV 
 
 3.54 
 
 
 1795 
 
 
 g 
 
 Nitrogen . . 
 
 Nitrogen 
 
 N 
 
 14.01 
 
 I to V 
 
 
 .96 
 
 213 
 
 196 
 
 o 
 
 
 Phosphorus 
 
 P 
 
 31.04 
 
 III-V 
 
 1.8 
 
 
 44.1 
 
 290 
 
 <! 
 
 
 *Arsenic 
 
 As 
 
 74.96 
 
 III-V 
 
 5.7 
 
 
 185 
 
 449.5 
 
 
 
 * Antimony 
 
 Sb 
 
 120.20 
 
 III-V 
 
 6.6 
 
 
 630 
 
 1460 
 
 
 
 *Bismuth 
 
 Bi 
 
 208.0 
 
 III-V 
 
 9.7 
 
 
 269 
 
 1485 
 
 
 
 ^Vanadium 
 
 V 
 
 51.0 
 
 III-V 
 
 5.9 
 
 
 1680 
 
 
 
 
 Oxygen 
 
 
 
 16.0 
 
 II 
 
 
 1.10 
 
 218 
 
 182 
 
 
 
 
 
 (15.97) 
 
 
 
 
 
 
 
 
 Hydrogen 
 
 H 
 
 1.008 
 
 I 
 
 
 .07 
 
 259 
 
 252 
 
 
 
 
 
 (1.) 
 
 
 
 
 
 
 Name 
 
 51 Argon 
 
 52 Caesium 
 
 53 Cerium 
 
 54 Columbium. . . 
 
 55 Dysprosium... 
 
 56 Erbium 
 
 57 Europium 
 
 58 Gadolinium... 
 
 59 Germanium. . . 
 
 60 Helium 
 
 61 Holmium 
 
 62 Indium 
 
 63 Iridium 
 
 64 Krypton 
 
 65 Lanthanum . . . 
 
 66 Lutecium 
 
 67 Neodymium. . 
 
 Very Rare Elements. 
 
 Atomic 
 Symbol Weight Name 
 
 A 39.9 68 Neon 
 
 Cs 132.81 69 Niton 
 
 Ce 140.25 70 Osmium 
 
 Cb 93.1 71 Praseodymium 
 
 Dy 162.5 72 Polonium 
 
 Er 167.7 73 Radium 
 
 Eu 152.0 74 Rubidium 
 
 Gd 157.3 75 Samarium 
 
 Ge 72.50 76 Tantalum 
 
 He 4.0 77 Terbium 
 
 Ho 163.5 78 Thorium 
 
 In 114.8 79 Thulium 
 
 Ir 193.1 80 Uranium 
 
 Kr 82.92 81 Xenon 
 
 La 139.0 82 Ytterbium 
 
 Lu 175.0 83 Yttrium 
 
 Nd 144.30 84 Zirconium.. 
 
 Atomic 
 Symbol Weight 
 
 Ne 
 
 Nt 
 
 Os 
 
 Pr 
 
 Po 
 
 Ra 
 
 Rb 
 
 Sa 
 
 Ta 
 
 Tb 
 
 Th 
 
 Tm 
 
 U 
 
 Xe 
 
 Yb 
 
 Yt 
 
 Zr 
 
 20.2 
 222.4 
 190.9 
 140.9 
 210.0 
 226.0 
 
 85.45 
 150.4 
 181.5 
 159.2 
 232.15 
 168.5 
 238.2 
 130.2 
 173.5 
 
 89.33 
 
 90.6 
 
REACTIONS 13 
 
 Electrons: Until 1900 all the elements had resisted all efforts to 
 break them up into simpler substances, and atoms were considered to be 
 the smallest divisions of matter. With the discovery of radium and other 
 radio-active substances, however, a new field for investigation was opened, 
 and subsequent discoveries indicate that the atom is divisible. These very 
 small particles are called electrons. It is thought that electrons, in some 
 intimate relation to the Ether, are the fundamental particles of which all 
 matter is composed. 
 
 SECTION VI. 
 CHEMICAL FORMULA AND REACTIONS. 
 
 Chemical Formulas of Compounds: The method of representing the 
 elements by symbols, together with the system of atomic weights, affords 
 a convenient and concise method of representing chemical compounds, or 
 to be more explicit, the molecules of chemical compounds. Thus, by 
 analysis, water is found to be composed of hydrogen and oxygen in the 
 proportion of eight parts of oxj^gen to one part of hydrogen by weight. 
 These facts are completely expressed by the formula H 2 O, which indicates 
 a molecule of a compound composed of two atoms of hydrogen and one 
 atom of oxygen, or, since the atomic weight of hydrogen is 1 and of 
 oxygen, 16, 2 parts of hydrogen to 16 parts of oxygen (1 to 8). Likewise, 
 the formula Fe 2 O 3 represents a compound, the molecule of which is made 
 up of 111.68 parts of iron to 48 parts of oxygen. 
 
 Molecules of Elements: In studying chemical changes in which 
 elements are set free, it is found that they are much more active at the 
 instant of their liberation than afterwards, and are, therefore, said to be 
 in the nascent state at that instant. This fact leads to the belief that 
 the instant an element is set free from its compounds it exists in the atomic 
 condition, but if there is nothing else present with which the atoms can 
 combine, they combine with each other to form molecules of the element. 
 This idea cannot be proven in the case of solids, but its correctness is easily 
 shown in the case of gases. From many facts, Avogadro was able to show 
 that equal volumes of all gases, under the same conditions of temper- 
 ature and pressure, contain the same number of molecules. Hence, 
 the molecular weight in grams of all gases give a constant volume of 22.32 
 liters, called the gram-molecular volume. Now, the weight of 22.32 liters 
 of oxygen=32 gms., of hydrogen, 2 gms., of nitrogen, 28 gms. Dividing 
 these weights by the respective atomic weights of the elements, the quotient 
 is 2 in each case. Hence, the molecules of these elements contain two atoms 
 each, and the correct formulas for these elements are O 2 , H 2 and N 2 , 
 respectively. 
 
 Chemical Equations: This system of symbols and weights also 
 simplifies the representation of chemical changes. Suppose it is desired to 
 represent the chemical change that takes place when a common substance, 
 like coal for instance, burns. Coal is largely made up of carbon; the element 
 which combines with it is oxygen in the air; an invisible gas, CO 2 , is 
 
14 
 
 REACTIONS 
 
 formed and diffuses into the air. This change, spoken of as a reaction, 
 is represented in the form of an equation; thus, C+O=COs. 
 
 Balancing Reactions: Since matter is conserved, there must be as 
 many atoms on one side of the equation as on the other. This is shown by 
 placing a 2 before O on the left side of the equation, thus, C+2O=CO 2 . 
 This process is called balancing. Thus, reactions tell not only the names 
 of the reacting substances and of the products formed, but also give the 
 proportions by weight, and in the case of gases, volume relations as well. 
 
 Radicals: In .the molecules of many chemical compounds, certain 
 groups of atoms appear to be more closely bound together than others in 
 the same molecule. In these groups the atoms composing them appear to 
 bear a fixed relation to each other, which remains unchanged during a 
 chemical reaction. Thus, in many wet reactions in which H^SO^ is 
 employed as a reagent, the sulphur and oxygen do not separate but 
 remain closely combined, as illustrated in the reaction that takes place 
 between this acid and barium chloride: 
 
 H 2 (S0 4 )+Ba Cl 2 =Ba (SO 4 )+2 HC1 
 Such groups of atoms are called radicals. 
 
 Ions and Electrolysis: In electrolytes these radicals are readily 
 identified as ions. From a study of the effect of dissolved electrolytes on 
 the boiling and freezing points of the water in which they are dissolved, 
 and on their osmotic pressures, evidence is obtained to show that each of 
 the dissolved molecules breaks up or dissociates into two parts. 
 
 The following simple experi- 
 ment may be employed to throw 
 additional light upon this sub- 
 ject. Into the U-tube of Fig. 2 
 is placed a solution of sodium 
 sulphate and some neutral 
 litmus, into which is immersed 
 two small platinum rods to act 
 as electrodes for an electric 
 current, as shown in the figure. 
 Upon closing the circuit, bubbles 
 of hydrogen are given off at the 
 cathode and bubbles of oxygen 
 at the anode, while the solution 
 about the cathode becomes deep 
 blue in color, showing it is basic, 
 and that about the anode be- 
 comes red, showing it to be 
 acid. These facts are explained 
 by assuming that the molecules 
 
 Pi a. 2. Showing decomposition of water and 
 formation of sodium hydroxide and sulphuric 
 acid from sodium sulphate by electrolysis. 
 
REACTIONS 15 
 
 of dissolved Na 2 SO4 dissociate into parts, called ions. This 
 
 + + 
 dissociation is indicated thus: Na 2 SO4=Na-fNa+SO4. The sodium 
 
 + 
 ions, Na, carrying a positive charge of electricity, are propelled by 
 
 the current toward the cathode, while the negatively charged sulphions, 
 SO 4 , go to the anode. Here they give up their charges and become 
 chemically active, decomposing the water thus: 
 2Na+2H 2 O=2NaOH+H 2 
 
 This experiment is but one example of electrolysis. Any inorganic acid, 
 base or salt may be substituted for the sodium sulphate; and any conductor 
 of electricity, such as iron, may be used instead of platinum. It is to be 
 noted, however, that if iron had been used in this experiment, the anode 
 would have been corroded away by the acid radical; thus, Fe+SO4= 
 FeSO4. Electrolysis has been advanced to explain the corrosion of iron. 
 
 Dry and Wet Chemistry : Chemical changes take place under constant 
 conditions. Substances that will react in one way under one set of con- 
 ditions will not react, or react in an entirely different way, under another 
 set of conditions. Some substances react simply by contact, as quick lime 
 and water. Many reactions will take place only in a water solution, while 
 many other substances, being insoluble in water, must be heated almost 
 to their point of fusion before they react. A study of reactions between 
 substances in solution is called "wet chemistry" while the study of reactions 
 brought about by heat is termed ''dry chemistry." However, under the 
 same conditions, the same substances will always produce the same results. 
 This fact is known as the law of constancy of nature. 
 
 Acids, Bases and Salts of Dry Chemistry: Most substances dealt 
 with in wet chemistry lose water when heated. This statement is par- 
 ticularly true of inorganic acids, bases and salts. Thus, in the case of 
 acids and bases, heat breaks up these compounds into water and oxides, 
 called anhydrides. Acids give acid anhydrides, and bases, basic 
 anhydrides. These anhydrides constitute the acids and bases of dry 
 chemistry. They have the same power of neutralization that their 
 corresponding wet compounds possess, and form neutral compounds to which 
 the term slag is applied instead of salt. Many salts, in crystallizing from 
 aqueous solutions, unite with, or better, take up a definite amount of water, 
 which does not go to form a new compound, but to form crystals, and is 
 called, therefore, water of crystallization. This water is held very loosely 
 by the molecule and is readily given up by it. In some crystals, like those 
 of washing soda, for example, this tendency is so pronounced that they give 
 up their water of crystallization to the air, if its humidity is low. Such 
 substances are said to be efflorescent. On the other hand, many dry 
 substances absorb moisture from the air and are, therefore, said to be 
 
16 
 
 REACTIONS 
 
 hygroscopic. A few substances will absorb enough water from a very 
 moist air to become wet and actually go into solution in the water they 
 absorb. These substances are said to be deliquescent. The following 
 reactions will serve to illustrate these facts in so far as they involve chemical 
 changes: 
 
 H 2 SiO 8 -fheat=H 2 O+SiO 2 
 Metasilicic Acid Water Silica, or silicic anhydride 
 
 Mg (OH) 2 +heat=H 2 O+MgO 
 Magnesium hydroxide Magnesia a basic anhydride 
 
 Na 2 SO 4 .10H 2 O+heat=10H 2 O+Na 2 SO 4 
 Glauber Salt Sodium Sulphate 
 
 (Crystallized) (Dry Powder) 
 
 In this connection a study of the following table will also prove helpful. 
 
 Table 2. Acids, Bases and their Anhydrides with Salts 
 Resulting from Neutralization. 
 
 
 
 
 Salt With 
 
 Name of Compound 
 
 Formula of 
 Compound 
 
 Formula of 
 Anhydride 
 
 
 Univalent Base 
 
 Divalent Base 
 
 Trivalent Base 
 
 Sodium hydroxide 
 
 NaOH 
 
 Na->0 
 
 
 
 
 Calcium hydroxide 
 
 Ca(OH) 2 
 
 Cat) 
 
 
 
 
 Ferric hydroxide 
 
 Fe(OH) 3 
 
 Fe 2 3 
 
 
 
 
 Sulphuric Acid 
 
 H 2 S0 4 orH 2 0-S0 3 
 
 S0 3 
 
 NaoO-SOaor 
 
 CaO-SOacr 
 
 (Fe 2 3 ).(S0 3 )a 
 
 
 
 
 Na 2 S0 4 
 
 CaSO t 
 
 orFe 2 (BO*) 3 
 
 Nitric Acid 
 Orthophosphoric Acid 
 Orthosilicic Acid 
 
 HN0 3 or 
 H-,0-No0 5 
 H 3 P0 4 or 
 3H 2 0-P.0 5 
 H 4 Si0 4 ~or 
 2H 2 0-Si0 2 
 
 N 2 5 
 Si0 2 
 
 Na.0-N 2 5 or 
 NaNOs 
 3Na.>0-P-.0 5 or 
 Na 3 P0 4 
 2Na.,0-SiO.>or 
 
 CaO-N.,0 5 or 
 Ca(N0 3 )o 
 (CaO)-P 2 O r , 
 orCa :! (P0 4 ) 2 
 (CaO) 2 -Si0 2 
 orCa 2 Si0 4 
 
 Fe 2 3 -(N 2 5 )s 
 
 orFe(N0 3 ) 3 
 
 orFePOt 
 ofFe 4 (Si0 4 )" 3 
 
 Kinds of Reactions : As already indicated, all reactions may be placed 
 under one of two heads; namely, those that liberate heat, called 
 exothermic, and those that absorb heat, called endothermic. A more 
 detailed classification, such as the following, is sometimes employed: 
 
 1. Direct combination (synthesis) 2H+O=H 2 O or 2H 2 +O 2 =2H 2 O. 
 
 2. Direct decomposition (analysis) 2HgO=2Hg+O 2 . 
 
 3. Simple replacement or substitution 2H 2 O+2Na=2NaOH+H 2 . 
 
 4. Double replacement or metathesis BaCl 2 +H 2 SO 4 =BaSO 4 +2HC1. 
 
 /3 Fe+4O=Fe 3 O 4 . 
 < ni n , ,-, ~, 
 
 (Fe CI 2 +Cl=re Cls. 
 
 /Fe 3 O 4 +8H=3 Fe+4H 2 O. 
 "\Fe Cl 3 +H=Fe C1 2 +HC1. 
 The two processes of oxidation and reduction are of great importance 
 in metallurgy. They have a triple meaning. Primarily, oxidation means 
 the taking on of oxygen by an element or compound, and reduction means 
 the giving up of oxygen. In the case of elements that form more than one 
 
 5. Oxidation < 
 
 6. Reduction < 
 
REACTIONS 17 
 
 compound, if the number of atoms of one that combines with a fixed number 
 of the other be increased, the process is oxidation; if decreased, reduction. 
 In metallurgy an element in the metallic state is said to be reduced. The 
 two processes are inseparable; when one thing is reduced, another is 
 oxidized. In metallurgical operations these two processes are of paramount 
 importance, for all the substances reduced constitute the metallic product 
 and all in oxidized form make up the slag. 
 
 Some Laws Controlling Chemical Reactions: In writing reactions 
 considerable knowledge of a specific character is essential. Thus, suppose 
 it is required to write the reaction that represents the action of iron brought 
 in contact with water. First, it will be necessary to know under what 
 conditions the substances are brought together, for at ordinary temperatures 
 no reaction will take place. At a high temperature, a reaction takes place, 
 and it is necessary to know that ferroso-ferric oxide and hydrogen are 
 produced, and the formulas of all these substances. This knowledge can 
 then be indicated thus: Fe+H 2 O=Fe 3 O4+H 2 . Balancing the reaction, 
 which is done by inspection and arithmetic, is the next step. Finally, 
 the reaction is reversible, for if instead of steam over hot iron, hydrogen 
 be passed over hot iron oxide, iron and water are the products. The reaction 
 is, therefore, correctly written thus: Fe 3 O 4 +4H 2 =*= 3Fe+4H 2 O, or 3Fe+ 
 4H 2 O =*= Fe 3 O4+4H 2 . Many reversible reactions, under conditions which 
 do not permit the products to escape from the field of action, do not pro- 
 ceed to completion, but reach a balanced condition after a time and seem 
 to stop, though as a matter of fact they are progressing in one direction as 
 rapidly as in the other, hence are described as being in dynamic equilibrium. 
 In practical chemical work it is usually desirable to have reactions go to 
 an end. As an aid in writing reactions the following laws may be found of 
 value: 
 
 A. The reaction of two or more substances will go to an end, that is, 
 will be complete, provided, 
 
 1. One of the products is volatile at the temperature of the reaction. 
 
 2. One of the products is insoluble in the solvent in which the reaction 
 
 takes place. 
 
 3. One of the products is a non-electrolyte, that is, does not ionize in 
 
 the solvent. 
 
 B. The speed of a chemical action in a given direction may be increased 
 by effecting a greater concentration of one of the reacting substances. 
 This is a simple, non-mathematical statement of the law of mass action. 
 
 C. Chemical reactions always tend to proceed in the direction that 
 will liberate the most heat, and without the addition of heat from an external 
 source those substances that have the greatest heats of formation will 
 tend to form. 
 
18 CHEMICAL NOMENCLATURE 
 
 SECTION VII. 
 
 CHEMICAL NOMENCLATURE. 
 
 General Principle: A brief description of the nomenclature of 
 chemical compounds will be found of great assistance to those not familiar 
 with the subject. The names of the elements first discovered, and, there- 
 fore, unfortunately, the more common ones, are not based on any principle; 
 but of the more recently discovered elements the metals have received 
 names ending in um orium, and the metalloids, in n or ne. Jn the naming 
 of compounds, however, the old names have been discarded and new ones 
 substituted. The system employed in assigning these new names is this: 
 The name of a compound should show the elements of which it is 
 composed, and as far as possible their relative proportions. 
 
 Terminology of Binary Compounds: The simplest compounds are 
 those composed of only two elements. The names of all such compounds 
 are made up of the name of the basic element, if one is present, succeeded 
 by the name of the acid element, which ends in ide: examples; ferrous (iron) 
 sulphide, FeS; sodium chloride, NaCl; calcium oxide, CaO. 
 
 In such cases as iron and sulphur where the same two elements combine 
 to form more than one compound, the compounds, when two in number, are 
 distinguished by changing the ending of the metallic part of the name from 
 ous to ic; thus, ferrous sulphide, FeS; ferric sulphide, FeS 2 ; stannous 
 chloride, SnCl 2 ; stannic chloride, SnCl 4 . Often, the prefixes mono-, di-, tri-, 
 tetra-, pent-, per are used, especially if the name will not permit the ending 
 ous and ic, or if more than two compounds are formed by the same two 
 elements. Carbon dioxide, CO 2 ; nitrous oxide, or nitrogen monoxide, N 2 O; 
 nitric oxide or nitrogen dioxide, N 2 O 2 ; (NO); nitrogen trioxide, N 2 Os; 
 nitrogen tetroxide or nitrogen peroxide, N 2 O4,; nitrogen pentoxide, N 2 Os 
 are examples. 
 
 Terminology of Ternary Compounds: The names of compounds 
 ,that contain three elements, provided they are not derived from acids, 
 may end in ide, also, in which case all three of the elements appear in the 
 name, as sodium aluminum fluoride, (NasAlFe), bismuth oxychloride, 
 (BiOCl). A few ternary compounds have names ending in te (from ter, 
 three), as potassium chlorplatinate, (K 2 PtCl ). 
 
 Terminology of Acids: Acids are composed of the acid-forming 
 elements in combination with hydrogen or with hydrogen and oxygen. 
 The name of a given acid is derived from the name of the acid forming 
 element. The best known acid of an element has the ending ic. Example, 
 chloric acid, HC1O 3 . Then the acid the molecule of which contains one 
 less atom of oxygen has the ending ous. Example, chlorous acid, HC1O 2 . 
 If the element also forms an acid containing one more atom of oxygen in 
 its molecule than the ic acid, it is designated by the prefix per. Example, 
 
CHEMICAL CALCULATIONS 19 
 
 perchloric acid, HClO 4 . Acids, like sulphuric (H 2 SO 4 ) and ortho phos- 
 phoric (H 3 PO 4 ), which contain more than one replacable hydrogen atom 
 are called, as a class, polybasic acids; and, in individual cases, the different 
 acids are referred to as di basic, tri basic, etc., because such acids may be 
 neutralized by more than one base at once. For example, potassium and 
 sodium may replace the two hydrogens in H2SC>4 to form sodium potassium 
 sulphate, NaKSO^ In all such double salts both the base forming 
 elements must appear in the name of the salt. 
 
 Terminology of Bases: The base forming elements form compounds 
 with hydrogen and oxygen in which these two elements appear as a radical, 
 OH, called hydroxyl. Hence, these compounds are called hydroxides in 
 wet chemistry. Thus, sodium hydroxide, Na OH, and calcium hydroxide 
 Ca (OH) 2 are examples. In dry chemistry the hydrogen and part of the 
 oxygen are driven off, leaving the oxides which still act as acids and bases. 
 
 Terminology of Salts : Salts take their names from those of the base 
 forming elements and the acids of which they are composed, changing the 
 endings of the acids. Salts of acids that end in ic change this ending to 
 ate, and those that end in ous to ite. Thus, sodium chlorate NaClO 3 , 
 derived from chloric acid, sodium perchlorate NaClO 4 , from perchloric 
 acid, and sodium chlorite, NaClO2, from chlorous acid, are examples. 
 Other systems of nomenclature are in use, but the ones just noted cover 
 the largest field. 
 
 SECTION VIII. 
 
 CHEMICAL CALCULATIONS. 
 
 Kinds of Problems: Chemical calculations constitute a very im- 
 portant branch of study in the training of the metallurgist. They are 
 of three general classes; namely, those involving weight, those involving 
 the volumes of gases, and those involving both weight and volume. Further 
 subdivisions of these classes may be made, the principles of which are best 
 taught from specific examples, as follows: 
 
 PROBLEMS INVOLVING WEIGHT ONLY. 
 
 Calculation of the Molecular Weight from the Formula: Problem! 
 The formula of copper sulphate is CuSO 4 . Find its molecular weight. 
 Solution: From the table of atomic weights and the formula, we find 
 Atomic weight of copper =63.57 Atomic No. of Combining 
 
 " sulphur=32.06 Weights Atoms Weights 
 
 " oxygen =16 63.57 x 1 = 63.57 
 
 32.06 x 1 = 32.06 
 16. x 4 = 64.00 
 
 Molecular weight of CuSO 4 = 159.63 Ans 
 
20 CHEMICAL CALCULATIONS 
 
 Calculation of Percentage Composition of a Compound from its 
 Formula: Problem: Find the percentage composition of a compound, 
 the formula of which is CuSO 4 . 
 
 Solution: 
 
 CU S OA 
 
 Molecular weight=63.57+32.06+(4 x 16)=159.63. 
 
 159.63(63.57 32.06 64.00 
 
 Percentage Composition=39.83% Copper, 20.09% Sulphur, 40.09% Oxygen. Ans. 
 
 Calculation of Formula from the Analysis of a Compound: 
 
 Problem: By analysis a pure compound is found to be composed of calcium 
 29.41%, sulphur 23.53% and oxygen 47.06%. What is its simplest formula? 
 
 Solution : 
 
 Parts in a Atomic Atomic Number of 
 
 Element Hundred Weights Ratios Atoms 
 
 Ca 29.41 -4- 40.07 = .735 K735= 1 Ca 
 
 S 23.53 -f- 32.06 = .735+ -K735= 1 S 
 
 O 47.06 m 16. =2.941 -K 735= 4 O 
 Formula of compound is CaSO 4 . Ans. 
 
 Calculation of Relative Weights from the Chemical Equation: 
 
 Problem: Five per cent, of a certain limestone is non-volatile impurities 
 and 95% is pure calcium carbonate. What will be the weight of lime 
 obtained from calcining 2000 Ibs. of this stone? 
 
 Solution: 
 
 Weight of impurities =5% of 2000= 100 Ibs. 
 " " calcium carbonate =1900 Ibs. 
 
 Reaction on calcining CaCOa =CaO+CO 2 
 
 Combining or atomic wts. . 40.00+12+ (3xl6)=40.+16+12+(2xl6) 
 Relative or molecular wts.. 100.00 = 56.00 + 44 
 
 100 Ibs. CaCO 3 gives 56 Ibs. CaO 
 
 1 " " .56 " " 
 
 1900 " " 1064 " 
 
 1064 Ibs. CaO+100 Ibs. non-volatile impurities=1164 Ibs. of lime. Ans. 
 
 PROBLEMS INVOLVING VOLUME ONLY. 
 
 Calculation of Relative Volumes of Gases: From Avogadro's 
 hypothesis, it is known that molecular weights of all gases give the same 
 volume under standard conditions of 0C and 760 mm. barometric pressure. 
 Problems involving volumes of gases only are, therefore, very simple to 
 solve, because the relative volumes are identical with the coefficients of 
 
CHEMICAL CALCULATIONS 21 
 
 the molecules, as will be evident from an inspection of the following 
 examples: 
 
 H 2 +C1 2 =2HC1 
 1 vol. hydrogen+1 vol. chlorine gives 2 volumes hydrochloric acid gas 
 
 CH 4 +2O2=CO2+2H 2 O 
 
 1 vol. methane +2 vol. oxygen gives 1 vol. carbon dioxide +2 vol. water 
 vapor 
 
 N 2 O 2 +O 2 =2NO 2 
 1 vol. nitric oxide +1 vol. oxygen gives 2 volumes nitrogen peroxide. 
 
 PROBLEMS INVOLVING BOTH WEIGHT AND VOLUME. 
 
 Indirect Method: This method necessitates finding the relative 
 weights of the gases involved, from which the volumes may be calculated 
 from the specific gravity, or the weight of a unit volume. 
 
 Problem: How many cubic feet of carbon dioxide measured under 
 standard conditions would be given off by 2000 Ibs. pure calcium carbonate 
 during the process of calcination? 
 
 Solution: 
 
 Reaction CaCO 3 = CaO + CO 2 
 
 40+12+48 40+16 12+2x16 
 100 56 44 
 
 100 Ibs. CaCO 3 give 44 Ibs. CO 3 
 2000 " " " 880 " CO 2 
 Wt. 1 cu. ft. CO 2 =. 1235 Ibs. 
 
 880H-.1235 = 7125 cu. ft. Ans. 
 
 Direct Method: The fact that molecular weights of gases give 
 constant volumes at standard conditions affords a simple direct method 
 for calculating volumes from the equation. If the weights are expressed 
 in grams, each gram-molecule of the gases involved represents 22.32 liters; 
 if in kilograms, each kilogram-molecule stands for 22.32 cubic meters; 
 and if in avoirdupois ounces, each ounce-molecule gives a volume of 22 32+ 
 cubic feet. By this method the problem above would be solved as follows: 
 
 2000 lbs.=32000 ozs. 
 
 ? cu. ft. 
 
 CaCO 3 = CaO + CO 2 
 40 + 12+(3xl6)_40 + 16 
 100 56 
 
 100 ozs. CaCO 3 gives 22.32 cu. ft. 
 
 1 " " " .2232 
 
 32000 " " * 7142 cu ft. Ans. 
 
22 DESCRIPTION OF ELEMENTS 
 
 SECTION IX. 
 
 A DESCRIPTION OF ELEMENTS COMMONLY MET 
 WITH IN THE MANUFACTURE OF STEEL. 
 
 Oxygen. 
 
 Occurrence: This element is most widely distributed in nature; 
 49.85% of the solid crust of the earth, 88.89% of water and 20.8% of air 
 is oxygen. In air it exists in a free state. In a combined state, it exists 
 in limestone, sand, marble, clay, quartz, iron ore, and many other substances. 
 
 Preparation: It is prepared by merely heating certain of its 
 compounds, some of which are mercuric oxide, potassium chlorate and 
 manganese dioxide; by the decomposition of water by electrolysis; and from 
 the air by purifying processes. 
 
 Properties: Oxygen is a colorless, odorless, tasteless gas, heavier 
 than air (sp. gr. =1.1056) and slightly soluble in water. At a low temper- 
 ature and a high pressure it is converted into a liquid which boils at 181 C. 
 
 The phenomenon of ordinary burning or combustion is due to the 
 combination of oxygen with other substances. It unites with many elements 
 to form a class of compounds, the oxides. It is necessary to life. Animals 
 die in an atmosphere of less than 16% oxygen. 
 
 Compounds: Some important oxides are: carbon dioxide, CO 2 , 
 carbon monoxide, CO, calcium oxide, CaO, magnesium oxide, MgO, ferric 
 oxide, FeaOa, and ferroso-ferric oxide, Fe 3 O 4 . The last two are important 
 as ores of iron. 
 
 Hydrogen. 
 
 Occurrence: Hydrogen does not occur in nature in a free state, but 
 combined with oxygen it forms water, of which it constitutes 11.11%. In 
 a combined state it occurs also in the bodies of plants and animals, hence, 
 in the volatile matter of coal, in petroleum, and in natural gas of which 
 it constitutes almost 25%. Water is always one of the products of 
 combustion when a fuel containing hydrogen is burned. 
 
 Preparation : It can be prepared by decomposing water with sodium, 
 potassium, hot iron, hot coke, or the electric current; by treating 
 certain metals with certain acids; and by treating aluminum with sodium 
 or potassium hydroxide. 
 
 Properties: Hydrogen is a colorless, tasteless, odorless gas, almost 
 insoluble in water. It can be converted into a liquid that boils at 252C. 
 It is the lightest substance known, being about Vis as heavy as air and 
 VIQ as heavy as oxygen. Its specific gravity, air standard, is .0696. It 
 is combustible and explosive. It combines with oxygen in the proportion 
 of 1:8 to form water. Its great tendency to combine with oxygen makes 
 it an intense reducing agent. 
 
DESCRIPTION OF ELEMENTS 23 
 
 Sulphur. 
 
 Occurrence: This element occurs free in the neighborhood of 
 volcanoes and in underground deposits, from which it may be prepared by 
 purifying processes. In combined state it is found as FeS2, FeCuS2, ZnS, 
 and PbS, the last three being valuable ores of copper, zinc, lead, respectively. 
 It also occurs as the sulphates CaSO4, BaSC>4 and PbSC>4, and in animal 
 and vegetable matter. Compounds of sulphur occur in iron ores, in 
 limestone, and in coal; and these are reduced in the blast furnace, when 
 a varying part of the sulphur combines with the iron, in which form it is 
 very undesirable, if present in large amounts, on account of its injurious 
 effects on steel and cast iron. 
 
 Properties: Sulphur is a brittle, yellow crystalline solid which melts 
 at 114.5, forming a straw colored liquid. It is allotropic, i. e., can exist 
 in different physical forms. These forms are prismatic, rhombic and 
 amorphous. When heated to a sufficiently high temperature, it combines 
 with oxygen to form sulphur dioxide, SOa, with iron to form ferrous sulphide, 
 FeS, and with most of the metals, forming sulphides. The sulphur in iron 
 or steel is in the forms of FeS and MnS, distributed almost uniformly 
 throughout the metal while in the molten state. Upon solidifying, however, 
 owing to the difference in density and fusion temperature between these 
 compounds and the metal, they may, under normal conditions, segregate 
 to some extent, causing some parts of the solidified mass to show a higher 
 content of this impurity than the average, or of the whole in the molten 
 state. With hydrogen it forms a gas, hydrogen sulphide, HaS, very 
 important in Chemistry. 
 
 Uses: Sulphur is used in the manufacture of matches and black gun 
 powder, also for disinfecting purposes and for vulcanizing rubber. Its 
 chief use, however, is in the manufacture of sulphuric acid; and the 
 amount of this acid consumed by a nation is a measure of its scientific 
 advancement. 
 
 Compounds: Besides compounds already mentioned, sulphur forms 
 several acids, one of which, sulphuric acid, HaO* SOg (H^SO^,) is a most 
 important compound. It is obtained by oxidizing sulphur dioxide, SO 2 , 
 which is given off as a gas from the roasting of FeS2, ZnS, CuS, and from 
 the burning of sulphur. 
 
 Carbon. 
 
 Occurrence: This element occurs free in nature in crystalline forms 
 as diamonds and graphite and in the amorphous form as coal. It is the 
 chief constituent of the bodies of plants and animals, of all natural fuels, 
 and of nearly all prepared fuels. It occurs in combined state in limestone, 
 magnesite, marble and other carbonate rocks. 
 
 Properties: Carbon is allotropic; diamond and graphite have been 
 mentioned. The common amorphorous forms are coal, lampblack, charcoal, 
 coke, bone black and gas carbon. Its density varies with its form. 
 
24 DESCRIPTION OF ELEMENTS 
 
 Compounds and Uses: Carbon forms many compounds with 
 hydrogen, called hydrocarbons, as methane CH4, ethylene CsEU, benzene 
 CeHe, acetylene CaH's, each of which is but the first member of a series 
 of related compounds. With oxygen it forms carbon dioxide, CO 2 which 
 is a product of combustion and of respiration. CO '2 is also given off when 
 carbonates, such as limestone, are heated. The reaction is, CaCO3=CaO 
 +CO2. Carbon monoxide is formed in combustion when the supply of 
 oxygen is insufficient for the formation of CO 2. Thus, in the blast furnace, 
 a fixed amount of air is blown against an excess of hot carbon, which act 
 results in this reaction: 2C+O2=2CO. Owing to its tendency to combine 
 with oxygen, forming CO 2, CO is a good reducing agent. So, the CO formed 
 before the tuyeres of the blast furnace reacts with the iron oxide thus: 
 
 3 CO+Fe 2 O3=3CO 2 +2 Fe. 
 
 Carbon alone acts as a reducing agent in the metallurgy of iron. 
 3C+Fe' 2 3 =2 Fe+3 CO. 
 
 Iron forms a carbide with carbon, the formula of which is FesC. In 
 pig iron it is also found uncombined in the form of tiny flakes of graphite, 
 hence the term graphitic carbon. Carbon has a marked effect upon iron. 
 The varying properties of steel and the many uses to which it can be applied 
 are due largely to the influence of this element. Carbon in steel, then, up 
 to a certain limit, is not to be considered as an impurity but as an essential 
 factor. 
 
 Silicon. 
 
 Occurrence: Next to oxygen, silicon is the most abundant element 
 in nature. It is the most important constituent of the mineral part of the 
 earth. Sea sand, quartz, jasper, opal and infusorial earths are almost 
 pure forms of SiO2. As silicates, it occurs in clay, mica, talc, hornblend 
 and feldspar. On account of its wide distribution it forms the chief impurity 
 of iron ore, as well as of nearly all natural mineral deposits. 
 
 Compounds: As already indicated silica, SiOa, is one of the chief 
 compounds of silicon. It also forms several acids, chief of which is silicic 
 acid, 2H2O.SiO2 (H^SiO^ which loses water when heated and forms SiOs. 
 H 4 SiO4=SiO'2+2H 2 0. 
 
 Thus, in whatever form silicon may occur in an ore, it is looked upon 
 as SiO2- This substance is the great acid of dry chemistry and at high 
 temperatures will neutralize any base with which it comes in contact. In 
 the blast furnace some of the silica (SiO2) contained in the charge is reduced 
 to silicon. The amount so reduced varies with the working conditions of the 
 the furnace, mainly the temperature. Once reduced, the silicon alloys 
 with the iron and becomes a part of the metallic bath. All but traces of 
 this silicon is re-oxidized and removed in the various processes of making 
 steel. However, a little is beneficial to steel, so it is sometimes added in 
 small amounts in the form of an iron alloy. 
 

 DESCRIPTION OF ELEMENTS 25 
 
 Nitrogen. 
 
 Occurrence and Properties: This element occurs in niter beds as 
 saltpeter, KNO 3 , and Chili saltpeter, NaNO 3 , also in organic compounds 
 and in coal. It is an odorless, tasteless, colorless gas that constitutes 
 about 78% of the atmosphere. 
 
 Compounds: With hydrogen it forms ammonia, NH 3 ; with oxygen a 
 series of oxides, N2O, NO, N2O 8 , N 2 O 5 and NO 2 ; and with hydrogen and 
 oxygen, an important acid, H2O-N 2 O 5 (HNO 3 ). It is a very inert element 
 and has very slight effects in the manufacture of steel. Nevertheless, its 
 presence in the air in so large amounts makes it an important factor in 
 blast furnace practice. 
 
 Phosphorus. 
 
 Occurrence: Phosphorus, always combined with other elements, 
 occurs widely distributed in limited amounts, particularly in soils. It is, 
 therefore, found in all iron ores. It occurs in deposits as phosphorite and 
 apatite, and is an important constituent of bone. 
 
 Properties and Compounds: While phosphorus belongs in the same 
 group of elements as nitrogen, it does not much resemble it from a physical 
 standpoint. It is allotropic and exists in two forms, as a pale yellow solid 
 that melts readily at the low temperature of 44.1C, and as a red form 
 quite different in properties. While it is a much more active element, it 
 closely resembles nitrogen chemically. It forms compounds with hydrogen 
 and oxygen, such as PHs and P2O 5 , and an acid, H 2 O.P 2 O 5 (HPO 3 ), called 
 metaphosphoric acid. It generally is found in nature as salts of 
 orthophosphoric, 3H 2 O.P 2 O 5 (H 3 PO 4 ), and pyrophosphoric, 2H 2 O.P 2 O 5 
 (H 4 P 2 O 7 ), acids. With iron it forms a phosphide, FeaP. It is completely 
 reduced in the blast furnace, hence all the phosphorus occurring in the raw 
 materials is found in the pig iron. In steel it is a very undesirable impurity, 
 but fortunately it is oxidized readily, when it can be neutralized with 
 lime and easily removed as part of a slag. 
 
 Calcium and Magnesium. 
 
 While these two elements belong to different groups, they are very 
 similar so far as the manufacture of iron and steel are concerned. With 
 few exceptions one may be substituted for the other without great 
 inconvenience. Their oxides are the more important bases of dry chemistry. 
 
 Occurrence and Chief Compounds : Both elements occur as insoluble 
 carbonates; limestone, marble, chalk and marl are forms of calcium car- 
 bonate, CaO.CO 2 (CaCO 3 ). Magnesite is magnesium carbonate, MgCO 3 . 
 When heated, both these compounds decompose into the oxides and carbon 
 dioxide, thus: 
 
 CaCO 3 =CaO-hCO 2 . 
 
 MgCO 3 =MgO-hCO 2 . 
 CaO represents quick lime, and MgO, magnesia. 
 
26 DESCRIPTION OF ELEMENTS 
 
 These elements also occur together as a double salt of carbonic acid, 
 calcium magnesium carbonate, CaMg (CO 8 ) 2 , commonly called dolomite, 
 which gives calcium magnesium oxide CaO- MgO when calcined. 
 CaMg (CO 8 ) 2 =CaO-MgO+2 CO 2 . 
 
 Uses of Lime and Magnesia: Lime, CaO, Magnesia, MgO, and the 
 double oxide, CaOMgO, are all very refractory. But on account of its 
 tendency to slake in air, CaO is not used as such. Practically, MgO is 
 the best basic refractory known, and calcined dolomite is the best available 
 substitute. 
 
 The oxides are reduced with difficulty, and on account of their cheapness 
 constitute the principal basic fluxes. As MgO is the leading basic 
 refractory, CaO is the leading basic flux. It combines with both silica and 
 phosphoric acid to form readily fusible slags, which have a lower density 
 than iron and consequently lie upon the surface of the metallic bath. 
 
 Aluminum. 
 
 Occurrence and Properties : This element in combined form is very 
 widely distributed, occurring as one of the constituents of feldspar, granite, 
 mica, cryolite, and all clays. It is reduced from the oxide, A1 2 O 8 , by an 
 electrolytic process, in which state it is applied to many uses. It has a 
 strong affinity for oxygen, violently reducing iron oxide, and on this account 
 it is added to steel as a deoxidizing agent. 
 
 Compounds: In its compounds aluminum displays decidedly basic 
 properties, forming salts with all the common acids except carbonic acid. 
 It forms neither a carbonate nor a sulphide. Aluminum hydroxide, 
 Al (OH) 8 , however, acts like both an acid and a base. When this compound 
 is heated, it loses water and forms alumina, A^Os- It is found in varying 
 amounts in all the raw materials that enter into the metallurgy of iron. 
 In the blast furnace it is never reduced. Its presence, however, has a 
 marked influence on the slag, affecting its fluidity and fusion temperature, 
 important considerations in blast furnace practice. In its purer states 
 alumina is a good refractory, but its scarcity prohibits its extensive use 
 as such. 
 
 Chromium. 
 
 Occurrence: This element is somewhat rare. In small deposits it is 
 found as chromite, Cr 2 O 8 -FeO. This substance is the best neutral refrac- 
 tory known. In its purer states it melts at about 2175 C. 
 
 Properties and Uses: Chromium is both acid and basic in character. 
 It is very important in the manufacture of alloy, or special steels. Its 
 chief effect is one of hardening, hence it is employed to increase the hardness 
 of projectiles, armor plate, automobile steel, and tool steels. 
 
DESCRIPTION OF ELEMENTS 27 
 
 Manganese. 
 
 Occurrence: This element occurs in nature as MnO 2 , its deposits 
 being somewhat limited in the United States. In very small amounts it 
 is widely distributed, and is found in nearly all raw materials of iron manu- 
 facture. About 75% of this manganese is reduced in the blast furnace, so 
 it is a constituent of all pig iron. But it is readily oxidized in the purifying 
 processes, and except for almost traces, that found in steel is added to it 
 in the process of manufacture. Its effect in steel up to 1.0% is good, 
 because it offsets the evil effects of oxygen and sulphur. Higher per- 
 centages, 7% to 15%, are employed to produce the special steel known as 
 manganese steel. 
 
 Iron. 
 
 Occurrence: This most important metal occurs in combined states 
 and, in slight amounts, in nearly all earthy matter, as clays, soils, sands, 
 etc. In deposits, it is found as the sulphide, FeS 2 , as silicates, as a con- 
 stituent of chromite, as the carbonate, FeCOs, and as the oxides, Fe 2 Os 
 and FegOv The compound last named is magnetic. 
 
 Properties: Pure iron, almost unknown, is somewhat unlike the 
 ordinary commercial forms. It is grayish-white in color and relatively soft 
 when compared with steel of high carbon content. It is malleable, ductile, 
 and magnetic. Its specific gravity is 7.78, and its melting point in its purest 
 commercial form is about 1520 C. The presence of certain elements, notably 
 carbon, silicon, phosphorus or sulphur, in the metal lowers the melting point 
 rapidly. 
 
 Compounds: Iron forms two series of compounds, the ferrous and 
 the ferric. The more important ferrous compounds are FeO, Fe (OH) 2 , 
 FeCl 2 , FeSC>4'7H 2 O; corresponding ferric compounds are Fe 2 O 8 , Fe (OH) 8 , 
 FeCl 3 , Fe2 (804)3. Most of these compounds are of the highest 
 commercial importance, and many will receive much fuller treatment as 
 this course advances. 
 
28 REFRACTORIES 
 
 CHAPTER II. 
 
 i 
 
 REFRACTORIES. 
 
 SECTION I. 
 
 NATURE OF REFRACTORIES. 
 
 Importance: The problem of obtaining refractories suitable for each 
 particular operation is one of supreme importance in the metallurgical arts, 
 especially in the manufacture of steel. They form the chief materials of 
 which all furnaces and retaining vessels are made, as well as flues and 
 stacks through which hot gases are conducted. This equipment is 
 expensive, and any failure in the refractories results in a great loss of 
 time, equipment and product, and, too often, in loss of life as well. 
 
 Requirements of Refractories: A refractory may be denned as any 
 substance which is infusible at the highest temperature it may be required 
 to withstand in service. In any particular application however, this definition 
 is incomplete, because the fact that a substance is infusible does not alone 
 determine its value as a refractory. An almost infusible brick, for example 
 may be so fragile as to be worthless, since bricks are generally required to 
 support a load in addition to resisting the effects of great heat. A perfect 
 refractory would meet the following requirements at any temperature: (1) 
 it would not fuse or soften; (2), it would not crumble or crack; (3), its contrac- 
 tion and expansion would be the minimum; (4), it would not conduct heat; 
 (5), it would be impermeable to gases and liquids; (6), it would resist mech- 
 anical abrasion; (7), it would not react chemically with substances in contact 
 with it. Needless to say, an absolutely perfect refractory has never been 
 discovered. However, there are a number of substances which closely 
 approach the first six requirements, at temperatures commonly employed 
 in metallurgical work, and whether or not they will meet the seventh 
 depends upon their chemical composition and the nature of the substances 
 with which they are in contact. 
 
 Classes of Refractories: Refractory substances, in common with matter 
 in general, are of three classes: namely, acid, basic and neutral. Recalling 
 the chemical action of acids and bases toward each other, it is at once 
 apparent that a refractory of an acid character is useless in contact with 
 a basic slag, and vice versa. In selecting a refractory for a specific purpose, 
 the first question to be decided is what class of refractory will be required. 
 Other factors affecting its life and usefulness are the amount of impurities 
 it contains and the uniformity of its composition; and, in the case of brick, 
 

 ACID REFRACTORIES 29 
 
 to these will be added strength, toughness, porosity, or other special 
 qualities. In the manufacture of prepared refractories these factors are all 
 under control, depending upon the selection of materials and the method 
 of manufacture. 
 
 SECTION II. 
 
 ACID REFRACTORIES. 
 
 Chemical Composition: Acid refractories owe their acid character 
 only to silica, SiO^, and are of two kinds, namely, those composed mainly of 
 silica and those composed of aluminum silicate, or clay. In the pure state 
 silica fuses at a very high temperature, about 1830 C, a temperature much 
 above that obtained in ordinary furnaces, but when heated in contact with 
 basic substances it forms silicates, some of which are easily fused. Hence, 
 in refractories composed of silica the presence of impurities, alumina as well 
 as the stronger bases, must be guarded against. As a refractory, silica is used 
 in the natural forms of sand and cut stone and in the prepared form of 
 brick. Sand (90%to 99.5% SiC>2) is used to make up the bottoms of acid 
 open hearth furnaces and of some types of heating furnaces. Ganister, 
 a very superior material for lining converters, is a highly silicious rock. It 
 has a silica content of about 98%. 
 
 Silica Bricks are prepared from quartzite rock found in Pennsylvania, 
 Wisconsin and Alabama. The rock is first crushed fine, then intimately 
 mixed with a binding material which acts as a cement to hold the particles of 
 silica together and to give the brick the necessary strength. For this purpose 
 either clay or lime, usually in the form of milk of lime, is used, the former 
 to produce quartzite brick and the latter, "silica" or ganister brick. The 
 mixture, in a moist condition, is next compressed and moulded into the shape 
 desired for the bricks, which are allowed to dry slowly and then are burned at 
 high temperatures, about 1500 C.,in large kilns. From seven to ten days 
 are required to complete the burning. Silica brick expands slightly when 
 heated. 
 
 Clay is a natural occurring earthy material which has the property of 
 plasticity when wet but becomes hard when burned. The ordinary varieties 
 are more or less impure silicates of aluminum, formed by the decomposition, 
 or weathering, of feldspathic rock, and contain high percentages (10% to 15%) 
 of combined water. They may be residual or sedimentary. Fire clays 
 are of two varieties, known as plastic and flint clays; the latter is very 
 hard, even when ground, but very refractory. The most refractory clays 
 are associated with the coal measures of Pennsylvania. 
 
 The impurities in clays are alkalies, due to undecomposed feldspar; 
 sand; gravel; iron oxide, silicate or sulphide; calcium and magnesium 
 silicates or" carbonates; titania; and organic matter. Of these impurities, 
 
30 BASIC REFRACTORIES 
 
 the basic oxides are the most harmful, as they lower the fusion point 
 decidedly. This is due to the fact that aluminum silicate combines with 
 bases, forming double silicates. 
 
 The process of making fire clay brick is similar to that for silica 
 brick. The clay, in a finely crushed condition, is moistened with a definite 
 amount of water and thoroughly mixed. When flint clay is being used, 
 some plastic clay is used as a binder. Upon being dried, clay begins to 
 shrink and continues to do so during the burning, when the combined water 
 is driven off and the brick becomes hard. Thus, a brick, 9 inches in 
 length after being burned, may measure from 9^ to 9^ inches when 
 moulded, depending on the mixture used. Calcined, or burnt clay is 
 employed in the mixtures to control the shrinkage. Once burned, the brick 
 ceases to shrink and permanently loses the property of plasticity, which 
 latter fact would indicate that the plasticity is due to combined water. 
 The refractory properties of a brick depend upon the nature and amount 
 of impurities and the ratio of silica to alumina. Besides its use as brick, 
 clay is important as a refractory mortar to be used in laying bricks in 
 furnaces and ladles, and as plaster where seamless linings are required. 
 
 SECTION III. 
 
 BASIC REFRACTORIES. 
 
 Magnesia, with a melting point of 2165 C is, for practical purposes, 
 the most satisfactory basic refractory. It is prepared by calcining the 
 mineral magnesite, a natural carbonate of magnesium. Large deposits are 
 somewhat rare. In this country very pure deposits had long been known 
 to exist in the states of California and Washington, but up to the outbreak 
 of the World War the entire supply was obtained from Austria and Hungary. 
 Now, however, the demand is supplied almost wholly from the State of 
 Washington. For this reason it is an expensive material, which fact accounts 
 for its not being used except where a basic substance of the highest 
 refractoriness is required. It makes an ideal brick for the construction of 
 basic furnaces, and is used for the inner courses of bottoms and walls to 
 slightly above the slag line. In a coarsely crushed form, described as pea 
 size, it is very desirable material for making up bottoms in basic furnaces, 
 as, mixed with a small percentage of basic cinder, it is readily fritted, forming 
 a solid mass that resists chemical and mechanical action of the charge and the 
 buoyant force of the bath. 
 
 Lime is even more refractory than magnesia, resisting the intense heat 
 of the oxyhydrogen flame, but on account of its slaking properties it is of 
 little practical value as a refractory. Mixed with magnesia it gives satis- 
 factory results. 
 
NEUTRAL REFRACTORIES 31 
 
 Dolomite, fortunately, furnishes such a mixture and occurs in this 
 country in abundant quantities. Upon calcining the mineral, a mixture of 
 lime and magnesia in the best proportions is obtained. It cannot be fritted 
 on a bottom as well as magnesite, and the lime content fastens upon it a 
 tendency to slake. In the steel industry it is used chiefly for making up 
 the banks of basic open hearth furnaces. 
 
 Bauxite is a natural form of the sesquioxide of aluminum, mixed with 
 varying amounts of earthy matter and the corresponding oxide of iron. 
 It usually contains one per cent, or more of titania, TiO 2 . It is but feebly 
 basic and, when free from silica, is highly refractory. In pure form alumina 
 melts at 2010 C., but the fusion temperature of the natural bauxite will 
 seldom exceed 1820C. Recent trials indicate that it may prove to be an ex- 
 cellent lining material, but its scarcity precludes its general use. 
 
 SECTION IV. 
 
 NEUTRAL REFRACTORIES. 
 
 The Ideal Furnace Lining is a neutral material, a substance that will 
 permit of changing from acid to basic, or basic to acid, processes on the same 
 lining. Two such substances are well known, but unfortunately the con- 
 ditions of natural deposits will not permit of their use except in restricted 
 quantities. These, are graphite and chromite. 
 
 Graphite: This substance is a natural product, though it can be 
 prepared artifically in small quantities. It occurs mixed with calcareous or 
 silicious rocks in Ceylon, Siberia, Austria, England, Brazil and New York. 
 It requires expensive purification. It is infusible even at the temperature of 
 the electric arc, but burns rapidly at that temperature, forming CO or CC>2. 
 At the temperature of the open hearth it would be very slowly consumed. 
 It is used in making special brick, crucibles, etc. Clay may be used as a 
 binding material. 
 
 Chromite most nearly approaches the ideal refractory. Experience 
 proves it to give equally satisfactory results in either an acid or a basic 
 process. Its fusion point, about 2180 C., is far above the highest working 
 temperature of the open hearth or blast furnace. It is difficult to set or 
 sinter. In a finely ground condition and mixed with the proper proportion 
 of slag also finely ground, it is used regularly in the open hearth to daub 
 ports and jambs and patch walls near the slag line. In the form of brick it 
 is used as dividing courses to separate acid from basic bricks, and in the 
 bottoms of soaking pits, because it is impervious to pit cinder. The binding 
 material for chromite brick is lime, or clay and lime. 
 
 Protection for Refractories: The fusion temperature of the materials 
 discussed is amply high to withstand the temperatures of carbon heated 
 
32 
 
 REFRACTORIES 
 
 furnaces, if resistance to heat were the only requirement. But the 
 refractory must possess strength, resistance to abrasion and corrosion, etc., 
 and as these properties decrease rapidly with increase of temperature, it 
 is desirable, in some cases necessary, to protect them as much as possible 
 from the heat. This end is accomplished by backing the brick work with 
 hollow metal forms through which water is kept constantly flowing. These 
 forms are made of cast iron, steel, copper, or bronze, depending upon their 
 use and position in the furnace, and may be in the shape of coiled pipes, 
 hollow boxes, or sprayed jackets. As this course progresses, these devices 
 will be frequently met with and their value demonstrated. 
 
 For purposes of comparison typical analyses of the various refractories 
 will be found in the following table: 
 
 NAME 
 
 Table 3. Chemical Analyses of Refractories. 
 
 PERCENT OF 
 
 
 Silica 
 Si0 2 
 
 Iron Oxides 
 
 Alum- 
 ina 
 
 A1 2 8 
 
 Lime 
 CaO 
 
 Mag- 
 nesia 
 
 MgO 
 
 Soda. 
 Na 2 
 
 Potash 
 K 2 
 
 Water 
 H 2 
 
 Ti- 
 tania 
 
 Ti0 2 
 
 Chro- 
 mic 
 
 Oxide 
 Cr 2 3 
 
 Mang- 
 anese 
 Oxide 
 MnO 
 
 Car- 
 bon 
 
 C 
 
 Fe 2 3 
 
 FeO 
 
 Canister 
 Low Grade 
 Silica Sand . . 
 High Grade 
 Silica Sand.. 
 Silica Brick . . . 
 Low Grade 
 Fire Clay.... 
 High Grade 
 Fire Clay.... 
 Low Grade Fire 
 Clay Brick... 
 High Grade Fire 
 Clay Brick... 
 Calcined 
 Magnesite . . . 
 Calcined 
 Dolomite 
 Bauxite 
 Bauxite Brick . 
 Chromite 
 Artificial 
 (Coke tar) 
 Graphite.... 
 Natural 
 Graphite 
 Brick 
 
 98.20 
 91.60 
 
 99.25 
 96.42 
 
 60.50 
 50.35 
 61.72 
 53.52 
 3.96 
 
 1.66 
 4.10 
 8.82 
 9.36 
 
 5.95 
 13.04 
 
 .30 
 3.48 
 
 .31 
 
 .50 
 
 2.35 
 .75 
 6.43 
 2.00 
 5.81 
 
 .94 
 3.20 
 
 6.30 
 
 .40 
 
 13.50 
 2.15 
 .44 
 
 .90 
 3.68 
 
 .20 
 .75 
 
 24.95 
 33.65 
 28.70 
 41.00 
 1.95 
 
 1.24 
 60.80 
 78.01 
 10.60 
 
 3.04 
 6.12 
 
 .15 
 .10 
 
 Trace 
 2.01 
 
 .25 
 .10 
 .46 
 .30 
 .40 
 
 55.01 
 .04 
 .98 
 Trace 
 
 .43 
 
 .10 
 .05 
 
 Trace 
 .08 
 
 .05 
 .05 
 1.04 
 .30 
 
 87.45 
 
 38.26 
 .04 
 4.41 
 21.06 
 
 .20 
 
 
 
 
 
 
 
 
 Trace 
 
 Trace 
 
 
 
 
 
 
 
 
 
 
 .06 
 
 
 
 
 .15 
 .10 
 .05 
 .90 
 
 .40 
 .05 
 .20 
 
 10.00 
 13.75 
 
 1.40 
 
 .80 
 1.60 
 1.60 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .10 
 
 
 30.08 
 
 1.62 
 1.16 
 
 
 
 
 
 
 
 
 
 
 
 43.97 
 
 .80 
 
 88.20 
 77.80 
 
 
 .43 
 
 1.95 
 
 
 
 
 
TESTING REFRACTORIES 33 
 
 SECTION V. 
 
 TESTING REFRACTORIES. 
 
 Trial Tests and Laboratory Tests: The best test fora refractory is 
 a trial test in which the material is placed in actual service under the most 
 trying conditions it will be expected to stand up under. As such tests can 
 seldom be made on material for new work and as there may be considerable 
 variation in raw materials and in methods of manufacture, laboratory tests 
 are necessary. Such tests are not always conclusive, owing to the difficulty 
 of obtaining laboratory conditions identical to those in actual practice. 
 They are, however, very useful for the purpose of comparisons, and, if the 
 conditions of the tests are sufficiently severe, the more serious defects will 
 be revealed. These tests are chemical and physical. From the chemical 
 analysis the composition of the material is determined and its quality is 
 judged. As the method of manufacture and the care with which it is carried 
 out affect the properties of the refractory, the chemical test should be, 
 and usually is, supplemented by physical tests. Chief among these tests 
 are the fusion or softening point, crushing strength, expansion and con- 
 traction, slagging, porosity, density, resistance to compression, impact, 
 abrasion and spalling tests. Each of these tests may be made in a compara- 
 tively simple manner, but care and judgment are required to see that the 
 conditions of the tests conform closely with those to which the brick are 
 to be subjected in actual service. On this account some of the tests 
 usually employed will not be applicable to the iron and steel industry, 
 while others must be modified to conform to its conditions. The tests 
 here described are those particularly suited to this industry. 1 
 
 The fusion temperature, in ordinary practice, is usually determined 
 by means of Seger cones. These are small triangular pyramids, 6 cm. 
 high, with a base of 2 cm. They are composed of aluminum silicates. Cone 
 number 28 contains ten parts silica to one part alumina and corresponds 
 to a temperature of 1630 C. The fusion temperatures of succeeding cones 
 up to number 40, which corresponds to a temperature of 1920 C, are increased 
 by decreasing the proportion of silica to alumina. For lower temperatures 
 varying amounts of alkali or lime are added. By this means the melting point 
 is so controlled, that a series of cones may be prepared with melting points 
 between the limits of 500 and 1900 . Upon being gradually heated to a 
 sufficiently high temperature, these cones will soften and slowly bend until 
 their tops touch the floor, which point is taken as their fusion point. In 
 making a test, a pyramid of the material to be tested, having the same shape 
 and dimensions as the standards, is placed in a furnace with two or more 
 standards having melting points estimated to be near that of the material 
 to be tested. As the temperature of the furnace is raised, the standard cone 
 that melts at the same time as the test will register the temperature of the f ur- 
 
 JSee Practical Methods for Testing Refractory Fire Brick by C. E. Nesbitt and 
 M. L. Bell. Proceedings of the American Society for Testing Materials. Vol. 
 jL. v II t 1016. 
 
34 TESTING REFRACTORIE^ 
 
 nace and the fusion point of the test. The softening temperature is considered 
 to be that at which the specimen bends, sags or puffs out of shape. Instead 
 of the standard cone, the more accurate pyrometer is coming into use for 
 making this test. 
 
 Resistance to Compression: The ability of brick to withstand 
 pressure at a high temperature is a very important property. This test 
 is made on a modified form of Brinell ball testing machine. The ball is 
 made of steel and is 2K inches in diameter. In making this test, the brick 
 is uniformly and slowly heated from atmosphoric temperature to 1350 and 
 held in the furnace at this temperature for three hours or longer, when it 
 is removed and placed flat under the ball of the machine, and a pressure 
 of 850 Ibs. is immediately applied, which is gradually and uniformly increased 
 at such a rate that a maximum load of 1600 Ibs. is attained at the end of 
 five minutes. The depth of the depression made by the ball is taken as the 
 measurement of the resistance of the brick to compression. 
 
 Expansion and Contraction : A brick must be prepared for this test 
 by grinding the ends so that they will be parallel to each other and at right 
 angles to the sides. Its length is then measured by means of a specially 
 constructed micrometer. The brick is next heated to the temperature at 
 which it is to be used, removed from the furnace and immediately measured. 
 The expansion or contraction is expressed in inches per lineal foot. 
 
 Slagging Test : By this test the impermeability of the brick to molten 
 slag is determined. The brick is prepared by drilling two circular cavities, 
 2^ inches in diameter, each at the intersection of the diagonals of the 
 rectangles formed by bisecting transversely the unbranded face of the 
 brick, to such a depth that the area of the greatest cross section is 1.7 
 inches. The brick is then heated as in the compression test. When the 
 temperature has reached 1350, 35 grams of a standard blast furnace slag is 
 placed in one cavity and 35 grams of a standard heating furnace slag in the 
 other. Both slags are pulverized to pass a 40 mesh sieve. The temperature 
 of 1350 is maintained for two hours after the slag is added, at the end of 
 which time the brick is removed from the furnace and, when cold, sawed 
 lengthwise so as to bisect both cavities, thus exposing the part of the brick 
 subject to slag penetration. The area penetrated by the slag is measured 
 with a planimeter and expressed in square inches. 
 
 Density: The density is determined from the weight of the brick in 
 air and its dimensions. This method gives the apparent specific gravity. 
 This test is greatly influenced by the method of manufacture, being affected 
 by both the amount of water used in pugging and the pressure in moulding. 
 While in manufacturing practice the amount of water added is determined 
 by the plasticity of the clay, investigations have shown that the moisture 
 content should be about 8% and the pressure about 1500 Ibs. per square inch 
 to secure the greatest density. 
 
TESTING REFRACTORIES 35 
 
 The Impact Test: This test is important in the case of brick to be 
 used in blast furnace tops, where they are subject to much impact from 
 lumps of ore, stone, and coke in charging. The test is greatly affected 
 by temperature. A brick heated to 260C. was found to be 20% weaker 
 than one of the same brand tested at 20 C., and 40% weaker when tested 
 at 540 C. In carrying out the test, the brick is first heated from 
 atmospheric temperature to 260 C., the temperature being raised gradually 
 through a period of one hour. The brick is then placed end up in a machine, 
 by means of which a steel ball 2 l /2 inches in diameter and weighing 2.34 
 Ibs. is dropped upon the longest axis of the brick from heights successively 
 increasing by two inches until the brick breaks. The height in inches of the 
 ball in the last test is taken as the measure of the resistance of the brick 
 to impact. 
 
 The Abrasion Test : This test aims to determine the wearing qualities 
 of 'the brick at the temperature to which it is subjected in actual service. 
 The brick is heated to the required temperature, and its end then pressed 
 against a carborundum wheel for a given time and with a fixed pressure. 
 The brick must be ground to uniform thickness and a preliminary cut made, 
 so the wheel will cut through the entire thickness from the beginning of 
 the test. The depth of the cut made on the hot brick measures the abrasion. 
 It is obvious that the test will be affected by the width of the cutting wheel, 
 the speed at which it revolves, also the grade and grit of the carborundum, 
 all of which must be fixed and constant. 
 
 Spalling Test: Spalling in a brick is usually produced by temperature 
 changes, often accelerated by mechanical pressure. The drawbacks with this 
 test are the fact that it must be made much more severe than the conditions 
 of actual service and that it concerns the brick only as made, thus neglecting 
 the effect of slag penetration and the resulting vitrification that often causes 
 brick to spall in service. The test is usually made on several bricks. The 
 bricks are dried at 100 C. for 5 hours or more, weighed, and then placed in 
 the door of a furnace, heated to 1350 C., so that one end only is exposed to the 
 interior heat of the furnace. The remainder of the door space, if any, is then 
 filled with other brick. After heating for one hour at this temperature the 
 bricks are removed, and each one is immediately plunged into two gallons 
 of water at 20 C. to a depth of four inches and held there for three minutes. 
 It is then withdrawn from the water, allowed to dry three minutes, and returned 
 to the furnace as before. This operation is repeated until the bricks have 
 been plunged ten times, when they are dried at 100 C. for five hours or more 
 and again weighed. The percentage of loss in weight, which is taken as a 
 measure of the spalling, is claculated from the original weight. 
 
36 IRON ORES 
 
 CHAPTER III. 
 
 IRON ORES. 
 
 SECTION I. 
 
 ORES AND THE IRON BEARING MINERALS. 
 
 Minerals and Ores: Any homogeneous inorganic substance that 
 occurs naturally in the solid state is called a mineral. A mineral, therefore, 
 may be either an element or a compound. While a few elements, like 
 gold and platinum, occur for the most part native, and others, like silver, 
 copper, mercury, sulphur and carbon, may be found both native and com- 
 bined, most minerals, of which some 800 varieties have been discovered 
 and named, such as quartz, feldspar, hematite, hornblende, calcite, mica, 
 etc., or their species, represent definite chemical compounds. Owing to the 
 many forces that are constantly at work in nature and the wide distribution 
 of some of the minerals, it is seldom a deposit consisting of but a single 
 mineral is encountered. It is of such natural deposits that the ores are 
 constituted. In general, then, an ore is defined as a mineral or a mixture 
 of minerals from which one or more elements may be extracted with profit. 
 
 The Iron Bearing Minerals: While there is a vast number of mineral 
 species that contain iron, there are only a few that are of any importance 
 commercially, because, in most cases, either the iron content is too low 
 to justify the extraction of the metal or the mineral itself does not occur in 
 sufficient abundance to make it available for use as an ore. Grouped 
 according to their chemical composition, the iron bearing minerals of chief 
 importance are divided into four classes; namely, the iron oxides, iron 
 carbonates, iron silicates, and iron sulphides. Of these, only the first 
 class may be considered as a factor in the manufacture of steel in the United 
 States. These oxides go to form a large number of minerals, which have 
 been grouped and named as shown in the following table: 
 
 Table 4. Chief Iron Bearing Minerals. 
 
 Chemical Name Mineralogical Name. 
 
 1. Ferroso-ferric Oxide Magnetite 
 
 2. Anhydrous Ferric Oxide Hematite 
 
 3. Hydrous Ferric Oxides Limonite and others 
 
 4. Ferrous Carbonate Siderite 
 
 5. Iron Silicates Chloropal and others 
 
 6. Iron Sulphides Pyrite and others 
 
MINERALS 37 
 
 Magnetite Group: The only important mineral of this group is 
 magnetite, chemical formula Fe 3 O4, composed of iron, 72.4%, and oxygen, 
 27.6%. The mineral is found in Arkansas, Pennsylvania, New Jersey, and 
 New York. It varies in color from gray to black, has a specific gravity 
 of about 5.0, and is magnetic. This last named property is taken advantage 
 of in locating ore bodies below the surface of the ground and in mechanically 
 purifying ores of this group by magnetic concentration. It is often found 
 closely associated with igneous rocks, when it is apt to contain appreciable 
 amounts of chromium or titanium oxides which cannot be removed from it 
 by magnetic concentration. The remaining magnetic ores of the United 
 States are, for the most part, of a low grade and require dressing, but 
 the magnetite ores of Sweden represent the purest ores in the world and 
 are of a grade approaching that of the pure mineral. 
 
 Hematite Group: The typical mineral of this group is hematite, 
 which contains the equivalent of 70% metallic iron, based on the chemical 
 formula Fe 2 O 3 . It furnishes the base of the world's most important ores. 
 Being associated with rocks of various geological periods, these ores occur 
 widely distributed, and in a variety of forms, which differ greatly in their 
 iron content. Many of these varieties are known, from their outstanding 
 characteristic, as red hematite, specular hematite, oolitic hematite, fossil 
 ore, etc. 
 
 Limonite or Brown Ore Group: The minerals of this group are all 
 hydrous ferric oxides, and may be represented, as a group, by the general 
 formula mFe 2 O 3 - n H 2 O. There are five of these minerals, and they have 
 been named, in the order of their progressive increase in water content, 
 turgite, 2Fe 2 O 3 -H 2 O; goethite, Fe 2 O 3 -H 2 O; limonite, 2 Fe 2 O 3 - 3 H 2 O; 
 xanthosidente, Fe 2 O 3 -2 H 2 O; and limnite, Fe 2 O 3 - 3 H 2 O. On a theoretical 
 basis the iron content of this series will vary from 52.31% to 66.31%. 
 These minerals are widely distributed throughout the United States. In 
 southern Virginia they make up the greater part of the available ores, all 
 of which are low in iron content and high in silica. 
 
 The Carbonate Group: The representative member of this group is 
 the mineral known as siderite, or iron carbonate, FeCO 3 , which contains 
 43.8% of iron. Owing to the fact that carbonic acid is dibasic, a part of 
 the iron required to neutralize it may be replaced by other metals, thus 
 giving rise to a series of minerals, such as iron-calcium carbonate, iron- 
 magnesium carbonate, etc. Some of the names commonly applied to these 
 ores are spathic iron ore, kidney ore, blackband ore, etc. The ore deposits 
 in which this group appears are of little commercial importance in the 
 United States. In England they make up the ores of the Cleveland district. 
 Usually, carbonate ores are calcined before they are charged into the blast 
 furnace. 
 
 The Mineralogical Make-up of Iron Ores: As was indicated at the 
 beginning, an ore deposit at best represents but a mixture of different 
 
38 IRON ORES 
 
 minerals, only a part of which will contain the element or elements sought. 
 All iron ores, then, may be looked upon as being made up of these two 
 parts: One part is composed of the iron bearing minerals, which represent 
 definite compounds of iron; the other part includes all the other substances 
 mixed with these compounds, and is known as the gangue of the ore. 
 Evidently, the richness of the ore, by which term is meant the proportion 
 by weight of iron to all other elements in the ore, depends on the composition 
 of the iron bearing minerals it contains and upon the amount of gangue 
 associated with them. In working up the ores, their physical condition 
 must also be taken into consideration. In this respect, they are subject 
 to the widest variation, ranging from soft clay-like or earthy matter to 
 hard compact masses. Both extremes tend to give trouble in the blast 
 furnace. Thus, the soft fine ores are so apt to choke up a furnace, not 
 designed to use them, that they were once considered practically worthless. 
 The successful smelting of these oreS represents one of the great achieve- 
 ments of American furnacemen. One objection to very fine ores, and one 
 that has not yet been overcome, is that they give rise to large amounts 
 of flue dust, which interferes seriously with the economical utilization of 
 the furnace gases. On the other hand, the very hard and dense ores, which 
 enter the furnace in the form of comparatively large lumps, are difficult 
 to reduce and require an excessive amount of fuel. 
 
 SECTION II. 
 
 VALUATION OF ORES. 
 
 Factors in the Valuation of Ores: Omitting relative property 
 valuations, prices of competitive ores, costs of transportation, and other 
 considerations of a purely business nature, the chief factors that determine 
 the value of an ore are its richness, its chemical composition and its access- 
 ibility. The richness of the ore will, of course, be made the basis for the 
 valuation. For this purpose a unit system is employed, a unit of iron 
 corresponding to one per cent. But the prices of ores do not rise and fall 
 parallel with the number of units of iron they contain, because the gangue 
 to be disposed of must also be considered. For example, suppose two 
 hematite ores containing 63% and 42% iron are being considered. In the 
 first, 90% of the ore is pure mineral, leaving only 10% as gangue to be 
 disposed of, but the second represents only 60% pure mineral with 40% of 
 its weight as gangue to be fluxed and transported. Next to richness comes 
 the consideration of the chemical composition of the ore as a whole, for 
 certain impurities, when present in only relatively small amounts, may 
 make a rich ore worthless. Without taking the time to consider all the 
 possibilities in this connection, the more common impurities in ore may be 
 classed as follows: 
 
IMPURITIES 39 
 
 1. Those impurities that are never reduced in the blast furnace 
 
 and so do not affect the composition of the iron are alumina, AlgOsj lime, 
 CaO; magnesia, MgO; and the alkalies, soda, (Na 2 O), andpotassia, (K 2 O). 
 All these substances, it will be observed, are strong bases, with the exception 
 of alumina which may be either an acid or a base. Therefore, the presence 
 of these substances in the ore may not be objectionable, for the lime and 
 the magnesia, in particular, are valuable as fluxes. Alumina, also, up to 
 about 5%, may play a useful part in regulating the blast furnace. The 
 alkalies for the most part are driven off with the flue dust, and with 
 modern appliances they may be recovered, when present in sufficient 
 amount to justify the installation of the necessary apparatus, so that they 
 may form a valuable by-product. 
 
 2. Those impurities that may be partially reduced in the furnace 
 and give elements that enter the p'ig iron are silica, or the silicates, 
 sulphates and manganese compounds. Of these, the silica, which term 
 includes both the free silica and the silicates, constitutes a large 
 part of the gangue of most ores, and as it requires an equal weight 
 of lime or magnesia to flux it, it must be considered in fixing the value of 
 an ore. Owing to the fact that the amount reduced in the blast furnace 
 is subject to control to a considerable extent and that the element is readily 
 removed during the process of purifying the pig iron, it is not considered 
 of much importance from the standpoint of its effect on the steel produced 
 from the iron. This attitude toward silica is just the opposite of that 
 displayed toward the sulphur compounds. All these compounds are reduced 
 in the furnace to sulphides, in which form the sulphur enters either the 
 metal as ferrous or manganese sulphides or the slag as calcium sulphide. 
 Now, there is a limit to the quantity of sulphur a given slag can absorb, 
 the highest figures given being less than 5%, and, naturally enough, the 
 nearer this limit is approached, the more difficult it becomes to keep the 
 sulphur out of the metal. Since even comparatively small amounts of 
 this element exert an evil influence in steel, and it can be removed 
 from the metal only partially and with much difficulty, the importance 
 of this element in fixing the value of an ore is evident. As to the 
 manganese compounds, the amount of this element that enters the iron 
 varies with the manganese content of the ore and takes place to the 
 extent of nearly 75% of the manganese charged. The per cent, of this 
 element is, therefore, considered in its relation to the iron content. An 
 ore is available for the manufacture of the ordinary grades of pig iron 
 when the manganese content does not exceed 2% of the iron content; 
 between 2% and 10%, calculated on the same basis, it is necessary to 
 mix the ore with others containing little of this element; but if the manga- 
 nese content is 15% to 20% of the iron content, then the ore becomes 
 available for the manufacture of spiegel. 
 
40 IRON ORES 
 
 3. The impurities always reduced in the furnace are all the com- 
 pounds of phosphorus, which element enters the pig iron only. While 
 this element is easily removed from the metal by basic processes, none at 
 all is eliminated by the acid processes, with the result that acid steels 
 contain a higher percentage of this element than the average of the charge 
 from which the steel is produced. This element, therefore, is the basis 
 for the separation of all ores into the two great classes, known as Bessemer 
 and basic. This division, like that for manganese, is made on the basis 
 of the relation of the phosphorus content to iron content of the ore. Since 
 it is desirable to produce Bessemer steel that will contain not more than 
 .100% of its weight as phosphorus, a true Bessemer ore would be one 
 whose phosphorus content plus the phosphorus content of the coke and 
 limestone required to smelt and flux it would produce a pig iron with a 
 phosphorus content not exceeding .090%. Allowing 10% for conversion 
 loss, such a pig iron would give a steel containing less than .100% of its 
 weight of phosphorus. Commercially, however, since commercial toler- 
 ances usually permit the phosphorus in the steel to rise as high as .110%, 
 no allowance is made for conversion, and a commercial Bessemer ore is 
 one whose phosphorus content plus some arbitrary figure, usually about 
 .015%, to allow for the phosphorus acquired from the flux and fuel, is 
 less than one one-thousandth of its iron content. Thus, the per cent, of 
 phosphorus in an ore containing 60 units of iron must be .045, or less, to 
 be classed as a Bessemer ore, for 1/1000 of 60=.060 and .060 .015=. 045. 
 Another method for determining the grade of an ore is explained by the 
 following example: 
 
 Question*. To what class does an ore containing 60% iron and .045% 
 phosphorus belong? 
 
 Solution: 
 
 045-=-.60=.075=per cent, phosphorus in the pig iron, acquired from the ore. 
 .020=Estimated per cent. phos. in pig iron, acquired from coke 
 
 and stone. 
 
 Ans. .095=per cent, phosphorus in the pig iron. Therefore; the ore is 
 of Bessemer grade. 
 
 Water or moisture is another factor to be considered in the valuation 
 of ores, because it adds to the weight of ore to be handled and transported. 
 The importance of this matter in fixing the value of an ore is seen at once 
 when it is pointed out that many of the soft ores of the Lake Superior region 
 carry as much as 12% of their weight as hygroscopic water, and a few as 
 much as, or more than, 15%. This moisture content for any particular ore 
 is much more nearly constant under varying weather conditions than might 
 be expected; but in the Case of different ores there is a wide variation, 
 ranging from .40% in some of the hard red hematites to 16.80% in a few 
 
IMPURITIES 
 
 41 
 
 of the soft red ores. These points are well illustrated by the table below, 
 the examples in which have been selected because they show about the 
 same iron content when dry. 
 
 TABLE 5. Analyses of Ores Illustrating Dry and Wet Basis. 
 
 
 
 
 
 Ai 
 
 c 
 
 a 
 
 8 * 
 
 i 
 
 1 
 
 02 
 
 g 
 
 2 
 
 ORE 
 
 STATE 
 
 | 
 
 8 
 
 & 
 
 1 
 
 f 
 
 |3 
 
 of 
 
 i 
 
 d O 
 
 M M 
 
 rl 
 
 ' a o 
 
 ra O 
 
 3m w 
 
 
 
 $ 
 
 
 ^ JQ 
 
 s 
 
 ^ 
 
 a 
 
 5' ' 
 
 
 feS 
 
 s m 
 
 A. (Marquette Range) . . 
 
 Dry 
 Natural. . 
 
 57.36 
 56.82 
 
 .137 
 .135 
 
 15.62 
 15.47 
 
 .08 
 .08 
 
 1.26 
 1.25 
 
 .68 
 .67 
 
 .33 
 .327 
 
 .007 
 .007 
 
 .03 
 .03 
 
 .944 
 
 B. (Missabe Range) .... 
 
 Dry 
 Natural. . 
 
 57.03 
 52.54 
 
 .042 
 .039 
 
 12.48 
 11.50 
 
 .56 
 .52 
 
 1.69 
 1.56 
 
 .21 
 .19 
 
 .32 
 .29 
 
 .010 
 .009 
 
 2.80 
 2.58 
 
 7.87 
 
 C. (Missabe Range) .... 
 
 Dry 
 Natural. . 
 
 57.06 
 47.47 
 
 .081 
 .067 
 
 7.33 
 6.09 
 
 1.72 
 1.43 
 
 1.00 
 .83 
 
 .30 
 .25 
 
 .40 
 .32 
 
 .010 
 .008 
 
 2.00 
 1.66 
 
 16.80 
 
 The marketing of the ores and all the metallurgical calculations 
 involving them are based on the analyses of samples dried at 100 C. It 
 will be observed that drying at this temperature may not drive off water of 
 crystallization and that in the case of the brown hematites a much higher 
 temperature than the drying temperature is required to drive off all the 
 combined water. 
 
 Accessibility; It is evident that the economic importance of an 
 ore deposit depends to a great extent upon its size and its location, both 
 geologic and geographic. Thus, an ore that is very desirable from the 
 standpoint of chemical composition and physical condition, may be so 
 located as to be practically inaccessible; or granting it can be made access- 
 ible, the amount of ore in the deposit may not justify the expense of opening 
 it up. On the other hand, a poor ore may be so conveniently located that 
 it may be concentrated at a profit. A thorough discussion of this topic 
 cannot be undertaken in the brief space allotted to this chapter. Suffice 
 it to say, that the working of any ore body under modern conditions presents 
 difficult engineering problems both in mining and in transportation. 
 Perhaps the best way to impart some understanding of these problems 
 is through a brief description of the ore mining operations of the Steel 
 Corporation, itself. With the exception of the Tennessee Coal, Iron & 
 Railroad Company, which obtains its ore from the Birmingham District 
 in Alabama, all the constituent companies of the Corporation depend 
 upon the Lake Superior district for their ore supply. 
 
42 IRON ORES 
 
 SECTION III. 
 
 THE BIRMINGHAM DISTRICT. 
 
 Location and General Geology: The Birmingham District includes 
 the area from which the furnaces at Birmingham, Ensley, and Bessemer 
 secure their iron ores, and is co-extensive with Birmingham Valley. This 
 Valley extends from the City of Birmingham in both a northeast and south- 
 west direction for a total length of about 75 miles and a width of about 
 six miles. The ore, which is a variety of red hematite, occurs in the Clinton 
 formation, which consists of shale, sandstone, iron ore, and a little ferru- 
 ginous limestone. Geological researches conducted by the government 1 
 indicate that the ore was formed at the same time as the rocks with which 
 it is associated. The valley lies within the area originally covered by 
 this formation, which, therefore, occurs on both sides and dips away from 
 it on each. But it is only in Red Mountain that the ore bed has been found 
 of sufficient thickness and purity to justify its being worked on a large 
 scale, and nearly all of the most productive mines are located in a section, 
 about 62 miles long, of this mountain between Narrow Gap and Sparks Gap. 
 
 Method of Mining: All of the red ore mines in the Birmingham 
 district were started as open cuts along the outcrop, and the product of 
 these surface mines, having been leached, were at first soft ore. At a 
 few points these simple mining operations are still carried on, but, owing 
 to the dip of the ore beds, all mines from which any large quantity of ore 
 has been taken are now completely underground and are operated by means 
 of slopes or inclines. At these greater depths the ore is very hard and 
 compact. On account of the fact that the southern portion of the ridge 
 is overlaid by more recent formations, the ore gradually becomes more 
 and more deeply buried on passing southward, and all the deepest slopes 
 are in the strip of mountains south-west of Birmingham. The deepest 
 slope at this southern extremity of the district extends downward on beds 
 whose average dip is about 22. The co-existance of the ore with 
 limestone and the proximity of coal beds suitable for. making coke give 
 this district an advantage over other districts of the country. The ore 
 contains phosphorus to the extent of about .8%, which is much higher 
 than other basic ores of the country . f By employing the duplex or the 
 triplex processes in refining the pig iron, a slag with a high phosphorus 
 content is produced that is available as fertilizer for agricultural use. 
 
 SECTION IV. 
 
 THE LAKE SUPERIOR DISTRICT. 
 
 Importance, Location and General Geology: During recent years 
 the Lake Superior district has provided approximately four-fifths of the 
 entire iron ore output of the United States, and there is nothing to indicate 
 but that the region will, for many years to come, continue to be the nation's 
 
 i. Sec. Bulletin U. S. Geol. Survey No. 315, 1907. The Clinton or Red Ores of 
 the Birmingham District, Alabama, by E. P. Burchard, also Bulletin U. S. Geol. 
 Survey No. 340. Investigations relating to Iron and Manganese, by E. F. Burchard, 
 A. C. Spencer, W. C. Phalen. 
 
IRON ORE RANGES 43 
 
 most important source of ore supply. This district, which surrounds Lake 
 Superior, contains certain isolated areas, or ranges, where bodies of iron 
 ore have been discovered. These ranges are scattered over the northern 
 part of the states of Michigan, Wisconsin and Minnesota and also a part 
 of the Canadian province of Ontario. Investigation has proven that these 
 ore bodies occur at certain well defined geological horizons and are 
 associated with certain rocks. Geologically, the Lake Superior deposits 
 are much older than the Clinton ores of the Birmingham district, being 
 associated with rocks of pre-Cambrian age. According to the conclusions 
 of those who have made a study of the area, the iron was originally deposited 
 as an integral part of certain sedimentary rocks. Following their deposi- 
 tion and solidification, these rocks were elevated and folded, after which 
 surface waters, bearing different compounds in solution, percolated through 
 them, and, through chemical action and solution, concentrated the iron 
 in the troughs which had been formed by the folding of the formation or 
 by the intrusive dikes which had cut across the strata. Much later, 
 with the retreat of the ice at the end of the glacial epoch, these ore 
 bodies were left covered by varying depths of glacial drift. In the order 
 in which they were opened the six chief areas, or ranges, lying within the 
 borders of the United States, are Marquette, Menominee, Gogebic, 
 Vermilion, Missabe, and Cuyuna. 
 
 The Marquette Range lies near the southern shore of the lake in the 
 state of Michigan, and a short distance west by south of the lake city of 
 Marquette, from which it takes its name. Besides Marquette, the towns 
 of Ishpeming, Negaunee, Champion, Republic and Gwinn are also included 
 within the area. The formation is narrow as compared with its length 
 and very irregular, but its general direction is from east to west. The 
 original outcrop was very conspicuous and was responsible for the early 
 discovery, in 1844, of the deposit, which, as indicated above, was the first 
 of the great ranges to be worked. It was opened in 1854, and up to 1916 about 
 119,292,000 long tons of ore had been taken out of it. The shipments for 
 that year were 5,396,007 tons. The ore is partly hematite of the red, soft 
 variety, but there are smaller amounts of magnetite and limonit'e and some 
 hard hematite. 
 
 The Menominee Range is also in the state of Michigan. It lies several 
 miles due south of the Marquette range and is, hence, nearer Lake Michigan 
 than Lake Superior. It includes the towns of Iron Mountain, Vulcan, 
 Norway, Florence, Alpha, Crystal Falls and Iron River. The principal 
 belt, composed mainly of hematite, extends in a direction from east to 
 west. Only a part of the range is productive, but up to 1916 more than 
 103,600,000 long tons of ore had been mined from it, and during that year 
 6,365,363 tons were shipped. It was opened in 1872. 
 
 The Qogebic Range lies almost due west of the Marquette range, and, 
 extending as a narrow belt in a direction from a point a little north of east 
 
14 
 
 IRON ORES 
 
IRON ORES 
 
 MAP 
 
 OF THE 
 
 LAKE SUPERIOR REGION 
 
 SHOWING 
 
 IRON RANGES 
 
46 IRON ORE RANGES 
 
 to a point a little south of west of its center, it is located partly in Michigan 
 and partly in Wisconsin. The area includes the towns of Hurley, Ironwood 
 and Bessemer. The original iron formation, which dips sharply toward 
 the north, rests on quartzite, and is cut by igneous dikes that extend at 
 almost right angles to the original quartzite. The dike and the impervious 
 strata thus combine to form troughs, in which ore bodies have been formed 
 by concentration. The ores, which are mostly soft and red, represent 
 partially dehydrated hematites, with subordinate amounts of hard, blue 
 hematite. The range was opened in 1884, and in 1916 there had been 
 produced from it more than 94,812,800 long tons of ore. The yearly ship- 
 ments for 1916 were 8,489,685 tons. 
 
 The Vermilion Range was opened the same year as the Gogebic 
 range. It lies in northeastern Minnesota, and includes the towns of Tower, 
 Soudan, and Ely. The whole district is one of complex folding, so the ore 
 deposits occur in narrow belts, which are enclosed on the bottom and sides 
 by original greenstones of Archean age and on top by the original iron 
 formation. As the pitch or slope is very steep, the outcrops are very small. 
 The ores are all hard and are composed of red and blue hematite. This 
 range had contributed a little more than 39, 526,800 long tons of ore by 1916. 
 The shipments for that year were 1,947,200. This range and the three 
 previously mentioned are known as the old range* to distinguish them 
 from the more recently discovered Missabe and Cuyuna ranges. 
 
 The Missabe Range is one to excite the interest of every one interested 
 in the manufacture of iron or steel, because from it comes the greater part 
 of the ore used for the production of pig iron today. It was opened in 
 1892 and up to 1916 about 406,855,200 long tons of ore had been taken from 
 its mines. The shipments for the year were 42,526,612 tons. It lies in 
 Minnesota, northwest of Lake Superior, and extends in an east and west 
 direction approximately 100 miles. The principal towns are Biwabik, 
 Eveleth, Virginia, Chisholm, Hibbing, Nashwauk, and Coleraine. The 
 iron formation is the Biwabik in the Upper Huronian. It lies along the 
 southern slope of a ridge that is known as the Giants, or Missabe, Range, 
 and has a gentle slope toward the south. The surface is covered with 
 glacial drift, and rock exposures are not common. This surface, originally 
 covered with forest, gave few signs to indicate the presence of ore bodies. 
 The slope of the iron formation is gentle, so most of the ore deposits are 
 flat-lying and have a large horizontal area compared with the deposits on 
 the other ranges. The impervious basement under the ore deposits is 
 formed by layers of slate or paint rock, interbedded with the iron formation. 
 The ores are mostly soft and hydrated hematites and limonite. They vary 
 in texture from very fine dust to fairly coarse, hard and granular ore. 
 Toward the western end of the district, layers of sand are often interbedded 
 with the ore, forming the so-called "sandy" ores, which require concen- 
 tration to form ore of commercial grade. The deposits are all compara- 
 tively shallow. 
 
IRON ORE MINING 47 
 
 The Cuyuna Range, which is the last range of any importance to be 
 discovered, was opened in 1911. It is located in Crow Wing County, Minne- 
 sota, about 100 miles west of Duluth. The principal towns in the district 
 are Deerwood, Crosby, and Brainerd. The range has no marked topo- 
 graphic features, the surface being level and covered with a heavy mantle 
 of sand. Since there are no surface indications to assist in the exploration 
 for ore, the presence of lines of magnetic variation must be depended upon 
 almost entirely. By drilling, these lines have been found to be associated 
 with belts of iron-bearing formations which trend in a northeasterly and 
 southwesterly direction. The formation is interfoliated with slate and 
 schist, and is usually steeply tilted. At some localities igneous intrusive 
 rocks occur. The ore deposits are usually lenticular in form. In certain 
 restricted areas of the range, particularly in the northern part, mangani- 
 ferous iron ores have been found. The deposits of these ores occur in 
 irregular pockets or lenses, and contain as high 45% manganese. Some 
 of these bodies of ore are being worked for their manganese content only. 
 In 1916 the yearly production had reached 1,716,218 tons, and the total 
 production, 4,897, 298 tons. 
 
 SECTION V. 
 MINING THE LAKE ORES. 1 
 
 Prospecting and Exploration: Since the Lake Superior ores occur in 
 pockets or distinct bodies and vary much as to character and location, 
 the actual mining of the ores is preceded by much work of an exploratory 
 character. This work includes prospecting and exploration. 
 
 Prospecting is the term generally applied to the quest for surface 
 indications of ore, or the conditions which would warrant the expectation 
 of finding ore in the vicinity. It includes such quest operations as geological 
 examination, dip needle work, shallow test-pitting, and trenching. The 
 ore bodies of the Missabe Range are non-magnetic, and dip needle prospect- 
 ing is therefore valueless. On the Cuyuna Range, however, magnetic 
 attraction as evidenced b'y the dip needle has been extensively employed 
 as a guide to the location of ore deposits; in other localities it has also 
 found limited application. 
 
 Drill Exploration: After the presence of an ore deposit is known or 
 suspected, resort is generally had to exploration by means of diamond of 
 churn drills. On the old ranges geological conditions generally make this 
 manner of ascertaining the exact limits of an ore deposit impracticable; 
 so, if two or three adjacent drill holes develop considerable depths of ore, 
 the sinking of a shaft for underground exploration, development and mining 
 is generally considered warranted. On the Missabe Range, however, the 
 flat-lying and comparatively shallow characteristics of the ore formation 
 warrant much more extensive drill explorations. On this range, then, an 
 ore body is almost invariably followed out with the drills, and its limits 
 
 i For further details concerning the mining of the Lake Ores, see Minn. School of 
 Mines Experiment Station Bulletin No. 1, Iron Mining in Minnesota, by Charles 
 E. Von Barneveld, University of Minnesota, Minneapolis, Minn. 
 
48 
 
 IRON ORES 
 
 
 PIG. 4. Open Pit Mining 
 
IRON ORES 
 
 
 FIG. 4. Open Pit Mining 
 
50 IRON ORE MINING 
 
 are determined to the point where the complete plan of development can 
 be worked out in advance of actual mining operations. 
 
 Methods of Mining: Both open pit and underground methods of 
 mining are employed in the mines of the Lake Superior District. On the 
 old ranges, where the ore bodies often extend to great depths and usually 
 lie at angles so steeply inclined to the horizontal that the surface exposures, 
 or outcrops, are small, underground mining methods are employed almost 
 without exception. On the Missabe Range the ore bodies are, as a rule, 
 flatlying with relatively large areas of outcrop, and open pit mining is, 
 therefore, general. Of course, there are many deposits on this range that, 
 on account of limited operating area, excessive depth of over-burden, or for 
 other reasons, must be mined by underground methods, and there are, 
 therefore, a large number of underground mines also. But by far the 
 greater part of the tonnage produced from the Missabe Range comes from 
 open pits. 
 
 Open Pit Mining: Before deciding whether an ore body should be 
 mined by underground methods or as an open pit, a detailed operating- 
 analysis is made of the proposition to determine by which method the ore 
 can be mined most economically. Estimates are made determining the 
 yardage of overburden, or "stripping", that must be removed to uncover the 
 ore body, the tonnage of ore which can then be mined by steamshovel, and 
 the additional tonnage which can be "scrammed" or "milled" in the pit 
 after the limits of steamshovel operation have been reached. Then the 
 cost of the entire operation, including interest-charges on the necessarily 
 large investment in stripping removal, is calculated and reduced to a final 
 cost per ton of ore recoverable. If this figure is less than the probable cost 
 per ton of underground mining, and if the other operating conditions are 
 satisfactory, open pit operation is, of course, deemed advisable. The 
 laying out of an open pit mine involves the following engineering problems: 
 first, outlining the area of ore which it will pay to strip,!, e., considering the 
 two factors of depth of ore and thickness of overburden; second, planning the 
 disposal of stripping which it will be necessary to remove to uncover the ore 
 body, for this material must often be hauled considerable distances from the 
 pits to dump grounds; third, locating the track systems outside the pit 
 for the transportation of stripping and the hauling of ore; fourth, designing 
 the system of railroad tracks within the pit that will make available the 
 maximum quantity of ore accessible by steamshovel, which designing 
 generally involves a series of switchbacks on limiting operative grades and 
 curvature; fifth, providing for drainings of the pit; and sixth, planning 
 in advance for the mining of the ore that cannot be mined by steam- 
 shovel. The general term open pit mining covers three recognized 
 methods of mining, i. e., steamshovel, milling and scramming. Steam- 
 shovel mining, of course, needs no description; it is simply the loading 
 of ore directly into railroad cars by steamshovel. 
 
IRON ORE MINING 
 
 51 
 
IRON ORE MINING 
 
 Milling is a term applied to a thoroughly well worked out system of 
 open pit mining, extensively prosecuted in the early days and still applied 
 under suitable conditions. It consists of the following operations: first, 
 the removal of the overburden from the ore body to be mined, this being 
 done by steamshovel; second, the sinking of a hoisting shaft or incline to 
 the bottom of the ore and the development of a system of underground 
 
IRON ORE MINING 53 
 
 tramming drifts tributary to the shaft and underneath the ore to be 
 mined; third, the putting up of a number of raises (vertical openings) 
 extending from the underground drifts through the ore; fourth, "milling" 
 or shoveling the ore into the raises, through which it is drawn into tram 
 cars operating in haulage drifts that lead to the shaft or incline, where it is 
 hoisted to the surface. The milling system of mining can well be applied to 
 small ore bodies which can be successfully stripped, but where the resultant 
 open pit areas are too small to permit of steamshovel operation; also, as a 
 sequel to steamshovel mining in larger pits where considerable depths of 
 ore remain after the limits of steamshovel work have been reached. 
 
 Scramming is a term applied colloquially on the Missabe Range to 
 the operation of recovering shallow pockets and hummocks of ore left 
 unmined in and around the open pits following the period of steamshovel 
 mining. It is a general term inclusive of hand work, scraper work, mining 
 with dragline excavators, etc., and is applicable generally to the operation 
 of "cleaning up" a pit after its period of real production has passed. 
 
 Advantages of Open Pit Mining: It is very apparent that open pit 
 mining, when feasible, offers decided advantages as compared with under- 
 ground methods. Probably the most evident of these is the possibility of 
 big production; in 1916 the Hull Rust Mine alone shipped 7,665,611 tons of 
 ore, more than 10% of the total mined in the United States during that 
 year, which, according to the U. S. Geological Survey, amounted to 
 75,167,672 tons. Where the overburden is light in comparison with the 
 depth of ore, and stripping charges are not heavy, open pit mining produces 
 low cost ore. It accomplishes a great saving in labor; the output per man 
 per day from the open pit mine is many times that from the average 
 underground mine. Aside from the skilled operators of the steamshovels 
 and locomotives, common labor only is required in open pit mining, while 
 in underground work the miner is a rather high class workman, and he 
 receives a relatively high wage. Owing to this latter condition, strikes 
 have never bsen able to interfere seriously, so far, with the output of 
 Missabe Range open pit mines. 
 
 Underground Mining Slicing: The system of underground mining 
 most generally in use in the mines of the Missabe Range, and in the soft-ore 
 mines of the old ranges, as well, is known as top=slicing and caving. The 
 development of a mine under this method is as follows : First, a shaft is sunk 
 to the bottom of the ore body, or to such depth in the ore as has been 
 determined as desirable. Second, after cutting a "station," pumproom 
 and pocket at the bottom of the shaft, a main haulage drift, or system 
 of haulage drifts, is driven out underneath the ore body. Third, raises are 
 put up from the haulage drifts at intervals of about 50 feet along the drifts 
 through the ore body to the top of the ore. Fourth, cross cuts are driven 
 from the tops of the raises to the limits of the ore body or the property 
 lines, the cross cuts being parallel and the same distance apart as the 
 
54 
 
 IRON ORE MINING 
 
 Fifth, beginning at the ends of the cross cuts farthest from the 
 raises, the ore is "sliced" out between cross cuts, trammed to the raises, 
 dumped into the latter, drawn off thru chutes into cars operating on the 
 main haulage level, hauled to the shaft, dumped into the shaft pocket and 
 hoisted to the surface, where it is either loaded direct into railroad ore 
 
 FIG. 7. Underground Ore Mining Square Set Timber. 
 
GRADING 55 
 
 cars, or (if in the Winter time) stockpiled for later shipment. A "slice" 
 consists of a room opened up between crosscuts, and may be one, two or 
 more sets wide depending on the tendency of the overburden to crush the 
 temporary timber supports. When the ore has been removed from the room 
 or slice, the supporting timbers are blasted out and the overburden allowed 
 to cave and fill it. Before blasting the timbers, however, boards are laid 
 over the floor of the room to prevent admixture of the caved material 
 with the ore below. While slicing and caving operations are proceeding 
 on the top level, the cross cuts to develop the next level immediately below 
 are being driven, and as soon as considerable areas of cave have been 
 developed on the first level, slicing under these areas is started on the 
 second level. Thus, the entire ore body is mined, slice by slice, and level 
 by level. Levels are generally about eleven feet apart, floor to floor. 
 Haulage of ore on main levels from chutes to shaft may be by hand, mule or 
 electric motor, depending on the size of the mine. On the sub-levels the 
 ore is hand-trammed in small dump cars, or for short hauls, in wheelbarrows, 
 from the slices to the raises. 
 
 Advantages of the Slicing System of Mining: The top-slicing-and- 
 caving system has many advantages. It gives a high percentage of ore 
 extraction. If desired, the ore from different working places can be 
 separated, and two or more grades can be produced from the same mine. 
 Development and mining operations are simple and safe, and can be carried 
 out along well defined plans worked out in advance. While the consumption 
 of mining timber is high, cheap inferior grades are used, and under ordinary 
 conditions this item of cost is not excessive. In common with most other 
 systems of mining, it possesses the disadvantage of a limited number of 
 working places; considerable handling of the ore is also necessary. 
 
 The depth of mine shafts on the Missabe Range rarely exceeds 350 
 feet. The average is probably between 250 and 300 feet. On the old ranges, 
 where the rock formations have been folded and tilted, mining operations 
 extend much deeper. Here mine shafts 500 to 1500 feet deep are common, 
 while in some cases mining operations are still in ore at depths well in 
 excess of 2,000 feet. 
 
 Grading the Ores: In the early days of iron ore mining, no grading 
 of the ore from analysis, such as prevails today, was made, and the ore 
 was known by the name of the mine that produced it. Then, the number 
 of mines was small, and the ore from any one mine was fairly uniform. 
 As the production increased, however, and the field of available ore was 
 broadened to include deposits previously regarded as unprofitable, it 
 became necessary, in order to simplify shipping, to grade ores according 
 to their composition, and, further, to mix ores differing in composition to 
 produce certain grades. Finally, it became quite common for one mine 
 to ship several different grades, and for the ore from several mines to 
 be grouped under one name. These conditions brought about a necessity 
 
56 IRON ORES 
 
 for knowing the exact composition of the various ores, and whether or not, 
 in the case of mixed ore, each cargo was of the grade guaranteed. This 
 grading is done by sampling the ore in the cars as fast as they are loaded 
 at the mine, in lots not exceeding ten, and making a rapid but accurate 
 analysis for the determining elements. From this analysis the class or 
 grade of the ore is fixed, and its allotment into a certain group can be made. 
 This is the work of the grader, who, from the analysis of the cars as sub- 
 mitted to him, makes a theoretical shipment in which the contents in 
 silica, iron, phosphorus and possibly manganese, the determining factors 
 in the value of an ore for its particular purpose, must fall within certain 
 predetermined limits. 
 
 Transporting the Ores: The Lake ores now supply all the furnaces 
 in Western New York, Western Pennsylvania, Ohio, Illinois, and Indiana, 
 as well as those in the ore producing states of Michigan, Wisconsin, and 
 Minnesota. In order to reach these markets, the ores must be transported 
 for distances varying from 300 to more than 1000 miles, depending upon 
 the locations of both the mine and the furnaces. The cost of transporting 
 this ore by rail alone would be a serious handicap to some furnaces, but 
 fortunately the chain of Great Lakes affords a cheap mode of transportation 
 for the greater part of the long as well as the short distances. Nearly all 
 the ore mined in the ranges, then, goes first by rail to a harbor on Lake 
 Michigan or Lake Superior, where it is loaded on ore carrying boats that 
 carry it either down Lake Michigan to Chicago or Gary or through Lake 
 Huron and Lake Erie to ports further south. For most of the ore, even 
 these lower lake ports are not ultimate destinations, and another haul by 
 rail is required to place it at the furnaces. Now, to return to the grading 
 of the ore, what was there referred to as a theoretical shipment might 
 better have been called a theoretical cargo or boat load. When the cars 
 containing this theoretical cargo, which may weigh from 3000 to 13,600 
 tons, depending upon the size of the boat, reach the dock at the shipping 
 port, they are unloaded into the dock pockets, one on top of the other, 
 three to six cars to a pocket, in such order as to mix the ore as much as 
 possible. The ore is then allowed to flow from these pockets into the 
 hatches of the steamer, thus, again mixing the ore. Then the boat 
 proceeds to her destination, where the ore is unloaded by electrical, 
 or otherwise operated, grabs, which process of unloading still further 
 tends to mix the ore and make it uniform. All this mixing of the ore 
 is not to be thought of as merely incidental to the operations, but 
 as a necessary course of procedure, for uniformity in the ore is a very 
 important requirement in the successful operation of the blast furnaces. If 
 the ore is unloaded at the works located on the lakes, it is sampled for 
 analysis during the unloading by an elaborate system -and is dumped 
 upon its appropriate stock pile; if for inland works, such as Pittsburgh, 
 it is placed in cars and ultimately reaches the works, where each car is 
 sampled, according to printed instructions common to all the works of 
 
IRON ORES 57 
 
 the Steel Corporation. The cars are then unloaded upon a stock pile from 
 which the ore can be used as needed, or directly into the furnace bins, 
 if the ore is needed for immediate use. 
 
 Mining and Grading in Winter: In winter the procedure as outlined 
 above has to be changed somewhat. During a part of November, and all 
 of the winter months of December, January, February and March, the ore 
 cannot be transported over the lakes because of the ice. On this account, 
 operations in the open pit mines of the Missabe District are suspended in 
 winter; but in all the underground works, both of the old ranges and the 
 Missabe, mining is continuous throughout the year, and the ore mined 
 during the non-shipping season must be stock piled. As this ore is removed 
 it is carefully sampled, and average samples are analyzed daily. These 
 analyses, supplemented by those made in the work of exploration that is 
 constantly carried on in advance of the mining, make it possible to calculate 
 the average composition of each stock pile at the beginning of the shipping 
 season in the spring. This stock, therefore, may be combined, if necessary, 
 with the ore direct from the mines to make up cargoes of definite and known 
 composition. 
 
58 FUELS 
 
 CHAPTER IV. 
 
 FUELS. 
 
 SECTION I. 
 
 SOME PRE- REQUISITES TO THE STUDY OF FUELS. 
 
 'Introductory: There are five basic materials upon which the 
 metallurgical arts depend; namely, ore, fuel, flux, air and water. Of 
 these one is as important as the other, for all metallurgical industry would 
 cease with the failure of any one. At one time all these were thought to 
 be inexhaustible, but recently it has been generally recognized that the 
 supply of the higher grades of ore are limited, and that the more suitable 
 fuels, at the present rate of consumption, must be exhausted in a compara- 
 tively short time. Representing the only source of energy under man's 
 absolute control, fuels are the foundation upon which a nation's progress 
 and prosperity depends. The subject is also a very large one. Hence, in 
 this chapter it is desirable to discuss briefly a few fundamental topics of 
 general interest, and more in detail, a few matters especially important in 
 the iron and steel industry. 
 
 Sensible and Specific Heat: Provided no change of state is involved, 
 the effect produced by imparting heat to a body is a rise in temperature 
 of the body, and if the body is made to give up heat, its temperature falls. 
 This heat, which is easily detected, is often spoken of as sensible heat. 
 The quantity of heat required to raise the temperature of a body 1C. is 
 called its thermal capacity. The amount of heat required to raise the 
 temperature of equal masses of different substances 1C. varies greatly, 
 and the thermal capacity of one gram of any substance, in other words, 
 the number of calories required to raise the temperature of one gramlC., 
 is called its specific heat. Specific heats are determined by the method 
 of mixtures, which is based on the law of heat exchange. This law states 
 that when bodies at different temperatures are brought into contact, 
 exchange of heat takes place until a uniform temperature for all is reached, 
 and that the heat lost by the hotter bodies equals in quantity that gained 
 by the colder ones. The specific heats of a few common metals follow: 
 Iron=.109 cal. Copper=.092 cal. Zinc=.093 cal. Mercury=.033 cal. 
 
HEAT LAWS 59 
 
 Latent Heat and Change of State: Changes of state are governed 
 by the following laws: 1 
 Laws of Fusion: 
 
 I. "Every crystalline substance begins to melt at a definite temper- 
 ture, which is invariable for each substance if the pressure is 
 constant." 
 II. "The temperature of a body, when slowly melting, remains constant 
 
 till the whole mass is melted." 
 III. "Substances that expand on solidifying have their melting points 
 
 lowered by pressure, and vice versa." 
 Laws of Evaporation: 
 
 I. "The rate of evaporation increases with rise of temperature." 
 II. "The rate of evaporation increases with the surface of the liquid 
 exposed." . 
 
 III. "The rate of evaporation is increased by a continual change of 
 
 air in contact with the liquid." 
 
 IV. ''The rate of evaporation is increased by diminishing the vapor pres- 
 
 sure, that is, by applying suction to the vessel inclosing the liquid" 
 Laws of Ebullition: 
 I. "A pure liquid has a definite boiling point, which is invariable 
 
 for that liquid under the same conditions." 
 
 II. The temperature of the vapor given off by a pure liquid near its 
 surface remains constant till all the liquid is vaporized if the 
 pressure remains constant. 
 
 III. "The boiling point of a liquid is raised by salts and lowered by 
 
 gases dissolved in it." 
 
 IV. "The boiling point of a liquid rises with increase of pressure and 
 
 falls with decrease of pressure." 
 
 It will be noted from these laws that when change of state takes place, 
 there is no change in temperature of the body, notwithstanding the constant 
 application or withdrawal of heat. The heat thus involved is sometimes 
 spoken of as latent heat, though heat of fusion and heat of vaporization 
 would appear to be the better terms to employ. The heat of fusion for water 
 is about 80 calories per gram, while its heat of vaporization is 535.9 calories 
 per gram under standard pressure. 
 
 Transmission of Heat : Heat is transmitted in three ways; namely, by 
 conduction, by convection and by radiation. Conduction is the transmission 
 of heat through a body without visible motion of the body, as through an 
 iron bar. When the heat is transmitted by mechanical motion of the 
 particles of matter, through air or water currents for example, it is called 
 convection. The distribution of heat through a blast furnace makes use 
 of these principles. Radiation refers to the transmission of heat, independ- 
 ently of matter, by means of waves in the ether. This is the means by 
 which the heat of the sun reaches the earth. These factors are important con- 
 siderations in the action of furnaces, stacks, ventilators and heating plants. 
 iSeeHigh School Physics and University Physics by Henry S. Carhart and Horatio 
 N. Chute, published by Allyn and Bacon, Boston and Chicago. 
 
60 FUELS 
 
 .Fuels and Combustion: Any chemical reaction by which light and 
 heat are evolved is called combustion. In the ordinary cases of combustion, 
 one of the reacting substances is the oxygen of the air. Therefore, fuels 
 are sometimes defined as substances which will burn in air and liberate 
 heat with sufficient rapidity to be applied to practical purposes. The chief 
 elements constituting fuels are carbon and hydrogen, though in certain 
 processes silicon, phosphorus, sulphur, manganese and the metals may 
 serve as fuel. In metallurgy the fuel is often required to act as a reducing 
 agent, in which cases the total heat produced will be derived from two 
 sources, namely, by combustion with oxygen of the air and by combination 
 with oxygen of the ore. 
 
 Fuels and Chemical Energy: Fuels represent potential energy, 
 which is given off as heat by chemical action. Therefore, the relations 
 of the weights of fuel, weights of air, and the amounts of heat evolved are 
 fixed quantities. The following reaction furnishes a simple illustration: 
 
 C+O 2 =CO 2 +Heat, 
 
 that is,12 gm.C+32 gm.O 2 =44gm.CO 2 +97200 cal. (Heat of formation of CO 2 ) 
 orl gm.C+2.666 gm.O 2 (11.51 gm. air)=3.666 gm.CO 2 +8100 cal. (Calorific 
 power of carbon). 
 
 The heat liberate is referred to in two ways. The chemist bases his 
 calculations on the total heat evolved to form a molecular weight in grams 
 of a given substance, in this case, 44 gms. CO 2 , which is called the heat 
 of formation for that compound. But the metallurgist and engineer 
 employ the heat evolved from a unit weight of fuel, in this case, one gram 
 of carbon, and refer to it as the calorific power of tfce fuel. It should be 
 observed that these terms take into consideration only the total amount 
 of heat evolved, irrespective of the time or speed of the reaction, the 
 duration of which may vary through wide limits. This point is important 
 in the attainment of high temperatures and is connected with the second 
 important factor affecting the combustion of fuels, namely, the temperature 
 to which the products of combustion may be raised by the heat evolved. 
 This is referred to as the calorific intensity of the fuels. Both the calorific 
 power and the intensity enter into the valuation of fuels. 
 
 Measurement of Calorific Power: As has already been noted, the two 
 practical heat units are the large calorie and the B. t. u. Metallurgists 
 express the calorific power in large calories per kilogram, which is numer- 
 ically the same as small calories per gram employed by chemists, while 
 engineers use B. t. u. per pound of fuel as the basis of their calculations. 
 The units are readily converted from one to the other; the relation with 
 respect to the quantity of heat they contain is expressed thus: 
 
 B! t. u. per lb.: Cal. per kilo=l : 1.8 
 
 Hence, to reduce Cal. per kilo, to B. t. u. per pound, it is only necessary 
 to multiply by 1.8. The factor for changing B. t. u. per lb. to Cal. per 
 kilo is .5555. 
 
CALORIFIC POWER 61 
 
 The Calorific Power of some common elements in simple oxidation 
 reactions is as follows: 1 
 
 Table 6. Calorific Power of Some Elements. 
 
 Element Reaction. Calorific Power in Calories per Kilo. 
 
 H 2H 2 +O 2 =2H 2 O Liquid 34500 
 
 H 2H 2 +O 2 =2H 2 O Vapor 29030 
 
 C C+O 2 =CO 2 8100 
 
 C 2C+O 2 =2CO 2430 
 
 Si Si+ O 2 =SiO 2 7000 
 
 Al 4A1+3O 2 =2A1 2 O 3 7270 
 
 P 4P+5O 2 =2P 2 O 5 5895 
 
 S S+ O 2 =SO 2 2196 
 
 Fe 3Fe+2O 2 =Fe 3 O 4 1612 
 
 N N 2 + O 2 =2NO 1541 
 
 Calculating Calorific Power: Given the heats of formation of the 
 reacting substances and of the products of combustion, it is possible to 
 calculate the calorific power of some fuels from their analyses. The calorific 
 power of a well made coke can be estimated approximately from the fixed 
 carbon, if the moisture and volatile matter are low. 
 
 By analysis, Fixed Carbon=87.0% 
 
 Calorific Power of Carbon=8100 Cal. per Kilo 
 
 .87 
 
 7047 Cal. per kilo.=calorific power of coke 
 
 This calculation for gases becomes more complicated, because the 
 calorific power is usually expressed in Calories per cubic meter or B. t. u. 
 per cu. ft. This fact requires a conversion from weight to volume, since 
 the calorific powers in the table are based on weight. This conversion, 
 however, is a simple matter, since a gram molecular weight of any gas has 
 a volume of 22.32 liters under standard conditions. To find the heat evolved 
 from a gas in calories per liter, it is only necessary to c divide the heat of 
 formation by 22.32. In the case of a blast furnace gas composed of CH4, 
 .2%; CO, 25%; H 2 , 3%; CO 2 , 12%; N 2 , 59.8%; the calorific power may be 
 found as follows: The reactions that may occur in complete combustion 
 are 
 
 Reaction (1) CH 4 +2O 2 =2H 2 O+CO 2 
 
 Heats of formation, 21700 cal. +0=2x58060 cal. +97200 cal. 
 
 Reaction (2) 2CO+O 2 =2 CO 2 
 
 Heats of formation, 2x29160 cal. +0=2x97200 cal. 
 
 Reaction (3) 2H 2 +O 2 =2H 2 O 
 Heats of formation, 2x58060 cal. 
 
 iSee Metallurgical Calculations by Joseph W. Richards, One Volume Edition 
 published by McGraw-Hill Book Company, New York. 
 
62 FUELS 
 
 From reaction (1) the total heat available from CH 4 = 
 
 per litel , 
 
 22.32 
 
 From (2) available heat from CO= 
 
 972^0 29160 ==3048+calories per liter> 
 22.32 
 
 From (3) available heat from H2= 
 
 H5?15_=2605.+calories per liter. 
 2x22.32 
 
 Total heat of gas available is, 
 for CH 4 .2% of 8585= 17.17 
 CO 25% of 3048=762.00 
 H 2 3% of 2605= 78.16 
 
 857.33 calories per liter at 0C. and760m.m. 
 pressure= (857.33 x.l 1236) =96.33 B. t. u. per cu. ft. 
 
 Practical Tests : All calculated results, however, are usually higher 
 than can be obtained in actual practice. Furthermore, with complex fuels, 
 like coals, the composition of which can only be guessed at, it is impossible 
 to make accurate calculations from the analysis, because no account is 
 taken of the heat required to decompose the fuel and gasify the products. 
 Many fuel chemists, however, have evolved formulas by which they are 
 able to determine very closely from the analysis the calorific value of a 
 coal as obtained by laboratory experiment. Nevertheless, experimental 
 methods are relied upon almost wholly to determine the heating power 
 of fuels. These tests may be practical, in which large quantities of the 
 fuel are consumed under conditions approximating closely those of the 
 process for which the fuel is to be used; or they may be laboratory tests, 
 in which small . quantities of fuel are burned, and the heat evolved is 
 measured. Practical tests may be made in specially constructed apparatus 
 or under boilers in actual service. These specially constructed apparatus 
 are in the form of large heaters through which water circulates and in which 
 the fuel may be completely consumed. From the amount of fuel consumed, 
 the weight of water heated, the rise in temperature of the water, the 
 difference in temperature between the in-going air and the products of 
 combustion, the calorific power may be accurately determined. 
 
 Laboratory Tests : For determining the maximum amount of heat a 
 given fuel is capable of generating, laboratory tests are more exact than 
 practical tests. Such tests are carried out in specially designed apparatus 
 called calorimeters. There are several makes of these instruments, but 
 
CONSERVATION OF HEAT. 63 
 
 the fundamental principles of all are the same. The process consists in 
 completely oxidizing the fuel in a space enclosed by a metal jacket (the 
 bomb) so submerged that the heat evolved is absorbed by a weighed 
 portion of water contained in a perfectly insulated vessel. From the rise 
 in temperature of the water, the heat liberated by one gram of the fuel is 
 calculated. The best types of calorimeters are those called oxygen bomb 
 calorimeters, in which the fuel is burned in the presence of compressed 
 oxygen. 
 
 Calorific Intensity: The calorific intensity is more precisely defined 
 as the degree of heat evolved by a given weight of fuel in perfect combustion 
 in air at 0C. and 760 m. m. pressure. The theoretical maximum temper- 
 ature depends on the calorific power, and the density and specific heats of the 
 products of combustion, and is inversely proportional to the time required 
 in producing it. In practice the temperature is also affected by the initial 
 temperature of the fuel and air, the amount of air used, amount of water 
 in fuel and air, and by radiation, conduction and convection of materials 
 of the furnace. As the attainment of high temperatures is necessary to 
 many metallurgical processes, all these factors are important, and special 
 devices have been invented to prevent waste, preheat air' and gas, and 
 eliminate moisture. 
 
 Methods of Conserving Heat: Owing to the high temperature at 
 which the products of combustion escape from furnaces and the large volume 
 of air necessary for the combustion of the great quantities of fuel used, 
 much of the heat generated in the furnace is carried away by the outgoing 
 gases and wasted, unless special methods are employed to recover this waste. 
 This can be done in several ways. The hot gases may be passed through 
 boilers and made to generate steam, conducted into other furnaces requiring 
 lower temperatures, or used to preheat the air and fuel and thus increase 
 the intensity of the heat in the furnace. Since high temperatures are 
 required in most metallurgical processes, the third option is generally 
 selected. The methods by which this preheating is accomplished depends 
 upon two principles, called the regenerative and the recuperative. In 
 the regenerative method the hot gases from the furnace are conducted 
 through expanded portions of the horizontal flues, almost filled with open 
 brick work, called the "checkers" from the manner of laying the bricks. 
 When the checkers have absorbed heat sufficient to raise their temperature 
 to nearly that of the gases, their connection with the stack is closed, and 
 the air, or air and fuel, if the latter can be preheated, is made to pass through 
 the checkers on its way into the furnace, thus taking on the stored up heat in 
 the checkers. With two such sets of checkers to a furnace this process is 
 made practically continuous by reversing the direction of the gases at short 
 intervals. In the recuperative method the checkers are replaced by a 
 
64 FUELS 
 
 system of pipes through which the out-going gases must pass. The in-going 
 air, being at the same time drawn in through the space around the pipes, 
 is heated in proportion as the waste gas is cooled. 
 
 Pyrometers: The measurement of high temperatures requires special 
 instruments called pyrometers, many of which are made self recording so as to 
 measure continuously the temperature of a furnace through long periods of 
 time. There are many types of these instruments, and only the principles 
 upon which some of the more important types are based will be briefly 
 described. 
 
 Specific Heat, or Water, Pyrometer : This is an old type of instrument, 
 and one that is still extensively used. The operation of the instrument is 
 carried out as follows: A weighed amount of metal of known specific heat 
 is placed in a furnace, and when it has attained the temperature of the 
 furnace, it is withdrawn and quickly dropped into an insulated vessel, con- 
 taining a definite amount of water and provided with a thermometer for 
 reading the temperature of the water. The rise in temperature of the 
 water is proportional to the weight of the ball, its specific heat, and the 
 temperature of the furnace. The first two factors being known, the 
 temperature of the furnace can be readily calculated. 
 
 Electric Resistance Pyrometers : Instruments of this type depend on 
 the fact that the electrical resistance of metals increases with rise in their 
 temperatures. In practice the metal will be platinum in the form of wire, 
 which will be inserted in one arm of a wheatstone bridge for measuring 
 resistance. A battery and a galvanometer for detecting difference in 
 potential, both being attached to the bridge, completes the apparatus. In 
 operating the instrument, the slide on the bridge is adjusted so that the 
 resistance of the two arms of the bridge are the same and the galvanometer 
 reading is zero. The "bulb" of platinum wire is now inserted in the furnace, 
 when, the resistance of the platinum wire being increased by the rise in 
 temperature, it is necessary to insert resistance in the other arm of the 
 bridge to keep the galvanometer reading at zero. The amount of the 
 resistance inserted measures the increase in resistance of the wire, which 
 can be interpreted in degrees of temperature. 
 
 Thermo-Electric Pyrometers: These instruments are both con- 
 venient and accurate, being very simple in construction. They depend 
 upon the fact that if two metals are in contact at a given point, any change 
 in temperature at that point causes an electric current, the intensity of 
 which is proportional to the change in temperature, to flow around a circuit 
 connecting their free ends. This current can be measured by inserting a 
 millevolt meter in the circuit. In practice the metals employed for high 
 temperatures are platinum in conjunction with an alloy of platinum and 10% 
 
PYROMETERS 65 
 
 rhodium or iridium in the form of wires, which are insulated from each other 
 by means of hollow tubes of refractory materials. For low temperatures the 
 baser metals may be used, such as iron and nickel-copper alloys. 
 
 Hot 
 
 Pl atinum 
 
 Junction /" Voltmeter 
 
 Pt. + 10% Iridium 
 
 FIG. 8. Diagram of Wiring for Thermo-Electric Pyrometers 
 
 Radiation Pyrometers are based on the law of heat radiation which 
 is briefly stated thus: The energy emitted by a highly heated black body 
 is proportional to the fourth power of its absolute temperature. Such 
 instruments consist of a millevoltmeter and a telescope which contains a 
 concave mirror reflector and a delicate thermo electric couple. By pointing 
 the telescope, from a certain distance, toward a highly heated surface, a 
 portion of the radiant heat is made to fall upon the mirror, which con- 
 centrates the rays upon the couple, causing it to generate a current that 
 can be measured by the millevolt meter. 
 
 Optical Pyrometers depend upon the relation of the intensity of light 
 emitted by an incandescent body and its temperature. In them the 
 intensity of the light from the hot body is compared with that of an incan- 
 descent lamp. The simplest form consists of a telescope containing the 
 lamp and a battery to supply the current. In making a determination, 
 the telescope is pointed at the heated body, and the current is adjusted 
 so that the intensity of light from the filament of the lamp matches that 
 from the body. From the adjustment necessary the temperature of the 
 body is determined. In improved forms of this instrument, a circular plate 
 of colored glass is inserted in the telescope between the lamp and the eye 
 in such a manner that the light from the lamp falls on one-half of this plate 
 and light from the body falls on the other. The two lights are matched 
 by varying the intensity of the light from the body with a diaphragm. A 
 second improvement is made by providing a special rotating prism by means 
 of which the lights from both the body and the lamp are varied in intensity. 
 The amount and direction of rotation necessary to match the lights measures 
 the temperature. 
 
 All modern pyrometers are constructed with graduated scales to read 
 in degrees, so that no calculations for converting the various relations into 
 temperatures are required. 
 
66 
 
 FUELS 
 
 SECTION II. 
 
 CLASSIFICATION OF FUELS. 
 
 Of the many ways of classifying fuels, that shown below in Table 7 
 is a very logical and simple one and most convenient for the purposes of 
 this chanter. It requires no explanation. 
 
 Table 7. Classification of Fuels. 
 
 , (Hard. 
 Wood 
 
 Carbon- 
 Hydrogen 
 Fuels 
 
 Natural 
 
 Peat 
 
 Soft. 
 
 New. 
 
 lOld. 
 
 Lignite 
 
 New. 
 
 Liquid 
 
 I-PJ., . /Coking. 
 ^ , Bituminous < ... 
 CoaH \Non-coking. 
 
 [Anthracite. 
 
 f Briquettes. 
 Prepared <{ Pulverized Coal, f Charcoal. 
 
 [ Carbonized Fuel { ~ . /Beehive. 
 v_yOke\ -p. , 
 
 \ (By-product. 
 
 Natural Petroleum . 
 
 Prepared 
 
 Gaseous 
 
 Distilled Oils. 
 Coal Tar. 
 
 Natural Natural Gas. 
 
 (Producer Gas. 
 Blast Furnace Gas. 
 Coke Oven Gas. 
 [Coal Gas. 
 
 Incidental 
 Fuels 
 
 Bessemer Converter 
 
 Silicon. 
 Phosphorus. 
 
 Sulphur Works<f^ astin S- 
 Smelting. 
 
 Iron. 
 
 Manganese. 
 Carbon. 
 
LIQUID FUELS 67 
 
 Plan of Study: In discussing the different fuels it does not seem 
 desirable to follow the order of the outline above. Some, like blast furnace 
 and coke oven gases, are best taken up in connection with the processes 
 that produce them, while others, like the distilled oils, are of so little 
 importance from a metallurgical standpoint that they cannot be more than 
 mentioned here. Concerning the others, which play more or less prominent 
 parts in metallurgical processes, it is the intention to dispose very briefly 
 of the less important first, so that the attention may be concentrated upon 
 the more important ones, which will be taken up at the last. 
 
 SECTION III. 
 
 INCIDENTAL AND LIQUID FUELS. 
 
 Incidental Fuels: Under this heading is included all substances 
 which incidentally act as fuels in certain processes. In the acid Bessemer 
 converter, for example, the oxidation of the impurities, silicon, manganese 
 and carbon, to which is added phosphorus and sulphur in the basic converter, 
 furnish heat necessary to keep the metal in a molten state during the blow, 
 and so perform the function of fuels. In the roasting and smelting of pyritic 
 ores the burning of a portion of the sulphur furnishes a great part of the 
 heat necessary for those processes. 
 
 Tar: The use of tar as a fuel is of recent origin, and offers a means 
 for the disposal of the excess quantities produced above that required by 
 the tar refiners. Having a low ash and sulphur content, it is well suited 
 chemically for use as open hearth and heating-furnace fuel. It is a viscous 
 fluid at ordinary temperatures. Hence, it must be kept at a relatively 
 high temperature both in the storage tanks and feed lines. This heating 
 is accomplished by means of steam pipes immersed in the tar. Special 
 burners, one type of which is shown in the accompanying figure, of the 
 steam or air injector type are required to burn tar properly. Tar carries 
 in suspension a great many small carbonaceous bodies, and on standing, 
 especially at the lower temperatures prevailing in storage tanks, these 
 grow into pitch-like bodies of considerable size, which clog up the burners 
 and the small pipe lines of the system. Its calorific power is between that 
 of coal and oil, 1600018000 B. t. u. per pound. This fact, coupled with 
 its low cost due to the increased production, tends to stimulate efforts to 
 use it wherever possible. 
 
68 FUELS 
 
 Tar/ 
 ijl 
 
 Air 
 
 FIG. 9. One Type of Tar Burner. 
 
 Petroleum: The only natural liquid fuel, and a material of the highest 
 commercial importance, is petroleum, a product obtained from reservoirs 
 deep in the earth. Its heating power is much greater than that of coal, 
 (The calorific power of crude petroleum=21000 B. t. u. per pound, Coal= 
 9000 to 15000 B. t. u. per pound) and it is obtained in immense quantities. 
 On distilling, it yields a high percentage of very valuable oils. On this 
 account it is used as a metallurgical fuel only where coal is scarce and 
 high in price. In using the oil special burners are required, as it must be 
 vaporized or atomized and properly mixed with air to insure complete 
 combustion. 
 
 Composition of Petroleum: Petroleum is a very complex mixture 
 of organic compounds. In small amounts it contains compounds of oxygen, 
 sulphur, and nitrogen, but principally it is composed of compounds of car- 
 bon and hydrogen. Its content of the former element varies from 84 to 
 87%, and of the latter, from 11 to 13%, depending upon the locality from 
 which it is obtained. 
 
 Hydrocarbons Generalized, Empirical and Structural Formulas: 
 
 These compounds of carbon and hydrogen found in petroleum are called 
 hydrocarbons. They are numbered by the hundred, and a study of their 
 composition has shown that they fall into a number of homologous series 
 which may be represented by generalized formulas as shown in Table 8. 
 In representing these compounds, the empirical, or simplest, formulas are 
 often found inadequate, because it frequently happens that two different 
 compounds will have the same empirical formula, and that in many com- 
 pounds the valence of carbon is apparently not a whole number. To 
 overcome this defect, the structural formula, which aims to show how the 
 molecules are built up, was invented. In these formulas the valence of 
 carbon, which is represented by 's, called valence bonds, is assumed to 
 be four in all cases, and it is also assumed that the atoms of carbon have 
 the power of uniting with each other to form nuclei to which other 
 elements may attach themselves. These formulas are also illustrated in 
 the following table: 
 
PETROLEUM 
 
 69 
 
 Table 8. The Different Homologous Series of Hydrocarbons. 
 
 Generalized 
 Formula of 
 Series 
 
 Names Applied 
 To Series 
 
 Names of First 
 Compound of Series 
 
 Empiri- 
 cal 
 Formula 
 
 Structural 
 Formula 
 
 Formula of 
 Second 
 Member 
 
 CnH2n+2 .. . 
 
 Methane, Paraffin, 
 or Chain Series. 
 
 ( Methane 
 { Marsh Gas 
 (Fire Damp 
 
 CH 4 
 
 H 
 H-C-H 
 
 A 
 
 H H 
 H-C C-H 
 H h 
 
 CnH 2n 
 
 Oleflne, Ethylene, 
 Unsaturated Open 
 Chain Series. 
 
 Ethylene. 
 
 C 2 H 4 
 
 H H 
 
 TT TT 
 
 H H 
 C=C 
 H CH 8 
 
 CnH2n-2.... 
 
 Acetylene Series. 
 
 Acetylene. 
 
 C 2 H 2 
 
 H-CsC-^H 
 
 H-C=C-CH 8 
 
 CnH2n-4 
 
 First member unk 
 
 nown. 
 
 
 
 
 CnH2n-6 
 
 Benzene, - 
 Aromatic or 
 Closed Ring Series. 
 
 Benzene. 
 
 C e H 6 
 
 H 
 
 6 
 
 H C C H 
 H-C C-H 
 
 V 
 
 H 
 
 6 
 / \ 
 H-C C-H 
 
 H-C C-CH 
 
 * 
 
 CnH2n-8 
 
 Not many membe 
 
 rs discovered. 
 
 
 
 
 CnH2n-io.. . 
 
 Not many membe 
 
 rs discovered. 
 
 
 
 
 So on to 
 
 
 
 
 
 
 CnH 2 n-ss... 
 
 Not many membe 
 
 rs discovered. 
 
 
 
 
 Hydrocarbons of the series CnHten+a make up the greater portion of 
 the paraffine base of petroleums, as is indicated in Table 9. Members of 
 the series CnH an are also constituents of many petroleums, while only a 
 few of the higher members of the series CnHan 2 and CnH2n 4, have 
 been found in oils west of Pennsylvania. The aromatic series, CnH2n e, 
 occur in small amounts in nearly all petroleums. Occurrence of members of 
 the other series is somewhat rare in petroleums, and are in small amounts 
 when found at all. 
 
 Fuel Oil and Other Products of Petroleum : The increasing demand 
 for gasoline and other petroleum products makes it very undesirable that 
 crude petroleum as obtained from the wells be used for fuel. Besides, gaso- 
 line in a fuel oil is dangerous on account of the increased danger of explosions 
 its presence entails. Fuel oil, then, is a very indefinite term that is applied 
 to any product of petroleum used for the production of heat or power. 
 There are no fixed specifications for it, and consumers order it to suit their 
 requirements. The usual grades have a calorific value of about 135000 B. t. 
 u. per gallon. The products from many of the oil refineries west of the 
 Mississippi River are gasoline, naphtha, kerosene and fuel oil, while Eastern 
 refineries usually carry the fractionation of the oil much farther, their output 
 being such products as gasoline, benzine, naphtha, kerosene, light machine 
 oil, automobile oils, cylinder oils, paraffin wax and tar, pitch, or coke. 
 
70 
 
 FUELS 
 
 SECTION IV. 
 
 GASEOUS FUELS. 
 
 Advantages of Gaseous Fuels: The many advantages possessed by 
 gaseous fuels make them ideal for many purposes. Owing to their gaseous 
 state, they require no labor in handling, and their freedom from foreign 
 matter eliminates ash and danger of contamination. As the temperature 
 is easily controlled, and the flame can be directed wherever desired, the 
 working conditions of a furnace may be kept very uniform. The kindling 
 temperature of gases is between 650 C and 700 C, and the speed of com- 
 bustion is practically instantaneous at that point, which fact makes it easy 
 to attain very high temperatures. 
 
 Table 9. The Paraffin Series of Hydrocarbons, the Members of 
 
 which are found in Natural Gas and Petroleum of the 
 
 Western Pennsylvania District. 
 
 STATE AT EMPIRICAL MELTING 5 
 
 ORDINARY NAME FORMULA 
 
 TEMP. 
 
 BOILING* 
 
 POINT POINT 
 
 Deg. C. Deg. C. 
 
 ,. [Methane CH 4 184.0 165.0 
 
 | Ethane C 2 H 6 171.4 93.0 
 
 . J Propane C 3 H 8 195 45.0 
 
 [Butane C 4 H 10 135 + .1 
 
 'Pentane C 5 H 12 130.8 36.3 
 
 Hexane C 6 H 14 94.0 69.0 
 
 Heptane C 7 H 16 97.1 98.4 
 
 Octane C 8 H 18 56.5 125.5 
 
 Nonane C 9 H 20 51.0 150.0 
 
 3 pecane C 10 H 22 31.0 173.0 
 
 Undecane CnH 24l 26.0 195.0 
 
 Dodecane C 12 H 26 12.0 214.0 
 
 Tridecane C 13 H 28 - 6.0 234.0 
 
 Tetradecane C 14 H 80 + . 5.0 252.0 
 
 Pentadecane . . C 15 H 32 10 270.0 
 
 Hexadecane C 16 H 34 18.0 287.0 
 
 Octadecane C 18 Hs 8 28.0 317.0 
 
 Eicosane C 20 H 42 37.0 
 
 Tricosane C 23 H 48 48.0 
 
 Tetracosane C 24 H 5o 51.0 
 
 Pentacosane C 25 H 52 53-54.0 
 
 Hexacosane C 26 H 54 55-56.0 
 
 Octocosane C 28 H 58 60.0 
 
 Nonocosane C 29 H 60 62-63.0 
 
 Hentriacontane C 8 iH 64 68.0 
 
 Dotriacontane C 82 H 66 70.0 
 
 Tetratriacontane C 84 H 7 o 71-72.0 
 
 Pentatriacontane C 85 H 72 75.0 
 
 See American Petroleum Industry by Raymond F. Bacon and William A. Hamor, 
 Published by McGraw-Hill Book Company, New York. Also Organic Chemistry by 
 A. F. Holleman, Fourth English Edition, published by John Wiley & Sons Inc , N Y. 
 
GASES 
 
 71 
 
 Natural Gas is the most remarkable fuel of all. Found in the earth 
 under high pressure and free from non-combustible gases, it represents a 
 perfect fuel. Upon being regenerated it undergoes partial decomposition 
 and is, therefore, never preheated, but with it the highest temperatures 
 that are practical are easily obtained by proper manipulation. The sup- 
 ply of this gas, formerly thought to be inexhaustible, is now declining rapid- 
 ly, and this fact, combined with its demand for domestic purposes, is forcing 
 its use in the metallurgical and other industrial arts to be abandoned- 
 Geologically, natural gas is closely associated with petroleum and undoubt- 
 edly is of similar origin. It is composed of the lower gaseous hydro- 
 carbons of the paraffin series, mainly methane. 
 
 Artificial Gases are manufactured from coal. The method of manu- 
 facture depends on the end sought. Thus in retort gases coal gas and 
 coke oven gas only the volatile products of the coal are utilized for 
 gas, while in the gas producer the whole combustible substance of the coal 
 is converted into gas. Of these, coal gas and by-product gas most nearly 
 approach natural gas in calorific power and efficiency. Producer gases 
 always contain a high percentage of non-combustibles. The advantages 
 in favor of the producer are that an otherwise poor fuel may be converted 
 into a desirable one, and that all of the fuel is gasified. As to blast furnace 
 gases, their utilization under boilers, in gas engines, and in blast furnace 
 stoves represents a saving that amounts to millions in a single year. A 
 detailed account of this fuel will be taken up later. Table 10 below shows 
 useful data in comparing the various gaseous fuels. 
 
 Table 10. Composition of Gaseous Fuels. 
 
 Representative Analyses 
 Percent by Volume 
 
 
 Natural Gas 
 
 Coke 
 
 Bench 
 
 Carbu- 
 
 Water 
 
 Producer Gas 
 
 Blasb 
 
 
 #1 
 
 #2 
 
 Oven 
 Gas 
 
 Coal 
 Gas 
 
 Water 
 Gas 
 
 Blue 
 Gas 
 
 #1 
 
 #2 
 
 nace 
 Gas 
 
 Carbon Dioxide, CO 2 
 
 .2 
 
 .2 
 
 1.7 
 
 2.0 
 
 3.0 
 
 3.8 
 
 5.0 
 
 10.6 
 
 12.9 
 
 Illuminants (asC2H 4 ) 
 
 .4 
 
 .5 
 
 3.0 
 
 3.7 
 
 10.0 
 
 
 
 .2 
 
 .4 
 
 o 
 
 Oxygen, O 2 
 CarbonMonoxide.CO 
 
 
 
 
 .3 
 .5 
 
 .1 
 3.5 
 
 .8 
 7.5 
 
 .5 
 34.0 
 
 .5 
 43.1 
 
 
 
 2'5.6 
 
 
 17.6 
 
 
 
 26.3T 
 
 Hydrogen, H 2 
 Methane, CH 4 
 
 
 
 77.7 
 
 
 94.5 
 
 53.9 
 34.6 
 
 48.5 
 33.0 
 
 35.5 
 12.0 
 
 47.5 
 .8 
 
 10.2 
 3.8 
 
 11.8 
 4.4 
 
 3.7 
 
 o 
 
 Ethane, C 2 H 6 
 
 19.4 
 
 
 
 not det'd 
 
 not det'd 
 
 not det'd 
 
 
 
 
 
 
 
 o 
 
 Nitrogen, N 2 
 
 23 
 
 4.0 
 
 3.2 
 
 4.5 
 
 5.0 
 
 4.3 
 
 55.2 
 
 55.2 
 
 57.1 
 
 *Net B.t.u. per cu.ft. 
 
 1027 
 
 868 
 
 518 
 
 512 
 
 465 
 
 277 
 
 148 
 
 135 
 
 94.8 
 
 Gross " " " " 
 
 1134 
 
 963 
 
 583 
 
 573 
 
 505 
 
 301 
 
 157 
 
 146 
 
 96 7 
 
 Sulphur per lOOOcu.ft 
 
 
 
 
 
 .8 Ibs. 
 
 not det'd 
 
 not det'd 
 
 
 
 .1 Ibs. 
 
 .15 Ibs. 
 
 
 
 *Does not include the latent heat of the water formed in combustion. 
 
 Principle of the Gas Producer : While there are many different types 
 of gas producer, the apparatus is essentially a vertical cylindrical shaft, 
 lined with fire brick, partially filled with coal when in use, through which 
 air, or steam and air, are forced at the bottom where combustion of the 
 non volatile part of the coal is continuous. In its upward passage the 
 carbonic acid gas formed by the combustion of the carbon of the coal is 
 reduced, in part, by the incandescent fuel, forming carbon monoxide. 
 Part of the water, if steam is used, is also acted upon by the hot coke, 
 forming carbon monoxide and free hydrogen, and some methane is 
 
72 
 
 FUELS 
 
 obtained from the distillation which the coal undergoes at first. 
 
 steam is used with the air the producer gives a gas which may 
 
 composition about as follows: 
 
 Carbonic Acid CO 2 5 to 9% 
 
 Carbon Monoxide CO 18 to 27% 
 
 Methane CH 4 2 to 4% 
 
 Hydrogen H 2 10 to 18% 
 
 Nitrogen N 2 48 to 55% 
 
 In case 
 vary in 
 
 FIG. 10. Sketch. Section Through Gas Producer. 
 
THE GAS PRODUCER 73 
 
 Factors Affecting the Efficiency of the Producer: The greatest 
 efficiency of the gas producer is attained when all the oxygen of the injected 
 air is caused to combine with carbon to form only carbon monoxide, provided 
 the excess heat thus generated is also made available. In practice these 
 results are never accomplished entirely, but efforts to attain them have 
 revealed the fact that they can be most nearly approached by carefully 
 regulating the temperature, by maintaining perfect uniformity of the 
 working conditions, and by injecting steam with the air. All these objects 
 are accomplished in fairly efficient degree in the Hughes mechanically 
 poked producer, a brief description of which follows: 
 
 The Hughes Producer as an Example of Mechanically Poked Pro- 
 ducer: This producer, a vertical section of which is illustrated in the 
 accompanying sketch, is a cylindrical steel shell, j^g" thick, lined with 
 9 inches of first quality fire brick, and closed at the ends with water sealed 
 tops and bottoms. When ready for use it sets with its base resting on five 
 wheels which are mounted on a frame carried on a concrete foundation. 
 By means of gears connected to a driving mechanism, it is rotated over 
 these wheels, the speed of rotation when in use being 1/10 r. p. m. The 
 top plate is a steel casting riveted to the charging floor, under which the 
 producer itself revolves. It contains the openings for the gas outlet, the 
 hoppers, the poker and the observation holes. There are two hoppers, 
 through which coal is fed to the producer, one on each side of the outlet, 
 but they are at different distances from the center of the producer to 
 help provide even distribution of the coal. A bell valve closes the 
 base of the hopper, and when this bell is dropped to dump the 
 coal into the producer, the hopper may be closed by sliding a circular 
 plate over its top. There are several holes three inches in diameter at 
 various points in the top seal for observing the condition of the 
 fire, and for poking out clinkers; these holes are closed with water 
 sealed caps. The poker is a round hollow steel casting with a forged 
 steel tip. It is six feet in length and tapers from eight inches in diameter 
 at the top to five inches at the tip. The poker and its trunnions are water 
 cooled, the water being admitted through the trunnions, then passing 
 through the poker to the top plate which is covered with the water to a 
 depth of five inches. From the top plate the water flows to the top seal, or 
 trough, around the top plate, then through a drain pipe to the water seal 
 in the ash pan. The top of the poker is enclosed in a gas tight mounting, 
 and is so mounted that the poker is swung back and forth through an arc 
 of about 35 by means of eccentric connections from the producer rotating 
 mechanism. A full stroke of the poker carries its tip from the center to 
 the side wall of the producer, and is timed to occupy 3.21 minutes, thus 
 allowing 3.1 strokes in one revolution of the producer. The result of the 
 rotating motion of the producer and the backward and forward action of 
 the poker is to produce a series of semi-ellipses, so that the poker covers, 
 in a period of 70.72 minutes, or 22 strokes, practically the entire area of the 
 
74 FUELS 
 
 shell. The bottom of the vessel serves as an ash pan, which must also 
 be water sealed. To form this seal the bottom is made in the form of a 
 circular trough, which is attached to the main shell of the vessel so that 
 its outer rim, or lip, extends several inches beyond this circumference of 
 the shell. Into this trough the sealing shell of the producer projects to 
 within five inches of the bottom. Since this construction leaves the central 
 portion of the bottom within the producer somewhat cone shaped, the 
 ashes are deflected toward this five-inch opening at the bottom of the trough, 
 where they may be removed through the water which flows from the top 
 and fills the trough to prevent the escape of gases. The steam blower is 
 inserted through the center of the bottom and extends some twenty inches into 
 the producer, where it is capped by a conical hood to prevent it from becom- 
 ing choked with the ashes. The mixture of steam and air is admitted just 
 beneath this hood through three rows of small openings to provide for 
 equal distribution of the blast. The ratio of steam and air is controlled 
 by the openings at the bottom of the blower, but the quantity of the mixture 
 admitted to the producer is regulated by the steam pressure, which may 
 be changed at will by the operator from the charging floor. 
 
 Conditions and Reactions: An understanding of the principle and 
 the operation of the producer is much clarified by a study of the reactions 
 and conditions prevailing in it while it is in use. A study of the conditions 
 show that there are three zones or belts of action in the producer, known 
 as the distillation or top zone, the reaction, or middle, zone, and the 
 combustion, or bottom, zone. Then, below these zones is the inactive, or 
 ash, zone. Thus, upon being charged into the producer, the raw coal is 
 first subjected to a distillation very much as in the process of coking. In 
 this top zone the volatile products are driven off, and the coal is converted 
 into a kind of coke, which will have then reached the reaction zone. Some 
 of this coke, passing through the reaction zone unchanged, reaches the 
 region just over and around the hood of the blower. Here it meets the 
 incoming air, and having been heated to above the kindling temperature, 
 combustion takes place, whereby all the remaining carbon is consumed 
 according to this simple reaction, C+O 2 =CO 2 +heat. The carbon dioxide 
 gas thus generated, together with the undecomposed steam and other gases. 
 rises at once into the reaction zone. Here the coke, having been heated 
 to a high temperature from the heat liberated by the above reaction, acts 
 as a reducing agent toward both carbon dioxide and water, thus, CO 2 +C+ 
 heat=2 CO and H 2 O+C+heat=H 2 +CO. It will be noted that both 
 these reactions absorb heat, but that only the second is under control and, 
 hence, available for lowering the temperature in the producer. The 
 reduction of all the CO 2 formed in the combustion zone has never been 
 brought about, so that a small quantity is always present in producer gas. 
 The relative amounts of CO, H 2 and CO 2 in the final gas depends to a 
 great extent upon the manipulation of the producer. As to the other com- 
 
SOLID NATURAL FUELS 75 
 
 ponents of this gas, the nitrogen, being introduced with the oxygen as air, 
 cannot be controlled, while the hydrocarbons such as CH4 and C2H4 repre- 
 sent products of the distillation. 
 
 Operation of the Hughes Producer : The ideal conditions in a Hughes 
 producer are realized when the combustion zone extends to about a foot 
 above the top of the blower hood; when the reaction zone is from one and 
 one-half to two feet thick; when the distillation zone is from one-half to 
 one foot thick; when the . conditions are such that the ash, the coke 
 and the coal occur in level zones; and when the amount of air and steam 
 are so adjusted that the fuel is properly -burned without excess of any of 
 the undesirable components in the final gas. The necessary air is injected 
 into the producer by a steam jet. Thus the steam serves the two-fold 
 purpose of injecting the air and of controlling the temperature in the pro- 
 ducer by absorbing heat, during its decomposition, which later appears as 
 potential energy in the gas. This lowering of the temperature, combined 
 with the disintegrating effect of the steam upon the ash, tends to prevent 
 clinkering. If too much steam be used, the temperature in the reaction 
 zone will drop below normal, the CO will be low and the percentage of 
 hydrogen will be high. This condition causes the gas to burn with a short, 
 intense non-luminous flame that has a detrimental effect upon the brick 
 work of the furnace in which it is used, especially upon the ports and roof 
 of the open hearth furnace. But the judicious use of steam may increase 
 the efficiency of the producer to 80 or 85% of the heating power of the fuel. 
 The experienced operator judges the quality of the gas by its appearance, 
 striving for a dense yellowish blue gas. The greatest trouble in operating 
 arises from the accumulation of unburned carbon and fine ash in the mains, 
 which must be cleaned out at regular intervals. The mains are brick 
 lined and are fitted with numerous dust catchers, or man holes, to afford 
 access for cleaning. 
 
 SECTION V. 
 
 THE SOLID NATURAL FUELS. 
 
 Analysis of Solid Natural Fuels: Upon examination all the solid 
 natural fuels are found to consist of combustible and non-combustible 
 materials. The combustible portion is composed mainly of carbon and 
 hydrogen, and the constituents of the non-combustibles are water and a 
 mixture of mineral substances called ash. By the gradual application of 
 heat without access of air, the water is first expelled, which is followed 
 closely by the combustible volatile matter, and there remains a non-volatile 
 mass composed of carbon, called fixed carbon, and ash. Upon admitting 
 air, the fixed carbon burns readily, leaving only the ash. A similar process 
 carried out so as to determine the amounts of these four classes of materials 
 is called a proximate analysis. There are two general methods for making 
 a proximate analysis, depending upon whether or not it is desired to separate 
 
76 
 
 FUELS 
 
 the volatile matter into its constituents. These are often referred to as 
 the American and the European methods, the latter being also called the 
 progressive distillation method. The determination of the percentages of 
 the various elements present in the fuel constitutes an ultimate analysis. 
 The following analysis of a coal by each of these three methods will illus- 
 trate all the points mentioned, and help to show the importance of the 
 chemical analysis. 
 
 Table 11. Analysis of a Solid Fuel, Coal, by the Three Different 
 
 Methods. 
 
 fAsh 
 
 . . .7.16% Proximate 
 
 /Ash.... 7.160% 
 * 6 \ Carbon 59.980 
 Tar 5.420 
 
 Fixed Carbon . . . 
 Proximate Volatile Matter.. 
 Analysis, 
 American Total 
 
 . 59.98 Analysis, 
 . 32.86 Progressive 
 
 Distillation 
 .100.00% A/r .-> , 
 
 Free NH 3 285 
 Comb. NH 8 . .041 
 Moist 4.765 
 
 /-\ -| f\ A / 
 
 Method Total Sulphur.. 
 Phosphorus .... 
 
 Method 
 . 1.02% 
 . .005% 
 
 Oxygen 1.046 
 Volatile Sulphur .313 
 Crude Benzol... 1.353 
 
 Total... 100.003% 
 
 Ultimate Analysis 
 
 Ash 
 C 
 
 H 
 N 
 
 .. 7.16% 
 . .. 79.41 
 . .. 5.14 
 . . . 1.46 
 
 O 
 
 s... 
 
 . . . 6.03 
 1.02 
 
 
 
 100.22% 
 
 Wood: This very interesting substance is composed mainly of 
 cellulose, CoH 10 O 5 , a compound formed in the tree or plant from water 
 taken up from the soil and carbon dioxide from the air. The change is 
 wrought mainly in the leaves of plants through the agency of sunlight. 
 Wood, then, represents potential energy, the original source of which is 
 the heat from the sun, and it, in turn, is the source of all the natural solid 
 fuels. It was the first fuel used by man, and for centuries was the principal 
 one. In metallurgy it has been supplanted by coal, though for some purposes 
 it is still used, mainly in the form of charcoal obtained by the destructive 
 distillation of wood. The calorific power of dry wood is about half that 
 of good coal. 
 
PEAT, WOOD, COAL 77 
 
 Peat is of little value as a metallurgical fuel. It finds extensive use 
 in Europe as a domestic fuel, and the better grades may be successfully 
 employed in gas producers. Its chief interest lies in the fact that it is the 
 first step in the formation of coal. Peat results from the decomposition 
 of wood substance out of contact with air. It is formed in swamps and 
 marshes from water plants of all kinds such as algae, mosses, sedges, rushes, 
 reeds, shrubs, like willows, and even trees. A species of moss called 
 sphagnum is especially important in the formation of peat. It grows on 
 the surface of still and shallow waters with only a small portion in air, 
 and as it grows the submerged portion extends farther and farther beneath 
 the surface until the bottom is reached. Starting growth near the shore 
 of shallow lakes, it gradually extends into a lake until the whole basin 
 is filled with soft carbonaceous matter, and a bog results. This growth is 
 followed by larger growths, until the former lake is packed with carbonaceous 
 matter. The accumulation being submerged, the carbon compounds of the 
 plants are slowly decomposed, by which process the carbon is isolated, 
 though a part escapes with hydrogen and oxygen as marsh gas and carbon 
 dioxide. The reaction is represented thus: 
 
 6C 6 H 10 O5 = 7CO 2 + 3 CH 4 + 14 H 2 O + C 26 H 20 O 2 
 
 Cellulose or Carbon Marsh Water Peat Substance 
 
 Wood Substance Dioxide Gas 
 
 In certain geological periods, particularly the carboniferous, the 
 conditions being more favorable for plant growth of this kind, the processes 
 described proceeded more rapidly than at present, with the result that 
 marshes of great depth and area were filled with vegetable growths. These 
 carbonaceous deposits were subsequently submerged through vertical 
 movements of the earth's crust, in which position they became covered by 
 deposits of sedimentary rocks. Later movements of the earth's crust 
 raised many of these deposits up out of the sea. In the meantime the peat 
 had been changed into coal. 
 
 Lignite and Brown Coal, geologically and chemically, occupy positions 
 intermediate between peat and coal. They were formed between the 
 Quaternary and Jurassic periods and are widely distributed. They have 
 low calorific power, and some kinds contain as much as 15% of water. 
 They are sharply distinguishable from peat, but grade into coal so gradually 
 that no one has attempted to distinguish between the oldest lignite and 
 the youngest coal. The relation between vegetable and mineral fuels are 
 more clearly shown by the accompanying table (12) and diagram (Fig. 11). 
 
78 
 
 FUELS 
 
 Table 12. Approximate Analyses of the Different Solid Fuels. 
 
 
 Air Dried 
 Wood 
 
 Air Dried 
 Peat 
 
 Lignite 
 
 Bituminous 
 Coal 
 
 Anthracite 
 Coal 
 
 
 c/ 
 
 c/ 
 
 o/ 
 
 c/ 
 
 c/ 
 
 
 7o 
 
 /o 
 
 /o 
 
 /o 
 
 /o 
 
 Vol. M 
 
 42-40 
 
 30-60 
 
 30-45 
 
 20-45 
 
 .5-6 
 
 Fixed C 
 
 39-41 
 
 11-40 
 
 45-50 
 
 40-85 
 
 85-92 
 
 Ash 
 
 .15-2 
 
 3-75 
 
 4-15 
 
 4-20 
 
 2-15 
 
 Moisture or 
 
 
 
 
 
 
 Water 
 
 20-25 
 
 6-20 
 
 10-15 
 
 1-6 
 
 .5-4 
 
 C. P. (Cals. 
 
 
 
 
 
 
 per Kilo). 
 
 4600-5000 
 
 2000-5000 
 
 3000-6000 
 
 7000-9000 
 
 9000-9500 
 
 90 
 80 
 
 70 
 60 
 
 3* 
 
 30 
 
 201 
 10 1 
 
 -tVate 
 
 Ash 
 
 Wood 
 
 Peat 
 
 Lignite 
 
 Bituminous 
 Coal 
 
 Anthracite 
 Coal 
 
 Graphit 
 
 FIG. 11. Graphic Representation of Transformation of Fuels. 
 
PEAT, WOOD, COAL 
 
 79 
 
 Periods 
 
 and 
 Chief Events 
 
 Order 
 of 
 
 Strata 
 
 Deposits 
 
 of 
 Valuable Minerals 
 
 Quaternary 
 
 Glaciers in N. E. 
 
 Peat in East 
 Lignite in West 
 Petroleum in Tex. and Lou. 
 
 Tertiary 
 Rocky Mountains 
 
 Lignite in West 
 
 Gold and Silver in West 
 
 Oil and Gas in Wyo. and Cal. 
 
 Cretaceous 
 
 Rapid Erosion, W 
 
 Lignite in West 
 Chalk in So. W. 
 
 Petroleum in Cal. 
 Colo, and La. 
 
 Wyo., Tex., 
 
 Mountains West 
 
 Jurassic 
 Warping of Surface E\^ , 
 
 Potomac Carbon Rock 
 Petroleum in Wyo. 
 
 Limestone Wyo. and Utah 
 
 Triassic 
 
 Coal in Va. 
 Petroleum in Wyo. 
 
 Slow Folding East 
 
 Surface Alternately 
 Rises and Sinks 
 
 Carboniferous 
 
 Sea Advances 
 
 JVOCK rtaii 01 i exas 
 
 Petroleum both East and Wcs1 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Coal 
 
 Gypsum 
 
 Limestone in Middle West 
 
 Pennsylvania District 
 
 Sea Retreats 
 
 Devonian 
 
 Sea Advances 
 
 Oil, Gas Pa.,W.Va.,0.,N.Y.Ind 
 Limestone, East 
 
 Limestone, East 
 i Sand Stone 
 
 Sea Retreats 
 
 Silurian 
 Land Submerged 
 
 Clinton Iron Ores, Ala. 
 Oil-Gas in N. Y., 0.. Ind., 111. 
 and Ky. 
 
 Rock Salt in N. Y. 
 
 Limestone 
 
 Zinc and Lead 
 
 Sand Stone 
 Limestone 
 
 Cambrian 
 
 Land Submerged 
 
 Huronian 
 
 Laurentian 
 
 Limestone 
 
 Lake Superior Iron Ores 
 N/ | Graphite 
 Igneous Rock 
 
 Carboniferous Period 
 Coal Beds 
 
 of 
 Western Penna. District 
 
 Name 
 
 Waynesburg 
 
 Sewickley 
 Redstone 
 Pittsburgh 
 
 Upper Frteport 
 Lower " 
 
 Upper Kittanning 4 ft. 
 
 Middle 
 
 _ Lower 
 
 Clarion 
 Brookville 
 
 Thickness 
 
 Max 
 
 10ft 
 
 6ft. 
 
 6ft. 
 
 9ft. 
 
 8ft. 
 
 5ft. 
 
 3ft. 
 8ft. 
 
 2ft. 
 5ft. 
 
 Min. 
 
 6 in. 
 
 3ft. 
 
 3ft. 
 
 5ft. 
 
 3 in. 
 
 FIG. 12. Diagram Depicting Geologic Periods in which Gas, Oil and the 
 Valuable Minerals are Found in the United States. 
 
80 FUELS 
 
 Coal : This mineral, on account of its availability, adaptability, and 
 high calorific power, has become the chief source of energy at the command 
 of man. Used both in its natural state and in prepared forms, it constitutes 
 the chief metallurgical fuel; and the high state of development of certain 
 processes, like that of the blast furnace, for example, have been possible 
 only through the peculiar properties of this remarkable substance. Its 
 origin and history is as remarkable as its properties, and though these 
 subjects belong to geology, they are of interest to the metallurgist because 
 they emphasize the need of conserving the fuel. While it has been deposited 
 in immense amounts, the supply is exhaustible and practically fixed, since 
 the rate of consumption is many times the rate at which it is being formed. 
 In this connection a study of Fig. 12 will be found interesting. 
 
 Bituminous Coal: All coal in the natural state may be looked upon 
 as being composed of coal substance, ash, and hydroscopic water. 
 Bituminous coals, on account of their peculiar properties, are the chief 
 source of metallurgical fuels. The coal substance of these coals is decom- 
 posed by distillation into carbon and a mixture of volatile compounds. 
 During this process some kinds fuse into a pasty mass, leaving at the end, 
 when all volatile matter has been expelled, a strong but porous mass called 
 coke. It is not known what the coking properties of coals depend upon. 
 Coals very much alike in physical appearance and chemical composition 
 may show widely differing coking qualities, while others differing in both 
 these respects produce cokes of equal quality. During the coking process, 
 some coals expand while others contract. This point is an important one 
 in by-product practice, because expansion wedges the coke in the oven, 
 making it difficult to remove, and causing damage to the oven walls. 
 
 Ash in Coal: The ash in coals is also an important factor in their 
 valuation. Aside from decreasing the calorific power, it affects the coal in 
 other ways. In steam coals the composition of the ash may be such that 
 it fuses at a low temperature, thereby forming large clinkers; or it may be 
 practically infusible, resulting in no clinker, with the result that a suitable 
 bed of coals cannot be kept on the grate, due to the fineness of the ash. 
 To cite a concrete example, a certain coal in the Pittsburgh District pro- 
 duced ash of approximately the following composition: SiO 2 , 45%; Al 2 Os, 
 24%; Fe 2 O 8 , 21%; CaO, 7%; MgO, 2%; P 2 O 5 , .6%. Such an ash is moder- 
 ately fusible, and so is most desirable. In the ash is found the phosphorus, 
 which determines whether coke made from a certain coal shall be used for 
 making Bessemer or basic iron. The sulphur is also important. In the 
 coal it is present mostly as FeS 2 , which, on being coked, is decomposed 
 into FeS or Fe-ySg and S, and on burning, it is completely changed, the 
 iron being oxidized to Fe 2 O 3 , or Fe 3 O 4 and the sulphur to SO 2 . Often a 
 seam of good quality coal is divided or cut horizontally by deposits of slaty 
 material known as bone coal, binder, horse back, etc., all of which must be 
 mined with the coal. Where these conditions exist, it is necessary to clean 
 
PULVERIZED COAL 81 
 
 the coal by picking, jigging, or washing. As a rule the purest coal is in 
 the middle of the seam. Phosphorus in particular occurs mainly at the 
 top. The top and bottom will always contain the highest percentages of 
 ash and sulphur. 
 
 SECTION VI. 
 
 PREPARED SOLID FUELS. 
 
 Powdered Coal : It has long been known that the combustion of finely 
 pulverized coal presents features similar to those encountered in burning 
 gases. When mixed with air and ignited, it explodes; and when it is blown 
 into a heated chamber with sufficient quantity of air,, complete and rapid 
 combustion approximating that of the fuel gases ensues. These facts led 
 to= the idea of using pulverized coal as a substitute for gaseous fuels. Though 
 first attempted about 100 years ago, no progress was made in its use until 
 recently, owing to the difficulties of securing the proper conditions, and also 
 to the abundance of other desirable fuels. Although still in the experimental 
 stage, it is now used very successfully and gives promise of replacing gaseous 
 fuels for metallurgical and many other uses. 
 
 Requirements: The use of powdered coal necessitates meeting the 
 following requirements: 
 
 1. With the apparatus now in use, only high volatile coals (volatile 
 
 matter over 30%) may be used. 
 
 2. The coal must be very finely pulverized. Approximately, 70% 
 
 should pass a 300 mesh sieve, 80% a 200 mesh, and 95% a 100 
 mesh. 
 
 3. The dust must be injected into the furnace in such a manner that 
 
 each particle enters the combustion chamber surrounded with 
 air. 
 
 4. If the coal is to be used in regenerative furnaces, special checker 
 
 work that will permit of easy cleaning is required. A high 
 percentage of ash is drawn out with the gases, which quickly 
 clogs ordinary checkers. 
 
 5. Careful regulation of draft to give a low velocity of the air and gases 
 
 is necessary to secure complete combustion, since the dust burns 
 more slowly than gases. This precaution also prevents rapid 
 clogging of checkers when the fuel is used in regenerative furnaces. 
 
 The pulverizing of coal makes it necessary to dry it thoroughly, and 
 necessitates the installation of special appliances for handling the dust, 
 which can be done only through pipes and tightly closed bins. Two general 
 
82 FUELS 
 
 methods of handling are available, namely, the worm screw and the 
 pneumatic. The third requirement calls for special burners, so constructed 
 as to permit of the regulation of the amounts of both air and dust and the 
 adjustment of the direction of the flame. The equipment will consist, then, 
 of a dryer, a pulverizer, separator, conveyors, bins, burners, and air com- 
 pressors, with the necessary motors or engines. Added to these, in many 
 cases, will be the special regenerators previously mentioned. 
 
 t 
 
 Advantages of Powdered Coal: It is adaptable for use wherever large 
 amounts of fuel are consumed, and many claims as to its advantages are 
 made. As compared with producer gas to fire open hearth furnaces, for 
 example, it is said to be as efficient and convenient as the gas and to give 
 a more regular supply of heat. It is cheaper to prepare than producer gas, 
 increases the production of steel 10% or more, and reduces the loss by 
 oxidation all without contamination of the steel from impurities in the 
 ash, if proper conditions prevail. 
 
 The Sharon Plant : The appliances for preparing and burning the coal 
 vary in form and method. A brief description of the installation at the 
 Sharon Works of the Carnegie Steel Company, which is equipped to supply 
 three 40-ton basic open hearth furnaces, may serve as an example of one of the 
 methods employed in its use. This plant was the first of the Carnegie 
 Company's to use this fuel in the open hearth. The installation is that 
 of the Raymond Bros, of Chicago, who employ an impact pulverizer with 
 an air-separating system. 
 
 Description of Pulverizing Plant: The building in which the pulver- 
 izing is done is separated from the open hearth building by about 75 yards. 
 Outside this building, is a small trestle storage bin to which the coal, 
 crushed to pass a one inch ring, is delivered from the cars. From this bin 
 the coal is Delivered by a motor driven belt conveyor to the elevated end 
 of a revolving cylindrical dryer, about 30 ft. long and 5 ft. in diameter, 
 inclined at an angle of about 10. By revolving this dryer, the coal is 
 caused to pass slowly toward the lower end, and in so doing it is stirred 
 and scattered in the cylinder, so as to be thoroughly dried by a forced 
 circulation of an atmosphere of hot gases from a small brick furnace located 
 at the elevated end of the furnace. These gases are conducted from the 
 furnace to the lower end of the dryer by a stationary flue, about 18 inches 
 in diameter and concentric with the external cylinder. Upon reaching the 
 lower end of the dryer, the coal is discharged into a hopper bin from which 
 it is elevated vertically a distance of about 20 ft., by means of belt buckets, 
 to a 25-ton storage bin. From this bin it is fed by gravity to the pulverizer, 
 through the opening of which it is mechanically fed at a rate adjusted to 
 the speed of the pulverizer, which has a capacity of 5 tons per hour. 
 Through a pipe, about 16 inches in diameter, leading from the top of the 
 pulverizer, the finest dust is pneumatically elevated to a cone shaped cyclone 
 
POWDERED COAL 83 
 
 separator, about 6 ft. in diameter and some 70 ft. above the pulverizer. 
 The mixture of air and dust enters at the top of the separator on a tangent, 
 and is, therefore, given "a swirling motion at the same time that its velocity 
 is reduced. The air, carrying very little dust, is forced out through a pipe, 
 about 24 inches in diameter, inserted in the center of the top of the separator. 
 Through this pipe the air is returned to the pulverizer, completing the 
 circuit. By this arrangement the dust in the air, on entering the separator, 
 is subject to the double effect produced by the whirling motion and the 
 reduction in speed namely, centrifugal force and gravity with the result 
 that it is precipitated upon the inclined wall of the separator and falls out 
 at the bottom through a rectangular chute (about 6"x8") leading vertically 
 down to a 12-inch screw conveyor, which carries the dust to the distributing 
 conveyor located above the open hearth furnaces. This conveyor, extending 
 at right angles to the 12-inch one, distributes the dust to six 9-ton bins, one of 
 which is located above and at each end of each furnace. The bottoms of 
 these bins are connected to small 4-inch screw conveyors, driven by variable 
 speed motors, which feed the dust to the burners at any speed desired. 
 Falling vertically through a 4-inch pipe, the dust passes through a 1-inch 
 opening into a 2-inch horizontal pipe where it is met at right angles by a 
 jet of compressed air. This jet blows the dust through the 2-inch pipe a 
 distance of about 8 inches into a 5-inch nozzle, some 16 inches long, where 
 it is mixed with a larger volume of air under the low pressure of 16 inches 
 of water. The velocity of this air is sufficient to carry the dust into the 
 furnace through a water cooled opening, where it comes in contact with 
 regenerated air and is completely burned. The air blown in through the 
 burner is about 25% of the total required for combustion. 
 
 Clairton and Homestead Plants: More recent and much larger 
 installations for powdered coal have been made at Clairton and Homestead. 
 The plant at Clairton id equipped to supply 5 of the 16 sixty-ton open hearth 
 furnaces, while the one at Homestead is designed to furnish fuel for the 
 whole of the No. 3 open hearth plant, which consists of twenty-four 60-ton 
 furnaces. At both these plants the drying, pulverizing and conveying 
 equipment is practically the same in kind as that used at Sharon, but a 
 different type of burner is used. While the burners at Sharon are 
 mechanically fed, at Clairton and Homestead they are wholly pneumatic. 
 The principle of the pneumatically fed burners is easily understood from a 
 description of the apparatus. The arrangement of the burner is shown in the 
 accompanying drawing, Fig. 14. It consists of a delivery pipe, 1^4, inches in 
 diameter by 2 feet 3 inches long, a compressed air nozzle, and a ' 'cross, ' ' which 
 is a small casting containing a 3-inch cubical cavity. The nozzle, Y% inches in 
 diameter inside, passes through one side of the cross, then across the cavity 
 and enters the delivery pipe inserted in the opposite side, so as to give an 
 injector effect that draws the dust through the one inch opening shown by the 
 dotted lines of the drawing. This opening is connected by a suitable pipe 
 to a cast iron feeder box attached to the bottom of the storage bin, located 
 
84 
 
 FUELS 
 
 some ten feet away. This box, about twelve by fourteen inches in cross 
 section, is in the form of an elbow. One end is bolted to the bottom opening 
 of the bin, while the other is closed with a steel plate. The pipe leading 
 to the burner is inserted in an opening on the top of the horizontal portion 
 
 of the box. With all connections between burner and feeder box tight, the 
 operation of the burner is very simple. By means of a valre, not shown 
 in the drawing, compressed air delivered under a pressure of about eighty 
 pounds may be admitted through the nozzle, when the suction draws the 
 
COKE 85 
 
 coal dust from the feeder box and blows it through the delivery pipe into 
 the furnace. It will be observed that the design of this burner is based 
 upon the principle that the quantity of fuel dust injected into the furnace 
 is controlled by the quantity of air passing through the injector. The 
 threaded hole in the top of the cross provides a means of attaching the 
 burner to its supports. 
 
 Coke. Coke is the residue that remains after certain bituminous coals 
 have been subjected to destructive distillation. Owing to its peculiar struc- 
 ture and physical, or mechanical, properties, it has become the chief metal- 
 lurgical fuel. All coke possesses a cellular structure, but there is a wide 
 variation in the degree of porosity for different cokes. Likewise the hardness, 
 brittleness and strength of coke is subject to wide variations. Coke for 
 blast furnace consumption should be of a porous character to admit of ready 
 combustion, and it must at all times be sufficiently strong to resist pressure 
 due to the heavy burden without crushing. In chemical composition, the 
 different cokes show a similar wide variation, though for metallurgical 
 purposes the fixed carbon, which is the only constituent sought, will con- 
 stitute 85 to 90% of the coke. The other constituents, roughly stated as 
 ash, sulphur and phosphorus, are impurities. The per cent, of phosphorus 
 in the coke determines whether the coke is suitable for making Bessemer 
 or basic iron. For the former grade, modern practice requires that the 
 phosphorus content of the coke must not exceed .018%. The sulphur content 
 ranges from .60% to 1.50%, though it is evident that both the sulphur and 
 ash should be kept as low as possible. 
 
 Methods of Manufacturing Coke: There are two methods for the 
 manufacture of coke, known as the beehive and the by-product, or retort, 
 process. In the beehive process, air is admitted to the coking chamber 
 for the purpose of burning therein all of the volatile products of the coal 
 to generate heat for further distillation. Incidentally, some of the fixed 
 carbon of the coal is also consumed. In the other method, the coking 
 chambers are air tight, and the necessary heat for distillation is supplied 
 from external combustion of the volatile products of the coal; and with 
 modern ovens, properly operated, only about half of the gaseous matter 
 from the coal is used in carrying on the coking process. While the beehive 
 process was, until recently, the leading method for the manufacture of coke, 
 it is fast being replaced by the by-product process. The processes of manu- 
 facture have very little effect on the quality of the coke, but if there is any 
 difference, the latter process has the advantage. There is, however, a 
 difference in appearance, due mainly to the difference in the coking temper- 
 ature of the two processes, that of the by-product being much lower than 
 the beehive. In general, beehive coke is silvery gray in appearance, while 
 by-product coke is of a much darker color. As examples of these two 
 methods of coking, the following brief descriptions of plants are to be taken 
 as typical of the best modern practice for each process. 
 
86 FUELS 
 
 SECTION VII. 
 
 THE BEEHIVE PROCESS FOR THE MANUFACTURE OF COKE. 
 
 The Continental No. 1 Plant of the H. C. Frick Coke Company 
 
 may be cited as an example of beehive coke practice. It is located near 
 Uniontown, Pa., in the southeastern part of the famous Connellsville coke 
 region. It consists of a coal mine and a coking plant of 400 beehive ovens. 
 
 The Mine: Since the coal bed here lies about 330 feet below the 
 surface, the coal is mined through a shaft. Although some gas is given 
 off as the coal is mined, it is prevented from collecting and thus becoming 
 dangerous by a very efficient system of ventilation, which permits the 
 installation of electrical appliances for lighting and the use of both pick 
 and machine methods of mining. The coal seam in this mine varies from 
 seven to nine feet in thickness, but in mining the coal, about four inches 
 at the bottom and from four to eight inches at the top, being high in ash, 
 sulphur and phosphorus, is allowed to remain in order to improve the quality 
 of the coke. Incidentally, this top discard also helps to support the gob, 
 an easily dislodged and treacherous slate-like formation lying between the 
 coal and the hard overlying rock deposit and forming the roof of the mine. 
 The average output of the mine is twelve hundred net tons per day. POT 
 transporting this coal through the mine underground, a combination system 
 of electric and rope haulage is employed from certain points, while horses 
 are used to distribute empty cars to and assemble loaded cars from the 
 various working places. From these assembling points the loaded cars 
 are moved by electric locomotives in trains of thirty cars each to a sub- 
 station, where they are attached to the rope haulage which pulls them to 
 the foot of the shaft. Here they are hoisted, one at a time, to the tipple 
 and automatically dumped into bins. From these bins the coal is loaded 
 by chutes into electric larries which convey it to the ovens some hundred 
 yards away. Each larry holds sufficient coal to charge one oven, and it 
 will be noted that run-of-mine coal is used for coking, no crushing nor 
 preparation of any kind being necessary. 
 
 Construction and Arrangement of the Ovens: As to the essential 
 features of construction, the name beehive is literally descriptive of the 
 form of the beehive oven. The dome-like chamber, built on a suitable 
 foundation, is constructed of highly refractory brick, has a flat but 
 slightly sloping bottom, an opening in the top, the "trunnel head," through 
 which the coal is charged and the products of distillation and combustion 
 escape, and an arched opening at the bottom, called the door, through 
 which air is admitted for combustion and the coke is watered and drawn. 
 In general, the dimensions of different ovens vary a great deal. The ovens 
 at this plant are each 12 feet 3 inches in diameter and 8 feet high from the 
 bottom to the top of the dome, inside dimensions. Of this height, the side 
 wall, built of fire brick, rises vertically a distance of 27 inches, and is capped 
 
BEEHIVE COKE 87 
 
 by the crown which is built of silica brick. Except in the case of the special 
 brick used about the openings, the brick in these walls are of standard 
 size and are laid ends in and out, thus making both the side wall and the 
 crown wall approximately 9 inches thick. Finally, this brick structure is 
 covered on the outside and up to the level of the "trunnel head," which is 
 14 inches in diameter, with loam or rough clay which acts as an insulator of 
 and a store house for heat. For retaining this loam covering, there are in 
 general three different arrangements of ovens as follows: (1) the bank 
 system, in which the ovens are built in single rows against a bank of earth, 
 natural or artificial, thus making it necessary to build but one retaining wall 
 along the front of the ovens; (2) the single block system, which consists 
 of a single row of ovens with retaining walls at both the front and back; 
 and (3) the double block system, in which the ovens, in a double row, are 
 built back to back or staggered with a retaining wall extending along the 
 front of each row. At this plant there are four double block batteries and 
 two batteries of banked ovens. 
 
 Waste Heat System : One battery of the banked ovens, forty in number, 
 is arranged for utilizing the waste heat from the products of combustion 
 to generate steam. For this purpose a large tunnel is constructed in the 
 bank some 10 feet back of the ovens and parallel to the battery. This 
 tunnel is connected to each oven by means of a small flue, which conducts 
 the hot gases out of the oven from an opening sufficiently above the side wall 
 to prevent its being closed by the largest charge of coal used. Each flue 
 is provided with a damper for closing off the draft during the period the 
 oven is being watered, drawn and charged. From the battery, the tunnel 
 passes to the boiler house, where branches conduct the hot gases through 
 the fire boxes and flues of the boilers which are connected to a common 
 stack, about 100 feet in height to cause the proper draft. During the coking 
 period, the "trunnel head" is necessarily kept tightly closed. Owing to the 
 increased draft, these ovens are inclined to run up a little higher temper- 
 ature than the ordinary oven, so that the temperature in the tunnel is 
 high, sometimes reaching 1500C. A maximum of about 800 horsepower 
 is generated from the waste heat from this battery of forty ovens. 
 
 Charging the Ovens: The ovens are charged as soon as practicable 
 after drawing, so that the stored up heat from the previous charge will be 
 sufficient to start the coking process. In the case of new work, the ovens 
 must be heated up gradually to a coking temperature by means of wood 
 and coal fires, after which period small charges of coal for coking are used 
 until the ovens reach normal working conditions. With the oven in 
 readiness for charging, the door is bricked up to within about one and one 
 half inches of the top; and the charge, which consists of six and one half 
 tons of coal for 48-hour coke and eight tons for 72-hour coke, the latter 
 being made over the week-ends, is dropped through the "trunnel head" from 
 the larry above, leaving the coal in a cone shaped pile in the oven. In 
 
88 FUELS 
 
 order to secure uniformity in the coking of the coal, this pile must be 
 levelled so that the coal wi'l lie in a bed of uniform depth over the entire 
 bottom of the oven. This result is attained by means of an electrically 
 operated leveling machine, which is moved from oven to oven on the 
 same tracks that the charging larry uses. The essential part of this 
 machine consists of a vertical rod and sleeve, on the lower end, or head, 
 of which is mounted two collapsible leveling arms. By means of suitable 
 gear connections with an electric motor, this apparatus, with the head 
 closed, may be mechanically lowered through the "trunnel head" upon the 
 apex of the pile of coal, when the rod and sleeve are made to revolve and 
 the leveling arms are at the same time slowly extended. These motions, 
 combined with the continued lowering of the head, distribute the coal to a 
 uniform depth in a very few minutes. In works not equipped with this 
 machine, the leveling is accomplished by means of a large long-handled 
 scraper, operated, by a laborer, through the door of the oven, which is 
 purposely bricked up to about only two-thirds of its height at the time of 
 charging. 
 
 The Coking Process begins very soon after the levelling is completed, 
 as the ovens retain enough heat in the brick of the walls and the loam backing 
 to start the distillation of the volatile matter of the coal. As more and more 
 heat is conducted through the walls from the hot loam backing, the tem- 
 perature of the interior of the oven soon reaches the kindling point for these 
 volatile gases, which, in the presence of the air admitted to the oven, ignite 
 with a slight explosion at first, then continue to burn quietly in the crown 
 of the oven, or, as small candle-like flames at the surface of the coking 
 mass, thus supplying heat to continue the process. The coking proceeds 
 from the top of the coal downward, so that the coking time depends mainly 
 upon the depth of the coal. The volume of volatile matter thus rapidly 
 approaches a maximum, which is maintained for a period, then declines to 
 practically nothing, hence the burning of this volatile matter must be 
 regulated by gradually closing up the opening at the top of the door for the 
 admission of air. This regulation is very necessary to maintain the 
 temperature at a maximum, and conserve coke, as an excess of air at the 
 beginning of the coking period tends to cool the oven, and later con- 
 sumes the carbon of the coke. The yield is also reduced by improper 
 leveling. If the coal is not of uniform depth to begin with, the thin portions 
 coke through before the thick, and some of the coke in the thin sections 
 is consumed while the coking of the thick portions is being completed. 
 On the other hand, if the process be stopped when the thin areas have 
 coked through, there will be a loss due to green butts on the thick areas. 
 It will be recalled that the coal assumes a semif used, or pasty, state during 
 the coking process. The result to be expected from such behavior is that 
 the coke would be found in a continuous mass, or cake, at the end of the 
 process; but, due to expansions and contractions of the mass in coking and 
 on cooling, the cake is ramified by a great number of irregular vertical 
 
BEEHIVE COKE 
 
 89 
 
 fissures, thus giving it a long columnar structure, in which the very irregular 
 columns extend from the top to the bottom of the cake. This structure 
 affords a second means by which beehive coke can be distinguished from 
 by-product. 
 
 Fia. 15. Ideal Section of Beehive Coke Oven Showing Watering Machine 
 in Use and Structure of Coke. 
 
 Watering and Drawing the Coke: As soon as the volatile matter 
 has ceased to be evolved, as indicated by a subsidence of the smoke at 
 the "trunnel head" and a decided shortening of the candle flames on the 
 surface of the coke, the coke should be drawn. In good practice, the charge 
 will be so regulated that this point is reached near the coking time assigned, 
 and the ovens will be drawn on a schedule. If any circumstances delay 
 the drawing, the doors of the ovens are sealed tight with clay, and the draft 
 at the "trunnel head" is reduced. However, it is very important that the 
 ovens be drawn on schedule, as a delay results in burning some coke and 
 in cooling the oven, so that it will not coke the next charge in the period 
 assigned. At the end of the coking time, then, the brick work closing the 
 door is torn out, and the coke is watered out. At Continental this watering 
 is accomplished by a self propelled spraying device. It consists of a tube 
 or pipe a few inches shorter than the diameter of the oven, pivoted at the 
 center to a feed pipe and perforated by two rows of holes on opposite sides, 
 starting from the center. The holes are arranged to throw jets of water 
 horizontally, which causes the pipe to revolve. Where this device is not 
 
90 FUELS 
 
 provided, the ovens are watered by spraying with a pipe on the end of a 
 hose in the hands of a laborer, who directs a stream of water through the 
 door of the oven. For drawing the coke, a Covington coke drawing machine 
 is employed at this plant. It is provided with a long arm fitted with a 
 head, flat on the bottom, but inclined on the top, and a pair of hinged ears, 
 or drawing lugs. Upon being pushed by motor into the oven, the head 
 moves in advance of the drawing lugs, which lie flat, and raises the coke 
 from the bottom of the oven. Upon the return, the lugs engage this loosened 
 coke and force it through the door in advance of the head. Here the coke 
 falls upon a belt conveyor running parallel to the ovens, and is carried to 
 the loading conveyor, which is inclined and extends at right angles to the 
 row of ovens. At the top of the loading conveyor the coke falls upon a 
 stationary screen, to separate the breeze, then slides down a chute into 
 a railroad car, and is ready for shipment. It is impossible to remove all 
 the coke with the machine, and what remains must be drawn by hand, so 
 while the machine moves forward to the next oven, a laborer cleans out 
 the oven with a long handled scraper, drawing the coke out upon the con- 
 veyor of the machine, which is more than long enough to span the distance 
 between the doors of two adjacent ovens. In straight hand drawing, the 
 coke is drawn out into the yard and forked into barrows, which are used 
 to wheel the coke into railroad cars. 
 
 Longitudinal Ovens: In order to adapt better the beehive oven to 
 the use of mechanical devices and effect a saving in labor, there appeared 
 in 1906 a modified form of the old Belgian oven, known as the Mitchell 
 oven. The essential features of this type of oven are a long narrow chamber, 
 rectangular in shape, with a flat tile bottom, an arched roof sloping towards 
 the ends, a "trunnel head" in the center of the roof, and two doors, one at 
 each end, which extend over the entire width and height of the oven ends. 
 These ovens are built side by side in blocks or batteries, and are charged, 
 controlled and watered like beehive ovens. The coke is pushed out of the 
 oven by a mechanical pusher upon a loading conveyor which is made to 
 screen the coke and drop it directly into railroad cars. 
 
 SECTION VIII. 
 
 THE BY-PRODUCT PROCESS FOR MANUFACTURING COKE. 
 
 General Features of the Process: The by-product process, being a 
 true disfillation process, involves the use of retort ovens. While there 
 are many modifications, these ovens may be said to consist essentially of 
 three main parts, namely, the coking chambers, the heating chambers, and 
 the regenerative chambers all constructed of brick. The retorts are 
 rectangular in shape, varying in general from 30 to 42 feet in length, from 
 6 to 10 feet in height, and from 17 to 22 inches in width, and are built in 
 batteries, of from 40 to 90 ovens, in which the coking chambers alternate 
 with the heating chambers. The coal is charged through openings in the 
 top of the oven, and the coke is pushed out one end by means of a power 
 driven pusher acting through the other end. All watering is done outside 
 
BY-PRODUCT COKE 91 
 
 of the oven. During the coking period, the ends of the oven are closed 
 by brick lined doors, while an opening in the top, connected with suitable 
 pipes, provides a means for the escape of the volatile products of the coal, 
 which must undergo several different treatments in order to separate the 
 many valuable products. As to the combustion chambers, these consist 
 of a great number of flues, in order to secure a uniform temperature through- 
 out the entire length of the oven, and may be either of the horizontal flue 
 or of the vertical flue type. While some of the older ovens employed the 
 recuperative principle for pre-heating the air for combustion, modern 
 practice demands the use of regenerative chambers, because the heat is 
 better conserved and less gas is required thereby to operate the oven. 
 In the arrangement of these regenerators, two plans have been employed 
 with about equally good results. By the first plan the regenerative 
 chambers, two in number, are placed longitudinally beneath a whole battery 
 of ovens, but in the second plan a small regenerator is placed under each 
 end of each oven. The latter has been employed in the most up-to-date 
 plants, because each oven is thus made more nearly an independent unit, 
 and the operation of the whole battery is not liable to be influenced by one 
 or two ovens that may be shut down for repairs or other reasons. 
 
 Advantages of the By-product Process: From the brief description 
 given above, it will be surmised that the initial cost of a by-product 
 installation is very great. Nevertheless, owing to its many advantages, 
 the method is rapidly becoming the leading process for the production of 
 coke in this country. These advantages as stated by Mr. Carl A. Meissner 
 are as follows: 
 
 1. "The by-product coke plant can be constructed at or near the blast 
 
 furnaces which are to consume its coke, and thus be under the 
 same management. 
 
 2. It is practicable to ship to it coking coals from any section within 
 
 a radius of a favorable freight rate. 
 
 3. Many coals not suitable for coking in beehive ovens become 
 
 available for by-product ovens by mixing with other coals and 
 are so used to make a first-class blast furnace coke. 
 
 4. Coking coals in by-product ovens permit of the full recovery and 
 
 use of the very valuable by-products and the gas. 
 
 5. The cost of making by-product coke at the iron and steel works 
 
 is considerably less than the cost of making beehive coke at the 
 coal mines and transporting the coke to blast furnaces, especially 
 when located some distance away from the beehive district. 
 
 6. The profits thus obtained give a substantial return on the invest- 
 
 ment in by-product coke plants, large though such investment 
 may at first appear." 
 
 The Plant of the Clairton By-product Coke Company is located 
 at Clairton, Pa., in close proximity to the Clairton Steel Works and Fur- 
 naces. This plant is the largest of its kind in the world. It consists 
 
92 
 
 FUELS 
 
 
 
 tf 
 
BY-PRODUCT COKE 93 
 
 of two units, the first of which was completed in the Spring of 1918. Each 
 unit consists of 768 ovens. The ovens are constructed in groups, or 
 batteries, of 64 ovens each, and in each unit they are arranged in two 
 parallel rows of six batteries each. As each oven has a capacity of 13.3 
 net tons of coal, more than 25000 tons of coal per day are required to 
 supply these two units when coking on a 19-hour schedule. From 
 this coal there are produced in the neighborhood of 16,700 net tons of fur- 
 nace coke; 500 net tons of domestic coke; 1,500 net tons of breeze and 
 dust, which is used to generate steam for the plant; 275,000,000 cubic feet 
 of gas (average thermal value 565 B. t. u.), 57 per cent, of which is sur- 
 plus not needed for heating the coke ovens at the plant and therefore 
 available as fuel gas for the mills; 285,000 gallons of tar; 650,000 pounds 
 of ammonium sulphate; 68,000 gallons of motor benzol, or 50,000 gallons 
 C. P. benzol, 11,500 gallons C. P. toluol and 10,500 gallons refined solvent 
 naphthas; and 10,000 pounds of crude naphthalene. The coal for the works 
 is obtained from the mines of the H. C. Frick Coke Company in the lower 
 Connellsville Field, and is known commercially as Klondike coal. These 
 mines are located near the Monongahela River, and the coal is transported 
 from the mines to the coke works by water, for which purpose more than 180 
 barges of 1,000 tons capacity each and ten steamers are employed. 
 
 Construction of the Ovens: The ovens of this plant are known as 
 the Koppers 500 cubic feet by-product oven. All parts of these ovens are 
 constructed almost entirely of the best grade of silica brick. To give the 
 coking chamber a volume of 500 cubic feet, each oven inside has a length 
 of 37 feet from face to face of the doors, a height of 9 feet 10 inches from floor 
 to roof, and a width that tapers from 17 inches at the pusher end to 19J 
 inches at the discharge end. Four "trunnel heads'* in the top provide 
 means for admitting the charge, while a separate opening at one end provides 
 an outlet for volatile matter. The oven is of the vertical flue type with 
 individual regenerative chambers. The heating chamber is composed of 
 a total of thirty vertical flues, which rise from the bottom of the chamber, 
 where they are provided with openings to the regenerative chambers and 
 to the gas mains, to a large horizontal cross-over flue on a level a little below 
 the top of the coking chamber. A dividing wall near the middle of the oven 
 separates this chamber, except the cross-over flue, into two parts with 
 sixteen vertical flues on the narrower end of the oven and fourteen on the 
 wider end. Each end, approximately each half, of the oven may thus be heated 
 alternately, and in practice the reversals are made, automatically every half 
 hour for each battery of sixty-four ovens, by means of a re versing motor con- 
 trolled by an electrical clock attachment. Two large underground flues, 
 one on each side, extending along in front of and parallel to the battery and 
 connected to the checker chambers by means of cast iron goose necks, furnish 
 means for the escape of the products of combustion. These flues lead to 
 a stack, which is located at one end of the battery and is 200 feet high in 
 order to furnish the draft necessary to draw the gases through their tortuous 
 course. An idea of the magnitude of the structure may be gained from the 
 
94 
 
 FUELS 
 
 fact that a single battery of these ovens contains the equivalent of about 
 2,500,000 nine inch brick. 
 
BY-PRODUCT COKE 
 
 95 
 
 Heating the Ovens: This construction may be further explained by 
 tracing the course of the gases when the ovens are in operation. The air 
 
 
 -2 
 
 is 
 
 55 
 
 II 
 
96 - FUELS 
 
 for combustion is admitted to the checker chamber through a capped 
 opening on the goose neck leading to the stack flue. From the top of the 
 regenerators it is delivered through individual openings into each of the 
 fourteen or sixteen vertical flues on the side of the oven where the combustion 
 is to occur. Likewise, the gas for combustion, which is conducted from the 
 gas main into a fire brick gas duct located below the vertical flues, is admitted 
 through individual fire brick nozzles to each of the vertical flues, about 
 10 inches below the air openings. Thus, the gas and air meet in the flues, 
 combustion occurs, and the hot waste gases are carried over to the opposite 
 side of the battery by the horizontal flues, then down the vertical flues, 
 through the checker work, out through the goose neck and into the large 
 flue that leads to the stack. In order to secure uniform heating of the 
 oven at all times, individual regulation of the draft in each vertical flue 
 is provided by means of a brick that may be pushed out over the top of 
 the flue to reduce the size of the opening. In the top of the oven an opening, 
 which is closed by a plug except as occasion demands it to be opened, pro- 
 vides access to the sliding brick and also to the gas nozzle in case it is 
 desired to change the amount of gas admitted. The total amount of air 
 admitted to each oven is controlled by an adjustable valve at an opening 
 in the goose neck. 
 
 Drying and Heating New Ovens: Great care is required in preparing 
 new ovens for their first charge. This preparation is carried out in two 
 stages, namely, a drying and a heating period, in both of which the tem- 
 perature of the ovens must be raised very slowly and uniformly, in order 
 to avoid uneven expansion and consequent cracking of the brick work. 
 Both operations are carried out by building fires in the coking chambers, 
 which are temporarily provided at each end and near the tops with a number 
 of small holes, less than two inches in diameter, that open into the combustion 
 chambers and thus furnish a passage for the products of combustion through 
 the flues and checkers to the stack. The drying operation is effected with 
 wood fires and occupies a period of two weeks or longer, during which time 
 the temperature of the ovens is raised to 250 F. Coal fires are then 
 substituted for the wood, and the heating period is begun. About four 
 weeks are required for this heating, during which time the temperature 
 of the ovens is raised at the rate of about 25 F. each day. The ovens 
 are then heated rapidly up to the coking temperature of 1700 F or more. 
 When available, gas may be substituted for the wood and coal for heating 
 the ovens. 
 
 Operation of the Ovens : Upon reaching the docks at the coke plant, 
 of which there are two to a unit, the coal is unloaded from the barges by 
 means of grab buckets (5 ton) which drop it into the hoppers of crushers. 
 These hoppers are provided with 2^-inch cataract screens, so that only that 
 portion of the coal that is too coarse for coking passes to the crushers. 
 Here this coarse coal is crushed to lumps 2J/ inches, or smaller, in size, and 
 falls, together with that from the cataract screen, upon a conveyor belt 
 and is carried to the eight bunkers, each of which is located above and 
 
BY-PRODUCT COKE 
 
 97 
 
 between two batteries of ovens. These bunkers have a capacity of 4,000 
 tons each, so that four bunkers contain, when filled, enough coal to supply 
 j . . . . . . . . -. one unit for 24 hours. From the bunkers 
 
 *k e coa ^ * s c ^ ar S e< i mto the ovens by 
 means of larry cars that travel length- 
 wise of the batteries and on top of the 
 ovens. Each larry holds a single oven 
 charge of 13.3 tons, and is so constructed 
 that the coal is measured both by vol- 
 ume and by weight. From the larry, 
 which has the form of four large funnels, 
 the charge is dropped into the oven 
 through the four "trunnel heads," the 
 doors of the oven having been previously 
 set in place and luted with a mixture of 
 loam or clay and coke dust. A recipro- 
 cating levelling bar, carried on the pushing 
 machine, is then inserted through a small 
 opening at the top of the door on the 
 narrower end of the oven, and the peaks 
 of coal are levelled to a uniform depth 
 of 9 feet, thus filling the oven to within 10 
 inches of the top. Finally, all openings 
 to the oven are closed and sealed, the 
 valve or damper to the gas collecting 
 main is opened, and the coking process, 
 which lasts for a period of 19 hours or 
 less, begins. The heat for coking being 
 supplied from the heating chamber by 
 conduction through the walls of the oven, 
 coking proceeds from both sides of the 
 oven toward the middle, with the result 
 that a marked plane of cleavage is 
 produced vertically down the center of 
 the whole charge. This fact gives to 
 the coke a short, block-like structure 
 that distinguishes it from beehive coke, 
 which, as previously noted, has a long 
 columnar structure. At the end of the 
 coking period the doors of the ovea are 
 moved to one side by mechanical devices 
 
 Fia. 19. Ideal Section of the 
 By-product Coke Oven Showing 
 Structure of Coke. 
 
 for the purpose, and the coke is pushed out of the oven from the narrower end 
 by means of a ram mounted upon the pusher previously mentioned. The 
 coke falls into a side-dump hopper car, is carried therein to a quenching, or 
 watering house, of which there is one at each end of a row of batteries, 
 is there watered by an overhead spray until well blackened, but still hot 
 
FUELS 
 
 enough to dry itself, and is then discharged into an inclined dock or bin. 
 Here it is allowed to become dry and to cool somewhat, after which period 
 it is permitted to fall upon a large belt conveyor and is carried up an incline 
 to the screening house. The coke then falls upon an incline screen, known 
 as the adjustable Grizzley bar screen. The bars usually being adjusted to 
 give a 1 A inch opening at the top and a M inch opening at the bottom, the 
 furnace coke is separated from the breeze and dust and drops into a railroad 
 car placed ready to receive it. The material that passes this screen may 
 be further divided by rotary screens into dust and domestic coke, which is 
 also loaded directly into cars. At this plant all the dust is used under 
 boilers to generate steam for use at the plant. The volatile products from 
 the coal pass out of the oven and are conducted through pipes to the by- 
 product plants, of which there are two, one for each unit. 
 
 SECTION IX. 
 
 THE BY-PRODUCT PLANT. 
 
 The Volatile Matter of Coal is a very complex mixture. It may be 
 roughly divided into three classes of substances, based on their state at 
 ordinary temperatures; namely, the fixed gases, or those substances that 
 are gases at ordinary temperatures, the liquids, and the solids. The fixed 
 gases are hydrogen, Ha; methane, CH 4 , also known as marsh gas; ethane, 
 C 2 H 6 ; propane, C 8 H 8 ; butane, C 4 Hi ; ethylene, C 2 H 4 ; small amounts of 
 propylene, C 3 H 6 ; butylene, C 4 H 8 ; acetylene, C 2 H 2 ; carbon dioxide, CO 2 ; 
 carbon monoxide, CO; hydrogen sulphide, H 2 S; nitrogen, N 2 ; oxygen, O 2 ; 
 and ammonia, NHs. The vapors that are liquid at ordinary temperatures 
 are benzene, C 6 H 6 ; toluene, C 6 H 5 CH 3 ; xylene, C 6 H 4 (CH 3 ) 2 ; carbon disul- 
 phide CS 2 ; and aqueous vapors. Among the vapors that are solid at ordinary 
 temperatures, are naphthalene, CioH 8 ; phenol, also known as carbolic acid, 
 C 6 H 5 OH; anthracene, Ci 4 Hi ; and many others, all of which, together with 
 heavy pitch-like substances, soot carbon, and small amounts of many of the 
 more volatile liquid compounds cited above, enter into and make up the tar. 
 
 Gas Mains and Coolers: All these substances pass out of the ovens 
 through up-takes at their narrower ends and into the U-shaped gas collecting 
 main that extends above and parallel to a battery. The gases and vapors 
 enter this collecting main at a temperature of about 400 C, and under 
 a uniform suction of about .078 inch (2 mm.) of water, which is kept 
 constant by means of a Northwestern gage governor and valve. From 
 the collecting main the gas is conducted by two pipes to a large main, 
 known as the suction main, which serves as a common main for one half of 
 a row of six batteries. This suction main leads to the primary coolers. In 
 passing through these mains, the temperature of the gases drops to about 
 75 C, at the inlet to the primary coolers. This reduction in temperature 
 causes much of the heavy tar vapors to condense in the mains, and it is 
 found necessary to maintain a heavy stream of new flushing tar (composed 
 
TAR AND AMMONIA 99 
 
 of tar, 50%, and ammonia liquor, 50%), flowing through the collecting 
 mains to keep them clear of pitch and carbon stoppages. The require- 
 ments for this flushing tar amount to approximately 450 gallons of tar and 
 weak liquor to be circulated through the gas mains for each ton of coal 
 carbonized in the ovens. The primary coolers are large rectangular tanks 
 provided with tubes through wjiich water circulates, while the gas, in its 
 passage through the cooling chamber, is brought into intimate contact 
 with these pipes. The gas leaves these coolers at a temperature of about 
 32 C. The cooling of the gas in its travel from the ovens through the gas 
 mains and primary coolers results in the condensation of about 95 per cent 
 of all the tar and water vapor. The condensation takes place about as 
 follows: 50 per cent in the collecting mains and cross-over mains, 35 per 
 cent in the suction main, and 10 per cent in the primary coolers. The con- 
 densing vapors carry with them all of the fixed ammonia in the gas which 
 amounts to about 15 per cent, of the total ammonia. 
 
 Separation of the Tar and Ammonia Liquor: These condensed 
 liquids, composed of about 70% tar and 30% ammonia liquor, are conducted 
 through pipes to two large tanks known as the hot drain tanks, or tar wells, 
 whence a small portion is pumped back into the gas mains as flushing tar 
 and the remainder to two separating tanks. In these tanks the tar and 
 liquor, the former of which has a specific gravity varying from 1.15 to 
 1.17 at 15 C while the latter is little, if any, heavier than water, are allowed 
 to separate by gravity, when the liquor is drained off into storage 
 tanks and the tar is pumped also into storage tanks. From these 
 storage tanks the tar, which is composed of water, 2%, pitch, 65%, 
 and heavy oil, 33%, and has a heating value of about 16,500 B. t. u. 
 per lb., is withdrawn to a small loading tank from which it is loaded by 
 gravity into tank cars as it is required for shipment; but since the semi- 
 direct process for the recovery of ammonia is employed, the ammonia 
 liquor, containing about 1.1% of ammonia, is pumped to ammonia stills. 
 Here, the liquor is brought into contact with steam heated lime water, 
 which liberates the ammonia. This ammonia gas is then conducted through 
 cast iron pipes back to certain points in the gas mains, where it is disposed 
 of in a manner to be described later. If desired, this ammonia gas may be 
 conducted into water to produce concentrated ammonia liquor. For oper- 
 ating these stills exhaust steam from various engines in the plant is used. 
 
 Compressors and Tar Extractors: After the gas leaves the primary 
 coolers, it enters a number of positive exhausters (Connersville Exhausters) 
 which produce a suction of 15 inches of water on the entering side and 
 compress the gas to a pressure equivalent to 50 inches of water on the 
 discharge side. This pressure is required in order to force the gas through 
 the apparatus succeeding, the first of which are the P. and A. (Pelouze 
 and Audouin) tar extractors. In each of these extractors the gas stream, 
 by means of a perforated plate, is broken up into innumerable small jets 
 which impinge upon the cold surface of a plate immediately behind the 
 
100 FUELS 
 
 perforated plate. The impact causes the very fine particles of tar to collect 
 on the impact plate, and the tar, thus accumulating, runs off the plate and 
 out of the apparatus through a sealed overflow at the bottom. In these 
 apparatus it is necessary to maintain a constant differential pressure of 
 about 8 inches of water, and since the holes in the perforated plate tend 
 to become closed by the more viscous of the tarry substances, thus causing 
 an increase of the pressure, special means must be employed to overcome 
 this tendency. At this plant the desired result is accomplished by lower- 
 ing the tar level in the bottom of the apparatus, thus exposing more holes 
 as those in use become clogged. The tar level is controlled by means of 
 a pressure gauge and automatic regulator attached to the gate valve 
 through which the tar passes in flowing out of the apparatus. The tar 
 extracted by this machine amounts to about 5% of the total tar originally 
 carried by the gas. 
 
 Recovery of Ammonia: The temperature of the gas, now about 38C., 
 having been raised about 6C. by compression in the exhausters, is brought 
 to about 66 C. by being forced through preheaters, which are cylindrical 
 steel tanks containing steam coils. This preheating is necessary to 
 prevent the accumulation of water in the saturators and to accelerate the 
 reaction, between the ammonia and the dilute sulphuric acid, that occurs 
 in them. These saturators, of which there are ten to a unit, are large 
 lead lined steel pots containing a 5% solution of sulphuric acid, through 
 which the gas is forced in tiny bubbles. This gas, it is to be noted, 
 contains all the ammonia recovered from the coal, for that which 
 was liberated in the ammonia liquor stills, previously described, has 
 been introduced into the gas mains just after the latter leaves the 
 preheaters. In this way all the ammonia given off by the coal in coking 
 is brought into direct contact with the dilute acid, with which it immedi- 
 ately reacts to form ammonium sulphate, (NH 4 ) 2 SO 4 . This salt dissolves 
 in the water with which the acid was diluted, but, when the baths become 
 saturated, it is precipitated and settles to the bottom, where it is forced 
 through syphon ejectors by means of compressed air to elevated draining 
 tables, also lead lined. From the draining tables, the salt is periodically 
 removed, placed in centrifugal dryers, and whizzed for fifteen minutes, 
 which process removes nearly all the water, the salt retaining about 
 2.0% of its own weight of moisture. The mother liquor derived from the 
 drying operations, as well as the wash water used to free the crystals of the 
 slightly acid mother liquor, flows back into the saturators, while the salt 
 is scraped off the copper screen plates of the centrifugal machines with 
 wooden paddles and delivered through a chute to a belt conveyor, which 
 carries it to a final dryer, where the moisture content, by means of hot 
 gases, may be reduced to .25% or less. The final drying prevents caking, 
 so that the salt will remain in a finely divided state for indefinite periods. 
 From the final dryer the salt falls into a pit, from which it is removed with 
 grab buckets to a storage pile, to be shipped later as required. 
 
BENZOI' PLANT 101 
 
 Debenzolating the Gas: In bubbling through the liquid in the 
 saturator, the gas tends to carry a little of the acid along with it. Hence, 
 from the saturator the gas passes into an acid separator. Its temperature 
 here is about 54 C. which is much too high for the complete separation 
 of the benzene and its homologues. Therefore, the gas is put through final 
 coolers where its temperature is lowered to 30 C. These coolers are tall 
 steel towers, about 100 feet in height. In them the gas is brought into 
 direct contact with cold water, which is introduced at the top, while the 
 gas enters at the bottom and leaves at the top. From these coolers the 
 gas is forced through three benzol, or oil, scrubbers in series. Like the 
 coolers, these scrubbers are large steel towers, in which the principle of 
 counter currents is employed throughout. They are filled with a kind of 
 checker work of wooden slats. A product from the refining of petroleum, (or of 
 tar), known as straw oil or wash oil, with a distilling temperature ranging from 
 270 to 370 C., is sprayed into the top of the washers, where it trickles down 
 over the wooden checker work and is thus brought into intimate contact 
 with the ascending current of gases. The oil absorbs the benzene, toluene, 
 xylene, naphtha and naphthalene, becoming saturated to the extent of 
 about 3%, and removing 92% or more of the total amount of these products 
 in the gas. The entire removal of the naphthalene at this point is of great 
 importance, because, if any remains in the gas, it crystallizes out and clogs 
 the gas lines. From the scrubbers, the oil carrying the benzene, toluene, 
 etc., is pumped to the benzol plant, which serves both units of the plant, 
 while the gas, now freed from all except its fixed gases, is divided, half 
 being sent to the fuel lines to heat up the ovens and half to the booster 
 station, where it is compressed by steam turbo-blowers and delivered to 
 the mills as surplus gas. The loss in heating power of the gas from a given 
 quantity of coal, due to the removal of the by-products, amounts to about 
 
 5.8%. 
 
 SECTION X. 
 
 THE BENZOL PLANT. 
 
 Light Oil: At the benzol plant the wash oil, carrying in solution the 
 benzene, naphthalene, and their homologues, is first delivered to the wash 
 oil, or separating, stills. In order to conserve as much heat as possible, 
 the inflowing wash oil is made to serve as a condensing liquid for the 
 vapors from the still. After leaving the condensers, or heat exchangers, 
 the oil passes to preheaters, or superheaters, in which its temperature is 
 raised to about 145C. and much of the benzol is vaporized. The oil then 
 passes to the stills, where it comes into direct contact with steam which 
 drives off the higher boiling oils and naphthalene. Since the wash oil has 
 a much higher boiling point than the hydrocarbons it is desired to recover, 
 only a small portion of it escapes from the stills. The vapors from the 
 stills, passing into the condensers, or heat exchangers, are cooled and con- 
 dense to form a liquid, known as "light oil," which flows from the bottom 
 of the condensers into storage tanks. The wash oil, which is not vapor- 
 ized, flows from the bottom of the stills and is conducted to heat exchang- 
 
102 t :>:.. FUELS 
 
 ers, and then to water coolers where its temperature is lowered to 30C. 
 From these coolers it is pumped back to the oil scrubbers, and can be used 
 repeatedly. However, there is a daily loss of approximately 2%. 
 
 Composition of the Light Oil: The light oil is pumped from the 
 storage tank to the crude still. The composition of a light oil is approx- 
 imately as follows: 
 
 Light Runnings (benzol and carbon bisulphide). 1.00% 
 
 Pure Benzol 57.00% 
 
 Pure Toluol 14.50% 
 
 No. 1 Refined Solvent Naphtha 4.50% 
 
 No. 2 Crude Heavy Solvent Naphtha 1.50% 
 
 Crude Naphthalene 40% by weight. 
 
 Wash Oil 12.00% 
 
 Construction and Principles of the Still : The crude still consists 
 of three sections. The lowest section is a horizontal cylinder provided 
 with steam coils, and has a capacity of 20,000 gallons; the second part, 
 called the fractionating column, is a vertical column mounted upon this 
 cylinder and composed of thirty-one bell sections for scrubbing the vapors 
 as they pass upward; the third part, mounted on top of the column and 
 called the dephlegmator, is a short horizontal cylinder, that contains a 
 number of water cooled pipes and acts as a partial condenser. The separa- 
 tion of light oil into its component oils is effected by taking advantage of their 
 different boiling points. As the temperature of the light oil is raised and ap- 
 proaches the boiling point of the first runnings, this liquid is vaporized and 
 passes up through the columns; a portion of the other oils with higher boiling 
 points is also vaporized, but is condensed before reaching the top of the col- 
 umn. There are thus two movements in the fractionating column and de- 
 phlegmator, the vapors going upward and the condensed oils flowing down- 
 ward into the still. The vapors are thus forced to pass through the return 
 oils which aid in condensing the vaporized oils of higher boiling points and 
 permit only the lighter vaporized oils to reach the top of the dephlegmator, 
 where they are condensed and flow from the still through a manifold into 
 the storage tanks. As the temperature of the still is further raised, the 
 benzol, toluol, etc., is successively vaporized and condensed, and flows from 
 the still. However, it is impossible to separate absolutely, the benzol from 
 the toluol, since, before the benzol is completely driven over, some toluol will 
 be carried along with it. It is likewise impossible to separate absolutely one 
 from another the other constituent oils. Hence, it is only aimed to separate 
 roughly the light oil into what are termed fractions. These fractions, desig- 
 nated in the order in which they are made, are as follows: Light Runnings, 
 Crude 90% Benzol, Crude 90% Toluol, Crude Light Solvent Naphtha, Crude 
 Heavy Solvent Naphtha and Still Residue. Each fraction is stored in a 
 separate tank. 
 
 Operation of the Crude Still: The details of the operation of the 
 crude still are as follows: 20,000 gallons of light oil are charged. The 
 
BENZOL PLANT 103 
 
 temperature is gradually raised, and approximately 1,600 gallons of oil are 
 vaporized and condensed. This product, known as the light runnings and 
 consisting of benzol containing approximately 3% carbon bisulphide, is 
 conducted into the light runnings storage tank. The next product is the 
 90% benzol. The still is continued on this fraction until a test shows 
 30% by volume will distill over at 100C. Ninety per cent, toluol is then 
 collected until a test shows that 10% will distill over at 130C. The next 
 product is the light solvent naphtha which is collected until the flow of 
 oil is very small, at which point, and continuing throughout the operation, 
 the still is maintained under partial vacuum. The oil in this fraction is 
 collected until 10% will distill over at 160C., when heavy solvent naphtha 
 is produced until a test shows that 90% will distill over at 205 C. The 
 residue in the still, consisting of wash oil and naphthalene, is drained into 
 the naphthalene pans, where, upon cooling, the naphthalene separates 
 as a solid. The wash oil is removed by a centrifugal machine, and 
 the naphthalene is washed with hot water. It is sold as crude naphthalene, 
 or refined and sold as C. P. naphthalene. 
 
 Washing the Products of the Crude Stills: Before the products of 
 the crude still are further treated for the separation of their component 
 oils, they are pumped to the washer and there agitated with sulphuric acid. 
 The object of this treatment is to free the oils from unsaturated hydrocarbon 
 compounds, paraffins and other impurities. These substances are acted 
 upon and polymerized by the acid, to form substances that have very high boil- 
 ing points. Some of these are insoluble in the oils and will settle out with 
 the acid, forming a sludge. Several thousand gallons of oil are transferred 
 to the washer and 66 Damne" sulphuric acid is added, the proportions being 
 about 920 pounds of acid to 5,000 gallons oil. The contents of the washer 
 are agitated for twenty minutes and allowed to stand for fifteen minutes, 
 when the sludge settles to the bottom and is drawn off. This acid sludge is 
 then heated in special pots with live steam, and the acid thus separated from 
 the carbon aceous matter, when it is used in the saturators to produce ammon- 
 ium sulphate. For the purest product the oil is then washed an additional 
 number of times in the same manner. After the use of the acid, the oil 
 is washed with 10% caustic soda solution until the last trace of acid is 
 neutralized. The oil is then transferred to the pure still. 
 
 The Pure Stills: The construction and operation of the pure still is 
 practically identical to that of the crude still. However, while the fractions 
 of the latter consist of a mixture of oils, the pure still is operated so that 
 one or more fractions may be pure compounds, as is shown by the following 
 data: 
 
 Still Charge 17,676 Gals. Washed 90% Benzol. 
 Fraction. Time Gals. Product. 
 
 1 12% hrs. 1440 R. R. (Rerun) Benzol Benzol containing 
 
 Carbon Bi-sulphide. 
 
 2 31 " 10950 C. P. Benzol. 
 
104 FUELS 
 
 Fraction. Time Gals. Product. 
 
 3 % hrs. 790 R. R. (Rerun) Benzol Greater portion 
 
 Benzol, part Toluol. 
 
 Residue 4496 Principally Toluol. 
 
 Fraction 2, being pure benzol, is not further treated, and is ready for the 
 market. The rerun (R. R.) benzol fraction, 1 and 3, and the residue are 
 stored in separate tanks. When a sufficient quantity of a R. R. fraction 
 has accumulated, the pure still can be charged with it and a C. P. 
 (chemically pure) product obtained, as is shown by the following data: 
 Still Charge 18,200 Gals. 
 R. R. Benzol. 
 Fraction. Time Gals. Product. 
 
 1 16% hrs. 2550 R. R. Benzol containing Carbon Bisulphide. 
 
 2 39 " 10900 C. P. Benzol. 
 
 3 10 " 1750 R. R. Benzol Greater portion Benzol, part 
 
 Toluol. 
 
 4 5 " 650 R. R. Toluol Part Benzol, greater portion 
 
 Toluol. 
 
 5 6K " 1300 C. P. Toluol. 
 
 Residue 1050 Toluol and Solvent Naphtha. 
 
 Fraction 2 and 5 being C. P., are ready for the market, the other 
 fractions are stored and, when a sufficient amount is collected, are distilled 
 like the R. R. Benzol charge just described. It is not essential that the 
 pure still be charged with a straight R. R., or 90% product, as a mixture 
 of the two is often distilled as follows: 
 
 Still Charge 6800 Gals. 90% Washed Toluol. 
 
 10150 " R. R. Toluol. 
 Fraction. Time. Gals. Product. 
 
 1 15M hrs. 1610 R. R. Benzol greater portion Benzol, part 
 
 Toluol. 
 
 2 29 " 5130 R. R. Toluol greater portion Toluol, part 
 
 Benzol. 
 
 3 24 6950 C. P. Toluol. 
 
 4 6M 400 R. R. Toluol greater portion Toluol, part 
 
 Naphtha. 
 
 2860 Residue Toluol and Naphtha. 
 
 It is thus apparent that the light oil will eventually be completely 
 worked up into its pure product. There is one exception, however, in that 
 the light runnings containing the carbon bisulphide cannot be separated by 
 fractional distillation. The crude carbon bisulphide benzol is the first 
 1600 gallons that come over , in the crude still. This is placed in a 
 separate tank, into which the forerunnings, or carbon bisulphide benzol, 
 from the pure still is also collected. When 20,000 gallons of this product, 
 containing approximately 3% carbon bisulphide, is collected, it is placed 
 
BENZOL, TOLUOL, NAPHTHA 105 
 
 in the 90% crude still. The first 4000 gallons condensed is benzol 
 containing 10% carbon bisulphide, the remaining portion is 90% benzol, 
 which is transferred to the crude 90% benzol tank. Only a restricted 
 market has so far been found for the 10% carbon bisulphide benzol. 
 
 SECTION XI. 
 
 SOME PROPERTIES AND USES OF THE RAW 
 BY-PRODUCTS FROM THE COKE WORKS. 
 
 Characteristics of Benzol, Toluol and Naphtha: While the names 
 benzol, toluol, xylol and naphtha are those commonly applied in commerce, 
 chemical names used to designate corresponding pure compounds are 
 benzene, toluene, xylene, cumene, etc. Naphtha is a mixture of several 
 compounds, including xylene, cumene and others, so it has no chemical 
 name. As previously indicated, these compounds are members of the 
 aromatic, or benzene, series of hydrocarbons represented by the general 
 formula C n H 2n _ e. The empirical formulas for benzene, C 6 H 6 , and toluene, 
 C 7 H 8 , represent individual compounds, but these formulas for xylene, 
 C 8 Hio, and cumene, C 9 Hi 2 , represent series of isomeric compounds, which, 
 though they are members of the same series and may have the same formula, 
 differ widely in properties. Thus the formula CgHio, may represent 
 
 CH 8 CH 3 
 
 6 6 
 
 / \ / \ 
 
 orthoxylene H-C C-CH 3 ; metaxylene H-C C-H; 
 
 H-C C-H H-C C-CHs 
 
 \ = \ S 
 
 c c 
 
 A it 
 
 CH 8 C 2 H 5 
 
 6 fc 
 
 / \ x \ 
 
 paraxylene, H-C C-H; or ethylbenzene, H-C C-H. 
 
 H-C C-H H-C C-H 
 
 VV \ S 
 
 c c 
 
 The chief physical properties of the first and more important members 
 of the series are given in the following table, which will also give some 
 idea of the method of naming the compounds: 
 
106 
 
 FUELS 
 
 Table 13. Some Members of the Benzene Series and 
 Their Physical Properties. 
 
 FORMULAS 
 
 NAME 
 
 State at 
 Ordinary 
 Temperature 
 
 Melting or 
 Freezing 
 Point C. 
 
 Boiling 
 Point 
 C. 
 
 Specific 
 Gravity 
 
 Empirica 
 
 Rational 
 
 C 6 H 6 . 
 C 7 H 8 . 
 
 CgHio 
 CgHia . 
 
 CioHi 4 
 and so o 
 
 CeHe 
 
 Benzene. . . . 
 
 Toluene or 
 Methyl- 
 benzene. . . 
 
 Ortho xylene 
 or Ortho- 
 dimethyl- 
 benzene. . . 
 
 Meta-xylene 
 or dimethyl- 
 benzene. . . 
 Para-xylene 
 or dimethyl- 
 benzene. . . . 
 
 Ethylbenzene 
 
 Hemimelli- 
 thene or v 
 Tri methyl- 
 benzene. . . . 
 Pseudocu- 
 mene or 
 as-Trimeth- 
 ylbenzene. . 
 
 Mesitylene or 
 s-trimethyl- 
 benzene. . . . 
 
 Normal 
 Propyl- 
 benzene. . . . 
 
 Cumene or 
 Isopropyl- 
 benzene. . . . 
 
 Prehitene or 
 Tetra- 
 methyl- 
 benzene. . . . 
 
 Clear Liquid 
 
 +5.4 
 92.4 
 
 28 
 
 54.8 
 +13 
 
 80.4 
 110.3 
 
 142.0 
 
 139.1 
 138.0 
 136.0 
 
 175.Q 
 169.5 
 165.0 
 159.0 
 153.0 
 
 fl5l 
 
 -{-} 
 
 -'(-I 
 I*"/ 
 
 f o1 
 
 -(?} 
 
 -{-} 
 
 \4/ 
 
 -{5} 
 
 -{?} 
 
 -{I-:} 
 
 {?} 
 
 /15\ 
 
 (f) 
 
 8 Jli!l 
 \ff 
 
 CeHs- CHs 
 C6H 4 . (CHs)2 . . . . 
 
 
 
 \ 
 
 C6H 5 . (C2H 5 ) .... 
 C6H 8 .(CHs)3.... 
 
 
 
 
 C6H5.C 3 H 7 
 
 
 
 CeHa- (CHs)4. . . . 
 
 C6H4.CH3.C3H 7 .. 
 tt to C25H44 
 
 Cymene or 
 Methyliso- 
 propyl- 
 benzene . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
BENZOL 107 
 
 Commercial Benzol: The term "benzol" is employed commercially in 
 connection With the various mixtures of the benzene hydrocarbons. Pure 
 benzene is usually marketed as "Chemically Pure", or "C.P.," benzol. The 
 mixtures are generally designated as 90 % washed benzol, 90% crude benzol, 
 80 % washed benzol, 80 % crude benzol, and so forth. The terms "washed" 
 and "crude " denote whether or not the product has been washed with sulphuric 
 acid to remove the various unsaturated compounds which are always present 
 in the crude product. As a rule, most of these products are washed to a greater 
 or less degree of refinement, according to the purpose for which they are required. 
 The terms " 90 % "or " 80 % "refer to the amount of-the fraction which will distil 
 over up to 100 C. The lower this percentage, the greater will be the amounts 
 of toluol and solvent naphthas present in the product. The 90% washed benzol 
 will contain approximately 80% benzol, 15% toluol, and 5 % of xylols and light 
 solvent naphthas. 
 
 Uses of Commercial Benzols: Benzols have been largely used as 
 solvents for fats, waxes, gums, and resins. Mixed with alcohol and ammonia, 
 benzols of these grades make an excellent cleanser for the removal of grease 
 and paint. Its solvent action on gums and resins makes it a valuable substance 
 in the manufacture of paints, varnishes, and lacs, especially of enamel, bronze 
 and aluminum paints, in which a neutral gum, or resin, is used to form the 
 bases. The solvent action of these grades upon rubber makes them valuable 
 in the preparation of cements and insulating varnishes. They dissolve sulphur 
 mono-chloride, and are hence used in the cold vulcanization of rubber. Heavy 
 benzol and solvent naphtha are employed in the preparation of enamels, 
 wood stains, varnishes, and waterproofing materials, such as the rubberized 
 cloth known as mackintosh. Certain grades may be used as a substitute 
 for turpentine in paints intended to cover resinous woods. Naphtha, in par- 
 ticular, is important as a rubber solvent in the manufacture of rubber goods, 
 and as a solvent for anthracene during the final purification of this substance 
 with sulphuric acid. It is also used in the cleaning of clothing. 
 
 Motor Benzol: A product corresponding approximately to 80% washed 
 benzol makes an excellent motor fuel for automobile engines, though up until 
 1919 it had not been extensively used as such in the United States. Inasmuch 
 as the commercial use of pure benzol and pure toluol is comparatively limited, 
 since the demand for explosives has decreased, fully 80% of all the benzol and 
 toluol now produced in the United States is used as a motor fuel. For this 
 purpose there is no necessity of separating the toluol from the benzol,and all 
 of these two substances, together with practically all of the xylol, are combined 
 in one product. This product is marketed as " Motor Benzol". Pure benzol, 
 freezing at 5.4C., congeals readily in cold weather, but the presence of the 
 toluol and xylol in motor benzol lowers this freezing point so that cars may be 
 satisfactorily operated during freezing weather. Gasoline to the amount 
 of about 25% mixed with this product makes a mixture suitable for the coldest 
 weather. Any carburetor adapted for gasoline can be used as well for motor 
 benzol, but more air, or oxygen, is required in the explosive mixture with this 
 
108 
 
 FUELS 
 
 
 So. .03 
 
 Q o 
 
 O 
 
 "* <#~ u \ iS 
 
 
 
 T 
 
 I 
 
 bfl 
 C, 
 
 I 
 
 t/3 
 
 O ' 
 8*8 
 
 
 
 w o 
 
 S a 
 
 8 
 
 c;.l 
 
 8T8 
 
 s -I ,K ! 
 
PURE BENZOL 109 
 
 fuel than with gasoline. When properly handled, motor benzol gives from 
 20% to 30% greater mileage than does gasoline. At the present time motor 
 benzol is made under the following specifications: 
 
 Color Water- white. 
 
 Distillation Start 78 to 82 C . 
 
 Dry not higher than 
 
 135 C. 
 
 Wash Test No 9, or better. 
 
 Sulphur content not to exceed 0.25% . 
 
 The wash test is made by agitating equal quantities, usually 20 cc., 
 of the oil and pure sulphuric acid in a glass tube and comparing the color 
 with that of numbered standards, the first of which, No. 1, is perfectly clear. 
 
 Properties and Uses of Pure Benzol, or Benzene : The chief physical 
 properties of this very interesting substance have already been given. 
 Concerning its chemical properties and uses, it may be said that it is one 
 of the most interesting and useful compounds known to the chemical 
 profession. This fact is more fully appreciated when it is known that it 
 is the base from which such drugs as phenol, hydroquinon, antipyrin and 
 acetanilid; such dye stuffs as resorcinol, benzidine, aniline, and indigo; and 
 such explosives as nitrobenzol and picric acid are prepared. The relations 
 of benzene to these compounds is best shown briefly by means of some 
 such diagram as that on the opposite page. 
 
 To illustrate the reactions by means of which some of the more 
 important compounds are derived from benzene, the following tables have 
 been prepared: 
 
 Table 15. Reactions Showing How Aniline and Benzidine Are 
 Derived from Benzene. 
 
 (H 2 SO4)=C 6 H 5 NO2+H 2 O ( 
 Nitro Benzene. 
 
 Nitro Benzene. Azobenzene. 
 
 C6H 5 'NO2+6HC1+ 3Fe=C 6 H 5 -NH2+ 3Fe C1 2 + 2H 2 O 
 Nitro Benzene. Aniline. 
 
 Azobenzene. Benzidine. 
 
 Substantive Cotton Dyes. 
 
110 FUELS 
 
 Table 16. Reactions Showing How Phenol, Picric Aci4 and 
 Resorcinol May Be Derived from Benzene. 
 
 TT 
 
 CG Hc+H 2 SO4=H 2 O + 
 
 Benzene Sulphonic Acid. 
 
 C 6 H= C 6 H3 + H 2 
 
 Potassium Benzene Sulphonate. 
 
 C 6 H5 >S 3+2KOH=K2 S03+C 6 H 5 OK+H 2 
 (Heated to fusion) Potassium Phenate; 
 
 2 C 6 H 5 OK+H 2 SO4=2C 6 H 5 OH+K 2 
 Phenol. 
 
 C 6 H 5 OH+H 2 S04=H 2 
 
 Phenol. Para Benzene Sulphonic Acid. 
 
 TT 
 
 OH-C 6 H4 >S 3+2KOH=K2S 3 + H 2 0+C 6 H 4 (OH) 2 
 
 Metadihydroxyl benzene 
 
 or Resorcinol 
 
 (Resorcine) 
 
 Base of many colors. 
 
 2C 6 H 5 
 
 Phenol. Picric Acid. 
 
TOLUENE 111 
 
 Uses of Toluene: One of the chief uses of toluene is found in the 
 manufacture of high explosives. These are prepared by nitrating toluene. 
 Both the di-nitro-toluene and the tri-nitro-toluene are used, but of these 
 the latter, often referred to as T. N. T., is more important. In intensity 
 of explosion it ranks below picric acid, derived from phenol, but is much 
 safer to handle, because the acid has the property of reacting direct with 
 metals to form picrates, which are very sensitive to shock, whereas T. N. T. 
 does not form dangerous salts with metals and is not sensitive to mechanical 
 shock. It is also replacing gun cotton, or nitrocellulose, in torpedoes, mines, 
 etc. 
 
 CH 3 
 
 I 
 
 C 
 
 / ^ 
 
 O2N-C C-NOg 
 
 II 
 
 Its formula is represented thus: H C C H 
 
 \ ^ 
 C 
 
 N0 2 
 
 Since the molecule of T. N. T. does not contain enough oxygen for 
 the complete combustion of the carbon atoms, it produces much smoke on 
 burning or exploding. This defect is overcome by mixing with it some 
 nitrate, preferably ammonium or lead nitrate. In addition to its use as 
 an explosive, toluene is also the base from which saccharin, benzoic acid, 
 many dyestuffs, and perfumes are prepared, as a glance at the accompany- 
 ing diagram will show: (See Table 17.) 
 
 Commercial Toluol and Solvent Naphtha: Commerical toluol, 
 often spoken of as 90% toluol, is a mixture composed mainly of toluol, 
 90% of which will distill at 120C and not more than 5% at 100C. In the 
 case of solvent naphtha, 90% distills at 160C and not over 5% at 130. 
 These mixtures are often used instead of benzol, because they evaporate 
 more slowly or because they have a higher flash point. Some of the 
 industries in which their use is found advantageous are the manufacture 
 of automobile tires, rubber cements, artificial leather, wood stains, paint 
 and varnish removers, paints (as a substitute for turpentine) and special 
 inks. 
 
112 
 
 FUELS 
 
 w a a 
 
 3-g; 
 
 w **+ 
 
 >* > 
 
 M -^ 
 W 
 
 33 35 
 
 -< 
 
 Q a 
 
 -t: 
 
 w a 
 
 z 
 
 5 
 
 i 
 
 w 
 
 a a 
 
 1!^ 
 
 g^ 5 
 
NAPHTHALENE 113 
 
 Uses of Naphthalene: Ci H 8 does not belong to the benzene series, 
 CnH 2 n 6, but is the first member of a series represented by the general 
 
 formula CnH 2 n 12, and by the structural formula H H 
 
 i i 
 C C 
 
 i\/\ 
 
 H-C C C-H 
 
 i ii i 
 H-C C C-H 
 \/\ = 
 C C 
 I I 
 H H 
 
 It is used as an antiseptic and insecticide, and is familiar to every one 
 in the form of moth balls. But its real importance lies in the fact that 
 it is the base from which many dyestuffs are prepared, chief of which is 
 indigo. The steps by which this important dye is derived from naphthalene 
 are about asfollows: (1) Naphthalene, Ci H 8 , is heated with fuming sulphuric 
 acid and mercury, which acts as a catalytic agent, whereupon there is 
 
 formed phthalic acid, C 6 H 4 (COOH) 2 , (2) which, on being heated, passes into 
 
 CO 
 
 phthalic anhydride, C 6 H4< or .>O (3) from which phthalimide, 
 
 v/U 
 
 CO 
 
 C 6 H4< >NH, is obtained with the aid of ammonia and heat 
 v^O 
 
 (4). By oxidizing phthalimide with bleaching powder, anthranilic acid, 
 C 6 H4-NH2-COOH is formed, (5) which is changed by treatment with 
 monochloacetic acid to phenyl-glycine-ortho-carboxylic acid, C 6 H 4 -COOH- 
 NH-CH 2 -COOH, (6) By fusing this compound with caustic soda, indoxyl, 
 C2H4NH-CO-CH2, is formed and is readily oxidized by the oxygen of the 
 air to indigotin, C 6 H 4 .NH.CO.C=C.CO.NH.C 6 H 4 , or indigo-blue. The 
 following table will serve to show the many other dyestuffs that may be 
 obtained from naphthalene: 
 
114 
 
 FUELS 
 
 Table 18. Showing Some Products Derived from Naphthalene. 
 
 NAPHTHALENE 
 
 C 10 H 8 
 
 Naphthionic Acid Phthalic Acid Nitronaphthalene ce and /3 Naphthols 
 C H4< COOH C ioH 7 N0 2 Ci H 7 OH 
 
 Dye Indigo Dye Explosives Dye Stuffs such as 
 Stuffs Stuffs Biebrich scarlet, 
 
 M a r t i u s and 
 Naphthol yellow. 
 
 Congo Red 
 
 Dye 
 
 Tar: As obtained from gas works as well as from by-product coke 
 plants, tar is a black, viscous, oily liquid, with a specific gravity that varies 
 from 1.15 to 1.20, that from coke works having a gravity of about 1.16. 
 It also varies a great deal in other respects, especially in composition. 
 It is a very complex substance; the number of its compounds have 
 been estimated at about 300, only some of which have been isolated in the 
 laboratory and but a very few in the commercial working up of the liquid. 
 In the crude state it may be used as a fuel and for road dressing, but, by 
 refining, it is made to yield a great number of products of great economic 
 and hygienic importance. The refining of tar forms an industry by itself, 
 which requires volumes to describe in all its details. Suffice it to say, 
 that the refining of tar is essentially a process of fractional distillation, 
 in which it is first separated roughly into several parts, which may 
 then be further rectified into purer substances as shown in the following 
 diagram, which is also made to indicate the uses to which the products 
 are applied. 
 
COAL TAR PRODUCTS 
 
 115 
 
 I 
 
 5 
 
 -I I 
 
 ^Jj. 
 
 1 u 
 
 S c 
 
 -2 >> 
 
 1 
 
 
 ?! 
 
 I* 5 
 
 1 
 
 il 
 
 -* 
 
 I'-ii 
 
 1 r*|J 
 
 iir 
 
 i 
 
116 AMMONIUM SULPHATE 
 
 Ammonia, as concentrated ammonia liquor (NH 4 OH and H 2 O), is 
 used in making anhydrous ammonia gas (NH 3 ) for refrigeration purposes 
 and the aqua ammonia of commerce, used for cleaning. It is also used in 
 the manufacture of baking soda and a large number of ammonium salts, 
 such as ammonium chloride and ammonium nitrate. The last named salt 
 is extensively used in the manufacture of explosives. By passing ammonia 
 and air over heated platinum black, the former is oxidized to nitric acid, 
 and large quantities of ammonium nitrate are now produced by this method. 
 Ammonia is used extensively in dye works, and a considerable amount is 
 consumed by chemists in analytical laboratories. 
 
 Ammonium Sulphate, (NH 4 ) 2 SO4, is a white crystalline salt, very 
 soluble in water and easily decomposed by heat, beginning at 140, into 
 NH 4 HSO 4 and NH 3 , and at a red heat into NH 8 , SO 2 and H 2 O. Unlike 
 most ammonium salts, it cannot be sublimed. As obtained from the coke 
 works, it is slightly discolored by small amounts of various impurities 
 which it is impossible to exclude entirely. It is used for a number of 
 purposes, including the preparation of ammonium persulphate and nitrate, 
 but its great field of usefulness is exhibited as a commercial agricultural 
 fertilizer. William H. Childs of the Barrett Co. who has made a careful 
 study of the use of the salt for this purpose speaks thus of it: 
 
 Use of Ammonium Sulphate as a Fertilizer: "Sulphate of ammonia 
 is extensively used in ready mixed fertilizers, which is the form generally 
 purchased by the American farmer. These usually contain acid phosphate 
 and potash, together with sulphate of ammonia, tankage, cotton-seed meal, 
 etc. Sulphate of ammonia is dry in its nature, and makes an excellent 
 mixture as far as mechanical condition goes, with the added advantage 
 that it does not react with the other fertilizer chemicals to cause loss of 
 nitrogen or reversion of the acid phosphate, both of which points are claimed 
 against nitrate of soda. The nitrogen in sulphate of ammonia is quick to 
 act, is not easily leached out of the soil, and it continues its action over a 
 considerable period, so that the growing plant is carried along to maturity 
 without setback. Its only disadvantage is the tendency to exhaust the 
 lime in the soil. While this point is apt to be urged by Agricultural Experi- 
 ment Station men, it is really of minor importance because the actual 
 amount of sulphate of ammonia in the usual fertilizer application is small, 
 and its nitrogen is relatively so much more beneficial to the growth of the 
 crop. The liming of the soil, which, of course, overcomes all objections, 
 is urgently recommended by all Experiment Station advisers, and in large 
 areas of the Eastern States is practically the foundation of profitable 
 agriculture. On the other hand, in some soils, as in those of Southern 
 California and parts of Texas, which tend to excess of alkali, the action 
 of sulphate of ammonia is peculiarly beneficial. In some soils sulphur is 
 lacking, so that the sulphur in sulphate of ammonia actually acts as a 
 plant food." 
 
FLUXES 117 
 
 CHAPTER V. 
 
 FLUXES AND SLAGS. 
 
 SECTION I. 
 
 FLUXES. 
 
 Smelting and the Functions of a Flux: Any metallurgical operation 
 in which the metal sought is separated, in a state of fusion, from the 
 impurities with which it may be chemically combined or physically mixed 
 is called smelting. Since botlj these conditions with regard to impurities 
 are usually present, smelting involves two processes; namely, the reduction 
 of the metal from its compounds and its separation from the mechanical 
 mixture. Many of these impurities may be of a highly refractory nature, 
 and if they were to remain unfused, they would choke up the furnace, retard 
 the separation of the metal and interfere in various other ways with the 
 smelting. To render such substances more easily fusible is the primary 
 function of a flux. Again, some elements, being reduced almost simul- 
 taneously with the metal, combine chemically with it, while other elements 
 and some radicals, chemically combined with the metal in the raw materials, 
 refuse to be separated from it, except there be present some substance for 
 which they have a greater chemical affinity. To furnish a substance with 
 which these elements and radicals may combine in preference to the metal 
 is the second function of the flux. 
 
 The Selection of the Proper Flux for a Given Process is, then, chiefly 
 a chemical problem and requires a knowledge of the chemical composition 
 of all the materials entering into the process. With this knowledge in 
 hand, the selection will be governed by well known physical and chemical 
 laws, chief of which is the action of acids and bases toward each other 
 and the fusibility of the various compounds thus formed. In general, if 
 the matter to be fluxed is basic, such as lime, magnesia and other compounds 
 of base forming elements, the flux must be acid, while if the impurities be 
 acid, such as silica and phosphoric acid, a basic flux will be required. In 
 most ores the impurities will belong to both classes with one or the other 
 class, usually the acids, predominating. In a few iron ores the two classes 
 of impurities are so well balanced as to render the ores self-fluxing, or by 
 proper mixing they can be made so. In order to control fusibility, a neutral 
 flux is sometimes required. The cost of the flux is also to be considered, 
 hence the natural deposits of greatest purity that are easy of access and in 
 
118 FLUXES 
 
 close proximity to the works are made use of. A brief discussion of the 
 fluxes of greatest importance in the iron and steel industry follows: 
 
 Acid Fluxes: Silica is the only substance that may be classed as a 
 strictly acid flux. For this purpose it is available as sand, gravel and 
 quartz in large quantities and in a sufficiently pure state. In blast furnace 
 practice, it is customary to employ acid open hearth or Bessemer slags or 
 ores of high silica content when it is desired to increase the acids in the 
 furnace, as,in this way,the metallic contents of these substances are recovered. 
 
 Alumina: Unlike silica, which is a strong acid under all conditions, 
 alumina may perform either the function of an acid or a base, depending 
 upon the conditions imposed. Thus, with silica, it forms aluminum silicate, 
 and with a strong base, such as sodium, sodium aluminate. A marked 
 peculiarity is its tendency to form, in conjunction with other bases, double 
 salts with polybasic acids. As a rule, double silicates are more easily 
 fused than those containing a single base. Alumina is seldom used inten- 
 tionally as a flux, but it is present in nearly all raw material, hence unavoid- 
 able. 
 
 Basic Fluxes : The chief natural fluxes of this class are limestone and 
 dolomite. In addition, iron and manganese oxides act as such in certain 
 processes where their performing this function is uncontrollable, as is the 
 case in the acid open hearth. Referring to limestone and dolomite as blast 
 furnace materials, there is a difference of opinion among furnacemen as to 
 their relative value as fluxes. Some hold that limestone is the better, 
 while others maintain that dolomite gives as good, if not better results, 
 their opinions usually being influenced by their training and by the extent 
 of their experience with these materials. The presence of magnesium in 
 limestone in small amounts has little effect, but as the content increases, 
 it may lower the fusion point of the resultant slag by the formation of 
 double salts. A high percentage (over 3%) of magnesia in blast furnace 
 slag renders it undesirable for cement, but for concrete, ballast, etc., it is 
 desirable, as it makes the slag harder. Aside from this objection, not 
 one of much weight, the factor that governs the choice between limestone 
 and dolomite is the cost per ton of available base. 
 
 Available Base: By available base is meant the amount of basic 
 substance that remains in the raw flux after the acids of its own 
 content are satisfied. Referring to the analysis of limestones on a 
 succeeding page, it is at once noticed that the total is not 100%. The 
 substance that is missing is carbon dioxide, which constitutes 44.0% of pure 
 calcium carbonate, and, being evoived as a gas, is seldom determined in 
 making an analysis. Using the Bessemer stone as an example and 
 remembering that the iron and phosphorus are completely reduced in the 
 
LIMESTONE 119 
 
 furnace, we have remaining SiO 2 , 3.43%; A1 2 O 3 , .86%; CaO, 51.45%; 
 MgO, 1.66%. If now, it is desired to produce a slag in which the 
 combined weight of the bases (CaO+MgO) equals the silica and alumina 
 (SiO 2 +Al 2 O 3 ) the available v base=(51.45+1.66) (3.43+.86)=48.82% 
 In a similar manner the basic stone will show 52.66% available base, if it be 
 calculated on the same slag basis. 
 
 Limestone, which term also includes dolomite belongs to the sedi- 
 mentary class of rock-formation and is widely distributed. Immense 
 deposits underlie most of the area drained by the Ohio and Mississippi 
 rivers. The best of these deposits are of more ancient origin than our 
 coal beds, belonging to the Mississippi, or early carboniferous, period and 
 previous geological periods. Some limestone is formed by chemical pre- 
 cipitation, but the greatest portion of these natural deposits in all prob- 
 ability originated by the accumulation of the remains of minute sea animals. 
 During succeeding periods these deposits were covered to a great depth. 
 They were later made accessible by processes of uplifting and erosion which 
 have exposed the strata in places. 
 
 Supply of Limestone: The main supply of limestone for Carnegie 
 Steel Company comes from the Altoona, New Castle, Martinsburg and Butler 
 County districts. In the Altoona and Martinsburg districts the beds have 
 been folded and broken in the uplifting and lie at various angles, while in 
 the other fields their position is horizontal or nearly so. In all the districts 
 mentioned the ledges representing the various deposits vary in their silica con- 
 tent, but by aprocess of combining the stone from the different ledges, the silica 
 content of the different car loads is kept very constant, rarely exceeding 
 5%. Great care is exercised to keep the stone as free from clay as 
 possible, as this is one of the disturbing elements in uniform blast furnace 
 operation. At most of the quarries, suitable crushing and screening 
 devices have been installed to size the stone and remove the more 
 silicious fines. 
 
 Action of Limestone in Furnaces: In the blast furnace, limestone is 
 not affected, excepting for the liberation of carbonic acid, until the lower 
 levels and higher temperatures in the furnace are reached. In the smelting 
 zone, or the regions just below, it combines with the gangue, forming slag, 
 and also unites with varying amounts of sulphur depending upon the con- 
 ditions of temperature and basicity of the slag. Ordinarily, about one-half 
 ton of limestone is required in the production of one ton of pig iron. In 
 the manufacture of steel, it plays the part of a purifying agent. Phosphorus, 
 in particular, cannot be removed, commercially at least, without it. Stone 
 with the lowest silica content is usually reserved for open hearth furnaces, 
 that with the lowest phosphorus for furnaces making Bessemer pig iron, 
 while stones with higher phosphorus content are used in furnaces making 
 
120 SLAGS 
 
 iron for basic open hearths. An analysis typical of each of these grades is 
 shown in the following table: 
 
 Table 20. Representative Analyses of the Three Different Grades 
 
 of Limestone. 
 
 Open Hearth Bessemer Pig Basic Pig 
 
 Silica 80% 3.43% 1.20% 
 
 Iron 10 % .30 % .60 % 
 
 Phosphorus 005% .006% .033% 
 
 Moisture 10 % .60 % .60 % 
 
 Alumina 16% .86% .70% 
 
 Lime 54.90% 51.45% 53.88% 
 
 Magnesia 47% 1.66% .68% 
 
 Neutral Fluxes : For the purpose of making slags more fusible without 
 changing their acidity or basicity, neutral substances having very low 
 fusion points may be used. This practice is common in basic open hearths. 
 Fluorspar is the substance generally used, though calcium chloride can be 
 substituted. 
 
 SECTION II. 
 
 SLAGS. 
 
 Slag is the name applied to the fused product formed by the action 
 of the flux upon the gangue of an ore and fuel, or upon the oxidized impurities 
 in a metal. As previously indicated, it results from the neutralization of 
 bases and acids, hence corresponds to the salts of wet chemistry. The 
 word cinder is used interchangeable with slag, but cinder is also applied 
 to refuse in a solid form. 
 
 Functions of Slags: On account of their fusibility, chemical activity, 
 dissolving power, and low density, slags furnish the means by which the 
 impurities are separated from the metal and removed from the furnace. 
 Incidentally, they perform other important functions. Lying upon the 
 molten metal, they serve as a blanket to protect it from the injurious action 
 of hot gases, and being poor conductors, they prevent over heating of the 
 metal and at the same time conserve its heat by preventing radiation. 
 Since they possess the power of dissolving oxides, they mark a sharp line 
 between reduced and unreduced material, and on this account serve to keep 
 the metal clean. 
 
 Importance of Slags: In the metallurgy of iron, the importance of 
 slags cannot be over emphasized. In the blast furnace they furnish the 
 only positive means of removing sulphur, and, as their fusion temperature 
 varies with their composition, they are one of the means by which hearth 
 temperature is regulated. On this account, the slag controls to the greatest 
 extent the quality of the iron produced. In the open hearth, particularly 
 in the basic process, the slag is the only means by which the impurities 
 
SLAGS 121 
 
 in pig iron, excepting carbon, are removed. To the metallurgist a know- 
 ledge of the properties of slags is essential. He understands their 
 chemical behavior, knows their formation temperatures, fusibility and 
 fluidity, and how to control these factors. 
 
 The Chemical Composition of Slags is within the control of the 
 metallurgist, and by varying it, slags of almost any set of properties desired 
 may be produced. Slags are mainly composed of two or more silicates in 
 which other substances are dissolved or suspended. In the blast furnace, the 
 slag will consist principally of calcium silicate, with a part of which the 
 magnesium charged into the furnace will be found as a double salt. The 
 same may also be the case with the small quantities of iron, manganese, 
 and traces of alkali found in these slags. The sulphur removed will be in the 
 form of CaS, which dissolves in this mixture. As to the state of alumina, 
 which usually makes up 12% or more of the slag, there is room for doubt. 
 By some it is considered as a base; by others, an acid. As noted under the 
 heading of fluxes, it has the properties of both an acid and abase. Hence, 
 being governed by the Law of Mass Action, it acts as a stabilizer to main- 
 tain a kind of equilibrium between acids and bases. In a highly silicious 
 slag it may side with lime to form a double silicate, while in a strongly 
 basic slag it takes the place of silica in neutralizing lime and magnesia. 
 Since it has very weak properties in either direction, it seems reasonable 
 to suppose that, in a case where lime and silica are in stable proportions, 
 all or a part of the alumina may play a neutral part and dissolve in 
 the slag. In ordinary blast furnace practice, however, the sum of the 
 silica and alumina (SiC^+A^Os) is considered as the acid of the slag, 
 while lime plus magnesia (CaO+MgO) is taken to represent the base. 
 
 Relation of Acids to Bases in Blast Furnace Slags: By chemical 
 analysis of blast furnace slags it is found that, usually, SiO2+Al2O3= 
 about 48% of the slag, the ratio of SiC>2 to A^Oa being about 2:1. After 
 deducting from the lime enough to satisfy the sulphur, the sum of the 
 remaining lime together with the magnesia (CaO+MgO) will also be about 
 48%. This relation of acid to base will generally vary through a range of 
 about 2%, any increase in one being followed by a corresponding decrease 
 in the other. The remaining 4% to 5% is made up of CaS and small amounts 
 of ferrous and manganous oxides. As previously indicated, the ratio of 
 lime to magnesia may vary somewhat without noticeably affecting the 
 properties of the slag. The following results of an analysis represent a 
 slag production by a furnace making basic iron. 
 
 Table 21. Showing Relation of Acids to Bases in Blast Furnace Slags. 
 
 Acids Bases 
 
 SiO 2 35.02% CaO 44.03% FeO 1.16% 
 
 A1 2 O 3 14.99" MgO 2.72 " MnO 1.08" 
 
 50.01" 46.75" S 1.35" 
 
122 SLAGS 
 
 Ratio of Acids to Bases in Open Hearth Slags : Final basic open hearth 
 slags contain a much higher percentage of bases than blast furnace slags 
 and a much lower percentage of acids. The lime and magnesia will always 
 be more than twice the silica, alumina and phosphorus. Some open hearth 
 furnacemen hold that the best results are obtained when the percentage f 
 of lime plus magnesia in the final slag is three times that of the silica. The 
 strong basic character of basic slags is necessary for the removal of phos- 
 phorus and the small and variable amounts of sulphur possible by the 
 process. If the percentage of lime is too high, the slag will be viscous 
 and retard the working of the heat. The following analysis is the average 
 tapping slags from thirteen basic furnaces. 
 
 Table 22. Relation of Acids to Bases in Basic Slags. 
 
 Acids, per cent. Bases, per cent. 
 
 SiO 2 20.74 CaO 40.90 
 
 A1 2 O 3 3.55 MgO 9.67 
 
 P 2 O 5 2.85 FeO 10.84 
 
 S .04 Fe 2 O 3 5.24 
 
 SO 3 .23 MnO 5.86 
 
 27.41 72.51 
 
 Acid to Base in Acid Furnaces: Both in the acid open hearth and 
 in the acid Bessemer processes the slags will consist, practically, of the 
 oxides of iron and manganese silicates. In both cases the silica, SiO 2 , is 
 usually about 50%; that in the acid open hearth is seldom higher than 
 52% or lower than 48%, while acid converter slag will sometimes contain 
 as high as 65%. The remainder of about 50% will consist of FeO and 
 MnO, together with small amounts of lime and magnesia and traces of 
 phosphorus and sulphur. Acid open hearth slags are self regulating as to 
 the amount of FeO and MnO. These oxides, in such slags, are always 
 present in such quantities that their combined percentage is equal to about 
 46% of the slag. 
 
 Electric Steel Furnace Slags: The composition of these slags is 
 affected by the kind of steel made and the method of refinement used. 
 However, the final slag of an electric steel, that is, the slag formed near 
 the end of the reducing period, should be very basic, very low in iron and 
 manganese content and show a goodly percentage of calcium carbide. The 
 following analysis may be considered as representing a good average finish- 
 ing slag for this process: 
 
 Silica, 17.90%; Iron, .32%; Lime, 61.51%; Magnesia, 7.47%; Manganese, 
 .35%; Sulphur, 1.30%; Calcium Carbide, .51%; Alumina, 6.00%. 
 
 'Acids Formed by Silicon: A study of slags is facilitated by a study 
 of the acids of silicon. There are a number of these acids which chemists 
 consider as being derived from orthosilicic acid, H4SiO4 or (H 2 O) 2 SiO 2 , 
 
SLAGS 123 
 
 through the loss of varying amounts of water, and are called polysilicic 
 acids. When orthosilicic acid, (H^O^-SiOjj, is set free from its salts, it 
 always forms H2O-SiO2, which is called, therefore, metasilicic acid or 
 normal silicic acid. The relations of these various acids are shown in the 
 following table: 
 
 Table 23. Acids Formed by Silicon. 
 
 Orthosilicic Acid Metasilicic Acid Polysilicic Acids Water 
 
 (H 2 O)2-SiO2 H2O. SiO2 + H2O 
 
 2(H 2 0) 2 Si0 2 . /(H20)3.(Si0 2 ) 2 + H 2 
 
 \ H 2 .(Si0 2 )2 H-3H 2 
 
 I (H a O)4- (Si0 2 ) 3 -f- 2H 2 
 
 3(H 2 0) 2 Si0 2 \ (H 2 0) 2 . (Si0 2 )3 + 4H 2 
 
 { H 2 O .(SiO 2 )3 +5H 2 O 
 
 Besides these acids, salts of other acids, which^do not fit into this table, 
 are known to exist, such as (H 2 O)4-SiO2. By substituting bases like CaO, 
 MgO, FeO, MnO, for H2O in these formulas, silicates such as are formed 
 in slags would be represented. In the case of sesquioxides, as Fe 2 O3 and 
 Al 2 Os, in which the valence of the metal is- increased to three, this sub- 
 stitution cannot be made on a basis of 1 for 1, but 2 for 3, and the formulas 
 are, therefore, more complicated. 
 
 So-called Acid and Basic Slags : In substituting a base, such as CaO, 
 in the formulas above, it will be observed that the relation of base to silica 
 varies in the different compounds through the wide range from 4 CaO 
 combined with 1 SiO 2 , (CaO)4-SiO 2 , to 1 CaO combined with 3 Si0 2 , 
 CaO-(SiO2)3. The salts of the meta-and the ortho-acids, (H2O)-SiO2 and 
 (H2O)2-SiO2, in which, for example, CaO is combined with 1 SiO 2 , and 
 2 CaO with 1 SiO2, appear to occupy positions of equilibrium or neutrality. 
 Any increase of 'lime increases the affinity of the slag for acids, whilst a 
 decrease in lime content causes the slag to exhibit acid properties. Slags 
 of this composition are also very fusible and flow readily. 
 
 Classification of Slags: Some metallurgists classify and name the 
 slags derived from the acids of silicon according to the ratio of oxygen in 
 the base to oxygen in the acid, as shown in table 24: 
 
 As illustrating the importance and value of this table, it is suffi- 
 cient to say that it is often made the basis of calculation for the theo- 
 retical burdening of the blast furnace. These calculations are necessary 
 in dealing with new and unfamiliar materials in order to determine the 
 proper proportion of fuel, ore and flux in the charge. From the type of 
 slag best suited to produce the kind of iron desired, the oxygen ratio is 
 fixed upon, which in turn determines the relation of acids to bases. Then, 
 with the analysis of the raw materials in hand, the impurities in each are 
 
124 
 
 SLAGS 
 
 combined according to this relation. As a result of this combination the 
 excess acids of fuel and ore are found, and the available base of the flux. 
 These quantities, being combined in accordance with the slag ratio, will 
 then, with the exception of variations in fuel consumption, fix the relations 
 of the three materials. 
 
 Table 24. Method of Classifying Slags. 
 
 Monoxide 
 Base 
 (CaO) 4 .SiO 2 
 (CaO) 2 .SiO 2 
 (CaO) 4 .(SiO 2 )3 
 CaO.SiO 2 
 (CaO) 2 .(SiO 2 ) 3 
 
 Sesquioxide Oxygen 
 Base in Base 
 
 (Al 2 3 ) 4 -(Si0 2 ) 3 2 : 
 (Al 2 O 3 ) 2 -(SiO 2 ) 3 1 : 
 (Al 2 O 3 ) 4 .(SiO 2 ) 9 2 : 
 (A1 2 3 ) .(Si0 2 ) 3 1 : 
 
 (Al 2 O 3 ) 2 .(SiO 2 ) 9 1 : 
 
 Oxygen 
 in Acid Name 
 1 Subsilicate 
 1 Monosilicate 
 3 Sesquisilicate 
 2 Bisilicate 
 3 Trisilicate 
 
 Fusibility 
 Fusible 
 Very Fusible 
 Very Fusible 
 Moderately 
 Less Fusible 
 
 Uses of Slags: While to the metallurgist slags represent refuse no 
 longer useful to his art, they may be applied to many purposes. Railroad 
 ballast, road building, roof covering, concrete work, Portland cement, 
 insulating materials, fertilizers, brick, and sand for mortar are some of 
 the avenues open for the economic disposal of slags. 
 
PIG IRON 125 
 
 CHAPTER VI. 
 
 THE MANUFACTURE OF PIG IRON. 
 
 SECTION I. 
 
 SOME INTERESTING HISTORICAL FACTS. 
 
 Early History of Iron: 1 While the sole purpose of this chapter is to 
 describe the manufacture of pig iron as carried on at the present time, one 
 or two of the many interesting topics presented by the historical aspects 
 of the subject will be found pertinent. A word as to the origin of the use 
 of iron will serve to emphasize the process of evolution through which this 
 wonderful industry has passed in attaining its present state of advanced 
 development. When iron was first used, no one knows, for that date belongs 
 to prehistoric times. Archaeological research can only establish that it 
 has been in use by man through a period of about four thousand years. 
 Evidence as to the extent of its use during the first three thousand years 
 of this period is lacking, but it is very probable that the metal was used 
 much more extensively than the few specimens uncovered would indicate. 
 The corrosive properties of iron make it, to the archaeologist, a perishable 
 substance that leaves no trail. ' If the use of iron on this continent were 
 to cease suddenly today, no evidence of its present extensive application 
 would be expected a thousand years hence. Therefore, only occasionally 
 is some implement or ornament found among ancient ruins. There is 
 doubtful evidence of its use by the Egyptians in building the pyramids, 
 about 4000 B. C. As to its use by the ancient Hebrews, by the Assyrians 
 about 1400 B. C., and, more recently, by the Greeks, there can be no doubt. 
 The Greeks were followed by the Romans who became somewhat proficient 
 in its metallurgy. These people, through their numerous and extensive 
 conquests, the success of which they no doubt owed to the use of metals 
 in making their instruments of war, spread the art of extracting and 
 fashioning it throughout Europe. Some knowledge of the metal, however, 
 preceded them, for Caesar, crossing the English Channel, found it in use 
 among the native Britons. During the Roman occupation, the industry 
 grew to one of importance in England. At that time it was obtained by 
 heating a mixture of ore and charcoal, probably in a flat bottom furnace 
 or forge, until there had collected a small body of pasty metal which was then 
 drawn and worked by hammering to make wrought iron. Such, briefly, was 
 the process until 1350, when the iron makers of Central Europe succeeded in 
 producing iron that would melt in the furnace and permit of casting. This 
 result they accomplished in a new type of furnace, built of masonry, which 
 enclosed a shaft or vertical opening in the form of two truncated cones placed 
 
 J See Metallurgy of Iron by Thomas Turner, published by Charles Griffln and Co. 
 Ltd., London. 
 
126 PIG IRON 
 
 end to end, in a crude way, the lines of the modern blast furnace. The lower 
 frustum came to be known as the boshes, the bottom, as the hearth. In this 
 furnace, ore, flux and charcoal were charged in at the top of the furnace, while 
 air, under very low pressure, was blown in at the bottom. This method, was 
 introduced into England about the year 1500 where, in 1619, coke was first used, 
 to be followed, 200 years or more later, by the introduction of hot blast. 
 In America the first iron works was established in Virginia on the James 
 River in 1619, and about 100 years later (1710-1715) the first furnace using 
 blast was built. Thence the industry spread, for the most part, westward. 
 
 Old American Furnaces: The furnaces of a period as recent as one 
 hundred years ago were what would now be called very crude affairs. 
 Portions of some of them are still standing, and one is within a two hour 
 ride of Pittsburgh. They were usually in the form of a truncated pyramid, 
 twenty to thirty feet high, and constructed of stone work which enclosed 
 a circular shaft, some four feet in diameter at the top and about eight feet 
 at the bosh. The hearth was either round or square in cross section. 
 The capacity ranged from one to six tons a day. By the year 1880, this output 
 had been gradually increased to nearly 100 tons per day, with a daily 
 coke consumption of nearly 300 tons. With all the basic principles in use 
 for so long a time, it is remarkable that so little progress was made. 
 About 1880, for reasons, which would be too lengthy to explain here, very 
 rapid advancement was made, so that now there are furnaces whose daily 
 output of pig iron exceeds 600 tons with a fuel consumption of less than 
 2000 pounds of coke per ton of iron produced. Attention has been called 
 to these facts here, because it is well to remember in beginning the study 
 of the modern blast furnace, that the present method for the extraction 
 of iron from its ores represents a pyrochemical process just attaining its 
 highest state of development. 
 
 The Importance of Iron: This topic needs no comment here. Pig 
 iron, besides being used directly in the form of castings, is the intermediate 
 from which all ferrous products are derived. Its importance is emphasized 
 by the reports of the yearly productions. 
 
 SECTION II. 
 COMPOSITION AND CONSTITUTION OF PIG IRON. 
 
 Constitution of Pig Iron: In the solid form, pig iron represents a 
 very complex mixture made up of uncombined elements, chemical compounds 
 and alloys. The amounts and relations of these constituents may vary 
 with conditions, so that the complexity of the mixture does not depend 
 wholly upon the number of elements present nor upon their amounts. Initial 
 temperature and rate of cooling are two of the most important factors 
 affecting the properties of pig iron. These matters are of great importance 
 when the iron is to be used for castings, and to understand them fully requires 
 a very extended study of the subject. This chapter has to do mainly with 
 pig iron as an intermediate product in the making of steel, so it will be 
 most profitable to discuss only the subject of its composition very briefly. 
 
ELEMENTS IN 127 
 
 Chemical Elements in Pig Iron: In addition to iron, the elements 
 commonly occurring in pig iron are carbon, silicon, manganese, sulphur 
 and phosphorus. Of these elements iron will constitute 91 to 94%, carbon 
 3.0 to 4.0%, silicon .50 to 3.00%, sulphur less than .065%, and phosphorus 
 .040% to 2.00% of the whole. 
 
 Carbon occurs in pig iron in two forms, called graphitic carbon and 
 combined carbon. Graphitic carbon is practically pure carbon, existing 
 in the iron in the form of tiny flakes which are distributed throughout the 
 mass. It forms in the pig iron during the process of cooling, because the 
 absorbing power of iron for carbon decreases as its temperature falls. 
 Carbon in this form gives to pig iron the grayish black appearance so often 
 seen. But in cooling, some of the carbon continues in combination with 
 the iron as a definite compound, Fe 3 C, 93.33% Fe and 6.67% C. Both 
 forms of carbon produce marked effects upon the properties of the iron. The 
 tendency of the graphitic is to weaken, while the combined carbon, up to 
 the limit of about .90%, strengthens it. In Metallography the compound 
 FesC is called Cementite, and to the uncombined iron is given the term 
 Ferrite. In cooling these two substances conduct themselves in a peculiar 
 way toward each other. In passing a certain temperature (about 700 C.) 
 they arrange themselves in layers in the definite amounts of approximately 
 seven parts ferrite to one part cementite. The resultant stratified segregate 
 will, therefore, contain approximately .85% C. Under the microscope these 
 stratifications present the appearance of mother of pearl, whence it is 
 named Pearlite. Pearlite is the strongest constituent of cast iron. When 
 heated, iron absorbs carbon, and from the fusion point this absorption 
 becomes very rapid. The limit, called the saturation point, beyond which 
 it will not absorb any more, varys with the temperature, and is also affected 
 by the amount of silicon present, a rise in the percentage of silicon causing 
 a corresponding decline in the carbon content. Silicon tends, also, to 
 decrease the combined carbon, and increase the graphitic. Manganese and 
 Chromium have the opposite effect. Rapid cooling tends to prevent the 
 formation of pearlite and graphite. In general, the more rapid the cooling 
 the less the graphitic carbon and the greater the combined carbon content 
 will be. 
 
 Silicon : In small quantities silicon has little direct effect on pig iron, 
 but an increase above 4% makes the iron very brittle, hence foundry iron 
 will seldom contain more than 3%. In iron for basic open hearth use the 
 percentage of silicon should not be higher than 1.25, as a high silicon content 
 tends to flux away the lining of basic furnaces very rapidly. For the acid 
 Bessemer process iron containing about 1.25% silicon is desirable, but the 
 content may vary from 1.00% to 1.50%. The oxidation of the silicon 
 in the Bessemer process produces a large quantity of heat, so that iron 
 containing a high percentage of this element is usually referred to as hot 
 iron. The amount reduced in the blast furnace is variable and depends on 
 conditions of slag and temperature. Its effect on the carbon has just been 
 
128 PIG IRON 
 
 noted. Since it tends to throw the carbon out of solution, silicon is 
 used to regulate the depth of chill in chilled castings. A content of one 
 per cent, silicon in ordinary low sulphur iron renders it difficult to obtain 
 a chill. Below this percentage the chilling properties of the iron are, roughly 
 stated, in inverse ratio to the amount of silicon present. 1 Silicon also 
 prevents blow holes, and tends to decrease the shrinkage in white irons. 
 
 Manganese alloys with iron in all proportions. An alloy containing 
 10 to 25% manganese is called spiegel. Alloys containing 40 to 80% man- 
 ganese are called ferro-manganese. Up to one percent manganese tends to 
 strengthen pig iron. It decreases the bad effects of sulphur, with which it 
 combines, replacing iron. Its presence opposes that of sulphur, so that, with 
 uniform raw materials, furnace conditions that give a high percentage of 
 manganese tend to decrease the percentage of sulphur. Hence, in reason- 
 able amounts of about one per cent, it is desirable, especially for basic open 
 hearth use, where it also aids in the elimination of sulphur. In Bessemer 
 practice iron with a manganese content of about .50% is desirable. The ele- 
 ment is oxidized, and unites with silica to form a slag that fuses at a com- 
 paratively low temperature and is very fluid, so that iron containing a high- 
 er percentage than that indicated by the latter figure gives rise to a condi- 
 tion in blowing known as a "sloppy" heat. As to whether manganese has 
 a good or a bad effect on cast iron,, there is much difference of opinion, 
 some considering it almost as a cure for all troubles and others 
 condemning it as a source of much trouble, especially in chilled castings. 
 While it tends to hold carbon in solution, chill produced by increasing the 
 manganese content alone is soft and tends to spall^. In moderate amounts 
 it is said to prevent cracking of the surface and also spalling to some 
 extent, especially in chilled rolls. Nearly 75% of the total amount of 
 manganese charged into a blast furnace is obtained with the metal. 
 
 Sulphur in pig iron is generally supposed to be injurious, though 
 recently the statement that the inferior qualities exhibited by high sulphur 
 iron is due entirely to its presence has been questioned. Nevertheless, as 
 sulphur in steel is considered undesirable and as the blast furnace affords the 
 only positive means of reducing it, pig iron containing less than .05% is 
 desirable for making steel by all the fusion processes. Sulphur with iron 
 forms iron sulphide, which is soluble in the metal and has a melting point 
 that is lower than the other constituents of the iron. According to some 
 authorities 1 , this sulphide in iron used for castings has a three fold influence. 
 First, it tends to hold the carbon in combined condition, hence can be 
 used to increase the depth of chill in chilled castings; second, its low 
 melting point causes it to segregate as the iron solidifies, thereby causing 
 the condition in castings known as bleeding; third, it increases the shrinkage 
 of the iron to a marked degree, thus increasing the difficulty of making 
 accurate castings and increasing the tendency to cracks which are a result 
 
 iSee The Principles, Operation and Products of the Blast Furnace, by J. E. 
 Johnson, Jr. Published by McGraw-Hill Book Company Inc., New York. 
 
GRADES OF 
 
 129 
 
 of the high shrinkage. The chill imparted by sulphur is a very hard one, 
 but is very brittle and somewhat unreliable. 
 
 Phosphorus is the only element entering the blast furnace over which 
 the skill of the furnaceman has absolutely no control. Its compounds are 
 completely reduced, so that all the phosphorus in the raw materials is found 
 in the metal. Therefore, its content must be regulated by proper selection 
 of raw materials. High phosphorus causes a slight brittleness in pig iron, 
 and has a marked effect upon the total carbon. Ferro-phosphorus containing 
 about 15% phosphorus is carbonless. Lesser amounts permit a pro- 
 portionate increase of carbon, so that the total carbon in an iron containing 
 .2% phosphorus may be as high as 4%. In this respect its action is not 
 selective, since the ratio of combined to graphitic carbon is not affected. 
 Phosphorus is known to form a compound, FesP, with iron, but it is able 
 apparently to combine with it in several proportions. Ferro-phosphorus 
 containing as much as 25% phosphorus is now manufactured? In iron for 
 casting, phosphorus exercises a beneficient effect. It tends to eliminate 
 blow holes, decreases shrinkage, and increases the fluidity. Above .5% 
 it begins to weaken iron, so the amount used will be governed by the use 
 to which the casting is to be applied. 
 
 Grading Pig Iron: Pig Iron is graded by chemical analysis. There 
 are several systems employed, many of which are somewhat elaborate. 
 The following table, which includes other important blast furnace products 
 as well as ordinary pig iron, shows one of the simplest methods of classi- 
 fication: 
 
 Table 25. The Metallic Products of the Blast Furnace. 
 
 RANGE IN PERCENT. OF 
 
 GRADE 
 
 Silicon 
 
 Sulphur 
 
 Phosphorus 
 
 Manganese 
 
 Total Carbon 
 
 No. 1 Foundry 
 No. 2 Foundry 
 No. 3 Foundry 
 Malleable Casting. . . 
 Forge 
 
 2.5 to 3.0 
 2.0 to 2.5 
 1.5 to 2.0 
 .75 to 1.5 
 About 1.50 
 
 Under .036 
 .045 
 .060 
 .050 
 1.00 
 
 .25 to 1.00 
 .25 to 1.00 
 .25 to 1.00 
 .2 
 1.0 
 
 Under 1.00 
 1.00 
 
 ; i.oo 
 
 "' 1.00 
 1.00 
 
 3.504.25 
 3.504.25 
 3.504.25 
 3.504.25 
 3.50 4.25 
 
 Acid Bessemer 
 Basic Bessemer 
 Low Phos. Acid Iron 
 Basic . 
 
 1.00 to 1.50 
 Under 1.00 
 2.00 
 1.25 
 
 .050 
 .050 
 .030 
 .050 
 
 0. 1 or less 
 2.00 to 3.0 
 .030 
 .100 to 1.00 
 
 About .50 
 Under .50 
 1.00 
 1.00 to 2.50 
 
 3.50 4.25 
 3.50 4.25 
 3.504.25 
 3.50 4.25 
 
 Spiegel 
 
 About 1 00 
 
 .050 
 
 .150 
 
 18.0 22.0 
 
 5.0 6.0 
 
 Ferro-Manganese, . . . 
 Ferro-Silicon 
 Silico-Spiegel 
 
 .50 to 1.00 
 8.0 to 15.00 
 8.0 to 15.00 
 
 .030 
 .070 
 .010 
 
 .10 to .30 
 .10 to .50 
 .15 
 
 78.0-82.0 
 15.00-20.00 
 
 5.0 7.0 
 1.002.00 
 
130 BLAST FURNACE 
 
 SECTION III. 
 
 A BRIEF OUTLINE OF THE PROCESS AND EQUIPMENT FOR 
 THE MANUFACTURE OF PIG IRON. 
 
 Trend of Modern Improvements : With the preceding brief summary 
 of the history, importance, and composition of pig iron in mind, the process 
 by which it is manufactured furnishes a theme of great interest. Apropos 
 of this idea, however, it is to be observed that a description of the modern 
 methods of manufacture is rendered difficult both by the complexity of the 
 details of the process and by its recent rapid development. As already 
 pointed out, the fundamental principles have remained unchanged since 
 the founding of the process, because experience has demonstrated that this 
 process is the most practical. All improvements, then, have been made 
 with the aim of increasing the production and at the same time decreasing 
 the cost. These objects have been attained to a degree almost approaching 
 perfection by the use of materials of greatest purity, selected through 
 chemical control, by increasing the size of furnaces, by economies in fuel 
 consumption, and by improved methods of handling the materials. The 
 result is that the small plants of 100 years ago have been succeeded by 
 complex and gigantic affairs. As the greatest changes have been brought 
 about since 1880, a comparatively recent date, the blast furnace plant is 
 just approaching the uniformity of perfection. Furthermore, since the 
 improvements have been contributed by a great number of men, it is not 
 to be wondered at that an inspection of the industry will reveal not only 
 different stages of development but also many different methods of attaining 
 the same end. The aims and fundamental principles being the same, 
 however, the numerous plants, while differing greatly in detail, will present 
 certain similarities in their gross features which may profitably be reviewed 
 before proceeding with the detailed description. 
 
 Essentials of the Process: Essentially, the present process for the 
 extraction of iron from its ores consists in charging a mixture of ore, fuel, 
 and flux in proper proportions through a specially constructed opening in 
 the top of a tall cylindrically shaped furnace called a blast furnace, while 
 heated air is simultaneously blown in near the bottom through openings, 
 called tuyeres, the nitrogen of the air together with the products of com- 
 bustion and reduction passing upward and escaping through openings at 
 the top. These parts of the process, being almost continuous ones, are 
 accompanied by the periodic removal of a part of the impurities in the 
 form of slag at an opening between the tuyeres and the bottom, and by a 
 like removal of metal through a larger opening at the bottom. In order 
 to carry out these operations on the large scale previously mentioned, it 
 is evident that extensive equipment is required. 
 
 Essential Equipment: The central feature in this equipment is the 
 furnace, which is provided with apparatus for hoisting the materials to the 
 top and with ladles for containing slag and molten metal, to which is some- 
 
EQUIPMENT 131 
 
 times added casting beds or pig machines for casting the metal into "pigs" 
 of convenient size, and slag granulating pits. Next in importance follows 
 the blowing engines for producing the blast, then the stoves for heating 
 it. Of great importance is the pumping station, the function of which is 
 to furnish the great quantities of water needed for steam, for cooling, etc. 
 As the gases that escape from the top of the furnace are combustible, 
 apparatus for their most efficient disposal is desirable. They are used to 
 heat the stoves and to generate power either by burning them under boilers 
 or in gas engines, in which case they must be cleaned of the large quantities 
 of flue dust which they carry out of the furnace. As the moisture in the 
 air affects the efficiency of the furnace, some modern plants will be provided 
 with apparatus for drying the blast. Referring again to the solid materials 
 of the charge, modern equipment requires a stock house, topped by bins, 
 in which the ore, fuel and flux may be temporarily stored and conveniently 
 removed for weighing or measuring before delivering it to the hoisting 
 device. Adjacent to the bins will be located the stock yard containing the 
 ore pile, which is spanned by the ore bridges. A car dumper, advantage- 
 ously situated, will complete this part of the equipment. Finally, the various 
 parts of the plant will be made accessible by a system of railways for trans- 
 porting the materials. 
 
 SECTION IV. 
 
 CONSTRUCTION OF THE BLAST FURNACE PROPER. 
 
 The Gross Features of the Furnace Proper: The modern blast 
 furnace is a tall circular structure, 90 to 100 feet high, built of fire brick, 
 reinforced externally by a close fitting steel shell and encasing internally 
 a circular space of varying diameters. This space is divided into three 
 main parts. The bottom section, called the hearth or crucible, is cylin- 
 drical in form and some 10 to 12 feet deep in the larger furnaces. The 
 second section, having an altitude of some 12 or 13 feet, is called the bosh. 
 It is in the form of a frustum of a cone, which, in an inverted position, tops 
 the crucible with its smaller base. Setting above the bosh in an upright 
 position with an altitude of about 70 feet, is the stack. Formerly its outline 
 was also that of a frustum of a cone, but recent studies of furnace lines 
 indicate that the slanting lines of the cone should be changed to the vertical 
 for a distance of 4 or 5 feet above the bosh, and for about 10 feet from 
 the lower bell at the top. The whole is now capped by the furnace top, 
 which completes the list of the gross features of the furnace proper. 
 
 The Foundation : Before proceeding with the details of the parts 
 noted above, the foundation should be considered. In view of the immense 
 weight which it is required to support, this part of the furnace is of great 
 importance, because any extensive settling of the furnace after it is in 
 operation would result in serious troubles and probably put it out of com- 
 mission. The depth of the foundation will vary with the conditions of the 
 
132 BLAST FURNACE 
 
 rock materials on which it is to stand. If these be sand .or clay, it may 
 be necessary to drive piling for a depth of many feet, and upon this begin 
 the foundation. On the other hand, if solid and firm rock underlies the 
 location for the furnace, an excavation to this rock is all that is required. 
 A proper bed having been found or otherwise provided, the foundation is 
 started and built up several feet with concrete, which-extends some distance 
 outward beyond the floor of the furnace. The remainder of the foundation 
 is then made up of common brick of good quality and strength, except the 
 space directly beneath the hearth and walls of the furnace, where firebricks 
 are used. 
 
 The Hearth or Crucible is the portion of the furnace which serves 
 as a receptacle for the molten metal and slag. It is constructed of fire 
 brick of the best quality; its wall is usually sixty inches or more in thick- 
 ness; and it may be protected in places with water cooled plates, if the fur- 
 nace is of recent construction. At the bottom the walls of the hearth are 
 usually stepped out into the interior of the hearth for four or five courses 
 of brick. This construction gives the bottom surface a slight basin shape, 
 and tends to hold the bottom brick in place. The hearth varies in diameter 
 and depth with the size and capacity of the furnace. In the larger ones, it is 
 about eighteen feet in diameter and eleven feet deep. Externally, it is 
 reinforced by a heavy metal jacket made of steel plates that are riveted 
 together, or of iron castings in segments that are jointed and bolted together. 
 Jackets are always cooled, those of cast iron by internal circulating 
 systems, and those of steel by external sprays. The upper diameter of this 
 jacket is smaller than the diameter at the base, so that the jacket will 
 better hold the walls of the hearth in place by offering resistance in oppo- 
 sition to the buoyant forces of the bath and slag. 
 
 The Bottom of the crucible is built of fire brick, sometimes in the 
 form of large blocks, which are laid on end with closely fitting joints in 
 order to prevent intrusion of metal. Bottoms vary in thickness from about 
 six feet in the smaller furnaces to about twelve feet in the large ones. The 
 bricks are almost entirely replaced in time by metal, which, collecting in 
 ,a solid mass, often weighing many tons, is known as the salamander. 
 
 Tapping Hole : Situated at some convenient point in the circumference 
 of the hearth and just above the top course of stepped-in brick is the tapping 
 hole or iron notch. If the bricks are not stepped-in, the opening will be 
 at the bottom. It may be a square opening in the brick about 8x8 inches or 
 an oblong or rectangular one 6x8 inches on the inside. The outside dimen- 
 sions may be somewhat larger to permit of easily inserting the tapping 
 tools. Proper provision is made for the protection of the hearth jacket 
 at this point. During the tapping of iron, the metal structures directly 
 above the tapping hole are protected with a splasher. 
 
CONSTRUCTION 
 
 133 
 
 01/5TLE 
 
 FIG. 20. t Vertical Section of a modern Blast Furnace. 
 
134 BLAST FURNACE 
 
 Cinder Notches: There is usually but one cinder notch. This 
 opening may be placed at any convenient point in the circumference of the 
 hearth at a sufficient height above the tapping hole to permit the collection 
 of the desired amount of iron between tappings. In the larger furnaces it 
 is about six feet from the floor of the hearth and four to five feet above the 
 tapping hole, being generally placed 45 or 90 from this opening. Unlike 
 the tapping hole, this opening is water cooled to protect the brick from the 
 fluxing action of the slag. Hence, the opening in the brick work is larger, 
 being about one foot in diameter inside and increasing to about two feet on the 
 outside. In this circular cone-shaped hole in the brick the cooling devices 
 are placed. These are castings, usually made of copper, and consist of acin=> 
 der cooler, an intermediate, or monkey, cooler, and a monkey.The cinder 
 cooler is in the form of a hollow frustum of a cone. It is about two inches 
 thick, and, between, its walls, proper provision is made for the circulation 
 of water. It is made to fit the hole in the brick work and is tamped securely 
 in place with fire clay. The opening in this cooler is then reduced by 
 inserting into its inner end the close-fitting intermediate cooler, which is 
 constructed like the cinder cooler, but smaller and shorter. Finally, the 
 still smaller monkey, through which water circulates also, is inserted, 
 reducing the opening to about two inches. A short iron rod, called a bott, 
 tapered to fit the monkey and attached to a long steel rod which serves 
 as a handle, is used to close the opening. The sizes of the three coolers 
 are regulated so that the large diameters of the monkey and intermediate 
 cooler fit the smaller diameters of the intermediate and cinder coolers, 
 respectively. Thus, the monkey, when in position, is wholly within the 
 furnace. In each of these castings and within their topmost quadrant, 
 when in position for service, are provided two threaded holes into which 
 the pipes for ingress and egress of the cooling water may be inserted. 
 
 Tuyeres: The tuyeres, from ten to sixteen in number, are distributed 
 symmetrically about the upper circumference of the hearth just below the 
 boshes. Their function is to provide passages for the blast. They also 
 determine the height to which the slag in the furnace may rise. In the 
 large furnaces, this height is about three feet. Fitted into the opening in the 
 brick, flush with the wall, both internally and externally, is the tuyere 
 cooler. It is similar to the cinder cooler, and set tight with fire clay. The 
 tuyere itself, of copper, presents an internal diameter of from four to seven 
 inches, while its external diameter is such as to permit it to fit snugly into the 
 smaller end, or nose, of the cooler and project several inches into the furnace. 
 Like the cooler, the tuyere is water cooled and is tapped at two places in 
 the top quadrant for the insertion of water pipes through which a copious 
 stream of water must be kept flowing to avoid burning it. 
 
 Tuyere Connections : With one end fitting closely against the tuyere, 
 is a horizontal cast iron pipe, about five feet long, called the blow pipe. 
 sometimes the ' 'belly" pipe. Through it the hot blast is delivered to the 
 
CONSTRUCTION 135 
 
 tuyere from the tuyere stock, to which the names leg pipe, boot leg and 
 pen stock are also often applied. The blow pipe may be slightly larger in 
 diameter at one end than at the other, and both ends are turned to fit into 
 slight sockets in the tuyere and the tuyere stock. It is held in place with 
 the smaller end fitting into the tuyere by pressure from the tuyere stock, 
 which is provided with a spiral spring and connecting rod, attached to the 
 hearth jacket, for this purpose. The tuyere stock curves upward immedi- 
 ately on leaving the blow pipe, and a hole in a lug on its under part gives 
 an opening through which the connecting rod passes through the coiled 
 spring to the hearth jacket where it is anchored. The heavy spiral spring, 
 which provides pressure and at the same time allows motion due to expan- 
 sion and contraction of the tuyere stock and blow pipe with changes in 
 blast temperatures, is placed between this lug and a large brass nut on the 
 other end of the connecting rod. The nut is made of brass so it will not 
 rust and be difficult to operate. In the outer part of the curve in the tuyere 
 stock, and in the center line of the blow pipe and tuyere is a small opening, 
 closed by the tuyere cap or "wicket," through which a small rod may 
 be inserted to clean out the tuyere without removing the blow pipe. 
 The wicket must be of such a form that it may be opened readily at 
 any time and still be practically gas tight. To meet these conditions 
 the wicket is constructed on the principle of the' ball and socket joint. 
 The ball is attached to the end of the short arm of a right angle lever and 
 is held tightly in the socket by a ball weight attached to the long arm. 
 A smaller opening, called the peep hole, in the tuyere cap is covered with 
 glass, which permits inspection of that portion of the interior of the furnace 
 directly in front of the tuyeres. Extending upward, the tuyere stock 
 connects with the nozzle of the goose neck, to which it is clamped by means 
 of lugs and keys that fit into seats of hanging bars. The goose neck then 
 turns at right angles and extends horizontally to the neck of the bustle 
 pipe to which it is fitted and securely fastened by flanges and bolts; or the 
 tuyere stock may lead at an angle to the horizontal directly to the lower 
 part of the bustle pipe. The bustle pipe is the large pipe, about four 
 feet in outside diameter, which encircles the furnace and distributes the 
 hot blast to the tuyeres. All these pipes, down to the blow pipe, are 
 lined with fire brick. The bustle pipe is fed by the hot blast main which 
 terminates at the stoves. It is of about the same size as the bustle pipe 
 and lined with nine to twelve inches of fire brick. 
 
 Boshes: As previously described the bosh is that part of the furnace, 
 just over the hearth, where the greatest diameter is attained. No shell 
 cr jacket covers the exterior of the bosh. A standard bosh is constructed 
 in the following manner: Starting at the top of the hearth, the brick 
 work, 30 inches in thickness, is stepped outward, externally, nearly six 
 inches for, each twelve inches of vertical rise. Each step-out is supported 
 by means of a heavy steel band, or a pair of bands, called bosh bands. 
 
136 BLAST FURNACE 
 
 Inserted in the walls of the boshes, through cast iron ''boxes," placed in 
 the brick spaces between pairs of bosh bands, will be found cooling plates, 
 called the bosh plates, in horizontal rows about two feet apart, measuring 
 vertically. The plates in each row will be about four or five inches apart, 
 and the plates in the different rows will be staggered vertically, breaking 
 joints like brick work. This construction adds to the cooling efficiency of 
 the plates. There are several different makes of bosh plates, but the more 
 common ones will be somewhat wedge-shaped, with a flat bottom and oval 
 top, and about four inches thick at the point of their greatest altitude. They 
 are hollow and have inserted in them, usually at opposite corners, two 
 pipes through which water flows continuously. These plates are necessary 
 to help protect the brick work, for, being just above the tuyeres in the 
 zone of fusion, the bricks here receive the highest heat of the furnace. 
 Formerly the plates extended almost through the wall in new work, 
 usually to within one course of brick, but it was found that this course of 
 brick is soon cut away after the furnace is blown in, so now the plates 
 extend entirely through the wall from the first. 
 
 Mantle: At the upper limits of the bosh is found the mantle, con- 
 forming to the shape of the furnace at that point and totally encircling it. 
 The mantle is made up of heavy steel plates and angles, upon which rests 
 the weight of the stack. It is supported by a series of cast iron pillars 
 or fabricated steel structures, which rest on foundations supported by the 
 main furnace foundation. This construction allows the entire bosh and 
 hearth to be removed without disturbing the rest of the furnace. 
 
 Shaft, or Stack, and In=Walls: The shaft comprises all that part 
 of the furnace which is located above the boshes. The wall of this shaft is 
 usually, in an imaginary way, divided into three almost equal parts, called 
 the upper, middle and lower inwalls. Up to this point, the construction 
 for all furnaces is fairly uniform as to general features. But as to the 
 inwalls three types are employed, namely, the thick, the intermediate, 
 and the thin wall. The construction necessarily differs for these different 
 types. Therefore, each type is best described separately. 
 
 Thick Wall Type: The inwalls of this type are about five feet thick, 
 and are inclosed in a heavy riveted steel shell about one-half inch thick. 
 The shell is usually made oversize to provide a small space between it and 
 the brick work, in which space is tamped lightly a packing of loam and 
 granulated slag, to allow for the expansion and contraction of the inwalls. 
 Thick walled furnaces are seldom water cooled above the bosh, and their 
 walls furnish the sole support for the top. 
 
 The Furnace Lines and Bosh Angles of furnaces of the thick wall 
 type differ somewhat. In modern blast furnace construction the lines of 
 
CONSTRUCTION 137 
 
 the furnace, by which is meant the lines formed by the inner edges of a 
 vertical section through the center, with their enclosed angles, are con- 
 sidered of great importance. In the old furnaces, the lines of the inwalle 
 were straight, and the boshes somewhat flat with corresponding sharp 
 angles. But with the fine ores from the Lake Superior district, experience 
 has taught that much better practice is obtained with more nearly vertical 
 lines. So, in the latest type of furnace the lower inwall will rise verti- 
 cally for several feet, the boshes will be steep, and the upper inwall will 
 drop vertically for a distance of about ten feet from, the stock line. Bosh 
 angles, that is, the angles at the top of the bosh, which its wall forms 
 with a horizontal from the center of the furnace, are now being increased 
 from about 75 to 80. These steeper boshes are proving to be an 
 important improvement. 
 
 Intermediate, or Setni=Thin, Wall Type: In the intermediate type, 
 the walls are about three feet thick, and to protect the brick as much as possible, 
 cooling plates, similar to bosh plates, are inserted in the lower inwall. They 
 may extend for a distance of from twenty to forty feet above the bosh. 
 The top in this type is sometimes supported by columns of fabricated steel, 
 but in the majority of cases it is supported by the walls as in the thick 
 walled type. The inwalls are surrounded by a steel jacket as in the thicker 
 type, the only difference being in the necessary openings for the coolers. 
 
 Thin Walled Type: In this type the inwalls are from nine to eighteen 
 inches thick, the top is always supported by structural work, and the shell must 
 be cooled throughout its entire length. This cooling may be done in three 
 ways. One method consists in spraying the jacket with water, conducted 
 through suitably arranged pipes and enclosed by a light "splash jacket" which 
 conforms to the size and shape of the stack. In the second method the 
 shell is encircled by a series of deep and narrow horizontal troughs through 
 which water is kept flowing from the top to the bottom of the furnace, 
 overflowing from each trough to the next succeeding lower one. Each of 
 three or four of these troughs drain separately to a common point where 
 the temperature of the water can be noted. In the third type, the entire 
 outer surface of the stack is kept covered with water by means of a spiral 
 trough which, slightly separated from it, encircles the stack from top to 
 bottom. This trough is kept full of water by a series of feed lines that enter 
 it at various points in the spiral. 
 
 Furnace Linings: The brick-work which forms the hearth, bosh and 
 inwalls of a furnace are referred to as its linings. All the brick used in 
 the construction of these parts are made of fire clay, and are of three kinds, 
 known as hearth and bosh brick, inwall brick and top brick, each of 
 which is made of such materials and in such a way as to render it best adapted 
 to the conditions it is to be subjected to in service. The hearth and bosh 
 brick are required to resist a very high temperature and the action of flux 
 
138 
 
 BLAST FURNACE 
 
 and slag; inwall brick must be able to withstand abrasion at a moderately 
 high temperature; and top brick, always at a comparatively low temper- 
 ature, must resist the impact and abrasive forces of the charges as they 
 are dropped into the furnace. These different qualities in the different 
 bricks are obtained by varying the method of manufacture especially with 
 respect to composition, degree of grinding and temperature of burning. 
 As to the former factor, three classes of materials, or clays, are available. 
 These are flint clay, plastic clay and calcined clays. Mixtures employed 
 by different manufacturers are by no means uniform, but the following 
 may be taken as fairly representative of good practice. 
 
 Table 26. Showing Data Relative to Fire Brick for Use in 
 Blast Furnaces. 
 
 PROPORTION OF 
 
 Kind of Brick 
 
 Flint Clay 
 
 Calcined Clay 
 
 Plastic Clay 
 
 Manner of 
 Grinding 
 
 Final 
 Temperature 
 of Burning 
 
 Hearth and 
 Bosh Brick 
 
 50 to 65% 
 
 20 to 35% 
 
 14 to 16% 
 
 Coarse 
 
 1350 C 
 
 Inwall Brick 
 
 40% 
 
 30 to 40% 
 
 20 to 30% 
 
 Medium 
 
 to 
 
 Top Brick. . 
 
 to 30% 
 
 30 to 50% 
 
 40 to 50 % 
 
 Fine 
 
 1450 C 
 
 All the materials, irrespective of the kind of brick, should be and are 
 of the best quality obtainable, and the brick are carefully inspected before 
 being put in place in the furnace. The three kirfds of brick are distinctly 
 marked by the manufacturer, so that the danger of wrongly placing a brick, 
 a top brick in the hearth, for example, may be avoided. Great care is 
 exercised with respect to brick, because the life of the furnace depends in 
 a large measure upon the lining, and the item of cost for brick is not a small 
 one. In the construction of one of the large modern furnaces, close to the 
 equivalent of 800,000 nine inch brick are required, and the average con- 
 sumption of brick is a little more than two brick for each and every ton of 
 pig iron produced. Fire bricks are always laid in a thin slurry composed 
 of fire clay and water. The slurry is applied by pouring it on the top of 
 each course with a dipper, and is followed immediately by the next course of 
 bricks, which are hammered into place to squeeze out all the fire clay except 
 that required to compensate the inequalities of the brick. 
 
CONSTRUCTION 139 
 
 Water Trough: Encircling the furnace bosh, inside of and above the 
 bustle pipe, will be found one or more water troughs into which the water 
 supplying the numerous cooling plates is discharged in visible streams, 
 thus providing means of determining when a plate is burned out. The 
 water, flowing from these troughs into a well, may be pumped back through 
 the cooling system, and thus be used over and over. 
 
 Tops: At the present time furnace tops are somewhat complicated 
 affairs. In olden times the tops of furnaces were left open, the escaping 
 gases being allowed to burn in the air. In the year 1814, these waste gases 
 began to be employed in France for the purpose of burning bricks and heating 
 small furnaces. The first attempt to heat the blast by utilizing these gases 
 was made in 1834, and consisted in laying, across the tunnel head, pipes 
 through which the air blast passed in going to the tuyeres. It was not 
 till 1845 that a plan was evolved by which the gases could be used to heat 
 the stoves. To effect this purpose, changes in top construction were neces- 
 sary. At first the gases were merely drawn off by the chimney draft of 
 the stoves through openings below the stock line. The arrangement, 
 known as the bell=and=hopper, or cup=and=cone, was not put into use 
 till 1850. It consisted in closing the top of the furnace by means of a large 
 circular hopper, the smaller opening of which was closed by the bell 
 that could be lowered and raised at will. With the bell raised against the 
 hopper, the materials were dumped into the hopper; then the bell was lowered, 
 and the charge dropped into the furnace. As large quantities of gas escaped 
 with each lowering of the bell, this device was improved by the double- 
 bell=and=hopper, which is of comparatively recent origin. Essentially, 
 this improvement consisted in placing a second but smaller bell-and-hopper 
 above the first, and providing a gas tight space of large size between the 
 two. The raw material, upon being hoisted to the top, is first dropped or 
 dumped into the upper hopper, whence it may fall into the larger hopper 
 below when the small bell is lowered. The small bell being raised against 
 ttfe upper hopper, the large bell is lowered, and the charge falls into the 
 furnace without the escape of gas. The bells are made of cast steel, in one 
 piece, and of such a slope, 45 to 55, as to permit the charge to slide off 
 readily. They are usually supported from their top centers by means of a 
 rod and a sleeve, each attached to a counterbalanced lever operated by 
 means of a steam or air cylinder or an electric motor, controlled from the 
 ground. The large bell is attached to the rod and the small bell to 
 the sleeve. The hoppers, of cast steel or iron, will generally be made up in 
 sections, which are securely bolted together. They are provided with 
 a removable, but gas-tight, flange and extension ring, also in sections, 
 which permit the bell to be readily removed in case a change is 
 necessary. The details of this construction differ somewhat to conform 
 to the introduction of improvements, with the type of hoist, and with the 
 ideas of the different builders. 
 
140 BLAST FURNACE 
 
 Stock Distributor: One of the alleged improvements in the bell- 
 and-hopper device is that of the stock distributor. There are several types 
 of these distributors employed, a description of which would not be profit- 
 able here. However, the object aimed at by such devices may be explained 
 thus: It is apparent that, in a mechanically filled furnace, when the raw 
 materials are dropped into the receiving bell, the larger lumps of ore and 
 stone will have a tendency to roll and thus collect either around the edges 
 or to one side or the other. The same things will also happen upon dropping 
 the charge into the furnace. This tendency results in more or less open 
 and continuous channels being formed through the materials and extending 
 from the top towards the bottom of the stack. These channels offer less 
 resistance to the passage of the blast than the remainder of the materials, 
 or the fuels, with the result that a disproportinate quantity of gas passes 
 through them. This condition, called channelling, results in higher tem- 
 peratures throughout these passages, with the consequent cutting away of 
 the walls where these channels come in contact with them. It is to over- 
 come this defect that the various devices formerly mentioned have been 
 designed. 
 
 Hoisting Appliances: The old time method of charging by hand 
 having been entirely superseded by automatic mechanical charging, there 
 are now in use two types of these devices, namely, the skip hoist and the 
 bucket hoist. In both cases there is an incline, a fabricated steel structure, 
 extending from the top of the furnace to or below the bottom of the stock 
 house; and over the tracks of this incline the materials charged into the 
 furnace must pass. In the skip hoist, the conveying vessel is a small open- 
 ended steel car, called a skip, that automatically dumps the materials upon 
 the little bell-and-hopper. Skip hoists are generally provided with double 
 tracks, so that while a loaded skip is passing up the incline an empty one 
 is descending. In the bucket hoist the solid materials are raised in a 
 bucket, suspended from a truck or carriage, that drops the charge into the 
 space above the large bell direct. When in position for dropping the charge, 
 the bucket, being itself provided with a small bell at the bottom, takes the 
 place of the little bell and hopper. During the time the bucket is filling 
 at the stock house, the opening left in the top is closed with a 
 special gas seal. 
 
 Top Openings: The smallest opening in the top of a furnace is the 
 try=hole. In operating a furnace it is necessary to be able to determine 
 the position of the stock line. This is done by means of the stock indicator, 
 which is a rod of steel passing through and fitting the try-hole loosely so 
 that one end rests upon the stock, while the other is attached to a small 
 steel cable that leads to the stock house or the cast house below. Some 
 stock indicators are automatic and self-recording. For the escape of the 
 gases, from two to four large openings, called offtakes, are provided. They 
 
CONSTRUCTION HI 
 
 pierce the furnace top just beneath the large bell. From these openings, 
 about four feet in diameter, lead fire-brick lined pipes which converge into 
 one large pipe called the downcomer or downtake. Into openings, either in 
 the offtake pipes or in special openings in the top, are inserted the explosion 
 doors. These doors, located usually in the ends of upright pipes arranged 
 so as to prevent ejection of material from the furnace, are really valves 
 which are adjusted, either by weights or by a mechanical means, to open 
 at a certain pressure. They are designed to relieve pressure and so prevent 
 possible injury to the top by slips in the furnace. The bleeder is a tall 
 vertical pipe, usually inserted on the higher surface of the offtake pipe 
 leading to the downcomer, to allow surplus gas to escape. It is closed 
 with a valve on the top, which opens automatically, and may also be 
 opened from the ground. Bleeders are usually lined with one course of 
 fire brick. 
 
 General Consideration for Top Construction: As previously 
 pointed out, there are many types of top, and the description above is 
 intended to give a general idea of the essential parts and their uses only. 
 The chief endeavor in top construction is to perfect the distribution of the 
 stock entering the furnace stack, and either eliminate or compensate for 
 as many irregularities as possible. However, in attaining this end, sim- 
 plicity must be considered, as any great amount of mechanism on the top 
 of a furnace is objectionable. It is important to prevent large material 
 from being thrown out of the furnace in case of slips, and as little dust as 
 possible should be carried out by the gases. Hence, in the most recent 
 construction, the offtakes enter vertical up takes, closed at the top by 
 explosion doors, and are taken off the furnace as high above the stock line 
 as possible, preferably at four points equally spaced on the circumference. 
 The downcomer connections are taken off part way up on these up-takes. 
 In locating the uptakes in furnaces of most recent construction, care is 
 taken to see that they do not enter the furnace directly over the tapping 
 hole, cinder notch, or the entrance of the blast main to the bustle pipe, 
 because, these being the most active points in the furnace, this 
 arrangement will tend to give a more even distribution of the gases 
 through the stock. 
 
 Runners: Though not given much prominence in blast furnace dis- 
 cussions, the runners, through which the slag and metal are carried away 
 from the furnace, constitute an essential part of the furnace proper. These 
 are metal castings in the form of deep troughs which are made in sections 
 laid end to end and buried so that their top edges are flush with the floor 
 of the cast house. The trough leading from the cinder notch will, of course, 
 be elevated. It forms an uninterrupted passage for the slag from the cinder 
 notch to the slag ladle or granulating pit. The metal runner is more com- 
 plicated. Beginning as a very deep trough at the tapping hole, it is inter- 
 rupted at the end of about 10 feet by the skimmer, a device for separating 
 
142 BLAST FURNACE 
 
 the metal from the slag that comes near the end of a cast. There are 
 two branches here, one for carrying away the slag and another for draining 
 the metal from this part of the skimmer trough after the cast. From the 
 skimmer the main trough is drained by branches leading to casting beds 
 on the floor of the cast house, or what is more common now, to hot metal 
 ladles on a track far enough below the floor of the cast house to permit 
 the metal to flow into them from above. Before casting, these troughs 
 are given a heavy coating of a loam or clay wash, which acts as an insulator, 
 protects the trough from the hot metal and facilitates the subsequent 
 cleaning up. Without this wash the hot metal would either chill in the 
 trough or melt it away. 
 
 SECTION V. 
 
 BLAST FURNACE ACCESSORIES. 
 
 The Stoves, of which there are nearly always four to a furnace, are 
 first in importance under the heading of accessories, being an absolute 
 necessity in modern blast furnace operations. This importance is due to 
 their function of heating the blast. The first stoves used were constructed 
 of iron pipes enclosed in a brick structure through which the blast passed 
 to the furnace, the gases from the furnace being burned as they circulated 
 outside and around these pipes, the recuperative principle. Then it was 
 found that the regenerative principle is much more efficient, so that now 
 stoves are built entirely of brick. Essentially they are brick walled cylin- 
 ders, enclosing a combustion chamber and a system of regenerative flues. 
 Externally, the brick wall of a stove is reinforced and supported by a steel 
 shell of riveted plates. The top of the stove is dome-shaped. Generally, 
 the stoves are as high and almost as wide as the furnace itself. They 
 vary in size with the size of the furnace. For the largest furnaces 
 they are approximately 100 feet in height and 22 feet in diameter. Inter- 
 nally, the combustion chamber will extend from the bottom to the top of the 
 stove, and may be located at the center, in which case they are called 
 center combustion stoves, or at the circumference, as in side combustion 
 stoves. The regenerative flues are filled with brick checker work, the 
 checkers being so laid as to form a system of vertical flues, from five to 
 nine inches square, which extend from the rider walls on the bottom to the 
 top of the stove. The arrangement of the flues also furnishes a means of 
 classifying stoves. Stoves in which the gases from the combustion chamber 
 pass through only one regenerative flue, are called two=pass stoves, while 
 in three=pass and four=pass=stoves they pass through two and three 
 regenerative flues, respectively. Two-pass and three-pass types are the 
 most common. Since the combustible gases are burned at the bottom 
 always of all stoves, it follows that in two-pass stoves the products of 
 combustion, passing through the checkers, must leave the stove at the 
 
STOVES 143 
 
 bottom, hence the opening to the stack on two-pass stoves is at the 
 bottom. At some plants, stoves of this type are provided with individual 
 stacks which rise along the side of the stove, but in other plants under- 
 ground flues from a number, usually a set of four, of such stoves empty 
 into a single stack, centrally located. In the three-pass type the hot gases 
 are returned to the top of the stove, there to escape through an opening 
 in the dome into a stack which tops the stove. 
 
 Stove Burners and Valves: Since the checkers in stoves are alter- 
 nately heated by the products of combustion of a gaseous fuel in their 
 passage to the open and cooled by the passage of the air or blast in an 
 opposite direction and under pressure, a system of burners and gas tight 
 valves are required. The burners are usually very simple in construction, 
 consisting of a movable gooseneck mounted on a rack attached to the 
 terminal of a vertical section of an underground gas flue in such a manner 
 that the horizontal portion extends into a gas port in the side of the stove. 
 A plate, or valve cover, is attached to the base of the goose-neck or the rack, 
 so that racking the goose-neck back and forward automatically closes and 
 opens the connection to the gas main. There are a number of moie com- 
 plicated and patented burners in use, which aim to increase the efficiency 
 of the stove 1 . The valves present such a variety of forms that a detailed 
 description of all cannot be attempted. The essential ones are named 
 from their location. The gas valve has already been described in con- 
 nection with the burner. The chimney valve is located at the base of the 
 stack, its office being to prevent the escape of air through that opening 
 when the stove is on blast. In three-pass stoves this valve is controlled 
 from the ground by means of chain, or cable, and pulleys. The cold blast 
 valve is located in the air line, which branches from the cold main from 
 the blowing engines at a point just ahead of its entrance into the stove. 
 As it is never subject to high temperatures, an ordinary form of gate valve 
 may be used. In the three-pass type of stove this valve will also be at 
 the top. The hot blast valve controls the exit of the blast from the stove 
 through which the air passes. The blast being highly heated at this 
 point, the hot blast valve must be constructed to withstand high 
 temperature. It is usually of the mush-room type, and water cooled. 
 
 Other Stove Openings: Besides the openings mentioned above, the 
 stove will be provided with a blow off valve to relieve the pressure and 
 provide partial escape of air on changing the stove from air to gas. This 
 valve may also regulate the admission of air for combustion. Numerous 
 clean out holes, through which the flue dust that collects in the stove 
 while it is being heated may be removed, will be placed at points most 
 desirable for the purpose. 
 
 iSee Modern Methods of Burning Blast Furnace Gas by A. N. Diehl. Year 
 Book of the American Iron and Steel Institute, 1915. 
 
144 
 
 BLAST FURNACE 
 
 r 
 
 \ 
 
 I 
 
 i 
 
 -1! 
 
STOVES 
 
 145 
 
 Stove Linings: Stove linings is a term that corresponds to furnace 
 lining, and includes all the brick work enclosed by the shell. As in the 
 case of the furnace, an expansion space of about two inches is left between 
 the circular brick wall and the shell. For these linings a strong yet porous 
 firebrick is required, because such brick absorb the most heat and also 
 give it up most readily. The brick need not be very refractory, for the 
 temperature in the stove is relatively low except in the combustion chamber, 
 where a brick possessing fairly high refractory properties is required. The 
 temperature of the hot blast is maintained at about 538 C. (1000 F.) 
 which marks the lowest temperature to which the hottest part of the stove 
 
 Fio. 21 a. Cross Section of Blast Furnace Stove Section C C of Fig. 21. 
 
146 BLAST FURNACE 
 
 can drop. With modern stoves, from 25 to 30% of the gas produced by 
 the blast furnace is required to maintain the blast at the correct temper- 
 ature. Of the remainder about one fifth, (12 to 20% of the whole) is used by 
 the blowing engines, so that a little more than half of the total gas produced 
 by the furnace is available as surplus for the generation of electrical power. 
 
 Dust Catcher and Gas Mains: From the down-comer the gas from 
 the furnace passes directly into the dust catcher. Its object, as implied 
 by its name, is to clean the gas as much as possible of the flue dust blown 
 over from the furnace, with which dust the gas is heavily laden. If this 
 dust is not removed, in part at least, it cakes upon the wal Is of the combustion 
 chamber and small flues of the stoves and, dropping down, necessitates frequent 
 cleaning. Besides, it acts as an insulator on the brick, preventing the full 
 absorption of heat. Similar conditions also prevail when the dirty gas is 
 burned under boilers. The dust catcher may be looked upon as a great 
 enlargement of the flue, or down-comer. Its diameter is often 20 feet or 
 more. It is brick lined, often has a dome shaped top, and a bottom in 
 the shape of an inverted cone. The principles involved in its construction 
 is that of greatly reduced velocity accompanied by sudden change in 
 direction. By this means the dust in the gas may be reduced sufficiently 
 to be used under boilers and in stoves with large flues very satisfactorily. 
 From the dust catcher the gas passes through a gas main that divides 
 into two branches one to supply the stoves and one to furnish gas to the 
 boilers for generating steam, which disposal will now apply to the older 
 and less progressive plants only. In up-to-date plants, the gas will be 
 subject to additional and more efficient treatment, after which it may be 
 used in the two ways mentioned or in internal explosion engines. This 
 additional cleaning of the gas is a necessity where gas engines are used, 
 and it is also claimed that gas for the stoves is cleaned at a profit, since 
 it eliminates the necessity of frequent cleaning of the stoves and permits 
 smaller checker flues, thus increasing the heating surface of the brick. 
 The matter, while past the experimental stage, is not yet fully developed 
 at all plants. As representing a highly developed method of gas cleaning, 
 the Duquesne works will furnish a good example. 
 
 Arrangement of Furnaces and Cleaning Plant at Duquesne: At 
 
 this plant there are six furnaces situated in a row, for the full length of which 
 extends a large gas main, called the rough gas main. The gas from all 
 these furnaces may enter this main after passing the dust catchers. From 
 this main, gas may be led to any part of the plant to be used in the raw 
 state, though it is primarily intended to supply the cleaning plant. The 
 flow of the gas through the main is controlled by water valves. A water 
 valve is a vertical cylinder with a cone-shaped bottom; in its center is a 
 vertical diaphragm reaching down over half way to the base of the valve. 
 Water can be admitted into the valve to a level somewhat higher than the 
 lower edge of this diaphragm. With the water level below the diaphragm 
 an outlet is provided so that a current of gas may be allowed to flow through; 
 
GAS CLEANING PLANT 
 
 147 
 
 when water rises high enough above the base of the diaphragm to resist 
 the pressure of the gas and prevent its passage under the diaphragm, the 
 valve is sealed. By means of these, any furnace can be shut off from the 
 system. The gas cleaning plant consists of two divisions called the Primary 
 and Secondary. The primary division receives and washes all the gas that 
 
 Blast Furnace 
 
 Dust Catcher* 
 
 Dust Catcher 
 
 Dustpocket 
 
 Fia. 22. Diagram showing Route of Gas from Furnace through 
 
 Gas Cleaning Plant to Boilers, Stoves and Gas Engines. 
 
 Carnegie Steel Company, Duquesne Steel Works. 
 
 is brought from the furnace by the rough gas mains and returns the washed 
 gas to the stoves or boiler houses when desired. The secondary division 
 
148 BLAST FURNACE 
 
 receives only such gas as is intended for use in the gas engines. 
 
 Primary Division: The primary cleaning of the gas is accomplished 
 by vertical water scrubbers and fans. It reduces the dust content of the 
 gas to .06 grain per cubic foot of gas under standard conditions. The 
 vertical water scrubbers are the most important part of the equipment in 
 respect to the amount of dust removed from the gas. There are nine of 
 them, and the gas connections for them are led to their bases from the 
 gas main through water-and damper- valves. These scrubbers are vertical 
 steel cylinders, unlined, 77 feet 6 inches high and 12 feet in diameter, and 
 are built of % inch steel plates. Gas is admitted at the base. Water is 
 admitted so as to fall against the gas, and, as the currents flow in opposite 
 directions, there is intimate mixture between them. The water is applied to 
 the gas in the form of a spray and when falling in the interior of the 
 scrubbers is like rain in that it is in small drops and thus presents the 
 greatest possible surface to the gas. 
 
 Methods of Scrubbing the Gas: Two methods of producing and 
 supplying the water spray have been used at this plant. The older method 
 employed a horizontal spray pipe located in the top of the tower and rotated 
 by a small electric motor. Water was supplied to it from a pipe inserted 
 at the top. The falling water was prevented from striking the shell by 
 a vane in the bottom of the spray pipe. Just below the spray pipe were 
 12 screens set together so as to break up the water stream as it fell, but 
 from that location the spray fell uninterrupted to the base, where it was 
 drained out through a gas seal. In the improved form, two series of seven 
 pipes each are inserted one above the other. The water is forced up through 
 screens placed at six foot intervals and must then fall back through them. 
 Motor-driven cut-off valves shut off the water from each pipe in turn, making 
 an area of low resistance over the pipe from which water has been cut off. 
 The gas rushes to this region; then water is turned on again and the gas is 
 deflected. A spiral motion results, giving a larger exposure of gas area to 
 cleaning water than would ordinarily result. The scrubbers use about 
 6,000,000 gallons of water per 24 hours. The temperatures of the gas 
 entering the scrubbers range from 300 to 600 F., and the pressure varies 
 from 8 to 16 inches of water. The water enters the scrubbers at river 
 temperature and at an average of 57.4 F. with a maximum of 84 F. and a 
 minimum of 33 F. The dust caught in the settling basin, which is built in 
 duplicate and extends from one end of the plant to the other, averages half 
 to three-quarters of a standard 50-ton hopper car a day. The gas passes 
 at a velocity of four feet per second up through the steel shell into a 
 pipe connecting with a 10 foot 6 inch main. The gas leaves at a temperature 
 varying from 96 F. to 37 F., or at an average of 68 F. The dust in the 
 gas is reduced from 3.5 grains per cubic foot to .22 grain at standard 
 conditions; the moisture in the gas entering the scrubbers is 34 grains per 
 cubic foot and on leaving is 8.5 grains. About 25 cubic feet of gas is 
 cleaned per gallon of water used. 
 
GAS CLEANING PLANT 149 
 
 The Fans : From the scrubbers, a large gas main, 10 feet. 6 inches in 
 diameter and about 40 feet above the ground, conveys the gas to a number 
 of fans that complete the primary cleaning of the gas. The connections 
 to these fans are provided with water valves. The fans are located in a 
 gas cleaning building, or fan house, and are four in number. Each fan has 
 a rated capacity of 84,000 cubic feet of gas per minute at 100F. The fans 
 raise the pressure of the gas to about six inches of water and thus give it 
 sufficient head to pass through the entire system of stoves, boilers and 
 engines; the furnace pressure alone is not sufficient to supply this head. 
 The gas leaves the fans at temperatures varying from 93 F. to 35F., or 
 an average of 69 F. By introducing water at several points into the shell 
 of each fan, the fans are made to serve as cleaners, and the dust content of the 
 gas is reduced from .22 grain per cubic feet to .06 grain per cubic feet. 
 
 Water Separator: From the fans, the gas passes through water 
 separators. These are made of two concentric steel cylinders which stand 
 in a vertical position. Tlie outer cylinder is much larger in Diameter than 
 the inner one and somewhat longer, so that the gas, entering at the top 
 of the outer cylinder and on a tangent, is given a downward spiral motion 
 and escapes at the bottom rising through the inner pipe. In this way, 
 the greater portion of the water, owing to its greater inertia, is deposited 
 by the gas current. From the water separator the gas enters the clean 
 gas main and is distributed to the stoves and boilers and also to the 
 
 secondary division. 
 
 
 
 The Secondary Division: This division furnishes gas for internal 
 combustion engines, which require gas almost as free from dust as the air 
 itself. It consists of four Theisens cleaners. This cleaner is a combina- 
 tion fan and cleaner. Externally it has a form approximately like that of 
 a large steel cylinder and is mounted horizontally. This outer cylinder is 
 stationary and encloses a similarly shaped but smaller revolving cylinder on 
 the shell of which is riveted twenty-four steel -vanes. These vanes project 
 12 inches from the shell of the inner cylinder and extend longitudinally in a 
 slight spiral to the circumference. At the receiving end the vanes project 
 beyond the end of the cylinder to form a drawing fan for receiving the gas, 
 while at the delivery end they terminate in blades, attached to the same 
 cylinder, that act as a booster fan for propelling the gas through the succeed- 
 ing apparatus. Water is admitted at low pressure through six pipes half 
 way up and on the side of the outer shell. This water is dashed to a spray 
 by the revolving vanes, and, being propelled in a direction opposite to that 
 of the gas, is thoroughly mixed with it, thus wetting the last small particles 
 of dust, which must, therefore, separate with the water. This water is 
 let out of the apparatus through a water seal at the bottom. The gas 
 flows through the shell and out into a water separator, thence to the gas 
 main leading to the gas engines. The Theisens cleaners have a rated 
 capacity of 14,000 cubic feet of gas per minute at standard conditions. The 
 gas leaves the Theisens at an average temperature of 64.2 F., or a maximum 
 
150 BLAST FURNACE 
 
 of 91 F. and a minimum of 35 F. The water enters at an average of 57.5 
 F. The dust in a cubic foot of gas is reduced from .06 grain to .009 grain. 
 45.44 cubic feet of gas is cleaned per gallon of water consumed. 
 
 SECTION VI. 
 
 EQUIPMENT FOR HANDLING RAW AND FINISHED MATERIALS. 
 
 The Boiler House, Power Plant, Pumping Station, Blowing Engines, 
 etc. while constituting a very vital part of the blast furnace equipment 
 present features of more interest to engineers than to metallurgists and 
 are therefore, best omitted from this discussion. * 
 
 Dry Blast : About 60% by weight of all the materials entering the 
 blast furnace is air. As air always contains moisture and since the decom- 
 position of water is an endothermic reaction, the heat absorbed by the 
 amount of water thus entering the furnace may be very great. It has 
 been estimated that during the month of July, for instance, the average 
 quantity of water, per hour, entering a furnace lising 40,000 cubic feet of air 
 per minute is approximately 224 gallons. That this quantity of water 
 may seriously affect the operation of the furnace is now well recognized, 
 and installations for drying the air have been made at a few plants. With- 
 out discussing in detail the apparatus used, the principle employed is that 
 of refrigeration. By cooling the air to a low temperature by drawing it 
 over a system of pipes cooled with brine, (a solution of common salt, NaCl2, 
 or calcium chloride, CaCl.2, which has a less corrosive action on the pipes), 
 which in turn is cooled with liquified ammonia, the moisture is condensed 
 and frozen on the pipes, leaving the air practically dry. 
 
 Cold and Hot Blast Mains: It is still the most common practice to 
 use undried air, which, compressed by the blowing engines, is forced 
 normally under the high pressure of about 15 pounds per square inch through 
 the cold blast main into the stoves, from which it issues highly heated; and 
 passing successively through .the hot blast main, the bustle pipe and the 
 tuyeres, begins its work in the furnace. In this connection one or two 
 apparently minor details of construction referring to temperature regu- 
 lation and pressure control should be noted. Leading around the stoves 
 from the cold blast mam into the hot blast main is a small pipe called 
 the by=pass. It provides a means of controlling the temperature of the hot 
 blast. The Snort Valve, also located in the cold blast main, is used to 
 reduce the pressure of the blast at the end of a cast while the tap-hole is 
 being stopped, or to release the pressure in case of a hanging furnace. 
 
 Appliances for Handling Ores, Coke and Stone: As was pointed 
 out in Chapter III, all ore is shipped over the Lakes from May until 
 December. Consequently the ore required to operate the furnaces during 
 the months intervening between shipping seasons must be stored until used, 
 either at the docks or at the w^orks. This storing of ore requires a stock 
 
 *For details on construction of the blast furnace and equipment, see Blast- 
 Furnace Construction in America by J. E. Johnson, Jr., published by McGraw-Hill 
 Book Company, New York. 
 
HANDLING RA W MA TERIALS 151 
 
 yard with suitable provision for an ore pile, rapid means of unloading cars, 
 and convenient and economical methods and appliances for handling large 
 quantities of ores. For unloading the cars, car dumpers have been 
 installed, while for piling and delivering the ore to the bins over the stock 
 house, ore bridges are employed. The ore, arriving at the works in train 
 load lots, is switched to a siding ahead of the car dumper, and the cars 
 are unloaded one by one in rapid succession. A car, being pulled up an 
 incline to the platform of the dumper, is bodily lifted and turned over so 
 as to empty its contents into large larry cars, or into bins, if small larry 
 cars are used. The dumper then resumes its former position, and the car 
 is pushed off the platform by the next car of ore to an incline, down which 
 the empty car moves to a car siding. Larry cars, designed for the purpose, 
 carry the ore to the ore pile, where the ore bridge picks up the ore and 
 dumps it in its proper place in the pile. The details of this operation will 
 vary much, but the general scheme is essentially as stated. Aside from 
 the mere storing of the ore other aims, while more or less incidental, must be 
 kept in mind. In order to obtain uniform conditions necessary to keep the 
 furnace operation under good control, it is desirable to mix the ore of each 
 kind or grade as much as possible; again, the uniformity of the ore may 
 be affected by dumping on large, sharply peaked piles, because dumping in 
 this manner causes a separation of the coarse and the fine material, which 
 always differ widely in chemical composition. For similar reasons, the 
 use of ore direct from hopper cars unloaded from the trestle into the bins 
 is undesirable. The ore is the only material stored, both the limestone 
 and the coke being brought in as required. All up-to-date plants are 
 provided with an ore trestle running out over the bins above the stock 
 house. The bins are used for storing smaller amounts of ore, fuel and flux, 
 which are then conveniently available for immediate use. The bins are 
 large hoppers, the bottom openings of which are closed in such a way as 
 to permit the withdrawal of fixed quantities of materials as desired. The 
 bins for the three materials are alike except that those used for coke are 
 provided with screens for removal of the "fines." In the most modern 
 plants the open tops of the bins are covered with a heavy grid-iron grating, 
 which serves the two-fold purpose of preventing accidents, resulting from 
 workmen falling into the bins, and the stoppage of the chutes below, due to 
 oversize pieces of material that might otherwise be dropped in. 
 
 Stock House Equipment: Under the trestle and bins is a large space 
 known as the stock house. Here are found the mechanical devices which 
 have superseded the old and original method of charging by hand. The 
 equipment will be different for each type of hoist. If the skip hoist is in 
 use the arrangement in general will be as follows: The skip tracks will 
 extend down beneath the floor of the house far enough to permit the lowering 
 of the skip beneath chutes, which lead from the floor so as to deliver the 
 materials into the mouth of the skip. These materials will be delivered 
 to the chute by means of a small trolley car, running on tracks that extend 
 
152 BLAST FURNACE 
 
 under the bottom openings of the bins. This car a small hopper car 
 is provided with a scale to weigh the ore and stone as it falls into the hopper 
 of the car. In this way any mixture of ores or stone desired may be 
 accurately made up by weight for charging. In the bucket hoist the bucket 
 itself is placed, on descending, upon the weighing car, which is transported 
 by trolley or dinkey from bin to bin for the different ores required in making 
 up the charge. Only the ore and stone are weighed, the coke being charged 
 by volume. 
 
 Disposal Equipment for the Iron: The old method of casting the 
 metal in beds of sand has, for many reasons, been replaced by casting 
 machines. Of the two types of these machines, the endless chain carrying 
 a series of parallel moulds or troughs with over-lapping edges is the one 
 most commonly used. In the operation of this machine, the molten metal 
 from the furnace is allowed to flow into ladles, which are pulled at once 
 into the casting house. Here, the metal is poured slowly into a trough 
 from which it flows onto two lines of moving moulds, which have been 
 previously prepared, to prevent sticking of the iron, by being either 
 "limed" or "smoked." The chains may carry the iron directly through 
 a trough of water, or dump the half cooled pigs upon a second conveyor to 
 be so cooled. A number of modifications of this machine are in use. 
 
 Equipment for Slag Disposal: The greater portion of the slag 
 produced cannot be used except as waste, so most of it will be transported 
 while molten to a convenient spot and dumped. When the slag is to be 
 used for certain purposes, as for making Portland cement, it is best granu- 
 lated. This condition is produced as the slag flows from the furnace by 
 allowing it to fall into a large concrete lined pit, partly filled with water 
 and known as the granulating pit. By forcing a small stream of water against 
 and from behind the stream of molten slag as it drops into the pit, the stream 
 of slag is broken up and the fineness of the slag is increased. Merely 
 allowing the slag to fall into the water is a much less effective method. 
 
 SECTION VII. 
 
 OPERATING THE FURNACE. 
 
 Blowing In : Upon being completed and provided with as much of the 
 equipment described above as is necessary or desired, the active career of 
 the furnace is begun. In blast furnace parlance, the process of starting a 
 furnace is called blowing in. It is carried out in three steps; these steps 
 are drying, filling and lighting. 
 
 Drying: Newly constructed furnaces and stoves, or new linings, must 
 be carefully and thoroughly dried before being put into operation. In the 
 case of a furnace fully equipped and ready to operate, the drying may be 
 accomplished by either wood fires built in the hearth or by gas. The heat 
 is applied very gradually, and the drying is continued for about ten days. 
 
OPERATION OF THE FURNACE 153 
 
 Filling: After the furnace is sufficiently dried, it is allowed to cool 
 slightly, and then the important process of filling is begun. While different 
 individuals will pursue slightly different methods, the general scheme will be 
 rather uniformly carried out. Briefly stated, it consists of first placing wood 
 and coke on the bottom to a height somewhat above the tuyeres, about which 
 fine kindling, shavings, oily waste or any material easily ignited is piled; 
 then following the wood with a large quantity of coke, mixed with enough 
 lime stone to flux its ash, and gradually introducing ore with the proper 
 amount of flux. Good practice requires that this initial volume of coke 
 should be about half the cubical contents of the furnace. Sometimes, to 
 get an easily fusible slag and a good volume of it, blast furnace slag may 
 be introduced ahead of the ore. These are called the blowing-in burdens, 
 and additions are made till the furnace has been filled to the stock line, 
 when it is ready for lighting. 
 
 Lighting: Starting the burning of the wood in the bottom of the 
 furnace may be done in several ways. If the space in front of the tuyeres 
 has been filled with light kindling wood, as is customary, oil is poured or 
 sprayed in at the tuyeres until the wood is thoroughly soaked with it. Then 
 with all the gas burners and valves in the gas mains and the bells closed, 
 the bleeder and explosion doors are opened, a light blast is turned on and 
 the wood ignited by inserting hot bars through the tuyeres. Often, instead 
 of the hot bars, a wood fire is built in the stove nearest the furnace, and 
 the oil is ignited by blowing sparks over with the blast. With a light 
 blast on, the wood soon burns away, and the stock begins to settle, after 
 which the blast pressure is gradually increased. Some furnacemen start 
 off, after the fires are well caught, with a fairly high blast pressure for a 
 few minutes, in order to drive the flames well in toward the center of the 
 furnace and consume the wood quickly, as it is thought that a better initial 
 settling of the stock is thus obtained. As soon as the stock gives signs 
 of settling, the blast pressure is reduced to that normally used for the 
 remainder of the blowing-in period, which is at first about % that used 
 when the furnace is in full blast. Up to this point a great deal of gas and 
 smoke escape from the furnace openings, and great care must be exercised, 
 for the gases contain a high percentage of carbon monoxide, and are very 
 poisonous. Great care is also required to prevent explosions, because 
 mixtures of furnace gas and air in a wide range of proportions are explosive. 
 Since the interstices of the stock in the furnace and all the gas mains are 
 filled with air to start with, an explosive mixture may be formed any time 
 soon after the lighting, and if this mixture should be ignited it might 
 cause serious damage. The difficulty is generally overcome by providing 
 outlets for the gas at the ends of the gas mains. These outlets are kept 
 open until all the air has been expelled, which condition is indicated by 
 the color and odor of the escaping gas. Both men and fires are kept 
 away from these openings until it is time to use the gas and the outlets 
 are closed. 
 
154 BLAST FURNACE 
 
 Heating the Bottom: Another feature connected with the lighting 
 of the furnace is heating up the bottom, which is warmed by the dry- 
 ing-out fires to only a slight degree as compared with the temperature 
 required to keep the slag and iron that form in a molten state. In order to 
 have the bottom at the proper temperatrue when slag begins to form, two 
 methods are employed, both of which involve leaving an opening at the 
 tapping hole so as to draw the flame downward from the tuyeres upon 
 the bottom. In the first method a round tapered wood plug, three or four 
 inches in diameter at the smaller end, is placed in the tap-hole and the 
 space about it is packed full and tight with clay. With the rise in tem- 
 perature due to the burning of the wood and coke in the bottom, the clay 
 sets, and this plug is then removed, which permits the flame from within to 
 shoot forth, thus heating up the runner outside as well as the bottom inside of 
 the furnace. When slag begins to flow from the tap-hole, the opening is 
 closed until time for tapping the first iron has arrived. In the second 
 method, an iron pipe, about four inches in diameter, is placed in the fur- 
 nace, before it is filled, so that one ead protrudes from the tap-hole out- 
 side of the hearth, while the other extends to the center of the furnace. 
 The space about the pipe where it passes through the wall of the hearth is 
 tamped with clay or ball stuff, which is also built up about the part of the 
 pipe within the furnace for a foot or so from the hearth wall. When the 
 furnace is lighted the gas flame is drawn to the center of the bottom to 
 pour forth from the exterior end of the pipe. This pipe need not be moved 
 until a fairly large flow of slag is attained, when it is drawn from the tap- 
 hole, which is immediately closed, as in the case of the wooden plug. 
 
 The heating of the stoves is another factor connected with the light- 
 ing of the furnace. The temperature of the hot blast when the furnace is in 
 full operation is 500 to 550 C. (930 to 1020 F.), and it is a great help if 
 the stoves can be heated nearly to this point for the lighting of the furnace, 
 especially as the furnace and filling are cold to the bottom. But this stove 
 temperature can be obtained by the use of gas only, so that in the case of 
 isolated furnaces where gas is not available before starting up the furnace, 
 the stoves must be heated as hot as possible for the lighting by means of 
 wood and coal fires. 
 
 Tapping: At the end of ten or fifteen hours after the blast is on full, 
 there will be a sufficient accumulation of slag to tap. This is done by re- 
 moving the bott from the monkey, and pricking through the solid slag clos- 
 ing the opening, if the cinder does not flow immediately. The bleeder is 
 closed after the first cinder is tapped, as the gas can now be used in the 
 stoves, and boilers or gas engines. It requires from 30 to 40 hours before much 
 iron accumulates. When the iron is ready to tap, a hole is bored by means 
 of a long auger or drill, electrically or otherwise operated, almost through 
 the clay plug of the tapping hole. During the boring, the dust is blown 
 put of the hole by a jet of compressed air. The splasher having been put 
 in place, the opening is then completed by driving a long pointed bar into 
 the furnace. When this bar is removed, the iron will usually flow out, at 
 
OPERATION OF THE FURNACE 155 
 
 first slowly. As the flow of iron progresses, the opening is enlarged and 
 the metal flows out rapidly. The iron will then flow out through the 
 runners under the skimmer to the ladles provided to receive it. During 
 the flow of the metal, samples of the iron for chemical analysis and fracture 
 tests are taken by collecting small spoonfuls from the main runner. The 
 slag, which follows the iron near the end of the cast, is stopped by the 
 skimmer, where it may be run off through a more elevated runner to the 
 slag ladle or granulating pit. When the iron has almost ceased to flow 
 from the tapping hole, and gases are pouring forth, the blower signals the 
 engineer to reduce the blast and opens the snort valve on the cold gas 
 main to relieve the pressure. Then, the iron and slag having been drained 
 from the skimmer, the clay gun, hung on a crane, is swung into the 
 opening, either by hand or mechanically. This gun is provided with a 
 steam cylinder which operates a rammer that forces a quantity of clay 
 mixed with a little coke dust into the tap hole. The clay forms a plug 
 that closes the opening. This plug of clay is then backed up with more 
 of the mixture, which is fed into the gun, through an opening for the 
 purpose, in the form of moist balls. As soon as the hole is stoppered, 
 the snort valve is closed and the furnace goes on blast till next tapping 
 time four, five or six hours afterward. 
 
 Care of Runners: After the tapping hole has been closed, from one to 
 three minutes being required, the troughs are emptied, and preparations for 
 the next cast are begun. The runners are cleaned carefully of both metal 
 and slag, and their inside surfaces are carefully brushed with a thick clay 
 or loam slurry which, when dry, protects the trough, an.d prevents the 
 iron from sticking to the runner. 
 
 Sampling the Iron: Sampling pig iron is a very important part of 
 every tapping. As the iron is graced by chemical analysis, care should 
 be taken to secure a sample for the chemical laboratory that will be repre- 
 sentative of the whole cast. This sample, therefore, is generally made 
 up of a number of equal portions taken from the main runner at the farther 
 side of the skimmer and at periods corresponding to the middle of each 
 ladle of metal in the cast. These samples may be in the form of shot made 
 by pouring the molten metal slowly into water or upon a cold iron plate, or 
 they may be small castings made by. pouring the metal into a mould. In 
 addition to these laboratory tests, samples called sand tests or chill tests, 
 according to the manner of casting them, are also taken. In these tests 
 the iron is allowed to cool in small moulds about two inches square in cross 
 section and four inches long. The moulds maybe either of sand as in "sand 
 tests" or of metal, when they are called "chill tests." When cold, the 
 small casting is broken with a hammer, and from the fracture thus exposed, 
 if the test has been cooled properly, the blower is able, generally, to judge 
 very closely as to the quality of the iron. Chill tests of all slags are 
 also taken and carefully inspected. 
 
156 BLAST FURNACE 
 
 Tapping Slag: In about two hours, the slag will have risen near the 
 tuyeres, and another flush will be necessary. If the iron is tapped six 
 times a day, only two flushings of slag are necessary between tappings, 
 but if the tapping is on a five hour schedule three flushings will be 
 required. 
 
 Changing Stoves: The temperature of a furnace at the hearth is a 
 matter of great importance, as this is one of the two main factors which 
 control the quality of the iron produced. One of the means of regulating 
 this temperature is by changing the slag composition, as has been suggested. 
 Another way by which quicker results may be obtained is by control of 
 the hot blast temperature. This may be raised or lowered by use of the 
 by-pass, and can be kept high by proper manipulation of the stoves. As 
 a part of the routine of blast furnace work, the tending of stoves is of im- 
 portance. They must be kept clean and be changed regularly and a't not too 
 long intervals. Usually but one stove at a time is employed for heating the 
 blast, and the stoves are changed once each hour. Thus, each stove is heating 
 for three hours. In changing stoves the hot stove must be put on the furnace 
 before the cold one is taken off. To put a stove on hot blast, the gas burner 
 is racked back from the gas port, and the blow off and chimmey valves are 
 closed. Then in quick succession the cold blast valve and the hot blast valve 
 are opened, when the blast is free to pass through the stove, which it does 
 in the direction opposite to that by which the stove was heated. The 
 cold stove is nov taken off, the procedure being the reverse of the above. 
 The cold air valve is closed, and then quickly, the hot blast valve. To 
 relieve the pressure in the stove, the blow-off valve is slowly opened, which 
 permits the chimney valve to be opened. The stove is then ready for the 
 gas, which is admitted by racking the burner forward. 
 
 Charging the Furnace: The charging of the furnace is a part of the 
 routine that must be done with great care and cannot be interrupted. The 
 furnace tends to empty itself rapidly, and constant vigilance is necessary 
 to keep the stack full. The proportions of the materials used is a pre- 
 determined quantity. Therefore, all the materials are carefully weighed 
 before charging into the furnace. The charging is usually done in rounds. 
 The basis of charging is the weight of fuel in each round. The fuel remains 
 a fixed quantity, and any variations in the charge are made with the ore 
 and flux. Usually the coke in the round is measured by volume and 
 not weighed, but, of course, the weight of the given volume in a 
 round is known. The weight of this coke unit varies at different plants, 
 because it is subject to no fixed rule, the opinions of furnacemen 
 differ as what it should be, and it is affected by the size of the 
 furnace and other conditions. The weights most often used are 10,000, 
 12,000, and 15,000 pounds. Under present conditions the weight of ore 
 in the rounds will approximate twice and the limestone half the weight 
 of the coke. The manner of charging the materials is also subject to 
 much variation. Often it will be found that all the coke in a round 
 
IRREGULARITIES 157 
 
 will be charged, followed by the ore and limestone mixed together. To 
 charge in this manner, each skip or bucket of coke is first dropped upon 
 the small bell or placed over the gas seal which is lowered to allow the 
 coke to fall upon the big bell. This operation is repeated until all the 
 10,000, 12,000 or 15,000 pounds of coke has been dropped upon the big bell, 
 which is then lowered, allowing the coke to drop into the furnace. The 
 ore and stone are then charged in the same manner. To illustrate the 
 variation to be expected in the manner of charging, the simple scheme 
 outlined above may be compared with the following, which was once found 
 in use at a certain plant. 
 
 1 skip of ore mixture of ores A. and B. Weighed. Small bell lowered. 
 1 skip of stone and ore mixture of stone and ore C. Weighed. Small 
 bell lowered. 
 
 1 skip of coke not weighed. Small bell lowered. 
 1 skip of coke small bell lowered. 
 
 Big Bell Lowered. 
 
 1 skip of stone and ore mixture of stone and ore C. small bell lowered. 
 1 skip of ore mixture of ores A., B. and C. small bell lowered, 
 1 skip of coke small bell lowered. 
 1 skip of coke small bell lowered. 
 
 Big Bell Lowered. 
 
 Some Irregularities of Furnace Operation: The blast furnace, even 
 in its highest development, is by no means the even-going, easily-regulated 
 monster the casual observer may take it to be. Although furnace operations 
 are under better control now than ever before, the furnacemen still refers 
 to his furnace in the feminine gender, because, he knows she is a fickle 
 maid capable of acting in most unexpected and astonishing ways. There- 
 fore, a full discussion of this subject would lead to possibilities and prob- 
 abilities almost without end. However, the subject lends itself to at least 
 one positive statement. It is this: there are few situations in life where 
 promptness and decision, forethought and good judgment, skill and experi- 
 ence are more needed than about a blast furnace in times of trouble. A 
 few of these troubles are here enumerated. 
 
 Slips are due to a wedging of the stock in the upper part of the stack. 
 They are thought to be caused by carbon deposition, which may, in some 
 cases, be more in volume than that of the ore. This deposition fills up 
 the interstices of the stock, so that the gas can penetrate it only with 
 difficulty. When this condition occurs the stock beneath the wedged 
 portion settles from that above, the blast pressure rises and the wedged 
 stock finally falls. The sudden release of pressure on the gases produces 
 
158 BLAST FURNACE 
 
 a result like that of an explosion. Slips of great violence have been known 
 to tear off the top and do very serious damage. 
 
 Scaffolding occurs near the top of the bosh. This condition is often 
 due to irregularities in the working of the furnace, the following explanation 
 often being suggested: If the zone of fusion is suddenly lowered, the pasty 
 mass at its top tends to adhere to the encircling wall, with the result that 
 an incrustation is formed which projects toward the center of the furnace. 
 This mass offers obstruction both to the gases and to the descent of the 
 stock. If this condition is not soon remedied, the blast gases will channel, 
 perhaps on one side, in which case serious damage to the lining would 
 result. Dynamite is sometimes necessary to break a scaffold. This 
 condition is often referred to as hanging. 
 
 Chimneying and Hot Spots: Chimneying is caused by the improper 
 distribution of the charge with the coarser material segregating to the 
 center of the furnace. The hot gases naturally seek the lines of least 
 resistance, and the principal reaction is up through this more open center, 
 with a corresponding slower movement of the finer and more compact 
 material along the side walls. With the coarser materials segregating next 
 the side walls, the more violent reactions are next the brick work, with a 
 cold column in the centre, and the condition is sometimes called pillaring. 
 If the latter condition becomes localized, the action of the stock and the 
 hot gases soon cut away the walls adjacent to the area affected, if the 
 condition continues for any length of time. Eventually, this may 
 develop a hot spot, showing on the shell. By the generous use of 
 water sprayed against the hot spot the furnace can sometimes be kept in 
 operation for a considerable length of time after a hot spot shows. In 
 either case the colder material from the inactive zone causes a cold hearth 
 and poor quality of iron. 
 
 Loss of Tuyeres and Chilled Hearth may be brought about by burning 
 out the coolers due to failure of the water and by filling during a slip. Bad 
 slips always throw a great deal of the molten slag up into the tuyeres, 
 blowpipes and tuyere stock where it immediately solidifies, necessitating 
 a shut down. The large amount of comparatively cold stock that drops 
 into the hearth from a severe slip may lower the temperature of the molten 
 iron and slag below the fusion point, thus producing a chill in the hearth. 
 When the tuyeres are finally opened in case of a bad chill, and the furnace 
 is on blast, if necessary, before the tap hole can be opened, the iron can 
 be tapped through the cinder notch after the removal of the coolers. 
 
 Uncertainties and Variables in Furnace Control: Besides the 
 irregularities just mentioned which affect and occur in the furnace itself, 
 there are many others which may arise from outside sources, because to 
 obtain uniform working of the furnace it is necessary that all the raw 
 
BANKING 159 
 
 materials, the limestone, the coke, the ore, and the air, be kept uniform, 
 a feat that is manifestly impossible. Again, the furnace plant may be 
 looked upon as a composite mechanism containing many vital parts, of which 
 the furnace itself is but the central figure. The prolonged failure of any 
 one of these parts, the boiler plant, the blowing engines, the pumping station, 
 the gas mains, the water lines, or the stoves, is sufficient to close down the 
 furnace. All of these things are of great interest to the furnaceman, but 
 their discussion cannot be undertaken in as brief a discourse as the present 
 one is intended to be. 
 
 Banking: Whenever it becomes necessary to close down a furnace 
 temporarily, it is banked. This is done by charging coke blanks, beginning 
 a few hours before banking. The Amount of the blanks varies with the time 
 the furnace is to be off. After the blanks have been charged, the furnace 
 is drained as "dry" as possible of iron and slag, the connections to the rest 
 of the plant are closed, the blast is shut off, the blow pipes and the tuyeres 
 are removed, and the openings are bricked up tight. The bleeders, explosion 
 doors and then the bells are opened, and the gas is allowed to pass out at 
 the top. The furnace may now remain inactive for several days or weeks. 
 In starting up, the furnace is filled up with coke and a little ore, the ashes 
 are raked out through the tuyere openings, the tuyere connections are 
 made and the blast is turned on. The same precautions with regard to gas 
 must be observed here as in blowing in. A week or more, depending upon 
 the length of the banking peroid, may be required for the furnace to 
 return to normal condition. 
 
 Blowing Out: In blowing cut, charging is merely stopped and the 
 stock is allowed to settle. Streams of water are allowed to flow into the 
 try holes to keep the top cool and prevent warping of the bells. As soon 
 as the stock line descends near the tuyeres, the blast is taken off, the tuyeres 
 are removed, and the rest of the stock is later removed with shovels. This 
 done, the career of the furnace is ended. 
 
 SECTION VIII. 
 
 THE BLAST FURNACE BURDEN. 
 
 Burdening the Furnace: The amounts of ore and stone charge 
 per ton (or other fixed quantity) of fuel is referred to as the burden, the 
 fuel or coke being constant in amount. Any increase in ore and stone above 
 the normal is spoken of as a heavy burden, while the reverse of this results 
 in a light burden. The regulation of the proportions of ore, flux and coke 
 is called burdening. It has two objects; namely, the most efficient oper- 
 ation of the furnace and at the same time the production of the grade of 
 metal desired. The subject is of the greatest importance in the operation 
 of a furnace, and is a problem that may be solved either by practical exper- 
 ience or by calculations based on theoretical considerations. With a 
 furnace well started and on familiar materials, practical knowledge only 
 
160 
 
 BLAST FURNACE 
 
 may be required to operate successfully. But in dealing with unknown 
 materials, theoretical burdening based on chemical analysis must be resorted 
 to. A full discussion of these matters would make this Chapter too 
 technical for the purpose it is intended. However, as illustrating the pro- 
 blems that confronts the blast furnace operator, the following will supply 
 concrete examples: 
 
 Given: A furnace with a certain capacity and raw materials of the com- 
 position shown in the following table: 
 Required: 1. To produce 1 ton (2240 Ibs.) pig iron with 2000 pounds 
 
 coke or less. 
 
 2. To produce metal containing silicon, less than 1.25%; 
 sulphur, less than .040%; manganese, as high as possible; 
 carbon and phosphorus, not specified. 
 To determine: 1. In what proportions the ores shall be mixed. 
 
 2. Weight of ore mixture in the charge. 
 
 3. Weight of flux in the charge. 
 
 Table 27.* Analysis of Raw Materials Used in the Blast Furnace. 
 
 ELEMENTS AND RADICALS 
 Dry Basis 
 
 js 
 
 
 ORES 
 
 Limestone 
 
 Coke 
 
 1 
 
 2 
 
 3 
 
 4 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 Silica 
 
 SiO2 
 Fe 
 Mn 
 P 
 A1 2 3 
 CaO 
 MgO 
 C 
 C0 2 
 S 
 
 SOs 
 
 Na2O- 
 K2O 
 TiO2 
 
 H2O 
 
 6.48 
 56.80 
 1.14 
 .082 
 3.22 
 .11 
 .14 
 
 9.04 
 
 54.84 
 1.09 
 .096 
 2.70 
 
 .18 
 .26 
 
 12.78 
 52.93 
 .83 
 .083 
 3.34 
 .25 
 .21 
 
 5.86 
 55.59 
 .16 
 .618 
 3.63 
 1.22 
 .87 
 
 3.43 
 .30 
 .08 
 .006 
 .86 
 51.45 
 1.66 
 
 4.40 
 1.35 
 .07 
 .030 
 2.80 
 .25 
 .15 
 90.00 
 
 .940 
 
 Un- 
 deter- 
 mined 
 Trace 
 
 Trace 
 1.50 
 
 Iron 
 
 Manganese 
 
 Phosphorus 
 
 Alumina 
 
 Lime 
 
 Magnesia 
 
 Fixed Carbon 
 
 Carbon Dioxide 
 
 
 
 
 
 41.43 
 
 Sulphur 
 
 
 
 
 
 Sulphuric Anhydride 
 Alkalies 
 
 .03 
 Trace 
 .018 
 14.06 
 
 .04 
 Trace 
 .009 
 14.50 
 
 .04 
 Trace 
 .012 
 15.00 
 
 .06 
 Trace 
 .019 
 12.10 
 
 .060 
 Trace 
 Trace 
 .50 
 
 Titania. . . . 
 
 Water, (Wet Basis) 
 
 *N. B. The preceding table is intended to give a complete list of the 
 elements and radicals which make up the solid materials entering the 
 furnace, or the charge. The gases, i. e., the blast, may be looked upon 
 as a mixture of oxygen and inert gases composed mainly of nitrogen. In this 
 mixture the oxygen content is 20.8% by volume or 23.2% by weight. 
 
BURDEN 161 
 
 Outline of a Method for Solving a Burdening Problem : In a general 
 way the solution of the problem is arrived at in the following manner: 
 From the physical condition of the various ores and the amount of each 
 on hand, their relative cost or other considerations, the furnaceman 
 first decides the approximate proportions in which it is desirable to use 
 the ores. From these proportions he is able to determine the average 
 composition of the ore mixture in each charge, the size of which he has 
 also decided upon. From this average he is able to calculate the amount 
 of ore required to produce one ton of pig iron, and the weight of the impuri- 
 ties therein. Then, since he must make one ton of iron with one net ton of 
 coke, or less, he is able to arrive at the total impurities in the ore and coke 
 required to produce one ton of iron. These impurities, he separates into 
 acids and bases, and then combines them according to the slag ratio of 
 acid to base which experience has taught is the best to produce the kind 
 of iron desired. This process gives the excess acids which must be fluxed 
 with limestone. From the analysis of the stone he determines the available 
 base, from which the amount of limestone required to flux the excess acids 
 in accordance with the accepted ratio can be found. The next thing to 
 consider is the slag volume, or the amount of slag to be made per ton of 
 iron, which experience has taught must be within certain limits to be con- 
 sistent with good furnace practice. If the volume of slag is very low, its 
 ability to remove sulphur from the iron may be seriously interfered with, 
 while if it is very high, the fuel consumption will increase above that desir- 
 able, because coke must be consumed to furnish the heat necessary to form 
 and fuse the slag. If the slag volume falls outside the limits which the 
 furnaceman's judgment from experience has set for it, he must begin all 
 over again, starting with a different mixture of ores, or different limestone 
 or coke. Evidently, with new materials the solution of the problem involves 
 a great deal of try-work with different combinations of the materials that 
 may be available. 
 
 The Burden Sheet: During the operation of a furnace, the burden may 
 be changed from time to time to meet the ever changing conditions. These 
 changes are governed by observation and by the analysis of the pig iron and 
 slag produced. An accurate record is kept of all changes made, and of the 
 weights and analyses of all materials charged, and for purposes of record and 
 of comparison between the theoretical and actual conditions, this data is all 
 assembled at certain times, usually once each week, placed on a burden sheet, 
 and the theoretical amounts of the various ingredients of the raw materials 
 and the products are calculated. To illustrate the calculations involved, the 
 following burden sheet is appended. The figures given are based on a single 
 charge instead of on amounts of materials used for any given length of time, 
 otherwise they represent actual conditions and show a typical charge for a 
 furnace making basic iron. In studying the sheet it should be kept in mind 
 that only the weights of the ores, cinder, scale, scrap, coke and stone, together 
 with their analyses, and the theoretical analyses of the pig iron are given to 
 
162 
 
 FURNACE 
 
 3 
 
 PH 
 
 JiCiC-^ 
 5 CN OCO 
 
 50-HrH 
 
 OM rf CO 
 
 rH CO rH rH 
 
 OOOO 
 
 10 ooo C5 
 
 CO 1C 1C CO 
 COOrHO 
 
 TJ<COCN 
 coo o 
 
 oq 
 
 05 rH 
 
 O rHlC 
 C5 00 i-l 
 
 d CO i-< O 
 
 3 o ^ : 
 
 1C 1^ 
 
 3 3 
 
 CN 
 
 05 CO 
 
 00 t> COO 
 
 *< co icos 
 
 CO rH (NCO 
 
 16 cow 
 
 o oo <N?o 
 rn q qo 
 
 3 o>o 
 
 O O 
 
 <N IN 
 
 oo oo 
 
 O> OJ 
 
 <N <N 
 
 5 * A 
 
 1C ^ ? ^ 
 
 10 - <"~8 
 
 1=1 
 
 1 = 
 s . 
 
 O 
 
 > 
 
 aaj 1 
 
 CO CO CO O 
 iOO30<N 
 
 o^ co i> o 
 
 1 
 o^ 1 
 
 O CO CO CO 
 COCNCN<N 
 
 00 CN CO IN 
 
 lo'oodco 
 
 O O J^CO 
 
 co 01 qc 
 
 o ^ >o >c 
 
 1C CN rH05 
 
 o oo *o-* 
 
 CO CO rH 1C 
 rH (N COrH 
 
 >c co t> t> 
 
 O O CN O 
 C$ <N 00 CN 
 
 tO 
 
 OOt>.(NlOO rH 
 
 co co <N 
 
 O5 CO CO 
 
 1 I. 
 
 oo i 
 
 ooco 
 
 '^-fo 
 
 FH .00 
 
 no 
 
 rH O 
 
 ic co 
 
 : : 
 
 00 CO COCO 
 t^iCC5(N 
 
 05 CO 000 
 
 oooo 
 
 o o oo 
 
 iO O CO 
 
 S 
 
 
 
 'iC CO 1C CO 
 
 . (N O rH O 
 
 Socoooo 
 OOrHO5 
 
 r-((NO5OOO 
 Tt< Tt* i 1 (N t> 
 
 CO CO 00 OOO 
 
 TAL 
 CHARGE 
 
 TO 
 CH 
 
 ed 
 
 lag 
 
 Reduce 
 Slagged 
 
CHEMISTRY 163 
 
 begin with, and that all the other figures are supplied by calculation. Also, 
 it will be observed that the analyses of the raw materials are based on these 
 materials in their undried, or natural, state. 
 
 SECTION IX. 
 
 CHEMISTRY OF THE PROCESS 
 
 Methods of Investigating the Reactions of the Blast Furnace: A 
 
 question, which the non-technical reader is usually much interested in, is 
 this: What changes do these numerous ingredients of the raw materials 
 undergo during their passage through the furnace? Now, the reactions as 
 they take place in the furnace are beyond the reach of thorough investi- 
 gation; but from a study of the chemical properties of these elements 
 and compounds under conditions similar to those existing in the furnace 
 and from the chemical composition of the products, which is easily obtained 
 by analysis, together with observations made in working with the furnace, 
 reliable conclusions as to the reactions that must bring about the various 
 transitions may be formed. Nearly all the data necessary to this study 
 has already been supplied. However, the following brief review of the 
 chemical properties of the elements under furnace conditions may profitably 
 be made. These properties are revealed by laboratory experiments properly 
 conducted under conditions approximating those of the blast furnace. 
 
 The Functions of Oxygen and Carbon : 20.8% of the blast by volume 
 is oxygen, which enters the furnace at a high temperature, about 530 C., 
 and, coming in contact with hot coke, immediately reacts with carbon 
 giving off heat thus: 
 
 (1) C+O 2 =CO 2 (+97200 cal.) l^ 
 
 In the presence of an excess of carbon at a high temjjferature, CO 2 is at once 
 reduced to CO, and 68040 cal. are absorbed (2) CO 2 +C=2 CO (68040 cal.) 
 The net heat, then, from (1) and (2) is 29160 cal. (+97200 68040=29160 cal.) 
 At moderately high temperatures the CO gas formed acts as a powerful 
 reducing agent and will liberate heat at the same time, thus: 
 (3) 3CO+Fe 2 O 3 =2 Fe+3CO 2 (+8520 cal.) 
 87480 cal. 195600 cal. +291600 cal.=8520 cal. 
 
 At temperatures ranging from 250 C. to 700 C., a dull red heat, the reduc- 
 tion of Fe 2 O 3 by CO may take place in three steps, the Fe 2 O 3 being successive- 
 ly reduced to FesO^ FeO and finally to Fe. That the reduction of the ore 
 in the blast furnace does take place in this way is shown by the fact that a 
 large part (60% to 75%) of the flue dust, ejected from the top of the furnace, 
 is magnetic, though only Fe2Os, may have been charged. The total heat 
 liberated by these three reactions would be a little more than twice and the 
 total iron reduced one-half as much as that in reaction (3) for equal weights of 
 CO. A very interesting reaction that takes place between Fe2Os and CO at 
 low temperature is (4), in which carbon is deposited and heat is liberated thus: 
 \ (4) 2 Fe 2 O 3 + 8 CO = 7 CO 2 +4 Fe+C (+55920 cal.) 
 391200 cal. 233280 cal. +680400 cal. 
 
164 BLAST FURNACE 
 
 Carbon is not deposited with magnetite and CO reacting together, but 
 it may be deposited at temperatures below 600 C by the action of 
 metallic iron upon CO, thus; Fe+CO=FeO+C. The CO 2 formed in the 
 preceding reactions and from the decomposition of limestone may act as 
 an oxidizing agent, absorbing or giving off heat, as shown in reactions 
 
 (5) (6) (7). 
 
 /v ' (5) Fe + C0 2 = FeO + CO ( 2340 cal.) 
 
 97200 cal. + 65700 cal. + 29160 cal. 
 
 (6) 3FeO + C O 2 = Fe 3 O 4 + C O (+5660 cal.) 
 197100 cal. 97200 cal. + 270800 cal. + 29160 cal. 
 
 (7) 3Fe + 4CO 2 = Fe 3 O 4 + 4CO (1360 cal.) 
 
 388800 cal. + 270800cal. + 116640cal. 
 
 These reactions will take place at temperatures ranging from about 350 C. 
 to 800 C., and their extent will be governed by the relative amounts of 
 CO 2 and CO in the furnace gas, obeying, in this respect, the law of mass 
 action. To be reducing the volume of CO in the gas mixture must equal 
 or exceed twice the volume of the CO 2 . 
 
 Carbon alone is also a reducing agent toward oxides of iron at 
 low temperatures (450 C. to 700 C.), but the reduction of ore by carbon 
 alone absorbs much heat as shown by the following reactions: 
 
 (8) 3Fe 2 O 3 +C = 2Fe 3 O 4 -f CO (16040 cal.) 
 586800 cal. + 541600 cal. + 29160 cal. 
 
 (9) Fe 3 O 4 + C=3FeO + CO (44540 cal.) 
 270800 cal. + 197100 cal. -f- 29160 cal. 
 
 (10) FeO + C=Fe + CO ( 36o40 cal.) 
 65700 cal. + 29160 cal. 
 
 Under proper conditions the reduction of Fe 2 O 3 by solid carbon may take 
 place in a direct way, thus: 
 
 Fe 2 O 3 +J?C=2 Fe+3CO (108120 cal.) 
 195600 cal. +87480 cal. 
 
 At very high temperatures say around 1500C. carbon in large excess 
 may reduce manganese, silicon and phosphorus oxides, the reactions being 
 represented thus: 
 
 (11) Mn 3 O 4 +C=3 MnO+CO Heat is absorbed. 
 
 (12) MnO+C=Mn+CO Heat is absorbed. 
 
 (13) SiO 2 +2C=Si+2 CO. Heat is absorbed. 
 
 (14) P 2 O 5 +5C=2P+5CO Heat is absorbed. 
 
 Some oxygen also enters the furnace as water vapor in the blast, where the 
 
 following endothermic reaction occurs: H 2 O+C=CO+H 2 . 
 
 The hydrogen formed may do work temporarily by reacting with iron 
 
 oxide and reducing it thus: (15) FeO+H 2 =H 2 O+Fe. 
 
 But the water BO formed is again decomposed as shown by the presence 
 
 of hydrogen in blast furnace gas. Therefore, the net energy result from 
 
 water vapor is a loss. 
 
 In this connection it should be noticed that, since carbon is the only 
 
CHEMISTRY 165 
 
 fuel employed, the carbon-oxygen reactions must be relied upon to furnish 
 the heat required in the process, and that only a few of these are heat pro- 
 ducing. Reactions (l)to (4) produce most of the heat absorbed by other modes 
 of reduction, also that required to dry the raw materials, to decompose the 
 limestone, to flux the impurities, to melt the iron and slag, and to replace 
 the waste. On this account they are among the most important reactions 
 occurring in the furnace. 
 
 Behavior of Nitrogen in the Furnace: Nitrogen and the other inert 
 gases of the air, totalling 79.2% of the blast by volume, pass through the 
 furnace, for the most part, unchanged chemically. Since they equal in 
 weight about six-tenths of all the other materials entering the furnace, 
 they play an important part in heat conduction, and make a source of 
 unavoidable heat waste. Some nitrogen, however, may react with alkali 
 carbonates and carbon to form salts of hydro-cyanic acid. 
 (16) K 2CO 3 +4 C+N 2 =2K CN+3 CO. 
 
 This reaction explains the small amount of cyanogen, CN, always 
 present in blast furnace gases. 
 
 Action of Phosphorus in the Furnace: Phosphorus enters the 
 furnace with the charge in the form of phosphates. At very high tem- 
 peratures and in the presence of coke (carbon) these compounds are com- 
 pletely reduced, as shown in reaction (14). Phosphorus reacts with iron 
 to form Fe 3 P, thus: (17) 3Fe+P=Fe 3 P. 
 
 This phosphide, being soluble in iron, becomes a part of the metallic 
 bath in the blast furnace. Hence, the phosphorus in the pig iron can be 
 controlled only through the selection of materials. 
 
 Disposition of Sulphur in the Furnace: Sulphur is carried into the 
 furnace mainly by the coke, though small amounts are found in both the 
 ore and the limestone. The greater portion contained in the coke enters 
 in the form of FeS, which, when melted, alloys with the iron in the furnace; 
 a smaller portion, in the form of sulphates, as CaSO 4 , enters as an impurity 
 in the ore, limestone and coke, and is reduced to sulphide at a low red heat 
 and in the presence of carbon. At a very high temperature and in the 
 presence of a very basic slag, or CaO, and carbon, the following reaction 
 may take place^Q (18) FeS+CaO+C=CaS+Fe+CO. Owing to lack of 
 proper conditions in the blast furnace, this reaction is never complete, so 
 a small portion of the sulphur remains in combination with the iron. This 
 iron sulphide, being soluble in iron, becomes a part of the metal. 
 
 Behavior of Silicon: Silicon enters the furnace as SiO 2 , some of 
 which may be combined with bases as silicates. At temperatures of about 
 1200 C., corresponding to the fusion zone in the blast furnace, the greater 
 portion of this silica combines with lime, CaO, and other bases to form 
 silicates, which have already been discussed under the heading of slags. 
 However, at a high temperature, such as exists in the hearth of the furnace, 
 and in the presence of carbon, silica is reduced, and the resultant silicon 
 
166 BLAST FURNACE 
 
 combines with the iron. (19) SiO 2 +2C=Si+2CO. 
 
 (19A) Fe+Si=FeSi. 
 
 Action of Calcium and Magnesium: Calcium and Magnesium enter 
 the furnace mostly as carbonates. Small portions may be in the form of 
 silicates, in which CaO and MgO are combined with SiO2, and may undergo 
 no chemical change in the furnace. The carbonates, however, are decom- 
 posed at temperatures above 800 C., liberating CC>2. 
 ! J (20) CaCO 3 =-CaO+CO 2 . (21) MgCO 3 =MgO+CO 2 . 
 
 At the proper temperature for their formation the caustic lime and magnesia 
 in intimate contact with Si(>2 will both combine with it to form slags. 
 
 Action of Aluminum: Aluminum, in the form of alumina, A^Oa, and 
 alumina silicates, is found in ore, flux and fuel. Neither alumina nor its 
 silicates are reduced under the conditions that prevail in a blast furnace. 
 Al2O 3 , as already pointed out, may exert a marked influence upon the 
 fluidity and fusibility of the slag. 
 
 Action of Less Abundant Elements: Titanium, potassium, sodium, 
 zinc, arsenic, copper and chromium, are elements, a few of which are present 
 in very small amounts in the materials used in the Pittsburgh district. 
 Titanium enters the furnace as titania, TiC>2, combined with some base. 
 Titania is similar to silica, SiC>2, except that it is more difficult to reduce 
 at temperatures attainable in the blast furnace, and all but traces of it, 
 which is found in the iron, passes out with the the slag. Under the con- 
 ditions prevailing in the furnace, titanium exhibits a slight tendency to 
 combine with carbon and nitrogen to form titanium cyano-nitride. This 
 substance is sometimes found in the salamander on the hearths of furnaces 
 being repaired. Here, it occurs in the form of small cubes that have the 
 appearance of copper. The alkalies, soda and potash, are found in nearly 
 all blast furnace slags, and when they are present in the raw materials 
 to a considerable extent, they are partly volatilized and driven over 
 out of the furnace with the flue gases, from which they may be separated 
 with an installation of suitable apparatus. Zinc is a very troublesome 
 element when present in blast furnace material. Its compounds may 
 be reduced in the lower regions of the stack; but, if so, the zinc is vol- 
 atilized, driven upward by the blast, and oxidized to zinc oxide, which 
 condenses on the walls of the colder part of the flues and in time closes 
 up the passages to such an extent as to seriously restrict the flow of the 
 gases. Zinc oxide also tends to combine with the alumina in the fire 
 brick lining of the furnace, causing the brick to expand with consequent 
 evil, or even disastrous, results. Arsenic acts very much like phosphorus. 
 All of its compounds are reduced, and the resultant elementary arsenic then 
 combines with iron to form iron arsenide which dissolves in the metal. 
 Copper compounds are readily reduced, yielding metallic copper, which 
 alloys with the iron. Chromium is separated from its oxides only with 
 great difficulty in the blast furnace, an exceedingly high temperature and a 
 special slag being required for the reduction of its oxides. 
 
REACTIONS 167 
 
 The Reactions Within the Furnace: With these facts concerning the 
 properties of the various ingredients of the raw materials in mind, the 
 changes that take place in the blast furnace are easily understood. The 
 accompanying chart (Fig. 23) gives a graphic representation of these changes, 
 showing the relative weights and volumes of materials, the reactions and the 
 temperatures at which the changes take place and the final disposition of the 
 products. In studying this chart, however, one important fact should be 
 kept in mind. It is this: Owing to the conditions prevailing within the fur- 
 nace, very few, if any, of the reactions will be complete, that is, use up all 
 the material at hand at the location indicated. Thus, the first reaction, 
 showing the reduction of the ore to metallic iron with the deposition of carbon, 
 affects only a part of the ore and gas. This condition, with but one exception 
 holds for all these reactions. Even limestone, which will decompose com- 
 pletely into lime and carbon dioxide at 1000 if given sufficient time, will often 
 reach the tuyeres as calcium carbonate. The fact that iron and manganese 
 oxides are not completely reduced is established by their presence in the 
 slag. The one exception to this rule is phosphorus. Its compounds, down to 
 small traces, are completely reduced. 
 
 The explanation for these statements is to be found only in a careful 
 study of chemical laws in connection with the conditions prevailing in the 
 furnace. Such a study reveals the fact that the reactions in the upper part 
 of the stack of the furnace are subject to conflicting tendencies. Thus, 
 there are in constant contact with the solid substances FesC>4, FeO, Fe, and C 
 the gaseous substances CO and COs. Of these, FesO^ and CO2 are oxidizing 
 agents. FeO and CO may act as either oxidizing or reducing agents, while 
 C and Fe are reducing agents. With these substances in contact at any given 
 temperature and in any given proportions as shown on the chart the reversible 
 reactions would proceed in a given direction until equilibrium should be estab- 
 lished, and no further change would occur until either the concentration of one 
 of the reacting substances or the temperature should change. Then the re- 
 actions, subject to the law for mass action, would proceed in a direction that 
 would again establish equilibrium. But the slow downward movement of the 
 stock to regions of higher and higher temperatures, the presence of an excess 
 of carbon and the rapid upward flow of the gases, which has the effect of giving 
 a constant surplus of CO and of carrying CO2 out of the field of action, tend 
 to prevent the establishment of equilibrium, and to force the reactions 
 to proceed in a direction that will result in the final reduction of the iron 
 oxides, with the consequent oxidation of either the C or the CO. These 
 same conditions, however, which tend to reduce the oxides of iron, prevent the 
 complete oxidation of the CO to CO2, because the CO, passing so rapidly 
 over the stock, does not have time to become wholly oxidized, and the presence 
 of the reducing agents, Fe, FeO, and C, tend to oppose the formation of CO2. 
 The escaping top gases, therefore, always show a large content of CO much 
 in excess of the CO2 content. In modern furnaces, operating according to 
 the prevailing practice with respect to coke consumption, the relative volumes 
 
168 
 
 BLAST FURNACE 
 
 I50000CU.FT ia72ILBS. = 2357LBS. 
 
 GASES 
 
 2OOLBS. - 
 DUST 
 
 26 LBS. 
 SiO* 
 
 C0 
 
 IOOLB5. 
 Fc 3 Q4 
 
 I 
 
 2977 LBS. 
 CO 
 
 23LBS. 1.6 LB5. 
 FeO MnO 
 
 1076 
 
 STONE 
 
 I DISTANCE FROM f 
 
 | BOTTOM IN FT. 58 
 
 85 
 
 2162 
 COKE 
 
 4333 
 ORE MIX 
 
 97-4 
 
 CaCO, 
 
 3167 
 
 I 
 732 
 
 C 
 
 60 
 
 190.3 39.2 451.2 54.6 
 AI,C\ MnO SiO, FeS 
 
 4.7 2O.S* 
 
 75 
 
 70 
 
 _65_ 
 
 * ' r 
 
 g Fe g O^tflCO * 4 Fe +C->-7CO a 
 
 fc 
 
 <co ^ 
 
 /C * |C f O 
 
 Fe*04 +\CO =^FeO+jCO e 
 
 eo 
 
 3FeO C0 2 ^Fe 3 p 4 .C 
 Fe + C0 FeO'+ CO 
 
 50 
 
 45 
 
 -l4Los.H --|96 LeS.CO 
 CaCQ, >c 
 
 40 
 
 MgCO A -( 
 
 ,277Lss. 
 
 FeO+C * 
 
 30 
 
 MnOC 
 
 TAPPING H 
 
 25 CaOAlpOj ^iQ^ ^ 
 
 J 
 
 9 44 LBS RP^ 3760.6 LBS 
 
 c'a 
 
 Sfl08LBsCO z +766LB&C - 3574 LBS CO 
 2042LBS 0+766L85C *280SLB5 C O 
 
 >9.44 LBS H 2 ^132 23LBS.CO 
 FeS+CaO + C ^CaS t Ke + CO 
 
 1 22^.5 LBS. SLAG =4O3.4LBaSO 4 ,l.2LDifeo, i3iLBs.Mn O 
 
 5CO + 2 Fe 3 P 
 
 2240 LBS.PtG IRON = 21 05.6 LBS.Fe> e7LBS.C, 
 
 FIG. 23. The Making of a ton of Pig Iron. A diagram showing the raw 
 cbanges that take place therein. 
 
REACTIONS 
 
 169 
 
 Z3 LBS, 
 
 675 8 LBS. 
 
 6O6 LBS. 
 
 .5LB. 4LB5. 3.6 LBS. ILB. 4OLBS. 1.3 LBS. 
 Ca 3 P 2 O fl Al a Q 3 CaO MgO C FeS 
 
 WEIGHTS OF RAW MATERIALS. 
 
 OF CHIEF 
 
 /APPROX. TEMP. 
 DEGREES CENT 
 275 
 
 RELATIVE WEIGHTS OF CHIEF IMPURITIES. 
 
 375 
 
 SOME IRON SESQUIOXIDE is REDUCED ANO CARBON DEPOSITED. 475 
 SOME IRON SESQUIOXIPE is REDUCED TO MAGNETIC OXIDE:. 550 
 
 SOME MAGNETIC OXIPE is REDUCED TO FERROUS OXIDE. 625 
 
 SOME FERROUS OXIDE is REDUCED TO METALLIC IRON. 
 
 SOME IHQN MAV BE RFQXIDIZE.D AND CARBON DEPOSITED. 7OO 
 
 775 
 
 CARBON DIOXIDE MAY BE REDUCED BY IRON 
 OR FERROUS OXIDE. 
 
 MUCH OF THE CARBON DIOXIDE is REDUCED BY CARBON. 
 
 875 
 
 COMBINED WATER REMAINING is DECOMPOSED. 
 
 975 
 
 LIMESTONE is DECOMPOSED. 
 
 TOTAL LIME FROM ORE AND STONE = 576.4 LBS. 
 
 CARBON is ABSORBED BY SPQNGV IRON. 
 
 1075 
 
 REDUCTION OF IRON OXIDES is COMPLETED 
 
 1175 
 
 PART OF THE MAMGANOUS OXIDE is REDUCED. 
 
 1250 
 
 LIME. ALUMINA, AND SILICA UNITE To FOR.M SLAG. 
 
 1350 
 
 FUSION ZONE FOR ALL SUBSTANCES BUT COKE 
 
 1550 
 
 COMBUSTION ZONE, (OXYGEN AHDV/ATEK OF THE AIR 
 COMBINE 'WITH CARBON OFTHE.COKE TO FORM HYDROGEN 
 AND CARBON MONOXIDE.) 
 
 1700 
 
 000 
 
 NEARLY ALL THE IRON SULPHIDE is CONVERTED INTO CALCIUM SULPHIDE. 
 
 I90i3 LBS. AliO 3 , 539.4LB3. C&Q, ZZZLes-MgO, 474 Las.Ca S 
 
 CIUM PHOSPHATE is REDUCED TO IRON PHOSPHIDE. 
 SOME SILICA is REDUCEI? FORMING IRON SIHCIDE. 
 
 .a2.4Les.5i, 2O. E LBS. Mn. 4.13 LBS. R .67LB.S 
 
 \wtR NOTCH.. 
 
 materials and the products of the blast furnace; their relative weights and the 
 
170 BLAST FURNACE 
 
 of CO and CO2 are approximately 2 to 1. In the lower part of the stack, 
 the temperature is so high that CO 2 cannot exist in the presence of carbon, 
 and any oxide reduced in this region results in the gasification of a proportion- 
 ate amount of carbon. This direct reduction of oxide by carbon is the most 
 inefficient mode of reduction, because it absorbs much heat, as shown by re- 
 actions (8), (9), and (10), and robs the tuyeres of carbon needed for combus- 
 tion. This mode of reduction, then, is one the furnaceman strives to avoid so 
 far as possible. 
 
 Tracing the Materials Through the Furnace: The ore, limestone 
 and coke, upon being charged into the top of the furnace, come in contact* 
 with an ascending current of hot gases (temperature about 275 C.). The 
 first change that takes place is the physical one of drying. The hygroscopic 
 water, being first driven off and carried out of the furnace by these gases, 
 is then followed by the water of crystallization. The stock, with its inter- 
 stitial spaces filled with an ascending atmosphere containing the reducing 
 gas CO, starts to descend toward the bottom of the furnace and to regions 
 of higher and higher temperatures. At different levels, then, chemical 
 reactions peculiar to the temperatures of these levels will occur. At first 
 only the oxides of iron and carbon suffer change, and the first reaction to 
 occur is number (4), in which carbon deposition takes place at a temperature 
 as low as 300 C. A large part of the remaining iron oxide, in the presence 
 of both C and CO, is next reduced in successive levels and temperatures as 
 follows: 
 
 fc fpo 
 
 3 Fe 20 3 +|^ =W 2 +2 Fe 3 4 , begins at 450 C. 
 Fe 3 O 4 +< =< +3 FeO, complete at 600 C. . 
 
 FeO +co == \CO +Fe ' begins at 7 C - 
 
 At about 800C. the free iron is subject to re-oxidation by CO2, as is 
 also the compounds FeO and Fe 3 O 4 though to a less degree, the chief action 
 being represented by reaction (5). At 800 C., or a little above, the decom- 
 position of limestone takes place, thus : CaCO3=--CaO+CO2- This reaction 
 is complete at 1000 C. At 900 C., carbon reduces CO 2 to CO. thus: 
 C+CO2=2CO, so that CO 2 does not exist below the 60 foot level. From 
 this level the mixture is one of gangue, quick lime, coke, spongy iron and 
 varying amounts of unreduced ore, all of which descend to the fusion zone 
 with very little change, if the absorption of carbon by the iron and the 
 action of carbon on the unreduced ore be excepted. At this level, which 
 is located at the top of the bosh, the lime combines with some of the gangue 
 and, with a little unreduced iron oxide and manganese oxide, forms a part of 
 the slag. The slag, such as is already formed, and the iron, both now in 
 the liauid state, trickle down through the interstices of the coke to the 
 
REACTIONS 171 
 
 hearth, where they become separated by gravity, forming these two layers; 
 a lower or metallic layer containing all reduced substances and an upper 
 or slag layer containing all unreduced matter. Here, since these two 
 layers are in contact with each other and with carbon of the coke, which 
 probably extends to the bottom of the hearth, or at least to within a few 
 inches of the bottom, reactions (11), (12), (13), (14), (17), (18), and (19), 
 known as hearth reactions, occur. 
 
 Conditions Affecting the Amount of Silicon and Sulphur in the 
 Metal: Reactions 13 and 18 should receive special attention here, because 
 they affect the quality of metal and are subject somewhat to the control 
 of the furnaceman. Number (13), SiO2+2C=Si+2CO, depends upon two 
 conditions, namely, temperature and basicity of slag. High temperatures 
 favor the reaction, while basic conditions of the slag retard it. Reaction 
 (18), FeS+CaO+C=CaS+Fe+CO, is also subject to the same influences. 
 Therefore, the conditions which tend to raise the silicon in the iron will 
 lower the sulphur content, provided the high temperature is obtained 
 without the use of an excessive amount of high sulphur coke. In both 
 these cases the extent of the reactions is governed by time. The longer 
 the time the farther they will progress. This fact results in a difference 
 in composition between the first and last metal in the same cast. Since 
 the iron on the bottom of the furnace crucible is formed four or five hours 
 before that on the top of the layer at the time of tapping, these reactions 
 will tend to advance farther in the first than in the last iron formed under 
 normal conditions. The first of the cast will, therefore, usually be found 
 to contain a higher percentage of silicon and a lower percentage of sulphur 
 than the last. This first iron is called "hot iron" on this account. 
 
172 THE BESSEMER PROCESS 
 
 CHAPTER VII. 
 
 THE BESSEMER PROCESS OF MANUFACTURING STEEL. 
 
 SECTION I. 
 
 THE CLASSIFICATION OF FERROUS PRODUCTS. 
 
 Introductory: In beginning this chapter it is desirable to decide the 
 question as to what constitutes steel. Owing to the many varieties of iron 
 now classed as steel, a concise and wholly satisfactory definition is well 
 nigh impossible. Attempts have been made to restrict the usage of the 
 term, but without success, because in defining any term, the name must 
 be taken as it is used. Therefore, since an adequate definition of steel is 
 lacking, a brief resume of the commercial products of iron may be profitable. 
 In beginning this survey, it is to be born in mind that the basis for the 
 preparation of the various ferrous products is pig iron and that this 
 substance, a direct product of the blast furnace, represents the crudest 
 form of commercial iron. All higher grades are the products obtained by 
 different methods of refinement and the degree to which this refinement 
 is carried. The ferrous products may, therefore, be placed under two 
 classes, namely, pig iron and refined iron. 
 
 Pig Iron and Cast Iron: As pointed out in the preceding chapter, 
 pig iron may vary, or be varied, very much in chemical composition and 
 constitution. This variation gives the different grades of pig iron and 
 determmes the use to which the metal can be applied. On cooling, the 
 crude forms first undergo a slight expansion, which is followed by a slight 
 contraction. This fact makes it particularly suitable for mould casting, 
 in which form it is called Cast Iron. Cast iron offers a high resistance 
 to crushing, but all forms of unrefined iron are lacking in tenacity, 
 elasticity and malleability. 
 
 Malleable Cast Iron: In the second class will be found a series of 
 products, which may be classified according to the initial method of refine- 
 ment. This refinement may be brought about in two ways, namely, one 
 in which the metal remains in the solid state throughout the process and 
 another in which the purification involves fusing the metal. Malleable 
 cast iron is an example of the first method. Products of this class are 
 obtained from crude pig iron of a certain composition chemically, which, 
 upon being cast into the desired form, is subsequently subjected to a com- 
 bined anneaMng and oxidizing process by which the malleability is 
 
WROUGHT IRON AND STEEL 173 
 
 developed. In carrying out the process, the clean casting is packed in iron 
 oxide and subjected to a temperature of about 700 C. for three or more 
 days, when it is allowed to cool in the furnace very slowly. By this treat- 
 ment the greater portion of the combined carbon is converted into graphite 
 that takes the form of very minute particles evenly distributed throughout 
 the casting, and so does not have the weakening effect that flakes of graphite 
 have. Some carbon, say twenty per cent of that originally present in the 
 iron, is oxidized and eliminated from the metal. It is said that a slight 
 reduction in the sulphur content also takes place. 
 
 Wrought Iron: At present there are two classes of iron products 
 recognized as being produced by the method of purification by fusion. To 
 these are given the names of wrought iron and steel. Wrought iron, as 
 indicated in the study of the blast furnace, may be produced directly from 
 the ore. This method, however, has now been superseded by the indirect 
 process, in which pig iron is melted in a reverberatory furnace, called a 
 puddling furnace, the hearth of which is lined with iron oxide. This treat- 
 ment results in the oxidation and consequent removal, from the metal, of 
 all but small amounts of carbon, silicon, manganese, phosphorus and sulphur. 
 The purification brings about a rise in the fusion temperature of the iron 
 above that of the furnace, and at this point the metal is removed from the 
 furnace in the form of pasty balls in which more or less slag is incorporated. 
 As much as possible of this slag is at once removed by hammering or 
 squeezing, after which the bloom thus produced is rolled into muck bar. 
 In this form it may be converted into steel as noted below, or subjected 
 to further treatment to produce merchant bar. Wrought iron is soft, 
 tough and very malleable. It welds easily, and is characterized by a 
 fibrous structure, due to the presence of the intermingled slag and the 
 mechanical treatment it receives. Various modifications looking to 
 improvement in the process of producing wrought iron have been devised. 
 
 Steel is the term applied to all refined ferrous products not included 
 under the classes described above. It is distinguished from pig iron by 
 being malleable at temperatures below its melting point, from malleable 
 iron by the fact thatit is initially malleable without treatment subsequent 
 to being cast, and from wrought iron by the circumstance of its manufacture. 
 In the case of wrought iron, the metal was in a fused state during a part 
 of the purifying process only, whereas the purification of pig iron to produce 
 steel takes place at a higher temperature, and the metal remains in the 
 molten state throughout the period of purification. From a chemical 
 analysis it is practically impossible to distinguish wrought iron from soft 
 steel, but the one, being obtained in a state of complete fusion and free 
 from slag, may exhibit physical properties very different from the other, 
 which is obtained in a semi-fused state and retains small amounts of the 
 slag incorporated with it. Between pig iron and steel, however, a marked 
 difference in chemical composition as well as in physical properties is 
 
174 
 
 BESSEMER PROCESS 
 
 observed. All three substances show a wide variation in chemical com- 
 position. The following table may be studied with profit. 
 
 Table 28. Chemical Relations of Pig Iron, Wrought 
 Iron and Plain Steel. 
 
 PER CENT. OF 
 
 Name 
 
 Iron 
 
 Carbon 
 
 Manganese 
 
 Sulphur 
 
 Phosphorus 
 
 Silicon 
 
 Pig Iron. 
 
 91 94 
 
 3.50 4.50 
 
 .502.50 
 
 .018 .100 
 
 .030 1.00 
 
 .25 3.50 
 
 Plain 
 
 
 
 
 
 
 
 Steel. . 
 
 98.1 99.5 
 
 .071.30 
 
 .30 1.00 
 
 .020 .060 
 
 .002 .100 
 
 .005 .50 
 
 
 
 
 (.03 .10 
 
 (.120) 
 
 
 
 
 
 
 as cast) 
 
 
 
 
 Wrought 
 
 
 
 
 
 
 
 Iron. . 
 
 99.0 99.8 
 
 .05 .25 
 
 .01 .10 
 
 .020 .100 
 
 .050 .20 
 
 .02 .20 
 
 This table would indicate that wrought iron is not the purest form of 
 commercial iron, as is often asserted. However, in wrought iron part of 
 the manganese, sulphur, phosphorus and silicon shown in the table above 
 may be derived from the incorporated slag, in which case they would exert 
 little influence upon the metal itself. 
 
 Methods of Making Steel : Formerly it was possible to make a much 
 finer distinction between wrought iron and steel than that indicated above. 
 Prior to 1856, there were but two kinds of finished steel; they were known 
 as shear steel and crucible, or cast, steel. Both were at that time manu- 
 factured from blister steel made by the cementation of wrought iron. Shear 
 steel was made by piling and welding blister steel bars into faggots, which 
 were then forged or rolled into strips or bands suitable for cutlery. Crucible 
 steel was produced by melting blister steel and scrap in graphite crucibles, 
 casting the fluid metal into moulds, and then forging these small ingots 
 into bars of the required size and shape. These products were distinguished 
 from wrought iron by the fact that they could be hardened and tempered, 
 and this property was, therefore, made the basis for a definition of steel. 
 But the introduction of the Bessemer and open hearth processes, with 
 their numerous grades of products, many of which can also be hardened 
 and tempered and all of which are quite different from wrought iron, neces- 
 sitated a revision of this definition for steel, because, lacking a better 
 name, the term steel was applied to the products from the new processes 
 also. Then, still more recently, the advent of the electric furnace added 
 another variety to the ferrous metals. Finally, the cementation process 
 
PRINCIPLES 175 
 
 has been superseded almost entirely by the crucible process, and the intro- 
 duction of alloying elements has made a definition based on the purity of 
 the metal inapplicable. It appears therefore that the only general definition 
 for steel that can be offered is one based on the method of refinement. 
 On this basis, then, steel is a ferrous metal, derived from pig iron or 
 wrought iron, which has been subjected to a refining process by complete 
 fusion. 
 
 General Principles of the Methods of Purifying Pig Iron: From a 
 commercial standpoint, the fundamental principle by which the purification 
 of pig iron is effected is that of oxidation in all cases, excepting the electric 
 furnace, which employs both oxidation and reduction. For the purpose of 
 purifying by oxidation two substances are available. These substances are air 
 and iron oxide, the application of which requires different types of apparatus. 
 The two chief methods of purification, then, represent attempts to meet 
 these requirements. These methods are known as the pneumatic, or 
 Bessemer, and the open hearth, or Siemens' processes. In both, the puri- 
 fication may be brought about by oxidation alone, in which case they are 
 called acid processes, or by oxidation in conjunction with strong bases, 
 such as lime, when they are designated as basic processes. By the first 
 class of process, only the elements carbon, silicon and manganese are 
 removed from the iron, while the second method also removes phosphorus 
 and, to a limited extent, sulphur. The basic Bessemer process has been 
 named after its inventors, the Thomas=Gilchrist, and the basic open 
 hearth is generally spoken of as the basic. Each of the five purifying 
 processes mentioned above, namely, the acid Bessemer, the basic Bessemer, 
 the acid open hearth, the basic open hearth, and the electric, produces 
 steel having certain peculiar properties, and with the exception of the 
 electric process, each requires pig iron of a composition different from 
 any of the others. Owing to the composition of iron ores available in 
 this country, the pig iron produced is best adapted for treatment by the 
 basic open hearth or the acid Bessemer process; hence, these are the 
 leading methods employed. 
 
 SECTION II. 
 
 PRINCIPLES AND HISTORY OF THE BESSEMER PROCESS. 
 
 Principles of the Process: Of all the processes for the purification 
 of pig iron, the Bessemer is the simplest. Essentially, it consists of blow- 
 ing air under pressure through a bath of molten metal contained in a 
 vessel constructed of proper refractory materials, whereby a portion of 
 the iron, all of the silicon and manganese, and then the carbon are suc- 
 cessively oxidized. The first three elements, upon combining with oxygen, 
 go to form a slag, while the carbon is eliminated in the form of the gases, 
 carbon-monoxide, CO, and carbon dioxide, CO2- As noted elsewhere, the 
 
176 BESSEMER PROCESS 
 
 oxidation of these elements are exothermic reactions, from which the heat 
 required to maintain the metal in the liquid state is derived. Since steel 
 produced in this way, without recarburization, contains deleterious oxides 
 which render it unfit for use, it is necessary to add deoxidizers to the metal 
 after blowing. This fact was not realized at first, and the history of the 
 process serves to emphasize its importance. 
 
 Some Incidents Connected with the Early History of the Process: 
 
 The history of this process also furnishes an example of the way in which 
 a method is developed, and illustrates the fact that the perfecting of a 
 process is seldom accomplished by one mind alone, but by many minds 
 thinking toward one goal. The method was almost concurrently, but 
 independently, originated by two men: one, an American named Wm. Kelly 
 of Eddyville/ Ky.;the other, an Englishman, the illustrious inventor, Henry 
 Bessemer. Although Kelly did not apply for patents until 1857, almost 
 two years after Bessemer's English patent was granted, his application 
 was allowed on grounds of priority, because he was able to prove that he 
 had worked out the idea as early as 1847. In the same year that he made 
 application for patents, Kelly erected a tilting converter for the Cambria 
 Steel Works at Johnstown, Pa. This vessel is still preserved. Lacking 
 financial means, however, Kelly was unable to perfect this invention, and 
 after much litigation with the Bessemer interests, a settlement was made, 
 whereby Kelly dropped out of the game. Bessemer, on the other hand, 
 in addition to conceiving the idea and putting it to trial, continued his 
 experiments in the face of great difficulties and many failures until he had 
 brought the process to a high degree of perfection. At first Bessemer 
 accidently employed only Swedish iron, which had a low phosphorus and 
 a high manganese content, and was very successful in converting it. Then, 
 it having been adopted by many manufacturers, the process failed when 
 applied to English irons which were high in their phosphorus and low in 
 their manganese content, and prejudice and opposition to the method 
 became so great among steel makers that, in order to save his process, 
 Bessemer was obliged to build a steel works himself. His plant, built at 
 Sheffield, began to operate in 1860. 
 
 Importance of Manganese : At the Sheffield works the process was used 
 at first to produce high carbon steels from Swedish pig iron only, because low 
 carbon steels, obtained by subjecting the metal to a full blow, were almost 
 invariably hot short, even when made from the excellent Swedish iron. 
 This defect was later overcome by the addition of manganese in the form 
 of spiegeleisen, the beneficial effects of which were first recognized by 
 R. Mushet as early as 1856. With the adoption of the use of manganese, 
 mild or soft steels produced by the process came into so great demand 
 that the former practice in blowing was abandoned in England, though it 
 is still employed in Sweden. The first Bessemer plant in this country was 
 erected in 1867. 
 
HISTORY 177 
 
 Thomas and Gilchrist: The removal of phosphorus by the use of a 
 basic lining and the addition of lime ta the bath was first conceived by 
 Thomas, who made known the success of his scheme in 1878. In the develop- 
 ment of this process, Thomas was assisted by his cousin, the chemist 
 Gilchrist, hence the name Thomas-Gilchrist. 
 
 Other Improvements : While the process was highly developed along 
 mechanical lines by Bessemer, himself, it remained for Alexander Holley, 
 an American Engineer, to introduce many improvements in the erection of 
 Bessemer plants. The most important of these was his invention of the 
 detachable bottom, which will be described later. Another important 
 invention was that of the hot metal mixer, since it furnished a ready supply 
 of molten metal of fairly uniform composition, thus allowing the process 
 to be operated much more rapidly and economically. This vessel, also to 
 be described later, was the invention of W. R. Jones of the Carnegie 
 Steel Company's Edgar Thomson Plant at Braddock. 
 
 Plan of Study: Before beginning a more minute description of the 
 process as it is carried on with these modern improvements, it is well to 
 note that the details of the operation will vary much in different plants 
 as well as in different countries. The description, therefore, must be either 
 very general in character, or be restricted to some one plant which will 
 suffice as an example for all. For the present purpose, it is best to follow 
 the latter course, and the Carnegie Steel Company's plant at the Edgar 
 Thomson Works is selected to serve as such an example. General features 
 of great importance may then be introduced in connection with the dis- 
 cussion of the various topics. At these works, the product from eleven 
 modern blast furnaces is available to supply both the open hearth plant 
 of fourteen 90-ton furnaces and the Bessemer plant of four converters, the 
 maximum capacity of which is twenty tons. 
 
 SECTION III. 
 
 EQUIPMENT AND ARRANGEMENT OF THE EDGAR THOMSON PLANT. 
 
 The Converter House: The four converters are arranged in a row 
 along one side of the converter building, which is located in one corner of 
 the works in close proximity to the rail mills. The converters, being of 
 the concentric type, tilt in two directions, in one direction for charging 
 and in another for pouring. On the charging side of the vessels the building 
 is erected three-story fashion. The ground floor extends under the 
 converters and offers space for the removal of bottoms, slag, etc. The 
 second floor, designated as the charging floor, is on a level with the 
 trunnions. From this floor all molten materials are charged into the 
 vessels. From the third floor, called the scrapping floor, all cold materials 
 are charged. Serving the four converters on the pouring side, are two 
 jib cranes for handling the steel ladles into which the metal is poured after 
 
178 
 
 THE BESSEMER PLANT 
 
EQUIPMENT 179 
 
 each blow. These cranes may be swung around so as to deliver the steel 
 to stationary teeming tables located in front of the teeming platform which 
 extends along the side of the building opposite the converters. Between 
 this platform and the tables, which support the ladles during the teeming 
 process, is laid a narrow gauge track, along which the ingot moulds, set 
 upon small cars, are moved during and after the teeming. This motion 
 in front of the platform is imparted by means of dogs attached to three 
 hydraulically operated cylinders that lie between the rails of the tracks. 
 All the operations of the converters and jib cranes are controlled from 
 two pulpits in opposite corners of the building and above the teeming 
 platform. Just back of the converter building and inter-communicating 
 with it through an open side beneath the charging floor, is the bottom 
 house, equipped with over-head cranes and buggies for handling bottoms. 
 Here the frequent repairs to bottoms are made. As these repairs must be 
 made with wet refractories, new bottoms require thorough drying before 
 being put into service. For this purpose six drying ovens, each large 
 enough to contain two bottoms, are provided. Built against one end of 
 the converter house, like the wing of a building, is a cupola house. It also 
 is inter-communicating with the converter building on its charging floor. 
 In the angle formed by the cupola and bottom houses are kept the stores 
 of cold pig iron, spiegel, and ferro manganese, while beyond these 
 will be found a building in which is housed suitable rock crushing machines 
 and Chilean mills for crushing and mixing the refractory materials used 
 in making up the various mixtures required for bottoms and repairs about 
 the plant. 
 
 The Larger Accessories: The cupolas, blowing engines, mixers and 
 strippers, are each separately housed and are located at various distances 
 from the converter house. All these accessories play a very vital part in 
 the process and in the operation of the plant, hence are deserving of special 
 consideration. 
 
 The Cupolas: In former times, when converter plants were operated 
 as independent units, detached and far removed from the blast furnaces, 
 cupolas were used to melt the cold pig iron preparatory to charging. At 
 this plant furnaces are used only for melting the pig iron and spiegel mixtures 
 employed as recarburizers. In construction a cupola is cylindrical in shape 
 and resembles a miniature blast furnace. Those at Edgar Thomson Works 
 are eight feet in diameter outside and some twenty feet in height, measur- 
 ing from the mantle to the charging doors. Like the blast furnace, there 
 is an opening at the bottom of the hearth for tapping out metal and another 
 for slag. Above these openings are inlets for ten tuyeres, through which 
 cold air, under a pressure of six to ten ounces, is blown by fans. Near 
 the top are the two large openings or doors for charging, directly opposite 
 each other and opening upon the charging floor. A little above these 
 openings, a contraction of four or five feet in the outside diameter of the 
 
ISO BESSEMER PROCESS 
 
 shaft forms the stack, also about twenty-four feet high, for the escape 
 o-f gases. The entire furnace rests vertically on a mantle which is 
 supported by a number of columns fixed upon a firm foundation. This 
 construction permits the use of the drop bottom, which facilitates the 
 removal of worn-out linings, the frequent repairing required by the lower 
 lining, and the rapid discharge of the stock in case of emergency. For the 
 sake of economy, the lining, or wall, is made of different materials. The 
 upper wall, for a distance of about four feet below the charging doors, is 
 made of fire brick and is nine inches thick. Below this brick work, the 
 wall is built of firestone and is gradually increased in thickness, forming a 
 kind of bosh above the hearth, the walls of which are about eighteen inches 
 thick. No attempt is made to cool these walls, so it is customary to 
 back up the firestone of the hearth-wall with fire brick in order to safe- 
 guard the steel shell. The shell, like that of the blast furnace, supports 
 and re-enforces the masonry. It is made of steel plates, which are riveted 
 together. 
 
 Charging the Cupola: The cupola charge is composed of coke, spiegel 
 and pig iron, in alternate layers of metal and coke, to the last of which 
 is added sufficient limestone to flux the ash. When the orders call for steel 
 with a high content of silicon, ferro-silicon is also added to the charge. 
 The ratio of coke to metal varies a little. At all times the amount of fuel 
 will be as small as possible, both for the sake of economy and to exclude 
 sulphur and phosphorus, which are absorbed by the metal, as much as 
 possible. Sulphur in the charge does not result in a rise in the sulphur 
 content of the molten spiegel, but in a waste of the^manganese, which reacts 
 with the ferrous sulphide to form manganous sulphide, and goes off wit/h the 
 slag. Ordinarily, the coke will be about 8% and the stone about 2^% of 
 the metallic charge. 
 
 The Blast: Just outside the converter house, on the pulpit side, 
 is the blowing room. Here are located three steam blowing engines of the 
 compound vertical type, which create the air blast for the converters. The 
 bla,5t from 'these engines is delivered into a common main, through which 
 it is conducted into the converter building, where it is distributed through 
 a manifold to lines leading separately to the four vessels. The admission 
 of air to the vessels and its pressure are nicely regulated by a system of 
 valves under the control of the blower. Thus, the pressure on the main is 
 maintained at about 25 pounds per square inch by means of a blow-off valve, 
 which is used to regulate the pressure while the vessels are charging or 
 pouring. In case some of the converters are not being operated, one or more 
 of the blowing engines is stopped. By means of a second valve, operated 
 from the pulpit by a screw control, a pressure of 18 to 20 pounds per 
 square inch is maintained on the line leading to each vessel. Under this 
 pressure the blast may be almost instantaneously admitted to or shut 
 off from the vessel by means of a third valve of the butter-fly type. A 
 fourth valve provides a means by which steam may be admitted to the 
 
EQUIPMENT 181 
 
 blast line as required. The limits of blast pressure to the converter are 
 about 10 and 25 pounds per square inch. The lower pressure is just 
 about sufficient to keep the metal out of the tuyeres in a normal charge, 
 while if the higher pressure be exceeded, large amounts of metal are blown 
 out of the converter. 
 
 The Mixers: The supply of molten pig iron is obtained from two 
 200-ton hot metal miners located near the blast furnaces and some three 
 hundred yards from the converter mill. They are large vessels constructed 
 of steel plates riveted together to form a shell, which is lined with silica or a 
 good grade of fire brick. The vessels at this plant represent the oldest type. 
 They have a rectangular horizontal section and discharge the metal by 
 tilting. This type has a slightly arched roof and a bottom which slopes 
 from the front or pouring end toward the rear. The axis of rotation is 
 located at the bottom near the center line of the vessel. Molten iron 
 from the blast furnaces is conveyed to the mixer in tipping ladles, from 
 which the metal is poured into the mixer through an opening in its top at 
 the rear end. In the opposite end another opening, provided with a spout, 
 permits the drawing off of hot metal as required by merely tilting the 
 mixer, thus permitting the metal to be weighed with a fair degree of exact- 
 ness. These mixer's are not provided with gas burners, as is customary, 
 for very little heat above that held by the metal is ever required to keep 
 the contents molten. 
 
 Importance of the Mixer: The hot metal mixer is almost indis- 
 pensable to a modern Bessemer plant, or for that matter, to any steel 
 making plant. Primarily the mixer serves as a storage place for the hot 
 metal from the blast furnace, and in performing this function bestows great 
 benefits. Thus, not only is the heat from the hot pig iron conserved, but the 
 metal delivered to the vessel, or vessels, is of a more uniform composition 
 than could be otherwise obtained. Again, since the capacity of the modern 
 mixer permits it to contain casts from several furnaces, iron low in some 
 elements may be mixed with some that is high in the same ingredients; 
 and so it is possible to extend the chemical limits of the iron receivable. 
 Mixers have been constructed of various shapes and sizes. Their capacities 
 will range from 150 to 1200 tons, but the tendency in all modern construction 
 is toward the larger size. Purification of the metal is said to take place 
 to a slight extent in the mixer, sulphur being the chief impurity removed. 
 The reduction in sulphur, however, is only noticeable when the manganese 
 content of the iron is high. This removal is at all times so small as to be 
 of minor importance. 
 
 The Stripper: One of the most efficient and economical inventions 
 contributed to the steel business is the stripper. As its name indicates, 
 it is a device whereby the moulds are pulled, or stripped, from the ingots 
 after the metal has cooled sufficiently to form a solid shell on their outside 
 surfaces. Those at the Edgar Thomson Works are of a late type and are 
 
182 BESSEMER PROCESS 
 
 electrically operated. A stripper of this type is in the form of a strong 
 over-head crane, from which is suspended a vertical arm, provided, in place 
 of a hand, with two jaws that fit over lugs cast on either side and near the 
 top of the mould. Operating between the jaws is a ram, or plunger, capable 
 of exerting pressure on the top of the ingot, while it is beingstripped, sufficient 
 to balance the pull. In stripping an ingot, the jaws engage the lugs and 
 exert a powerful pull upward, while the ram, having been inserted through 
 the top of the mould, holds the ingot on the stool till the mould is loosened. 
 The mould is then raised high enough to clear the ingot and placed upon 
 an empty car standing, ready to receive it, on a track next and parallel to 
 that on which the stripped ingot stands. Electric strippers, owing to the 
 fact that they are travelling, possess a decided advantage over the older 
 type of hydraulically operated machines, which are stationary. 
 
 The Casting Equipment includes the teeming ladles, ingot moulds, 
 stools, and cars. The teeming ladle, which acts as a container for the 
 finished steel while casting, is a large cup-shaped vessel made of steel and 
 lined with a few inches of "ball stuff." As slag is liable to spoil the ingots 
 if allowed to flow into the moulds, steel cannot be poured from a vessel 
 by tipping, but must be teemed from a small hole in the bottom. For 
 opening and closing this hole, the vessel must be fitted with a stopper 
 that can be operated from the teeming platform. This stopper consists of 
 a steel rod, protected with fire-clay sleeves, to the lower end of which 
 is fastened a stopperhead, made of plumbago bonded with clay, that fits 
 neatly into a nozzle placed in the bottom of the ladle. The upper end of 
 this stopper is fastened to a goose neck that fits over a vertical sliding 
 bar attached to the outside of the ladle. This bar is provided with a lever 
 by which it may be raised or lowered, causing a like movement of the 
 stopper. To guard against a leaking, or "running," stopper the nozzle 
 may be filled with dry sand or loam, which is held in place by a sliding 
 plate on the outside. When the ladle is ready to teem, this sand is easily 
 punched out of the nozzle after removing the plate. 
 
 The Ingot Moulds into which the finished metal is teemed are made 
 of cast iron and may be of almost any convenient form and size to suit 
 the respective blooming mills. At these works, the standard moulds are 
 about 6 feet high, have a square section of 23% inches at the bottom, with 
 corners slightly rounded, 'and taper sufficiently to allow the mould to be 
 stripped readily from the ingot. The moulds are open at both ends, and, 
 when ready for teeming, rest, big end down, on heavy cast iron plates, called 
 stools. The stools are mounted in twos on small cars, or buggies, which are so 
 constructed that their sides form aprons that protect both the track on which 
 the cars run and their own running gear from splattering by hot metal during 
 the teeming of the metal from the steel ladle. The care of the moulds is 
 very important, since defects here are very likely to show up in the finished 
 material after rolling. Their sides must be kept smooth and clean, and 
 the teeming must be done so as to avoid splattering their sides, if possible. 
 
CONVERTER CONSTRUCTION 183 
 
 After being stripped, the moulds are inspected and, if their condition is 
 satisfactory, may be used again. So, after cooling to a point where the 
 hand may be held against them, they are cleaned, then sprayed inside and 
 around the tops with a clay wash to prevent the steel from sticking, and 
 marked for size of ingot required. As 'soon as the clay wash is dry, the 
 mould is ready to receive the molten steel. Usually about seventy ingots 
 may be cast in one mould before it is scrapped. 
 
 SECTION IV. 
 
 CONVERTER CONSTRUCTION AND REPAIRS. 
 
 General Features Pertaining to Converters: As to the form and 
 size of converters, methods of admitting the blast, and removing the steel 
 after blowing, there are several possible arrangements, and converters have 
 undergone many modifications. Thus, as was originally the plan, the air 
 might be admitted horizontally through the wall of the vessel near the 
 bottom, in which case the vessel would be of the side=blowing type; or 
 a blast of air will be forced upward through openings in the bottom of the 
 vessel, to which method the term bottom blown is applied. To facilitate 
 the charging of materials into the vessels and the removal of metal from 
 them, converters are now always constructed so that they may be rotated 
 on their shorter axis through arcs of varying size. Such converters are of 
 the tilting type. The first vessels were of the fixed type, the metal being 
 tapped through a hole in the wall at the bottom. In both size and shape, 
 vessels still vary much. The capacity will range from 5 to 25 tons. The 
 vessels at Edgar Thomson are 11 feet in diameter, outside and measured 
 along their axis of rotation, and almost 18 feet long. As to form, converters 
 were at first somewhat pot-shaped. Attempts to design vessels that would 
 retain heat and prevent the ejection of materials has led to two general 
 forms, each with its own advantages. In both forms the upper diameters 
 are shortened, forming the nose of the vessel and leaving a small opening 
 in the top, which forms the mouth. The body may retain the form of 
 the cylinder, as in the straight=sided type, or be narrowed at the bottom 
 also, in which case the body has a curved contour and somewhat resembles 
 an egg in shape. The mouth may be located at the top concentric with 
 the bottom and in a plane parallel to it, or it may be placed to one side, 
 in which case the opening lies in a plane at an angle to the bottom and is 
 then called eccentric. The vessels at Edgar Thomson Works are all of 
 the bottom blown, curved body, tilting, concentric type. 
 
 Parts of Converter: For convenience in constructing the vessel to 
 allow for contraction and expansion and for making repairs later, these 
 converters, as are all bottom blowing types, are constructed in three 
 separate parts, known as the nose, the body, and the bottom. The shell 
 for each of these parts is made of heavy steel plates, all firmly riveted 
 
184 BESSEMER PROCESS 
 
 together. The nose section is bolted to the body, but the bottom is held in 
 place against the lower edge of the body by linked key bolts. The links of 
 these key bolts fit over lugs on the body, while the key bolts themselves 
 fit between lugs on the bottom, making it easy to key the two parts 
 firmly together. When it is necessary to replace an old bottom with a 
 new one, the keys can be very quickly knocked out or driven in with 
 sledges in the hands of the workmen. The shell for the body is, itself, 
 made up of three parts, known as the nose section, the journal section, and 
 the shoulder section. The journal section is made up of a heavy band 
 to which the two trunnions that support the vessel are attached. All 
 these parts are firmly bound together by a great number of long key bolts 
 attached to the shoulder and nose sections, respectively. The trunnions 
 rest on bearings in a frame work which is supported by cast iron columns. 
 On the end of one of the trunnions, both of which are hollow, is the 
 connection, made through a packed joint, to the blast line. From capped 
 openings on this same trunnion between the bearing and the vessel, a 
 copper goose neck leads to the bottom of the vessel, thus forming a 
 continuous passage for the blast from the main, which is stationary, to the 
 wind box, which must move with the bottom of the vessel. To the other 
 trunnion is attached a pinion which meshes with a toothed rack that slides 
 horizontally. By means of a double acting hydraulic cylinder, the piston 
 of which is connected to this rack, the vessel may be rotated through an 
 arc of 270, the pinion and rack being geared so that the vessel may be 
 completely inverted for dumping slag or relining the vessel. All this 
 mechanism is carefully covered to protect it from slag, dust and other dirt. 
 
 Lining of the Converter: The lining for the shell may be composed 
 of any first class silicious refractory material. At most works a highly 
 silicious sandstone, known as firestone, is used, while a mixture, composed 
 of about five parts crushed ganister and one part best quality fire clay and 
 called ball stuff, serves as a kind of mortar. The lining varies in thickness 
 from ten to sixteen inches for the different parts of the vessel, being 
 thickest on those parts subject to the greatest wear. When lining a new 
 vessel or relining an old one, the bottom is detached, the vessel is inverted 
 and the lining is begun in the nose. The method pursued in starting the 
 lining will then depend largely upon the materials available and the shape 
 of the vessel. At the Edgar Thomson Works, the customary procedure is as 
 follows: A wooden frame, some five feet square and with a hole in the center 
 of- the same shape and size as the mouth of the vessel, is laid on suitable 
 cross pieces and then suspended from the vessel so as to press firmly against 
 the nose and in such a position that the hole is superimposed upon the 
 mouth. In this way a ledge upon which to begin the wall is formed. Upon 
 this ledge is placed a three inch layer of ball stuff, which is followed by a 
 course of large, flat, undressed firestone, set in on edge. All the inter- 
 stices are rammed full of wet ball stuff, so that the stones are securely 
 keyed into place and the side and top present a smooth surface. Upon this 
 
CONSTRUCTION OF THE CONVERTER 185 
 
 nose wall, which is about sixteen inches thick and thirty inches high, 
 the body wall is built. It consists of two courses. A thin course of split 
 brick is laid next to the shell, while within this, the inner course, about 
 twelve inches thick, is built up of rough blocks of firestone laid in a 
 mortar of ball stuff. The stones for the top course of this wall are cut 
 to shape and keyed in so as to hold the wall in place when the vessel is 
 righted and also to form a smooth joint, or shoulder, against which the 
 bottom is to fit. The lining is now completed by plastering the interior of 
 the vessel with ball stuff, after which it is carefully and thoroughly 
 dried. The coat of plaster, aside from giving a smooth surface, protects 
 the stone and overcomes its tendency to spall. To prepare the vessel for 
 use, the bottom is put on, the vessel is inclined, and then heated to a high 
 temperature with natural gas fires in the vessel itself. In case of a 
 shortage or absence of gas, coke or wood may be substituted for the gas. 
 With careful patching this part of the lining may last for several weeks, 
 or even months, of continuous running. 
 
 The Bottom of the converter warrants special mention. It is the part 
 of the vessel subject to the greatest wear and seldom lasts longer than 
 twenty heats, when it must be removed for repairs and replaced by 
 another. This change can be made with a delay of less than twenty 
 minutes, and is carried out in the following manner: The bottom of the 
 wind box is removed while the vessel is pouring, then as soon as the slag 
 is dumped, the converter is righted, and a small but strongly built truck 
 provided with a hydraulic jack, or lift, is run beneath it. The water 
 connection having been made with the hydraulic cylinder, the pressure is 
 applied to the jack, which raises a small table against the bottom. In some 
 plants the jack is placed beneath the track, in which case the whole truck 
 is raised. The keys are next knocked out, which leaves the bottom free 
 to descend with the table or the truck. The truck is then pulled into the 
 bottom house, where an overhead crane picks up the bottom and carries 
 it to one side. By reversing this procedure, a new bottom is soon in 
 place, and the converter is ready for charging. 
 
 Relining the Bottom : The repairing of the old bottom is immediately 
 begun. What remains of the old tuyeres and filling is quickly cooled with 
 water, and that on the bottom is loosened with suitable tools, when it may 
 be removed from the bottom by dumping it with the crane. The removal 
 of this material makes it easier to inspect the construction of the bottom. 
 The shell is made of heavy steel plates riveted together in the shape of 
 a shallow bowl with an open bottom. Closing this opening from within 
 the bowl, is the false bottom, a flat circular casting, with openings through 
 which the tuyeres may be inserted. It is a little larger in diameter than 
 the opening which it closes, thus making it unnecessary to fasten it in any 
 way. It supports the bottom stuff in which the tuyeres are packed. 
 Covering this same opening from without is the tuyere plate, a similar 
 casting containing bevelled openings into which the tuyeres fit when in 
 place. This plate is prevented from making a tight joint with the bottom 
 
186 BESSEMER PROCESS 
 
 by means of the splice plates that hold the riveted plates together. Thus, 
 an open space about one inch in depth is left between the tuyere plate and 
 the false bottom. The plate forms the top of the wind box, the two being 
 firmly bolted to each other and to the bottom with the same bolts. The 
 side of this wind box is a large casting, oval in shape, and about twelve 
 inches in depth. The bottom of the box is a steel plate which is firmly . 
 keyed to the casting to make an almost air tight joint when the vessel is 
 blowing. Connecting the wind box with the interior of the bowl, are 
 nineteen to twenty-one circular bevelled holes, through which the tuyeres 
 are inserted. The tuyeres are cylindrical bricks, flared for a distance of 
 about six inches from one end. They are about thirty inches long, seven 
 inches in diameter, and each one contains about twelve holes, one-half inch 
 in diameter and extending longitudinally. To place a tuyere, the flare is 
 covered with a mortar, composed of fire clay and Portland cement, and 
 the tuyere is inserted upward through the opening in the bottom, where 
 it is held in place with clamps until the filling has been put in. When all 
 the tuyeres have been thus placed in position, the top of each is covered 
 with a metal plate to keep dirt out of the tubes, some bottom stuff 
 is placed on the bottom in the space around the tuyeres, and on this 
 large tiles are set in as reinforcement to the tuyeres. The space remaining 
 about the tuyeres and brick is then tamped full with more of the bottom 
 stuff, which is a moist mixture composed of 28 parts crushed ganister, 12 
 parts blue fire clay, 3 parts ground brick bats, 3 parts old bottom stuff 
 and 4 parts coke dust. The bottom is then pushed into a drying oven, 
 fired with coke oven gas, and carefully dryed, then finally baked for several 
 hours. The time required to dry and bake a bottom properly is about 
 forty-eight hours, though bottoms will often be used at the end of thirty- 
 six hours. Upon being required for use, it is withdrawn from the oven, 
 and a heavy layer of a stiff clay mixture is placed around the upper edge to 
 form a tight joint with the shoulder of the vessel when the bottom is in 
 place. The mortar is then sprinkled heavily with coke dust, after which 
 the bottom is put into service as previously described. The function . of 
 the coke dust is to prevent the bottom from cementing itself to the 
 shoulder joint. When in service the position of the bottom is such that 
 the long axis of the oval wind box is parallel to the axis of rotation of the 
 vessel. The advantage of this shape is obvious, for it is easily seen that 
 with the wind box in this position a greater volume of metal may be held in 
 the vessel while in the horizontal position without filling the tuyeres than 
 would be possible with a round box, which is the form used on eccentric 
 vessels. The double bottom, mentioned above, is also of great advantage. 
 Since the space between the upper and lower plates connects with the outside, 
 it not only gives warning of a worn out tuyere, but also prevents the wind 
 box from being filled with hot metal in case of a break out. When a 
 tuyere becomes defective or badly and dangerously worn during a blow, 
 it may be plugged by turning the vessel down, removing the wind box lid, 
 and stopping its openings with clay. 
 
PURIFYING THE METAL 187 
 
 SECTION V. 
 
 THE CONVERTER IN OPERATION PURIFYING THE METAL. 
 
 Charging the Vessel: With the bottom fastened in place and the 
 vessel at the proper temperature, it is ready for the charge. The charging 
 is a matter of much importance. Besides being a factor in determining 
 the composition or grade of steel produced with respect to phosphorus and 
 sulphur, it also offers a means of controlling the temperature during the 
 blow. It must always be predetermined by the blower, who has charge 
 of the blow. In acid practice the charge consists of molten pig iron, 
 to which is added cold pig iron or steel scrap in amounts sufficient to meet 
 the heat requirements of the blow. As the only source of heat is the 
 oxidation of the iron, silicon, manganese and carbon, the composition of 
 the pig iron is important, and the blower must be kept informed in advance 
 as to the composition of the iron. In operating a hot vessel on hot iron 
 there is much more heat generated than is required to keep the metal 
 molten, in which case the temperature may be kept under control by 
 charging steel scrap. Scrap may be added to the heat at any time during 
 the first part of the blow. The addition of scrap also has the advantage 
 of increasing the output. In beginning on a new lining or a new bottom, 
 or after a delay, the vessel will be cold and will, itself, absorb much heat, 
 which condition precludes the use of cold materials in the charge. In an 
 attempt to lessen the loss of iron through oxidation and shorten the time 
 of a blow, roll scale or other oxides of iron are often charged. ' Such additions 
 may reduce the blowing period by about one-third, and are made regularly 
 at some plants, but this is not the practice at Edgar Thomson. On making 
 a steel that requires a high sulphur content, like screw steel, the required 
 amount of this element, in the form of pyrite, may be added along with 
 the molten metal. The blower, having been informed as to the require- 
 ments of the rolling mills, decides upon the charge best suited to the con- 
 ditions, then sends an order to the mixer for a certain weight of pig iron. 
 At these works this amount will vary from 30,000 to 36,000 pounds. The 
 molten iron is weighed at the mixer as it is poured into the ladle, the truck 
 of which sets on a scale platform. Some coke breeze is then thrown upon 
 the molten metal to keep it from skulling the ladle, when it is taken by 
 a dinkey to the charging floor of the converter. Here, the vessel is turned 
 down to a horizontal position, so as to bring the tuyeres well above the 
 bath, and the molten iron is poured into the mouth of the vessel by slowly 
 tipping the ladle. The scrap is added from the scrapping floor shortly after 
 the vessel is brought to the vertical position. 
 
 The Blow: Immediately after the vessel has received the charge of 
 molten metal, the blast, under a pressure sufficient to prevent the metal 
 from flowing into the tuyeres and also force the air through the liquid, is 
 turned on, and the vessel is racked to the vertical position. With this act 
 the air of the blast is forced to pass up through the molten mass, and 
 chemical action between the oxygen of the air and the various ingredients 
 
188 BESSEMER PROCESS 
 
 of the metal immediately begins. This oxidation takes place in successive 
 stages, each of which, provided the blow is a normal one, produces its 
 own peculiar effect in the metal and upon the kind of matter ejected from 
 the mouth of the vessel. Their order, therefore, may be followed by the 
 naked eye or through colored glasses. As the vessel is righted a shower 
 of sparks is emitted from its mouth. Then a stream of dense brown fumes 
 pours forth, to be succeeded shortly by a dull red, short, pointed flame 
 that protrudes from the mouth of the vessel. This action occupies but 
 five or six minutes, when this flame is gradually replaced by a short luminous 
 one that plays about the mouth. This flame soon begins to increase, both 
 in length and luminosity, until it has reached a maximum length of thirty 
 feet or more, which it maintains steadily for about eight minutes. During 
 this period, known as the boil, a dull roaring coming from the vessel may be 
 heard. This noise is caused by the violent agitation of the bath by the 
 blast and the rapid generation of carbon monoxide gas within it. Just 
 before the end of the blow, the flame begins to drop, or "die," that is, it suddenly 
 becomes less luminous, giving an effect similar to that to be expected if a 
 smoked glass or a cloud were placed between it and the eye; and if it is being 
 observed through blue glasses, purple streaks are visible in it. If the blow 
 should be continued, this flame would disappear entirely, but the metal is 
 always poured before this point is reached. Thus, the entire time required 
 to convert fifteen to eighteen tons of pig iron into steel is only about fifteen 
 minutes. 
 
 Controlling the Blow: The appearance of the flame just described 
 serves as an index to the change going on in the vessel, and so is very 
 important to the blower, upon whom rests the responsibility for the proper 
 operation of the vessel. He is also held accountable for the quality of 
 the steel he produces. He has an assistant who turns the vessel for charging 
 and pouring and operates the ladle crane, but the control of the process 
 is in the hands of the blower himself. He must decide the best proportions of 
 hot metal and scrap to use, regulate the temperature, determine the time 
 for turning down and over-see the recarburizing of the blown metal. AS 
 to the kind of recarburizer and the amount to use per ton of steel, he receives 
 instructions from his superintendent's office. Factors that enter into the 
 making up of the charge have already been explained. The importance of 
 a high temperature was also alluded to as necessary to keep the bath molten. 
 In this connection it remains to be pointed out that temperature is an 
 important factor in controlling the blow, and so exerts an influence on the 
 quality of the product. As it is impossible to regulate the charge so as 
 to meet all the variations in the conditions, other means of regulating 
 the temperature must be resorted to during the blow itself. To raise the 
 temperature after a blow is in progress, the vessel may be turned so as 
 to expose a few tuyeres above the metal. The combustion of the carbon 
 monoxide gas over the bath generates heat, which raises the temperature 
 of the vessel and consequently of the metal also. This method wastes 
 some metal, as iron is excessively oxidized. Ferro silicon is also used for 
 
PURIFYING METAL 189 
 
 this purpose, the oxidation of the silicon being the source of heat in this 
 case. Either method is expensive and should be avoided. With rapid 
 working, and with iron of proper grade, cold heats are the exception, 
 occurring mainly on new linings or in the first blow on a new bottom. To 
 lower the temperature is a much easier matter. The vessel may be tilted 
 and allowed to cool by radiation, or cold metal in the form of steel scrap 
 may be added, if the heat is not too far advanced. A more convenient 
 method is that of introducing steam with the blast, and as it is very con- 
 venient, it is often employed. The water coming in contact with the 
 highly heated metal is decomposed according to the following reaction: 
 H2O+Fe=FeO+H2. Steam thus introduced is not very efficient because 
 the oxidation due to air is not retarded and very little, if any, heat can be 
 absorbed. However, less heat is generated in oxidizing iron with water than 
 with air, besides, steam is easily controlled, is always at hand, and can be 
 introduced in varying amounts without delay to the blow or turning the 
 vessel. The blower will, then, keep close watch on the flame, and introduce 
 steam during the blow as often as required to hold the temperature at the 
 proper level. The speed of the blow, and, indirectly, the temperature, may 
 be controlled to a limited extent, also, by varying the blast pressure. In 
 this connection it should be stated that there are so many variables con- 
 nected with the operations that no uniform method can be established. 
 Even with metal of uniform composition and other conditions apparently 
 alike, two consecutive heats made to the same specification will seldom 
 require the same manipulation. Thus, the success of the entire operation 
 depends upon the judgment of the blower. 
 
 The End of the Blow: Owing to the rapidity of the reactions and other 
 peculiar conditions, the composition of the steel cannot be well regulated by 
 stopping the blow. While it is possible to blow a heat to approximately any 
 carbon content desired, the method is not practiced in America, because it 
 slows down the operation too much. It is much cheaper and surer, therefore, 
 to blow full, and add both carbon and manganese with the recarburizer. 
 Xhis is the practice at Edgar Thomson. At these works, if the blow 
 is stopped at the first indication of the drop of the flame, it is said to 
 be turned down young; if continued till the drop is pronounced, the blow is 
 full. In either case the silicon will have been completely eliminated, while 
 only small amounts of manganese and carbon will remain. The residual 
 manganese may be as high as .06 or .08%, depending upon the extent of the 
 blow and the percentage in the pig iron. If the blow is turned down young, 
 .08% to .10% carbon will remain, while in a full blow this amount is 
 decreased to .03 or .04%. The percentage of phosphorus and sulphur is 
 slightly higher than in the original pig iron, owing to a loss in weight due 
 to oxidation and elimination of the silicon, carbon, manganese and part of 
 the iron, and also to the ejection of metallic iron from the vessel. The 
 total loss will amount to something between 8% and 10% of the charge, 
 nearly half of which is oxide of iron and manganese which can be recovered 
 by using the slag in the blast furnace. 
 
190 
 
 BESSEMER PROCESS 
 
 SECTION VI. 
 
 FINISHING OPERATIONS CONVERTING THE PURIFIED METAL INTO STEEL. 
 
 Deoxidation and Recarburization must always immediately follow 
 the blow. At Edgar Thomson this is done in the ladle as the metal is being 
 poured, though at certain other plants some of the additions are made in 
 the vessel. In general the objects sought are: 1st., control of the carbon 
 content; 2d., deoxidation of the steel; and 3d., introduction of elements, 
 such as manganese, to improve the quality of the steel. The following 
 table shows the difference in the analysis of steel before and after recar- 
 burizing, and partly illustrates the many grades produced. 
 
 Table 29. Showing Chemical Relation of Purified Metal to 
 Different Grades of Steel. 
 
 Kind of 
 Steel 
 
 Per Cent. 
 Carbon 
 
 Per Cent. 
 Manganese 
 
 Per Cent. 
 Sulphur 
 
 Per Cent. 
 Phosphorus 
 
 As Blown.. . 
 
 .03 to .10 
 
 Trace to .06 
 
 .03 to .06 
 
 .08 to .100 
 
 Skelp 
 
 Not over .08 
 
 .30 to .40 
 
 Not over .06 
 
 Not over .100 
 
 Sheet Bar. . 
 
 " " .10 
 
 .30 to .50 
 
 " .06 
 
 " .100 
 
 Screw Steel. 
 
 " .08 
 
 .60 to .80 
 
 Not under .085 
 
 " .100 
 
 Special 
 
 
 
 
 
 Billet Steel 
 
 .25 to .30 
 
 .40 to .50 
 
 Not over .085 
 
 '4 .100 
 
 Light Splice 
 
 
 
 
 
 Bar 
 
 .08 to .10 
 
 .35 to .60 
 
 " .06 
 
 " .100 
 
 Rail Steel. . 
 
 .30 to .50 
 
 .70 to 1.10 
 
 " .06 
 
 " .100 
 
 Needless to say, the different grades of steel require different methods 
 of recarburizing to meet the requirements, which fact calls for different 
 recarburizers. The various recarburizers and deoxidizers most commonly 
 employed are ferro manganese, spiegel, anthracite coal, ferro-silicon, 
 and pig iron, analyses of representative samples of which are given in the 
 subjoined table. 
 
 Table 30. Analyses of Representative Samples of Deoxidizers 
 and Recarburizers. 
 
 
 Per Cent. 
 Iron 
 
 Per Cent. 
 Carbon 
 
 Per Cent. 
 Manga- 
 nese 
 
 Per Cent. 
 Sulphur 
 
 Per Cent. 
 Phos- 
 phorus 
 
 Per Cent. 
 Silicon 
 
 Per Cent. 
 Ash 
 
 Ferro Manganese. . 
 Spiegel 
 
 11.95 
 73 20 
 
 6.50 
 5 00 
 
 80.40 
 40 
 
 Trace 
 
 .160 
 100 
 
 1.00 
 1 10 
 
 
 Ferro-Silicon 
 
 86 70 
 
 2 00 
 
 50 
 
 050 
 
 080 
 
 10 60 
 
 
 Pig Iron. . . . 
 
 93 05 
 
 4 50 
 
 67 
 
 047 
 
 088 
 
 1 67 
 
 
 Anthracite Coal. . . 
 
 
 85.50* 
 
 
 
 
 
 4 50 
 
 
 
 
 
 
 
 
 
 *Fixed carbon only. 
 
FINISHING THE BLOW 191 
 
 Loss of Recarburizer and Deoxidizer: In adding the recarburizers, 
 a loss always takes place, for which an allowance must be made. The 
 amount of this loss is fairly uniform under similar conditions, and is deter- 
 mined by experience. In the case of manganese it amounts to about 20% 
 of the manganese added for full blown heats, in which the per cent, of 
 manganese does not exceed .60. The loss is somewhat less, not over 15%, 
 if the blow is stopped young. The loss varies, also, with the amount of 
 manganese added, increasing rapidly as the per cent, in the steel is raised 
 above .60. Similar data is required in using ferro-silicon and anthracite 
 coal, the loss of carbon in using the latter being about 50% of the total 
 amount added. 
 
 ^Examples of Recarburizing: Some simple examples of recarburizing 
 will illustrate the methods employed. 1. Suppose it is required to 
 produce a soft steel, such as the skelp shown in the table above. The 
 metal will be given a full blow to reduce the carbon content to about .04%, 
 and hot ferro manganese will be added in sufficient quantity to raise the 
 per cent, of this element to .40. A simple calculation, if proper allowance 
 is made for both residual manganese and manganese lost, will show that 
 this amount of ferro will raise the per cent, of carbon to .08. 2. In the 
 case of a medium soft steel, say .20% to .25% C., .40% to .50% Mn., the 
 recarburization after a full blow may be made with molten Spiegel mixture 
 containing about 12% Mn. and 5% C., or, as it is difficult to handle and 
 weigh small amounts of molten metal, coal and ferro-manganese are more 
 often used. In the latter case, the blow may be turned down young. 3. 
 In the case of a rail heat the blow is turned down young, and recarburized 
 with molten spiegel mixture. Molten pig iron and ferro manganese could 
 also be used, but this is not the practice at the Edgar Thomson Bessemer 
 plant. At this plant the cupola charge for the spiegel mixture used to 
 recarburize rail heats consists of spiegel, ferro silicon, and pig iron, in 
 proportion to produce a mixture containing 12% Mn., 4.50% C., and 1.50% 
 Si. For determining the quantity of deoxidizer and recarburizer to add, 
 the blower is provided with a set of factors, one for each grade of steel 
 produced, the numerical values of which are fixed by experience. Thus, 
 for sheet bar the factor giving the amount of 80% ferro manganese to add, 
 is .0045, but for skelp it is .0055 because this steel is blown very full. 
 Similarly, factors for finishing rail steel with spiegel are given. 
 
 Ladle Reaction : The addition of the recarburizer is usually followed 
 by a violent boiling of the metal in the ladle, causing much slag to be 
 thrown out over the sides. This is often referred to as the spiegel reaction 
 or ladle reaction. With the addition of the recarburizer, precautions 
 will be taken to mix it thoroughly with the metal. At Edgar Thomson 
 this mixing is accomplished in the case of rail heats by using molten 
 
192 BESSEMER PROCESS 
 
 spiegel and pouring it into the ladle with the metal from the vessel, 
 and in the case of soft steels by poling the metal in the ladle. 
 
 Teeming: The history of the heat may now be resumed. Soon after 
 the recarburizer has been added, the pouring of the metal will have been 
 completed. The converter is then inverted, and the slag which did not 
 flow out with the metal is dumped upon a small flat car beneath the vessel, 
 which is then ready for the next charge. While this is going on, the steel 
 has become quieter in the ladle and has been raised to the proper level 
 by the steel crane, which then transfers it to the teeming table in front 
 of the pouring platform. Here the teeming hole in the bottom of the ladle 
 is opened by removing the small plate and digging out the sand, when 
 the metal may be allowed to flow at will by raising and lowering the stonper 
 lever. The metal is now teemed, consecutively, into four ingot moulds, 
 which have been prepared as previously described. As each mould is filled 
 to the mark, the next is moved under the nozzle by means of the "dog," 
 hydraulically operated and provided for the purpose. During the teeming 
 of each ingot of soft or medium soft steel, small pieces, about four ounces in all, 
 of aluminum may be added, as the judgment of the teemer directs, to assist 
 in further deoxidizing the steel. This metal will always be added if the 
 steel is very wild, which condition is often found in soft steel made by 
 this process. After all the steel has been teemed into the moulds, the 
 little train is pushed along the track to the end of the teeming platform, 
 where the ingots are allowed to cool. If the ingots show a tendency to 
 grow in the moulds, the tops may be sprayed with water, and heavy caps 
 of cold iron will be placed on them. This treatment is intended to chill the top 
 and stop the growing, which invariably increases the number and size of the 
 blow holes and pipe in the top of the ingot. Growing is peculiar to soft 
 steels; rail heats seldom exhibit this tendency. When the ingots have 
 cooled sufficiently to form a thick, strong shell on the outside, they are 
 taken to the stripper, where the moulds are at once removed. This done, 
 they are ready for the soaking pits, which are more properly treated under 
 rolling mills. 
 
 Sampling the Steel for Chemical Analyses: A sample for chemical 
 analysis is taken during the teeming of each heat. This matter is of much 
 importance, and has received the attention it deserves. The sample is 
 obtained when half of the ladle of steel has been teemed by holding a large 
 steel spoon beneath the nozzle and allowing a small stream of the metal 
 to flow therein until the spoon is full. This metal is then poured from the 
 spoon into a specially constructed mould where it is allowed to cool or set, 
 after which it is stamped with the heat number and is then taken to the 
 chemical laboratory for analysis. Everything has been done to insure 
 this sample is truly representative of the whole heat, which is seldom true 
 of samples taken in other ways. 
 
CHEMISTRY OF 193 
 
 SECTION VII. 
 
 CHEMISTRY OF THE PROCESS. 
 
 The Order of Elimination of the Elements : As previously indicated, 
 the heat required for the process is generated by the oxidation of the iron 
 and the metalloids, silicon, manganese, and carbon. An examination of 
 the blow will show that, during the first period, the oxygen of the blast 
 attacks first the iron, then, both directly and indirectly, as will be explained 
 shortly, the silicon and manganese, producing exothermic reactions which 
 rapidly increase the temperature of the bath. The converter gases during 
 this period are mainly nitrogen with some carbon dioxide and traces 
 of oxygen and hydrogen. These reactions produce no flame, since all the 
 products of the oxidation are solids, but with the rise in temperature, 
 carbon begins to be oxidized to carbon monoxide, which will burn at the 
 mouth of the vessel to 'carbon dioxide and produce a flame outside the 
 vessel. So, the heat generated by combustion of the CO to CO2is wasted. 
 The rapid generation of CO in the metal produces the "boil," and the 
 increasing speeds at which the formation of this gas takes place causes 
 the flame to grow to a maximum size and finally subside with the elimi- 
 nation of the carbon. The escaping gases during this period consist mainly 
 of nitrogen and carbon-monoxide with small percentages of carbon dioxide 
 and traces of hydrogen. Thus, at no time during the blow, except for a 
 short period at the beginning, does any but traces of the oxygen of the air 
 escape from the bath uncombined, though the layer of metal is but some 
 twenty inches thick, and the volume of the blast is more than 6000 cubic feet 
 per minute. This fact is not surprising, if it is remembered that the temper- 
 ature of the bath from the first is much above the kindling temperature 
 for any element in the bath, and that the blast is delivered by the tuyeres 
 almost in the form of a spray. Under these conditions, the combination 
 of these elements with oxygen must be almost instantaneous, resulting in 
 all the oxygen being consumed at the mouth of the tuyere. Concerning 
 the brownish fumes ejected by the converter, especially at the beginning 
 of a blow, various suppositions have been advanced to account for them. 
 It has been suggested that they may be volatile compounds of iron and 
 manganese with carbon, which, upon coming in contact with the air at the 
 mouth of the vessel, are immediately oxidized, the metallic oxides producing 
 the brown color. Analysis of deposits made by this fume have been made, 
 and they are found to be composed roughly of one part ferrous oxide, two parts 
 silica and three parts manganese oxide. Manganese is volatile at a compara- 
 tively low temperature, which fact may account for a part of the fume, 
 but with respect to iron and silicon or silica, the most plausible explanation 
 is that they are carried out mechanically in a finely divided state by the 
 blast. 
 
 The Laws and Conditions Governing the Reactions in the Con- 
 verter: A review of the laws of chemical action and of the conditions of 
 
194 BESSEMER PROCESS 
 
 the blow will render an explanation of the changes that take place to bring 
 about the results enumerated above very easily understood. If reference 
 be made to the laws controlling chemical action in Chapter I., it will be 
 found that, under normal conditions of blowing metal, only two are applicable 
 to the matter under consideration. One of these, the law of mass action, 
 states, in effect, that the rate or speed of a chemical reaction may be 
 increased by increasing the active masses,' or amounts, of the reacting 
 substances; and the other law says that when chemical reactions take 
 place without the aid of heat supplied from an external source, those sub- 
 stances which have the greatest heats of formation, that is, those that 
 give off the most energy, will tend to form. As to the conditions, these 
 can be very briefly and simply stated. They are, that at the beginning of 
 the blow the bath represents a solution of approximately 4 parts carbon, 
 1.5 parts silicon, 1 part manganese, .1 part phosphorus and .05 parts sulphur 
 in 93.35 parts iron at a temperature that is several'degrees, say 100, above 
 the fusion point of the mixture; but this temperature is rapidly raised 
 during the first part of the blow. The phosphorus and sulphur are not 
 affected, so they need not be considered, but the elimination of the other 
 impurities presents an interesting study. 
 
 Reactions of the First Period : Chemical knowledge does not tolerate 
 the idea that these impurities are oxidized by the action of oxygen directly, 
 but indicates that the reactions occurring at the beginning of the blow are 
 governed by the law of mass action. According to this law, iron, by far 
 the most abundant element present, is first oxidized almost to the entire 
 exclusion of the other three elements. This reaction, which liberates a 
 large amount of heat, is represented thus! 
 
 (1) 2 Fe+O 2 =2 FeO (+131400 cal.) 
 2 (65700 cal.) 
 
 With the oxidation of the iron to FeO, this oxide, being soluble in the metal, 
 is distributed throughout the bath, and the oxidation of the silicon and 
 manganese take place in the order of the heats of formation of their oxides, 
 as shown in the following reactions: 
 
 (2) 2 FeO +Si=SiO 2 +2Fe (+64600 cal.) 
 Heats of formation: 2(65700 cal. ) + (196000 cal.) 
 
 (3) FeO+Mn=MnO+Fe (+25200 cal.) 
 Heats of formation: 65700 cal. +90900 cal. 
 
 With the oxidation of silicon and manganese, a slag is immediately formed 
 by the combination of silica with the excess FeO and the MnO according 
 to the following: 
 
 (4) FeO + SiO 2 = FeO.SiO 2 (+9300 cal.) 
 Heats of formation :-^65700 cal. 196000 cal. +271000 cal. 
 
 (5) MnO + SiO 2 = MnO.SiO 2 (+5400 cal.) 
 Heats of formation: 90900 cal. 196000 cal. +292300 cal. 
 
CHEMISTRY OF THE PROCESS 195 
 
 In comparing the ratio of acids to bases as determined by actual analysis 
 with ratios calculated from formulas, evidence is obtained that the slag 
 is made up, in part at least, of trisilicates, in which case these reactions 
 would be represented thus: 
 
 (4 A) 2 FeO+3 SiO 2 =(FeO) 2 -(SiO 2 ) 3 
 (5 A) 2 MnO+3 SiO 2 =(MnO) 2 -(SiO 2 ) 3 
 
 This slag, itself a solution of the two silicates thus formed, will dissolve 
 some of the FeO, and being mixed with metal by the violent agitation of 
 the bath, will also help to oxidize the impurities. Reactions (1) and (4) 
 also account for the rapid wearing away of the bottom. This period is 
 then preeminently one of slag formation. Some carbon, especially 
 toward the end of the period, however, may be oxidized directly to CO 
 and then to CC>2 by the FeO, thus: 
 
 (6) 2C+O 2 = 2 CO (+29160 cal.) 
 
 29160 cal. 
 
 (7) FeO +CO =Fe+CO 2 (+2540 cal.) 
 65700 cal. 29160 cal. +97200 cal. 
 
 These reactions account for the presence of both CO 2 and CO in converter 
 gases during the first part of the blow. At the beginning of the blow, CO 
 is subject to reduction by both silicon and manganese, especially if the 
 iron contains a high per cent, of these elements. 
 
 (8) 2 CO+Si=SiO 2 +2C (+137680 cal.) 
 2(29160) cal. + 196000 cal. 
 
 (9) CO +Mn=MnO+C (+71740 cal.) 
 29160 cal. + 90900 cal. 
 
 Reactions of Second Period: With the elimination of silicon and 
 manganese, reaction (8) and (9) cannot take place. Furthermore, the rise 
 in temperature brings about reaction (10) or (10A). 
 
 (10) FeO +C=Fe+CO (36540 cal.), 
 65700 cal. +29160 cal. 
 
 (10A) FeO+Fe 3 C=4Fe+CO ( 45000 cal.) 
 65700 cal. 8460 cal. +29160 cal. 
 
 These reactions, in conjunction with reaction (6), rapidly burn out the 
 remaining carbon. According to the law involving heats of formation, these 
 reactions should not take place, and it becomes necessary to explain certain 
 apparent exceptions. At ordinary temperatures the law has no exceptions, 
 but at elevated temperatures it holds true through certain ranges of tem- 
 perature only, so that, as the temperature in any particular case is raised, a 
 point, which may be called the critical temperature, is reached, where the 
 energy supplied from the external source overbalances that absorbed by the 
 reaction. The law then becomes reversed, and the reaction proceeds in a 
 direction that will absorb the excess heat. This fact suggests the possi- 
 bility that with hot iron, that is, iron high in silicon, which is also initially 
 
196 BESSEMER PROCESS 
 
 at a very high temperature, it might be possible to eliminate the carbon 
 before the silicon and manganese could be oxidized, and the testimony of 
 the older and more experienced operators of converters is to the effect that 
 just such a result as this has often occurred when the conditions noted 
 were present. Furthermore, in the elimination of the carbon, the law 
 of mass action here becomes prominent again, for with the elimination of 
 the silicon and manganese the active mass of the ferrous oxide rapidly 
 increases until a second equilibrium is established, this time with carbon. 
 The reaction is also probably influenced by the volatility of the carbon 
 monoxide, one of the products of the reaction. Comparatively little heat 
 is available in the bath during this period. The net heat generated is the 
 difference between the heat of formation of FeO (65700 cal.) and that 
 absorbed in reaction (10) (36540 cal), or 29160 cal. which is the heat of 
 formation for CO. The carbon reaction occurs concurrently with the 
 phenomenon commonly spoken of as the boil. There is, during this period, 
 very little, if any, iron oxidized above that required to eliminate the carbon, 
 so each volume or molecule of oxygen in the blast will produce two volumes 
 or molecules of CO. The converter gases, therefore, show a high content 
 of CO, very little CO2 and a marked decrease in N2- The phosphorus 
 and sulphur suffer no oxidation from the action of FeO until all but traces 
 of carbon is eliminated, and then only in the presence and under the 
 influence of a strong base, such as lime. If the loss in weight in the bath 
 be taken as 10%, then steel made from iron containing .045% sulphur and 
 .089% phosphorus would show approximately .050% sulphur and .100% 
 phosphorus immediately after the blow. These percentages are affected 
 but slightly by the recarburizer. 
 
 Chemistry of Recarburizing and Deoxidizing: The importance of 
 this part of the operation is more fully appreciated when it is recalled that 
 the Bessemer process was made a commercial success only through deoxi- 
 dizing with manganese. This element, then, plays a very vital part, the 
 effect in the product most evident being the prevention of that combination 
 of hot and cold shortness commonly spoken of as rottenness. It is due to 
 the presence in the metal of iron oxide, FeO, which is dissolved by molten 
 iron. This oxide is reduced by metallic manganese, thus : (a) FeO+Mn= 
 MnO+Fe. As MnO is not soluble in iron to any appreciable extent, reaction 
 (a) will result in ridding the steel of all but traces of metallic oxide. Carbon 
 may act as a deoxidizer, according to some authorities, as shown by 
 reaction (b). 
 
 (b) FeO+C=Fe+CO. 
 
 However, the evolution of CO gas that produces the violent boiling of the 
 metal in the ladle, which boiling often continues also in the ingot mould after 
 the metal has been teemed, is probably caused by CO and other gases passing 
 out of solution in the metal as the latter cools. Small amounts of these gases 
 retained by the steel produce the blow holes previously alluded to. It is 
 
CHEMISTRY OF THE PROCESS 197 
 
 to be noted, also, that manganese offsets the evil effects of sulphur as will 
 be explained in a later chapter. The silicon as well as the carbon and 
 manganese in the ferro or spiegel and pig iron will also serve as a deoxidiz- 
 ing agent. Besides, silicon, by attacking CO, prevents the formation of 
 blow holes. For the greatest effectiveness a considerable excess of silicon 
 and manganese over that required by their respective reactions should 
 be used. This is one of the reasons why the manganese in steel will range 
 from .30 to 1.00%. Additional losses of the manganese in the recarburizer 
 are likely to occur by reacting with the silicate of iron oxide, thus: 
 
 (c) FeO 'SiO 2 +Mn=MnO *SiO 2 +Fe. 
 
 A study of Bessemer slags shows that a slight decrease of iron oxide without 
 a corresponding increase of MnO takes place on recarburizing. This cir- 
 cumstance is usually explained by assuming that one of the following 
 reactions takes place: 
 
 (d) FeO+C=Fe+CO or 
 
 (e) ,2FeOSiO 2 +CO=FeO (SiO 2 ) 2 +CO 2 +Fe. 
 
 As was pointed out under the head of teeming, other deoxidizing agents 
 may be added to the ingot as the metal is being teemed. The one most 
 commonly employed is aluminum. This element is one of the most powerful 
 deoxidizers known. Upon being heated to a sufficiently high temperature 
 it will react violently with all the metallic oxides, and also many others. 
 
 (f) 3 FeO+2Al=Al 2 O 3 +3Fe. 
 
 Various alloys are beginning to be used for this purpose, also. One of 
 the best is known as "A. M. S." metal; it is an alloy of aluminum, man- 
 ganese and silicon, and is said to be very efficient for this purpose. 
 
198 THE OPEN HEARTH PROCESS 
 
 CHAPTER VIII. 
 
 THE BASIC OPEN HEARTH PROCESS. 
 
 SECTION I. 
 
 SOME GENERAL FEATURES OF THE SIEMENS PROCESS. 
 
 Early History of the Process: The ever increasing demand for steel, 
 which even the phenomenal success of Bessemer was not able to meet 
 entirely, soon led many other inventors into the same field. But the only 
 process which was destined to become a rival of the Bessemer was developed 
 through the invention of the regenerative principle by that prolific inventor, 
 William Siemens. In this connection it may be of interest to note that 
 Siemens first developed and employed this principle in the construction of 
 steam engines, but while several of these engines were built and put into 
 use, they were finally abandoned because of the severe wear on the heating 
 chambers caused by the high temperature attainable. But it was shown 
 that a great saving of fuel and very high temperatures could be obtained 
 by the use of the principle, and at the suggestion of his brother, Frederick, 
 Siemens then turned his attention to the application of the principle for 
 producing high temperatures in furnaces. The first experimental furnace 
 was built in 1858, when it was developed that, with large furnaces especially, 
 many difficulties were to be overcome, if the full efficiency which the use 
 of the principle promised was to be obtained. After two years or more 
 of experimentation, Siemens fell upon the plan of gasifying the fuel prior 
 to burning it in the furnace, when he found that most of his difficulties 
 had been overcome. The first furnace burning gaseous fuel, patented in 
 1861, was used for making glass. Here, the great advantages of the furnace 
 in economy and regularity of working were fully proven, and it was not 
 long until it was adopted in other industries, also. Some of these early 
 uses of the furnace were for zinc distillation, for puddling, for reheating 
 iron and steel, and for melting crucible steel. Siemens then turned his 
 attention to the manufacture of steel in his furnace, and, though many trials 
 were made at many different works, he met with only indifferent success. 
 Finally, like Bessemer, he found it necessary to erect a steel works of his 
 own in which the success of the process could be demonstrated. These 
 works were located at Birmingham, England, and were at first employed 
 in a remelting process by which steel of the best quality was obtained 
 from such scrap as old iron rails, plates, etc. In the meantime, Siemens 
 was busy developing an idea of decarbonizing pig iron for making steel by 
 means of iron ore, and by the year 1868 he had proved that this process 
 could be successfully employed. Siemens next turned his attention to 
 
PRINCIPLES 199 
 
 evolving a method whereby steel could be produced directly from the ore, 
 thus dispensing with the blast furnace. In this feat he actually succeeded, 
 but the cost of production was many times that of producing steel from 
 pig iron. Nevertheless, he continued his experiments until his untimely 
 death in 1883 put an end to his endeavors. He died firmly believing that 
 his direct process would eventually supplant all conversion methods. Sub- 
 sequent endeavors have shown this idea is wrong, and that his pig and ore 
 process is the most practicable and economical. 
 
 Principles of Siemens Pig and Ore Process: Briefly, the method of 
 Siemens was as follows: He used a rectangular covered furnace to contain 
 the charge of pig iron or pig iron and scrap, and provided most of the heat 
 for the chemical reactions by passing burning gas over the top of the 
 materials. The gas, with a quantity of air more than sufficient to burn 
 it, was introduced through ports at each end of the furnace, alternately 
 at one end and the other. The gaseous products of combustion passed out 
 of the port, temporarily not used for entrance of gas, into chambers partly 
 filled with checker brick, which absorbed some of their sensible heat, and 
 from these chambers out through a stack. After a short time, the gas and 
 air were shut off at the one end and introduced through the heated checker 
 chambers at the opposite end, absorbing some of the heat stored in the 
 bricks. These gases then entered the furnace with a high sensible heat 
 and gave a higher temperature in combustion than could be obtained 
 without preheating. In about twenty minutes, the course of the gas and 
 air was reversed, so that they entered through the port first used; and 
 a series of such reversals, occurring every fifteen to twenty minutes, was 
 continued until the oxidation had reached the desired point. The elements 
 in the iron attacked by the oxygen of the air and of the iron ore fed in were 
 carbon, silicon and manganese, all three of which could be reduced to as 
 low a limit as in the Bessemer process. Thus, as in all the other 
 processes for purifying pig iron, the basic principle of Siemens process was 
 that of oxidation. But in other respects it was unlike any other process. 
 True, it resembled the puddling process in both the method and the 
 agencies employed, but the high temperature attainable in his furnace 
 permitted him to secure in the liquid state a perfectly malleable metal 
 which could be cast into ingots and be free of slag. In this respect the 
 same result was produced as in the Bessemer process, but by a different 
 method and through different agencies, both of which imparted to it many 
 advantages over the older pneumatic process. 
 
 Advantages of the Process: These advantages may be briefly 
 summed up as follows: 1. By the use of ore as an oxidizing agent and by 
 the external application of heat, the temperature of the bath is made 
 independent of the purifying reactions, and the elimination of the impurities 
 can be made to take place gradually, so that both the temperature and 
 the composition of the bath are under much better control than in the 
 
200 OPEN HEARTH PROCESS 
 
 Bessemer process. 2. For the same reasons, a greater variety of raw 
 materials can be used and a greater variety of products can be produced 
 by this than by the Bessemer process. 3. A very important advantage 
 is due to the increased output of finished steel from the same amount of 
 pig iron, which means that fewer blast furnaces are required to produce 
 a given tonnage of steel. 4. Finally, with the development of the basic 
 process, the greatest advantage of the Siemens over the Bessemer was 
 revealed through the elimination of phosphorus. Comparing the basic 
 open hearth with the Thomas-Gilchrist process it is to be noted that, 
 due to the different temperature conditions, phosphorus is eliminated in 
 the former before the carbon, whereas it is not oxidized in the latter process 
 until after the carbon, in what is known as the after-blow. Hence, while 
 the basic Bessemer process requires a pig iron with a phosphorus content 
 of 2.00% or more in order to maintain the temperature high enough for 
 the after-blow, the basic open hearth permits the use of iron of any phos- 
 phorus content. In the United States this fact is of the greatest importance, 
 since, for reasons already explained, it makes available immense ore deposits 
 which could not otherwise be utilized. For this reason the basic open 
 hearth process has become the leading method in this country. 
 
 Mechanical Changes and Improvements in Siemens Process: As 
 
 would be expected, many variations of the process, both mechanical and 
 metallurgical, have been worked out since Siemens first put his method 
 into operation. Along mechanical lines various improvements in the 
 design, the size and the arrangement of the parts of the furnace have been 
 made. Originally, the furnace had a capacity of only four or five tons, 
 but now the size ranges from 40 to 100 tons capacity, and in new plants 
 the capacity will seldom be less than 75 tons. But the greatest departure 
 from Siemens' original plan was made by the invention of the tilting or 
 rolling furnace. These furnaces are of two types, and are known as the 
 Campbell and the Wellman furnaces, respectively. In each case the 
 furnace is built of brick, which are held firmly in piace by a strong frame- 
 work of steel, and is mounted upon rollers or rockers, thus permitting it 
 to be tilted either forward or backward. In the Wellman type the hearth 
 and ports are built solid, so that both move together; and as tilting the 
 furnace breaks the connections with the regenerator flues, the furnace can 
 be fired only when in an upright position. This fault is overcome in Camp- 
 bell's invention, in which the hearth only is movable, and the center of 
 rotation is coincident with the center line of the ports. By the use of 
 water cooled castings, fairly tight joints are made between the hearth and 
 the flues, so that the furnace may be tilted in either direction, forward or 
 backward, without turning off the gas and air. 
 
 Metallurgical Improvements: The hearth of Siemens' furnace was 
 of acid brick construction, and the bottom was made up of sand essentially 
 as in the acid process of today. Later on, in order to permit the charging 
 
IMPROVEMENTS 201 
 
 of limestone for the removal of phosphorus, the hearth was constructed 
 with a lining of magnesite brick, which were covered with a layer of burned 
 dolomite or magnesite to replace the sand of the acid furnace. These 
 furnaces were, therefore, designated as basic furnaces. The pig and scrap 
 process was originated by the Martin Brothers. By substituting scrap 
 for the ore in Siemens' pig and ore process they found it was possible so 
 to dilute the charge with steel scrap that little oxidation was necessary. 
 Since the time of the Martins, these processes have undergone various 
 modifications, chief of which are those known as the Talbot, the Campbell, 
 the Bertrand-Thiel, and the Monell processes. By using a basic lined 
 tilting furnace in which a large bath of the purified molten metal is always 
 retained, Talbot succeeded in hastening the oxidation of the silicon, manga- 
 nese, phosphorus and carbon to such an extent that the operation is made 
 more nearly continuous, and the time between heats or tappings is greatly 
 reduced. Campbell's tilting furnace permits him to apply the pig-and-ore 
 process to molten metal, because, by tilting the furnace forward, the frothing 
 of the bath, produced by the violent reactions, is prevented from throwing 
 the slag through the doors as would be the case in a stationary furnace. 
 By a combination process, he also aims to make acid steel from basic pig 
 iron. In a basic lined furnace he eliminates the silicon, manganese, phos- 
 phorus and a little of the carbon, then pours this semi-purified metal into 
 an acid furnace, where the remainder of the carbon is worked down as in 
 the regular acid process. The Bertrand-Thiel process is applied to pig 
 iron with a very high phosphorus content, and makes use of the two-period 
 scheme of purification, also. In the first period, the furnace is tapped in 
 order to separate the metal from the slag, which contains such a high 
 percentage of phosphoric acid, P2O5, that it is valuable as a fertilizer. 
 The metal is then poured either back into the same furnace or into another 
 basic furnace for the final purification. -In developing his process, Monell 
 had the same objects in mind as Talbot, namely, the rapid conversion of 
 basic iron into steel; but he wished to avoid the reservoir of molten metal, 
 and make his process adaptable to the stationary furnace. He accom- 
 plished his object by first charging limestone and ore into a basic furnace, 
 heating these until the batch became pasty, then adding molten pig iron, 
 when the silicon, manganese, and phosphorus were rapidly oxidized and, with 
 the lime, formed a slag that, as the carbon began to be oxidized, foamed 
 up and ran from the furnace through slag notches provided for the purpose. 
 
 The Process for the Pittsburgh District: Most of the pig iron 
 available in the Pittsburgh district contains a fairly high percentage of 
 phosphorus, and the mills produce considerable scrap. Hence, the furnaces 
 are practically all basic the Carnegie Steel Company no longer operates 
 any acid open hearth furnaces and a combination of the pig-and-ore, pig-and- 
 scrap, and Monell processes is employed. It has been briefly described as 
 follows: Limestone is charged on a basic bottom, ore is charged on top 
 of the stone, and scrap on top of this; if molten pig iron cannot be obtained 
 
202 OPEN HEARTH PROCESS 
 
 in sufficient quantity to complete the charge, some cold pig is charged 
 with the scrap; and the entire mass is heated in the furnace for about two 
 hours, or until the scrap is white hot and slightly fused. Molten pig iron 
 is then added, when a lively reaction occurs, in which almost all of the 
 silicon, manganese, phosphorus and part of the carbon are oxidized, the 
 first three forming compounds that slag with the iron oxide, and join the 
 iron and lime silicates that are already melted. About 80% of this slag 
 is drawn off by the end of two or three hours more. The ore acts on the 
 carbon for three or four hours longer, during which time, and continuing 
 afterwards, the limestone is being decomposed by the heat, and its CO 2 
 is bubbling up through the bath and exposing part of the metal to the flame, 
 thus oxidizing it and completing the purification started'by the ore reaction. 
 What is known as the lime action, or boil, lasts two or three hours longer; 
 and then, if the charge was calculated correctly, the carbon content will 
 be somewhat greater than that at which the metal is to be tapped. 
 Ordinarily, of course, the carbon is too low or too high, in which cases 
 more pig or more ore must be added. In about another hour the carbon 
 content will have been reduced to the proper amount for tapping, which is 
 usually about .10%. 
 
 SECTION II. 
 
 EQUIPMENT FOR A MODERN BASIC OPEN HEARTH PLANT. 
 
 The Modern Plant: Besides the furnaces themselves, the modern 
 open hearth plant requires considerable additional equipment. Thus, there 
 must be provided ladles for containing molten metal; moulds for ingots; 
 cranes and charging machines for handling materials; boxes for the solid 
 materials; dinkeys or electric-engines for hauling the materials; a stripper 
 for removing the moulds from the ingots; a great number of small articles, 
 like shovels, wheel barrows, rabbles, etc.; and, finally, apparatus for pre- 
 paring or controlling the fuel supply. In addition the more modern plants 
 will be provided with a mixer, a calcining plant, and also spiegel-cupolas, 
 if liquid recarburizers are used. A brief description of the more essential 
 items enumerated above is given herewith. 
 
 Calcining Plant: At most of the plants there are cupolas for roasting 
 dolomite. These furnaces are cylindrical in form, and each one is so placed 
 that one base forms the bottom, the other the open top of the furnace. 
 In the usual construction the cupola is made up of an outer shell of boiler 
 pi ate, one-half inch thick, and a double refractory lining made of two courses 
 of brick, the inner one being of first quality and the outer one next to the. 
 shell of second quality fire brick. On the floor of the furnace, there is a 
 cone shaped casting which deflects the burnt dolomite, in its descent, 
 toward the circumference, where it may pass out through openings provided 
 for the purpose in the base of the cupola. For fuel, coke is employed, and 
 it must be burned by an air blast. This blast is supplied at a pressure 
 
PLANT EQUIPMENT 203 
 
 of five to six ounces. A bustle pipe distributes the air and is provided with 
 about eight tuyeres, or connections, with the cupola. Built on the charging 
 platform and extending about three feet above it, there is a seat upon which 
 the charging bucket is deposited by a crane and from which its contents, 
 through a bell and hopper arrangement in the bottom of the bucket, may 
 be dropped into the cupola. Usually the ratio of materials in the charge is 
 two buckets of dolomite to one bucket of coke. Thus, the fuel consumption 
 is about 15% of the weight of dolomite calcined. The operation of burning, 
 in which the CC>2 is expelled from the stone, leaving calcium and magnesium 
 oxides (CaO and MgO) in place of their carbonates (CaCO3 and MgCOa), 
 requires from ten to twelve hours. The burning period is controlled by the 
 rate at which this material is withdrawn. As fast as the dolomite is 
 burned it is shoveled out at the bottom, crushed to pass a half-inch mesh, 
 then conveyed to the loading bins, whence it is later taken as required 
 to the open hearth. 
 
 Fuels: For fuels in open hearths, natural gas, coke oven gas, producer 
 gas, powdered coal, fuel oil and, sometimes, tar are used. The choice of 
 fuel depends largely upon the location of the plant. Natural gas is, of 
 course, preferable when it can be obtained, as it is of uniform composition, 
 is easily controlled and imparts no sulphur to the bath, while coke oven gas, 
 producer gas and powered coal are of varying composition and quality and may 
 impart some sulphur to the bath. The great demand for petroleum pro- 
 ducts has made fuel oil too costly, while the supply of tar is so limited that 
 it is not always available as an open hearth fuel. There remains, then, as 
 the principle substitutes for natural gas, only powdered coal, coke oven gas 
 and producer gas, all of which have already been discussed in the chapter on 
 fuels. (See Chapter IV., Section 6). In either case, coal of the best grade 
 obtainable is desired, especially with respect to sulphur, as there is always 
 danger of its being absorbed by the bath. In using coke oven gas from 
 which the benzol has been removed, it is found necessary to burn with it some 
 substance like tar, for example, that will impart luminosity to the flame, as 
 otherwise the poor visibility within the furnace makes it difficult for the 
 melter to control the temperature. In the new plants of the Carnegie Steel 
 Company, the producers are connected in sets of four five in one or two 
 cases and each set serves two furnaces, the gas mains from the producers 
 being arranged so that any one or all of the producers in a set may discharge 
 into either one or both of the furnaces. 
 
 Fuel Consumption: It is a difficult matter to arrive at any conclusion 
 as to the amount of heat required to produce a ton of steel. It is subject to 
 a number of conditions such as kind of fuel used, condition of the furnace 
 and checkers, continuity and rate of production, and the care and intelligence 
 with which the furnace is operated. In actual practice the number of heat 
 units per ton of steel produced is found to vary from 5,000,000 B. t. u. to 6,000,000 
 B. t. u. when natural gas is used, and from 6,000,000 B.t.u. to 8,500,000 B.t.u. 
 when other fuels are employed. An inspection of the fuel records of several 
 
204 OPEN HEARTH PROCESS 
 
 different plants and a consideration of averages for long periods of time, vary- 
 ing from six months to a year, indicate that the fuel consumption per gross 
 ton of steel produced, to be considered good practice, should be about as follows: 
 natural gas, 5,000 cubic feet; coke oven gas, 8, 000 cubic feet, with 16 gallons of tar; 
 tar alone, 45 gallons; producer gas equivalent to 600 pounds of coal ; and powered 
 coal, 500 pounds. 
 
 Hot Metal Mixer: The advantages of the hot metal mixer have 
 already been discussed in connection with the Bessemer process; and 
 although the conditions in the open hearth plant, where large quantities 
 of metal are needed at irregular and uncertain intervals of time, are exactly 
 the reverse of those in a Bessemer plant, which requires small quantities 
 of metal at short and comparatively regular intervals, these advantages 
 are as applicable to the one case as to the other. In order that these 
 advantages may be realized to the fullest, open hearth mixers should have 
 a large capacity. One large mixer of 1000 or 1200 tons capacity is to be 
 preferred to two of 500 tons capacity. 
 
 Spiegel Cupolas: In plants manufacturing large quantities of medium 
 high carbon, high manganese steels, such as is used for railroad rails, for 
 example, the use of spiegel for recarburizing may be advantageous, in which 
 case cupolas for melting the spiegel mixtures are an important adjunct 
 to the open hearth plant. In construction and operation, these cupolas 
 are similar to those already described for the Bessemer plant. For collect- 
 ing and weighing the different ingredients of the charge, a larry car equipped 
 with a multiple beam scale is most convenient. In charging, the metallic 
 parts of the burden, consisting usually of spiegel and pig iron, are 
 charged into the cupola together, while the coke, with which is mixed 
 enough limestone to flux its ash, is charged separately. The proportion 
 of coke required in each round will vary somewhat, but in good practice it 
 will seldom exceed seven per cent, of the weight of the metallic part of the 
 charge. To secure greater uniformity and provide an ever-ready supply of 
 molten recarburizer, the cupolas attached to the most modern plants are 
 provided with a small mixer. By means of brick and clay lined runners the 
 metal from the tap hole of each cupola is conducted directly into this mixer, 
 from which definite amounts may be taken as desired. For weighing the 
 recarburizing metal, a track scale, on which the transfer ladle may rest 
 during the pouring, is placed on the track directly in front of the mixer. 
 The manganese content of the molten recarburizer is varied to suit the 
 requirements of the different grades of steel by varying the amount of 
 pig iron with which the standard spiegel is diluted. Varying the proportion 
 of pig iron to spiegel also changes the carbon content of the mixture slightly. 
 With a given weight of standard spiegel, the more pig iron charged the lower 
 the carbon content of the mixture will be, as can readily be seen by 
 comparing the analyses of these materials. 
 
 The Steel Ladles: The ladle for receiving the steel is made of boiler 
 plate and is lined with two courses of brick each 2y 2 inches thick. The first 
 layer, next to the shell, is usually of fire brick, while the second layer is of 
 
PLANT EQUIPMENT 205 
 
 white river brick. Both courses are laid on end in a motar of fire clay, to 
 which a little loam is sometimes added. The capacity of the vessel is depen- 
 dent on the amount of steel to be handled in each heat, which in turn is fixed 
 by the capacity of the open hearth. The opening at the bottom of the ladle 
 is provided with a fireclay nozzle about two inches in diameter, which may 
 be closed by a stopper. The stopper is made of [clay bonded graphite and 
 is mounted on a rod, protected by fireclay sleeve brick, that reaches to the 
 top of the ladle; there it is connected to a sliding bar on the outside that can 
 be raised or lowered by a lever near the base. Both the stopper and the 
 nozzle must be replaced after each heat. Great care is necessary both in 
 placing the nozzle and in setting the stopper in the nozzle, for a bad fit 
 results in a running stopper, which may cause a great waste of metal. To 
 prevent the steel from chilling about the stopper, powdered coal is often 
 thrown into the depression around the nozzle just before tapping a heat. 
 
 The Stripper: The action of this machine has already been described 
 in connection with the Bessemer process. The ingots must all be sufficiently 
 cooled before stripping, so' that there will be no danger of breaking the 
 solidified wall of metal. After being stripped, the ingots are then ready to be 
 sent to the soaking pits previous to the rolling. To strip an ingot, it is only 
 necessary, in the majority of cases, to place the jaws of the stripping machine 
 under the lugs on the mould and apply the lifting force, when the ' mould 
 will slip from the ingot and can then be raised to a sufficient height to 
 transfer. It is only at times, usually due to a defective mould or to 
 metal being splashed over the top edges from a running stopper, that the 
 moulds are not slipped off easily, and then the plunger is rested on top of 
 the ingot as the mould is drawn upward. When this treatment fails to 
 loosen an ingot, it is sent to the mould yard where more time is available 
 for extracting it and where it may be subjected to various treatments 
 according to the means at hand and the cause of its sticking. 
 
 Moulds: After the ingots of each heat are stripped the empty moulds 
 are stored in the mould yard until they are sufficiently cool to be drawn back 
 to the open hearth for another charge, and during the wait they are washed 
 inside with clay slurry, the water of which is quickly evaporated by the 
 heat of the mould, leaving it covered with a thin coating of the clay. Any 
 damaged moulds that cannot be used are charged as cold iron, as they are 
 cast from a good grade of Bessemer pig iron. Many types and sizes of 
 moulds are used. Heavy moulds chill the surface of the steel quickly 
 and hasten the solidification, which always proceeds from the wall of the 
 mould toward the middle of the ingot. Since steel contracts on solidifying, 
 there is a cavity left directly under the top surface of the ingot, as the 
 metal in this location is the last to solidify. This cavity, called the pipe, 
 is responsible for the production of a great deal of scrap in rolling the steel. 
 There are numerous methods to reduce the size of the pipe and to keep it 
 as near the top as possible, but it cannot be entirely eliminated. The 
 principle of most of the devices is to keep the top of the ingot molten longer 
 
206 OPEN HEARTH PROCESS 
 
 than the bottom, so that the molten steel on top will flow into the cavity 
 as fast as it forms and thus lessen the extent of the pipe. The Gathman type 
 of mould depends upon uneven thickness of mould wall to effect the same 
 result. By having the mould thin at the top and thick at the bottom, 
 the thin top has a less chilling effect on the molten steel at the top, 
 which, therefore, is the last to solidify. A similar method consists in 
 lining a removable top of the mould with brick or clay, thus preventing 
 rapid conduction and radiation. Some try to keep the steel at the top of the 
 ingot fluid by a coke, a charcoal or a gas fire. Many other more complicated 
 devices have been invented, also, but their use involves much additional 
 expense. Besides, piping is regarded by many as a necessary evil, and the 
 safest way to avoid it is to allow a proper discard from the top of the ingot, 
 which discard is cut off at the blooming mill shears. Armor plate ingots 
 are sometimes cast in specially constructed hard sand moulds. Ingots 
 for this material are always bottom cast, two ladles being poured at the 
 same time, which are followed frequently by a third, pouring directly into 
 the mould. There are standard sinkheads for all armor plate ingots. 
 
 The Charging Machine: Of all the labor saving devices employed 
 about the open hearth plant, none have brought a greater saving of money 
 and time than the charging machine. Indeed, it may be looked upon as the 
 most essential part of the equipment, for if the charging were done by hand, 
 the time thus lost, especially in the case of the large furnaces, would be 
 so great that this feature would appear as a serious drawback to the process. 
 There are several types of these machines, but the ones most generally 
 employed are of the low ground type. They consist of two main parts. 
 First, there is the bottom truck made up of a very strong steel frame-work 
 and mounted on flanged wheels which travel on a very wide gage track 
 laid in front of the furnace. Next, there is the charging carriage, which 
 moves over a track, laid on the frame of the truck, at right angles to the 
 direction of motion of the truck itself. On this carriage is mounted a 
 kind of lever, the long arm of which extends toward the furnace and is 
 known as the charging bar. The charging bar is hollow to provide space 
 and bearings for the locking bar, about which it can be made to revolve, 
 and is shaped on the end to fit into the socket of the charging box. The 
 charging bar is thus capable of giving eight different primary motions, 
 or any number of resultants of these motions. In operating the machine, 
 the charging boxes rest on buggies running on a narrow gage track between 
 the machine and the furnace. First, the truck of the machine is moved 
 so that the charging bar is directly opposite the charging box to be emptied, 
 then the carriage is moved forward to bring into position the end of the 
 charging bar, which is then dropped ( into the socket on the end of the 
 charging box and locked in position by advancing the locking bar until its 
 front end projects into a hole provided for the purpose in the socket of the 
 box.- Now, the machine is made to serve for a shifting engine, and, by 
 moving the truck, the whole train of charging boxes may be moved along 
 in front of the furnace, so that the box engaged is brought directly opposite 
 
PLANT EQUIPMENT 207 
 
 the door. The charging bar is then raised, carrying the box with it, and 
 by a forward motion of the carriage the box is passed into the furnace, 
 where, by rotating the charging bar, the box is turned upside down and 
 its contents deposited. By reversing these motions the box is then placed 
 upon the buggy again. The charger will pick up and empty a box in less 
 than a minute. As the capacity of the box is more than one ton, for even 
 the lightest materials of the charge, it is possible to charge even the very 
 large furnaces in less than an hour. Charging machines are always elec- 
 trically operated. 
 
 Charging Boxes : The charging boxes are made of cast steel or of five- 
 eighth inch boiler plate with cast steel ends. One end of the box is provided 
 with a socket opening from the top so that the T section on the end of the 
 charging machine peel, or arm, may readily be inserted and withdrawn from it. 
 The boxes have a capacity of sixteen cubic feet or more, and into them all 
 solid material for the furnace charge is placed. For transporting the boxes 
 from place to place about the works, buggies are provided. These buggies 
 are of standard or narrow gage type, are made of cast iron and accom- 
 modate three to four boxes each. 
 
 Stock Yard : There is usually one stock yard to each plant. In it are 
 kept the stores of limestone, ore, cold pig and, sometimes, scrap, from 
 which materials the cold charges for the furnaces are made up. The ore 
 and limestone are loaded into the boxes from chutes or by grab buckets, 
 depending upon the manner of storing; the cold pig from railroad cars or 
 from a stock pile by means of a magnet; while the scrap is brought from 
 the rolling mills ready loaded in the charging boxes, or if it is delivered 
 in railroad cars, it is transferred to them by magnets. Soft ore is 
 generally used to make up a charge on account of its lower cost, because 
 there is no advantage in charging lump ore on the bottom, and it is not 
 necessary to have as pure an ore as is the lump ore. After a charge is 
 made up, the buggies are pushed over platform scales and weighed by 
 the weighmaster, who records the weights in a stock book. The stock- 
 yard men take into account the amount of pig iron to be used and, to a 
 certain extent, the carbon contents of the scrap charged and add to the 
 charge what ore they think will be necessary. Reports giving the 
 weights of all materials are made out in triplicate, one of which is sent 
 to the melter, so that after it reaches the furnace front the melter-foreman 
 or his first helper may vary the ore charge to suit their plans or ideas. 
 
 Arrangement of the Plant: The furnaces of the plant are always 
 enclosed and covered by an immense steel building, and in the modern 
 plant the furnaces are arranged end to end in a long row along the center 
 of this building. That part of the floor of the building along the front, 
 or charging side, of the furnace is called the charging floor. On this floor 
 and next to the furnaces is laid a narrow gage track on which the solid 
 materials are conveyed to the furnace for charging, while, back of the narrow 
 gage track, there extends a very wide gage track, with a spread of about 
 
208 
 
 OPEN HEARTH PROCESS 
 
THE OPEN HEARTH FURNACE 209 
 
 twenty feet, for the charging machines. The space above this floor and 
 the furnaces is spanned by two or more electric overhead cranes. The 
 remaining floor space of the building lying along the tapping side of the 
 furnace is called the pouring floor, and is also spanned by electric cranes. 
 These two floors may be on the same level, as in some of the older plants; 
 but all the new plants are of the two=level type, that is, the pouring floor 
 lies some twelve to eighteen feet below the level of the charging floor. 
 The pouring platforms, six to eight feet wide and about eight feet high, 
 are located along the outer edge of the pouring floor. The mixer and 
 cupolas are often located at one end of the open hearth building, as this 
 arrangement permits the transfer of the hot metal to be made with the 
 cranes. However, in large plants this arrangement would be inconvenient, 
 as it would interfere with the work of the cranes, so the hot metal is carried 
 to the different furnaces on a track laid on the charging floor. With this 
 arrangement the mixer may be located at any convenient point. When 
 producer gas is used for fuel, the producer plant is built back of the open 
 hearth plant, parallel to the charging floor. The calcining plant, stock 
 yard, and mould yard are located at points as convenient to the open 
 hearth house as possible. The stripper should be placed so that the steel 
 is always advancing toward the soaking pits of the blooming mill, though 
 this matter is but a question of convenience. 
 
 SECTION III. 
 
 CHIEF FEATURES OF BASIC OPEN HEARTH CONSTRUCTION. 
 
 Parts of the Open Hearth Furnace and Their Arrangement: An 
 
 open hearth furnace consists of the furnace proper, containing the covered 
 laboratory, hearth or bath, in which the charge is placed; ports for admitting 
 the gas and air over the charge; regenerative chambers, containing checker 
 brick for storing up heat from the products of combustion and imparting 
 it to the cold gas and air; flues and uptakes, connecting the checker chambers 
 with the furnace proper; slag pockets, which are located at the base of the 
 uptakes; flues, leading from the air and gas supply (if producer gas is used) 
 to the checker chambers, with connections to the stack; valves for regulating 
 the direction of flow of gas, air and waste gases; and the stack itself. The 
 furnace proper is located on the level of the charging floor and rests on 
 a concrete foundation. The slag pockets, checker chambers, flues and 
 valves are all located in a cellar, on a level about fifteen feet below the 
 charging floor in houses of the one level type, or on the first floor level in 
 the two level type. The checker chambers are not located under the furnace 
 proper but under the charging floor in front of it, and the stack is placed 
 a short distance beyond nearer the gas producers. The base of the stack 
 flue sets on a level with the bottom of the checker chambers, but the 
 stack proper begins at the charging floor level. 
 
 The Furnace Proper: The furnace, itself, is a rectangular brick 
 structure, supported on the sides and ends by vertical steel buck-stays in 
 the form of channels or slabs, four to five and one half inches thick and 
 
210 OPEN HEARTH PROCESS 
 
 \ 
 
 eleven inches wide, and bound together at their tops, both longitudinally 
 and crosswise, by stays and tie rods. The most recently constructed 
 furnaces have a capacity rated at 100 tons. Such a furnace is 
 approximately eighty feet in length and twenty feet in width, outside dimen- 
 sions over all. Ten sets of buck-stays on the front and rear sides, and 
 four or six sets on the ends are required to furnish the requisite support 
 against expansion of the brick work. The buck-stays are held in place by 
 12-inch channels placed at their tops. These channels extend entirely 
 around the furnace, those along the sides being securely tied with bolts 
 and clamps to those crossing the ends of the furnace. The front and 
 rear buck-stays are united by tie-rods which are two and one-half inch 
 steel rounds; the ends of these extend through the buck-stays and are 
 threaded to receive nuts to hold them in place so that they can be tightened 
 and loosened according to the expansion and contraction of the furnace. 
 The foundation under the furnace proper is built of concrete, and is of 
 such depth and shape as to bear the superimposed load with reasonable 
 safety. It is usually in the form of two large piers, with an arched 
 opening separating them. The furnace proper comprises the hearth, the 
 side walls, and the roof. 
 
 The Hearth is constructed as follows: On top of the concrete foun- 
 dation is placed a layer of three feet or more of second quality fire brick, 
 and upon these is laid a two foot layer of first quality fire brick in which 
 a number of 15 inch I-beams are placed to act as a bottom anchorage for 
 the vertical buck-stays which surround the furnace; a nine inch layer of 
 magnesite brick is then laid on top of the firebrick, and upon these bricks, 
 a bottom is made up approximately ten and one-half inches thick with a 
 mixture of burned magnesite, 75%, and ground basic slag, 25%, which is 
 sintered into place. Dolomite may be substituted for the magnesite, but 
 in this case the bottom must be much thicker than when magnesite is 
 used. When complete, the hearth has the form of a shallow dish whose 
 sides extend up to' the level of the charging doors. In order to obtain this 
 shape the succeeding courses of magnesite brick are stepped back, until 
 the normal thickness of side wall, about thirteen and one-half inches, is 
 reached. The exact hearth dimensions, inside, between fifteen and sixteen 
 feet in width and about forty feet in length, are dependent upon the desired 
 maximum capacity of the furnace and incidental features. The depth is 
 such that the bath of molten metal will be from twenty to twenty-four 
 inches deep The back wall of the hearth is pierced at its exact center for 
 the tapping hole, which is about eight inches in diameter and is provided 
 on the outside with a removable cast iron lip for receiving the end of the steel 
 spout, the function of which is to conduct the molten steel from the furnace 
 to the steel ladle at the time of tapping. The slag hole is placed about 
 fifteen feet from the tap hole and near the upper edge of the hearth. It is 
 surrounded with magnesite brick and is provided with an iron casting at 
 its base for the attachment of the cinder spout. 
 
 The Walls are begun on the top course of magnesite brick that sur- 
 
CONSTRUCTION OF FURNACE 211 
 
 rounds the upper edge, or brim, of the hearth. They are built of silica 
 brick, are about thirteen and one-half inches thick and extend to a distance 
 of about eight feet above the charging floor level. The back wall is built 
 up solid except for the tapping hole and slag hole, but the front wall 
 contains the arched doorways for charging. The doors are usually five 
 in number, the middle one being in the middle of the furnace, and are so 
 placed that their bases, or sills, are a few inches above the slag line. 
 Each opening is provided usually with a water cooled cast steel frame, 
 placed between two buck-stays to which it is fastened, and is closed by a 
 water cooled, fire brick lined cast iron or steel door that may be lifted 
 vertically either by hydraulic or electric power. A wicket, or peep hole, 
 is placed on the center line near the bottom of each door. 
 
 The Roof over the hearth is made of silica brick, about twelve inches 
 thick, and is arched from front to back only in the newer furnaces, but 
 sometimes, also from end to end in older types. The roof is built inde- 
 pendent of the walls, and rests on skew back brick set in water cooled 
 skew back channels that are riveted or bolted to the buck-stays of the 
 furnace. 
 
 The Bulk Heads which form the ends of the hearth below the ports 
 were originally built of solid brick work and were a source of much trouble, 
 as they burned out rapidly. These difficulties were all avoided by replac- 
 ing much of this brick work with a large, hollow cast steel box with open 
 ends to provide for air cooling. The inside surfaces of the bulk heads 
 are made of magnesite brick. 
 
 The Ports : The air and gas ports at each end of the furnace are built 
 at an angle to the bath, so that the flame is directed against the bath and 
 away from the roof. In order to protect the roof and promote the mixing 
 of the gas and air, the air enters above the gas. The bricks separating 
 the two ports are protected by a water cooling system. Where natural 
 gas is used, unless the furnace was constructed for producer gas, there is 
 but one port at each end of the furnace, the gas entering on each side of 
 either end of the furnace behind a bridge wall, which extends across the 
 furnace in front of the well, or up-take, from the checkers. This wall is 
 usually made of magnesite brick, though chrome brick is well suited for 
 the purpose, and is about thirteen and one-half inches thick and nine inches 
 high. 
 
 The Up-and-Down-Takes are the vertical flues which connect the 
 air and gas ports with the slag pockets and the fan-like flues leading to 
 their respective checker chambers. They are built of silica brick and are 
 not water cooled. In producer gas fired furnaces there is a pair of up-and- 
 down-takes for air at each end of the furnace, the two in each pair being 
 at opposite sides of the furnace. The up-and-down-takes for the gas rise 
 with their centers coincident with the center line of the furnace and may 
 stand out with three walls exposed, designated as the dog-box type, or 
 be built in between the air flues. For a 100-ton producer gas fired furnace 
 these flues are each four feet by three feet, inside dimensions. 
 
212 OPEN HEARTH PROCESS 
 
 Arrangement of Up-and-Down-Takes for Natural Gas, Coke Oven 
 Gas, Powdered Coal and Tar: The construction of the up-and-down- 
 takes in a natural gas fired furnace is much simpler, as it is only necessary 
 to have one up-take for the air at each end of the furnace. This up-take 
 in modern furnaces is circular, with a diameter of about six and one half 
 feet, and is therefore called the well. The air, as it rises, is deflected 
 downward toward the bath by the port, which is arched from front to 
 back but is straight, longitudinally, with a downward slope toward the 
 hearth. This roof is usually about nine inches thick, except near the 
 neck where it joins the roof of the furnace. Here it increases to twelve 
 inches on account of this point being subjected to the greatest wear. 
 The bridge wall previously described, which crosses the port adjacent 
 to the up-take, causes the incoming air to roll down past the opening from 
 the gas pipes, one of which enters at each side of the port. Thus, there are in 
 all four pipes to a furnace. These gas pipes have a diameter of four inches. 
 The main supply pipe usually passes over an entire row of furnaces. 
 A branch line, provided with a meter, a valve, and a three way cock, leads 
 to each furnace, where it again branches into the four inch pipes which 
 enter the ports. For tar and powdered coal the same construction as for 
 natural gas has been employed, because, so far, these substances have been 
 used only as a substitute for the latter fuel when the supply became low. 
 Both these substitutes are introduced into the furnace by inserting the 
 nozzles of the burners through small openings in the brick work closing 
 the ends of the furnace, one burner at each end of the furnace being required. 
 
 Slag Pockets: The slag pockets are chambers at the bottom of the 
 up-and-down-take flues. Their functions are to serve as flues to conduct 
 the gases to and from the checkers and to catch any solid matter carried 
 over with the products of combustion, thereby preventing most of this 
 slag material from reaching the checkers and clogging them up. The 
 pockets are designed large enough so that only in extreme cases do they 
 have to be cleaned out more than once every run. In the 100-ton producer 
 gas fired furnace, they are about three feet six inches wide, and eight feet 
 high. The two at each end of the furnace are separated by a three foot 
 silica brick wall. The outside walls are two feet seven and one half inches 
 thick for the air, and three feet for the gas side; the former have an inside 
 lining of silica brick, set against first quality fire-brick, while the latter is 
 made of silica brick only. The floor and roof are covered inside with 
 silica brick, the latter being arched on a radius of half the width of the 
 pockets. One end of each pocket merges into a short, fan-like flue, called 
 a neck, which leads to the top of its regenerator chamber. 
 
 Regenerators for Producer Gas: The regenerators, of which there 
 are two pairs to a furnace, are built out in front of the furnace and under 
 the charging floor, about half below and half above the casting floor 
 level. They are separated from the furnace by a distance of about 
 four feet. Each pair is made up of one checker chamber for gas and one for 
 
CONSTRUCTION OF FURNACE 
 
 213 
 
214 
 
 THE OPEN HEARTH PROCESS 
 
 Fia. 26. Longitudinal Vertical Section of 100-Ton Open Hearth Furnace, 
 
 S/L/CA 
 
CONSTRUCTION OF FURNACE 
 
 215 
 
 Showing Slag Pockets, Flues, Ports and Hearth. 
 
216 OPEN HEARTH PROCESS 
 
 air, arranged so that the outer wall of the gas chamber is nearly in line 
 with the end of the furnace. The gas chamber is always smaller than the air 
 chamber, because the larger volume of air is necessary to burn the gas and 
 assist in the oxidation of the bath. The total space actually occupied 
 by the checkers in all four chambers is from 120 to 150 cubic feet per ton of 
 furnace capacity. For a 100-ton furnace the volume of the checkers in 
 the air chamber is between 3500 and 3600 cubic feet while the corresponding 
 volume in the gas chamber is between 2500 and 2600 cubic feet. 
 
 The gas chambers on such a furnace, measured inside, are about 
 thirty-one feet long, eight feet wide and sixteen and one half feet high from 
 the bottom to the base of the roof, which is arched to rise twenty-three 
 to twenty-five inches higher. The air chambers are of the same length 
 and height, but are about eleven and one half feet wide, and the arch 
 in the roof rises about thirty-four inches. The walls of both gas and air 
 chambers are built usually with nine or thirteen and one half inches of 
 common brick on the outside and thirteen and one half inches of first 
 quality fire brick on the inside, and are reinforced on the two sides and 
 the free ends by channel buck stays and tie rods. At some plants the 
 two chambers in a pair are built en bloc with a single dividing wall 
 between them. In this plan of construction the dividing wall is about three 
 feet thick and is built entirely of first quality fire brick. The floors of the 
 chambers are started usually with a nine inch layer of concrete, which is fol- 
 lowed with a heavy coat of tar as a water proofing. On the tar is laid 
 another nine inch layer of concrete, then four and one half inches of common 
 brick and four and one half inches of first quality fire brick. On this floor, 
 are laid nine inch fire brick withe walls, which divide the gas and air 
 chambers longitudinally into three and four flues, respectively, to a height 
 of about four feet. These walls are spanned by fire brick tile, size, 
 3"xl2"x31", and on these tile, the checker work, of best quality fire brick 
 size, 4M"*4J$'xlOi", is begun and continued to within about three and 
 one half feet of the top of the gas chamber, or to within about four feet 
 of the top in the air chambers. The arched roofs of the regenerators 
 are of fire brick and thirteen and one half inches thick. The checker 
 work is separated from the flues leading to the slag pockets by a solid 
 wall which rises to their top. This wall aids much in preventing slag 
 and dust from being carried into the checkers. But in spite of all 
 precautions, some dirt is carried over into the chambers, which causes 
 them to become choked eventually, when the furnace must be closed down 
 until the checkers are cleaned or replaced. 
 
 Regenerators for Natural and Coke Oven Gases: At some plants 
 where natural gas or coke oven gas is used, as for example at Homestead 
 and Clairton where natural gas was originally the only fuel employed, the 
 gas chamber, in addition to the regular air chamber, is utilized to preheat 
 the air. The original idea was to construct the regenerative chambers so 
 that in case gas producers were built, the change in fuels would not necessi- 
 
CONSTRUCTION OF FURNACE 217 
 
 tate a rebuilding of the furnace; but, now, the chambers for natural gas 
 are all being constructed in this manner, because it was found that when 
 two air checkers at each end of the furnace are employed, better results 
 are obtained than when only one is used. Where one large checker is 
 operated, the air and stack gases, instead of flowing to all parts of the 
 chamber, tend to take a direct course through the center, thus markedly 
 decreasing the efficiency of the chamber. 
 
 Regenerators for Powdered Coal : The use of powdered coal for fuel 
 introduces a serious difficulty in the operation of the regenerators because 
 of the large percentage of ash that is carried over into the chambers by 
 the draught. This fume soon clogs the ordinary checkers to such an extent 
 that they are no longer efficient. Two different types of regenerators, 
 namely, the arched and columnar types, designed with the idea that they 
 would permit the ash to be cleaned out of the chamber without tearing 
 out the brick work, have been tried; but as the ash fuses upon the bricks, 
 these schemes are impracticable and have consequently been abandoned 
 in favor of the old style of construction. But instead of the usual checker 
 brick a large tile, measuring about 24"x9"x4", has been substituted, which 
 gives larger openings for the passage of the gases. This construction 
 appears to be much more satisfactory than either of the others that have 
 been mentioned. 
 
 Flues and Valves: While the openings into the slag pockets are at 
 the top of the checker work, the openings for the ingress and egress of 
 gases at the opposite end of the checker chamber are at the bottom. Here 
 the small flues formed by the withe walls open into a large one which leads 
 to the stack flue in the case of the air chambers, or, in the case of the gas 
 chambers, to a three-way water sealed valve, one of the best types of which is 
 represented by the Ahlen valve. Another branch of this valve leads to 
 the stack and the third to the gas main. These valves, together with the 
 dampers and mushroom valves in the flues from the air chambers, supply 
 the means by which the reversals of the flame are made. In the modern 
 furnaces, these valves and dampers are connected so that the reversal of 
 the air and gas currents take place simultaneously. All the valves and 
 dampers are controlled from the charging floor. Since natural gas, and 
 also coke oven gas, cannot be preheated without decomposing them, the 
 valve system in furnaces using these fuels, as well as those using powdered 
 coal, is much simpler than for those using producer gas. 
 
 The Stack : The stack for each furnace must be of such size and height 
 as to supply sufficient draught to the furnace. It is lined with first quality 
 fire brick, and usually has an inside diameter of 5 feet and a height of from 
 140 to 160 feet above the charging floor. The shell is made of Muich boiler 
 plate. It usually rests on a concrete foundation, on the same level as the 
 floor of the checker chambers, and at this level it has openings for flues 
 from the gas and air chambers, as previously described. For controlling the 
 draft a damper is placed in the main flue at its entrance to the stack. With new 
 or clean checkers this damper partly closes the main flue, but as the checker 
 becomes clogged, it is raised from time to time as required. 
 
218 OPEN HEARTH PROCESS 
 
 SECTION IV. 
 
 OPERATION OF A BASIC OPEN HEARTH PURIFYING THE METAL. 
 
 Furnace Attendants and Their Duties: For the work on each 
 furnace, three men, a first helper, a second helper and a cinder-pit-man, are 
 needed, and besides these, there is a foreman, called a melter foreman, in 
 charge of a number of furnaces. Ordinarily, the first helper has charge of the 
 furnace except at the tapping of a heat. He informs the charging machine 
 operator of the amount of ore the charge will require and how and where 
 to place the various parts of the charge; he regulates the heating of the 
 furnace; runs off the slag; directs any repairs necessary during the operation; 
 and has charge of working the heat, that is, making the necessary additions 
 of ore, pig, spar, etc. to prepare the steel for tapping. But when the heat 
 is ready to tap, the melter foreman takes charge. The first helper, subject 
 to the supervision of the melter, actually taps the heat, and, after doing 
 so, he directs the repair of the bottom and helps make up the banks and 
 clean up the steel spout. The second helper is next in charge; he keeps 
 a supply of dolomite, feed ore, fluorspar, ferro-manganese and ferro-phos- 
 phorus on hand and places the solid recarburizing additions on the platform 
 convenient to the ladle. He helps to work the heat, digs the plug out of 
 the tapping hole when the heat is ready to tap, keeps the tapping hole 
 open and clean while the furnace is being rabbled, and assists in making 
 up the banks of the furnace preparatory to recharging. He also attends 
 to the plugging of the tapping hole, relines the steel spout after each heat 
 and cleans up around the furnace. The cinder-pit-man attends to the 
 cleaning of the pits, from which the slag and metal must be removed after 
 each heat. In addition, he assists in making bottom at his own furnace 
 and all the others under his melting foreman. The melter, or foreman, 
 usually has charge of a group of six or seven furnaces. He takes charge of 
 any furnace in his group when any serious difficulty arises, and he always 
 has charge of the tapping of the heat. He receives an order for the kind of 
 steel desired from the steel distributor, so when the tapping time of a heat 
 is near, he orders the recarburizer and moulds necessary, and takes charge 
 of the furnace when the carbon is but a few points above the tapping point. 
 He decides when the heat is ready, gives the order to tap, and directs the 
 addition of the recarburizers. He gives the order for lifting the ladle when 
 the steel is out of the furnace, superintends the teeming of the steel, and 
 inspects the bottom of the furnace after the heat is out. 
 
 Preparation of the Furnace for Its First Charge: Starting a new 
 furnace is an operation that requires a great deal of care in order to avoid 
 injuring the brick work and to prevent explosions, especially when producer 
 gas is used for fuel. The complete preparation of the furnace may be said 
 to take place in four stages, known as drying, heating, making bottom 
 and washing. The drying is begun very slowly with wood or gas fires, 
 and requires about twenty-four hours, during which time all the con- 
 nections to the stack on both ends of the furnace are left open. The 
 
CHARGING RAW MATERIALS 219 
 
 temperature is then gradually increased for about another twenty hours. 
 When the furnace has almost reached a red heat inside, the products of 
 combustion are led off through only one set of checkers for three or four 
 hours, then gas is turned on carefully and the real heating is begun, 
 which requires about twenty-four hours more. During this time, the 
 flame is reversed at intervals of about an hour at first, then more often, 
 in order to heat up both sets of checkers evenly and uniformly. When a 
 slag-melting temperature has been reached, finely ground magnesite is 
 thrown into the furnace to cover the joints between the magnesite brick, 
 and a little finely ground basic cinder is scattered on top of it. About 
 twelve hours is required for these additions to fuse and make the bottom 
 solid. The making of the bottom is then begun. For this purpose burned 
 magnesite is much preferred, but as this substance is sometimes very 
 expensive, calcined dolomite is employed as a substitute. With the 
 former, the procedure is about as follows: A mixture of burned magn- 
 esite, 75%, and basic cinder, 25%, both ground to pass a half inch screen, 
 is scattered over the bottom and sides of the hearth to a depth of about 
 a half inch, and allowed to sinter. At the end -of about three hours, the 
 gas is turned off, and another layer of the mixture is thrown in; and this 
 procedure is repeated, at the same intervals of time, until the bottom 
 and banks have been built up to the desired thickness of about eleven 
 inches, which occupies about ten days in all. The tapping hole is next cut 
 through from the outside, to terminate on the bottom, and is then filled 
 up with burned dolomite, held in place by a cap of clay on the outside. 
 The furnace is then ready for the wash heat. About twenty tons of basic 
 cinder is charged and melted. This melt is rabbled up against the banks 
 so that every part of the hearth is made solid, and is then tapped out. 
 Burned dolomite is now piled on top of the banks as high as possible, 
 when the furnace is ready to receive its first charge. 
 
 Charging: The first charge consists of limestone, scrap, and cold pig 
 iron; neither ore nor hot metal are used on a new bottom until it shows 
 it is not absorbing iron and is absolutely solid. Trade heats, of approxi- 
 mately half scrap and half hot metal arechargedfor the first half dozen heats, 
 after which the percentage of hot metal is increased as rapidly as possible to 
 the normal. The materials in the charges vary for different kinds of heats, 
 but in general, a so-called Monell heat requires from 75% to 100% hot metal 
 and a trade heat less than 75% hot metal. In each case, the remainder of the 
 metallic part of the charge consists of scrap, while the ore and limestone are 
 varied to suit the conditions. An example of each as used on a 100-ton funrace 
 follows: 
 
 Monell. Trade. 
 
 Limestone .... 20000 pounds. Limestone .... 17000 pounds. 
 
 Ore 40000 pounds. Ore. . . 10000 pounds. 
 
 Scrap 45000 pounds. Scrap 95000 pounds. 
 
 Pig Iron 165000 pounds. Pig Iron 115000 pounds. 
 
220 OPEN HEARTH PROCESS 
 
 In place of ore, briquettes, made from blast furnace flue dust, or heating 
 furnace cinder may be substituted. The charge, with the exception of the 
 molten iron, is brought to the furnace in the charging boxes previously 
 mentioned, and charged by machine. Hot metal is brought, either from 
 the mixer or from the blast furnace direct, in ladles and is then poured 
 into the furnace through a runner that is introduced at one of the doors 
 for the purpose. Other additions in small quantities are thrown in by 
 hand through the doors. At one plant, furnaces with removable tops are 
 provided, in order to make it possible to charge very large pieces of scrap 
 which would not pass through ordinary doors. At all plants advantage is 
 taken during repairs to old furnaces to charge such large scrap through 
 the top before the roof is put on. As to the grade of the materials in the 
 charge, it is preferable to have an iron low in sulphur and silicon because 
 the former element is only partly removed in the furnace, if at all, and the lat- 
 ter, upon being oxidized to silica, rapidly cuts away the banks. A manganese 
 content between 1% and 2% is also desirable, as it assists somewhat in the 
 removal of the sulphur. As there is almost a complete elimination of 
 phosphorus in the process, the quantity of this element in the charge is 
 not of great importance up to one per cent. As previously indicated, 
 the pig iron, in order to save time and conserve heat, is charged in the 
 molten state whenever possible. 
 
 The Order of Charging the Raw Materials: As to the order of 
 charging, the limestone is always charged first for these reasons: If it 
 were charged on top of the scrap, for example, it would act as an insulator 
 and thus prolong the melting period; it would all go to make up a part 
 of the first slag, which would be too thick and viscous to work well; it 
 would be drawn off with this slag in the run offs, thus leaving very little 
 lime in the furnace to hold the phosphorus in the latter stages of the refine- 
 ment; and, finally, the benefits to be derived from the lime boil, to be 
 described later, would be lost. Upon the limestone, will be charged the 
 ore, or briquettes, which, if any is needed, will vary in amount according 
 to the nature of the rest of the charge and the heating capacity of the 
 furnace. In order to hasten oxidation, ore may also be added from time 
 to time during the later stages of the process. The scrap is next charged, 
 and if cold pig iron is used, it is charged with the scrap. The gas, which is 
 usually but partly turned on during the charging is then turned on full, and 
 the first or melting stage begins. If hot metal is to be charged, it is not added 
 until the melting period is well advanced. 
 
 Melting Down the Charge: Heat is imparted to the charge partly 
 through radiation from the incandescent particles in the flame. The fuel 
 should, therefore, burn with a full long flame reaching almost from end to 
 end of the furnace. But the flame should never extend through the ports 
 and down-takes, as it would then rapidly fuse the brick of those 
 flues and waste the fuel. For the same reason, the flame should 
 be directed downward from the port and not be allowed to impinge on the 
 
PURIFICATION PEROIDS 221 
 
 roof. The light scrap and pig iron, if any is added to the solid charge, 
 begin to melt first. During the melting much of these materials is oxidized, 
 so that there is formed both molten metal and oxides, which trickle down 
 over the scrap to the bottom. A slight amount of molten slag and metal 
 is thus present, on the bottom of the furnace, before the hot metal, 
 i. e., molten pig iron, is charged. Reversals of the flame should occur 
 every fifteen to twenty minutes during this period, and care must be taken 
 not to overheat the roof, for too high a temperature will cause the bricks 
 in a new roof to spall, and those in an old one to fuse. Silica brick frequently 
 sweat, that is, fuse slightly, but this condition does no harm and indicates 
 a favorable temperature in the furnace. Care must be taken with the roof 
 and checkers in a new furnace, especially, and the temperature must be 
 kept relatively low for the first ten heats or more, after which time the 
 gas may be gradually increased until the full working temperatupe is 
 attained. 
 
 The Addition of the Hot Metal: The molten metal can usually be 
 added in about two hours after the charging of the solid materials is begun. 
 The exact time for adding this metal is governed by the temperature of 
 the solid charge. Evidently this temperature should be above, or at least 
 as high as, that of the melting point for pig iron. This statement does not 
 imply that the scrap, which has a much higher melting point than pig 
 iron, should be completely melted. Indeed, a delay in the addition of the 
 molten metal until the scrap is all melted may be very undesirable, for 
 net only would the scrap be excessively oxidized, but the high temperature 
 combined with the excess oxides present would result in a too violent 
 reaction, and much foaming of the bath and loss of metal due to the rapid 
 generation and evolution of carbon monoxide would result. 
 
 The Purification Periods: The purification of the hot metal, after 
 it is introduced into the furnace, is brought about through the oxidizing 
 influence of the iron oxides and the fluxing properties of the limestone. 
 While both oxidizing and fluxing reactions are actually taking place in the 
 furnace at the same time, the action of the iron oxides must be considered 
 as preceding that of the limestone, for the acid impurities must first be 
 oxidized before they can be neutralized, or fluxed, by the bases. It is 
 evident that the fluxing action may immediately succeed the oxidation, 
 but the conditions set up by the manner of charging the limestone tends 
 to retard its calcination and thus to separate the two actions. Now, the 
 carbon monoxide generated by the action of the iron oxides upon the carbon 
 of the pig iron is at first evolved in a manner quite different from that 
 of the same gas formed later on in the process or of the carbon dioxide 
 from the calcination of the limestone, and this difference is indicated by 
 the way in which the bath is agitated. Hence, the fui nacemen have fallen 
 into the habit of speaking of the purification as taking place in stages, 
 known as the ore boil, the lime boil, and the working period. The third 
 
222 OPEN HEARTH PROCESS 
 
 is the stage that follows the complete calcination of the limestone. In 
 order that the reader may understand what is implied by these terms, 
 the changes that occur during the purification of the metal are discussed 
 under these three headings. 
 
 The Ore Boil : Proper chemical testing will show that the purification 
 of the molten iron begins immediately after it is charged into the furnace, 
 and, with the exception of carbon, the oxidation of which is not completed 
 till the heat is ready to tap, progresses very rapidly. So, in about two 
 hours practically all of the silicon and the greater part of the manganese 
 will have been oxidized, and the former, then in the form of silica, will 
 have been neutralized, some with lime, but the greater portion with the 
 oxides of iron and manganese, and will have become slag. Some of the 
 sulphur, also, will have been oxidized, particularly if the sulphur content 
 of the hot metal was high, but as the oxides of this element are volatile 
 and the high temperature tends to decompose the sulphites and sulphates, 
 only a part of the oxidized sulphur is retained by the slag, and the remainder 
 is carried off with the products of combustion. A very small portion of 
 this element finds its way into the slag as sulphides, probably as manganese 
 sulphide. With the silicon and manganese, the phosphorus is also rapidly 
 attacked by the iron oxide, which not only oxidizes it, but neutralizes the 
 resulting oxides of this element. These iron phosphates, which are easily 
 reduced, likewise pass into the slag, where the iron oxide is replaced with 
 lime, thus forming the calcium phosphates, which are very stable 
 compounds. During all this time the carbon is also being slowly oxidized. 
 This action, which at first takes place near the surface of the metal, results 
 in the evolution of carbon monoxide in the form df tiny bubbles, which 
 become entangled in the viscous slag and cause it to foam. Consequently, 
 the slag, thus permeated with little gas cells, occupies much more than its 
 natural space in the furnace. Carbon dioxide is also evolved by the lime- 
 stone, which begins to be calcined more and more rapidly as the temperature 
 at the bottom rises; but as the gas resulting from the decomposition of 
 the limestone escapes in relatively large bubbles, it causes Very little of 
 the foaming. 
 
 The Run off: When the slag level has been raised to a height a little 
 above that of the bottom of the openings for the doors and is threatening 
 to break through the dolomite dykes built up just inside these openings, 
 the dolomite with which the slag hole is dammed is cleaned out of this 
 opening, and the excess slag is allowed to flow through the cinder spout 
 into the cinder pit or into a slag pot placed below to receive it. This 
 tapping of slag is known as the run=off. The bases in this first slag 
 are composed chiefly of iron and manganese oxides, the lime and magnesia 
 being relatively low. It is not unusual for these slags to contain iron 
 as oxide equivalent to 30% metallic iron, and as they constitute about 
 40% of the total slag formed in the process, they represent the source of 
 greatest loss of metal for the entire process. Practically all of the iron 
 contained in the run-off is in the ferrous condition. 
 
PURIFICATION PERIODS 223 
 
 The Lime Boil: The action of the ore and other iron oxides, formed 
 by the oxidizing flame, upon the carbon will be somewhat violent for two 
 hours or more, during which time the scrap will have been almost com- 
 pletely melted down. Gradually, however, as the carbon content decreases 
 and the temperature of the bath rises, the ore boil subsides, or changes 
 its character, and, the calcination of the limestone becoming more rapid, 
 the lime boil is in the ascendency. The lime boil is characterized by a 
 rising of the lime to the top of the bath and by a violent bubbling of the 
 bath, caused by the rapid evolution of carbon dioxide gas from undecom- 
 posed limestone which still remains on the bottom, and also in part by 
 the continued oxidation of carbon in the molten metal. These activities 
 play important parts in the process. Thus, not only does the violent 
 bubbling caused by the evolution of the carbon dioxide gas agitate the 
 metal and slag, thus mixing them and exposing the metal to the oxidizing 
 influence of the flame, but a part of the gas, at least, unites, directly or 
 indirectly, with the carbon remaining in the iron to form carbon monoxide. 
 Furthermore, by rising to the surface, the lime may replace iron and 
 manganese oxides in the phosphates, sulphates and silicates present and 
 thus become a part of the slag, while any excess lime is also taken up and 
 goes to increase the basicity of the slag. This property of the slag makes 
 it more capable of retaining both the phosphoric and silicic acids together 
 and renders the former less liable to be reduced. During this period the 
 flame in the furnace should be reversed more frequently (about every 15 
 minutes) in order that the temperature of the bath eventually will be well 
 above the melting point of the decarbonized metal. 
 
 The Working Period: Since all the impurities except carbon have 
 now been eliminated, the operations during this period aim at regulating 
 the properties of slag, adjusting the carbon content of the steel, and raising 
 the temperature of the bath to the point where the steel may be tapped 
 from the furnace and cast into ingots before it begins to solidify. In order 
 for the steel to be teemed without difficulty, this temperature should be 
 at least 167 C. (300 F.) above its fusion point. Both the chemical and 
 the physical properties of the slag play most important parts in the basic 
 process. In order that it may protect the metal against contamination 
 by sulphur from the flame, retain the impurities, especially phosphorus, 
 and promote the elimination of the carbon, the slag must contain a large 
 quantity of active oxidizing agents, except at the end of the period, and 
 must be strongly basic at all times. But even with the chemical com- 
 position of the slag properly adjusted, its activity will depend upon the 
 fluidity to a great extent. The reagents at the disposal of the operator 
 for regulating these properties are iron oxide, limestone, dolomite and 
 fluorspar. The iron oxide is usually in the form of lump ore, though heating 
 furnace cinder formed on a magnesite bottom may be used. 
 
 Methods of Working the Heat : There are two general 
 methods of working heats, and briefly described, they are as follows: 
 
224 OPEN HEARTH PROCESS 
 
 The first method is somewhat like the Bessemer, that is, the carbon content 
 of all heats is reduced to a common point, about .10%, when the steel will 
 be tapped and the per cent, of carbon will be raised to that desired by 
 the addition of recarburizers. In the second method the carbon is caught 
 on the way down, that is, the carbon content is reduced to a point slightly 
 under that required, to allow for the carbon contained in various additions, 
 and the bath of steel is then tapped. Medium and low carbon steels are 
 usually worked by the first method, while high carbon steels maybe worked 
 by either. 
 
 Testing for Carbon: So toward the end of the lime boil, or earlier if 
 it appears that the carbon content of the bath is dropping rapidly, the first 
 helper will begin taking tests in order to follow the progress of the heat. These 
 tests he takes by securing a small test-spoon full of the metal, which he 
 pours into a small rectangular mould. As soon as the metal has solidified 
 in the mould, it is removed by jarring the mould while in an inverted 
 position; the test piece is nicked in the center, rapidly cooled with water, and 
 then, while still warm enough to dry itself, it is broken with a heavy 
 sledge hammer. From the fracture thus exposed, the carbon content, 
 which determines how the heat is to be treated, can be very accurately 
 estimated. In order that the temperature of the bath may be raised to a 
 point sufficiently high for tapping by the time the carbon is reduced to the 
 point aimed at, it is desirable that the carbon content of the bath at the end 
 of the lime boil should be forty to fifty hundredths of a per cent. (40 to 50 
 points) higher than that desired at tapping. 
 
 Control of Carbon and Temperature: If, as occasionally happens, 
 the carbon is nearly all removed while the bath is yet too cold to tap and pour 
 successfully, it is difficult, on account of its inactivity, to bring the heat 
 up to the proper tapping temperature without danger of burning, or over- 
 oxidizing the steel and unduly increasing the wear on the roof of the furnace. 
 A heat working under such conditions is known as a sticker. To prevent 
 this over-oxidizing, pigging up is resorted to, that is, the carbon content 
 is held, or kept constant, by adding pig iron, which also aids in raising the 
 temperature by producing a little boil in the bath. Usually, however, there 
 will be fifty to eighty points of carbon to be removed from the bath after the 
 lime boil. Therefore, as soon as the first helper sees that the lime is about 
 all up, he will first take a test, then see that all lumps of unfused matter, 
 or nigger heads, are melted and that the slag is sufficiently fluid. To 
 bring about the rapid melting of the unfused bodies and increase the fluidity of 
 the slag, fluorspar sufficient for the purpose will be added. Then to hasten the 
 elimination of the carbon, it may be necessary to ore down, that is, additions 
 of ore or heating furnace cinder will be made from time to time as required to 
 reduce the carbon content. After each addition of oxide has had time to act, a 
 test is taken. During the last half hour, in some cases the last hour, the heat 
 is in the furnace, no ore will be added. Some foremen erroneously believe that 
 the elimination of carbon at this point may be hastened by stirring the bath 
 
WORKING THE HEAT 225 
 
 with a long steel bar, a process known as shaking down, while others 
 will merely allow the metal to lie in the hearth undisturbed. In the case 
 of low carbon heats the flame will now be reversed in the furnace about 
 every ten minutes in order to raise the temperature, and as soon as the 
 tests show that the carbon is within three or four points of the desired 
 content, the melter, or foreman, is notified. He takes additional tests for 
 the carbon content and also for temperature, orders the recarburizers, 
 inspects the furnace, ladle, etc., and completes the arrangements for tapping 
 the heat. 
 
 Judging the Temperature of the Bath : For judging the temperature 
 of the bath, two very simple tests are employed by the furnacemen. One 
 of these tests consists of quickly inserting the end of a long steel bar or rod 
 into the bath of metal and slowly moving it from side to side until the 
 part immersed in the metal melts off. Then the bar is withdrawn, and 
 from the appearance of the hot end the condition of the bath with respect 
 to temperature may be judged. Thus, if the bath is too cold, this end of 
 the rod will be pointed; if too hot, it will show nicks on the sides near the 
 end; but if the temperature is right, the end of the rod will have melted off 
 so as to leave a clean, square end. The second method depends upon the 
 quite evident fact that the higher the temperature of a fluid the longer 
 it will remain fluid in contact with cold surroundings. It is carried out 
 simply by quickly withdrawnig a test-spoonful of the molten steel from 
 the bath and at once pouring it, rather slowly, but at a fixed rate of 
 flow, out of the spoon. The operator judges the temperature of the steel 
 by the way it flows and by the extent and thickness of the skull it leaves 
 in the spoon. By long practice with these methods the workmen become 
 very expert in making these relative determinations of temperature. 
 
 Tapping: The furnace should be manipulated so that a tapping tem- 
 perature is reached before the carbon content has been reduced to the 
 tapping point, as otherwise some difficulty will be experienced with high 
 carbon steels in holding the bath, if the carbon is to be caught on the way 
 down, while with low carbon steels, it will be difficult to reach a tapping tem- 
 perature or the metal will be over-oxidized, with the result that it will tend 
 to be both hot short and cold short unless deoxidizers such as spiegel, f erro- 
 manganese, or pig iron, are added. Prolonging the life of the heat at this point 
 in order to reduce the sulphur content is very bad practice for the double 
 reason that the removal of the sulphur is uncertain and the cure is worse than 
 the disease. The proper temperature for tapping low carbon heats is 1600 
 C., or a little higher, while for heats in which the carbon is caught on the 
 way down, the tapping temperature may be about 100 C., lower. To ac- 
 complish the tapping, the second helper digs out from the rear the mud plug and 
 most of the dolomite with which the tapping hole is closed, after which the hole 
 is opened by driving outward the dolomite remaining in it by inserting a 
 tapping rod through the wicket of the center door in the front of the furnace. 
 The steel then flows through the hole out of the furnace and down the 
 
OPEN HEARTH PROCESS 
 
 spout into the ladle. Since the tapping hole is on a level with the bottom 
 of the hearth, the greater part of the steel is out of the furnace before any 
 slag appears, and this fact permits of recarburization in the ladle. It is not 
 advisable for the recarburizing materials to be allowed to come into 
 contact with the slag, since some of the phosphoric acid in the slag may 
 be reduced and the phosphorus re-enter the steel. The tapping spout and 
 ladle are so placed as to direct the stream of molten metal a little to one 
 side of the center of the ladle, as the swirling motion tends to mix and make 
 more homogeneous the contents of the ladle. 
 
 SECTION V. 
 
 FINISHING THE HEAT MAKING STEEL FROM THE PURIFIED METAL. 
 
 Methods of Finishing the Steel: The process of finishing the steel 
 consists in making such additions as are required to produce the kind and 
 grade of steel desired, and with few exceptions these additions are made 
 immediately before and after tapping the heat. The methods of making 
 the necessary additions to produce the various kinds and grades of steel 
 differ somewhat, not only for the different grades but in different works 
 making the same grades. For example, the ferro manganese and spiegel 
 are preferably added to the steel in the ladle and in the molten state, but not 
 all plants are at present equipped to melt these materials, and ferro manganese 
 is still generally added in the solid form. For the plain steels the methods of 
 making additions for carbon and manganese may be briefly stated in the 
 following tabulated form. 
 
 High Carbon Steels. (C.60% to 1.30%) 
 
 Method I. Carbon is caught on the way down; ferro-manganese 
 is added, and coal, if needed, in the steel ladle. 
 
 Method II. The steel is tapped with the carbon at .10%; molten 
 spiegel mixture is added in the steel ladle. 
 
 Method III. The carbon in the bath is eliminated to .10%; sufficient 
 molten pig iron is added in the furnace at the time of tapping to raise the carbon 
 content almost to the point desired. Then ferro-manganese may be added to 
 the ladle to make up the deficit in the carbon, and supply the manganese 
 deficit left by the pig iron. 
 
 Medium Carbon Steels. (C. .30% to .70%) 
 Method I. The steel is tapped with a carbon content of .10%; molten 
 
 spiegel mixture is added in the steel ladle. 
 
 Method II. The carbon in the bath is eliminated to .10%; molten pig 
 
 iron is added in the furnace and ferro-manganese in the ladle as in III for high 
 
 carbon steels. 
 
 Low Carbon Steels. (C. less than .40%) 
 
 Method I. For dead soft steels the carbon content is reduced as low 
 as possible without danger of over-oxidizing the steel, and ferro-manganese 
 is added in the steel ladle. 
 
FINISHING THE HEAT 227 
 
 Method II. The carbon content is reduced to .10% and ferro-manga- 
 nese is added alone, or ferro-manganese and coal are added in the steel 
 ladle. In case a large quantity of ferro is required, some furnacemen 
 prefer to add a part of it in the furnace just before the tapping hole is opened. 
 
 Method III. For finishing very low carbon steels neither molten spiegel 
 nor molten pig iron are used on account of the difficulty of weighing and keep- 
 ing molten small quantities of these materials. But in large furnaces, pig 
 iron may be used, if the carbon content is more than .20%, as in Method II for 
 medium carbon steels. 
 
 Other Elements: The additions for other elements will be made about 
 as follows: Copper is added in the solid form to the steel fifteen to twenty 
 minutes before the heat is tapped. Sulphur is always added before the ferro 
 additions. Ferro silicon is necessarily added in the steel ladle; while it is the 
 common practice to add ferro vanadium by dropping it into the stream of 
 molten metal in the runner, or spout, as it is flowing into the ladle. As 
 nickel is chemically negative to iron, none is lost in the furnace, so nickel steels 
 are made by charging nickel steel scrap, then adding pig nickel in sufficient 
 amount to make up the deficiency, as shown by chemical analysis, about 
 thirty or forty minutes before tapping. The same practice may be 
 employed in the case of steel requiring copper and chromium, but a 
 comparatively large part of the latter element is lost through oxidation 
 in the furnace. Chromium is added in the form of ferro chromium. 
 
 Some Features that Make the Finishing of the Steel Difficult: It 
 
 requires considerable experience to finish steel properly, for there are a 
 number of circumstances that tend to complicate the operations. For 
 example, the addition of ferro, on account of its carbon content, will always 
 slightly raise the carbon content of the steel, though it is primarily added 
 to increase the manganese. Similar conditions also prevail in the use of 
 other ferro alloys and pig iron. Again, there is always a loss of the 
 elements added, except in the case of copper and nickel, and this loss, 
 different for each element, will vary with any one under different con- 
 ditions. Hence, the efficiency of these substances is never 100%. Furthermore, 
 it is seldom any of the elements can be obtained in pure form for this purpose, 
 and the substance containing the element sought may vary in its content of 
 that element. The various substances used, their efficiencies, and the amount 
 of each element present in the bath at the tapping of a normal heat are 
 indicated in the following table. 
 
228 
 
 OPEN HEARTH PROCESS 
 
 Table 31. Data Relating to Materials Used in Finishing Steel. 
 
 Material Added 
 
 Element 
 Sought 
 
 Percentage of Element 
 in Bath 
 
 Efficiency 
 When Added in 
 
 Furnace 
 
 Ladle 
 
 Pig Nickel . . 
 
 Ni 
 Cr 
 Cu 
 
 S 
 
 c 
 c 
 
 Mn 
 Mn 
 Mn 
 C 
 P 
 Si 
 V 
 
 .00 unless Ni. scrap used 
 .00 " Cr. " 
 .00 " Cu. " 
 .040 
 Any desired 
 
 .10 to .20 
 .10 to .20 
 .10 to .20 
 .10 to Any desired 
 .010 
 .00 
 .00 unless V. scrap used 
 
 98% 
 80% 
 99% 
 Never 
 
 95% 
 50% 
 50% 
 Never 
 
 Never 
 
 66%-70% 
 44%- 50% 
 Never 
 
 85%-90% 
 80%-90% 
 100% 
 75% 
 6570% 
 90% 
 
 Ferro Chromium 
 
 Pig Copper 
 
 Stick Sulphur 
 Anthracite Slack 
 
 Pig Iron 
 
 Ferro Manganese 
 Spiegel 
 
 
 Ferro Phosphorus 
 Ferro Silicon 
 Ferro Vanadium 
 
 
 In the following table will be found an analysis typical of each of these 
 substances. 
 Table 32. Representative Analyses of Materials Used in Finishing Steel. 
 
 
 Fe 
 
 % 
 
 C 
 
 % 
 
 Mn 
 
 % 
 
 P 
 
 % 
 
 S 
 % 
 
 Si 
 % 
 
 V 
 
 % 
 
 Cr 
 
 % 
 
 Ni 
 
 % 
 
 Cu 
 % 
 
 Ferro Manganese 
 ' Phosphorus 
 " Silicon, Electric 
 " Blast Fee 
 ' Chrome 
 
 13.03 
 79.97 
 49.44 
 87.465 
 22.05 
 42.65 
 75.68 
 
 6.80 
 1.10 
 .55 
 1.52 
 6.36 
 1.58 
 4.39 
 
 79.35 
 .18 
 .016 
 .42 
 .35 
 6.25 
 19.13 
 
 .16 
 18.00 
 .075 
 .080 
 .003 
 .010 
 .053 
 
 .65 
 .018 
 .015 
 .79 
 1.06 
 .028 
 
 .66 
 .10 
 49.90 
 10.50 
 
 .48 
 10.49 
 
 .72 
 
 37.96 
 
 69.96 
 
 
 
 
 
 
 97.00 
 
 99.00 
 
 Vanadium 
 
 Spiegel 
 
 Pig Nickel... 
 
 Pig Copper . 
 
 
 
 
 .17 
 
 
 Stick Sulphur . . 
 
 
 
 100.00 
 .030 
 
 
 Pig Iron 
 
 94.00 
 
 3.80 
 
 1.00 
 
 1.00 
 
 
 Given the weight and composition of the metal in the bath, the desired 
 composition of the finished steel, and the composition and efficiencies of the 
 substances to be added, the calculation of the amounts of the various additions 
 is a simple problem in arithemetic. 
 
 Teeming: As soon as the stream from the furnace no longer contains 
 any steel, the spout, or runner, is removed, and the steel ladle is lifted by 
 the crane and carried to the pouring platform, where the steel is teemed 
 into the ingot moulds ready to receive it. Teeming is not to be confused with 
 pouring. While the latter logically refers to the way the metal is let out of 
 the ladle, usage has made pouring synonymous with casting, which refers to the 
 manner of introducing the metal into the ingot mould. Thus, if the metal 
 
TEEMING AND SAMPLING 229 
 
 is introduced into the mould through its top, the resulting ingot is said to have 
 been top poured or top cast; but if through its bottom by means of runners, 
 the ingot is said to have been bottom poured or bottom cast. Teeming and 
 top pouring are accomplished in the following manner : The ladle is placed with 
 its nozzle over the center and about a foot from the top of the first mould in the 
 mould train, when the stopper is raised and the steel flows through the 
 nozzle into the mould below. In teeming the heats, care must be 
 taken that neither the stream of metal nor any part thereof be allowed 
 to strike the sides of the moulds, for these splashes of metal will adhere 
 to the mould, which quickly chills them, and, being coated on their surfaces 
 with a film of oxide, they may cause ingot defects which later appear as 
 slivers in the rolled steel. As the first mould is filled, the stream is stopped, 
 and by means of a hydraulic pusher the train is moved forward so as to 
 bring the next mould under the nozzle of the ladle. At some plants the 
 teeming is done from an overhead crane, which moves the ladle from mould 
 to mould. After the ladle has been emptied of steel, the slag remaining 
 in it is dumped into cinder cars and ultimately conveyed to the cinder yard. 
 In the meantime, the mould train is hauled to the stripper. On soft steel 
 and special heats, unless there is a high percentage of manganese or silicon 
 present, aluminum is thrown into the moulds, about two ounces to each ton, 
 in order to further deoxidize and quiet the steel. Aluminum is especially 
 effective in overcoming wildness because of its strong tendency to combine 
 with oxygen. Of this small amount added practically none remains in the 
 metal, so that this aluminum exerts no influence as an alloy in the steel. 
 
 Sampling: The sampling of the heat for chemical analysis is accom- 
 plished when the heat is half teemed by slackening the stream from the 
 ladle, whilst a spoon of suitable size is held under the nozzle and filled with 
 the molten metal, which is immediately poured into a test mould specially 
 designed for the purpose. The test mould may be of either one of two 
 types, which careful and extensive experiments have shown to give test 
 pieces most uniform in composition and most free from blow holes. One 
 of these types is a split mould that gives a test piece having a section \Y^ 
 inches square and a length of nearly 5 inches, with a flared opening about 2 
 inches deep to facilitate pouring. The other type is a small cup shaped 
 mould that gives a test piece 3% inches in diameter at the top and 2^ 
 inches at the bottom with a depth of 2^ inches. Upon being taken from 
 the mould, the test piece is immediately stamped with its heat number, 
 and is then delivered to the chemical laboratory for analysis. Experiments 
 have shown that a sample taken in this way most nearly represents the 
 average composition of the heat. 
 
 SECTION VI. 
 
 KEEPING THE FURNACE IN REPAIR. 
 
 Preparation of the Furnace for the Next Charge: After the runner 
 is lifted and thus detached from the furnace, the cinder and any steel that 
 remains in the furnace flow out of the tapping hole into the cinder pit. 
 
230 OPEN HEARTH PROCESS 
 
 The second helper must keep the tapping-hole open until everything that 
 can be removed from the furnace has flowed out. Fluorspar is usually 
 thrown in on the slag left to be sure that it flows out and does not build 
 up on the bottom of the furnace. Often, holes will be found in the bottom, 
 due to the intrusion of steel below the surface, which, boiling there, brings 
 up part of the basic material forming the bottom. Slag and steel are found 
 in these holes after tapping and must be rabbled out, so that the bottom 
 can be properly repaired. After all the steel and slag are removed from 
 these holes, they are filled up with dolomite. The gas is left on to keep 
 the slag and steel fluid during this process; but is shut off as soon as the 
 repairs to the bottom have been completed. Proceeding to the next step, 
 the second helper and cinder-pit man remove the steel that has chilled in 
 the tapping hole, rake out and free the hole of iron and close it up with 
 dolomite. A plug of clay is used to seal up the outside of the hole and hold 
 the dolomite in place. The banks, which have been cut by the slag from 
 the heat just out, are repaired by throwing burned dolomite on them (3000 
 to 4000 Ibs. is used after each heat in a 100-ton furnace); and the furnace 
 is then ready for charging again. 
 
 Furnace Troubles: In the operation of a furnace, troubles of a very 
 serious nature may occur at any time, unless the furnace is watched closely, 
 and carefully handled. These troubles present so many possibilities and 
 are so varied that space will permit of little more than an enumeration 
 of some of the more serious ones here. Thus, the tap-hole may break out 
 prematurely if it is not properly tamped and capped, or it may become 
 hopelessly clogged if it is not properly cleaned after each heat. Sometimes, 
 sections of the bottom become detached and rise, due to the buoyant force 
 of the metal, and when this occurs the heat must be tapped at once, and 
 no more heats may be charged until the damaged bottom is repaired. The 
 ports require constant attention to prevent them from building up or melting 
 down, and thus changing the angle of the flame, which would then tend to 
 over-heat some part of the furnace and would be rendered less effective 
 in heating the bath. Leaks may occur in the walls of the up-and-down- 
 takes, which result in the gas being burnt in part before it reaches the 
 hearth. The walls and roof often wear out long before the rest of the 
 furnace needs repairing. Roofs usually last for about 300 heats. The roof 
 can be repaired in a few hours, and a cave-in of the roof is of a serious 
 nature only when it falls in near the end of a heat. The most disastrous 
 mishap that can occur to a furnace is a break-out. Breakouts may be 
 caused by several things. A hole near a bank may not have been noticed 
 or may have been insufficiently repaired, in which case the steel works 
 down into it and gradually makes it deeper, until, finally, the metal finds 
 its way through the wall and out of the furnace. Sometimes, owing to a 
 thin spot on the banks or to slag having reached above them and worked 
 down into them, the slag gradually cuts its way out through the walls, in 
 which case it is usually followed by steel, as the hole soon becomes low 
 enough to reach the bath. Such mishaps are also known as break-outs 
 
FURNACE REPAIRS 231 
 
 and are always of a serious nature. Once a break-out occurs, the tapping 
 hole should be opened immediately, and as much as possible of the steel 
 gotten into the ladle or cinder pit. The spread of cinder and metal upon 
 the floor where the break-out has occurred can be limited usually by 
 throwing dolomite around it. Finally, after about 600 or 700 heats, the 
 checker work has become so badly clogged and the brick work is so eaten 
 away, that it becomes necessary to close down the furnace for general 
 repairs, during which the greater part of the brick work may be torn out 
 and rebuilt. 
 
 Repair Materials: It is evdent that, for making up the bottom and 
 for doing the repair work about a furnace, much depends upon the materials 
 employed. Great care must always be exercised to see that they are of 
 the right chemical composition, and best suited for the work in nand, as, 
 otherwise, the best of workmanship in making the repairs will go for naught. 
 Therefore, a few remarks in this connection should be of interest. 
 
 Dolomite is found in local deposits similar to those of limestone. Like 
 the latter it varies in composition through quite wide ranges, but that 
 suitable for open hearth work will have, after being calcined, approximately 
 the composition shown by the following chemical analysis: Silica, SiC>2, 
 1.66%; Iron Oxide, Fe 2 O 3 , .94%; Alumina, A1 2 O 3 , 1.24%; Lime, CaO, 
 50.01%; Magnesia, MgO, 35.26%. 
 
 Magnesite: The magnesite used before the European war was im- 
 ported from Madelein and Budapest, Austria, and was brought to the mill 
 already burned and ground. It was used on the furnace bottom, banks and 
 ports, and in repair work. An average analysis of thirty-eight cars of the 
 imported material is as follows: 
 
 SiO 2 Fe Mn A1 2 O 3 CaO MgO Ig.Loss 
 
 2.07% 6.00% .37% 1.63% 3.81% 84.11% .52% 
 Since the outbreak of the war, however, deposits of this material in 
 California and Washington have been opened, and this domestic supply 
 promises to replace permanently the imported material. This magnesite is 
 purer than the imported, and for that reason it does not sinter or bond so 
 readily, but by mixing a little of the proper fluxing material with it, this 
 drawback has been easily overcome. 
 
 Chrome Ore: Chrome ore is still imported, as the limited deposits 
 so far discovered in the United States and Canada are of an inferior grade. 
 It is received in the form of small lumps. It is ground and mixed, in a wet 
 pan, with one-half magnesite, and is used in repair work where a neutral 
 substance is required, such as in patching flues, tapping holes, ports, etc. 
 An analysis of an average sample of a satisfactory grade of this ore gave 
 these results: 
 
 SiO 2 FeO MnO A1 2 O 3 MgO Cr 2 O 3 Ig. Loss 
 9.02% 13.50% .80% 10.82% 19.89% 42.66% 2.92% 
 
232 OPEN HEARTH PROCESS 
 
 Besides these materials, some ganister may be employed at some of the 
 works, while all plants will use large quantities of loam and of fire clay for 
 lining furance spouts and ladles, for making up stoppers, and for other 
 repair work of minor importance. 
 
 SECTION VII. 
 
 CHEMISTRY OP THE BASIC PROCESS. 
 
 Some of the Principles and Conditions Involved: Having followed 
 the procedure of making steel by this process, the reader should be interested 
 in a discussion of a subject, which to the metallurgist, at least, represents 
 the most interesting and profitable part of the study, namely, the chemistry 
 of the process. In beginning this study it should be recalled that the 
 purification of pig iron, which is the first of the two main steps in making 
 steel, includes the elimination from the metal of the four elements, silicon, 
 manganese, phosphorus and carbon, and that the principle by which this 
 elimination is effected is that of oxidation. In basic open hearth processes, 
 the elimination of sulphur may also take place to a greater or less extent, 
 depending upon the amount present, but is never to be considered seriously 
 as a principal objective. It now remains to be pointed out that this 
 oxidation, when brought about indirectly, that is, through the interaction 
 of these elements with oxygen bearing compounds, as is the case in this 
 process, involves two other principles as well. These are the principles 
 of reduction and neutralization, for it is manifestly impossible under these 
 conditions that one substance can be oxidized without another's being 
 reduced, and it develops, as will be shown later, that this interaction is 
 made possible through the immediate neutralization of the oxidized sub- 
 stances. While these principles and the reactions by which the purification 
 is brought about are, when considered separately, very simple and can be 
 easily understood, they are somewhat difficult to follow in the actual 
 working of the furnace, because they are here occurring simultaneously 
 and, therefore, tend to mask each other in the effects they produce. For 
 this reason it is best to consider the subject, first, from the standpoint of 
 the chemical properties of the elements affected and of the oxygen com- 
 pounds of these elements under the conditions of the basic open hearth 
 process. 
 
 Properties of Iron and Its Oxides: One of the most marked of the 
 chemical properties of metallic iron is its tendency to combine with oxygen. 
 Even at ordinary temperatures this tendency is very marked, as is seen 
 from the ease and quickness with which it combines with oxygen and water 
 to form the familiar compounds known commonly as iron rust. These 
 
CHEMISTRY OF THE PROCESS 233 
 
 compounds are but the hydrated sesqui-oxide, or per-oxide, of iron contain- 
 ing varying amounts of combined water, as represented by the formula 
 Fe2O3*xH2O. This tendency of iron and oxygen becomes stronger as the 
 temperature rises, so that at a temperature ranging from 800 to 900, or 
 higher, the combination becomes very rapid, and a compound quite different 
 from those composing rust is formed. It is commonly known as Scale, 
 and is represented by the formula FesO^ or FeOFe2C>3. These facts help 
 to explain why iron is seldom found in nature uncombined, and the two 
 compounds, represented by the formulas given above, together with the 
 carbonate of iron, constitute the valuable part of all the ores of iron. The 
 ore used in all our furnaces is the red hematite, which for the purpose of 
 this discussion, may be considered as being composed of the sesqui-oxide, 
 Fe2C>3, and gangue. Besides free oxygen, certain compounds may, at high 
 temperature, serve as sources of supply of oxygen to iron. Among these 
 are carbon dioxide and water vapor, which constitute the chief products 
 of combustion in any case of burning a fuel in the presence of an excess of 
 oxygen, such as normally exists in an open hearth furnace. The heating 
 of free iron to these high temperatures in contact with either free oxygen 
 or steam always results in the formation of FeaO^ according to the following 
 reactions: 3Fe+2O 2 =Fe 3 O 4 , 3Fe-r-4H 2 O=Fe 3 O4+4H2. But at a tem- 
 perature of 1000 C. or more, with iron in contact with carbon dioxide, another 
 and less common oxide, FeO is formed, thus Fe+CO2=FeO-hCO. Ferrous 
 oxide, FeO, and ferroso-ferric oxide, FeaO^ may be formed in the furnace 
 in other ways, also, one of which is by the progressive reduction of Fe2C>3. 
 If Fe2(>3 be heated to a high temperature it loses oxygen and is converted 
 into FeaO4. This change takes place at temperatures between 1100 and 
 1200 C., some 500 degrees below the maximum temperatures of the open 
 hearth. If FeO be formed under conditions even only slightly oxidizing, 
 it passes into FesO4. Another difference in the properties of these oxides, 
 which is of great importance in considering the chemistry of the open hearth, 
 is seen in their power to neutralize acids. Thus, while both Fe2Os and FeO ex- 
 hibit very marked basic properties and combine rapidly with acid oxides, 
 FesO^ cannot be induced to form corresponding salts at the temperatures that 
 prevail hi the open hearth. Pure scale, FeaO^, fuses at about 1450 C., a temper- 
 ature easily attainable in the open hearth, and it dissolves readily in either 
 iron or calcium silicates. Ferrous oxide, FeO, is soluble in both the molten 
 iron and the slag, and though the amount that remains dissolved in the metal 
 when solid is small, being seldom present to an extent greater than .315% 
 the equivalent of .07% oxygen, its effects are very harmful, as it produces 
 both red and cold shortness in the metal. With metallic manganese it gives 
 the following reaction: FeO+Mn=Fe-f-MnO. MnO is not soluble in the 
 molten metal, which fact assists in accounting for the efficiency of manganese 
 as a deoxidizing agent. 
 
 The Importance of Ferrous Oxide, FeO, in the Part Played by the 
 Oxides of Iron in the Process: From what has been said concerning the 
 properties of the three oxides of iron, it is evident that ferrous oxide, FeO. 
 
234 
 
 OPEN HEARTH PROCESS 
 
 is the principal, perhaps the only, direct oxidizing agent in the open hearth 
 process. Although iron sesquioxide, FesOa, may be charged into the 
 furnace, much of this oxide is transformed by the heat into ferroso-ferric 
 oxide, FesO4, before it has an opportunity to become active. Besides, 
 since the impurities are held in solution by the metal, either the oxidizing 
 agent must dissolve in the metal, a condition that is not true for either FesO4 
 or FesOs, or the oxidation of the impurities must occur at the surface of contact 
 between metal and slag. That conditions in the open hearth during the melt- 
 ing period tend to form an abundant supply of ferrous oxide can be shown by 
 an analysis of the first slag formed, as is illustrated by the following analyses 
 of samples of this slag taken just before the introduction of the hot metal. 
 
 Table 33. Analysis of First Slag Formed in 
 Open Hearth Heats 
 
 
 Si0 2 
 
 FeO 
 
 Fe 2 3 
 
 MnO 
 
 P 2 5 
 
 A1 2 3 
 
 CaO 
 
 MgO 
 
 S 
 (S+S0 8 
 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 % 
 
 /o 
 
 Slag from. Fee., 
 
 
 
 
 
 
 
 
 
 
 No. 9 
 
 8.54 
 
 61.05 
 
 11.10 
 
 2.31 
 
 .26 
 
 1.98 
 
 9.13 
 
 5.48 
 
 .16 
 
 Slag from Fee., 
 
 
 
 
 
 
 
 
 
 
 No. 15 
 
 1.00 
 
 78.24 
 
 15.31 
 
 .81 
 
 .14 
 
 .37 
 
 2.70 
 
 1.12 
 
 .25 
 
 Concerning the neutralizing powers of these oxides, FeO must also act 
 as the initial base, as will be shown later, but in the slag Fe2Os is also 
 capable of acting as a base. It is important to emphasize these facts here 
 because of their influence on the chemical action of the other elements 
 eliminated during the purification period, for their action depends on the 
 conditions, which it is, therefore, essential to define. Briefly, the chief of 
 these conditions is that at the time of introducing the hot metal there is 
 present in the furnace a slag that is very rich in ferrous oxide. 
 
 Properties of Silicon and Its Oxide, Silica: Silicon forms but one 
 compound with oxygen under the conditions prevailing in the open hearth, 
 and this compound is silica, SiC>2. The tendency of silicon to combine 
 with oxygen is even stronger than that of iron, due to the greater heat of 
 formation of its oxide, so that it is capable of reducing any of the oxides 
 of the latter, and upon this fact depends the elimination of this element 
 from the molten metal. As to which of the oxides of iron is the active 
 agent in the oxidation of silicon, there can be little doubt but that ferrous 
 .oxide, FeO, is the principal one that suffers direct reduction by the silicon, 
 for, as already indicated, the silicon, either as an alloy or a compound of 
 
CHEMISTRY OF THE PROCESS 235 
 
 iron, is distributed throughout the mass of molten metal, and it is necessary 
 that either the oxidizing agent also dissolve in the liquid or the silicon diffuse 
 to the surface of the metal in order that the molecules may be brought into that 
 intimate contact required to effect a reaction. The oxidation of the silicon, 
 then, is represented by the following reaction: Si-j-2FeO=SiO2+2Fe. 
 Silicon, however, is a very strong acid, so that if this reaction occurs, the silica 
 will immediately combine with ferrous oxide to form a bisilicate, a tri-silicate 
 or some still more acid salt, depending upon the relative amount of base 
 available. Since this oxidation takes place with only a limited supply of ferrous 
 oxide present, it may be assumed that the salt requiring the least base, such as 
 the tri-silicate, would be formed. If so, the reaction would be as follows: 
 3 SiO 2 -f-2FeO=(FeO) 2 -(SiO 2 ) 3 . If the bisilicate is formed, the reaction 
 would be represented thus: SiC>2-|-FeO=FeO'SiO2. Since the neutral- 
 izing action must always instantly follow the oxidation, it is perhaps best 
 to represent the change by a single reaction, which can be done by com- 
 bining the first two, thus: 3Si+8FeO=(FeO) 2 ' (SiO 2 ) 3 +6Fe. The ferrous 
 silicate is in the fluid state, for its fusion point is below that of the metal. 
 It is of lower density than the iron, and, therefore, rises to the surface, 
 where it temporarily forms a part of the slag. On the way to the slag it 
 may undergo a change as noted later under manganese. Once in the slag, 
 this ferrous silicate is capable of undergoing many changes. The first of 
 these changes is probably due to the ability of the silica to take on additional 
 base. Having been formed in a region where there is only a limited supply 
 of base, the silica could not be neutralized to the extent it is capable. But 
 now, having diffused into the slag, where there is abundance of bases, this 
 trisilicate may become a monosilicate with either a monoxide or a sesqui- 
 oxide base. The formation of the monosilicate with the monoxide base, 
 FeO, may be represented thus: (FeO) 2 -(SiO 2 ) 3 -|-4 FeO=3(FeO) 2 -SiO 2 , 
 The ferrous oxide thus combined cannot act as an oxidizer and, therefore, 
 becomes inactive. However, it may be made available through the action 
 of lime and magnesia. Both these bases are capable of replacing ferrous 
 oxide in the manner indicated by the following reactions: 
 
 2 Mg0 MgO) 2 -Si0 2 +2 F 
 
 The ferrous oxide thus liberated is now subject to reduction, and available 
 to the bath for further use. The sources of supply of lime and magnesia 
 are the stone charged into the furnace and the lining of the furnace itself. 
 These alkaline earth silicates constitute the major portion of all the final 
 slags formed in the process, but these slags are never free of iron, for they 
 hold its oxides in solution. 
 
 Properties of Manganese and Its Oxides: While manganese com- 
 bines with oxygen in several different proportions to form an equal number 
 of different oxides, under the conditions that exist in the basic open hearth 
 only one of these oxides is formed, namely, the manganous oxide, or 
 protoxide of manganese, MnO. Like silicon, the manganese in the charge, 
 
236 OPEN HEARTH PROCESS 
 
 being alloyed with iron,must be oxidized largely through the agency of ferrous 
 oxide. But as silicon is capable of reducing manganese oxide, there appears 
 little chance of oxidizing the latter until the former element has been largely 
 eliminated from the bath. However, there is much evidence to show that 
 at least a part of the manganese finds its way into the slag long before all 
 the silicon has been oxidized. This fact is explained by the assumption 
 that manganese is capable of replacing iron in the silicates of iron, thus: 
 FeO'SiO 2 +Mn=MnO-SiO 2 +Fe.or (FeO)2-(SiO 2 )3+2Mn=(MnO)2 > (SiO2)3 
 +2 Fe. With this idea in mind, it is easy to conceive the simul- 
 taneous elimination of both these elements, in which elimination the ferrous 
 oxide plays the part of oxidizing agent, only, and manganese fulfills the 
 office of the base for the neutralization of the silica. Such a change, 
 involving the simultaneous oxidation of silicon and manganese, is represented 
 by the following reaction: 3 FeO+Si+Mn=MnO-SiO 2 +3Fe. When this 
 silicate of manganese reaches the slag, it is subject to the same changes as 
 are the corresponding iron oxide silicates, the manganese oxide being 
 eventually set free by lime and magnesia. This free oxide of manganese, 
 being insoluble in the metal, remains in the slag as such as long as the 
 latter is rich in iron oxides; but if the slag should be depleted of its oxides, 
 then manganous oxide is liable to reduction, in which event the resulting 
 metallic manganese returns to the bath. Another property of manganese, 
 though it is of little importance in ordinary open hearth operations, may 
 be mentioned. It refers to the ability of manganese to replace iron in 
 combination with sulphur. Thus, all the sulphur contained in the pig iron 
 or steel scrap going into the furnace may be considered as being combined 
 with this element and in the form of manganese sulphide. This substance 
 is slightly soluble in the slag as well as in the metal, and this fact accounts 
 for the presence of small amounts of sulphide found in open hearth slags. On 
 the surface of the slag, in contact with an oxidizing flame, manganese sulphide 
 is subject to oxidation according to this reaction: 2 MnS+3O 2 =2 MnO+ 
 2 SO 2 . The sulphur dioxide, SO 2 , thus formed is a gas and may escape from 
 the furnace with the products of combustion. However, it is evident that 
 the quantity of sulphur removed in this way must be very small. 
 
 Sulphur and Its Oxides: Owing to the peculiar properties of sulphur 
 and its oxides, they are subject to a number of conflicting influences, under 
 the conditions of the open hearth process, that render the removal of this 
 element very uncertain. As an element, sulphur combines directly with 
 iron to form iron sulphide and is easily oxidized to form oxides, SO 2 and 
 SOs, both of which are gaseous acid anhydrides, and, when neutralized, 
 form sulphites and sulphates, respectively. At temperatures far below the 
 lowest working temperature of the open hearth, the sulphites and sulphates 
 of the heavier metals, like iron, for example, decompose to form either the 
 sulphide or the oxide of the metal and sulphur dioxide. At temperatures 
 relatively low for furnace operations, like that of the puddling furnace, both 
 manganese and iron sulphides are readily oxidized by the higher oxides of 
 
CHEMISTRY OF THE PROCESS 237 
 
 these elements, such as Fe2O3, forming oxides of the metals and SC>2 which 
 escapes, as a gas, from the furnace. At the higher temperatures of the open 
 hearth there are a number of factors that operate against the elimination 
 of the sulphur in this way, among which may be mentioned an increased 
 tendency of iron to combine with sulphur, an increase in the reducing power 
 of the molten iron, the fact that CO gas is capable of reducing SC>2, and the 
 probability that there is little Fe2Os available to do the work. Fe3O4 
 may replace Fe2Os in the oxidation, but it is very improbable that FeO 
 is capable of producing the same result. Unlike the sulphates of the heavy 
 metals, the sulphates of the alkaline earths, such as calcium sulphate, are 
 not decomposed by heat alone, at least not by any temperature attainable 
 in the open hearth. Therefore, once the sulphur is oxidized and thus com- 
 bined with lime, there is some chance of its being held by the slag. How- 
 ever, iron is capable of decomposing the sulphate of lime, thus, CaSO4+ 
 4 Fe=FeS+CaO+3 FeO, in which case the iron sulphide dissolves in the 
 iron. As evidence that such a reaction may take place, several instances 
 may be cited in which steel has been ruined, for the order it was intended, 
 through charging old boiler tubes, containing much boiler scale, with the 
 scrap. The presence of oxides in the slag tend to hold this reaction in 
 check, so that it takes place to an appreciable degree only when the slag 
 is burdened with an excessive amount of this sulphate, and even then it 
 can occur only at the surfaces of contact between metal and slag. 
 
 Sulphur From the Fuel: Another source from which sulphur may be 
 imparted to the metal is the fuel. That fuel carrying compounds of sulphur 
 may be responsible for a portion of the sulphur content of steel is a well known 
 fact, but through what reactions the transfer is brought about does not appear 
 to have been satisfactorily explained. Now, it has been established by J. 
 B. Ferguson, writing in the Journal of the American Chemical Society, Novem-' 
 ber, 1918, that "CO andSO 2 react between 1000 C. and 1200 C. to form CO 2 
 and sulphur vapor and traces of carbon oxysulfide in mixtures rich in CO." 
 In the early stages of an open hearth heat, just after the addition of the molten 
 pig of the charge, these conditions as to temperature and presence of CO gas 
 prevail at the surface of the charge, and any sulphur vapor that may be formed 
 as above will readily be taken up by the exposed molten or solid metal of the 
 charge, forming iron sulfide. If there is taken into account the action of man- 
 ganese toward sulphur, there are, then, two agencies that act feebly to eliminate 
 sulphur from the metal, and two that are active, also feebly, in returning 
 it, or of introducing it. The stability of the calcium sulphate, however, 
 acts as a guard against the introduction of the element, except under some 
 such unusual condition as that noted above. 
 
 Phosphorus and Its Oxides: This element is very easily oxidized, 
 when in the free state, by oxygen alone, and forms several oxides, of which 
 only one, phosphorus pentoxide, P2O 5 , need be considered here, because it 
 is the only one formed under the conditions prevailing in the open hearth. 
 Like sulphur, phosphorus occurs in the metal as a definite compound, iron 
 
238 OPEN HEARTH PROCESS 
 
 phosphide, Fes?, and like silica, the oxide, P2O 5 , is an acid, which must 
 be neutralized as soon as it is formed. The reaction by which it is removed 
 from the metal is, therefore, probably most nearly correctly represented 
 by the following expression: 2 Fe 3 P+8 FeO=(FeO) 3 .P 2 O5+ll Fe. Silica 
 has the power of replacing ^2^5 in the ferrous phosphate, thus exposing the 
 latter oxide to reduction, so that phosphorus is never permanently removed 
 from the metal until the silicon has been practically all eliminated. This 
 power of silica also accounts in part for the fact that phosphorus is not 
 eliminated by any of the acid processes for making steel, for the proportion 
 of this compound in the slag effectually prevents the formation of the 
 phosphate. In the basic process the abundance of bases present in the 
 slag is more than sufficient to satisfy the silica, so that the ferrous phosphate 
 is not only permitted to form, but on reaching the slag it is converted 
 into a much more stable calcium phosphate, probably the tri-calcium 
 phosphate, (CaO)3*P2O 5 . Even this salt is, relatively speaking, easily 
 reduced. Phosphorus, therefore, is held by the slag only so long as the 
 latter is maintained strongly basic and at least moderately oxidizing. 
 
 Carbon and Its Oxides: Owing to the peculiar chemical and physical 
 properties of carbon and its oxides, the elimination of this element gives 
 rise to phenomena distinctively different from those of the elements just 
 reviewed. In that review it was pointed out that the oxidation of those 
 elements gives compounds which are liquids under the conditions of the 
 open hearth, that is, they are slag forming elements. But the oxidation 
 of the carbon, which is represented by the reactions C+FeO CO4-Fe and 
 Fe3C+FeO=CO+4 Fe, gives rise to the gas carbon monoxide, and, owing 
 to the conditions under which it takes place, produces the phenomenon 
 known as the ore boil. Thus, since the carbon, either as a compound or 
 as an element, is dissolved in the metal, and the iron oxides, in the slag, 
 the region of greatest activity, at the beginning of the oxidation, is located 
 near the surface of contact between the two liquids. The generation of 
 the carbon monoxide here gives rise to innumerable tiny bubbles of the 
 gas, which immediately rise into the slag; but owing to the small size of 
 the former and the viscosity of the latter, their immediate escape is 
 hindered, so that they find their way to the surface very slowly. They 
 thus collect in the slag, increasing its volume and imparting to it the appear- 
 ance of foam. In the course of time, the highly oxidizing condition of the 
 bath has disappeared with the consequent lowering of the carbon content, 
 and both oxide and carbon are so reduced in amount that the oxidation 
 no longer takes place rapidly and near the surface of the metal; so the slag 
 loses the foamy appearance. Indeed, as the silicon, manganese, phosphorus 
 and part of the carbon have been oxidized, the bath of metal is becoming 
 depleted of its reducing agents, so that more and more ferrous oxide pene- 
 trates or is dissolved by the metal, which fact, together with the decom- 
 position of the limestone, gives rise to the formation of large bodies, or 
 bubbles, of carbon monoxide deep down in or near the bottom of the layer 
 
CHEMISTRY OF THE PROCESS 239 
 
 of metal. These bubbles rise through the metal rapidly, so that when they 
 strike the slag, it is not given time to part, but is lifted into the atmosphere 
 of the furnace and thrown to one side. 
 
 The Action of the Limestone: It is interesting to note to what 
 extent the decomposition of the limestone on the bottom of the furnace 
 may contribute to this action and to the elimination of the carbon. The 
 reaction representing this decomposition is CaCO3=CaO-{-CO2. Now, the 
 carbon dioxide gas is no sooner set free than it is attacked by iron, thus: 
 CO 2 +Fe=FeO+CO. The FeO is then available for the oxidation of the 
 carbon in the metal, thus: FeO+C=Fe+CO. By combining these two 
 reactions it will be observed that, from each and every volume of carbon 
 dioxide, CC>2, liberated, two volumes of carbon monoxide, CO, result. The 
 CO derived from the decomposition of the limestone, as well as that from 
 the action of dissolved FeO, escapes to the surface as just described. Any 
 ore that might have remained at the bottom of the furnace up to this period 
 of the carbon elimination would also contribute to the violence of this 
 action. At the surface of the slag the carbon monoxide discharged by 
 the bath may burn to carbon dioxide, which escapes with the products of 
 combustion from the flame. The boiling of the bath thus plays a very 
 important part in the process. When the slag is thrown aside by the bubbles 
 of gas, t he metal is exposed to the action of the flame, and though this 
 exposure is but momentary in each instance, the large number of such 
 exposures result in the formation of a considerable quantity of ferrous oxide 
 in this way. But the greatest benefits are derived from the agitation of 
 the bath. It is easily seen how this agitation must result in a mixing of 
 the slag and metal, thus increasing the area of the reacting surfaces, while 
 the stirring effect on the metal itself should not be overlooked. Thus, the 
 metal lying near the bottom, which is the coldest and most impure, is 
 brought upward to be heated and exposed to the oxidizing influences, so 
 that both the temperature and composition of the bath are kept more uniform 
 than they could otherwise be maintained. 
 
 Effect of Carbon Elimination on Slag Composition: In conclusion, 
 it should be pointed out that the effect of the carbon elimination upon the 
 slag is to reduce its content of iron oxides. By proper regulation of the 
 conditions this reduction of the oxides in the slag may be brought down to 
 the point where the total iron content of the slag will be about 10% of its 
 weight, of which iron about 7/10, or 70%, will be in the ferrous condition. 
 The advantages of adding ore to the charge, as may now be readily seen, 
 are to increase the speed of the purification and to decrease the waste 
 of metal, which is more expensive than ore. 
 
 The Order of Elimination of the elements just reviewed, with the 
 exception of sulphur, is the same as the order in which they have been 
 discussed, namely, silicon and manganese, phosphorus, and lastly carbon. 
 Some reasons why the first three elements are eliminated in this order 
 have been mentioned under their respective headings, but nothing has been 
 
240 OPEN HEARTH PROCESS 
 
 mentioned that would appear to cause carbon, which is capable, under 
 proper conditions, of reducing the compounds of all these elements, to be 
 the last element oxidized in the open hearth. The explanation for this 
 difference in the chemical properties of carbon is connected with the fact 
 that its reducing power increases as the temperature rises. Again, chemical 
 action, when it occurs independently of external influences, always takes 
 place in the direction that will liberate the most energy, as was pointed 
 out in Chapters I and VII. The oxidation of silicon, manganese, and 
 phosphorus and the neutralization of the resulting oxides are exothermic 
 reactions, whereas the carbon reaction is endothermic. The elimination 
 of the four impurities thus takes place in accordance with the amounts of 
 heat evolved or absorbed. As an example of these laws, let the elimination 
 of silicon and carbon be compared. These reactions with the heat, or 
 energy, values involved, are as indicated in the following expressions: 
 
 (l)Si+2FeO= Si O 2 +2Fe (+64600 cal.) 
 
 2(65700) +196000=64600 
 (2) C+FeO=CG+Fe (36540 cal.) 
 65700+29160= 36540 
 
 Reaction (1) shows that in the oxidation of one gram of silicon approxi- 
 mately 2307 cals. (64,600H-28=2,307) of heat are evolved, while in the 
 oxidation of one gram carbon, as shown by reaction (2) 3, 045 cal. of heat are 
 absorbed. If now the reduction of silica and carbon monoxide by carbon 
 and silicon, respectively, be compared as in reaction (3) and (4), it will 
 be seen that, whereas carbon absorbs heat in reducing silica, silicon reducing 
 CO, liberates heat. (3) SiO 2 +C=2 CO+Si( 137640 cal.) (4) 2 CO+Si= 
 
 196000 +2x29160 2x29160 
 
 Si(>2+C(+137640 cal.) It is evident, then, that the oxidation of the carbon 
 +196000 
 
 cannot be complete until the silicon has been eliminated. At high tem- 
 peratures, such as may prevail in parts of the blast furnace or in the electric 
 furnace, for example, the heat absorbed in reaction (3) is supplied from 
 external sources, which addition of energy causes the carbon to act as a 
 reducing agent toward the silica. What has been said with respect to 
 silicon also holds true in the case of manganese and phosphorus. In the 
 basic open hearth the temperature rises gradually, so that carbon has no 
 opportunity to act as a reducing agent toward oxides of these elements. 
 
 Factors Opposing this Order of Elimination: What has been 
 written above should not be taken to mean that each element is completely 
 and successively eliminated in the order mentioned, for there are other 
 laws, such as the law of mass action, for example, that operate to bring 
 about the elimination of these elements simultaneously. The oxidation of 
 the carbon, for example, evidently begins shortly after the hot metal has 
 
CHEMISTRY OF THE PROCESS 241 
 
 been added to the charge, and certainly before the manganese and phos- 
 phorus have been entirely disposed of. What is implied is that the elimi- 
 nation of each element in the order named is successively in the ascendency 
 until eventually only the carbon, in part, remains to be oxidized. When 
 this element has been practically all removed, the bath of metal no longer 
 contains reducing agents and is subject to over-oxidation by absorption of 
 ferrous oxide, FeO, up to the point of saturation in equilibrium with the 
 slag. This fact explains why it is undesirable to make ore additions to 
 the slag just previous to tapping, and also why the heat, unless deoxidizing 
 agents are added to the bath, should not be held in the furnace for more than a 
 few minutes after the carbon content has been lowered to .10%, which figure 
 is within about .03% of the minimum carbon content for this process. 
 
 Resume: All that should now be required in order that the chemistry 
 of this process may be fixed clearly in mind, is a rapid review of the subject 
 matter included under the heading of Operation of the Furnace, which will 
 now appear in a new light. To begin this review, picture a furnace in the 
 course of operation which has received its charge of solid materials for, 
 say, a Monell heat. The first effect on this charge will be an increase of 
 temperature. The limestone, ore and the lining of the furnace all being 
 basic in character, will remain inactive at first, and continue so until they 
 will have absorbed sufficient heat to raise their temperature to the point 
 where decomposition begins. For limestone this temperature is about 
 850 C., while the ore will not give up its oxygen until its temperature is 
 near the fusion point, about 1400 C., unless it comes in contact with reducing 
 agents. The absorption of heat by the ore and stone is hindered by the 
 scrap charged upon them. This material, being a good conductor of heat 
 and exposed to the flame, absorbs heat very rapidly, and as soon as the 
 temperature rises above the thermo-critical range, oxidation of the iron 
 begins, this giving rise to the formation of scale. The melting point of 
 this scale is so near that of the metal, that it may remain on the surface 
 until the metal itself begins to melt. It is understood, of course, that the 
 impurities contained in the scrap will suffer oxidation with the iron. These 
 fluids will trickle down over the colder material beneath and will eventually 
 reach the ore on the bottom of the furnace. Here, together with additional 
 oxide derived from the ore and some silica, lime, etc., collected from various 
 sources, this molten scale will go to make up the first slag. This slag, poor 
 in silica, but exceedingly rich in iron oxides, especially ferrous oxide, and 
 containing some lime also, is well constituted for the work it has to do; 
 and with the addition of the hot metal, the purification may begin at once. 
 Thus, the silicon, manganese and phosphorus will have been practically 
 eliminated from the metal within two hours after the molten metal is 
 charged. The extent and character of the purification of the metal at 
 the time of the run off with the resulting change in the composition of 
 the slag are indicated in the following table of analyses. 
 
242 
 
 THE OPEN HEARTH PROCESS 
 
 Table 34. Analyses of Hot Metal and Slag Before Charging and at 
 Time of First Run=off. 
 
 ANALYSIS OF METAL. 
 Per cent, of 
 
 PARTIAL ANALYSIS OF SLAG. 
 Per cent, of 
 
 Heat 
 
 No. 
 
 
 C 
 
 Mn 
 
 P 
 
 S 
 
 Si 
 
 SiOa 
 
 FeO 
 
 FesOs 
 
 MnO 
 
 CaO 
 
 MgO 
 
 PaOs 
 
 80s 
 
 S 
 
 1 
 
 1 
 2 
 3 
 
 4 
 
 5 
 
 Pig Iron Before 
 Charging 
 
 At Time of Run Off.. 
 
 3.85 
 2.39 
 2.41 
 2.45 
 
 2.80 
 
 3.74 
 
 1.55 
 .05 
 .02 
 .02 
 .01 
 
 .01 
 
 .198 
 .022 
 .053 
 .068 
 .015 
 
 .043 
 
 .035 
 .040 
 .060 
 .049 
 .043 
 
 .037 
 
 1.04 
 .04 
 
 4.72 
 19.19 
 25.18 
 23.68 
 15.74 
 
 19.30 
 
 66.67 
 32.86 
 17.39 
 26.33 
 45.16 
 
 42.59 
 
 not 
 det'md 
 
 5.22 
 4.07 
 7.10 
 7.50 
 
 5.91 
 
 1.30 
 12.97 
 13.16 
 13.54 
 6.19 
 
 6.84 
 
 18.00 
 18.38 
 17.88 
 12.22 
 11.34 
 
 12.41 
 
 2.00 
 6.11 
 12.18 
 10.14 
 5.23 
 
 3.84 
 
 .78 
 1.11 
 
 .090 
 .090 
 .165 
 
 .182 
 
 .029 
 .027 
 .037 
 
 .063 
 
 Briquettes Instead of 
 Ore Used 
 
 
 
 From this point the reader should be able to continue this review 
 through the oxidation of the carbon unaided, and in doing so, he will have 
 fixed in mind the chemistry of the process much more firmly than if he 
 but read the inadequately expressed thoughts of another. As a further aid, 
 what is said in Chapter V concerning final open hearth slags should be 
 referred to. 
 
ELECTRIC PROCESS 243 
 
 CHAPTER IX. 
 
 MANUFACTURE OF STEEL IN ELECTRIC FURNACES. 
 
 SECTION I. 
 
 INTRODUCTORY. 
 
 The Plan of Study: To understand the process of manufacturing 
 steel by means of the electric furnace requires some knowledge of electrical 
 phenomena. To the question as to what electricity is, no very satisfactory 
 definition can be given. The modern idea is, that it is the fundamental 
 of which all matter is composed, because there is evidence to indicate that 
 electrons are but corpuscles of negative electricity, which, together with 
 nuclei of positive electricity, go to form atoms. That it is either a form 
 of energy or an agent for transmitting energy is evident, and the various 
 phenomena attending it may be due to the production of strained conditions 
 in the Ether, somewhat similar to the effect of heat upon water in the 
 generation of steam. However, the question is of no importance except to 
 distinguish the thing from the phenomena produced by it, which are of 
 great importance. To those who have not been able to devote much time 
 or study to the subject, these phenomena appear as deep mysteries and are 
 difficult to understand. This study may, therefore, be appropriately intro- 
 duced by a brief explanation, presented in as simple a manner as possible, 
 of such of the phenomena and laws of electricity as apply to the subject 
 of steel manufacture by this method. To be of the greatest help to those 
 unfamiliar with electricity, it is necessary to begin with the fundamentals 
 and build up such a structure as the limits of space and time will permit. 
 
 Force, Work, Energy and Potential: In the industrial world the 
 fundamental or prime factor is work. It is defined as the operation of 
 overcoming resistance through space, or as the production of effects upon 
 bodies. That which is the immediate cause of these effects is force, which 
 is more accurately defined as that which causes, or tends to cause, a change, 
 in the motion of a body, in either velocity or direction. Thus, a column 
 may exert a powerful force in supporting part of a building, because it 
 tends to change the direction of motion the overburden would have if free 
 to fall; but it does no work, because no effect is produced. Force is measured 
 by the product of the mass of the body it acts upon and the acceleration 
 (rate of change of motion) it produces. In the centimeter-gram-second 
 (C. G. S.) system the absolute unit of measurement is the dyne, which 
 is the force required to produce an acceleration of one centimeter per. sec. 
 per. sec. in a mass of one gram. On the foot-pound-second (F. P. S.) system, 
 the unit is the pound, which is the force exerted by gravity on a definite 
 
244 ELECTRIC PROCESS 
 
 mass of matter. A similar unit on the centimeter-gram-second system is 
 the kilogram, which is the force exerted by gravity on a mass of one kilogram. 
 That which imparts to a body the ability to do work is energy. Both are, 
 therefore, measured by the same unit. In the foot-pound-second 
 system this unit is the foot-pound, or the work done by a force of one 
 pound acting through a distance of one foot. In the centimeter-gram-second 
 system a large unit is called the kilogram-meter, while a small unit, one dyne 
 acting throu gh a distance of one centimeter, is called the erg. The j ou le,equal 
 to 10,000,000 ergs, is a more practical unit. Thus, if a weight of 10 Ibs. is 
 lifted against gravity to a distance of 5 ft., 50 foot-pounds of work has been 
 done on that body, and 50 foot-pounds of energy was expended, and the 
 same amount of energy is stored up in the body raised in the form of 
 potential energy, which imparts to this body the ability to do work. In 
 practice the body lifted would be said to have its potential raised, or a 
 difference in potential has been effected between this position of the body 
 and (the same body in) its former position. 
 
 Power: It will be noticed that work is independent of time. The 
 time rate of doing work is called power. In the foot-pound-second system 
 the unit is the horse power, (h. p.), which equals 33,000 foot-pounds in one 
 minute or 550 foot-pounds in one second. It is based on experiments in which 
 it was found that the work an average draft horse can perform continually 
 without over-exertion is equivalent to lifting a weight of 150 pounds vertically 
 while travelling at the rate of 2.5 miles per hour. In the centimeter-gram- 
 second system the unit is the watt, which is that power that will do one 
 joule of work in one second. The large unit equals 1000 watts and is 
 called kilowatt. This unit is employed in electrical work. 1 kilowatt= 
 1.34 h. p., or 1 h. p., =746 watts=.746 kilowatts. Since energy is conserved, 
 power can be supplied only by creating a difference in potential. 
 
 Transmission of Energy: In the mechanical world it is often desir- 
 able, for economical reasons, to create this potential difference at some 
 central point, known as the power station, from which the power may be 
 distributed by proper means to various other points and applied as required. 
 For the transmission of energy there are four agencies, namely, gases, such 
 as steam; fluids, such as water; electricity; and the Ether. In certain 
 respects the characteristics exhibited by any one of these agencies in use 
 is similar to each of the others. In the first case the difference in potential 
 is maintained by making use of the potential, or chemical energy, of fuels 
 to generate steam, which may be conducted through pipes to impart motion 
 to engines and do work upon matter, or to give up its energy as heat. 
 Similarly, water may be made to transmit energy by causing it to flow 
 through pipes from high levels to lower ones. Somewhat analogous to the 
 flow of the water, is the passage of the electric current along a wire. 
 In each case means must be taken to maintain the flow by keeping up a 
 difference in potential. In the case of water, a pump could be inserted in 
 a circuit for returning the water to the higher level as rapidly as it flows 
 
ELECTRICAL UNITS 
 
 245 
 
 downward. In practice a pump would be impracticable, but the sun 
 accomplishes the same thing by vaporizing water so that it rises into the 
 atmosphere to fall again as rain, and thus complete the circuit. In electric 
 circuits this difference in potential is maintained by means of the electric 
 battery, the static machine, or the dynamo, the last of which will be briefly 
 described later. The close similarity between these two cases will be 
 readily observed by a study of the following table of analogues, in which 
 the water is assumed to be flowing through a horizontal pipe at the point 
 of examination: 
 
 Table 35. Hydraulic Electric Analogues. 
 
 Functions of 
 
 the Currents 
 
 Hydraulic 
 
 Electromagnetic 
 
 Hydraulic 
 
 Electric 
 
 Units 
 
 Units 
 
 Pressure 
 Quantity 
 
 E. M. F. or 
 
 Voltage 
 
 Pressure per sq. 
 in. or Head in 
 feet 
 Pound 
 
 Volt. 
 Coulomb. 
 
 Rate of flow 
 Friction. . 
 
 Amperage 
 Resistance to con- 
 
 Pounds per second 
 Loss in Head, 
 
 Coulombs per sec. 
 or Amperes. 
 
 Work 
 
 duction 
 Electrical Energy 
 
 No unit 
 Foot-pounds. . 
 
 Ohm. 
 Joule. 
 
 Rate of Work 
 
 Wattage . 
 
 Horse-power or 
 
 Watt or 
 
 
 
 watt 
 
 Volt Ampere . 
 
 One important point of difference between the transmission of water 
 and electric current is evident; namely, that whereas water passes through 
 a hollow tube, electrifications pass along solid bodies, usually wires. It 
 is also common knowledge that electrifications will pass along some sub- 
 stances very easily and only with difficulty, or not at all, along others. 
 No substance is so good a conductor as not to offer some resistance to the 
 transfer. Substances that offer little resistance are called conductors; 
 those in which the resistance is great, non-conductors or insulators. In 
 the following table, the substances named are arranged in the order of 
 their conductivities: 
 
 Table 36. Relative Conductivity of Various Substances. 
 
 Conductors 
 
 Metals 
 
 Graphite 
 
 Acids 
 
 Salt water 
 Linen 
 Cotton 
 Dry wood 
 Paper 
 
 Silk 
 
 India rubber 
 
 Porcelain 
 
 Air 
 
 Glass 
 
 Sealing-wax 
 Rubber 
 Vulcanite 
 Insulators 
 
246 ELECTRIC PROCESS 
 
 Electromotive Force (E. M. F.): While a definition of the various 
 electrical units at this time would be out of place, occasion should be made 
 to explain electrical pressures. Just as hydraulic pressure might be called 
 water-moving force, so pressure produced electrically is called electromo- 
 tive force (e. m. f.). As indicated above, the unit of measurement for 
 electromotive force is the volt, which is also the unit for measuring difference 
 in potential. In practice electromotive force and difference in potential 
 are different things. Electromotive force refers to the total electrical 
 pressure existing in a circuit, whereas difference in potential is merely the 
 difference in electrical pressure between two points on the circuit. 
 
 SECTION II. 
 
 THE DEVELOPMENT OF ELECTROMOTIVE FORCES OR 
 
 "GENERATION OF CURRENT." 1 
 
 Methods for Setting up Electric Currents: As already indicated the 
 difference in potential between two points, which is necessary to produce an 
 electric current, may be created by different methods, of which the most 
 common and useful are the following: 1. By friction, as in the electro- 
 phorus, or electrostatic machine. 2. By chemical action, such as that 
 which takes place in the electric battery, or voltaic cell. 3. By electro- 
 magnetic induction, as in the dynamo. In all these cases, energy must be 
 expended to produce the difference in potential, and the points of different 
 potential must be connected by a conductor. Electrostatic machines, while 
 they produce a high electromotive force, generate only a small quantity of 
 electricity in a given time. Current produced in this way has, therefore, 
 a very limited use. In the voltaic cell these conditions of current are 
 reversed, the amperage being high and the voltage low. The dynamo can 
 be made to give current high in both voltage and amperage. It is, there- 
 fore, the most useful of all electrical machines, and is the source from 
 which the current required by the electric steel furnace is obtained. This 
 being the case, a discussion of the principles involved in the working of this 
 machine should be both interesting and instructive. To an observer not 
 familiar with the subjects of magnetism and electricity, the dynamo appears 
 as a machine for changing motion into electrical energy, and, in a way, 
 this idea is correct. But the generation of the current is not due to motion 
 alone. Other phenomena, known as magnetism and induction, are involved, 
 and to understand the generation of the current these must be studied first. 
 
 Magnetism : The attractive force of magnets upon iron is well known. 
 Upon investigation, it is found that every magnet possesses two poles, 
 designated as North and South, N. and S., or + and ,from which lines 
 of force issue, and that these lines of force protrude into the space surround- 
 ing the magnet and extend from pole to pole. In studying the action of one 
 magnet upon another, the following laws are observed: 1. Like magnetic 
 
 *For a fuller study of the generation, transmission and utilization of the elec- 
 tric current see the standard text books on Physics, also Practical Electricity by 
 Terrel Croft and Applied Electricity for Practical Men by Arthur J. Rowland. 
 Published by McGraw-Hill Book Company, Inc., New York. 
 
MAGNETISM 
 
 247 
 
 poles repel one another; while unlike poles attract one another. 2. Lines 
 of force having the same direction, i. e., issuing from like poles, repel each 
 other; those of opposite direction attract. All these facts are illustrated 
 in the accompanying figures, which are diagrams made from photographs 
 of actual conditions. 
 
 Fro. 28. Diagrams of Sections through the space surrounding magnets showing lines 
 of force between like (A) and unlike (B) poles. Anyone may study these lines of 
 force for himself by securing two bar magnets, some iron filings, and a piece of 
 paper. The paper is laid upon the magnets and the filings are sprinkled upon it. 
 The filings are thus magnetized and arrange themselves in lines or curves corre- 
 sponding to the lines of force. 
 
 Magnets and Magnetic Substances: Magnets may be natural or 
 artificial. Most magnets are artificial and may consist of straight bars or 
 rods as shown in the figure, or be curved, as is the case with horse shoe 
 magnets. These magnets may be either permanent or temporary. Per- 
 manent magnets are made of hard steel and retain their magnetism 
 indefinitely, whereas temporary magnets are made of soft steel and are 
 magnets only so long as they are under the influence of a magnetizing force. 
 The space surrounding the magnet through which lines of force pass is 
 called the magnetic field, and the number of lines is referred to as the 
 magnetic flux. Substances that are attracted by a magnet or are mag- 
 netized when placed in a magnetic field are called magnetic substances. 
 While it can be shown that most substances are affected by magnetism, 
 iron, its alloys, and one oxide, Fe3C>4, are the only substances available for 
 use. These lines of force cannot be insulated, for they pass through all 
 
248 
 
 ELECTRIC PROCESS 
 
 substances, but by the use of a permeable substance, like very soft steel 
 or iron, they may be deflected from their course and concentrated in the 
 mass of iron as shown in A of Fig. 29. 
 
 FIG. 29. Magnetic Permeability and Induction. 
 
 A. The soft iron washer encloses a space through which there are no lines of force. 
 B. Cut in two, the washer becomes a magnet by induction. 
 
 Magnetic Fields and Electric Currents: These magnetic lines of 
 force are closely associated with the electric current, for it is easily shown 
 that every current bearing conductor is surrounded by a magnetic field, 
 the lines of force in which form circles concentric with the conductor. 
 These lines of force have a definite direction as in the case of ordinary 
 magnets, and this direction bears a fixed relation to the direction of the 
 current, as shown in A of Fig. 30. This fact is made use of in producing 
 the electromagnet, by coiling the conductor, properly insulated, about a 
 core of soft iron, as shown in C of Fig. 30. Such a coil is known as a helix 
 or solenoid, and has poles similar to magnets. The various relations 
 between flow of current, lines of force, and poles of the helix are shown 
 in the figure. The wide application of these facts cannot be discussed here, 
 except in so far as they have to do with the development of an electromotive 
 force, or, as this is more commonly referred to, the generation of the electric 
 current. In connection with the generation of currents, this question might 
 arise: If a current produces a magnetic field about a conductor, will the 
 production of a magnetic field about a conductor result in a current? This 
 question is now to be answered by a brief study of electromagnetic induction. 
 
ELECTROMAGNETISM 
 
 Electromagnetic Induction : The easiest way of bringing about such 
 a condition as noted in the preceding paragraph is described in the following 
 experiment: If a coil of many turns of fine copper wire is connected to a 
 delicate galvanometer and one end of a bar magnet is thrust into it, the 
 galvanometer needle will be deflected, showing that a current is set up in 
 
 Battery 
 
 FIG. 30. Magnetic Fields About Current Bearing Conductors. 
 
 A. Lines of force about a wire carrying a direct current. 
 
 B. Lines of force in the current bearing helix. 
 
 C. The Electro-magnet. The lines of force pass into the soft steel 
 
 bar, which becomes a magnet by induction. 
 
 the coil As long as the magnet remains stationary, no current will pass. 
 Upon suddenly withdrawing the magnet, however, a current will again pass, 
 but the direction of this second current is opposite to that of the first. If 
 these operations be repeated with the other pole of the magnet, similar 
 currents will be induced, but in directions opposite to those obtained when 
 the first pole is used. Instead of moving the magnet, the coil can be moved 
 with a like result, and in place of the magnet a solenoid can be substituted. 
 In the last case it may not be necessary to move either the solenoid or the 
 
250 
 
 ELECTRIC PROCESS 
 
 coil, as current can be set up in the coil by breaking the circuit or other- 
 wise interrupting the current in the solenoid. From these facts it would seem 
 that the sole cause of the current is the change in magnetic flux. Further 
 study of these phenomena reveals the fact that the current induced is affect- 
 ed by the speed with which the magnet may be inserted and withdrawn, and 
 the number of wires in the coil. In the case of the solenoid a third factor 
 is introduced, as the current carried by the solenoid itself affects the 
 induced current. Furthermore, no current is set up in the coil unless the 
 motion is such that its wires cut the lines of force produced by the exciting 
 elements: no current is generated if the conductor moves along the lines 
 of force. All these facts are summed up in the following laws: 
 
 Laws of Electromagnetic Induction: 1. Any change in the number 
 of lines of force passing through a closed conducting circuit induces a current 
 in that circuit. (2) The direction of the induced current is always such 
 that its magnetic field opposes the motion which produces it (Lenz's law). 
 (3) The electromotive force of the induced current is directly proportional 
 to the rate at which the number of lines of force are increased or decreased, 
 or, the rate at which the lines of force are cut. 
 
 The Dynamo: Coming now to the practical application of these 
 principles, it is found that, of the many electrical appliances depending 
 upon them, the dynamo and the transformer are of chief interest in a study 
 of the electric furnace. The dynamo is designed to convert mechanical 
 energy into electrical energy. The steam engine, for example, does work 
 on a dynamo, and the dynamo produces an electric current. This current 
 contains all the energy that was received from the engine except a small 
 percentage which was lost in heat and friction. There are three essential 
 parts of a dynamo: 1st., the field magnets; 2d., the armature; 3d., the 
 collecting brushes. The magnets are used to create a strong magnetic 
 field between the two poles, that is, a field in which there are many lines 
 of force. 
 
 Revolving Loop 
 of Copper Wire 
 
 Magnet"^ 
 
 Brushes 
 
 FIG. 31. Diagram Illustrating Essential Parts and Principle of the Dynamo, 
 a. Direction in which loop revolves. b and b'. Direction of current through loop, 
 c. Directi on of lines of force, dandd'. Direction of current in external part of the circuit. 
 
KINDS OF CURRENT 251 
 
 Hence, in all large dynamos electromagnets are used. The armature 
 consists of coils of wire wrapped around a soft iron core; it is mounted so 
 as to rotate in the magnetic field and cut lines of force. The current is 
 thus generated in the manner described under induction. The essential 
 parts of a dynamo are shown in the diagram of Fig. 31, which illustrates 
 the arrangement for a drum wound armature. The north-seeking pole of 
 the field magnet is marked N; the south-seeking pole, S. The armature 
 is here a single loop of wire. The ends of the loop are connected to rings 
 which rest on collecting brushes. When the loop is rotated, it cuts the 
 lines of force, causing a current to flow out to the brushes and through the 
 external part of the circuit. The direction of this flow bears a fixed 
 relation to the lines of force as explained under induction and as shown in 
 the figure. In practice the following rule is employed to determine the 
 direction of the current in the armature. Extend the thumb, first finger, 
 and middle finger of the right hand in such a manner that each will be at 
 right angles to the other two. Place the hand in such a position that the 
 first finger will point in the direction of the lines of force (N. to S.) and 
 the thumb in the direction in which the conductor moves. The middle 
 finger will then point in the direction in which the current flows. 
 
 SECTION III. 
 
 KINDS OF CURRENT. 
 
 Alternating Current: The current produced in the dynamo 
 armature is not a constant one and travelling in one direction, such as is 
 obtained by means of batteries, but is an alternating one, as a further study 
 of Fig. 31 will show. Let the coil be rotated as indicated by the curved 
 arrow above it. While the side at b is moving downward in front of the 
 pole N., the side at b' will be moving upward in front of the pole S. An 
 application of the law for direction shows that the currents thus induced 
 by the two branches of the coil cutting lines of force will flow in opposite 
 directions in relation to each other, but in the same direction from end 
 to end around the coil. This current continues as long as b and b' are 
 moving in the same direction across the lines of force that is, during 
 one-half of one complete rotation. Then, as the rotation continues, the 
 sides of the coil cut the lines in an opposite direction, and the current is 
 completely reversed. Each time the coil is turned half way around, the 
 direction of the current is changed. Each end of the ceil is attached to 
 a ring. The rings are attached to the axis of the coil and rotate with it. 
 Brushes slide upon the rings and conduct the current out upon the line of 
 the external circuit. The line current is thus changed in direction twice 
 during each complete rotation of the coil or armature. Each change in 
 direction is called an alternation. Two alternations constitute a cycle 
 that is, a circle or series of changes which will be repeated in the next cycle. 
 The time required for one cycle is the period. The number of cycles in 
 
252 
 
 ELECTRIC PROCESS 
 
 one second is the frequency. The frequency is one-half the number of 
 alternations in one second. In modern forms of alternators the frequency 
 is seldom more than 60 nor less than 25 cycles per second. To produce 
 60 cycles in a second with a 
 machine like that shown in Fig. 31, 
 the armature would have to make 
 3600 rotations in one minute. To 
 avoid such high speed, a number 
 of pole pieces are arranged in a 
 circle around the armature, making 
 what is called a multipolar ma- 
 chine. These poles are wound so 
 that north and south-seeking poles 
 alternate in position. A cycle is 
 then produced in the circuit by the 
 passage of any two adjacent poles. 
 If there are ten poles, there are 
 five cycles during one rotation of 
 the armature. A current of 25 
 cycles per second would mean 50 
 alternations per second or 3000 per 
 minute; on a two pole machine it 
 would require 1500 r. p. m.; on a 
 10 pole, 300 r. p. m. The latter 
 is about as Iowa speed as a machine 
 can be operated upon economically. 
 These are some of the reasons 
 why it is not practical to operate 
 alternators on cycles over 60 nor 
 under 25. 
 
 Graphic Representation of 
 Alternating Current: The surges 
 of the current are approximately 
 harmonic, hence the electromotive 
 force may be represented by the .- 
 sine curve, as shown in Fig. 32. i 
 This figure shows one complete ! 
 wave or cycle, as taking place in ^ ; 
 second of time, and going from zero 1 
 through a positive maximum and " 
 back to zero, where it continues to 
 a negative maximum to return 
 
 again to zero, thus completing a cycle. The different stages of vibration 
 represented in the graph are spoken of as phases. Thus b and b' are in 
 the same phase, but b and h' are in opposite phases, etc. 
 
KINDS OF CURRENTS 
 
 253 
 
 Direct Currents : While all dynamos produce alternating current in the 
 armature, this current may be changed to a direct one by means of a com- 
 mutator properly connected to the armature. How this is done can be 
 understood by a study of the accompanying figure. 
 
 Legend: 
 
 a. a. Armature Loops 
 b,b,b,b. Brushes 
 c,c. Commutator 
 Direction of Rotation 
 " " Current 
 " " Magnetic 
 Lines of Force 
 
 FIG. 33. Diagram Illustrating how Current is Rectified by means of a Commutator 
 For reasons to be given later, direct currents are not used in furnaces for refining 
 iron and steel. 
 
 Polyphase Currents: In order to increase the efficiency and output 
 of alternators, recourse is had to polyphase current. These currents 
 are the result of attempts to put to economic use the interpolar space, or 
 surface of armature core, which is only partly filled in by conductors in 
 single phase machines. In any given machine of this kind, which 'involves 
 certain given mechanical and magnetic losses, approximately only half of this 
 space is utilized. This waste of space and consequent inefficiency can be 
 removed by utilizing the space devoted to the core of the coil for the winding 
 of other coils, and thus form a second armature of an equal number of coils 
 overlapping the former and utilizing the same magnetic field. In this 
 second armature the phase of the current would be a quarter period, 90 
 degrees, ahead or behind that of the first, and so would require four wires to 
 carry the currents to and from the generator. The output of the machine, 
 however, has been doubled. Further attempts to increase output have 
 resulted in electrical engineers going even further and constructing triple 
 armatures, in which the phase of the currents generated differ by equal 
 intervals of % of a period, or 60 degrees. But this scheme then leads to 
 considerable elaboration, if the circuits be totally independent with six 
 conductors, and very little advantage could be shown to exist over the 
 two-phase with four lines. However, in cases where it is possible to arrange 
 that the demand in the different circuits be approximately equal and evenly 
 distributed, the three phase system can be worked out to great advantage 
 by using only three conductors. By referring to the accompanying diagrams 
 (Figs. 34 and 35) it will be seen that, at any moment whatever, the sum of 
 the electromotive forces in the three circuits is zero. In other words, the 
 electromotive force on one line is always equal to that on the other two, 
 
254 
 
 ELECTRIC PROCESS 
 
 and opposite in kind. Thus, at any instant one of the three wires is to be 
 looked upon as the return wire for the other two. 
 
 FIG. 34. Diagrams Illustrating The Methods of Generating the Three Kinds of 
 Alternating Current. 
 
 The Two Schemes of Wiring for Three Phase Current: The coils 
 of the three phase current alternator, as well as those of the apparatus 
 that is to consume the power, may be connected in two different ways 
 with each other as shown in the following diagrams. In these diagrams 
 the three phase generator is considered as a compound machine made up 
 of three separate machines. In such a machine, the separate units may 
 
KINDS OF CURRENTS 
 
 255 
 
 be wired independently as shown in Fig. 35A. Instead of three return wires, 
 one common return wire may be employed, as shown by the dotted line R 
 in the figure. But it is found that, with balanced loads on each system, R is 
 a neutral line, that is, no current flows through it; consequently, it is omitted 
 as in the figure. The points m and n are therefore called neutral points. 
 The form of this diagram suggests the name for the scheme, which is known 
 as the star, or Y, connection. In the other schemes the coils of the dynamo 
 are connected in a series, and the power is led off through mains connected 
 to points in the connecting lines. Fig. 35B shows diagrams intended to 
 illustrate the scheme for this connection. On account of the similarity 
 to the Greek letter A, this scheme is known as the delta connection. 
 It is also called the V and mesh connections. It will be observed that the 
 windings of the delta connection form a closed circuit, and the generator there- 
 fore appears to be short circuited. However, such is not the case, because, 
 owing to the fact that the currents generated by the three windings differ 
 in phase by 120, as shown in Fig. 34C, the resultant of the three voltages 
 is at every instant zero, hence no current can flow by way of the short circuit. 
 With this plan of connections the load may be balanced or unbalanced as each 
 coil or set of windings supplies its own load. 
 
 As generated 
 A. Diagram of Star (or Y) Connections. 
 
 L -" ' 
 
 As generated 
 
 58 Volts 
 
 As consumed 
 
 As consumed 
 
 B. Diagram of Delta (Mesh or V) Connections. 
 
 Fio. 35 Diagrams Illustrating the Two Methods of Winding and Wiring for 
 Three- Phase Currents. 
 
 These connections are of importance in understanding how the power 
 of 3-phase currents is transmitted. They also affect the voltage employed, 
 
256 ELECTRIC PROCESS 
 
 the relations being as shown in the preceding diagrams. The Duquesne 
 furnace, for example, operates on 104 volts with star connections and 180 
 volts with the delta. This care in explaining three phase current has been 
 taken, because it is the most common kind of current and is used in the 
 Heroult furnace, the details of which will be explained later. 
 
 SECTION IV. 
 
 TRANSMISSION OF THE CURRENT. 
 
 Ohm's Law: With this brief explanation of the generation of the 
 current, it will be well to turn next to the problem connected with its trans- 
 mission. Here, also, it is necessary to begin with the simplest essentials. 
 The most fundamental idea in this connection can be briefly stated in the 
 form of a law, known as Ohm's law. This law states that the strength 
 of current passing along a conductor varies directly as the electromotive 
 force, or drop in potential, and inversely as the resistance. This law is 
 generally put in the form of a formula, thus: 
 
 E Electromotive Force 
 
 or Current= ; 
 
 R Resistance 
 
 In place of these letters the units which are used in measuring the quantities 
 they represent may be substituted, thus: 
 
 V Volts 
 
 A= . or Amperes=; 
 
 O Ohm's 
 
 By the use of this formula any one of the three quantities may be found, 
 provided the other two are given. It may be applied to an entire circuit 
 or to a part of a circuit. 
 
 Resistance of Conductors : Referring now to resistance of conductors, 
 it will be recalled that, as previously stated, all conductors offer resistance 
 to the passage of the current. This resistance can be calculated by applying 
 the following law, which has been developed by many experiments. It is 
 stated as follows: The resistance to the flow of an electric current along 
 a given conductor at a given temperature varies directly as the length 
 and inversely as the area of its section, and is different for different 
 
 substances. As a formula, the law is expressed thus: R= g- where R= 
 
 resistance, l=length, d 2 =square of the diameter, which is directly pro- 
 portioned to the area, or the area itself, and K=specific resistance of the 
 substance. In the foot-pound-second system, l=length in feet, and d= 
 diameter in mils, or thousandths of an inch. A circular mil is the area 
 of a circle one mil in diameter. So if the diameter of a wire is expressed 
 
TRANSMISSION OF CURRENT 257 
 
 in mils, d 2 =the area in circular mils. The circular mil is not to be con- 
 fused with the square mil which is the area of a square having sides 1/1000 
 inch long. The area of a circle in square mils, the diameter of which is 
 
 expressed in mils = . In this connection the following formulas may be 
 
 found of assistance: 
 
 1. sq. mils=cir. mils X 0.7854 2. cir. rnils= 
 
 .0000007854 
 
 In wire, a length of one foot and a diameter of one mil constitutes a mil- 
 foot. The specific resistance of any substance is the resistance of one mil- 
 foot. In the following table are the specific resistances of a few substances, 
 determined for this system, at 60 F. In the metric system, l=meters, 
 d 2 =sq. millemeters, and, of course, K has a numerical value different from 
 that by the foot-pound-second system, as shown in the table. 
 
 Table 37. Specific Resistance of Various Substances. 
 
 60 F. C. 
 
 Foot-Pound-Second Centimeter-Gram-Second 
 
 (F. P. S.) System (C. G. S.) System 
 
 Silver 9.53 .017 
 
 Copper 10.35 .018 
 
 Aluminum .03 to .05 
 
 Gold 13.34 
 
 Zinc 36.46 .06 
 
 Iron .10 to .12 
 
 Platinum 58.00 .12 to .16 
 
 Steel 63.00 .10 to .25 
 
 Molten Steel 1.66 to 2.2 at 1700 C. 
 
 Nickel .15 
 
 Lead 127.2 .22 
 
 German Silver 135.0 .15 to .36 
 
 Mercury 616.16 .94073 
 
 The Ohm, the unit of resistance, may now be defined. It has been 
 established by law to be the resistance offered by a mercury column 
 1.063 m. long of 1. sq. mm. cross section, and at the temperature of C. 
 Originally, the length of the mercury column was one meter. 
 
 Effect of Temperature on Conductors : In connection with the effect 
 of temperature upon conductors, it should be noted that there are two 
 classes of substances, known as conductors of the first class and conductors 
 of the second class. In conductors of the first class, which includes all 
 the metals, the resistance increases with a rise in temperature, hence their 
 conductivities decrease. This is shown in the case of steel, which at 15 C. 
 
258 
 
 ELECTRIC PROCESS 
 
 has a specific resistance of only .10 to .12, whereas at 1700 the specific 
 resistance of the molten metal is about 1.66. In the case of carbon and 
 conductors of the second class, these relations are reversed, so that they 
 become better and better conductors as their temperatures rise. As all 
 the refractory materials that go into the construction of furnaces belong 
 to this class, this matter must be carefully considered in building electric 
 furnaces. 
 
 Resistance in Series and Parallel. In the preceding paragraph, only 
 single substances were considered in the circuit. In the actual distribution 
 and use of power, it is generally necessary to divide circuits, in which 
 divisions different materials or machines will also make up a part of the 
 circuit. This division and connections can be made in two ways, namely, 
 series and parallel, as shown in the following diagram illustrating two 
 methods of light wiring. 
 
 r' r" r'" 
 
 L 
 
 W 1 ' ' ^J ^ 
 
 HI 
 
 A. Series r , 
 
 U : 9 r 'f^ 
 
 n - L 
 
 i ^^^ 
 
 T 
 
 i 
 
 
 B. Parallel 
 FIG. 36. Diagram Illustrating Different Methods of Wiring. 
 
 In Fig. 36 A, the resistance of the line is r and of each of the lamps it 
 is r', then the whole resistance, R, is the sum of these four, or 
 
 R=r+r'+r"+r'" or R=r+3r'. 
 
 Resistance of conductors connected in parallel is not found so simply. 
 It is deducted from the law of conductivities, which states that the conductiv- 
 ity of a combination of conductors is equal to the sum of the conductivities 
 of the conductors singly. Now, it is self evident that the conductivity is 
 
 represented by -j- The resistance in the parallel connections above would 
 R. 
 
 then be found by solving the following: 
 11111 
 
 This matter is of importance in dealing with electric furnaces, as shown 
 
TRANSMISSION OF THE CURRENT 259 
 
 in the following schematic diagram of the wiring of Heroult and Girod 
 furnaces, which marks the chief difference between these two furnaces. 
 
 FIG. 37. Wiring Diagrams for Heroult and Girod Electric Steel Furnaces. 
 
 These diagrams show the Heroult furnace, on the left, to be connected 
 in series and the Girod in parallel. 
 
 Currents Through Divided Circuits : If the different parts of parallel 
 connected conductors have equal resistance, then equal currents will flow 
 through these parts, as is evident from Ohm's law, thus, 
 
 i=5 r=5 
 
 R R' 
 
 As the current is always delivered at a constant voltage, E is the same 
 in both cases; hence, if R=R', 1=1'. If the resistances are not the same, 
 then I would not equal I'. The current flowing in each conductor, however, 
 can be found from the following law: Currents which flow parallel to each 
 other vary inversely as the resistance of the parallel connected conductors. 
 
 Self Induction, Impedance, Power Factor: In dealing with alter- 
 nating currents, Ohm's law does not hold, for the alternations cause self- 
 inductance in the conductor, that is, they generate currents opposite in 
 direction, or kind, to that of the main current; and this inductance also 
 offers resistance to the current in addition to that offered by the conductor. 
 The sum of these two forces opposing the passage of the current is called 
 impedance. If impedance be substituted for resistance, then Ohm's law 
 holds for alternating currents, also. Again, any self-induction in the circuit 
 causes a difference in the phase between the electromotive force (voltage) 
 and the current (amperage), so that the latter either forges ahead of or 
 lags behind the former, and the two do not reach their maxima and 
 minima together. Consequently, their product at these points, which 
 represents the energy available for consumption, is less than the total 
 energy supplied. The ratio, expressed in per cent., between the useful 
 voltage, or that which is required to overcome the resistance of the con- 
 ductors and to produce the heat or do any other work desired, and the 
 actual voltage required is called the power factor. 
 
 Heat Developed in Conductors: Since all conductors offer resistance 
 to the flow of the current, work is done in overcoming this resistance. In 
 doing this work part of the electrical energy is converted into heat energy, 
 just as work is done and heat developed in overcoming friction. This heat 
 
260 ELECTRIC PROCESS 
 
 can be calculated by means of Joule's law, who found from many experiments 
 that the heat developed by a current flowing through a conductor is directly 
 proportional to the time, to the resistance, and to the square of the current. 
 Mathematically stated, this law is H=K I 2 RT, where H=heat, I=current, 
 R=resistance, T=time, and K=a constant, which Joule found to be .24. 
 
 Tjl 
 
 Since from Ohm's law, I=-R or E=IR, H=.24 IET colories, when I is 
 
 measured in ampres, E in volts, and T in seconds. The heat thus developed 
 is of much importance in the transmission of current, for if the current be 
 sufficiently large this heat may raise the temperature enough to burn off 
 the insulation, or even to melt the wire. This heating can be overcome 
 in two ways. In the first method, the diameter of the conductors could 
 be increased, which would decrease the resistance and increase the carrying 
 capacity. That this method has its limits is evident, due to the immense 
 weight of wire required in some cases where large current (amperage) is 
 required, as is the case with electric furnaces. The second method is 
 applicable to alternating current only and illustrates both the adaptability 
 of the electric current in general and one advantage of alternating current 
 in particular. From Joule's law it is evident that a current of large voltage 
 may be carried on a given wire provided the density, i. e., amperage per 
 circular mil or square millimeter be kept low. Since power=amperes X 
 volts, or for alternating current Power=amperes X volts X power factor, 
 this can be done without reduction in power. But since the furnace requires 
 a current of low voltage and high amperage, such a current could not be 
 used unless means be taken to change this high-voltage-low-amperage 
 current into one of low voltage and high amperage. How this is done can 
 be learned from the following description of the transformer. 
 
 The Stationary Transformer: In this instrument the desired trans- 
 formation is effected by electro-magnetic induction already discussed. Of 
 course, then, only those currents which are started and stopped or increased 
 and decreased in rapid succession, or those in which the direction of the 
 current is changed many times in a second, can be transformed. Such 
 current is furnished by the alternator. In structure the transformer con- 
 sists of two coils of wire side by side with a core composed of many sheets 
 of soft iron, or a special silicon steel, packed together. The coils must, not 
 have any metallic connection with any part of the instrument. The first 
 coil, that through which the main current flows, is called the primary. 
 The second coil, or that in which the current is induced, is called the 
 secondary. 
 
 Kinds of Stationary Transformers : Transformers are of two kinds 
 step-up and step-down. The step-up transformer increases the voltage and 
 decreases the amperage. The step-down produces the opposite effect. 
 The change depends on the relative number of turns of wire in the primary 
 and secondary coils. If, for example, there are 100 turns in the primary 
 and 1000 turns in the secondary, the voltage will be increased 10 times 
 
UTILIZATION OF THE CURRENT 261 
 
 and the amperage decreased 10 times. This is then a step-up transformer. 
 A step-down transformer would reverse these conditions, throughout. If 
 there are the same number of turns in both coils, the current will not be 
 changed except as it may be affected by the transformer efficiency, which 
 should be about 98%. 
 
 In regard to the power of the transformed current it will be seen that, 
 since, whenever the transformer increases or decreases the voltage, it 
 decreases or increases the amperage, the number of watts will be a constant 
 quantity. Suppose there is a current of 100 volts and 10 amperes flowing 
 through the primary. The power is then 1000 watts. If the transformer 
 raises the pressure to 500 volts, the strength of the current will fall to 2 
 amperes, but the power of the current is still 1000 watts. A good trans- 
 former gives out nearly all the energy that is put into it. A small 
 percentage, varying from 2% to 5% of the voltage, is converted into heat. 
 Usually this heat is prevented from collecting by immersing the coils 
 in cylindrical tanks of oil so constructed as to form a circulating system 
 through pipes extending externally from the top to the bottom of the 
 cylinder. These pipes act like a hot water radiator and serve to keep the 
 whole bath and its contents cool. The oil also serves as an insulator. 
 For steel furnaces three phase 25 cycle current is stepped down from 6600 
 volts to 104 on the star connection or 180 on the delta, the latter of which 
 is seldom used on molten charges. 
 
 SECTION V. 
 THE UTILIZATION OF THE CURRENT IN ELECTRIC FURNACES. 1 
 
 Effects Produced by Electric Current: The heating and magnetic 
 effects of electric currents have already been touched upon in connection 
 with generators, conductors and transformers. In order to understand 
 how the electric current is utilized in electric furnaces it is necessary to 
 study these and other effects from a slightly different standpoint; namely, 
 their effect upon the metallic charge in the furnace itself. In this con- 
 nection it may be truly said that there is but one other effect produced 
 by the electric current, and this effect is that of bringing about chemical 
 action. To the question as to what chemical action is caused by the current 
 in the bath of steel, the correct answer is: There is none. As this answer 
 is not in accord with chemical effects produced by currents in other 
 metallurgical operations, an explanation may be necessary. 
 
 Chemical Action Produced by the Electric Current: As pointed 
 out in Chapter I, electrolysis is brought about when electric currents are 
 passed through liquids under certain conditions, some well known examples 
 being the dissociation of water and the electrolytic separation of aluminum. 
 In these instances, however, it will be observed that these chemical changes 
 occur only when direct current flows through the electrolytes. If alter- 
 
 !For a full discussion of electric furnaces See The Electric Furnace by Alfred 
 Stansfleld, Published by McGraw-Hill Book Co., Inc., New York, Electric Furnaces 
 in Iron and Steel Industry, by C. H. Vom Baur, published by John Wiley & Sons, 
 New York, and Electric Furnaces for Making Iron and Steel by Dorsey A. Lyon 
 and Robert M. Keeney Bureau of Mines Bulletin 67. 
 
262 ELECTRIC PROCESS 
 
 nating current is used, then the direction of the current is constantly 
 changing, and no action, such as noted above, can take place. Furthermore, 
 such action would be harmful in carrying out the electro-thermal process 
 for steel. While it might be possible with direct current to purify the 
 metal by removing the impurities as sulphides, silicides, and phosphides, 
 there would be no way of controlling the process so as to prevent the 
 reduction of lime, alumina and other oxides, the elements from which 
 would then find their way into the metallic bath. This reduction would 
 result in a more impure product than the raw material. Designers of 
 electric furnaces for the iron industry will, then, use alternating current 
 exclusively and strive in every way to prevent any electrolytic action 
 that might result in electrolysis. 
 
 Electrical Units of Measurements: The subject of electrolysis offers 
 an opportunity to define another of the primary units used in electrical 
 measurements. The ohm has already been defined in studying resistance. 
 It now remains to explain how the value of the ampere is fixed. If a current 
 be made, by means of suitable electrodes, to pass through a solution of 
 silver nitrate, metallic silver will be deposited upon the cathode, or positive 
 pole, and the amount of silver thus deposited in a given time will be pro- 
 portional to the strength of the current. This fact has been made the 
 basis for fixing the value of the ampere. The legal definition reads as 
 follows: The ampere is that current which when passed through a 15% 
 neutral solution of silver nitrate will deposit .001118 grams of silver in one 
 second. The volt is then legally defined as the e. m. f . which will cause 
 a current of one ampere to flow through a resistance of one ohm. In fixing 
 the value of these units, it was arranged so that the power possessed by a 
 current of one ampere under a pressure of one volt is just equal to one watt. 
 Hence the power of the current in watts equals the product of the amperage 
 and voltage, or W=VXA. The watt=hour is the energy expended in one 
 hour when the current is one ampere and the voltage, or pressure, one volt. 
 Hence, 60 watts used for one min. or one watt used for 60 min. will give one 
 watt hour. A kilo=watt=hour=1000 watt hours. 
 
 The Magnetic Influence of the Current can be of but slight 
 importance to the metallurgist. To the designer of induction furnaces they 
 are of double importance. On account of a certain motor effect which 
 they produce in the molten metal, these forces cause what is known as the 
 pinch effect which will often break the circuit. In arc furnaces this motor 
 effect is present in the immediate vicinity of the electrodes causing a slight 
 motion of the bath. 
 
 Heating the Bath : It is to be understood, then, that the only use to 
 which the current is applied in the manufacture of steel is for the generation 
 of heat. It may be well, therefore, to consider briefly the heating pos- 
 sibilities of the current in connection with its practical application to this 
 purpose. A little thought shows that these possibilities are only three in 
 number, and may be called heating by direct resistance, heating by indirect 
 
METHODS OF HEATING 
 
 263 
 
 resistance, and heating by means of radiation from an arc or arcs. A brief 
 discussion of the three methods, which is also made to serve as a means 
 of describing the principles of the many makes of furnaces, follows: 
 
 Heating by Direct Resistance: In the method of heating by direct 
 resistance, the necessary heat to keep the bath molten is produced by making 
 use of the resistance of the iron itself. In some respects this method would 
 appear to offer some advantages. 1. Since heating is effected by the 
 passage of the current through the liquid metal, the heat would be uniformly 
 distributed throughout the mass. 2. Since the heat generated is pro- 
 portional to the square of the current, the amount of heat could be regulated 
 by changing the resistance of the furnace. 3. Very low voltage currents 
 could be used. But the disadvantages outweigh the advantages, as will 
 be realized better if they are set along side the advantages, thus: 1. As 
 the specific resistance of iron is low, the high temperature required could 
 be obtained only by the use of very high amperage, which would require 
 exceedingly heavy connections. 2. This draw-back could be overcome by 
 increasing the length of the bath and decreasing the cross sectonal area; 
 but such an arrangement has been found impracticable on account of the 
 large area the bath then covers, which results in great heat losses and 
 precludes the easy change of slags and handling of the metal. Hence, all 
 early attempts to employ direct resistance for heating proved failures. 
 The problem was, however, eventually solved by the invention of the type 
 of furnace known as the induction furnace. In these furnaces the bath is 
 made to form the closed secondary circuit of a transformer. This secondary 
 can then consist of but one coil, as shown in the following diagram 
 illustrating the principle of the induction furnace as invented by Colby 
 and Ferranti and later improved and adapted to the manufacture of steel 
 by Kjellin. 
 
 FIG. 38. Diagram Illustrating Principle of Induction Furnaces. 
 Vertical Section of Kjellin Furnace. 
 
 The furnace includes a magnetic circuit, C, formed of the usual 
 laminated sheet iron, as in a transformer core. Electric energy is supplied 
 to a primary coil, D, while the secondary circuit is formed by the ring of 
 
264 ELECTRIC PROCESS 
 
 metal under treatment, contained in the annular cavity, A, forming the 
 crucible. Now, when energy, in the form of alternating current is supplied 
 to the primary coil, D, it creates a varying magnetic flux in the laminated 
 iron core, which in turn induces a current in the closed secondary circuit 
 consisting of metal in crucible A. The current density in the secondary 
 bears a fixed ratio to the number of turns in the primary coil D, so that 
 it is possible by a suitable variation in impressed voltage to subject the 
 metal to an extremely heavy current density, the heat being thus produced 
 in accordance with Joule's law simultaneously throughout the entire mass 
 of the metallic bath. Because of the limited contact area between slag and 
 metal, this type of furnace does not readily lend itself to refining processes, 
 if the form shown in the figure be adhered to. However, this form was 
 later changed so as to give a central hearth of considerable size. Those 
 who have designed furnaces with the object of improving the Kjellin type 
 are Frick, Hiorth, Harden, Greene, and others. On account of the low 
 specific resistance of iron, it is difficult to reach high temperatures in these 
 furnaces. They are, therefore, not well adapted for desulphurizing oper- 
 ations in which sulphur is removed as sulphide. 
 
 Indirect Resistance Heating offers a second possibility. Instead of 
 depending upon the resistance of the bath alone to furnish the heat required, 
 some other conductor having a high resistance might be built into the 
 furnace in such a way that the heat generated in it would be absorbed by 
 the material to be heated. This is the principle employed in many 
 laboratory furnaces and in large furnaces for manufacturing carborundum. 
 But in applying the method to the manufacture of steel several insurmount- 
 able difficulties are presented. 1. The resister can not be composed of 
 carbon and in contact with the metal, on account of the absorption of this 
 element by iron. 2. If a suitable resister of another material could be 
 found, it could not be placed in the bath, because it would then be in 
 parallel with the metal. The only way these difficulties can be overcome 
 is by the use of crucibles to contain the molten metal. The impractic- 
 ability of this method is at once evident, and furnaces of this type 
 designed for manufacturing steel have met with no success. 
 
 Arc Heating: The use of the electric arc, which has been mentioned 
 as the third possibility for producing h,eat, is the simplest and the most 
 practical of all and has found the widest application in the steel industry. 
 Some information as to the nature of electric arcs should, therefore, be 
 interesting. In beginning, a distinction is to be made between electric 
 sparks and electric arcs. While the air is practically a non-conductor, it 
 is possible to create such a high difference in potential between two given 
 points as to cause a current to jump the gap and establish equilibrium. 
 Such conditions occur in electrical storms, and lightning is an example of 
 electric sparks. In arcs the current also passes through the air, but it 
 will be observed that a much smaller voltage is required to form an arc 
 
METHODS OF HEATING 
 
 265 
 
 than is needed to cause sparks. The most common example of the arc is 
 the arc lamp. Here the arc is made between two carbon electrodes, but 
 in order to strike an arc it is necessary to bring the electrodes into contact, 
 after which a gap may be gradually produced and the arc still maintained. 
 If the gap becomes too wide, however, the arc will break, hence means of 
 regulating the distance between the electrodes must be provided. 
 Evidently the air is not the conductor in arcs as in sparks. All these 
 phenomena are explained by assuming that some of the electrode material 
 is vaporized by the heat of the arc, and that these vapors serve as a con- 
 ductor of the current. Some idea of the intensity of the heat of this arc, 
 which gives the highest temperatures yet attained, is to be had from the 
 fact that carbon vaporizes at about 3500 C. 
 
 Methods of Applying the Arc in Arc Furnaces: Furnaces of this 
 type, then, depend almost wholly upon this high temperature of the electric 
 arc for the heat required by iron and steel baths. The following diagrams 
 will illustrate the three possible methods of applying the arc and heating 
 the bath. 
 
 ^ 
 
 A B \^ O 
 
 FIG. 39. Diagrams Illustrating the Three Ways of Employing the Electric Arc in 
 Steel Furnaces. 
 
 In each of these three cases the bath is directly beneath the arc or arcs 
 and receives its heat mainly by radiation. All three possibilities are 
 practical and have been successfully applied. So these same figures also 
 illustrate the principles of the three furnaces of the arc type. 
 
 The Stassano Furnace is represented in principle by Fig. 39A. The 
 distinguishing feature of this furnace is that the current does not pass 
 through the metal or slag, the heating being accomplished entirely by 
 radiation. At first, Stassano built his furnaces so that they could be rocked 
 or rotated in order to agitate the bath, but as this feature did not prove 
 to be of any advantage, it has now been abandoned. His furnaces are 
 now of the tilting type. In practice three phase current is generally used, 
 and the three electrodes enter the furnace at an angle through the walls. 
 This plan has the effect of placing a limit to the size of the furnace, and 
 so few of these furnaces with a capacity greater than two tons have been 
 built. From an electrical standpoint, the furnace possesses the important 
 advantage of uniform power consumption, thus avoiding harmful fluxuations 
 in current. 
 
266 ELECTRIC PROCESS 
 
 Girod Furnaces: The furnace scheme shown in Fig. 39B has been 
 developed by several inventors. It was first successfully introduced by 
 Girod for the purpose of manufacturing ferro-alloys. As shown in the 
 figure the electrodes are inserted in both the top and the bottom of the 
 furnace, thereby connecting electrodes, slag and molten metal in series. 
 Heat is thus produced in three ways, theoretically at least. By means 
 of an arc at the top, the greater portion of the heat is generated. After 
 forming the arc, the current, in its downward courses, must pass through 
 the layer of slag which, through the heat of the arc above, becomes a con- 
 ductor of the second class, after which the current is conducted by the 
 molten metal to the electrodes at the bottom. In furnaces using a single 
 phase current, these bottom electrodes are four or six in number and are 
 equally spaced about the periphery of the hearth. In the three-phase 
 furnaces there are four upper electrodes, two of which must be in parallel, 
 and sixteen bottom electrodes. In all cases the bottom electrodes are made 
 of steel and are water cooled. Other designers of this type of furnace are 
 Keller, Gronwall, Nathusius, Stobie and Soderburg. 
 
 The Principle of the Heroult Furnace is diagrammatically repre- 
 sented in Fig. 39C. The practicability of this principle is shown by the 
 fact that the Heroult electric steel furnace heads the list of such furnaces 
 in use for the manufacture of steel. This popularity of the Heroult furnace 
 is due to the fact that the application of this principle gives a furnace of 
 the greatest efficiency combined with simplicity of construction and adapt- 
 ability to many different uses. Details of the construction of this furnace 
 will be given later. At present it is desired to explain only the method 
 of heating. All the electrodes, as indicated by the figure, are suspended 
 from supports over the roof, through which they project to within an inch 
 of the surface of the slag. As the electrodes are so far separated from each 
 other as to prevent arcs between them, several resistances are introduced 
 in series. For example, let the current be considered as passing from A 
 to B. Then as the current jumps the gap at the foot of A, it forms an arc 
 and passes into the slag, which also has a high resistance. The metal, 
 having a much lower resistance, then acts as a conductor for the current 
 to the region directly beneath the foot of electrode B, where the current 
 must again pass through the layer of slag and form a second arc as it jumps 
 the gap between slag and electrodes. It is evident that practically all 
 the heat is formed by the arcs above the slag, which acts as a shield to 
 the metal and protects it both from the carbon vapors thrown off from 
 the foot of the electrode and from the exceedingly high temperature at 
 this point. Portions of this heat is next imparted to the metal through 
 the slag, where it may be distributed to all parts of the bath by conduction 
 and convection. The distribution of the heat is thought to be aided by 
 a slight motor effect produced by the current upon the metallic bath. 
 Furnaces of this type using single phase and three phase current are in use. 
 The only change necessary to be made for three-phase current is the insertion 
 
METALLURGY 267 
 
 of a third electrode. In the development of these furnaces Heroult stands 
 almost alone, though slight modifications have been introduced by Chaplet 
 and Anderson. 
 
 Some General Conclusions: From what has been said, the following 
 facts are evident: 1. That the only part the electric current plays in the 
 manufacture of steel is in the production of heat. 2. That for producing 
 this heat there are really but two methods available, which has resulted 
 in two successful types of furnaces, namely, the induction type and the 
 arc type. 3. , That from a strictly metallurgical point of view no one of 
 these types represents any marked advantage over the other, in their 
 present high state of development. This statement has no relation to 
 claims of the inventors to mechanical and electrical points of excellence in 
 their respective apparatus. What advantage electric heating has over 
 other methods of heating is now to be discussed. 
 
 SECTION VI. 
 
 GENERAL FEATURES PERTAINING TO THE METALLURGY OF STEEL MADE 
 BY ELECTRO-THERMAL PROCESSES. 
 
 Advantages of Electric Heating: To the metallurgist the electric 
 method of heating is an ideal one for the following reasons, which are 
 characteristic of it: 1. It makes heat available very quickly and at will, 
 and gives an unusually high temperature. 2. The heat may be regulated 
 very nicely, which fact permits a charge to be brought to any temperature 
 desired and to be maintained steadily at that temperature. 3. Being the 
 cleanest of heating agents, it exerts no deleterious effect upon the material 
 heated by the evolution of harmful gases. 4. It permits oxidizing 
 reducing or neutral operations to be carried on at will. How these 
 characteristics of electric heating work to the advantage of the metallurgist 
 and permit him to obtain a product of the highest quality from raw 
 material s of any grade will be understood from a study of the topics to 
 follow. I n this connection the chemical possibilities are of first importance. 
 
 Refining Procedure: The first effect of the characteristics noted 
 above is to bring the whole operation of refining metal under complete 
 control. The electric furnace is to the metallurgist what the casserole and 
 crucible are to the chemist. The bath, then, represents a mixture of com- 
 pounds and elements, any one of which may be removed at will by the 
 use of the proper reagents. The condition may be illustrated by assuming 
 a furnace is to be charged with the crudest of raw materials, cold pig iron, 
 and then showing how each impurity is removed. A study of the practice 
 of many plants shows that the following procedure would be carried out: 
 After the furnace has received its metallic charge, an oxidizing flux of lime 
 and iron ore will be added. A part of this flux may be charged ahead of 
 the metal, exactly as in the case of the basic open hearth. After the charge 
 
263 ELECTRIC PROCESS 
 
 has been melted down, the furnace will be tilted slightly and the slag which 
 has formed will be carefully raked off. This addition of flux and removal 
 of slag will be repeated as often as may be necessary. A cleansing flux 
 of lime alone will then be added and raked off. During this period the 
 temperature should be kept low, because phosphorus is not readily oxidized 
 at high temperatures in the presence of carbon. The bath will then be 
 covered with a flux consisting of about 5 parts lime, 1 part sand or other 
 form of silica, 1 part fluorspar, and M P ar * of carbon in some convenient 
 form, such as coke, coke carbon, old electrode, etc. The furnace will then 
 be tightly closed, and the temperature raised. In the case of induction 
 furnaces, it will be found necessary after about two hours to introduce 
 small portions of ferro silicon, silico-aluminum, or silico-spiegel. These 
 alloys act energetically as deoxidizers and form a fluid slag which rises 
 to the surface. In the Heroult furnace carbon is the only deoxidizer used. 
 After the steel shall have been thoroughly deoxidized, any carburizing 
 material or alloys will be added to bring the steel to the desired com- 
 position. When all such material will have been melted, and sufficient 
 time will have elapsed to permit them to mix with the bath, the molten 
 steel will be poured as a finished ingot product. This process divides 
 itself into three distinct periods, namely, an oxidizing period, a reducing 
 period, and a finishing period, a combination of conditions impossible of 
 attainment in any other process. The action brought about during each 
 of these periods, and the reasons for using the reagents employed may now 
 be discussed. 
 
 The Oxidizing Period: It is evident that the action of the oxidizing 
 flux must result in the removal of silicon, manganese and phosphorus in a 
 manner similar to that of the basic process. The important distinction 
 between open hearths and electric furnaces should be noted here. It is 
 this: All the oxygen introduced into the electric furnace must be in the 
 solid form as observed above, and the amount can be easily controlled, 
 whereas the air admitted to open hearths furnishes an unlimited amount 
 of oxygen that cannot be controlled. Therefore, since these three elements 
 are easily oxidized at low temperatures before the carbon, the reaction 
 may be stopped and the bath held at almost any carbon content desired. 
 With respect to phosphorus, it has been suggested that this element may 
 be removed as phosphide by means of some metal, as calcium. While this 
 scheme is theoretically possible, it is still impracticable, for it is a difficult 
 thing to find a metal that would not alloy with iron in preference to com- 
 bining with phosphorus, or whose phosphides would not so alloy. Quite 
 frequently, traces of phosphorus are found in the reducing slags, but this 
 method of eliminating phosphorus, though often attempted, has not been 
 made successful beyond the removal of very small amounts and at great 
 expense. Hence, the only sure way of removing this element is to oxidize 
 it to phosphoric acid, neutralize with lime, and rake off the resulting slag. 
 In refining purer materials than pig iron, where the removal of silicon 
 
METALLURGY 269 
 
 and manganese would not be required, the oxidizing slag is called the 
 dephosphorizing slag. Considerable sulphur is removed during this period 
 in the electric furnace, especially where an ore high in manganese is used, 
 whereas in the basic furnace, its removal is a very uncertain quantity. 
 
 The Reducing Period : This is the period in which the electric furnace 
 exhibits its great superiority over other modes of refining iron. During 
 the period, the bath is almost completely deoxidized by means of carbon 
 alone, and the removal of sulphur is positive and can be made almost com- 
 plete. Its entire removal seems to be impracticable, as a content of less 
 than .010% is obtained only after prolonged and expensive treatment. 
 The flux added is, therefore, called either the desulphurizing or deoxidizing 
 flux. 
 
 Oxygen : Oxygen occurs in steel principally as FeO, as has been stated 
 in previous chapters, in which its harmful effects were also dwelt upon. 
 Its removal may be represented by the following equation in which M 
 may represent a great number of suitable elements: 
 FeO+M=MO+Fe. 
 
 The deoxidation may be brought about in the induction furnace only 
 by means of the special deoxidizers previously noted, whereas in the Heroult 
 furnace carbon alone may be employed. The use of carbon alone has been 
 objected to because it was argued that the use of this element would give 
 the oxide, CO, which is a gas, and the formation of this gas in the metal 
 is objectionable. That steel at high temperature either combines with or 
 dissolves this gas is fairly well established, for by experiment it has been 
 shown that a given steel has a higher melting point in an atmosphere such 
 as nitrogen than it has in an atmosphere of carbon monoxide. There is 
 also good reason to believe that this gas is again liberated at certain tem- 
 peratures on cooling. If this be true, then deoxidizing with carbon may 
 give opportunity for formation of blow holes and other defects in the ingots. 
 Some metal, then, whose oxide is a solid that will easily come to the surface 
 as slag is to be preferred for this purpose. This metal must not be volatile 
 at high temperatures and must dissolve or alloy with the iron. The metals 
 that best meet these requirements are the ferro-alloys of manganese, silicon, 
 vanadium, titanium, and metallic aluminum. For many reasons manganese, 
 silicon and aluminum have proved the most satisfactory deoxidizers. How 
 the carbon, acting through one or more of these elements, may be used 
 to accomplish the deoxidation without injury to the steel is well illustrated 
 by the practice at Duquesne, to be described later. 
 
 Removal of Sulphur: According to the statements of authorities 
 upon the subject, sulphur may be removed in five ways: (1) As calcium 
 sulphide, formed by the action of lime and carbon on ferrous sulphide at 
 the high temperature of the arc furnace; (2) As calcium sulphide from 
 the reaction of lime, ferrous sulphide, and calcium carbide at the higher 
 
270 ELECTRIC PROCESS 
 
 temperatures of the arc furnaces; (3) As calcium sulphide through the 
 reaction of lime, ferrous sulphide, and silicon at the lower slag temperatures 
 of the induction furnace; (4) As calcium sulphide through the reaction of 
 calcium fluoride, ferrous sulphide, and silicon; (5) As iron sulphide from 
 the action of ferrous oxide on calcium sulphide. The reactions illustrating 
 the removal of sulphur in electric furnaces are as follows: 
 
 (1) FeS+CaO+C=Fe+CaS+CO. Occurs in arc furnaces only. 
 
 (2) 3 FeS+2 CaO+CaC 2 =3 Fe+3 CaS+2 CO. in arc furnaces only. 
 
 (3) 2FeS+2CaO+Si=2Fe+2CaS+SiO 2 . Used with induction furnaces. 
 
 (4) 2CaF 2 +2FeS+Si=2CaS+SiF 4 +2Fe. May occur in either induc- 
 tion or arc furnaces. 
 
 (5) FeS+CaO == CaS+FeO. May occur in arc furnaces. 
 
 An inspection of these reactions shows that certain requirements must 
 be met before the elimination of sulphur can be brought about. Thus, 
 in all cases a highly basic clag is essential, and in no case can desulphur- 
 ization be effected before deoxidation of the metal and slag has been accom- 
 plished. The importance of the presence of elementary carbon or silicon 
 is evident, and both these elements tend to react with iron or manganese 
 oxides rather than with sulphides and lime. Furthermore, reaction (5) 
 is reversible, acting from right to left in the presence of very slight oxidizing 
 influences. Because of the relation in concentration maintained between 
 oxides in the metal and oxides in the slag, both deoxidation and 
 desulphurization of the metal may be brought about by additions 
 of carbon to the slag, if the temperature is sufficiently high. In 
 arc furnaces this method is employed almost exclusively; but induction 
 furnaces, owing to their lower temperatures, require that desulphuri- 
 zation be effected by the use of silicon as shown in reaction (3) and 
 (4), both of which take place at much lower temperatures than (1) and 
 (2). In the arc furnace complete deoxidation of metal and slag is 
 recognized by the presence of calcium carbide in the slag. 
 
 The Finishing Period: With the dephosphorization and subsequent 
 deoxidation of the bath, the contents of the furnace may be brought to a 
 neutral or slightly reducing state, when- the final additions may be made 
 for finishing the steel to specifications. In order to save time it is a common 
 practice to make some additions before the desulphurizing period com- 
 mences, especially if the additions are to be made in large quantities, 
 which would chill the bath if ad'ded all at one time. As the conditions 
 are reducing or neutral, loss of alloying elements is reduced to a minimum, 
 and the composition of the steel can be controlled with precision. The 
 
METALLURGY 271 
 
 ability to finish the steel in the furnace gives the electric process another 
 advantage over any of the older processes, also, as this practice tends to 
 produce greater uniformity in the steel. 
 
 Some Comparisons: When deoxidation of the bath is complete the 
 contents of the electric furnace represents the nearest approach to perfect 
 chemical equilibrium that has yet been attained in other large metal- 
 lurgical operations. In the converter and the open hearth the metals are 
 subjected to the action of air and gas; in the crucible the metal takes up 
 carbon and silicon; but in the electric furnace, the action of the metal on 
 the basic lining being very slight, there is no exchange of elements 
 between metal and slag, if traces of phosphorus be excepted. 1 Of course, 
 if the dephosphorizing slag of the oxidizing period has not been completely 
 removed before deoxidization begins, some of the phosphorus compounds 
 in this slag, as well as some in the new slag, will be reduced. However, 
 with careful watching, this gathering-up of phosphorus by the metal may 
 be entirely avoided. As to the relative merits of the various types of 
 electric furnaces, the results obtained are about equal. However, it is 
 evident that arc furnaces are well suited for reducing processes, while 
 induction furnaces lend themselves most readily to oxidizing purposes. 
 
 Fluxing Materials used in the electric furnace should be as pure as 
 possible and free from injurious amounts of sulphur or phosphorus. The 
 lime and fluorspar should not contain more than small amounts of magnesia, 
 as it makes the slag less fusible, which fact is of great importance when 
 the high basicity of the slags is considered. For recarburizing, a material 
 low in volatile matter, phosphorus, and ash should be used. At Duquesne, 
 anthracite coal is chiefly employed. 
 
 General Manufacturing Practice: As indicated previously, the 
 electric furnace may be used to refine pig iron direct. But as the major 
 portion of the refining may be accomplished much more cheaply by one 
 of the older methods, direct refining is not economically practiced at present. 
 Two courses of procedure, therefore, remain. In the one, cold scrap iron 
 and steel of any grade as to quality is remelted and refined to produce 
 steel of high quality, while in the other the electric furnace is made part 
 of a duplexing process, whereby it is used in conjunction with one of the 
 older methods of refining to produce superrefined plain steels or alloy 
 steels. This second plan is the one used by the U. S. Steel Corporation. 
 At the South Chicago Works, the electric furnace is used in conjunction 
 with Bessemer converters and tilting basic open hearth furnaces, and may 
 be used in the finishing stage of either a duplex or a triplex process; but 
 in either case the steel finished in the electric furnace is taken from the 
 open hearth. At Duquesne the duplexing is in connection with basic open 
 hearths only, and to furnish a concrete example of the construction and 
 workings of the electric furnace this plant will be described. 
 
 Lyon and Keeney, Electric Furnaces for making Iron and Steel. Bureau of 
 Mines Bulletin 67, Page 125. 
 
272 
 
 ELECTRIC PROCESS 
 
CONSTRUCTION OF FURNACE 
 
 273 
 
274 
 
 ELECTRIC PROCESS 
 
 FIQ. 42. Heroult Electric Furnace. Vertical Section Through Tower and Furnace. 
 
DUQUESNE PLANT 275 
 
 SECTION VII. 
 
 THE DUQUESNE PLANT FEATURES PERTAINING TO ITS CONSTRUCTION. 
 
 Equipment: This plant was completed early in 1917, and is located 
 at one end of No. 2 open hearth plant and in the same building. The 
 special equipment of the plant may be said to consist of one 20-ton Heroult 
 tilting furnace with a transformer station and testing laboratory, three 
 charging ladles, three pouring ladles and one traveling over-head crane 
 for charging. Other equipment such as teeming cranes, molds, cars, etc. 
 are part of the regular equipment of the open hearth plant. The open 
 hearth floors are on two levels. The electric furnace, therefore, standing 
 in line with the open hearths, is elevated and so constructed that it may 
 be charged from the charging floor level, and tapped into the pouring ladle 
 on the ground floor thirteen and one half feet below the charging level. 
 The transformer station is in a brick building just outside the open hearth 
 building, and about eighteen feet from the center of the furnace. Three 
 transformers are used, the combined capacity of which is 3500 K. W. 
 From the transformers the current is conducted by means of bus bars to 
 the corner of the transformer house nearest the pouring side of the furnace. 
 Thence it is carried along a large number of heavy copper cables to the 
 furnace. Wood bars, tied together like sections of a picket fence, are used 
 to insulate the three lines of cables from each other. This arrangement 
 furnishes the flexible connections necessary to permit the tilting of the 
 furnace. 
 
 Construction of the Furnace Shell: The furnace is circular in form, 
 and has an outside diameter of approximately sixteen feet. The shell is 
 made of plate steel one inch thick, riveted together. This shell may be 
 considered as being made in these three parts: a channelled band for the top, 
 which is removable and made up of riveted plates; a side wall which is 
 cylindrical in shape; and a bottom, which is shaped somewhat like a hopper. 
 The bottom and side walls are riveted together. The shape of the bottom, 
 front to back, is made to conform to heavy steel rockers, which rest on two 
 heavy castings that serve as tracks. The rockers and tracks are provided 
 with teeth, which mesh into each other and thus prevent the furnace from 
 creeping. The weight of the furnace is supported by two brick piers upon 
 which the tracks, or stationary racks, are bedded. The tops of these piers 
 are about five feet above the ground. Attached to the rockers at the back 
 of the furnace, are two crank shafts which in turn are connected to two large 
 gear wheels, some five feet in diameter. These large wheels are geared to a 
 140 h. p. motor, which provides the power for tilting the furnace. 
 
 The Furnace Lining is made up of three layers of different materials. 
 Next to the shell, in the bottom and sides, is placed a four and one half 
 inch layer of fire brick laid edgewise, and upon this is laid a continuous 
 bottom and side wall of magnesite brick. The bottom course is nine 
 
276 
 
 ELECTRIC PROCESS 
 
 nn nn n n 
 
CONSTRUCTION OF FURNACE 277 
 
 inches thick, while the side wall is thirteen and one half inches in thickness. 
 Upon the bottom brick is then sintered a layer, about thirteen inches thick, 
 of dead burned magnesite, which is banked on the sides to a safe distance 
 above the slag line. 
 
 The Roof is slightly dome-shaped, twelve inches in thickness, and made of 
 silica brick set in on end. The first course next the channelled band is 
 made up of large skew-back brick. Thus, the roof is made self-supporting. 
 As a roof lasts for forty to seventy heats only, two extras are held in 
 reserve, ready to be placed in case a roof fails unexpectedly. In the roof, 
 three openings are made for the electrodes. Each opening corresponds to 
 one vertex of a equilateral triangle, each side of which is about six feet long, 
 and the center of which is the center of the roof. When in place, the roof 
 sets so that one vertex points toward the vertical guides for the 
 electrodes, which are on the side of the furnace next the transformers. 
 While in use, the top is bolted to four brackets on the shell to prevent 
 the top from slipping when the furnace is tilted. 
 
 Controlling the Electrodes : Through the three openings, noted above, 
 the electrodes extend into the furnace for a distance of about four feet. 
 In order to make the electrodes adjustable, they are attached to horizontal 
 arms that project out over the furnace from heavy vertical rods arranged 
 to move up and down within vertical guides. At the top of these rods and 
 properly insulated from them, the cables that carry the current from the 
 transformer house to the furnace are bolted and welded to bus bars which 
 lead to the electrode holders. Thus, the same motion is imparted to both 
 the electrode holders and the bus bars. Each of these rods is supported 
 and moved by means of a cable attached to its base and leading over a 
 small drum geared to a small electric motor. These motors act through 
 automatic regulators, which serve to keep the end of the electrode at the 
 proper arcing distance from the bath. By reversing a switch on the switch 
 board, these motors may be operated independently of the regulator. 
 Furthermore, the lifting device is provided with a hand wheel, so as to be 
 operated like a common windlass, whereby the electrodes may be regulated 
 by hand. 
 
 The Electrode Holders are made in two parts, both of which are in 
 the form of a two pronged fork. The upper part is of copper and makes 
 the connection between the electrodes and the bus bars, which are securely 
 bolted to it. The electrodes are held between the two prongs, and since 
 the distance between these prongs is about twenty-four inches, contact 
 blocks must be used for electrodes less than twenty-four inches in diam- 
 eter. A right-and-left screw bolt connects the ends of the two prongs, 
 which enables the holder to be opened and closed at will and permits the 
 electrode to be securely clamped in place. By this arrangement, electrodes 
 of any size up to twenty-four inches diameter may be used. The lower 
 part is made of steel and acts as a support for the upper part. These two 
 parts, carefully insulated from each other, are held together by means of 
 
278 
 
 ELECTRIC PROCESS 
 
CONSTRUCTION OF FURNACE 279 
 
 insulated bolts. Finally, this lower part is fastened to the horizontal arm, 
 previously mentioned, by an insulated flange joint. The upper prongs are 
 water cooled. 
 
 The Electrodes used at the plant may be of graphite, twelve inches in 
 diameter, or of amorphous carbon, twenty-four inches in diameter, and are 
 received in sections six feet long. By means of threaded holes in the ends 
 of the electrodes and headless screws of the same material to fit, these 
 pieces may be joined together so as to give a continuous feed of electrodes 
 to the furnace. In this way the great waste of electrodes from unused ends 
 is avoided. These threaded holes also serve a useful purpose in removing 
 the electrodes and changing them in the holders. As the carbons are 
 constantly burning away, this change is frequently necessary. Since the 
 electrodes are rather heavy a six foot length of graphite electrode weighs 
 426 Ibs. the use of the crane is made necessary. By means of a linked 
 pin of steel, threaded to fit the hole, the electrodes can easily be handled 
 with the crane hook. Experience seems to indicate that graphite electrodes 
 are best for molten charges but that amorphous electrodes are very well 
 suited for use in melting cold charges. A water cooled copper ring encircles 
 each electrode at its entrance into the furnace. 
 
 Furnace Openings: Besides the holes in the roof for the three elec- 
 trodes, the furnace has three openings in the side wall, all located within 
 an arc of 180 on the circumference. One is the tapping hole, a small opening 
 through which the steel is poured into the steel ladle. This opening is 
 provided with a spout, so constructed as to act as a slag skimmer when 
 the furnace is tapped. Opposite the tapping hole is the charging door, 
 through which the molten metal from the open hearth is charged, while 
 half way between these two openings is located a second charging door, 
 which can be used only for charging solid materials by hand. Both these 
 charging holes are closed with neat fitting brick lined doors, which are 
 lifted by means of compressed air cylinders in much the same way as open 
 hearth doors. 
 
 SECTION VIII. 
 
 OPERATION OF THE FURNACE. 
 
 Practice at Duquesne Plant: The Duquesne furnace is used to make 
 plain steels of any carbon content and many different alloy steels. For all 
 the steels made in the furnace the raw material charged is finished basic 
 open hearth steel, hence the refining in the electric furnace consists of 
 deoxidizing and desulphurizing only. Unless the specifications on the 
 basic steel should coincide with those for the electric, which is seldom the 
 case, additions are made to the charge either in the transfer ladle or in 
 
280 "ELECTRIC PROCESS 
 
 the furnace. No additions of any kind are made in the pouring ladle or at 
 the time of tapping, as is the common practice in making open hearth 
 steel. A good example is furnished in the case of carbon. When it is 
 necessary to raise the per cent, of this element, as is the case on orders 
 calling for a higher per cent, than that of the open hearth heat, the amount 
 required above that supplied by other additions is added, in the form of 
 anthracite coal, to the steel in the transfer ladle, in the furnace, or a part 
 in both. Additions of other elements are, as a rule, added after deoxida- 
 tion of the metal is well advanced or completed. This ability to finish 
 the heat in the furnace is a decided advantage in favor of the electric 
 process, as a more homogeneous product is thus obtained. 
 
 Charging: Omitting the mechanical and electrical features, the 
 operation of the furnace, in general, is carried out as follows: Approxi- 
 mately twenty tons of a suitable open hearth heat is teemed from the 
 steel ladle into the charging ladle, for the electric furnace. This charge is 
 then conveyed by a dinkey, over a narrow gauge track, to the furnace, 
 into which it is poured through a portable spout. During the pouring, a 
 test for chemical analysis is taken, and upon the results of this analysis is 
 based the approximate amount of carbon and manganese to be added. An 
 increase of three to five points in the manganese content of the steel usually 
 occurs during the deoxidizing period, and must be allowed for. If a medium 
 or high carbon heat is being made from a low carbon open hearth heat, requiring 
 the addition of a large amount of carbon, the greater portion of this 
 element is added in the form of anthracite coal, which is thrown into the 
 furnace as the heat is being charged. This procedure is necessary to 
 insure that the coal will be absorbed by the steel. 
 
 Deoxidizing: As soon as the charging has been completed, the elec- 
 trodes are adjusted, and the current is turned on. Since the charge usually 
 freezes over on top, especially in the case of low carbon steels, nothing 
 further is done until this solidified layer is completely melted. As soon 
 as the bath is in a state of complete fusion, the first slag mixture, consisting 
 approximately of four parts lime and one part fluorspar or one part clean 
 sand for low carbon heats, is added; for high carbon heats, the mixture 
 may contain about one-third part coke dust. Soon after this addition, the 
 second sample for chemical analysis is taken to determine the per cent, of 
 carbon and manganese in the bath, and while these determinations are 
 being made further additions of the first slag mixture takes place. Samples 
 of the slag taken at this time are usually brown in color and contain vary- 
 ing amounts of manganese oxide, which fact, shows that the iron oxide in the 
 steel is being reduced by the manganese present. A decided brown color 
 can be taken to indicate that the deoxidation of the steel is well advanced. 
 A second slag mixture composed of suitable proportions of lime, fluor spar, 
 sand and coke dust is now added. Soon after the addition of this mixture 
 the slag becomes less vitrious, shows a tendency to slake, and begins to 
 

 DUQUESNE PRACTICE 281 
 
 fade in color. If the heating be continued long enough, with proper addi- 
 tions of the carbonaceous flux, samples of the slag will slake when cold and 
 become gray, or even white, in color. Such behavior of the slag indicates 
 that the deoxidation of the bath of metal and slag has been completed. 
 This condition is also determined by means of the water test. If at this 
 time a small sample of the slag while hot be immersed in a little water, 
 the odor of hydrogen sulphide can usually be detected, and, if deoxidation 
 is complete, the smell of acetylene gas can also be detected. 
 
 Finishing the Heats: During the deoxidation of the bath the results 
 of the second chemical analysis have been reported, and if the per cents, 
 of carbon and manganese are satisfactory, any alloys that may be required 
 by the order are added as soon as the slag condition will permit. If the 
 carbon content should be too low, it is raised to the required point by the 
 addition of a proper weight of cold, very low phosphorus pig iron. The 
 bath is chilled somewhat by the addition of the alloys, especially if they 
 are added in large amounts, and about three-quarters of an hour is required 
 to heat the bath up to the tapping temperature. Besides, in order to give 
 the alloys time to mix with the steel, no additions are made for thirty 
 minutes before tapping except in the case of 50% ferro silicon, which is 
 added ten to fifteen minutes before tapping to avoid losses of the 
 element. 
 
 Tapping and Teeming: When enough time has elapsed to melt all the 
 alloys or other additions, slaking and water tests are made on the slag, and 
 if these indicate a satisfactory condition of the slag, the heat is tapped. In 
 tapping the heats, care is taken to prevent slag from running into the steel 
 ladle with the metal. The special skimmer with which the tapping hole is 
 provided for this purpose has already been mentioned. The steel is teemed 
 very carefully, being usually box poured. In this method of teeming the 
 stream of molten metal from the ladle flows into the middle of a box made 
 in three compartments. From the middle compartment the steel overflows 
 into the two end ones, which are provided with nozzles. This arrangement 
 permits these nozzles to be carefully centered over two ingot moulds before 
 the pouring is begun. Special care is taken in preparing the ingot moulds, 
 so as to prevent ingot defects due to bad moulds. The steel is allowed 
 to stand two hours in the moulds, so as to insure that solidification is 
 complete before it is stripped. 
 
 Scrap Heats: Besides the refining of molten open hearth steel, which 
 has just been described, the furnace is occasionally used to make steel 
 from scrap. When using scrap, two methods may be followed. Thus, 
 the charge may consist of cold scrap or be made up of scrap and molten 
 steel. When scrap alone is used, it must be small, and the coarser material 
 is charged first with the finest on the top. Even then the power fluctuations 
 are great, and some difficulty is experienced in melting the scrap. These 
 
282 ELECTRIC PROCESS 
 
 difficulties are over-come for the most part by starting the furnace on a 
 short charge of molten steel and then adding the scrap to this charge. 
 This latter method is the one employed most often at this plant. After 
 the melting period the procedure is then the same as that already described 
 for molten charges. An examination of the following tables will give a more 
 concrete idea of the method of handling the different kinds of steel made 
 in the electric furnace. 
 
 Table 38. Showing History of a Heat of Low Carbon Plain Steel 
 Made in the Electric Furnace. 
 
 Analysis of molten charge steel as finished at open hearth: C.=.09%; 
 Mn.=.38%; P.=.014%; S=.043%. 
 
 Order: C. .10/.15%; Mn.=.30/.50%; P. under .035%; S. under .040%; 
 Si. under .04%. 
 
 Time Additions 
 
 8:10 Charge, 46700 Ibs. Test. C.=.06%; Mn.=.33%; P.=.010%; 
 
 S.=.038%. 
 8:35 Power on. 
 
 9:35 1st slag mixture. Lime, 500 Ibs., fluor spar, 150 Ibs. 
 9:50 485 Ibs. pig iron added. 
 10:00 % second slag mixture, Lime, 750 Lbs., fluor spar, 125 Ibs., coke 
 
 dust, 100 Ibs; sand, 100 Ibs. 
 10:05 Sample for chemical analysis. 
 10:20 104 Ibs. ferro manganese added cold. (Laboratory report shows 
 
 C.=.08%;Mn.=.19%.) 
 
 10:30 Chemical analysis, C.=.10%; Mn.=.32%. 
 10:45 50 Ibs. Ferro manganese added cold. 
 10:50 300 Ibs. pig iron added cold. 
 10:55 y% second slag mixture. 
 11:00 34 third slag mixture Lime, 100 Ibs.; fluor spar, 30 Ibs.; coke dust, 
 
 75 Ibs. 
 
 11 :30 34 third slag mixture. 
 11 :40 34 third slag mixture. 
 11 :45 34 third slag mixture. 
 12:00 20 Ibs. coke dust. 
 12:30 30 Ibs. coke dust. 
 12:35 50 Ibs. ferro manganese added cold. 
 12:50 Heat tapped. 
 
 Final analysis: C.=.12%; Mn.=.35%; P.=.007%; S.=.035%; 
 
DUQUESNE PRACTICE 283 
 
 Table 39. A Heat of High Carbon Plain Steel. 
 
 Molten charge Steel as finished at open hearth C.=.19%; Mn.=.28%; 
 P.=.016%; S.=.041%; 
 
 Order: C.=.95/1.Q5%; Mn.=.20/.35%; P.=under.030%; S.=under.030%; 
 Si.=.10/.25%. 
 
 Time Additions 
 
 3:20 150 Ibs. Anthracite Coal added to ladle. 
 
 3:30 Charge, 52400 Ibs. Test: C.=.29%; Mn.=.23%; P.=.012%; 
 
 S.=.042%. 
 
 660 Ibs. Coal added in furnace as heat is being charged. 
 3:35 155 Ibs. coal added. 
 3:40 Power on. 
 
 5:35 % first slag mixture. Lime, 1200 Ibs. ; fluor spar, 200 Ibs. ; coke dust, 
 100 Ibs; sand, 100 Ibs. 
 
 ' m. i A i /front door, C.=.85%; Mn.=.29%. 
 6:15 Chemical Analysis< . , , ' 0> ' 
 
 \side door, C.=.85%; Mn.=.28%. 
 
 6:15 % first slag mixture. 
 
 6:30 ^s first slag mixture. 
 
 6:40 % first slag mixture. 
 
 6:50 }4 first slag mixture. 
 
 7:00 % second slag mixture. Lime, 100 Ibs.; fluor spar, 25 Ibs.; coke 
 
 dust, 15 Ibs. 
 
 7:05 1700 Ibs. pig iron. 
 
 7:10 Y$ second slag mixture. 
 
 7:15 Y^ second slag mixture. 
 
 7 :20 175 Ibs. 50% ferro silicon. 
 
 7:40 Power off, and heat tapped. 
 
 Final analysis, C.=.97%; Mn.=.29%; P.=.007%; S.=.016%; Si.=.19%. 
 
284 ELECTRIC PROCESS 
 
 Table 40. A Low Carbon Alloy Heat. 
 
 Molten Charge as finished at Open Hearth, C.=.09%; Mn.=.34%; 
 P.=.015%; S.=.034%. 
 
 Order: C.=.12/.20%; Mn.=.40/.70%; P.=under .040;% S.=under 
 .040%; Si.==under .20%; Cr.=.40/.80%; Ni.=1.00/1.75%. 
 
 Time Additions 
 
 10:55 Charge, 50800 Ibs. Test, C.=.08%; Mn.=.29%; P.=.010%; 
 S.=.042%. 
 
 11:15 Power on. 
 
 12:00 First slag mixture. Lime, 500 Ibs.; fluor spar, 150 Ibs.; 
 12:30 % second slag mixture. Lime, 750 Ibs.; fluor spar, 125 Ibs.; coke 
 dust, 100 Ibs; sand, 100 Ibs. 
 
 . /front door, C.=.08%; Mn.=.24%. 
 
 12:45 Chemical Analysis< . , , ' ' ' ' Q 
 
 [side door, C.=.08%; Mn.=.22%. 
 
 12:50 200 Ibs. ferro manganese; Mn.=78%. 
 
 12:55 446 Ibs. ferro chrome; Cr.=70%. 
 
 1:00 730 Ibs. nickel; Ni. =99%. 
 
 1:30 Yz third slag mixture Lime, 50 Ibs.; coke dust, 75 Ibs. 
 
 1 :45 3^ third slag mixture. 
 
 1:50 30 ^3S. ferro manganese and 60 Ibs. 50% ferro silicon. 
 
 2:00 Heat tapped. 
 
 Final Analysis, C.=.20%; Mn.=.51%; P.=.009%; S.=.028%; 
 Si.=.03%; Cr.=.54%; Ni.=1.28%. 
 
DUQUESNE PRACTICE 285 
 
 Table 41. A High Carbon Alloy Heat. 
 
 Molten charge Steel as finished at open hearth C.=.20%; 
 Mn.=.35%; P.=.010%; S.=.035%. 
 
 Order: C.=.95/1.05%; Mn.=.35/.50%; P.=under .030%; S.=under 
 .030%; Si.=under .03%; Cr.=1.35/1.65%; V.==over .18%. 
 
 Time Additions 
 
 11:48 150 Ibs. anthracite coal added in transfer ladle. 
 
 11:50 Charge, 54500 Ibs. Test, C.=.07%; Mn.=.34%; P.=.006%; 
 
 S.=.040%. 
 12:05 Power on. 
 
 12:50 % first slag mixture Lime, 1300 Ibs. ; fluor spar, 160 Ibs; sand, 125 Ibs. 
 'Front door, C.=.89%; Mn.=.34%. 
 
 1:20 Chemical anal ysiss ,-,., , ^ nn ~ -, _~ 
 (Side door, C.=.90%; Mn.=.35%. 
 
 1:25 % slag mixture. 
 
 1:35 y& second slag mixture. Lime, 150 Ibs.; coke dust, 150 Ibs. 
 
 2:00 1180 Ibs. ferro chrome, Cr.=70%. 
 
 2:10 to 2:40 Remainder of second slag mixture added at intervals of ten 
 
 minutes. 
 
 2:55 353 Ibs. ferro vanadium. V.=35% 
 3:05 60 Ibs., 50% ferro silicon. 
 3:30 Heat tapped. 
 
 Final analysis, C.=.99%; Mn.=.35%; P.=.006%; S.=.017%; Si.=.07%; 
 Cr.=1.44%;V.=.22%. 
 
286 ELECTRIC PROCESS 
 
 SECTION IX. 
 
 THE CHEMISTRY OP THE PROCESS. 
 
 Deoxidation of the Bath: The tracing of the chemical changes that 
 take place in the process employed at this plant furnishes an interesting 
 study. Since the charge is finished open hearth steel containing the usual 
 amount of manganese, it is to be expected that this element would play 
 a most important part in the deoxidation of the steel. The fact, plainly 
 shown by the preceding records, that the manganese is usually several 
 points lower in the charging test than in the open hearth steel is proof of 
 its deoxidizing action, which would be expected to continue in the furnace. 
 The removal of oxygen, then, is represented by the following reaction: 
 FeO+Mn=MnO+Fe. The MnO, being less soluble in the molten metal 
 than FeO, rises to the surface and becomes a part of the slag. This action 
 is identical with that in the ladle in finishing open hearth steel, but the 
 result is not the same in the two processes for two reasons: First, for want 
 of time, the deoxidation is not completed in the ladle, whereas in the electric 
 furnace it is complete. Second, the MnO in the electric furnace comes 
 under the reducing action of the carbon contained in the slag mixture and 
 is reduced, thus: MnO+C=Mn+CO. The carbon monoxide, CO. being a 
 gas, becomes a part of the atmosphere of the furnace, and the manganese 
 returns to the bath, as is indicated by the fact that the per cent, manganese 
 in the steel usually rises after the addition of the carbonaceous flux. 
 
 Desulphurizing the Metal: With the elimination of oxides from the 
 slag, the lime, under the influence of the extremely high temperature of 
 the arc which prevails under and in the immediate vicinity of the electrodes, 
 begins to be reduced by the carbon of the coke dust, with the consequent 
 formation of calcium carbide, thus: CaO+3C=CaC2+CO. It is at this 
 point that desulphurization takes place. Since manganese sulphide, like 
 the oxide, is less soluble in iron than ferrous sulphide, it is probable that 
 this element also aids in this process. Whether manganese sulphide or 
 ferrous sulphide plays the most important role in the sulphur reactions is 
 difficult to decide, but that the removal of the sulphur from the bath takes 
 place through the formation of calcium sulphide, which is insoluble in molten 
 iron, cannot be doubted. These reactions are two in number and may be 
 written thus: 
 
 /FeS , /Fe 
 
 \MnS- 1 -'- - Mn+ CaS + C 
 
CHEMISTRY 
 
 287 
 
 With metallic oxides present in the slag, neither of these reactions will take 
 place, ae the oxide would react with the calcium sulphide, thus: 
 
 CaS+ 
 
 /FeO /FeS 
 
 "\MnO~~\MnS" 
 and with calcium carbide according to this reaction: 
 
 Hence the presence of CaC2 in the slag is a guarantee that the bath is 
 deoxidized. In the water test the presence of considerable quantities of 
 CaC2 in the slag is detected by the odor of acetylene gas, which is generated 
 in accordance with the following reactions: CaC2+H2O=C2H2+CaO. 
 These facts are further illustrated by the following analysis of heats and 
 slags: Slags AZ No. 3 and AZ No. 5 illustrate an undesirable and a desirable 
 slag, respectively. The relatively high iron and manganese content of AZ 
 No. 3 indicate incomplete deoxidation, with a low sulphur and calcium 
 carbide content resulting. 
 
 Table 42. Analysis of Tests from Electric Furnace Heats. 
 
 Heat Number 
 
 Testa 
 
 Carbon 
 
 % 
 
 Mang. 
 
 % 
 
 Phos. 
 
 % 
 
 Sul. 
 
 % 
 
 Si. 
 
 % 
 
 
 Primary 
 
 .16 
 
 .35 
 
 .010 
 
 .036 
 
 
 AZ No. 1, (O. H. Heat 83023) 
 
 1st Preliminary 
 
 .80 
 
 .35 
 
 .006 
 
 .023 
 
 
 
 2d 
 
 .80 
 
 
 
 
 
 
 Final 
 
 .82 
 
 .35 
 
 .010 
 
 .020 
 
 .16 
 
 
 Primary 
 
 .20 
 
 .44 
 
 .012 
 
 .042 
 
 
 AZ No. 2, (O. H. Heat 88026) 
 
 Preliminary. . . 
 
 .61 
 
 .43 
 
 .010 
 
 .025 
 
 
 
 Final 
 
 .67 
 
 .60 
 
 .009 
 
 .024 
 
 
 
 
 Charging 
 
 .18 
 
 .41 
 
 .005 
 
 .024 
 
 .. 
 
 
 No. 1 Electrode 
 
 .20 
 
 .60 
 
 .004 
 
 .020 
 
 . . 
 
 AZ No. 3, (0. H. Heat 84598) 
 
 No. 2 
 
 .19 
 
 .60 
 
 .006 
 
 .020 
 
 
 
 No. 3 
 
 .21 
 
 .59 
 
 .005 
 
 .020 
 
 
 
 Final 
 
 .22 
 
 .58 
 
 .004 
 
 .021 
 
 
 
 Charging 
 
 .19 
 
 .34 
 
 .010 
 
 .032 
 
 .<; 
 
 
 No. 1 Electrode 
 
 .22 
 
 .53 
 
 .007 
 
 .026 
 
 . . 
 
 AZ No. 4, (O. H. Heat 101523) 
 
 No. 2 
 
 .22 
 
 .54 
 
 .009 
 
 .023 
 
 
 
 No. 3 
 
 .22 
 
 .54 
 
 .009 
 
 .023 
 
 
 
 Final 
 
 .20 
 
 .47 
 
 .006 
 
 .029 
 
 
 
 
 Charging 
 
 .28 
 
 .36 
 
 .009 
 
 .040 
 
 
 AZ No. 5, (O. H. Heat 101524) 
 
 No. 1 Electrode 
 
 .34 
 
 .54 
 
 .007 
 
 .017 
 
 
 
 No. 2 
 
 .32 
 
 .55 
 
 .006 
 
 .015 
 
 
 
 Final 
 
 .33 
 
 .52 
 
 .007 
 
 .020 
 
 
288 ELECTRIC PROCESS 
 
 Table 43. Partial Analysis of Final Electric Furnace Slags. 
 
 Heat No.'s 
 
 Silica 
 
 AZ No. 3 
 
 . . 19.58% 
 
 AZ No. 4 
 
 20.48% 
 
 AZ No. 5 
 
 18.40% 
 
 Iron 
 
 1.09 
 
 .72 
 
 .32 
 
 Total Lime 
 Magnesia 
 ]Vlanganese 
 
 . . 60.38 
 . . 10.32 
 .62 
 
 61.82 
 8.40 
 .60 
 
 61.26 
 7.27 
 .35 
 
 Sulphur 
 
 .85 
 
 .80 
 
 1.30 
 
 Calcium carbide . . 
 Alumina . . 
 
 .32 
 3.04 
 
 .19 
 3.61 
 
 .46 
 5.95 
 
 Difficult Specifications: While the electric furnace affords means of 
 making steel to very difficult specifications and with a greater degree of 
 accuracy than is possible in other processes, yet it has its limitations. 
 The case of sulphur furnishes an example. From what has just been said, 
 the importance of carbon in the' elimination of both oxygen and sulphur is 
 evident. In low carbon steels the lowering of the sulphur content to any 
 considerable degree becomes a difficult problem, because, if the flux be made 
 highly carbonaceous, a necessary condition, there is danger that the steel 
 will absorb carbon from the slag, and thus raise the carbon content of the 
 steel above the requirements. In the high carbon stee.s, the absorption of 
 carbon by the steel can be allowed for and presents no difficulty, so a more 
 highly carbonaceous flux may be used than with low carbons, and a greater 
 removal of sulphur results. As previously indicated, the elimination of 
 sulphur may be brought about by the use of silicon, as shown in the following 
 reactions : 
 
 But while it is claimed that little more than the theoretical amount of 
 silicon is required, in actual practice a residue of silicon in the steel is 
 unavoidable. Hence, this method could not be employed to produce silicon 
 free steel. Similar to the method of desulphurizing with carbon, these 
 reactions will take place only after the bath has been completely deoxidized. 
 For low carbon steels, then, the limit for sulphur should be .040%, while for 
 high carbons this limit could be reduced to .030%. To guarantee lower 
 limits than these would mean increased cost in production out of proportion 
 to the benefits to be derived, except in the case of steal that is to be used 
 for certain special purposes. Since the phosphorus is removed in the basic 
 open hearth, the same range in per cent, of this element as is customary 
 to allow in open hearth steel should be allowed in electric steel. As to 
 alloying elements, the variation in the composition of the alloys used 
 makes it desirable to secure as wide a range as possible in the specifications 
 for such elements. 
 
PROPERTIES OF ELECTRIC STEEL 289 
 
 SECTION X. 
 
 PROPERTIES AND USES OF ELECTRIC STEEL. 
 
 Properties of Electric Steel: We can deal with this topic in a no 
 more fitting way than to quote from impartial investigators. The following 
 is taken from a paper by Messrs. Lyon and Keeney of the Bureau of Mines: 
 "For many years all high grade steels were manufactured by the crucible 
 process, but since the advent of the electric furnace there has been a gradual 
 adoption of that furnace for refining steel. For the complete refining of 
 the highest grades of steel the use of the electric furnace is now thoroughly 
 established. Any products that can be made by the crucible process can 
 be made by the electric furnace, and in most cases with cheaper raw 
 materials and at a low cost. In the electric furnace complex alloy steels 
 can be made with precision. The high temperatures attainable facilitate 
 the reactions, and alloys need not be used so largely for the purpose of 
 removing gas. Very low carbon steels can be kept fluid at the high tem- 
 peratures. Steels free from impurities and of great value for electrical 
 apparatus can be made. With the electric furnace large castings can be 
 made from one furnace, whereas in the crucible process steel from several 
 crucibles must be used. For small castings, which require a very high 
 grade metal free from slags and oxides, electrically refined steel is especially 
 adapted. The electric furnace gives a metal of low or high carbon content 
 as desired, hot enough to pour into thin molds, and steel free from slags 
 and gases." 
 
 "There is now a tendency among customers of rail and structural steel 
 to require a higher grade steel at an increased price rather than steel of 
 acid Bessemer or even of basic open hearth grade at a lower price. With 
 the high cost of power that now prevails throughout the steel centers of 
 the United States the electric furnace cannot compete profitably with either 
 the acid Bessemer or the basic open hearth process in manufacturing steel of 
 like grade from pig iron. It is in combination with either of these processes 
 that the electric furnace seems destined to be prominent in steel manu- 
 facture." 
 
 "Experiments conducted by the United States Steel Corporation during 
 the past four years show that, as compared with the acid Bessemer and 
 basic open hearth processes, the electric process has the following advan- 
 tages: A more complete removal of oxygen; the absence of oxides caused 
 by the addition of silicon, manganese, etc; the production of steel ingots 
 of 8 tons weight and smaller that are practically free from segregation; 
 reduction of the sulphur content to 0.005 per cent., if desired; reduction 
 of the phosphorus content to .005 per cent, as in the basic open hearth 
 process, but with complete removal of oxygen.'* 
 
290 
 
 ELECTRIC PROCESS 
 
 "About 5,600 tons of standard electric rails from electric furnace steel 
 have been in service in the United States for the past two years (prior to 
 1914). These rails have been subjected to all sorts of weather and to 
 temperatures as low as 52 F. It seems that rails made by the basic 
 electric process can be made softer than by either the acid Bessemer or 
 basic open hearth processes and yet show highly satisfactory wearing 
 qualities." 
 
 "No steel rails made by the basic electric process in service in this 
 country have been broken. Electric furnace steel of a given tensile strength 
 has a slightly greater elongation than basic open hearth steel and is some- 
 what denser than basic open hearth or acid Bessemer steel.'! 
 
 The results of some comparative tests, made at South Chicago of 
 electric furnace steel for plates and basic open hearth steel for plates were 
 as follows: 
 
 Table 44. Comparison of Mechanical Properties of Electric and 
 Open Hearth Plate Steel. 
 
 ELECTRIC 
 
 OPEN HEARTH 
 
 Carbon 
 Content 
 Per Cent. 
 
 Ultimate 
 Strength 
 per Sq. In. 
 Pounds 
 
 Elongation 
 on 2 Inches 
 Per Cent. 
 
 Carbon 
 Content 
 Per Cent. 
 
 Ultimate 
 Strength 
 per Sq. In. 
 Pounds 
 
 Elongation 
 on 2 Inches 
 Per Cent. 
 
 0.08 
 
 59,194 
 
 27.25 
 
 0.08 
 
 51,690 
 
 32.00 
 
 .12 
 
 64,080 
 
 26.05 
 
 .12 
 
 56,510 
 
 29.70 
 
 .16 
 
 69,220 
 
 25.25 
 
 .16 
 
 52,901 
 
 28.61 
 
 .20 
 
 72,853 
 
 22.82 
 
 .20 
 
 58,294 
 
 28.82 
 
 .24 
 
 69,540 
 
 23.12 
 
 .24 
 
 63,560 
 
 26.25 
 
 The results show a 15.5 per cent, increase in ultimate strength and 
 11.3% decrease in elongation for electric steel, as compared with open hearth 
 plate steel of approximately the same chemical composition." 
 
 Illinois Steel Company's Tests on Rails : The Illinois Steel Company 
 have conducted a series of experiments from which it was shown that electric 
 steel is considerably more ductile at low temperatures than either the open 
 hearth or the Bessemer steel. In these tests about 900 pieces of electric, 
 open hearth, and Bessemer rails were tested at temperatures ranging from 
 70 F. to 50 F., and the results indicated that while all these steels 
 showed a marked decrease in resistance to shock as the temperature was 
 lowered, the electric steel was relatively more ductile than either of the 
 other two. The following table is a summary of the results obtained with 
 
USES OF ELECTRIC STEEL 
 
 291 
 
 two open hearth and two electric heats of similar composition chemically, 
 as shown by analysis, and may be taken as typical of the general results 
 obtained. 
 
 Table 45. Comparison of Tests on Open Hearth and Electric Steel 
 Railroad Rails at Different Temperatures. 
 
 Number of Tests, 242. 
 
 Temp 
 Deg'sF. 
 
 AVE. No. BLOWS TO 
 BREAK RAILS 
 
 DEFLECTION BEFORE 
 BREAKING 
 
 ELONGATION IN 12 IN. 
 
 Elec. 
 
 O.H. 
 
 Comparison 
 
 Elec. 
 
 O.H. 
 
 Comparison 
 
 Elec. 
 
 O.H. 
 
 Comparison 
 
 
 
 
 O.H. 
 
 Over 
 
 E. 
 
 E. Over 
 0. H. 
 
 Inches 
 
 Inches 
 
 O.H. 
 
 OverE. 
 
 E. Over 
 O.H. 
 
 Inches 
 
 Inches 
 
 O.H. 
 
 Over E. 
 
 E.Over 
 0. H. 
 
 -j-60 
 
 3.48 
 
 3.64 
 
 5% 
 
 
 2.96 
 
 3.36 
 
 14% 
 
 
 .808 
 
 .929 
 
 15% 
 
 
 
 
 
 4.41 
 
 3.82 
 
 
 15% 
 
 1.41 
 
 1.23 
 
 
 15% 
 
 .404 
 
 .397 
 
 
 2% 
 
 30 
 
 4.55 
 
 2.24 
 
 
 103% 
 
 1.46 
 
 .58 
 
 
 152% 
 
 .420 
 
 .201 
 
 
 109% 
 
 40 
 
 3.31 
 
 2.03 
 
 
 65% 
 
 .91 
 
 .43 
 
 
 112% 
 
 .297 
 
 .141 
 
 
 111% 
 
 Uses of Electric Steel: As to the uses of electric steel, little need be 
 said except it may be used with confidence wherever a steel of higher 
 quality than that furnished by the open hearth process is desirable. 
 So far, the demand for this steel has been in advance of the supply, and 
 its application is becoming more and more general. Among those with 
 whom it has found favor and by whom it is now being used, are manu- 
 facturers of automobiles, motorcycles, motor accessories, areoplanes, 
 machinery, engines, agricultural implements, tools, guns and munitions, 
 and by the railroads. The fact that it is being used more and more in 
 place of crucible steel is evidence of its superior quality. 
 
 Summary: In order to cover the whole subject of electric steel 
 making satisfactorily in so brief a manner, it has been necessary to treat 
 the subject from several different view points, which method is likely to 
 be confusing, with the result that the reader may have lost sight of some 
 important points. In order that these points may receive proper emphasis, 
 the following summary of the chapter is appended: 
 
 1. Of the many furnaces in use no one can be said to possess any great 
 advantage over any other from a metallurgical point of view, with the 
 exception that higher temperatures may be obtained with furnaces of the 
 arc type than with the induction type. 
 
 2. The only effect of the electric current is in the production of heat. 
 
292 SUMMARY 
 
 3. The electric process is the only one in which impurities are not 
 added to the steel by the operation. 
 
 4. The electro-thermal process affords the only positive means of 
 desulphurizing and deoxidizing steel simultaneously and in the same 
 operation. 
 
 5. It permits the addition of all alloying elements while the steel is 
 in the furnace. 
 
 6. It provides a way for remelting alloy steel scrap and producing a 
 product of high quality without loss. 
 
 7. Steel produced by this process exhibits some unusual wearing 
 qualities. 
 
 8. In quality, steel made by this process equals that of the best grades 
 of crucible steel. 
 
 9. Much larger quantities of metal may be treated in one operation 
 than is possible by the crucible process. 
 
 10. It gives a product that is uniform in quality for any given heat. 
 
 11. Steels refined in the electric furnace are freest from segregation. 
 
 12. Steels made in the electric furnace are free from slag and other 
 inclusions. 
 
 13. Electric steel is comparatively more ductile at low temperatures 
 than Bessemer or open hearth. 
 
 14. Considering the various methods from an economical point of 
 view, the duplexing process in which the electric furnace is used in con- 
 junction with the basic open hearth combines the greatest capacity and 
 efficiency with highest quality of product. 
 
DUPLEX PROCESS 293 
 
 CHAPTER X. 
 
 THE DUPLEX AND TRIPLEX PROCESSES. 
 
 SECTION I. 
 
 GENEBAL FEATURES OF THE DUPLEX PROCESS. 
 
 What the Duplex Process Is : The term duplex process may be applied 
 to a combination of any two processes for manufacturing steel, but it is 
 customary among the steel men of this country, at least, to restrict the 
 term to mean only a combination of the acid Bessemer and the basic open 
 hearth process, in which the latter plays the part of a finishing process. 
 Briefly described, the method, as usually carried out, consists, first, of 
 blowing molten basic iron in the converter until the silicon, manganese and 
 a part of the carbon have been oxidized and then transferring this semi- 
 finished metal to a basic open hearth furnace, where, through the agencies 
 of iron oxide and lime, the phosphorus and the remainder of the carbon 
 to be removed are oxidized. The steel is then finished, recarburized and 
 deoxidized, according to the usual open hearth practice. This combination 
 of processes may be made in other ways, also. One plant, for example, 
 in order to produce a very low phosphorus Bessemer steel for a certain 
 order, first oxidized the silicon, manganese, and phosphorus in the open 
 hearth, and then, by mixing this very high carbon steel with Bessemer 
 iron in suitable proportions, succeeded in blowing out the carbon in the 
 converter, thus reversing the customary procedure. But as stated in the 
 beginning, the duplex process refers to the combination in which the finishing 
 operation is conducted in and from a basic open hearth furnace. 
 
 Advantages and Disadvantages of the Process: In the northern 
 district of the United States the chief advantages of the process, when 
 there is a pressing demand for steel, is that of the increased tonnage which 
 it produces in a given time. Thus, while the product is similar in quality 
 and of the same grades as basic steel, the time of the open hearth operation 
 is shortened by about half; for, whereas one open hearth furnace will turn 
 out an average of about fifteen heats in a week of straight running by the 
 ordinary way, the same furnace operated as a duplexing unit will produce 
 about forty heats in the same period. This shortening of the time of a 
 heat saves fuel and tends to prolong the life of the furnace, as does, also, the 
 elimination of the silicon in the converter. The process does not require 
 the use of scrap, which fact may also be an advantage to some makers. 
 Offsetting these advantages, however, are the double conversion cost and 
 the decrease in yield, due to the increased oxidation, both of which may be 
 very serious drawbacks to the economical production of steej. In dull 
 
294 DUPLEX PROCESS 
 
 times, especially, the extra costs of maintaining two separate units may 
 more than counter- balance the gain from the increased output. 
 
 Methods of Duplexing: While the details of the process vary widely 
 in the different plants, there are two general methods of carrying out the 
 duplexing process: Thus, the purification in the converter may be carried 
 cut to the point where the metal is fully blown and represents a high phos- 
 phorus steel which may then be mixed with pig iron in the open hearth, 
 thus taking the place of steel scrap. In this method either a stationary 
 or a tilting furnace can be utilized. But the more common method is the 
 &UQ, already mentioned, in which the carbon is only partially eliminated 
 In the converter, and the purification then completed by the continuous 
 process, which is most conveniently carried out in a Talbot tilting furnace. 
 A brief description of these furnaces will simplify the description of the 
 process to be given shortly. 
 
 The Talbot Furnace: The object aimed at in the design of these 
 furnaces is to permit the removal of any quantity of slag or metal or the 
 addition of molten metal, oxidizing agents and flux at any time during the 
 working of the charge. They are, therefore, necessarily of the tilting type, 
 and are built upon racers and rollers which rest upon the foundation in a 
 manner similar, in a general way, to that of the large mixers of recent con- 
 struction. They are rectangular in shape, and of about the same proportions 
 as an ordinary open hearth furnace as to length and width,but they have a much 
 greater depth, which increases their capacity for containing molten metal. 
 The frame work must be of much stronger construction than that for the 
 ordinary open hearth in order to avoid twisting stresses and vibrations 
 which would be very harmful to the brick work. Only that section of the 
 furnace comprising the hearth, side-walls and roof is made tilting; all the 
 ports and flues are stationary, and, together with the checker work, of the 
 same construction as for the stationary furnaces. In the best types of con- 
 struction, these furnaces are so placed and the racers and rollers so formed that 
 the center of rotation of the furnace coincides with the center line of the ports, 
 so that all its parts always remain in the same relation no matter in what 
 direction or to what degree the movable portion of the furnace may be 
 tilted. By means of water cooled metal joints, the clearance between the 
 movable and stationary parts of the ports is kept very small, so that the 
 heating of the furnace may continue even during the tapping of a heat. 
 On the pouring side, these furnaces usually have but one opening, a tapping 
 hole located above the slag line and provided with a lip or spout for directing 
 the stream of molten metal into the steel ladle. As in the case of the 
 stationary furnace, doors for introducing the materials into the furnace 
 are located in the front side. But unlike the stationary types, the slag 
 notches are also placed in front, usually one on each side of the middle 
 door, and, of course, at a lower level. Since Talbot's method is but a 
 modification of the basic open hearth process, the furnaces are, as a matter 
 of course, provided with basic linings. 
 
METHOD OF OPERATION 295 
 
 SECTION II. 
 
 OPERATION OF THE PROCESS. 
 
 An Example of the Duplexing Process: Perhaps the best way to 
 describe the duplexing process is through an example. For this purpose 
 the method employed by a large steel manufacturing company in the North, 
 whose plant represents one of the most recent installations, is selected. 
 Their duplexing plant consists of three 20-ton converters and three Talbot 
 tilting furnaces, each of which has a capacity of approximately 200 tons. 
 While the practice may be varied somewhat to suit conditions, the process 
 at this plant is usually carried out about as follows: 
 
 Preparing the Furnace for Charging: The process may be said to 
 be continuous for a week, for each week-end the tilting furnace is thoroughly 
 drained, the bottom and slag lines are made up, the ports are cleaned and 
 repaired, and everything is made ready for the week's campaign. Of course, 
 during the interval of this campaign the front and back wall must be 
 attended to and such minor repairs made as are found necessary and there 
 is time for. About 6 p. m. Sunday, after the flues have been burned out 
 and the gas is once more on the furnace, the work of preparing the slag is 
 begun. Four boxes of calcined limestone and three of roll scale are charged 
 and melted down. These amounts are then repeated, and when again molten 
 the same amounts are again charged, the total being twelve boxes of lime and 
 eight to nine boxes of roll scale. The average weight of a box of the 
 lime is 2000 Ibs. and of a box of roll scale 3000 Ibs. Considerable care 
 is given by the melter to the preparation of a good slag, for, as in all open 
 hearth work, the success of the process depends on the slag. 
 
 Charging Molten Metal from the Converters for the First Heat: 
 
 At midnight, or shortly afterwards, the metal is ordered from the Bessemer 
 department. An average analysis of the mixer metal is as follows: 
 
 Total Carbon 3.85 per cent. 
 
 Silicon 1.55 " 
 
 Manganese 67 " 
 
 Sulphur 040 " 
 
 Phosphorus 365 " 
 
 The weight of this metal taken for a Bessemer heat is about 40,000 Ibs., less 
 the weight of the scrap in the converter. Two Bessemer heats, blown in 
 different vessels, are poured into a transfer ladle and taken to the tilting 
 furnace. When starting up, the first ladle contains metal blown down to 
 contain 0.60 per cent, carbon, which is allowed to remain to give a little 
 action, or boil, in the furnace. This first ladle is poured into the open hearth 
 furnace about midnight. It is followed by a ladle of "soft" metal, 
 that is, metal blown down to 0.05% or 0.07% carbon, and then by a 
 
296 DUPLEX PROCESS 
 
 "kicker," or a ladle of high carbon steel. This metal is blown down to 
 from 1.5% to 2.0% carbon, and when charged into the open hearth produces 
 a vigorous reaction, or boil. The metal and slag are thoroughly mixed 
 together by this boil, and during this reaction the phosphorus is largely 
 removed from the metal bath and passes into the slag. When the action 
 has subsided, another "soft" ladle and a "kicker" are charged. Then, 
 if the bath is found to be low in carbon, another kicker ladle is added to it, 
 but if high in carbon another "soft" ladle is charged. In this way a bath 
 of metal of about 200 tons is produced. The charge is then worked down 
 like an ordinaty basic open hearth heat until ready for tapping, which is 
 usually at about 3:30 a. m. 
 
 Tapping and Recarburizing the First Heat : When the bath is ready 
 for tapping, the tap hole is opened and plugged with wet sacking. The 
 furnace is then tilted for pouring. Before the sacking is burnt through, 
 the slag is up along the back wall so that clean metal free from slag comes 
 from the furnace. Only enough slag is drawn off at the end to cover the 
 steel in the ladle properly. Some of the steel made in the Talbot furnaces 
 is super refined by the electric process, but by far the greater portion is made 
 into the ordinary commercial grades which is recarbonized and deoxidized 
 in the ladle as for similar grades made in stationary furnaces. 
 
 Preparing the Furnace for the Second Heat: After the first heat 
 is tapped, there is a bath of about 100 tons of metal with a carbon content 
 of about .15% still in the furnace, covered with the tapping slag. Two 
 boxes of lime and two boxes of scale are charged, and two boxes of burnt 
 dolomite are used along the slag line, around the doors, etc., as found 
 necessary. Then two "soft" ladles of blown metal are charged, and two 
 more boxes of lime, which is followed by a "kicker." During the reaction, 
 the furnace is tilted slightly forward and slag is allowed to flow from the 
 front of the furnace through the slag spouts, which are under the doors 
 directly on each side of the center door. The slag falls into slag cars stand- 
 ing on the tracks below. Practically all the slag taken from the furnace 
 is removed in this way, for, as mentioned before, when tapping a heat only 
 enough is taken to cover the metal in the ladle properly. When the reaction 
 is over, another box of lime is generally charged, and the bath is worked 
 down in the usual way. Very often, another box of lime is spread over 
 the slag shortly before tapping, so that five to six boxes of lime are used 
 per heat, but as a rule only two boxes of scale are used he*re. After the heat 
 is tapped, this procedure is repeated, enough slag being taken from the 
 front of the furnace at the time of the reaction to maintain a constant and 
 proper volume of slag in the furnace. The average time for tapping one 
 heat to tapping the next is about three hours. 
 
TRIPLEX PROCESS 297 
 
 Closing Down the Furnace for the Week End: About midnight on 
 Saturday the furnace is drained. The bath is worked down, so that after 
 the heat is tapped there are not more than forty to sixty tons in the 
 furnace. Then this residue of metal is tapped and made into soft steel, 
 for which there is a constant demand, by making the proper additions of 
 ferro-manganese and recarburizer. 
 
 The Slag: At the high temperature at which its removal is effected, 
 phosphorus is easily reduced, so in order to oxidize, flux and hold the phos- 
 phorus in the open hearth slag, it is necessary that the latter be very basic 
 and highly oxidizing, as an analysis shows. The average composition of 
 the slag is about as follows: Silica, SiO 2 , 6.35%; ferrous oxide, FeO, 21.65%; 
 ferric oxide, Fe 2 O 3 , 6.90%; manganese, Mn, 1.12%; phosphorus, P, 3.25%; 
 alumina, A1 2 O 3 , .97%; lime, CaO, 44.07%; magnesia, MgO, 8.04%. The 
 high percentage of iron oxides, which are equivalent to approximately 24.0% 
 metallic iron, gives the impression that the process is wasteful of iron, which 
 is true, but due to another cause. While the percentage of iron oxide is 
 high, it does not exceed that of the run off slags of the open hearth process, 
 and the total volume of slag is much less than in the straight open hearth 
 process, so that the loss of iron here is perhaps less than in the latter process. 
 The chief loss is at the converters, and there can be no doubt but that 
 the double conversion loss exceeds the single loss in the straight open hearth 
 process. This matter assumes its chief importance as it relates to the 
 conservation of the iron ore supply. 
 
 SECTION III. 
 
 COMBINATION PROCESSES IN THE SOUTH. 
 
 The Duplex Process in the South: In the southern district the 
 conditions of steel manufacturing are very different from those in the 
 North, and many additional reasons for the use of the duplex process there 
 are to be found. First, in the South there is no pig iron that is suitable 
 for the Bessemer process manufactured there, whereas, in the North, 
 Bessemer iron is relatively abundant. Second, there is no low phosphorus 
 iron or spiegel commercially available for recarburizing in the southern 
 district as there is in the North, and this lack makes it necessary in 
 manufacturing high carbon steels in the South to catch the carbon on the 
 way down. Third, the phosphorus content of the basic iron in the south- 
 ern district, which averages about .80%, is very high as compared with 
 the phosphorus content of basic iron in the North, the average for which 
 is about .25%. In the manufacture of high carbon steel from high phos- 
 phorus pig iron the duplex process offers exceptional advantages for catch- 
 ing the carbon high, thus reducing the amount of coal or coke-dust 
 required to a minimum and avoiding rephosphorization from the high 
 phosphorus slag. Another advantage of the process, when iron with a 
 
298 TRIPLEX PROCESS 
 
 high phosphorus content is used, is that it permits the making of a slag 
 which contains a high percentage of phosphoric acid and is therefore suit- 
 able for use in the manufacture of fertilizers. This slag is a valuable by- 
 product from one of the southern plants. 
 
 The Southern Triplexing Process: In operating the duplex process 
 in the South, it has been found that, owing to the high phosphoric acid 
 content of the slag, it is difficult to prevent the reduction of some of the 
 phosphorus after recarburizing. This rephosphorization of the steel occurs 
 mainly in the ladle, particularly in the portion of the metal in direct contact 
 with the mass of floating slag, and is most noticeable in the last two or 
 three ingots from each ladle of steel teemed. In order to overcome this 
 defect and at the same time increase the production of basic slag for phos- 
 phate fertilizer, one plant has developed a triplex process in which two 
 basic open hearth units are required to finish the metal after blowing in 
 the converter. Briefly, the process is as follows: After blowing, the metal 
 is transferred from the converters to primary basic tilting furnaces where 
 it is treated with lime and the other necessary oxides for dephosphorizing 
 it. Here the phosphorus content in the metal is reduced to about .07%, 
 when it is poured into ladles and transferred by specially constructed, heavy, 
 extra-wide-gage trucks to a finishing unit composed of an equal number of 
 similar furnaces. In these furnaces the phosphorus content of the metal 
 is brought below .04%, when the steel is finished in the ladle by the addition 
 of the necessary recarburizer and deoxidizers, and any alloys required by 
 the specification. It is said that this process does not reduce the capacity 
 of the plant and materially improves the uniformity and quality of the 
 steel produced. 
 
TESTING OF STEEL 299 
 
 PART II. 
 
 THE SHAPING OF STEEL. 
 CHAPTER I. 
 
 THE MECHANICAL PROPERTIES OF STEEL. 
 
 SECTION I. 
 
 SOME GENERAL REMARKS PERTAINING TO THE TESTING OF STEEL. 
 
 The Factors that Affect the Mechanical Properties of Steel : There 
 are four factors that may affect the quality and the mechanical properties 
 of steel; namely, the method of manufacture, or refinement, the chemical 
 composition, the mechanical working it is subjected to, and the heat treat- 
 ment it receives. The first of these factors is discussed in the first part 
 of this book. The others are now to be taken up, and frequent use of the 
 terms employed in the mechanical testing of steel will be made in the pages 
 to follow. Hence, although in the natural order of manufacturing steel 
 physical tests follow the shaping, it seems well to take up this subject now 
 in order that the reader may be more familiar with the terms employed 
 in connection with these tests, when there is occasion to refer to them. 
 
 The Two Objects in the Testing of Steel: In the early days it was 
 the custom to order steel according to use, that is, the purchaser asked 
 the steel maker for a certain quantity of steel suitable for a certain purpose, 
 and the manufacturer then furnished steel of a kind or grade he considered 
 most suitable for the purpose. With the growth of the steel business, this 
 custom proved inadequate to the conditions and was superseded by the 
 practice of ordering steel to specifications, which appeared to be, and is, 
 a much more satisfactory arrangement for all parties concerned. The 
 consumer naturally decided that he should know better than anyone else 
 what the requirements were, and the manufacturer, in turn, was very glad 
 to be relieved of the responsibility he assumed under the old system. Now, 
 the only basis upon which the consumer of steel or his engineers originally 
 had to work in determining specifications was experience. Thus, providing 
 that a certain steel had proved satisfactory for a certain purpose, he desired 
 for the new work steel as nearly like the old as possible. With this progress 
 came the need for testing. Then as undertakings involving the use of steel 
 increased in magnitude, it was discovered that steels made by the same 
 methods are subject to considerable variation. Furthermore, in order to 
 
300 TESTING OF STEEL 
 
 obtain the requisite amount of steel, it is often necessary to use, for the 
 same purpose, steels made in different ways. And again the need for 
 testing was felt in order to secure uniformity in the materials. This 
 testing developed along two lines, namely, physical and chemical. 
 
 Relative Importance of Physical and Chemical Testing: It is 
 
 evident that, to the consumer of steel, its mechanical properties are of first 
 importance, because it is these properties that determine whether or not 
 a particular steel is suitable for the purpose he intends it. In all cases, 
 then, such as structural steels, in which the material is put in service as 
 received from the manufacturer, the customer does well to order his steel 
 to physical specifications only. In cases where the steel is to be heat 
 treated or is to undergo other treatment in the hands of the customer, 
 then it should be ordered to a chemical specification only. Since the method 
 of manufacture influences the properties of the metal, the kind of steel, 
 whether Bessemer, basic, acid, or electric, should be and is, usually, specified. 
 But for a great many reasons, for a discussion of which time and space are 
 not available, it is unfair to ask the manufacturer to make steel to order 
 in which all three factors are specified. Suffice it to say, that in the one 
 case the customer should be satisfied to get the kind of steel ordered with 
 the required physical properties, irrespective of the means, chemical or 
 otherwise, which the manufacturer may have found it necessary to employ 
 in order to supply metal with the properties called for. In the other case, 
 the purchaser is interested only in obtaining steel properly made and of 
 the proper kind and composition, because with such steel the original 
 physical properties will be replaced by new ones due to the subsequent 
 working or treatments. From the view point of the consumer, then, the 
 relative importance of the physical and the chemical test depends upon the 
 conditions that surround each individual case. But to the manufacturer, 
 chemical testing is of prime importance, because it offers a means of control 
 whereby he is able to produce the steel with a greater degree of certainty. 
 For a description of the methods employed in chemical testing the published 
 standard methods of the Steel Corporation are available. 
 
 Nature of Physical Testing: It should at all times be borne in mind 
 that the results obtained by any method of physical testing are not absolute, 
 but relative. Obviously, the only sure test is actual service, and it 
 is just as evident that such tests are impracticable. Therefore, the test 
 must be carried out with a small piece of material, the structure and con- 
 dition of which are likely to be different from that of the section, taken as 
 a whole, from which it was cut. A second objectionable feature is found 
 in the difficulty of subjecting this piece to the same conditions that it 
 would be subjected to in actual service. Attempts have been made to 
 analyze these conditions with the idea of classifying the forces steel is 
 required to overcome in service, so that in testing it might be subjected 
 to the same kinds of forces. With respect to the effect they tend to produce, 
 
PULLING TEST 301 
 
 forces have been classified as (1) tensional, or forces tending to put the 
 material under tension, that is, pull it asunder; (2) compressional, or forces 
 that tend to compress the piece in one or more directions; (3) torsional,or 
 forces tending to twist the material; and (4) shearing, or forces that tend 
 to cut the material across its section. With respect to the manner in 
 which the forces are applied, the following classification has bean made ; 
 1. Static stresses, which are the result of the gradual application of a 
 steady or constant load. 2. Fatigue stresses, such as result from the 
 repeated application of a load or loads. 3. Impact stresses, in which the 
 metal is subjected to a sudden blow. 4. Dynamic stresses, which are 
 repeated impact, or vibratory stresses. 
 
 From these are selected the class of force or stress the steel is likely 
 to be required to withstand in service, and the tests will be arranged accord- 
 ingly. Added to these are a number of miscellaneous tests, such as hardness 
 tests and tests to determine relative resistance to wear or penetration. 
 
 SECTION II. 
 
 THE TESTING OF STRUCTURAL AND OTHER SOFT STEELS. 
 
 The Pulling Test: A test that is most commonly applied to steel, 
 and one that is always used, and almost to the exclusion of all others, for 
 testing structural steels, is that commonly spoken of as the pulling test. 
 As the name implies, the chief aim in this test is the determination of the 
 tensile strength of the steel, but incidental to the carrying out of the test 
 much additional information as to other mechanical properties of the 
 sample of steel is obtained. The technical terms employed in testing to 
 indicate these properties are tensile strength, elastic limit, elongation, 
 reduction of area, and modulus of elasticity. The exact meaning of these 
 terms is best explained in connection with a description of the method of 
 making the test. 
 
 Procuring the Test Pieces: Except in one or two cases where it is 
 desirable to modify the usual procedure, the test piece, or sample, is sheared 
 from scrap ends cut from the material as it comes from the rolls. This 
 piece is about eighteen inches in length and two inches in width, and, 
 except in the case of sheared plates from which both longitudinal and 
 transverse pieces are sometimes taken, its long axis is parallel to the 
 direction of rolling. As a rule, the test piece is taken from a position not 
 too close to the rolled edge, but in the case of bars of small sectional area the 
 entire section of the proper length may be taken. The piece is then stamped 
 near the ends with the heat number and any other data necessary to 
 identify it. 
 
302 TESTING OF STEEL 
 
 Preparation of the Test Piece: The working or shearing of the test 
 piece puts it in a state of strain and produces a great number of incipient 
 cracks on the edges, so that if it were pulled in this condition it would 
 fail too easily, and the results of the pulling would not indicate the real value 
 of the properties determined. To eliminate these cracks, the edges of the 
 piece are milled off as shown in the accompanying sketch, and the milled 
 
 S 
 
 k -3"--*! L o,, _ _* V 3/; -- 
 
 18" --- --> 
 
 Thickness as Rolled Scale: J"=l" 
 
 Fig. 45. Diagram Showing the Usual Form of the Test Piece Used in Pulling 
 Structural Steels. Occasionally the edges are milled parallel for the full length 
 of the specimen. 
 
 edges are filed smooth. Then if the piece is a heavy, thick one or is an 
 alloy steel, such as nickel steel, it is allowed to rest for a period, the length 
 of which will depend upon the conditions and the kind of steel. This resting 
 is necessary to allow the steel to relieve itself of the condition of strain 
 which the working has set up in it. This condition seriously affects the 
 ductility of the steel, as is shown by the fact that some test pieces, par- 
 ticularly of alloy steels, show a marked improvement in elongation after 
 resting, but with little, if any, change in the tensile strength. In the case 
 of soft steels of moderate thickness, the resting period is not so important, 
 as the working does not produce severe strains, and such steels recover 
 quickly from strains. After the test piece has been machined and filed 
 and otherwise made ready for testing, its dimensions are taken. From a 
 point estimated to be the middle of the machined portion of the piece, 
 two spaces of two inches each are laid off, with a double pointed punch, 
 longitudinally along the bar and in both directions from the center punch 
 mark, thus making a distance of eight inches between the two punch marks 
 that are the farther from the center one. This space fixes the length of bar 
 that is later to be the basis for calculating the percentage of elongation. 
 Finally, with screw micrometers the width and thickness of the test piece 
 are taken and recorded. A careful operator will measure these dimensions 
 in three or four places to determine to what extent they are uniform. The 
 test piece is then ready to be pulled. 
 
PULLING TEST 303 
 
 Pulling the Test: The testing machine may be likened to a beam 
 scale or weighing machine, and if, instead of lying upon the platform of the 
 scale, the test piece be thought of as being attached by one of its ends 
 under the platform with its other end free for the attachment of some 
 device for exerting a vertical pull, the analogy is almost exact. In modern 
 machines the pulling device is either an electric motor, attached through 
 gears and screws to a gripping device that clamps one end of the test piece, 
 or an hydraulic ram connected directly with the clamping device. With 
 such an arrangement the amount of the pull is registered in pounds on the 
 graduated beam of the machine. This beam is provided with a travelling 
 weight, which, by means of a screw and worm drive actuated by hand, 
 may be rolled out along the beam as the pull increases, thus keeping the 
 beam in the neutral position and registering the amount of the tension at 
 all stages of the pull. When the test has been properly marked and 
 measured, it is placed in the machine in a vertical position and securely 
 clamped in the gripping boxes, or shackles. Then, the pulling device is 
 started at low speed, and the weight is cautiously moved out along the beam, 
 the operator keeping it in a perfectly horizontal position with the free end 
 midway between the upper and lower beam-end stops. The continued 
 action of the pulling machine is an indication that the test piece is stretching 
 slightly. That such is actually the case could be proved by stopping the 
 machine and measuring with delicate instruments the distance between 
 the two extreme space marks and comparing this measurement with the 
 original length. As the pulling is continued uniformly, the beam weight 
 is advanced at a uniform rate, indicating that the piece is obeying the law 
 of stress and strain; but a point is soon reached where the beam suddenly 
 drops, indicating that, without any increase in the load, there has been a 
 sudden increase in the length of the test piece. In testing parlance this 
 point is called the elastic limit, the yield point, or the point of permanent 
 set. To be accurate, the point reached immediately before this sudden 
 stretch, or give, occurs marks the true elastic limit, while the drop of 
 the beam marks the yield point. A reading of the weight indicated by 
 this position of the beam weight is therefore taken, and the pulling is con- 
 tinued as before, with the exception that the speed of the pull may be 
 increased somewhat. As the beam rises again, it is necessary to advance 
 the beam weight much more rapidly than before, which fact indicates a 
 more rapid stretching of the test. In a short time another point is reached 
 where the beam suddenly drops for the second time, but here, though the 
 pulling is continued, the beam will not rise again. A second reading is 
 therefore made, and the weight recorded is taken as a measurement of the 
 tensile strength of the test piece. Finally, as the machine continues to 
 elongate the specimen, the point of rupture is reached, and the piece breaks 
 apart. In practice no reading is taken at the breaking point, but if it were, 
 it would be necessary to reverse the direction of motion of the beam weight, 
 
304 
 
 TESTING OF STEEL 
 
 because the breaking load is generally less than the tensile strength. 
 The latter, therefore, is usually referred to as the ultimate strength, 
 maximum stress, or maximum load. 
 
 Graphic Representation of Tests : The pulling of a test is admirably 
 illustrated by means of a graph, which is also a great aid in understanding 
 the relations of the various terms employed in designating the points 
 described above. The following graph, while not absolutely accurate and 
 to scale in some of its parts, will serve to illustrate the scheme for preparing 
 graphs and to make clearer the description of the pulling of the test piece. 
 The diagram requires no explanation. 
 
 75000 
 
 70000 - 
 
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 
 Elongation in tenths of an inch 
 
 FIG. 46. Graph Representing the Pulling of a Structural Steel Test Piece. 
 
 Reasons for the Points of Yield and Maximum Stress: No very 
 satisfactory reason for the occurrence of the yield point has yet been 
 
PULLING TEST 305 
 
 advanced. Some think that it is due to some rearrangement of the mole- 
 cules. Again, grain structure may be the cause. Since steel is made up 
 of small grains or crystals, it appears reasonable to suppose that they have 
 at the time of their formation assumed a form and an arrangement that is 
 most natural, and that they will offer resistance to any force tending to 
 change this form or arrangement. This resistance is made up of two forces 
 of attraction, namely, one that tends to keep the grains in contact and 
 another that tends to preserve the arrangement of the molecules within 
 the grain. At the elastic limit this resistance is just balanced by the 
 tension, but under any greater tension, deformation of the grains begins, 
 and the structure "gives" suddenly, becoming at the same time longer and 
 smaller in cross section. Up to the elastic limit the slight stretch may 
 be due to a partial rearrangement of the grains. When the tension is 
 removed the natural arrangement is restored, with the result that the 
 piece immediately assumes its initial form and size. When subjected to 
 tension under this limit the body remains in an elastic condition, and the 
 deformation it undergoes is called elastic deformation. Above the yield 
 point the grains are undergoing deformation, that is, they are in a way 
 destroyed, and the piece of metal reacts more like a plastic than an elastic 
 body. Therefore the body is said to be undergoing plastic deformation. 
 This change in grain form is continuous, and requires an ever-increasing 
 force, or stress, to make it so. The condition is strictly analogous to cold 
 working, which will be discussed later. Consequently, the piece becomes 
 stronger, but at the same time it is becoming longer and correspondingly 
 smaller in cross section. Ultimately, as the necking of the piece becomes 
 pronounced, the loss in strength due to the decreased area of the section, 
 plus the external stress applied, balances and then exceeds the maximum 
 stress that the cold working can develop. From this point, then, the 
 external force necessary to balance the forces of attraction between the 
 grains becomes less and less as the area continues to decrease rapidly. 
 Finally, the maximum deformation of the grains is reached, when, if the 
 tension applied externally exceeds the forces of attraction tending to keep 
 the grains together at the point of least cross sectional area, the piece is 
 fractured at that point. 
 
 Examination of Test After Pulling: After fracture the two parts of 
 the test piece are removed from the machine, and the fractured ends are 
 fitted together as neatly as possible for the measurements to follow. The 
 distance between the extreme punch marks is now measured. The difference 
 between this distance and the original space of eight inches gives the elon- 
 gation for the piece, which is properly recorded. An examination of the 
 piece shows that while it has been reduced in section throughout its length, 
 the reduction is most pronounced in the region of the fracture, where the 
 piece underwent the characteristic deformation known as necking before it 
 broke. It is here, as near to the fractured ends as possible, that the width 
 and thickness are again measured in order to ascertain the reduction in 
 
306 TESTING OF STEEL 
 
 area. Finally, the fractures are designated as angular, cup-shaped, half 
 cup, and irregular. Very little importance can be attached to the form 
 of the fracture, but some inspectors believe that the cup shaped fracture 
 indicates more nearly perfect uniformity in the material than the other 
 forms. 
 
 Calculating the Results of the Test: The results obtained in the 
 test are for the given piece only, and, in order that the results from different 
 tests may be comparable, they must be calculated to a common basis. 
 Tensile strength and elastic limit are always expressed in pounds per square 
 inch in the United States, in tons per square inch in England, and kilograms 
 per square millemeter in France and other countries using the metric 
 system. The elongation is expressed as the percentage of increase on the 
 original length of the bar. In the United States this length for structural 
 and other low carbon steels is usually eight inches, as formerly stated, but 
 other lengths, as ten and twelve inches, ten centimeter, etc., may be used. 
 For this reason it is always important that the original length of the bar 
 be stated, as the percentage of reduction on two inches, for example, would 
 be much greater than that based on eight inches because of the pronounced 
 local contraction, or necking, at the point of fracture. This variation in 
 the length of test pieces is made, because the relation between the length 
 and the thickness of the test affects the elongation. Hence, in order that 
 tests of different thicknesses may be comparable, the ratio between the 
 thickness and length is kept constant by varying the length. The ideal 
 thickness for a length of eight inches is about three-fourths inch. Over or 
 under this thickness, the specification is usually modified either by chang- 
 ing the length or by making the proper allowance from the elongation as 
 determined. The reduction of area is expressed in percentage contraction 
 of area of the cross section as compared with the original area of the cross 
 section. An example will serve to clear up any doubtful points that may 
 not have been made clear in this explanation. 
 
 Table 46. Data on the Pulling Test Represented by the Graph 
 
 of Fig. 46. 
 
 Dimensions of Piece 
 Before Pulling After Pulling 
 
 Length i V ,,.,. 8.0 inches 10.20 inches 
 
 Width 1.41 " 1.00 " 
 
 Thickness : 860 " .60 " 
 
 Area 1.213 sq. in. .60 sq. in. 
 
 Readings on pulling bar: Elastic limit=44600 Ibs. ; Ultimate strength= 
 74600 Ibs. 
 
PROPERTIES DETERMINED 
 
 307 
 
 Calculations: 
 
 Elastic Limit 
 
 Ultimate Strength 
 Elongation 
 
 44,600 
 1.213 
 
 74,600 
 1.213 
 
 10.208 
 
 = 36770-lbs. per sq. in. 
 = 61500- " " " 
 = 27.5% in 8 inches 
 
 Reduction of Area = 
 
 91 Q r ^ An 
 - '- 
 1.213 
 
 = 50.5% 
 
 The Modulus of Elasticity, or Young's Modulus: The Modulus is 
 seldom determined in practical work, as it involves the determination of 
 the absolute increase in length of the test bar up to the elastic limit, which 
 is a quantity so small as to be very difficult to determine accurately. It 
 may be defined as that force, expressed in pounds, tons, or kilograms per 
 unit of cross section area, that would stretch a test piece to twice its original 
 length, when applied to one end of it and acting in the direction of its length. 
 
 It may be found by applying the following formula: M= where M= 
 
 Ax 1 
 
 modulus of elasticity, f=force applied, L=original length, A=original 
 sectional area, l=increase in length. For low carbon steel Young's Modulus 
 is about 29,000,000 Ibs. 
 
 Relative Importance of the Mechanical Properties as Determined 
 by the Pulling Test: From what has been said it is plain that the 
 quantities that may be taken as indicative of the strength of the material 
 are tensile strength and elastic limit; but while most engineers will insist 
 on both tensile strength and elastic limit being given and a few are content 
 to get the tensile strength only, it is evident that, in its practical application, 
 the elastic limit will determine the working strength of the steel, that is, 
 the maximum load a given piece may carry with safety. In the elastic 
 range the body stretches and recovers with increase and decrease in the 
 load; in the plastic range it cannot recover but continues to elongate 
 as long as tensional stresses are applied, and in this condition it is a very 
 unsafe material to use. The percentage reduction of area and the percentage 
 of elongation are, when considered together, an index of the ductility of 
 the metal, for both are required to give an idea of the amount of the 
 deformation before rupture. However, engineers are divided in their 
 
308 TESTING OF STEEL 
 
 opinions as to the relative importance of the two factors. In a general 
 way, it may be stated that the reduction of area is regarded as 
 more reliable than elongation, because, as previously explained, the quantity 
 denoting the latter is affected by the ratio of the length of the test piece 
 to its cross sectional area. In many foreign countries this ratio is always 
 specified. 
 
 Bending Tests: On certain classes of material, bending tests are made 
 in addition to the pulling test. These tests, simple in character, consist 
 merely in bending test pieces similar to those used for pulling, and similarly 
 prepared, through certain specified arcs. The bending is usually on cold 
 material, but some orders call for hot bending tests, also. Such tests are 
 employed to make sure that the steel is not cold short or hot short and, 
 in a way, to indicate the ductility of the metal. 
 
 SECTION III. 
 
 THE TESTING OF THE HIGHER CARBON AND HEAT TREATED STEELS. 
 
 Kinds of Tests Applied to the Higher Carbon and Heat=treated 
 Steels : For testing the class of material referred to under this heading, 
 a large number of different tests have been devised. These tests may be 
 classified under the headings of tensile tests, compressive tests, torsional 
 tests, impact tests, and hardness tests. Of these, the tensile, impact, and 
 hardness tests are the ones most frequently met with, and will, therefore, 
 be described later. Of the others the torsional test is perhaps the most 
 important, as it is largely used in the testing of steel for automobiles. 
 It consists in twisting a small round specimen of steel held in a suitable 
 machine until rupture occurs. In it a test piece of standard size is used, 
 and values for this piece corresponding to the elastic limit and ultimate 
 strength, but expressed in inch-pounds, are obtained very much as in the 
 tension test; the amount of distortion, however, is given in degrees. The 
 compression test is carried out by means of a machine similar in con- 
 struction to the pulling machine. The test piece may be in the form of 
 a small cylinder or a one inch cube. The elastic limit under compression 
 is determined, and the distortion is indicated by the decrease in length. 
 
 The Tensile Test: The test for determining the tensile strength of 
 the higher carbon and heat treated steels is carried out in a manner similar 
 to that for the softer steels, but since the material is so much stronger and 
 the items made from such steels do not lend themselves to the same method 
 of sampling, the test piece is much smaller than that employed in the case 
 of structural steels. This specimen is in the form of a small round, as 
 
HIGH TENSILE STEELS 
 
 309 
 
 shown in the accompanying figure, and is often obtained by boring with 
 a hollow drill about midway between the center and outside surface of 
 the section sampled. 
 
 2i" 
 4" 
 
 FIG. 47. Drawing Showing Usual Size and Form of Test Piece Used in Pulling High 
 Tensile Steels. The ends may be of any form desired but the central machined 
 portion must be as shown in the figure. 
 
 I mpact Test : While several different types of machines for measuring 
 the resistance of steel to impact have been invented, the results obtained 
 with any of these machines so far have not been considered very reliable, 
 as widely varying results may be obtained on the same steel tested on 
 the same machine. In practice, therefore, the nearest approach to an 
 impact test is what is commonly and correctly called the drop test. It 
 is applied to full size pieces of rails, to axles, and to other sections. It 
 consists in allowing a specified weight to drop from a specified height a 
 specified number of times upon the sample, which is supported at two points 
 on a heavy anvil or block resting upon strong springs. All three of these 
 factors may vary greatly with different classes of material and with the 
 different ideas of the engineers. While it does not measure absolutely 
 any property of the metal and is to be considered as comparative or 
 qualitative only, it is, nevertheless, one of the most useful of practical tests, 
 for it determines, in a crude way, the ductility and homogeneity of the metal 
 and its resistance to shock. In the case of axles, and other round bodies, 
 the deflection from a given weight may be kept constant for different sizes 
 by varying the height, for since the strength of such a section varies as 
 the cube of the diameter, for equal deflections, the height varies as the 
 cube of the diameter of the specimen at its center. 
 
 Hardness Tests: The best known and the most widely used instru- 
 ments for measuring the hardness of metals are the Shore scleroscope and 
 the Brinell ball testing machine. The Shore instrument consists of a small 
 diamond-faced tup enclosed in a glass tube which is provided with a suction 
 bulb, whereby the tup may be raised to the top of the tube and dropped from 
 a definite and fixed height. To make a determination, the instrument is 
 
310 TESTING OF STEEL 
 
 held in the vertical position with the lower end resting upon a smooth 
 and highly polished spot on the surface of the metal to be tested, when 
 the tup is allowed to drop by compressing the bulb. The height of the 
 rebound, which may be read on a scale inscribed on the tube, is taken as 
 a measurement of the hardness. Notwithstanding the fact that the results 
 obtained by this instrument are sometimes very erratic, especially if the 
 surface of the different spots tested have not been properly and uniformly 
 polished, it is a valuable instrument for comparing the surface hardness 
 of different parts of a body that is too large to be tested in any other 
 way. It also possesses the advantage that the tests may be made upon 
 the finished article without injury to the article itself. 
 
 Brinell Hardness: The Brinell hardness test measures the ability of 
 the metal to resist penetration by a small ball when propelled by a gradually 
 applied force. It consists in pressing a hardened steel ball into the surface 
 of the specimen under test by means of a fixed load gradually applied. 
 The instrument consists essentially of a small hydraulic press, which is 
 operated by a small hand pump and is provided with a pressure gauge 
 for reading the pressure, and a special contrivance for automatically 
 holding the pressure when it has reached a maximum of 3000 kilograms. 
 The piston of the press, which acts vertically downward, is provided 
 on its end with a hardened steel ball, ten millimeters in diameter, by 
 means of which an impression may be made on the smooth surface of the 
 specimen, which rests on a firm but an adjustable base. The operation of 
 the instrument is very simple. The specimen, the surface of which has been 
 planished with a file, a whetstone, emery wheel or similar means, is laid on 
 the base and is then brought in contact with the ball by turning a small 
 wheel for adjusting the base, or platform. By operating the hand pump 
 until the maximum pressure is attained and maintained for about one 
 half minute, the steel ball is pressed into the surface; then the pressure 
 is relieved, the base is lowered, and the diameter of the impression 
 made in the specimen is measured by means of a microscope fitted 
 with a millimeter scale, vernier, and cross hair. From this diameter the 
 spherical area of the impression may be calculated, which, divided into the 
 maximum load of 3000 kilograms, gives the hardness number. The formulas 
 for making these calculations may be combined into a single formula, thus : 
 
 H=- 
 
 where P=3000 Kilograms pressure, r=5 mm., radius of the ball, D=diameter 
 of the impression, and H=the hardness number. In practice it is most 
 convenient to have a table, such as that shown below, prepared, from 
 which the number may be obtained direct from the diameter of the 
 impression. 
 
BRINELL TEST 
 
 311 
 
 Relation of Brinell Number to Tensile Strength: It is both a 
 curious and a significant fact that the Brinell hardness number bears a 
 close relation to the ultimate strength, as may be seen from an inspection 
 of the following table, which was prepared only after comparing results 
 obtained upon thousands of specimens, to which both the Brinell and the 
 pulling tests had been applied. This relation, it will be observed, is 
 approximately 500, and holds for all grades of carbon steel whether they 
 be heat treated or in their natural state as forged or rolled. For this 
 reason the Brinell test is applicable to the rapid testing of steel from which 
 samples for the tensile test cannot be obtained. 
 
 Table 47. Brinell Hardness Numbers and Estimated Tensile Strength 
 for 3000 Kilogram Pressure on a 10 MM. Ball Testing Machine. 
 
 Diam. 
 of Im- 
 pression 
 in m/m 
 
 Hard- 
 Number 
 
 Ultimate 
 Pounds 
 per Sq. In. 
 
 Diam. 
 of Im- 
 pression 
 in m/m 
 
 Hard- 
 ness 
 Number 
 
 Ultimate 
 Pounds 
 per Sq. In. 
 
 Diam. 
 of Im- 
 pression 
 in m/m 
 
 Hard- 
 ness 
 Number 
 
 Ultimate 
 Pounds 
 per Sq. In. 
 
 2.00 
 
 946 
 
 465100 ;; 3.35 
 
 332 
 
 162700 
 
 4.70 
 
 163 
 
 80100 
 
 2.05 
 
 878 
 
 442100 
 
 3.40 
 
 321 
 
 157800 
 
 4.75 
 
 159 
 
 78300 
 
 2.10 
 
 857 
 
 421600 
 
 3.45 
 
 311 
 
 153100 
 
 4.80 
 
 156 
 
 76600 
 
 2.15 
 
 817 
 
 402000 ij 3.50 
 
 302 
 
 148600 
 
 4.85 
 
 153 
 
 74900 
 
 2.20 
 
 782 
 
 383700 i 3.55 
 
 293 
 
 144300 
 
 4.90 
 
 149 
 
 73300 
 
 2.25 
 
 744 
 
 366600 3.60 
 
 286 
 
 140200 
 
 4.95 
 
 146 
 
 71700 
 
 2.30 
 
 713 
 
 350600 | 3.65 
 
 277 
 
 136200 
 
 5.00 
 
 143 
 
 70200 
 
 2.35 
 
 683 
 
 335700 3.70 
 
 269 
 
 132400 
 
 5.05 
 
 140 
 
 68700 
 
 2.40 
 
 652 
 
 321600 3.75 
 
 262 
 
 128800 
 
 5.10 
 
 137 
 
 67200 
 
 2.45 
 
 627 
 
 308400 I 3.80 
 
 255 
 
 125300 
 
 5.15 
 
 134 
 
 65800 
 
 2.50 
 
 600 
 
 295900 1 3.85 
 
 248 
 
 121900 
 
 5.20 
 
 131 
 
 64500 
 
 2.55 
 
 578 
 
 284300 
 
 3.90 
 
 241 
 
 118700 
 
 5.25 
 
 128 
 
 63100 
 
 2.60 
 
 555 
 
 273300 
 
 3.95 
 
 235 
 
 115500 
 
 5.30 
 
 126 
 
 61800 
 
 2.65 
 
 532 
 
 262900 
 
 4.00 
 
 228 
 
 112600 
 
 5.35 
 
 124 
 
 60600 
 
 2.70 
 
 512 
 
 253100 
 
 4.05 
 
 223 
 
 109700 
 
 5.40 
 
 121 
 
 59400 
 
 2.75 
 
 495 
 
 243800 
 
 4.10 
 
 217 
 
 106900 
 
 5.45 
 
 118 
 
 58200 
 
 2.80 
 
 477 
 
 235000 
 
 4.15 
 
 212 
 
 104200 
 
 5.50 
 
 116 
 
 57000 
 
 2.85 
 
 460 
 
 226600 
 
 4.20 
 
 207 
 
 101600 
 
 5.55 
 
 114 
 
 55900 
 
 2.90 
 
 444 
 
 218700 
 
 4.25 
 
 202 
 
 99100 
 
 5.60 
 
 112 
 
 54800 
 
 2.95 
 
 430 
 
 211200 
 
 4.30 
 
 196 
 
 96700 
 
 5.65 
 
 109 
 
 53700 
 
 3.00 
 
 418 
 
 204100 
 
 4.35 
 
 192 
 
 94400 
 
 5.70 
 
 107 
 
 52700 
 
 3.05 
 
 402 
 
 197300 
 
 4.40 
 
 187 
 
 92200 
 
 5.75 
 
 105 
 
 51700 
 
 3.10 
 
 387 
 
 190800 
 
 4.45 
 
 183 
 
 90000 
 
 5.80 
 
 103 
 
 50700 
 
 3.15 
 
 375 
 
 184600 
 
 4.50 
 
 179 
 
 87900 
 
 5.85 
 
 101 
 
 49700 
 
 3.20 
 
 364 
 
 178800 
 
 4.55 
 
 174 
 
 85800 
 
 5.90 
 
 99 
 
 48800 
 
 3.25 
 
 351 
 
 173200 
 
 4.60 
 
 170 
 
 83900 
 
 5.95 
 
 97 
 
 47900 
 
 3.30 
 
 340 
 
 167800 
 
 4.65 
 
 166 
 
 82000 
 
 
 
 
 Pressure 
 
 =Hardness Number. 
 
 Area of Impression 
 
 Tensile in Kg. per Sq. MM.=Coefficient .346 x Hardness Number. 
 
 Factor to Convert Kg. per Sq. M/M to Lbs. per Sq. In.=1422.3 
 
312 MECHANICAL TREATMENT 
 
 CHAPTER II. 
 
 THE MECHANICAL TREATMENT OF STEEL. 
 
 SECTION I. 
 
 METHODS AND EFFECTS OF MECHANICALLY WORKING STEEL. 
 
 Methods of Shaping Steel: After the separation of the metal from 
 its ores, which in modern practice is accomplished by means of either the 
 blast furnace or a form of electric furnace, and its purification in the 
 Bessemer converter, open hearth, puddling furnace, or electric furnace, the 
 third step in the metallurgy of iron is the reduction of the large bodies 
 of metal thus produced to the various forms and sizes required by the 
 many uses to which it is to be put. In general this shaping may be brought 
 about either by pouring the metal while in a molten state into moulds, 
 which act is called casting, or by mechanically working it. Since by all 
 the methods of purification, puddling excepted, the metal is obtained in 
 the fluid state, casting would appear to be the simplest and cheapest 
 method of shaping; but for forming articles of very small section, it is 
 evident that this method is impracticable; nor is it used, unless unavoid- 
 able, to form the larger sections in which the mechanical properties of the 
 metal must be developed to the highest degree. Some shapes on account 
 of their size or their intricate design require casting, while others are cast 
 because they require no great strength in service and the cost of production 
 only is to be considered. A lack of strength and ductility in castings is 
 inherent, and is due to chemical and physical phenomena that accompany 
 the solidification of the molten metal, something about the nature of which 
 will be explained later in connection with the cooling of ingots. Suffice it 
 to say now that the weakness of castings is due chiefly to any or all of 
 three causes, namely, blow holes, segregation, and crystallization. 
 
 Benefits of Mechanical Working: On the other hand, mechanical 
 shaping improves the quality of the metal by forcing its particles into 
 more intimate contact, closing up cavities, and by refining its crystalline 
 structure, and so has important functions aside from the mere reduction 
 to form and size. The change in properties that may be attributed to the 
 process of mechanical working is a marked one, for the strength, ductility 
 and hardness are all affected. Of these properties the strength is always 
 increased by the working, the hardness may or may not be markedly 
 increased, while the ductility, i. e., elongation and reduction in area, may 
 be either increased or decreased, depending on the conditions of the 
 working. The amount of change in each of these properties for a given 
 steel of a certain chemical composition is affected by the amount of work 
 done and by the temperature at which the working is carried on. 
 
HOT AND COLD WORKING 313 
 
 Hot and Cold Working: In the mechanical treatment of the metal, 
 the first distinction to be made is that of hot and cold working. The 
 study of metallography has shown that the term hot working of steel should 
 be applied to the working of it at temperatures above its upper critical 
 range, the temperature of which varies, inversely with the carbon content, 
 from 700 to 900 C., while all work done at temperatures below this 
 range should be called cold working. It will be shown in the next part of 
 this book that a sharp change in structure due to working takes place as 
 the critical temperature of the steel is passed. This change is due mainly 
 to the fact that above this range iron exists in an allotropic crystalline 
 form, the gamma form, in which carbon dissolves to form a homogeneous 
 mixture, while below it the metal assumes the alpha form and is a crystal- 
 lized aggregate of ferrite and cementite. A metallographic examination 
 of specimens shows that the result of working this aggregate structure is 
 one of permanent distortion, or strain, and one in which the properties of 
 the metal are deeply affected, as indicated by the different physical 
 tests. The elastic limit, tensile strength, and hardness are increased, 
 while the ductility is reduced. The extent of this change varies according 
 to the temperature, and is most marked when the working is done 
 at or below atmospheric temperatures. Cold working becomes less 
 effective as the critical temperature is approached, which is due to the 
 increase in molecular energy and the resulting loss of rigidity by the solid. 
 It should be noted that below the critical range no refinement of the granular 
 structure can be accomplished by working. In hot working, the grain size 
 is decreased, and the metal is subject otherwise to mechanical refinement, 
 the extent of which depends not only on the amount of work done and size 
 of the section but on the temperature above the critical range at which 
 the work is finished. However, if, after a working, the metal be heated 
 above this finishing temperature, as is often the case in the actual rolling 
 of steel, the grain refinement of the previous working may be partly or entirely 
 destroyed, depending upon the temperature to which the piece is reheated. 
 Owing to the plasticity of the metal at the higher temperatures, the dis- 
 tortion due to working above the critical range does not produce a per- 
 manent strain in the structure of the solid. After each distortion the 
 structural components, the crystals or grains, are free to return to the shape 
 and arrangement peculiar to their state of equilibrium. Besides, since the 
 steel takes on a new structure and a new condition is born upon cooling 
 through the critical range, any internal tension set up by the working 
 is relieved by the rearrangement that takes place in passing through this 
 range. This fact gives another reason for the superior quality of hot 
 worked material over castings, which are subject to immense internal 
 tension, or stresses, set up by physical phenomena that accompany the 
 solidification and by the forces of contraction due to the unequal rates of 
 cooling between the exterior and interior of a casting. Such severe stresses 
 do not occur in hot worked material. Referring to the more common prac- 
 
314 
 
 MECHANICAL TREATMENT 
 
 
 
 
 1. Cast steel. Carbon .35 per cent. Magnified 100 
 diameters. 
 
 
 3. Cold worked hypo-eutectoid steel. Carbon 0.30 
 per cent. Magnified 100 diameters. 
 
 FIG. 48. Showing Effects of Working 
 
HOT AND COLD WORKING 
 
 315 
 
 2. Hot worked steel. Carbon 0.50 per cent. Finishing 
 temperature high. Magnified 100 diameters. 
 
 4. Hot worked steel. Carbon 0.50 per cent. Fin 
 ishing temperature low. Magnified 100 diameters 
 
 upon the Grain Structure of Steel. 
 
316 METHODS OF WORKING 
 
 tice of working the steel on its initial heat, that is, working it before it 
 has cooled much below the temperature of solidification after having 
 been cast subsequent to manufacture in the molten state, careful study 
 has developed the fact that it matters little, so far as the effect on the 
 refinement wrought by the working is concerned, whether the ingot has 
 or has not been allowed to become completely cold before being brought 
 to the required temperature for working. Therefore, while the idea is 
 contrary to the popular notion, the primary object in the steel workers mind 
 should be the improvement in quality of the material he is working, 
 while the shaping of the material may be looked upon as a secondary 
 object. 
 
 SECTION II. 
 
 SUMMARY OF THE HISTORY AND PRINCIPLES OF WORKING STEEL. 
 
 The Three Methods for Mechanically Working Steel: With 
 reference to the manner of applying pressure to steel during mechanical 
 working there are three possible methods; namely, hammering, pressing 
 and rolling, all of which are extensively used at the present time. The 
 shaping of steel by either of the first two methods is called forging. As 
 an introduction to the study of the rolling of steel, a brief resume of the 
 history, principles, and effects of each of these methods will not be out 
 of place and may be found quite interesting. 
 
 Hammer Forging: Hammering was the first method employed by 
 man in shaping the metals. The first forging was done by hand hammers 
 wielded by the workmen. The first power hammer, known by the name 
 of tilt hammer, was built in England, and was a crude affair com- 
 pared with the steam hammers now used. It consisted of a beam of wood 
 hinged at one end and provided with an iron hammer head at the other. At 
 an intermediate point, engaging cams on a revolving shaft alternately 
 raised the free end and allowed it to fall on a bottom die fixed upon a suitable 
 foundation. Thus the top die could be parallel to the bottom die in only 
 one position, and the larger the piece to be forged the less power there 
 was available to forge it. The first steam hammer was built in France 
 in 1842. It consisted of a two piece frame constructed so as to support, 
 directly over a die or anvil, a steam cylinder, to the piston rod of which 
 was attached a tup, or hammer head. By admitting steam into the cylinder 
 below the piston, the hammer was raised for a distance equal to the stroke 
 of the cylinder, and then allowed to drop upon the anvil or bottom die. 
 This hammer had the advantage of always keeping the top and bottom 
 dies parallel, but was still lacking in one important particular. Its power 
 being derived from the inertia of the falling tup, the hammer had the least 
 power when it was most needed, that is, when pieces of large diameter or 
 of great thickness were being worked. This fault in the single acting 
 hammer was corrected by the invention of the double acting hammer, in 
 
EFFECTS OF FORGING 317 
 
 which steam is admitted at the top of the piston and employed on the 
 downward stroke as well as for lifting the tup. The first double acting 
 hammer was built at Midvale, Pa., in 1888. 
 
 Principles and Effects of Hammering: The principles of the hammer 
 are that of an instantaneous application of pressure applied to a relatively 
 small area. The strains set up in the metal are compressive and take 
 place in a vertical direction in the region below the area subjected to the 
 force of the blow. The crowding of the metal into one region, however, 
 causes a small portion of the blow to be transmitted in horizontal directions. 
 The suddenness of the blow tends to localize the effect and confine the 
 refinement to the exterior. This fact results in a high degree of refinement, 
 provided the amount of reduction is great or the section worked is a thin 
 one, and is one of the reasons why it is possible to make some hammered 
 material superior to rolled material. The resistance of the metal to 
 deformation under shock, combined with the intermittant action of the 
 hammer, makes shaping by hammer a slow process. 
 
 The Forging Press: The press is an English invention, dating from 
 the year 1861. It was introduced into this country about the year 1887. 
 It consists essentially of a hydraulic cylinder supported by one or two 
 pairs of steel columns which are anchored to a single base casting of great 
 weight and strength. The ram of the cylinder points downward and carries 
 an upper forging bitt vertically opposite a similar lower and stationary 
 bitt which rests on the base casting to which the columns are attached. By 
 admitting water under pressure to the cylinder at its top the upper pallet 
 is forced down upon the material to be forged, which rests upon the lower 
 pallet. The pressure is applied slowly and is gradually increased to a 
 maximum which may be maintained till the metal yields. By means of 
 small auxiliary cylinders the ram is lifted after each application of pressure. 
 The pressure exerted by the forging press is very great. In practice it 
 is found that the lowest pressure that can be employed to be effective at 
 a full forging heat is about 1.2 tons per square inch, but the pressures 
 employed in actual work will often reach 13.26 tons per square inch. 
 
 The Effect of Pressing : The press differs very much from the hammer, 
 both in action and the effects produced. Unlike the instantaneous appli- 
 cation of the pressure as in the case of the hammer, the action of the press 
 is so slow that a kneading of the metal takes place, and the strain, instead 
 of being confined to the surface, penetrates deep into the material. An 
 illustration cited by Messrs Harbord and Hall will serve to demonstrate 
 the difference in the effect produced by the two methods of working. 1 
 
 "If tests are taken from the outer parts of a gun forging which has the 
 center trepanned out, little difference is found in the strength of the 
 material, whether the forging was done under the press or under the 
 hammer, provided the latter was sufficiently heavy for its work; the 
 
 Vol. II. Page 855. Published by J. B. 
 
 JSee The Metallurgy of Steel. 
 Lippincott Company, Philadelphia, Pa., 
 
318 METHODS OF WORKING 
 
 press showing, if anything, slightly better results. If, however, the test 
 pieces are taken from the cores which have been cut out of the center of 
 the forgings, the difference in the results is so very marked as to have in 
 duced all the best makers of heavy steel forgings to install presses in- 
 place of, or in addition to, their large hammers." 
 
 Advantages of the Press: Aside from its increased beneficial effect 
 upon the material, the press has many advantages over the hammer, some 
 of which it may be of interest to cite. The absence of shock in the press 
 is a decided advantage both in the construction of the machine and in the 
 working of material. The cost of working material under a press is less 
 than with the hammer because the output is greater, the press reducing 
 faster than the hammer, fewer men and less skilled labor are required, 
 and the fuel consumption per ton of output is less. A much greater propor- 
 tion of the total work put into a press is transmitted to the metal than is 
 the case with the hammer. Much of the energy of the latter is dissipated 
 through being absorbed by the spring in the anvil block and by the earth. 
 For certain work, however, this impact gives the hammer two advantages : 
 first, it serves to remove scale; second, it enables the hammer to strike 
 forgings in molds with greater ease than the press. The difficulty of 
 retaining water under the extremely high pressures required by presses 
 gives the hammer an advantage, but this advantage is offset by its greater 
 liability to breakage. 
 
 Rolling: Of all the known methods of shaping steel from the cast 
 material, that of rolling, as introduced by Henry Cort in 1783, though 
 perhaps not producing the best quality in certain classes of product, has 
 come to be the most extensively employed. Though Cort is rightly credited 
 with being the father of modern rolling, the use of this principle in shaping 
 metals antedates his mill by many years. Thus, there are records to 
 show that in the year 1553 a Frenchman employed rolls to produce sheets 
 of uniform thickness for the stamping of gold and silver coin. In Sweden 
 rolls were employed to produce certain steel sections prior to the year 1751, 
 and even at that time the assertion was made that as much as twenty 
 times more bars could be reduced in a given time than could be shaped 
 under the tilt hammer of those days. This fact, coupled with the great 
 efficiency of the rolling method, is responsible for the universal adoption 
 of rolling as the favorite method of shaping. The rapid growth in the 
 production due to the ever increasing demand for iron and steel, made it 
 imperative that the most rapid method of shaping be employed. From 
 the days of Cort to the present time, the rolling mill has kept^pace with 
 the growth in production and has passed through a surprisingly rapid 
 process of development, not only in size and power but in design and in 
 the shapes of sections turned out. This development, together with the 
 introduction of numerous appliances for handling the material mechanically 
 during the rolling, has multiplied the capacity of the mills many times. 
 
PRINCIPLES OF ROLLING 319 
 
 Some modern mills, like the rod mill, will now turn out a hundred times 
 as much tonnage in a given time as a mill of the same size and working 
 on rods of the same size could have done fifty years ago. 
 
 Principle and Effect of Rolling: The process of shaping steel by 
 rolling consists essentially of passing the material between two rolls revolv- 
 ing at the same peripheral speed and in opposite directions, i. e., clockwise 
 and counter clockwise, and so spaced that the distance between them is 
 somewhat less than the height of the sectidn entering them. Under these 
 conditions the rolls grip the piece of metal and deliver it reduced in section 
 and increased in length in proportion to the reduction, except for a slight 
 lateral spreading which is almost negligible in some sections. The extent 
 of the spread will be found to depend mainly upon the amount of reduction 
 and width of the piece. Thus in rolling plates, the total spread may be 
 less than that of the first pass in the reduction of two-inch billets, 
 especially if the percentage reduction in sectional area in the latter is 
 great. The nature of rolling may be best explained by means of the follow- 
 ing diagram. 
 
 I ~ "^^e-- 
 
 Fia. 49. Diagram Illustrating the Nature of Rolling. 
 
 Let O A C and (V A' C' be two plain motionless rolls which are being 
 forced into the bar AA' E'E by means of pressures applied vertically from 
 F and F'. The force exerted by the resistance of the bar will act along 
 the radii of the rolls, as OA, OC, O'A', and O'C'. The resultant of all 
 these forces will be in the vertical lines O B and O'B'. Vertical compression 
 of the metal will, however, occur only between the points B and B'. At 
 all the other points between AC and A' C' the metal is forced away from 
 the rolls and the bar is elongated. If now the rolls are made to revolve, 
 the lower one in a clockwise direction and the upper in a counter clockwise 
 direction, the piece is reduced in size and elongated as shown by figure 
 A A' D'D. This turning of the rolls introduces a second force, which 
 acts in the direction of tangents to the arcs AB and A' B' and is equal to 
 the force of friction and therefore proportional to the pressure between the 
 rolls and the piece. The result of this force is to subject the piece to a 
 
320 METHODS OF WORKING 
 
 longitudinal pull in the direction of B to D, this pull being at its maximum 
 at B and at its minimum at A. The compression, however, is at its 
 maximum at A and its minimum at B. The net result of this double action 
 is to cause the metal to flow forward so that the piece, reduced in size, 
 is delivered at a higher velocity than the peripheral speed of the rolls, 
 the evidence for which is found in the fact that the marks on the piece 
 caused by a depression or elevation on the rolls is farther apart than the 
 circumference of the rolls. A slight retardation of the forward speed of the 
 piece on the entering side may take place, but this point has not been 
 very well established as a fact. 
 
 Rolling Compared with Hammering and Pressing: It is a very 
 difficult matter to institute a fair comparison between the effects of rolling 
 with those of hammering or pressing. Each method has a field of its own 
 with rather well defined boundaries. Thus, many shapes are so intricate in 
 design that rolling them is out of the question, and so they must be formed 
 under the hammer or the press. A crank shaft and a hammer head serve 
 as examples of these classes of shape, which can be produced in no other 
 way, unless by casting, when they would then be lacking in the strength, 
 ductility and soundness imparted by working. That the hammer and the 
 press are both under better control than rolls is evident, and being slower 
 and more expensive to operate than rolls, these tools are used on material the 
 cost of manufacture of which is a secondary matter. Hence, extraordinary 
 care and attention is given to all phases of the working of forged articles. 
 Rolls on the other hand are cherished for their speed, and tonnage is always 
 a factor in rolling. There is, however, a small area in which the field of 
 operation of all three instrumentalities overlap, as in the shaping of billets 
 or blooms from ingots. In working these billets the different effects 
 produced by the three methods become visible as the piece is shaped. 
 When an ingot is hammered, the shape imparted to the section is very 
 liable to look like A in Fig. 50. If the blows are light and delivered at 
 high velocity, the upper surface only will be elongated as shown at B in 
 the figure. These facts show that the impact is almost entirely absorbed 
 by the surface of the metal, and to obtain the best effect from hammering 
 it is necessary to continue the work until the section has been reduced to 
 one of relatively small size. The deeper penetration of the work of the 
 press is shown by the rounded corners of the section represented by the 
 figure at C. Any cavity at the center of the piece is closed under the 
 action of the press, whereas the tendency of the hammer would be to enlarge 
 it. The effect of rolling is influenced very markedly by the temperature. 
 In the first place the temperature, in order to secure the greatest efficiency 
 from the rolls, is likely to be higher than that required for either hammering 
 or pressing. In a piece uniformly heated, the flow of the metal is slightly 
 faster at the two surfaces than in the center in the smaller sections, while 
 in larger sections the flow at the surface may be very much greater. The 
 effect of the additional plasticity imparted by even a slight rise in tern- 
 
COMPARISON OF METHODS 
 
 321 
 
 perature upon the flowing properties of the metal is plainly visible in the 
 results obtained in rolling ingots under the two conditions as illustrated 
 by the figures at D and E. The fishtailing of the piece represented at D 
 shows that the flow of the metal is faster at the surface due to the lack 
 
 Hammering 
 
 B 
 
 Heavy Strokes 
 
 Light Strokes 
 
 Pressing 
 
 Rolling 
 
 D 
 
 Center cooler than surface Center hotter than surface 
 
 FIQ. 50. Diagram Illustrating the Effects of Hammering, Pressing and Rolling. 
 
 of plasticity at the colder center. The reverse of these conditions is shown 
 at E with the corresponding difference in effect. In this case the effect 
 of the working has penetrated to a much greater depth and extent than 
 in the previous case. The amount of draught and the speed of rolling are 
 also important factors in producing these effects, a more thorough dis- 
 cussion of which will be taken up later. 
 
 Rolling and Pressing Ingots: The notion most prevalent among 
 steel men, however, is that the tendency of rolling is to produce a more 
 superficial effect than either hammering or pressing. That this notion is 
 correct with respect to pressing is indicated by the precautions taken in 
 casting large ingots for armor plate that is to be rolled. Figure 51 shows 
 the difference in shape of ingots for the press and for the rolls. The concave 
 and diamond shaped sides of the ingot for rolling are formed to prevent 
 the loss due to fishtailing, as already explained. Under uhe press the two 
 surfaces of a square sided ingot are slightly rounded, but, in rolling, a 
 square sided ingot would make a concave sided plate which in many cases 
 progresses to such an extent as to cause actual overlapping. It is admitted 
 by nearly every one, however, that with very slow rolling and carefully 
 regulated temperature that the quality of rolled material may be made 
 the equal of that reduced under the press. 
 
 For Pressing For Pressing For Rolling 
 
 FIG. 51. Shapes of Ingots for Pressing and Rolling Armor Plate. 
 
322 THE ROLLING MILL 
 
 CHAPTER III. 
 
 ESSENTIALS OF ROLLING MILL CONSTRUCTION AND 
 OPERATION. 
 
 SECTION I. 
 
 THE ROLLS THEIR PREPARATION AND ARRANGEMENT. 
 
 Parts and Equipment of the Simplest Type of Rolling Mill: After 
 the rolls, themselves, two in number in the simpler types of mill, the next 
 most essential part of the mill are the chocks or bearings which support 
 the ends of the rolls and permit them to be turned without displacement. 
 The chocks in turn are kept in place by means of the housings, which 
 together with the adjusting screws also furnish a means by which the 
 distance between the rolls is regulated. These parts constitute a stand 
 of rolls. The housings are bolted to shoes which rest upon a firm foun- 
 dation, to which they are always securely bolted. Next in im- 
 portance are the parts which connect the mill with the driving shaft. 
 First, there are the spindles that transmit the power from the pinions 
 to the rolls, to both of which they are connected by means of loosely fitting 
 coupling boxes. The pinions, supported in housings similar to the roll 
 housings, are gears, one of which is driven through a driving spindle in line 
 with one of the rolls. They serve to impart opposite motions to the rolls. 
 The last part of equipment essential to the mill is the prime mover, which 
 in modern mills may be a steam engine or an electric motor. As to other equip- 
 ment, reheating furnaces are first in importance. Large mills must also be 
 provided with roll tables for handling the material. A discussion of the 
 driving apparatus is an engineering subject which lies beyond the intended 
 scope* of this book and will receive no further mention here. All the 
 remaining parts, however, should be studied somewhat in detail. 
 
 The Rolls and Their Parts: Of the essential parts of the rolling 
 mill the rolls furnish a subject of great interest. There are three parts to 
 a roll; namely, the body, which is the part on which the rolling is done; 
 the necks, or the parts which rest in the chocks and furnish the surface 
 upon which the pressure is applied for reducing the size of the piece; and 
 the wobblers, one at the outer end of either neck or of both necks, which 
 are formed by notching the prolongation of the neck of the roll. Over the 
 wobblers the coupling box for driving the roll is fitted. In the case of 
 plain rolls, such as are used for rolling plates and, in part, for other flats, 
 these are the only parts of the roll. In the case of rolls for other material, 
 grooves are cut into the surfaces of the rolls to form the section required. 
 
THE ROLLS 
 
 323 
 
 A groove in one of the rolls or a combination of grooves in the two rolls, 
 which at the line of contact forms an opening corresponding to the shape 
 of the section desired, is called a pass. The three most common passes 
 are shown in the accompanying figure. 
 
 I Open Box 
 
 Closed Box 
 
 .ill Diamond {90 ) IV Gothic 
 
 FIG. 52. Showing Different Types of Passes For Roughing and Semi-finishing Mills 
 I, III and IV are spoken of as open passes while II is called a closed pass. 
 In the closed pass the piece is buried in one of the rolls so that three sides are 
 enclosed by the groove A, the fourth side being closed by the tongue or former 
 D, on the other roll. Collars are represented at O and Ci. Passes I and II 
 are commonly spoken of as box passes, while III is called the diamond pass, 
 and IV the Gothic pass. 
 
 The Manufacture of Rolls is a separate industry, and the art of roll- 
 making is not widely known even among the users of rolls. When the work 
 the rolls have to do is considered, together with their effect upon the product 
 of the mill, the importance of good rolls is better appreciated. Most rolls 
 are castings, yet they must be ductile to withstand the shock produced 
 as the piece enters them; strong to resist sufficiently the great pressure 
 applied to their ends; hard to give them good wearing qualities; and sound, 
 so that they may not develop surface defects which would leave their marks 
 on every surface rolled on them and cause the material to be rejected. 
 To secure these qualities the best of materials and the greatest of skill 
 are required in their manufacture. The materials are of three kinds, namely, 
 cast iron, steel and alloy mixtures. From these materials four kinds of 
 rolls are produced. They are known as sand rolls, which are made of pig 
 iron; chilled rolls, also of cast iron; steel rolls, made of steel by casting; 
 and "adamite" rolls, which is a trade name for a metal produced by mixing 
 steel with pig iron containing certain percentages of chromium and nickel, 
 or by mixing steel and the ferro alloys of these elements with the proper 
 amount of an ordinary pig iron of high grade. As an example of how rolls 
 are made, some of the processes as carried out by one of the leading manu- 
 facturers of rolls will be briefly described. 
 
 The Sand Roll: Sand rolls are cast in a sand mold. The sand used 
 is a loamy sand of a special kind obtained only from deposits left in old 
 water courses. This sand contains sufficient clay intimately mixed with 
 the silica to form a firm bond and yet be refractory enough that it will 
 not fuse at the temperature of the molten iron. The mold is prepared by 
 ramming this sand, moistened a little, into a half flask, and then sweeping 
 
324 THE ROLLING MILL 
 
 the sand from the half flask with a sweep, the outline of which is similar 
 to the contour of the roll. Two such half flasks are required for each roll, 
 each one containing one-half of the roll divided longitudinally. After 
 sweeping and smoothing, the half molds are coated inside with a plumbago 
 or other carbonaceous dressing and carefully dried. Just before casting, 
 these two parts of the mould are firmly clamped together and are set in 
 a vertical position for pouring, for which purpose a casting pit is provided 
 for large rolls. Thus, one end of the roll forms the bottom of the casting, 
 the other end the top. The top is capped by a cope to provide a deep sink- 
 head, which is cut from the roll after casting. The gating to the mold 
 enters the flask at the bottom neck of the roll and on a tangent, so that 
 a swirling action is imparted to the molten metal as it rises in the mold. 
 In this way all dirt and other foreign matter is forced to the center, which 
 condition insures the outer portion of the roll will be composed of cle*an 
 metal. 
 
 The Materials Used in Sand Cast Rolls are charcoal iron and roll 
 scrap. The mixtures are melted in coal fired reverberatory furnaces. 
 The bath, sealed off from outside air, is separated from the grate by a 
 bridge wall, over which a non-oxidizing flame sweeps and furnishes the heat 
 for melting. In the melting a little carbon, silicon, and manganese are 
 removed from the metal, and by the time the charge is melted a highly 
 silicous slag has formed, which protects the metal from any further action 
 that might be produced by the flame. As soon as the metal is melted, 
 fracture tests are taken, by means of which the metallurgist in charge 
 is able, from long experience, to determine when the bath is of the right 
 composition to produce the kind of roll desired. The molten metal is 
 tapped from the furnace into a small tilting ladle, which is carried by 
 overhead crane to the molds, and the metal is poured into the gate over the 
 lip of the ladle. The pouring is very rapid and must be continuous, as the 
 slightest interruption would ruin the casting. After the metal has 
 solidified and cooled sufficiently, the mold is removed, and the roll is 
 cleaned of the adhering sand, when it is ready to be machined to the size 
 and shape required. 
 
 Chilled Rolls: Rolls of the chilled type are made up of three layers 
 of metal, each of which represents a type of the same original metal. The 
 interior of these rolls is composed of grey iron, which is enclosed by a 
 cylinder of mottled iron, and outside of this a similar layer of white iron, 
 called the chill. This composite structure is procured by taking advantage 
 of the peculiar properties exhibited by pig iron on cooling from the molten 
 state. In this state iron holds in solution all the carbon which it contains 
 at a given temperature. In cooling some of this carbon separates in the 
 form of crystals of graphite, which is distributed throughout the mass; 
 the remainder is spoken of as combined carbon, the effect of which is to 
 increase the hardness of the metal. The separation of the graphite depends 
 mainly upon the rate of cooling, so that if the iron is cooled very suddenly 
 
CHILL ROLLS 
 
 325 
 
 all the carbon may be retained in solution as combined carbon, which 
 renders a chilled iron that is dense, white, intensely hard, and capable of 
 receiving a very high polish. In making these rolls only the body of the 
 roll is given a chill. This chilling of only a part of the roll is effected by 
 making that part of the mold corresponding to the necks and wobblers of 
 sand, while that part destined to form the body is made up of a heavy 
 cast iron ring, usually built up in sections which are carefully turned at 
 the joints and bored out true inside. After giving the inside of the mold 
 a coating of the carbonaceous wash, they are warmed to remove moisture, 
 
 The line in the cut marks the limit of clear chill. When depth of chill is 
 designated, it is assumed to mean clear chill. 
 
 FIG. 53. Method of Measuring Depth of Chill on Rolls. 
 
 then assembled, and the casting is made as for sand rolls. The rapid 
 cooling, caused by the absorption of the heat by the cold casting in contact 
 with the molten metal, causes the chill on the outer surface of the roll, 
 the depth and hardness of which is controlled by varying the composition 
 of the molten iron. Chilled rolls, once they are formed, cannot be softened 
 or hardened by heat treatment, as such treatment would destroy the chill 
 A patented chill is now in use. It is made in the form of a ring com- 
 posed of segments of solid metal on the inside and a water cooled ring 
 on the outside. This construction has the effect of causing the mold to 
 become smaller, as it is warmed by the heat from the molten metal, thus 
 subjecting the roll to a high pressure, which is said to give a more even 
 chill and a denser and tougher material than the common chill. The chill 
 
326 THE ROLLING MILL 
 
 is measured by the least depth of clear chill as shown in the accompanying 
 photograph, while the analysis of each of the three regions here depicted 
 is given in the following table: 
 
 Table 48. Analysis of Different Parts of a Chilled Roll. 
 
 TOTAL, COMB D. GRAPH. 
 CARB. CARB. CARB. 
 
 Chill 3.00 3.00 90 .04 .200 .25 
 
 Mottled 3.00 2.25 .75 .90 .04 .200 .25 
 
 Grey 3.00 1.00 2.00 .90 .04 .200 .25 
 
 Difficulties in Making Chilled Rolls: The greatest of skill and 
 experience are required in the making of chilled rolls. The process of 
 chilling causes the different parts of the roll to cool at different rates and 
 sets up stresses in the casting which make it liable to crack and break. 
 The range of temperature at which the metal may be poured is very narrow, 
 while a very slight change in the chemical composition of the metal will 
 sometimes produce a marked effect upon the chill, changing both the depth 
 and the hardness. The size of the roll also affects the nature and extent 
 of the chill. Besides, the roll in use is, subject to great pressure, uneven 
 stresses, uneven heating, over heating, and sudden cooling, all of which 
 tend to cause the chill to crack and spall. This tendency to spall is over- 
 come by the manufacturer to some extent, but careful handling of the roll 
 in use is essential also. Large rolls are especially difficult to cast properly. 
 The largest chilled rolls are made for rolling plates, and a very tough 
 chill is required. The chills for one of the largest of these rolls weighs 
 105,000 pounds and the roll itself requires 80,000 pounds of metal to cast 
 it, while the total length of the mold is twenty-three feet. A large 
 percentage of these rolls are lost in casting, due to the cracking of the roll 
 at places where the different sections of the chill are joined. Small chilled 
 rolls are used in guide, rod, hoop and bar mills, and for a variety of purposes, 
 but chilled rolls for shapes are very difficult to make owing to the fact that 
 the collars in such rolls are liable to bind in the chill and crack off. All 
 these factors tend to make chilled rolls very expensive, but a much greater 
 tonnage is obtained from them than from any other kind, and their use is 
 imperative where a very fine finish is required to be imparted to the product. 
 
 Steel Rolls are cast in sand in much the same way as sand rolls. In 
 this case, however, ganister sand mixed with a little fire clay to act as a 
 bond is used, because the higher temperature of molten steel will heat 
 any but the most refractory sands to their fusion point. Steel rolls are 
 stronger and more ductile than sand rolls. The deflection of a steel roll 
 under a given load is only about half as much as that of a common sand 
 roll. Besides, they may be annealed, when they become almost unbreak- 
 able. They cannot be permanently hardened, because any hardening by 
 heating and quenching is removed by contact with the hot metal, the heat 
 from which produces the same effect as a drawback. On account of their 
 
ROLL DESIGNING 327 
 
 unavoidable softness, then, steel rolls do not wear well, and hence cannot 
 be used on finishing stands. For blooming mills and roughing stands of 
 other mills where great strength is required, these rolls are invaluable, 
 and they are used in such stands almost exclusively. Occasionally, where 
 a good finish on the product is not required, steel rolls will be used on the 
 finishing stands. The material used is, for the most part, acid open hearth 
 steels varying from .40% to .65% in carbon content. When steel rolls are 
 used for finishing, the carbon content is increased to 186%, and sometimes 
 to as high as 1.25%. 
 
 Other Rolls : In an attempt to overcome the defective softness of 
 steel rolls and at the same time retain their great strength and toughness 
 the alloyed mixture previously referred to as "adamite," has been 
 developed. These rolls are being used with considerable success, and seem 
 to hold promise of even greater efficiency. Forged steel rolls have also 
 been tried and found to be very satisfactory, but their high cost prohibits 
 their use except where exceptional strength is required. 
 
 The Size of Rolls : In length of body, rolls vary from one to seventeen 
 feet, and in diameter from seven to forty-eight inches. The largest rolls 
 are used on the plate mills, the smallest on the small hand guide mills. 
 On account of the smaller surface exposed to pressure, small rolls cut into 
 the metal with greater ease than large ones and so require less power to 
 do the same work. Therefore, the heavier the rolls, the heavier must be 
 the machinery throughout the mill. The factor most important in determin- 
 ing the size of the roll is that of strength, and for the sake of safety rolls as 
 large as practicable will be employed; first cost is of secondary importance. 
 The resistance of a plain roll to transverse stress is proportional to the 
 cube of its diameter, and inversely proportional to the length of its body. 
 The diameter of the roll at the base of the deepest groove determines the 
 strength of a grooved roll; so, for grooves of the same depth, one set of 
 rolls may be many times stronger than another only one or two inches 
 smaller in diameter. The size of rolls is expressed by writing the diameter 
 and the length of body in inches, with the X sign between. Thus 42" X 
 60" means that the roll is forty-two inches in diameter and sixty inches 
 long in the body. 
 
 Roll Design: Designing the rolls was originally one of the duties of 
 the mill superintendent, the roll turner or the roller, but the demands 
 upon the mills in the way of new sections made it necessary to place this 
 work in the hands of men specially trained to the work, so that, now, roll 
 designing is a distinct profession. There are few rules in the trade, and 
 the roll designer must depend mainly upon experience for guidance. It is 
 seldom, therefore, that two roll designers will be found to develop a section 
 in precisely the same way. That exceptional* ingenuity and extreme 
 resourcefulness is required in this profession is attested by the wonderfully 
 intricate shapes these men are turning out, and that, too, with the most 
 astonishing accuracy. 
 
328 THE ROLLING MILL 
 
 Methods of Procedure in Designing Rolls: Given a new section to 
 evolve, the roll designer proceeds in some such manner as follows: From 
 a drawing of the section, if he has decided it is one that can be rolled success- 
 fully, he will have a templet made of the exact dimensions of the section, 
 and from this templet another for the finishing pass in which an allowance 
 of about .015 inch per inch of dimension of the finished piece is made for 
 contraction of the metal in cooling from the finishing temperature to 
 atmospheric temperature. He must then decide on the proper size of rolls 
 to use, which determines the mill that is to roll the section. This decision 
 made, he has given the approximate size of the billet or bloom from which 
 to begin, the number of sets of rolls, and the number of passes in which 
 the work must be done. Having given, now, the first and last passes with 
 their dimensions, and the total number of passes, he may begin the design 
 of the intermediate passes. This he does by drawings which are begun 
 by setting off a "construction line" or ' 'pitch line" as it is sometimes 
 called. This line locates the center of gravity, or the center of figure, of 
 the various passes and is usually placed midway between the axis of rotation 
 of the two rolls. 
 
 Difficulties in Designing Rolls: Having drawn the pitch line, the 
 roll designer then proceeds to mark off the passes from billet to finishing 
 pass, and in doing so he has a multitude of things that must be kept in 
 mind, some of which are: 1. The method of shaping is one of squeezing, 
 spreading, and bending. 2. The total amount of reduction is best dis- 
 tributed among the various passes as evenly as possible, excepting the 
 finishing, which is reserved to true up the shape. 3. All sides of the 
 piece should be thoroughly worked. 4. The piece should not enter two 
 successive passes in the same position, as otherwise the metal will be 
 squeezed out between the roll and form what is known as a fin. 5. Since 
 they weaken the roll very much, deep cuts into a roll should be avoided. 
 
 6. The passes should be so shaped as to eliminate side thrust on the rolls. 
 
 7. A piece will not enter a pass in the rolls if all its dimensions are larger 
 than the pass. 8. The thin parts of a section cool faster than the heavier 
 parts, and must, therefore, be formed in the last passes. 9. Sections that 
 require deep grooves in the rolls are difficult to roll successfully on account 
 of the difference in the peripheral speed of the bottom and the top of the 
 groove. The part of the roll having the greatest diameter elongates the 
 piece more rapidly than the part having the smallest diameter and tends 
 to cause the piece to twist and curl on leaving the rolls. This difficulty 
 can be overcome by using rolls of slightly different diameters, by raising 
 or lowering the center of mass of the piece from the pitch line or by reduc- 
 ing the amount of reduction on the part that elongates the more rapidly. 
 10. The draught on the various parts of a section must be properly 
 proportioned, as otherwise the piece will contain waves or be distorted in 
 other ways. 11. He must also keep in mind that all kinds of steel do not 
 work alike, and what can be done with open hearth steel, for instance, 
 
TURNING AND DRESSING ROLLS 329 
 
 would be impossible with Bessemer and vice versa. With these difficulties 
 to contend with, even highly experienced roll designers may fail on the 
 first trial at a new section. In that case an entirely new set of rolls may 
 be required, which adds much to the expense of rolling the section. Besides 
 questions, such as those above, that affect the shaping of the material, 
 the roll designer is also expected to consider time and cost. So, he will 
 endeavor to avoid roll changes or other operations that will delay the work 
 or add to the cost of the rolling operation. Thus, it will be found that in 
 most mills one set of roughing rolls will be used to produce a great number 
 of different sections. This has the effect of giving the designer a fewer 
 number of passes with which he forms the shape, and adds much to the 
 difficulty of his task. 
 
 Turning the Rolls: Having designed all the passes for the rolling of 
 a given section, a set of templets, one or more for each pass, is made. These 
 templets are to be used in turning the roll, for which purpose a special set 
 of tools may be required. In the roll shop, the rolls are first centered. 
 Various methods may be used for finding the center. When this point has 
 been located, a lead hole may be made with a ratchet drill, and then widened 
 out to the proper angle with a reamer to a depth of about % inch. The roll 
 is then placed in the necking lathe, when, supported by the center holes, 
 the necks may be turned to exact size, or they may be machined to near 
 the exact size and finished by grinding and polishing. Since the center 
 holes are liable to wear down irregularly if used throughout the process 
 of turning, the body of the roll is turned in another lathe in which 
 the roll is supported by chocks that fit the necks. Here the roll 
 is turned down to size, and the passes cut in to fit the templet supplied 
 by the roll designer. When one roll is completed, it is placed in chocks 
 higher up in the housing, and the second roll is placed below it, where it 
 may be turned with the finished roll as a guide, so that the two parts of the 
 passes may be made to fit exactly. With ordinary tools, chilled rolls are 
 seldom turned with a surface speed of more than fifty-six inches per minute, 
 but with tools made of high speed tool steel this speed may be increased 
 to seventy-two inches per minute. Speeds twice as great as these may be 
 employed for turning the other kinds of roll. 
 
 Dressing the Rolls: After a set of rolls has been in service a variable 
 length of time, the passes become worn to such an extent that they no longer 
 produce the section to the required dimensions, and they must then be 
 replaced by another set. In most cases these worn out rolls may be turned 
 again, or dressed down, so as to give the correct size once more, or if the 
 section is of such shape that this refitting is impossible, the passes may be 
 enlarged to produce a section similar in shape to the first one but of greater 
 weight. This wearing of the rolls is one reason why rolling tolerance is 
 required on all materials. 
 
330 THE ROLLING MILL 
 
 Types of Mills: Before proceeding farther it may be well to explain 
 that there are two main types of mill, referred to as Two=high and Three= 
 high mills. As the names indicate the classification is based on the manner 
 of arranging the rolls in the housings, a two-high stand consisting of two 
 rolls, one above the other, and a three-high having three rolls thus arranged. 
 In all three-high mills, each roll revolves continuously in one direction only, 
 whereas in two-high mills the direction of the rolling may be in one direction 
 only, or in opposite directions at different intervals, in which case they 
 are called reversing mills. 
 
 In the old days before the invention of the three-high mill or the 
 reversing engine, if it was desired to pass the bar more than once through 
 the same stand of rolls, the catcher returned the piece to the roller by 
 placing it on the top of the upper roll, which carried it in the direction 
 opposite to that in which it moved at the bottom of the roll. Mills in 
 which this practice prevailed were called pull=over or drag=over mills and 
 are to be looked upon as the fore-runner of the reversing mill. In the first 
 mill of the reversing type a ratchet gear furnished the means for reversing 
 the mill. Pull-over mills are still in use, and are the mills most often 
 employed for rolling sheets. Another kind of two-high mill is the continuous 
 mill, which consists of several stands of rolls arranged in tandem and 
 propelled with a single engine. Guide, loop and the so called Cross 
 country mills are made up of several two-high stands and one or more 
 three-high stands. Guide mills are small hand mills consisting of several 
 stands of rolls in a train. They take their name from their having metal 
 guides to support the piece as it enters the various passes. In many guide 
 mills it is the practice of the catchers, in order to save time, to start the 
 piece through each of the passes before it is through the preceding one, 
 thus forming a loop. After the institution of this practice it was found 
 that the loop could be made by means of a tube or trough, called a 
 repeater, and thus dispense with the catchers. Such a contrivance is a 
 part of many modern bar and strip mills. The cross country mill is made 
 up of several stands of rolls, arranged in trains or trains and tandem sets. 
 The bar, propelled mechanically by means of live rolls, transfers, etc., 
 must reverse its course two or more times to pass through the various sets 
 of rolls from the furnace to cooling tables. These mills represent one of the 
 latest and most efficient types. Combination mills are those in which the 
 roughing or major part of the reduction is done in continuous rolls and the 
 shaping in a guide or loop mill. The Universal Mill is one, which, in addi- 
 tion to the horizontal rolls, usually arranged two-high but occasionally three- 
 high, is provided with vertical rolls, all set in one housing. These mills origin- 
 ally contained but two vertical rolls on one side only of the horizontal rolls, but 
 in modern mills there are two sets of vertical rolls, one set on either side of the 
 horizontal ones. The mill is used for rolling plates and eye bars that 
 require rolled edges. Besides these types, there are many special mills, 
 usually named from the inventors, such as the Gray mill for rolling beams 
 
THE CHOCKS 331 
 
 and H-sections; the Wenstrom mill, a kind of universal mill for rolling bars; 
 Sack's mill for rolling shapes, also a development of the Universal mill; and 
 the Schoen mill, which rolls car wheels. Opportunity will be given later to 
 become better acquainted with most of these mills. 
 
 SECTION II. 
 
 PARTS OF THE MILL ESSENTIAL TO THE OPERATION OF THE ROLLS. 
 
 The Chocks : As previously indicated, the chocks furnish the bearings 
 in which the necks of the rolls turn. They are usually made in two parts. 
 The surface in contact with the neck is made of brass, bronze, or white 
 metal, which can be replaced as necessary. The use of these alloys is 
 necessary in order to reduce friction, which is much less between metals of 
 different kinds due to difference in size of the molecules or grains, and to the 
 tendency of the softer metals to flow. The approximate composition of 
 some of the more common of these alloys is given in the sub-joined table 
 of analyses. 
 
 Table 49. Composition of Bearing Metals. 
 
 NAME OF ALLOYS % COPPER % ZINC % TIN % ANTIMONY % LEAD 
 
 Red Brass 85 15 .' 
 
 Yellow Brass.... 65 35 
 
 Bronze, No. 1 ... 85 . . 15 
 
 Bronze, No. 2. .. 82 15 3 
 
 White Metals .... to 6 10 to 15 12 to 20 65 to 80 
 
 Since 1915, a new white metal composed of lead, about 98.5%, and 
 sodium, about 1.5%, has been used with much success. As all these metals 
 are soft and not very strong, it is necessary to carry them in castings, which 
 are set into the housings. These castings are box-like in shape, each one 
 containing on one side a semi-circular groove corresponding to, but larger 
 than, the necks of the rolls. In order to reduce their weight, they are cored 
 out, and may be made of either iron or steel. 
 
 The Arrangement of the Chocks in two-high mills is a simple matter. 
 Two chocks under the necks of the bottom roll, and two similarly placed 
 above the top roll furnish the main bearings. In case the top roll is 
 adjustable, light bearings must also be placed under its necks to make it 
 possible to support this roll. In the heavy mills hydraulic jacks or balance 
 weights, placed under the mill, are connected by vertical rods to the lower 
 chocks and serve to lift the roll as desired; in the small mills, screw 
 bolts extending through the housing serve the same purpose. The exposed 
 half of the neck of the lower roll will usually be covered to protect it from 
 scale, etc. The arrangement of the chocks in three-high mills is more 
 
332 THE ROLLING MILL 
 
 difficult. The simplest way is to place double groove chocks between the 
 top and middle and the middle and bottom rolls, and then set them in the 
 housings one above the other, so that all the adjusting made necessary by the 
 wearing away of the brasses and the material of the rolls, themselves, may 
 be made with the large set screws in the top of the housing. But this 
 arrangement causes the bottom bearing to wear down rapidly and increases 
 the power required to drive the mill, due to the additional friction induced 
 on this bearing by the weight of the two upper rolls and their chocks. 
 This fault may be overcome in two ways: (1) By making the bottom 
 roll fixed and supporting this extra weight on the shoulders of the chocks 
 themselves, the distance between rolls may then be regulated with shims, or 
 "liners," by adding or removing the shims as the bearings wear down. 
 (2) A better way, and the one most often employed in modern mills, is 
 to make the middle roll fixed, in which case the bottom roll is raised and 
 lowered by means of an adjusting wedge attached to a screw in the 
 housing which permits it to be moved back and forth with a wrench from 
 the outside of the housing. Other methods of adjusting this roll are in 
 use also. A method of supporting each roll separately by means of hooked 
 screws and cross bars has also been developed, the details of which would 
 be unprofitable to study here. In all mills, two-high as well as three-high, 
 the top chocks are held down by means of two strong screws which work in 
 threaded holes or nuts in the tops of the housings. 
 
 The Function of the Chocks is not only to furnish bearings for the 
 rolls vertically but to prevent their movement laterally as well. This 
 lateral displacement of the roll is prevented by the inner edge of the bearing 
 which is formed to fit against the shoulder of the roll. Adjustments for 
 wear in this direction are provided for by adjusting screws which extend 
 through the side of the housing and bear on the ends of the chocks. This 
 lateral adjustment is a matter of great importance in rolling sections that 
 require grooved rolls, the reason for which is self evident. 
 
 The Housings: There are two housings for each stand of rolls, they 
 may be made of either iron or steel, the choice of materials depending 
 upon the size of the mill, the strength required, and the preference of the 
 management. They are castings of an O-orU-form, each enclosing a space, 
 called the window, which serves as a receptacle for the chocks. Housings 
 may be either closed topped or open topped; in the former, the base, 
 the two legs, and the top are all cast in one piece, while in the latter the 
 top may form a separate part which can be removed. The base of the 
 housing is cast with a projection on each side, the two forming the feet 
 of the housing. In the bottom of each foot is cut a groove which fits over 
 a girder, called a shoe, running parallel to the rolls. Suitably shaped 
 bolts then serve to clamp the foot of the housing to the shoe, which is 
 firmly fastened to the foundation by means of long bolts. This method 
 permits the housing to be moved laterally, and much facilitates the plumb- 
 
HOUSINGS AND PINIONS 333 
 
 ing and lining up of the mill. The tops of the two housings in a set are 
 prevented from spreading apart by means of suitable tie rods, or the tops 
 of both housings may be cast in one piece. Similarly, tie rods will usually 
 be placed at the bottom. Recesses or other openings are cast in the 
 inside of each housing to receive the supports for the guards and guides/ 
 these supports being usually in the form of square bars which extend from 
 housing to housing in front of the rolls. The immense pressure applied to 
 the rolls between the top and bottom of the housing acts as a stretching 
 force on the uprights of the housings, and is an important factor in deter- 
 mining the reduction that can be effected in one pass and also the exactness 
 with which the thickness of the piece is controlled. 
 
 The Adjusting Equipment for the rolls has already been located and 
 partly described in the preceding paragraphs. In addition it should be 
 pointed out that in large mills, in which the top roll is adjusted during the 
 rolling, power must be supplied to operate the screws. To provide for the 
 transmission of the power, the top part of each screw, which is made square 
 or hexagonal for a distance slightly greater than the rise of the roll, passes 
 through the core of a pinion. These pinions may then be turned directly or 
 indirectly with a horizontal hydraulic cylinder located at a proper height, 
 usually on top of the housings of the driving pinions; or, by means of a 
 worm shaft and the proper worm gears, the screw down may be effected 
 with a small electric motor. In small mills where the adjustment is only 
 occasional, the screws will be operated by hand by means of spanner 
 bars. In all cases the compression of these screws is unavoidable and 
 combined with the stretch of the housings produces the spring of the mill, 
 which in some cases is surprisingly great. 
 
 The Pinions: An important part of the mill is the pinions. They 
 arc broad faced steel gears located between the prime mover and the rolls. 
 Their functions are to divide the power, which is delivered by the engine 
 or motor through a single shaft or driving spindle, usually spoken of as the 
 leading spindle, among the rolls and to control their direction of rotation. 
 They run in bearings contained in a pair of housings similar to those for the 
 rolls, and should be completely and tightly covered to protect them from 
 dust and dirt which would cause them to wear out rapidly. They need 
 to be well lubricated, and the present practice of giving them an occasional 
 dressing of pine tar, plumbago, and tallow, or other mixture of grease, is 
 giving way to the better plan of having the housings cast in one piece so 
 as to form an oil bath at the bottom in which the bottom pinion is partly 
 submerged. Pinions are of three kinds, based on the arrangement of the 
 teeth. In the oldest form the teeth ran straight across the face, but 
 eventually it was found that a smoother running pinion results if the face 
 be divided into two parts and the teeth of the two halves staggered, i. e., 
 set in so that the teeth in one half are in line with the space in the other. 
 This design gives an effect like that which would be obtained if the pitch 
 
334 THE ROLLING MILL 
 
 were decreased. This scheme was also found to effect a saving in power. 
 Still another improvement results from the use of pinions with helical or 
 "herring bone" teeth, which also tend to eliminate vibration in the pinions, 
 as some parts of the teeth are always in contact, thus making the trans- 
 mission of the power continuous. This presence of jar when each tooth 
 comes into action has an effect on the material, as in certain classes of 
 material the old form of pinion was found to produce marks on the bar by 
 the jarring of the teeth meshing being transmitted to the rolls. In all 
 mills except plate mills, the distance from center to center of the pinions 
 determines the size of the mill. 
 
 The Connections: Each roll, except in the case of friction driven 
 rolls, is connected to its pinion by means of spindles. They are usually 
 made of cast steel and are fitted at each end with wobblers like those on 
 the rolls. The connections between pinions and spindles and rolls and 
 spindles are made with coupling boxes. The coupling box is a hollow 
 cylindrical casting, the space in which corresponds in section to that of the 
 wobbler, one end of the box fitting over the wobbler on the roll and the 
 other over that of the spindle. In order to safe-guard the mill, the coupling 
 boxes are usually made the weakest part of the mill. In some mills this 
 weak spot is the leading spindle, which connects the pinions with the engine 
 or motor, instead of one of the coupling boxes. Since the spindle must be 
 put in place with the two coupling boxes on it, the length of the spindle 
 must be a little more than twice the length of the box. In mills, the top 
 roll of which moves up and down through a great distance, the upper 
 spindle is thrown out of line horizontally. As it is very difficult to 
 operate with a spindle more than 15 out of level, this angle must be kept 
 within the allowable limit by increasing the length of the spindle. In 
 such cases the ends of the wobblers are cut from a section of a sphere to 
 give them the rounded form necessary to permit them to work at different 
 angles, and the spindle is supported by means of saddles which rise and 
 fall with the roll and hold the spindle in place. 
 
 Guides and Guards: In order to prevent collaring and to insure 
 that the piece enters and leaves its pass in the correct position, guides 
 are employed. These guides vary in form and size to fit the conditions. 
 In some cases they are merely grooved fore-plates; in others they are blunt 
 edged plates set up in front of the collars, dividing the space in front of the 
 rolls into a series of pigeon holes; in large mills rolling heavy sections, 
 they may take the form of grooved rollers; in the smaller mills like the 
 guide mills, they are trumpet shaped castings that fit close up to the roll 
 and have exit openings to conform to the shape and size of the section 
 of the entering piece; in other mills, like the continuous mill, they may 
 be so constructed as to twist and thus turn the piece between two successive 
 passes. Guides may be employed on both sides of a pass, in which case 
 they are designated as entering guides and delivery guides. They are held 
 
ROLL TABLES, ETC. 335 
 
 in place by means of the rest bars previously mentioned in connection with 
 the housings. Guards are devices employed mainly on the delivery side of 
 the mill to control the direction of the piece after leaving the pass. Reversing 
 and three-high mills will be provided with guards on both sides of the mill. 
 
 Additional Equipment: In addition to the parts of the mill already 
 described, every mill must be provided with suitable appliances for heating 
 and handling the materials and disposing of the product. The various 
 furnaces for the heating of the raw materials will be described later. As 
 to the handling of materials, it is evident that reliance on man power places 
 such restrictions upon the size and output of the mill that, in the case of 
 certain small mills and of all the larger mills, the appliances for handling 
 the materials mechanically are to be considered as essential parts of the 
 mill. The number of these appliances is so great and the kinds are so 
 varied that a detailed description of all of them is impossible, and since 
 it would be unprofitable to describe only a few forms, little more than an 
 attempt to mention some of the more important ones will be made here. 
 For getting heavy material in and out of furnaces and delivering it to the 
 mill, electrically operated charging and drawing machines are used. These 
 machines are of two general types, namely, those that travel on overhead 
 tracks and those that move on a track laid on the mill floor. For handling 
 material during the rolling process, roll tables are provided for large mills, 
 while various forms of appliances, called repeaters, are used on small mills. 
 Roll tables consist of a frame work and a number of rolls, which may be 
 rotated at will, mounted thereon. For operating these rolls, the steam 
 engine has been replaced by the electric motor in all new mills and also 
 in most of the old ones. Roll tables may be either stationary or movable. 
 Movable tables, often used on three-high mills, may be of the tilting, the 
 lifting, or the traveling type. Tilting tables are mounted on an axis of 
 rotation, which may permit one or both ends, depending on the location 
 of the axis, to be raised and lowered, whereas lifting tables always remain 
 in horizontal positions and may be moved in up-and-down directions only. 
 Traveling tables, which are now to be found only on the old mills, move 
 along on tracks laid on either side of a roll train, and may be of the tilting 
 type also. After the rolling, cooling beds, of which there are many types 
 and forms in use, must be provided to receive the material from the mill. 
 The straightening of the material, which irregularities of the rolling and 
 cooling often makes necessary, is done in roll, or machine, straighteners 
 or by means of gag presses. In order to keep up with the mill only the most 
 rapid methods of cutting are permissible. For this purpose only two 
 instrumentalities are available, namely, the saw and the shear, either of 
 which may be used on hot or cold material. Shears are of three general 
 types; namely, the alligator, used only for cutting light weight material; 
 the guillotine, which is employed generally for all classes of work; and the 
 flying shear, designed to cut billets or bars while they arc in motion. The 
 
THE ROLLING MILL 
 
 power employed to operate the shears is hydraulic for the heavier materials, 
 such as slabs and large blooms, while steam and electric power are used 
 for all other work. 
 
 SECTION III. v 
 
 SOME GENERAL FEATURES PERTAINING TO OPERATION OF THE ROLLING MILL. 
 
 The Mill Force: Of equal importance with the equipment of a mill, 
 are the men who operate it and the organization and system back of them. 
 Under the general superintendent of the steel plant there may be a number 
 of rolling mill superintendents, each of whom will have charge of a group 
 of mills turning out similar products. As his assistants, the mill superin- 
 tendent selects foremen, each of whom are responsible for the successful 
 operation of one or two of the mills. Below the foreman the mill is divided 
 into departments, with a man at the head of each, who is charged with 
 the performance of a certain part of the work. Thus, there is the heater 
 who has the heating of the material to look after; the roller, who superin- 
 tends the actual rolling process; the engineer who tends the engine, or an 
 electrician, if motors are used for running the mill; and the shearmen, 
 whose duty is to see that the product is properly cut. Besides these, other 
 departments, such as the machine and the electric shop, the inspection and 
 shipping departments, play important parts in the mill operation, though 
 they do not come under the direct authority of the mill superintendent. 
 When it is remembered that the failure of any one of these may close down 
 the whole mill, the importance of system and of the personnel of the organi- 
 zation is more fully appreciated. 
 
 Duties of the Roller: So far as the product of a given mill is con- 
 cerned, it would appear that the roller and roll designer are the chief figures. 
 Co-operation between these two men is essential, for in a measure their 
 interests are identical; the roll designer decides how the work is to be 
 done, and the roller sees that it is done properly. The latter will, therefore, 
 concentrate his attention upon the product, and with caliper, gauge, or 
 templet will take frequent measurements to make certain the material is 
 being rolled true to the dimensions specified. He will keep a sharp look-out 
 for underfills, overfills, fins, guide marks, collar marks, laps and any other 
 rolling defects, and make the necessary adjustments to correct them. 
 
 Fins: It is the intention to discuss the defects of materials in con- 
 nection with the rolling of each particular class of product, but in all rolling 
 where grooved rolls are used, the occurrence of fins is so liable to happen 
 that it is well to consider them here, more especially since there will be 
 occasion to use the term frequently. Fins are formed when the section is 
 too large for the pass it is entering, or whenever, in designing the pass, 
 proper allowance has not been made for the spread of the material, thus 
 
OPERATING FEATURES 337 
 
 causing the metal to flow out between the flat bodies of the rolls on each 
 side of the groove. If this fin is thin and wide it will be folded over without 
 welding and form a lap, when the piece, after turning, has been sent through 
 the next pass. Besides, fins may be dangerous, for if the rolls are very 
 close together any spreading of material between them is likely to break 
 them. 
 
 The Different Passes and Stands in mills that roll finished shapes 
 are given class names. Thus the first rolls the piece enters in the mill are 
 used mainly to reduce the size of the bloom or billet, and the piece generally 
 leaves them in the same shape it entered. These passes are called the 
 roughing rolls and the stand or stands is spoken of as the rougher, or 
 roughers. If the succeeding stand merely carries this reduction further, 
 it is called the pony rougher. The stands and passes in which the actual 
 shaping of the piece is done are called the strands, usually numbered 1, 2, 
 etc. The pass next to the last is called the planisher, but since in roll 
 designing this pass may be looked upon as the first pass leading back from 
 the finished section to the bloom, some designers call this pass the leader. 
 The last pass is always called the finishing. 
 
 Factors Affecting the Rolling Operation: In the rolling of steel 
 there are five factors to be considered, namely, the temperature of the 
 steel during the rolling, the chemical composition of the metal, the speed 
 at which the rolls are revolved, the draught in each pass, and the diameter 
 of the rolls. Furthermore, these factors should be considered from the 
 three different standpoints of power, or energy, required to deform the 
 steel; their effect upon the rolling properties of the metal, that is, the way 
 it will spread, bend and flow in the rolls; and their effect on the quality of 
 the finished product. All these matters have not been fully investigated, 
 and our knowledge concerning them is somewhat meager, but in order to 
 invite attention to these subjects, a brief summary of what is known about 
 these factors is appended. 
 
 Effects of Temperature: The influence which the working of steel at 
 different temperatures may have upon the quality and properties of the 
 product has already been discussed under the caption of Hot and Cold 
 Working, (Chap. II, Sect. 1.). Relative to the power or energy require- 
 ments and the rolling properties of the metal, it is to be observed that the 
 higher the temperature is raised the more plastic the steel becomes. Thus, 
 while, a .10 per cent, carbon steel, for example, will give a tensile strength 
 of about 50000 pounds at atmospheric temperatures, at 600C it will break 
 under a pull of 20000 to 25000 pounds per square inch, . at 700 C under a pull 
 of about 11000 pounds, and at 800 C under a pull of about 6000 pounds. 
 Between 800 C and 900 C a distinct discontinuity in the tensile strength 
 occurs, with the result that at 900 C the tensile strength will suddenly 
 increase to nearly 9000 pounds. From this point the strength decreases 
 
338 THE ROLLING MILL 
 
 with rising temperature, being about 6500 pounds at 1000C., about 4600 
 pounds at 1100C., about 3000 at 1200C., and approaching zero at 1460C., 
 the fusion point. From these facts it would appear that the higher the 
 temperature of the steel the easier will it be deformed. But there are other 
 features that tend to keep both the initial and final working temperatures 
 within certain well defined limits. Since steel assumes a semi-fluid state 
 at temperatures somewhat below its fusion point, heating to within 
 less than 200 C of this point exposes it to the danger of overheating or 
 so-called "burning". For dead soft steels the initial temperature should 
 not exceed 1250C, and for high carbon steels, (1.00% to 1.20% carbon) 
 this temperature should not exceed 1050 C. In order to secure the greatest 
 refinement of grain, either the initial temperature or the speed of rolling 
 should be adjusted so that the finishing temperature of the rolling will be 
 above, but as near the critical range of the steel as possible. 
 
 The Effect of Chemical Composition need be considered here only 
 from the standpoint of energy required and rolling properties. As to the 
 energy required, experiments have shown that slightly more work is 
 required to roll a steel containing 1.00% carbon than for steels containing 
 only .10% carbon. Whether this difference was due entirely to the lower 
 temperature at which the higher carbon steel was rolled or also partly to 
 the higher content of carbon could not be determined. In the hope that 
 it would help to solve this question, an experiment was performed with 
 the object of comparing the tensile strength of high carbon and low carbon 
 steels at rolling temperatures. For this purpose three steels having a 
 carbon content of .10 percent, .22 per cent, and 1.10 per cent, but otherwise 
 of approximately the same composition, were selected. The results from 
 pulling the first have already been given. The mechanical properties of 
 the other two were compared at 900 C only. The average results obtained 
 from pulling ten pieces of each under similar conditions at the initial 
 temperature of 900 C are as follows: 
 
 
 
 Elongation Reduction 
 
 Carbon Content 
 
 Tensile Strength 
 
 in 8* of Area 
 
 .22% 
 
 13,500 Ibs. 
 
 110% 94% 
 
 1.10% 
 
 18,800 Jbs. 
 
 58% 83% 
 
 These results would indicate that the higher carbon steel is somewhat less 
 plastic at rolling temperatures than the lower carbon steel. Therefore, it 
 would require more energy for rolling, and would not spread or elongate 
 as readily as the steel of lower carbon content. As to the effect of the 
 other elements in plain steel, phosphorus may produce effects similar to 
 those of carbon, sulphur tends to produce red shortness, while manganese 
 tends to offset the effects of sulphur and oxygen and improve the rolling 
 properties. ' Open hearth steel, which is low in its phosphorus content, 
 tends to spread more in the rolls than does Bessemer steel, which is higher 
 in phosphorus. The rolling properties are still more strongly affected by 
 
OPERATING FEATURES 339 
 
 certain alloying elements, such as nickel and chromium. The difference in 
 rolling properties produced by difference in chemical composition may not 
 be noticable in the rolling of the simpler sections, but may cause much 
 trouble in the rolling of complicated sections with wide thin flanges or legs. 
 Thus, in one instance of such a complicated section, it was found that while 
 the section rolled perfectly with one heat of steel, it was imperfectly formed 
 when rolled from another heat in which the carbon content was ten points, 
 the manganese fifteen points, and the sulphur two points higher. 
 
 The Effect of Speed : The speed of rolling is the factor which has received 
 the least attention from the viewpoint of its effect upon the material. It is 
 the consensus of opinion among steel workers, however, that the speed of 
 rolling undoubtedly has an influence upon the quality of the product. It is 
 evident that the faster a piece of steel is deformed the less time the 
 molecules have in which to adapt themselves to the deformation and the 
 greater their resistance to the deformation. Consequently the stretching 
 effect of the rolling, as previously explained, increases in undue proportion 
 to the compression. If the speed of the rolls were increased sufficiently 
 they would then have a greater tendency to slip on the piece, and the 
 stretching effect would tend to become a tearing effect. However, it is 
 not speed alone that produces this effect but draught and speed together. 
 
 Draught: Draught is the difference in sectional area between one 
 pass and the next succeeding one, and is usually expressed in per cent. 
 While there are cases where as high as a 50% reduction occurs, and one 
 case in which a 70% reduction is made in a single pass, the draught will 
 seldom exceed 36%. These heavy draughts take place in the roughers, 
 where most of the reduction occurs. In the strands the reduction will be 
 as evenly distributed as possible, and will sometimes be as low as 10%. 
 If the leader is used as a planishing pass very little reduction is effected 
 there either, while very little reduction, with a few exceptions, ever occurs 
 in the finishing pass. This pass being intended to give an exact finish to 
 gauge and to true up the piece, it is important that very little work be done 
 in it in order that it may be subjected to as little wear as possible. Since* 
 a piece must be finished before it has lost its heat and has cooled below the 
 rolling temperature, which cooling is very rapid in the case of small sections, 
 the rolls must be run at a speed that will carry the piece through the rolling 
 before it has become too cold. The number of passes and the size of the 
 piece to start with controls the draught. Aside from these features the 
 desire to increase output acts as an incentive to increase both the speed 
 and the draught to the limit the material will stand. There are many 
 considerations, however, that operate to hold down both speed and draught 
 below that which will do injury to the steel. One of these is the additional 
 power required for very rapid reduction; another is the severe strain on the 
 rolls and other machinery when the piece enters the rolls at high speed 
 and with large draught. In mills composed of several stands, especially 
 
340 
 
 THE ROLLING MILL 
 
 iti the case of the continuous mill, the speed of all preceding stands is deter- 
 mined by the speed of the finishing stand. In hand mills, the speed is 
 restricted to the highest velocity, about six hundred feet per minute, at 
 which the catchers can grasp the piece with the tongs. The magnitude of 
 the draught is restricted by the limiting angle at which the rolls will "bite', 
 the piece on entering. This angle is found, by experience, to be about 30' 
 
 FIG. 54. The Limiting Angle of Rolling. 
 
 Above this angle the resultants of the forces of compression have receded 
 so far from a parallel to the line joining the centers of the rolls that they 
 exert a push on the piece, and the resultant of the forces of elongation is so 
 nearly vertical that the horizontally inclined component due to friction 
 only is not sufficient to balance this backward push and drag the material 
 between the rolls. In order to increase this limit, a series of horizontal 
 and well rounded grooves, called ragging, are often cut in the surface of 
 the roll, giving it the appearance of a half formed cog wheel. Since these 
 grooves leave ridges in the material, they can be resorted to only in bloom- 
 ing mills, billet mills, or roughing stands. Even then the grooves must be 
 cut with considerable care in order to prevent these ridges being folded 
 over into laps in succeeding passes and rolled into the material, to appear 
 as seams in the finished product. 
 
 The Effect of Diameter of Rolls: From a study of Fig. 54, it will be 
 seen that the larger the roll diameters are the greater will be the draught 
 that may be taken without exceeding the limiting angle of rolling. For the 
 same draft, however, a large roll gives a greater roll surface area in contact 
 
EFFECT OF SIZE OF ROLLS 341 
 
 with the metal than a small one and therefore requires more pressure to 
 force it into the metal, thus putting a greater tension on the housings and 
 requiring more energy to drive it. The large roll gives an affect more like 
 that of pressing than the small roll, and, with the draft and speed properly 
 regulated, the effects of the rolling can be made less superficial with the 
 large roll. The large roll tends to cause the metal to spread more than 
 the small roll. Hence, the size of the rolls is a factor to be considered in 
 designing rolls for flats and other products in which the spread of the metal 
 may affect the dimensions of the finished article. 
 
342 PREPARATION OF STEEL 
 
 CHAPTER IV. 
 
 PREPARATION OF THE STEEL FOR ROLLING. 
 
 SECTION I. 
 
 INGOTS AND THEIR DEFECTS. 
 
 Preparation of Ingots: In order that the large bodies of metal 
 refined at one time by the various methods of steel making may be obtained 
 in a convenient shape for rolling, it is necessary that these large bodies 
 be divided into smaller ones, called ingots, of a uniform shape and size. 
 These conditions are obtained by pouring the metal while it is still molten 
 into moulds of the desired dimensions, where it may be allowed to solidify 
 in part or in whole before the mould is removed. Before rolling begins, 
 however, the ingot must have been allowed to solidify throughout, and the 
 whole mass should be of uniform temperature. But in cooling naturally, 
 these conditions are not fulfilled, because the outside of the ingot, being 
 the part from which the heat is removed the most rapidly, is the first to 
 solidify. With this fact in mind, it is easily understood how, in any case 
 of natural cooling, the interior is the last to drop to any given temperature. 
 In fact, the moulds are stripped from many ingots while the central portion 
 is yet in the liquid state. This fact was early recognized by steel workers, 
 and so it was originally the custom to strip the ingots as soon as possible 
 and place them in a tightly covered hole or pit in the ground, where the 
 heat from the interior of the ingot was slowly conveyed to the outside by 
 conduction, and sufficed not only to heat up the colder exterior part of the 
 ingot but also to supply heat to the pit, which, with careful manipulaton, 
 was sufficient to maintain a rolling temperature. This process was called 
 soaking, hence the name soaking pit. In order to bring the soaking under 
 better control and make it adaptable to varying conditions, means for 
 supplying additional heat was introduced, so that the modern soaking pit 
 is in reality a kind of heating furnace, a detailed description of which will 
 be given later. 
 
 Ingot Defects: A prerequisite to faultlessly finished material is 
 perfect ingots, and by a perfect ingot is meant one free from all cavities 
 or openings and made up of material that is homogeneous throughout. 
 Unfortunately, the natural laws that govern the solidification of the liquid 
 metal operate against both these requirements, and develop the well known 
 natural defects in ingots called piping, blow holes, segregation and crystal- 
 lization. Added to these are other defects, both incidental and accidental, 
 
INGOT DEFECTS 343 
 
 such as checking, scabs, and slag inclusions. A brief discussion of these 
 defects follows; but an understanding of their causes requires a study of the 
 laws that control the cooling of ingots. 
 
 The Nature of the Cooling of an Ingot: The ingot moulds in common 
 use are tall box-like shapes made of cast iron; they are open at both ends, 
 one of which is a little smaller than the other to give the mould a little 
 taper; and have a square or rectangular section slightly rounded at the 
 corners. In use, one end of the mould, usually the larger, is closed by 
 the stool on which the mould stands in an upright position. Immediately 
 the molten steel is poured into this mold, the metal next to the mold and 
 stool is chilled by contact with the cold surfaces and solidifies on the bottom 
 and sides to form what is called the skin of the ingot. As more and jnore 
 heat is absorbed by the mold, this skin grows in thickness, but due to the 
 increase in the temperature of the mold and 'the insulating effect of the 
 skin, itself, it grows at a rapidly reduced rate, until the process becomes 
 so slow that it can be considered as a normal cooling. The cooling then 
 takes place by a dissipation of the heat through this skin along lines per- 
 pendicular to the surface of the solidified shell, which acts as the conductor, 
 with the result that this shell gradually grows in thickness, the growth 
 progressing toward the center until all the metal is in the solid state. 
 The laws of freezing which the metal obeys, combined with this manner 
 of freezing, gives rise to the natural defects enumerated above. 
 
 Pipes: One of the most noticeable effects of the freezing is the pro- 
 duction of a more or less cone-shaped cavity at the top of the ingot, known 
 as the pipe. Pipes are the result of the contraction of the metal on 
 solidifying in the manner just described. This contraction amounts to 
 about two-hundredths of the linear dimensions of the ingot, and if the 
 manner of cooling did not set in play forces which oppose the contraction, 
 no pipes would form, and a perfect ingot would measure one-fiftieth smaller 
 than the mould in all its dimensions. In ordinary ingots much of 
 this difference in volume is represented by the pipe. Since the skin and 
 the more slowly formed walls built up by the cooling are rigid, the void 
 left by contraction is filled by the metal in the central portion that 
 still remains fluid, the force of gravity directing the flow downward at 
 all times. After solidification of the metal is complete, further contraction on 
 cooling tends to open this pipe still farther towards the bottom, because 
 the exterior, being the colder, is the more rigid and is capable of 
 stretching or tearing the more plastic interior. The greater portion of 
 the surface of this cavity is likely to become more or less oxidized, 
 and, since the oxidized portion is not welded up in the rolling, the pipe 
 will appear in the smallest rod or wire into which this part of the ingot 
 may be rolled. Aside from injuring material, pipes are liable to cause 
 accidents in rolling, so the steel maker is very anxious to get rid of them 
 as early as possible. 
 
344 
 
 PREPARATION OF STEEL 
 
 PIG 55. Split Ingots Showing Various Forms and Degrees of Pipe. 
 
INGOT DEFECTS 
 
 345 
 
 FIG. 55 Continued. 
 
346 PREPARATION OF STEEL 
 
 Methods of Reducing Waste due to the Pipe: Obviously, the only 
 way of avoiding this pipe is by discarding the portion of the ingot affected. 
 Various schemes for reducing the waste due to this cause have been and 
 are being tried, and some of them are fairly successful, among which the 
 most promising seems to be the so-called hot top mould. As explained 
 in connection with the open hearth process, in one form of these moulds 
 the ingot is cast with the smaller end down, while the larger end is sur- 
 mounted with a short mould which is lined with refractory and non-con- 
 conducting material, such as clay. This lining reduces the size of the top 
 section and keeps the top of the ingot in the molten state until the ingot 
 proper has solidified. Thus, the pipe is brought up into the cope, or sink 
 head, which is of much smaller section than the ingot, and the waste due to 
 the cropping is decreased accordingly. This ingot is stripped by first 
 removing the insulated top section, gripping the sink head with tongs and 
 then lifting the ingot out of the mould. In a patented form of this mould, 
 known as the Gathman mould, a similar effect is produced by decreasing 
 the thickness of the mould at the top. Since the heavy part of the mould 
 causes a more rapid cooling than the thin portion, the metal at the top is 
 the last to freeze. 
 
 Blow Holes: In the molten state iron, or steel, is capable of dissolving 
 large volumes of gases, such as oxygen, carbon-monoxide, nitrogen and 
 hydrogen, this solvent power increasing with the temperature. The iron 
 probably unites with all the oxygen immediately it is dissolved, hence it 
 is retained by the metal in the solid state if chemical means are not 
 employed to remove it. In case of the other gases, however, no such stable 
 combination takes place, and they are largely thrown out of solution just 
 previous to the time when solidification of the metal occurs. As the metal 
 is in a more or less plastic condition at this time, the last gases thus liber- 
 ated may not be able to escape from the body of the metal, in which case 
 they collect in bubbles, as a gas will in making its way out of any fluid. 
 Each bubble will then form a small cavity in the metal which is known as 
 a blow hole. These holes will vary in size from those visible through a 
 microscope to large pockets, the dimensions of which can be measured in 
 inches. The smallest ones are liable to occur just below the skin of the 
 ingot where the rapidly cooling metal gave the tiny bubbles of evolved 
 gases time neither to escape nor to collect in larger bodies. Here the gases, 
 unable to escape upward on account of the very viscous nature of the 
 metal, form tube-like cavities that extend at right angles to the skin wall 
 of the ingot and toward the center. In the rolling of the steel, the blow 
 holes are closed up and welded together provided their surfaces have not 
 been oxidized, in which case they will not weld arid will produce defects 
 in the finished articles. Blow holes near the center of the piece, known 
 as deep seated holes, are less liable to oxidation, hence are the least harm- 
 ful. But the small blow holes beneath the skin of the ingot are liable to 
 be exposed to the air, or be filled with liquid oxide of iron, in which case 
 
INGOT DEFECTS 
 
 FIG. 56. Transverse Sections of Ingots Showing Blow Holes 
 
348 PREPARATION OF STEEL 
 
 they produce seams in the finished articles. Blow holes are an ever 
 present menace. But correctives may be employed successfully, and they 
 are seldom a source of serious damage in steel that has been properly 
 worked in the process of manufacture and thoroughly deoxidized at the 
 time of recarburizing. In this respect the use of aluminum in the mold at 
 time of casting has been found to be very effective. Blow holes have the 
 effect of reducing the size of the pipe, and on this account are to be de- 
 sired if they are deep seated. Various attempts to overcome both blow 
 holes and pipes mechanically by means of subjecting the metal to compre- 
 ssion while in the molten state have been tried, but the expense of 
 operating these appliances more than outweighs the good derived, 
 especially since the steel discarded on account of pipe is available for use 
 as scrap in the open hearth. 
 
 Crystallization is the property possessed by iron, in common with 
 many other substances, of forming crystals on solidifying. The size of 
 the crystals depends on the composition of the steel and the rate of cool- 
 ing; in general, the slower the cooling the larger the crystals will be. It 
 is plain that the temperature at casting and the size and shape of the 
 ingot and mold control the rate of cooling. If the crystals are large, the force 
 of cohesion among the crystals is decreased by the increased area of contact 
 and the larger size of their cleavage planes. The effect of unduly large 
 crystals is to make ingots liable to tear in rolling. This condition is 
 sometimes called ingotism. The deformation and refinement of the 
 crystals in rolling prevents their effect showing up in finished product that 
 has been properly worked. 
 
 Segregation: Steel is a mixture of various compounds and elements; 
 some of these are to be looked upon as impurities because of their detri- 
 mental effects, but others are necessary to impart the properties most 
 desired in the product. While in the molten state these solid ingredients, 
 like the gases just mentioned, are held in solution by the iron, a power 
 which it does not possess to the same degree at temperatures below its 
 freezing point. Some of these ingredients freeze at a lower temperature 
 than the iron. Furthermore, the solution, following the laws of selective 
 freezing, undergoes a series of changes and recombinations with the formation 
 of various eutectic solutions, which increase the number of substances that 
 solidify normally at much lower temperatures than pure iron. With such 
 an aggregate, it is easy to see how the process of solidification results in 
 an isolation of the ingredients. Those substances having the highest 
 melting, or freezing points, of course, are the first to crystallize. This 
 separation, then, has the effect of concentrating the solution of the sub- 
 stance having the lower freezing points in the mother liquor. This process 
 continues until the mother liquor is made up only of that substance that 
 has the lowest freezing point, when it, too, will freeze, forming in the ingot 
 a solid mass very different in composition from the metal that crystallized 
 
INGOT DEFECTS 349 
 
 out at the beginning. Under such conditions, it is to be expected that the 
 substances with the low melting points would be found in one spot or locality 
 in the ingot and that this spot would be located near the top and center 
 of the ingot, that is, at the bottom of the pipe, where the metal was the 
 last to freeze; and to some extent, this is what actually does occur, so 
 that this central position of the ingot is spoken of as the line of segregation. 
 That the condition is not even more pronounced is due to the closing in, or 
 entrapping, of the small pockets of the mother liquor during the freezing 
 and to the high viscosity of the fluid. Like the pipe and the blow hole, 
 segregation cannot be overcome, but by rapid cooling and the use of 
 aluminum to quiet the metal it may be lessened somewhat. In conclusion 
 it should be remarked that this is one of the fundamental reasons why 
 a range in chemical and physical specifications is imperative, at least from 
 the manufacturer's point of view. 
 
 Checking and Scabs: If the surface of a mold is very rough, or 
 contains cavities, so that resistance is offered to the natural contraction 
 of the steel, transverse cracks in the skin of the ingot may result. 
 However, in spite of all precautions that may be taken cracks in ingots 
 will occur, and a study of this matter indicates that this defect is more 
 liable to occur in certain grades of steel than ir others, and particularly 
 in those steels in which the carbon content is between .17% and .24%. These 
 cracks become oxidized and subsequently produce a seamy product. 
 Scabby material is often caused by improper pouring. If care is not taken 
 to prevent it, the metal may be splashed against the side of the cold mold 
 during the pouring. These splashes tend to stick to the mold, and, 
 becoming oxidized on the surface, will then appear as scabs on the ingot 
 after it is stripped. These defects very often show up after rolling in the 
 form of seams and slivers; in plates they will form serious surface defects. 
 Such defects are entirely avoided by bottom casting the ingots. A cracked 
 mold that must be forcibly drawn by the stripper may produce similar 
 defects. 
 
 Slag Inclusions: Various explanations are offered to account for the 
 presence of slag particles held within steel. Slag inclusions may be due 
 to an improperly finished bath, or in case the furnace practice has been 
 good, they may be caused by slag being stirred into the steel and mechani- 
 cally held by it while the heat is being poured. They may also be due 
 to dirt in the ladle or molds, or they may be the result of slag forming 
 reactions that occur during the deoxidation of the metal in the ladle or 
 molds. The latter cause would seem most conducive to the formation 
 of the minute slag particles that occur so frequently in nearly all steels. 
 Slag particles if given the opportunity, will rise to the top of the metal in 
 the ladle, but for lack of time small particles do not always do so. Slag 
 particles remaining in steel after it has been teemed have little opportunity 
 to rise, because the chilling of the steel by the mold is so rapid, and they 
 
350 
 
 PREPARATION OF STEEL 
 
 FIG. 57. Split Ingots 
 
INGOT DEFECTS 
 
 351 
 
 r- 
 
 
 r 
 
 Ml 
 
 Showing the Region of Greatest Segregc 
 
352 PREPARATION OF STEEL 
 
 are, therefore, entrapped. Hence, deoxidation in the mold should be 
 controlled by good judgment and avoided when possible. Large slag 
 inclusions are an original source of blisters in finished products. 
 
 Size and Shape of Ingots: With regard to the size of ingots a number 
 of factors operate to control both the size of the section and the length. 
 First of these is pouring cost. It is obvious that the cost of teeming a 
 50-ton heat into two-ton ingots will be much greater than if the same heat 
 is cast into four-ton ingots due to increased number of molds required, 
 the increase in scrap produced, and the longer time consumed in stripping 
 and charging into the soaking pits, etc. The cost of rolling may be in- 
 creased, also, for a long ingot may be rolled with the same number of passes 
 as a short one of the same section. The product desired, also, is a factor 
 in determining the size and the shape of the ingot. In plate mills, for 
 instance, which operate independently of slabbing mills, the size of the 
 plate to be rolled determines the size of the slab ingot. The blooming 
 and slabbing mills and their equipment, once installed, fix a limit to the 
 size of the ingot both as to section and length. As to their shape, ingots 
 may be of any convenient form, though for rolling they are usually 
 square or rectangular in section with rounded corners. These forms are 
 easiest on the steel, as the flat sides offer the least resistance to contraction 
 on cooling and the rounded corners prevent rapid cooling along the edges, 
 which would result in cracks from subsequent contraction on cooling. For 
 forging large rounds, a round ingot with a corrugated surface is used. 
 The corrugations permit expansion and contraction of the ingot without 
 the danger of developing cracks that are liable to occur in the surface and 
 interior of ingots cast in a perfectly cylindrical mould. The taper on in- 
 gots is to facilitate the stripping. To express the size of a rectangular 
 ingot the dimensions of its largest section are always given, unless other- 
 wise specified. Thus, a 23}/ inch ingot means it is 23^2 inches square 
 at the butt; an 18% x 21^ inch ingot means it is rectangular in section and 
 
 x 21^ inches at the butt. 
 
 SECTION II. 
 
 THE CONSTRUCTION OF THE SOAKING PIT. 
 
 General Features of the Soaking Pit: The soaking pit of modern 
 construction is so built that it can be used either as an old time pit or as a 
 heating furnace. Briefly, soaking pits are deep chambers, or underground 
 furnaces of square or rectangular sections, heated by the regenerative 
 principle and opening at the top. As to size, they are built large enough 
 to contain four, six, or eight blooming mill ingots per hole, in an upright 
 position. The older furnaces contain four ingots per hole, while the capacity 
 of the most recent ones is eight ingots. The increase in size is due mainly to 
 the economy in fuel which is obtained by the use of large pits. While the 
 details of pit construction may vary somewhat at different works, yet the 
 form and principle of all are alike. Therefore, it is sufficient to study but 
 one, which may serve as an example of all. For this purpose, a six ingot 
 furnace at Duquesne will be described somewhat in detail. 
 
THE SOAKING PIT - 353 
 
 Arrangement of the Pits: For heating the ingots for the two 
 blooming mills at these works, the 38-inch and 40-inch mills, there are 11 
 rows of pits, or to be more exact, 11 furnaces of 4 holes each. The holes are 
 numbered 1 to 44, inclusive, No. 1 to No. 20 and No. 37 to No. 40 serve 
 the 38-inch mill; the other 16, No. 21 to No. 36, inclusive, serve the 40-inch 
 mill. In case a furnace for the 40-inch mill is off for repairs No. 5 furnace, 
 containing holes 17, 18, 19 and 20, may be substituted. The first nine 
 furnaces are built to contain six 22" x 22" ingots per hole, but numbers 10 
 and 11 are constructed to hold eight ingots per pit. This gives a pit cap- 
 acity for the 40-inch mill of 96 ingots, and for the 38-inch mill, 184 ingots. 
 The furnaces with the exception of No. 1, are built in groups of two each. 
 From center to center of each two adjacent furnaces, the distance is thirty- 
 three feet. 
 
 Equipment for Handling Ingots: Spanning these soaking pit 
 furnaces are electric traveling cranes, two of which are over furnaces No. 10 
 and No. 11, and four are over No. 1 to No. 9, inclusive. These cranes are 
 Morgan 6-ton machines, and are equipped with Westinghouse motors as 
 follows: 50 h. p. on the bricTge, 50 h. p. on the hoist, 10 h. p. on the 
 trolley, and 5 h. p. on the tongs. The main hoist is operated by a gear 
 hoist and shafting rack. The tongs are connected up with a drum on a 
 lifting arm, giving a vertical movement of about nine and one-fourth feet. 
 The tongs are actuated by means of a curved groove in the main hoist so 
 that their distance apart may be varied. The tongs are equipped with four- 
 inch bits, giving a distance between the two bits, when in the closed or 
 lowered position, of sixteen inches and when in the raised or open position a 
 distance of nineteen and five-eighths inches. Thus the largest ingot that can 
 be gripped is one about eighteen inches at the top. When larger ingots, such 
 as the 22" x 22" size, are to be handled it is necessary to remove one bit. 
 
 Construction of the Pits: In detail the construction of a six-ingot 
 soaking pit furnace is as follows: Each furnace or pit contains four rect- 
 angular holes, eight feet long, five feet three inches wide and eight feet 
 seven inches deep. These holes are built side by side in the furnace, and 
 are separated only by firebrick walls. Each hole has two air regenerators, 
 one on each side, so that in connection with each furnace there are 
 eight regenerators. The holes are closed by firebrick covers, each cover 
 being supported on four wheels which roll on cast steel rails lying on the 
 division wall between the pits and fastened at their ends to the I-beams 
 supporting the platform about the pits. The walls enclosing the checkers 
 and those supporting the pit proper rest on a concrete foundation twelve 
 inches thick. These walls are built of firebrick, faced on the outside 
 with river brick. The outside walls are about eighteen feet high, and the 
 river brick wall directly under the pit is about eight feet high. The top of 
 this river brick wall is protected by cast iron coping plates. Placed 
 vertically on these coping plates and extending up into the firebrick 
 bridgewall are cast iron end plates. On the coping plates rests also the 
 
354 
 
 ROLLING OF STEEL 
 
 Fia. 58. Cross Section Drawing of a Four-Hole Six-Ingot Soaking 
 Pit Furnace 
 
THE SOAKING PIT 355 
 
 i 
 
 steel I-beams supporting the cast iron pans which form the bottom of the 
 pits. A pan is made in two sections with a semi-circular hole on the 
 inside edge of each section, so that when they are fitted together there 
 will be a circular opening about ten inches in diameter through which the 
 cinder can be removed from the pit. Each section rests on three ten-inch I-- 
 beams. About one foot on each end of the pan is flared upward at an angle 
 of 45. There are five cooling boxes resting on the I-beams, three inside, 
 or intermediate cooling boxes, and two end ones. The inside cooling boxes 
 are in the division walls between the pits, and the end ones are in the end 
 walls of the two outside pits. Each of these boxes rests on two 
 I-beams. The boxes are hollow, being open on the bottom, and the lower 
 half of the end of the bottom is cut off at an angle of about 45. Thus, 
 these cooling boxes and pans form with the end plates a triangular space 
 through which the air circulates and forms the air cooling system for the 
 bridgewall. On the top of the cooling boxes and near each end, there is 
 a four-inch hole, and when the walls between the pits are built up, this 
 opening is extended up to the surface so that a circulation of air is main- 
 tained through it. The boxes are cast iron, one inch thick, eight feet nine 
 inches long, twenty-eight inches high and are fourteen inches wide inside 
 at the bottom and four and three-fourths inches inside at the top. The 
 pans are lined with nine inches of firebrick set in firebrick mortar. The 
 cinder hole is built up of nine-inch side-arch brick. The pit, then, 
 to above the slag line, is lined with chrome brick, as these are the only 
 bricks that are not fluxed by the slag formed. Next to the coolers, however, 
 there is one four and one-half inch course of firebrick, laid in fireclay, but 
 on these there is a four-inch course of chrome bricks, laid dry, and at the 
 front and back of the pits, at the bridge wall, this chrome brick lining is 
 nine inches thick. There are, in all, seven courses of chrome bricks, which 
 brings the wall even with the top of the cooling box. Above this level, 
 the walls are built entirely of fire brick. The side of the bridgewall next to 
 the checker chambers is capped with heavy firebrick tiles (13^ x 6 x 2%") 
 giving a width to the firebridge of twenty-two and one-half inches. The 
 chrome bricks are heavily coated with silica slurry, which affords an 
 added protection. The use of this slurry is especially important on the 
 front and back, for the bridgewall is the weakest part of the furnace, 
 because it is subject to the intense heat of the products of combustion 
 leaving the pit and, therefore, has a tendency to crack and spall. The 
 firebrick part of the walls inside of the pit are coated with a slurry of 
 fire-clay. This coating fills up the cracks and forms a glaze which protects 
 the bricks. 
 
 The Air Regenerators are about six feet square in horizontal section 
 and seventeen feet four inches deep. The sewer which conducts the air 
 from the air valve to the two chambers on the same side and farthest from 
 the stack is beneath the air sewer for the two pits nearest the stack. Thus, 
 the regenerator chambers and, hence, the pits are fired in pairs. The bottom 
 
356 PREPARATION OF STEEL 
 
 sewer is two feet seven inches high and the arch is nine inches thick. There- 
 fore, the regenerator chambers for the two pits, on each side, nearest the 
 stack are three feet four inches less in height than the back chambers. 
 Between the two regenerators on the same air flue, starting at a height of 
 about five feet above the bottom of the chambers, there is an eighteen inch 
 firebrick wall separating the two adjacent chambers. The two pairs of 
 regenerators on the different air sewers are entirely separated from each 
 other. The bottoms of the chambers are separated into four spaces (four- 
 teen and one-half inches wide) by firebrick withe walls. These walls 
 extend back into the air sewer to the air valve and provide for the even 
 distribution of the air. On these withe walls there is one course of firebrick 
 tiles on which the checkers rest. The withe walls are three feet seven 
 inches high in the two pairs of regenerators farthest from the stack and 
 two feet four inches in the other two pairs, thus making the height of the 
 first mentioned checkers nine feet six and one-half inches. In the two pairs 
 of chambers farthest from the stack, the checkers extend up to within 
 thirty-one inches of the bridge wall and in the other two pairs to within 
 eighteen inches. The top row of checker bricks are laid so that the openings 
 are in the same direction that the air must take in entering the pit. The 
 roof of the regenerator chambjer is arched, being built of firebrick thirteen 
 and one-half inches thick, but the arch over the bridgewall, for a distance 
 of twenty-seven inches, is of silica brick nine inches thick and laid in silica 
 slurry on top of which is a thirteen and one-half inch firebrick arch. 
 Silica brick is used in this construction because it is very refractory and 
 can withstand a heavy load when highly heated. The whole furnace is 
 securely tied together by cast iron corner binders, tie rods, and buckstays. 
 Each of the four outside corners of the furnace has a corner binder fifteen 
 feet ten and one-half inches high, and the four corners directly under the 
 pits have binders eight feet one inch high. These binders have a twelve 
 inch flange, two inches thick, provided with lugs for the tie rods. The 
 binders are connected by two inch tie rods. Along the two outside walls 
 of the regenerators there is a structural steel buckstay, the ends of which 
 are connected, across the furnace end, by two inch tie rods. 
 
 The Pit Covers consist of four iron castings which are bolted together 
 and are held rigid at the center by a cast iron separator. Inclosed in this 
 frame is a firebrick arch. As already stated, to these castings are fastened 
 the wheels on which the covers roll. To each separator is fastened a steel 
 piston rod connected to the piston head in a hydraulic cylinder. These 
 cylinders furnish the means by which the covers are moved. In some 
 plants the covers are moved by lowering the tongs into a special box in 
 the separator casting and then moving the crane in the direction desired. 
 The hydraulic cylinders operate under a water pressure of 500 Ibs. per 
 square inch. They rest on cast iron stands fastened to the floor beams, 
 and the bearings for the cylinders are placed about the center of the 
 cylinders, thus making them free to rotate about this point. This con- 
 
THE SOAKING PIT 357 
 
 struction is made necessary, because, as the furnace becomes old, the 
 dividing walls between the pits sink and the rails bend. Therefore, the 
 connection of the piston rod and the separator has a constantly varying 
 elevation due to the different elevation of the rails, and it is necessary to 
 have the cylinders on a rocker so that they -may follow this motion and 
 constantly adjust themselves in line with the piston. The extreme stroke 
 of the cover piston is nine feet nine inches. In order to make the covers 
 fit nicely the tops of the pits are surrounded by floor plates of cast iron. 
 
 Fuel and Air Valves, etc.: These pits were built to use natural gas 
 for fuel, but this fuel has been replaced by coke-oven gas. When natural 
 gas was used for heating the pits, it was admitted through the roof of the 
 regenerator chamber by means of two three-fourth inch pipes twenty-one 
 inches long. These pipes were placed at such an angle (about four and one- 
 half inches of slope per foot) and distance from the pit that the flame did 
 not play directly on the face of the ingot, and a reducing atmosphere could 
 be maintained inside the pit. The pipes, or burners, were twenty-one 
 inches apart and were fed from a one and one-fourth inch pipe which in turn 
 was connected up to a four inch gas manifold supplying the gas to the four 
 chambers on a particular side. For coke oven gas it was thought it would 
 be necessary to modify this scheme of firing somewhat in order to make 
 the conditions suit the difference in the heating properties of these gases, 
 but trial runs indicate that satisfactory results are obtained if the coke 
 oven gas is burned in the same manner as the natural gas. For re versing 
 the direction of the air there are two thirty-inch Ahlen sliding valves, and 
 for reversing the direction of the gas a three-way valve. Each set of valves 
 consists of two cast iron bed plates, two cast iron sliding plates, two 
 hydraulic cylinders, and two hoods. The distance from center to center 
 of the hoods is five feet three and one-half inches. The bed plates are 
 bolted together and the cylinders are bolted to them. Each of these bed 
 plates has three openings connected to flues, of which the two outside ones 
 lead to the regenerators and the center one to the stack. These flues are 
 twenty-two inches wide, the division walls being nine inches thick and the 
 wall between the two sets of valves twenty-two and one-half inches thick. 
 As to the sliding plates they are also bolted together and the two are then 
 connected directly to the piston of the hydraulic cylinder. The total 
 length of the sliding plate is eleven feet one inch. Each plate has six open- 
 ings, two pairs of which are used as air dampers, while the other two form, 
 with the hood, a part of the flue to the stack. The hood rests on the sliding 
 plate, and both the hood and the sliding plate are water cooled. The 
 hydraulic cylinder has a stroke of two feet seven inches and a plunger 
 diameter of six inches. 
 
 Stack=Flues and Stack: The flues leading from the valve to the 
 stack are three feet eleven inches high and twenty-two inches wide. In 
 these flues are the stack dampers. These dampers are hand operated by 
 means of a chain and a counter-weight. They slide in a guide frame made 
 
358 PREPARATION OF STEEL 
 
 in the form of a casting set in the brick work. The stacks are one hundred 
 three feet eight inches high, and consist of a riveted steel shell and a lining 
 of brick work. The plates of which the shell is made are one-fourth inch 
 thick at the bottom of the stack and one-eighth inch at the top. The 
 outside diameters of the shell at the top and bottom are respectively four 
 feet six inches and five feet ten inches. The lining consists of a four and 
 one-half inch course of firebrick. 
 
 The Course of the Gases Through the Pits: Of the two thirty-inch 
 air valves, the one nearest to the pits is for the two front pits and the other 
 for the two back pits. Thus, for the two front pits, the air enters the inside 
 valve through the top sewer, goes through the two front regenerator 
 chambers, the two front pits, down through the opposite checkers, through 
 the top sewer, through the inside reversing valve, then past the right hand 
 stack damper to the stack. For controlling the sliding valve so that the 
 two hoods work in unison, a double-acting Critchlow valve is provided. 
 Each of the sliding plates is provided with two air dampers so that it is 
 impossible to shut off the air from the two adjacent pits with the valves 
 in either position. The gas on all four pits is reversed by one three-way 
 valve, but on each side of this valve there are four other valves, so that 
 the gas can be shut off separately on any pit. To reverse the direction of 
 the gas and air, the gas is shut off, then the air reversed, and after it the gas 
 is reversed. 
 
 Eight Ingot Pits: The main difference between the six ingot and the 
 eight ingot pits, aside from increased dimensions, lies in the fact that the 
 air regenerators are provided with a special division wall. This wall is 
 built parallel to the end of the pit and extends to the top of the checkers, 
 the object being to retain any cinder running over from the pit in the first 
 few checker openings, thus preventing the choking of all but a few of the 
 checker spaces, and maintaining a higher efficiency of the regenerators. 
 Also, these pits are provided with two cinder holes instead of one as for 
 the six ingot holes. The furnaces are spaced forty-one feet four inches from 
 center to center, and the holes are five feet three inches wide and ten feet 
 seven inches long, and are spaced eight feet three inches from center to 
 center. The covers on these pits can be separated about two-thirds of the 
 distance from the back. On these covers the piston is connected to the 
 end casting of the cover frame instead of to the separator, and there is a 
 separator on each portion of the cover. When closed, the separators fit 
 together and are fastened thus with hooks. With this arrangement it is 
 possible to move the entire cover or, by unhooking, the back portion only 
 may be moved. 
 
 Making Up the Bottom of the Pit: Bottom making is done by a 
 group of men called the bottom-makers, who are provided with shovels, 
 long handled pokers and cutters. To clean out a pit, the pit cover is pulled 
 back slightly, and a shield is drawn up over the front of the pit. The cinder 
 hole in the bottom of the pit is then opened with the poker, and by means 
 
SOAKING INGOTS 359 
 
 of the cutters the cinder is shoved out through this hole. After the cinder 
 has been removed, a piece of iron sheeting is put over the hole, and then 
 the process of making bottom is begun. Coke breeze from the blast furnace 
 coke bins is used to make up the bottom. Coke breeze is used because 
 it absorbs and makes fragile the molten oxide that runs off the ingot, pro- 
 tects the brick work, and helps to maintain a reducing atmosphere in the 
 furnace. The depth of the coke on the bottom should be maintained at 
 about eighteen inches. To provide the desired depth, making up for the 
 coke and cinder that are pushed out each time, requires for each bottom 
 six to eight wheelbarrow loads of breeze, which weighs about 250 pounds 
 per barrow load, for the six ingot pits; for the eight ingot pits ten to twelve 
 wheelbarrow loads, or about 2500 pounds, is required. This coke is thrown 
 in from the front, the ends being made up first, the sides next, and lastly 
 the center. The bottom is made up so that there will be two troughs into 
 which the ingots may be placed. The object in providing these troughs 
 is to keep the ingots away from the walls so that they will have a better 
 chance to heat but at the same time will not be placed so near the center 
 that they will be hit directly by the flame. During the time that the 
 bottoms are being made up, the gas and air are shut off, and the stack is 
 made to draw air through the cover opening so that the heat is drawn away 
 from the men. Before charging ingots on a new bottom, the coke breeze 
 should be allowed to become well heated throughout, as a cold bottom in 
 the pit allows the butts of the ingots to remain cold, and when ingots are 
 put through the mills in this condition they are liable to break the rolls. 
 
 SECTION III. 
 
 SOAKING THE INGOTS FOR ROLLING. 
 
 Charging the Ingots: Ingots should always be charged into the 
 soaking pits in an upright position, which explains the peculiar construction 
 of the pits. There are two reasons for this method of charging. First, 
 the best practice for the care of ingots demands that they be stripped as 
 soon as possible after pouring and delivered to the soaking pits before they 
 lose much of their original heat, because the hotter they are charged the 
 quicker will they reach the rolling temperature, and little fuel for reheating 
 will be required. In pursuance of this practice, most ingots, except high 
 carbon, high sulphur and alloy steels, which are allowed to become solid 
 throughout before stripping, reach the pits while their central portions 
 are still molten and must, therefore, stand in an upright position until this 
 portion has become solid, as otherwise the extent of the pipe might be in- 
 creased and its position would be changed. Second, by charging ingots 
 vertically more surface for ingress and egress of heat is exposed to the 
 atmosphere of the furnace, thus causing them to come to a uniform rolling 
 temperature much quicker than would be the case if they were placed in 
 any other way. 
 
360 PREPARATION OF STEEL 
 
 Heating the Ingots: From what has been said, it is easy to surmise 
 that great injury can be done in the heating of the ingots. This injury 
 consists of under-heating, over-heating, uneven heating, or worse than all, 
 burning. Of these, under-heating and over-heating are the least harmful 
 to the steel; the former increases the power required for rolling and decreases 
 the time permissible for the rolling; the latter, by increasing the grain 
 size and lessening the force of cohesion, makes the steel tender and liable 
 to crack. Uneven heating increases the difficulties of rolling very much. 
 A cold butt of an ingot, for example, may cause a roll to be broken. Burning 
 may range from extreme over-heating to a temperature just below the 
 melting point, where the more fusible constituents melt and run out of the 
 ingot, forming cavities that, on rolling, result in defects that will be cause 
 for rejection of the material. In the case of thin skinned ingots, severe over- 
 heating may have a like result by exposing the blow holes. Besides these 
 general precautions, different conditions and different grades of steel require 
 different treatment. . As an example of the point in question a summary 
 of the soaking practice as carried out at Duquesne is given herewith. 
 
 Week-End Charges: If hot steel is charged Saturday evening just 
 before the mill shuts down, it will be allowed to soak until one or two 
 o'clock Sunday morning, when gas will be admitted for about an hour. 
 Soaking will then be continued until, the day turn comes out at seven o'clock 
 a. m. But if cold steel should be charged before the week-end shut-down, 
 gas is admitted for three or four hours, the flame being reversed at intervals 
 of from one-half to one hour; and the steel is then allowed to soak until 
 Sunday morning. During the soaking, the stack and air dampers are kept 
 closed. 
 
 Soaking Hot and Cold Ingots: To bring hot steel to the required 
 rolling temperature requires approximately the same amount of time as 
 the interval between the time the heat was tapped and the time it was 
 charged into the pits. Hot special steel of medium carbon content must 
 be in the pits about one and one-half hours and spring steel about one hour. 
 Thus, the period, from the time the heat is tapped at the open hearth until it 
 can be rolled, is about three hours for Duquesne special, about two hours 
 for spring steel, and one and a half hours for ordinary steel. To heat six 
 cold soft steel ingots in the 6-ingot pits requires about six hours. For 
 about four hours after the pits are charged, the gas and air may be admitted 
 on each side alternately for half hour periods. The period of reversal 
 should then be cut to fifteen minutes. Towards the last, as the temperature 
 of the steel approaches the rolling temperature, the period of reversal may 
 be cut to five or ten minutes, for the more frequent the reversal the more 
 even will be the temperature of the pits. Cold steel is very rarely 
 charged in the ingot pits, the practice being followed only after a 
 shut-down when the mills start operating at the same time as the open 
 hearth, for at such a time there is no hot steel on hand. The period 
 required for heating cold steel in the eight ingot pits is about eight hours. 
 
SOAKING INGOTS 361 
 
 Soft steel is heated to a temperature of about 1200 C. (2200 F.) With 
 both high and low carbon steels, should only four or five ingots be charged 
 in the smaller pits, the period required for heating would be greatly reduced. 
 This is true especially for the low carbon steel, for with only four ingots 
 to a pit it is possible to prepare the cold steel in four hours. By charging 
 only six ingots in the large pits, the period may be reduced to about six 
 hours. Before a cold high carbon heat (.70% carbon or over) is charged, 
 the pits should be cooled for about a half hour, for if these ingots are heated 
 rapidly they are liable to crack. After the pits have been cooled, the ingots 
 are charged, and sometimes the covers are left open for a half hour, so 
 that the steel will be heated very slowly. The period between reversals 
 should not be as long as for low carbon cold steel, and so at first the reversals 
 for steel of this grade are made at intervals of a half hour or less, and during 
 the balance of the time the period between the reversals is about ten minutes. 
 The rolling temperature of spring steel is about 1090 C. (2000 F.) 
 
 Soaking Hot Spring Steel: This grade of steel is charged in a hot 
 pit. The gas may be admitted for a half hour, the flue being reversed every 
 five or ten minutes; the steel should then be allowed to soak for fifteen 
 minutes, and then gas should be admitted for about fifteen minutes to bring 
 up the temperature of the outside of the ingot. This steel should be ready 
 to roll in about an hour. While the steel is soaking, in addition to shutting 
 off the gas, it is best to shut off the air supply, also, for the effect of the 
 hot air on the ingot is to oxidize or even to burn it. If the steel is very 
 hot when charged, it should be allowed to soak for a half hour before gas 
 is admitted; then gas and air should be admitted for about a half hour, 
 with reversals every five or ten minutes. The steel should then be soaked 
 for a half hour without air, and then, just before drawing, the temperature 
 of the outer part of the ingot should be brought up by admission of gas 
 and air again. 
 
 Soaking Low Carbon Hot Steel: Hot low carbon steel ingots may 
 be heated without danger for a half hour, the direction of the gas and air 
 being reversed every fifteen minutes. The steel should then be allowed to 
 soak for fifteen minutes, and before drawing the outside temperature should 
 be raised. Since there is not as much danger of burning this steel as there 
 is with the high carbon grades, it is not always necessary to close the air 
 dampers during the time the steel is soaking. 
 
 Soaking Medium Steels : The practice with respect to medium steels, 
 .30% to .60% carbon, is to heat the ingots to about the same temperature 
 as for low carbon steel. The steel, if charged hot, should be ready to roll 
 in one hour after charging. 
 
 Soaking Screw Stock: High sulphur steel is charged as quickly as 
 possible after stripping. The time required to heat it is about one and a 
 half hours. The steel is heated to dripping, that is, until the scale melts 
 and flows readily from the surface, and is rolled when in that condition. 
 Owing to the high sulphur content, it is necessary, to maintain good practice 
 
362 PREPARATION OF STEEL 
 
 in rolling, to heat it very hot, so that the steel will be very plastic. This 
 condition is obtained only at a very high temperature, about 1240 :C. 
 (2240 F.). Since these screw stock ingots are heated until they are dripping 
 a large amount of liquid cinder is always formed, so that it is necessary 
 to add a little coke in the pits after every heat of this kind to absorb this 
 cinder. 
 
 Soaking Alloy Steels: Nickel steel is heated to about the same 
 temperature as spring steel, 1090 C. Chrome vanadium is heated to about 
 1250 C. Copper steel is heated according to its carbon content in much 
 the same way as carbon steels. 
 
 Drawing the Ingots: The craneman draws the ingots from the pits 
 according to the orders of the heater. Usually, a definite order is followed; 
 at Duquesne the regular method is to draw the two front ingots from each 
 of two holes, then the two middle ones from each of the two holes, and then 
 the two back ones from each of the two holes. The operation is then repeated 
 on the next two holes. However, the operation may be varied; the two 
 front ones in each of four holes may be drawn, thus affording more time for 
 the middle ingots to heat while the others are being drawn. To transfer 
 the ingots from the pits to the blooming mill tables three pot cars, two of 
 which are extras, are provided. These cars are operated by 19 h. p. 
 Westinghouse motors. At the 38-inch mill they are controlled and dumped 
 from the manipulator of the mill, but on the 40-inch mill, the man that 
 operates the pit covers controls the movement of the car. When the car 
 receives an ingot it is run to the first table roller, and there the car is tipped, 
 when the ingot falls upon the table. 
 
 Heat Balance of Pits: That the soaking of ingots is an expensive 
 process is evident from the equipment required. The cost of the up-keep 
 of this apparatus is high, and the efficiency is very low, even on up-to-date 
 furnaces, as the following heat balances as determined by experiment on 
 some Duquesne furnaces using natural gas will show: 
 
 Table 50. Data Relative to the Efficiency of Soaking Pit Furnaces. 
 
 Sensible Heat in Steel Charged 619,701 B. t. u. 
 
 Heat of Combustion of the Gas 781,808 
 
 Heat Carried in by Regenerated Air 384,970 
 
 TOTAL 1,786,479 
 
 Heat in Steel when Drawn 773,597 B. t. u. 
 
 Heat in Gases Entering Stack 173,862 
 
 Heat Given up to Regenerators 386,329 
 
 Radiation and Unaccounted for Losses 452,691 
 
 1,786,479 
 
 ,,., ^ ffi . Heat Absorbed by Steel 153,896 R0f 
 
 Pit Efficiency= = =8.6% 
 
 Total Heat Delivered to Furnace 1,786,479 
 
DISPOSITION OF INGOTS 
 
 363 
 
 Disposition of Ingot Products: Since the ingot is the starting point 
 for all mechanical working, it is interesting to trace the material through 
 the various processes that the steel undergoes to produce the many articles 
 in which it is used. For this purpose the following table has been prepared, 
 and requires little by way of explanation. Each dash marks a reheating 
 of the material, while each word means a mill, or a set of mills, where 
 work is done to produce the article named. 
 
 Table 51. Disposition of Steel from Ingots. 
 
 Universal Mill Plate. 
 Armor Plate. 
 
 Bloom, Sheet Bar Sheets. 
 Shaped Bloom, Large Shapes. 
 
 (Universal Mill Plates. 
 Slabs j Eye Bars. 
 
 [Sheared Plates. 
 fRods. 
 Bars. 
 
 Bloom, Billet { Bands. 
 Hoop. 
 [Small Shapes. 
 
 fRods. 
 
 Ingots \ Small Shapes. 
 
 Billet Bars. 
 Bands. 
 
 Seamless Tube. 
 . Sheet Bar Sheet. 
 
 Rectangular BloomH Structural Shapes. 
 Rails. 
 Rail Joints. 
 Q1 . /Tube. 
 Skelp -(pipe. 
 Forgings. 
 (Wheels. 
 Cylindrical Blooms { Circular Shapes. 
 
 [Shell. 
 Large Forgings. 
 
364 THE ROLLING OF STEEL 
 
 CHAPTER V. 
 
 THE ROLLING OF STEEL BLOOMS AND SLABS. 
 
 SECTION I. 
 
 INTRODUCTORY. 
 
 Outline of the Plan of Study: Rolling mills are somewhat like 
 houses. Thus, while they are alike as to gross features, they differ greatly 
 as to details of construction. Just as the architect will strive to impart 
 individuality to a building, so the rolling mill engineer and builder will 
 endeavor to introduce new ideas looking to greater improvements in con- 
 struction; and just as it is desirable to adapt a building to its location and 
 surroundings, so is it found necessary to alter the details of mill construction 
 to suit the conditions, local and otherwise. The result of all these influences 
 on mill construction has been to produce such a variation in mills that 
 there are no two mills exactly alike. Evidently, to describe all the details 
 of mills and their operations is well nigh an endless task; yet it is desirable 
 that the reader be given an opportunity to become so well acquainted with 
 the rolling of each product that he will be more or less familiar with the 
 more essential details of its production and thoroughly understand the 
 conditions under which it is produced. The plan decided upon as best 
 to pursue is this: An attempt will be made to describe the rolling of as 
 many products as possible, and in doing so the order followed will be 
 from the rolling of material from ingots, to semi-finished products, to 
 finished products, as indicated in the previous diagram. In this con- 
 nection one mill rolling the material in question will be described, as well 
 as the operation in detail; after which the product itself will receive special 
 attention. In describing mills, the details of one mill of each type or class 
 will be given. As a sort of working outline of the plan, the following classi- 
 fication of mills will give an idea of the ground to be covered and the 
 order in which the subjects are to be treated. The general discussion 
 preceding this part of the study should supply information to fill in any 
 gaps that may occur in the studies to follow. 
 
BLOOMS AND SLABS 365 
 
 Table 52. Classification of Mills. 
 
 A. Mills Rolling Material from Ingots. 
 
 1. Semi-finishing Mills: 
 
 a. Blooming (Cogging) Mills. 
 
 b. Slabbing Mills. 
 
 2. Finishing: 
 
 a. Universal Plate Mills. 
 
 B. Mills Rolling Material from Blooms and Slabs. 
 
 1. Semi-finishing: 
 
 a. Billet Mills. 
 
 b. Sheet Bar. 
 
 c. Skelp. 
 
 2. Finishing: 
 
 a. Plate Mills: 
 
 i. Sheared, 
 ii. Universal. 
 
 b. Rail Mills. 
 
 c. Structural Shape Mills. 
 
 d. Wheel Mills Schoen Mill. 
 
 e. Wheel and Circular Shape Mills. 
 
 C. Mills Rolling Material from Billets. 
 
 1. Merchant Mills. 
 
 a. Guide Mills. 
 
 b. Bar Mills. 
 
 c. Hoop or Strip Mills, etc. 
 
 Blooms, Slabs and Billets: As a preliminary step toward forming 
 steel into the various sections which its many uses require, the heavy 
 ingots, except in certain plate mills and some large shape mills, are first 
 roughly reduced, in mills especially designed for the purpose, to much lighter 
 but still very simple sections, as the round, the square and the rectangle. 
 When the ingot has been reduced to the dimensions of a square between 
 one and one-fourth inches and six inches it is cut into convenient lengths, 
 called billets ; if these pieces are six inches square or larger, they are known 
 as blooms : and if reduced to rectangular forms but with widths which are 
 less than twice the thickness and within the dimensions specified for the 
 square, the same names apply. But if the width far exceeds the thickness 
 of the rectangular section, then it is called a slab. If the output of the mill 
 is mainly blooms, it is called a blooming mill in the United States or a 
 cogging mill in England; if billets, a billet mill; and if slabs, a slabbing mill. 
 The blooming and slabbing are the largest and strongest mills used to roll 
 steel, if the mills that roll heavy armor, of which there are no longer any 
 in this country, be excepted. The reasons for the existence of these mills 
 are evident. 
 
THE ROLLING OF STEEL 
 
 SECTION II. 
 
 SOME GENERAL FEATURES PERTAINING TO BLOOMING MILLS. 
 
 Size of Blooming Mills: The size of blooming mills is popularly 
 supposed to be based on the diameter of the rolls, or on the distance from 
 center to center of the rolls. Both these quantities are constantly changing, 
 due to the wearing of the rolls, which affects their diameters, and to the 
 fact that they are adjustable. The size is, therefore, based on the distance 
 from center to center of the pinions, which corresponds to the distance 
 from center to center of the rolls and also to their diameters, and is always 
 constant. The blooming mills in use at the present time will range in size 
 from twenty-eight to forty-six inches. The older mills are the smaller, 
 because it was formerly the practice to cast the ingots much smaller than 
 at present, and large mills were not required. The size of ingots having been 
 gradually increased for the reasons already pointed out, the size of the 
 mills designed to roll them were necessarily increased also. This size seems 
 now to have approached a standard, and most mills of recent construction 
 have rolls in the neighborhood of forty inches in diameter. 
 
 Types of Bloomers, Their Advantages and Disadvantages : Bloom- 
 ing mills are of three general types, namely, reversing, continuous, both 
 of which are two=high, and three=high. Of these, two-high reversing and 
 three-high mills are the most common. As an example of the continuous 
 blooming mill, the billet mill at Gary, Ind., is cited. It consists of nine 
 stands of rolls arranged in tandem and separately driven by electric motors. 
 Since its blooms are delivered directly to a continuous billet mill, the 
 reduction of the ingot to billets is made in one continuous operation. The 
 chief advantage in this arrangement is an extraordinarily large output. 
 As to reversing and three-high bloomers, each of these types has its 
 advantages and disadvantages, some of which it may be of interest to 
 enumerate here. The main advantage of the reversing mill over the three- 
 high lies in its greater flexibility. Thus, the top roll being adjustable, 
 various sizes of blooms, billets or slabs can be rolled on one set of rolls, and 
 the draught can be regulated to suit steel at different temperatures and of 
 different grades. Even different methods of reducing the ingot may be 
 employed with the same rolls. On long lengths the two-high mill is to be 
 preferred on account of the greater ease with which such material can be 
 handled, while the simplicity of the roll design is also a factor in favor of 
 these mills. On the other hand, a reversing mill is a much more expensive 
 mill than a three-high mill. In the first place the tonnage is much lower. 
 On two-high forty inch mills the average output is about 2000 tons per 
 twenty-four hour day, to produce which about 2500 tons of ingots are 
 required, while a three-high mill of the same size will roll almost twice as 
 much steel. Again, the power equipment of the reversing mill is costly 
 and the loss of power is great. Reliable tests show that the total power 
 
BLOOMS 
 
 367 
 
368 THE ROLLING OF STEEL 
 
 developed by a reversing mill engine is distributed about as follows: 
 
 27% used in overcoming idle friction of the engine parts. 
 9% used in overcoming pinion and spindle friction. 
 
 13% used in overcoming roll journal friction. 
 
 21% used in overcoming the acceleration of the parts in reversing. 
 
 30% used in actually deforming the steel. 
 
 In three-high mills where the rotation is in one direction only, there 
 is no acceleration loss, besides, by the use of a flywheel, lighter engines 
 than those used on the reversing mill may be used to do the same work. 
 Efficiency tests on three-high mills show that about 85% of the total power 
 developed by the engine is used in driving the mill, and the idle friction 
 of the mill parts is about 15%, thus leaving nearly 70% of the motive 
 power developed available for deforming the steel. 
 
 Drive for Reversing Mills: Since the lengths dealt with on blooming 
 mills are relatively short, the speed of the rolling is slow, but as the 
 material is heavy and the pull is great, though the draughts are only 
 moderately heavy, great power is required. Hence, most of the older 
 reversing blooming mills are indirectly driven, that is, they are connected 
 to the engine through large gears which enable the engine to travel at a 
 higher speed than the mill. The power is thus multiplied by a number 
 equal to the speed ratio. This speed ratio will vary in the different mills 
 from as high as three to one to as low as one to one, while many mills, of 
 which the thirty-eight inch bloomer at Homestead and the forty inch 
 mills at Clairton and Duquesne are examples, are direct driven. A few 
 reversing mills installed since 1914 are driven by reversing electric motors. 
 These motor installations are very complicated. They consist of a main 
 motor and a motor-generator set, which prevents the acceleration loss 
 peculiar to the steam driven reversing mill and raises the efficiency of the 
 mill considerably. 
 
 SECTION III. 
 AN EXAMPLE OF REVERSING MILLS THE 40" MILL AT DUQUESNE. 
 
 The Engine for this mill, which is driven direct, is of the twin tandem 
 compound condensing type, but is operated non-condensing. It was made 
 by Mackintosh-Hemphill & Co. Its size is 44" x 70" x 60" and its maxi- 
 mum horse power is rated at 20,000. The maximum torque at the circum- 
 ference of a thirty inch roll is 465,000 inch-pounds. The engine may run at 
 any speed up to 140 r. p. m., but the maximum speed during rolling is about 
 130 r. p. m. The exhaust steam is discharged into a feed water heater. 
 The throttle is controlled from the pulpit located about thirty feet in front 
 of the rolls and directly over the roll tables. 
 
 Driving Connections: A cast nickel-steel crab of six pods is keyed 
 onto the crankshaft of the engine; it is four feet five inches in diameter. 
 Over it, and held in place by wooden stretcher blocks is fitted the large 
 end of a cast nickel steel compound coupling box, which, at this end, is 
 three feet two inches in diameter. The driving spindle from this coupling 
 
TWO-HIGH BLOOMING MILL 369 
 
 is five feet eleven and one-fourth inches long and is supported by a cast 
 steel coupling carrier resting at its four corners on standard spiral car 
 springs of 12500 pounds capacity, which stand eight and one-fourth inches 
 high when free, seven and one-fourth inches at 4700 pounds load, and six 
 and nine-sixteenth inches when fully compressed. The springs in, turn rest 
 on cast steel seats bolted to special cast iron shoes which are anchored to 
 the shoes carrying the pinions and housings. The carriers are lined with 
 one inch of babbitt metal. The mill end coupling box is two feet four and 
 one-half inches in diameter and is cast to fit over the four pods of the engine- 
 end wobbler of the bottom mill pinion. 
 
 Pinions and Pinion Housings: The pinions are of the staggered 
 straight tooth type and are made of nickel steel, approximately of a com- 
 position as shown by the following analysis: Carbon, .29%; manganese, 
 .66%; phosphorus, .020%; sulphur, .030%; and nickel, 2.97%. Their average 
 life is 253,575 tons of steel rolled. The top and bottom pinions are similar 
 and hence interchangeable. Each one is ten feet six inches long over all 
 and four feet ten inches between the necks, which are twenty-one inches in 
 diameter. This diameter is further reduced to twenty and one-half inches 
 at the wobblers. The pitch diameter is forty inches, and the number of 
 teeth is fourteen. The pinions run in solid cast steel babbitted bearings, 
 the bottom of which are beveled to fit on the sills of the window of the 
 cast iron pinion housings. The pinion housings are bolted to the mill shoes, 
 are nine feet six inches high, and have windows 2' 7W wide by 1' 6%" 
 deep; the windows are lined with one and one-fourth inch forged steel 
 liners held in place by stud-bolts through the housings. A cast steel housing 
 cap, to which is attached the hydraulic cylinder used for lowering the top 
 roll of the mill, is fastened over the housings by means of key bolts. The 
 bearings are each two feet six inches wide and twenty inches from 
 front to back and may be adjusted by set pins reaching through the housings. 
 The top bearings rest directly on top of the bottom bearings unless plate 
 liners are used in between them to get the proper pitch for the teeth. The 
 bearings are held down tight by keying the cap on tight and using liners 
 between it and the top bearing if necessary. 
 
 Spindles and Coupling Boxes: Over the mill-end wobblers of the 
 pinions are fitted cast steel coupling-boxes uniting the wobblers with the 
 spindles. The coupling boxes, of cast steel, are two feet six inches in diam- 
 eter and twenty-two and one-half inches wide, and cast to fit over the four 
 pods of the spindles. The bottom and top spindles are each ten feet long 
 and twenty-one inches in diameter where they rest on their carriers. The 
 bottom spindle is provided with wobblers twenty and one-half inches in 
 diameter and two feet in length, and is nineteen inches in diameter at the 
 center between the two carrier bearings. On the top spindle the wobblers 
 are nineteen inches in diameter, thirteen and seven-eighths inches long and 
 are curved at the ends to permit the mill end to ride up or down with the 
 top roll. Twenty-three inches at the center of the top spindle is turned 
 
370 THE ROLLING OF STEEL 
 
 smooth to a diameter of twenty-one inches to give a bearing for the spindle 
 carrier, which is movable. The bottom spindle rests at two points near 
 its ends on two stationary spindle carriers bolted to the mill shoes. The 
 carrier for the top, or vibrating, spindle consists of a cradle formed by two 
 cast steel arms hung at their engine ends from two supporting rods pivoted 
 on spring-supported bolts which pass through supporting brackets bolted to 
 the pinion housings. In the center of the carrier is a rest for the spindle, 
 and on its mill end the carrier is supported by the carrier bearing for the 
 top roll, being fastened to this bearing by a forged steel pin. A coupling- 
 box similar to those used with the pinions fastens the bottom spindle to 
 the bottom roll; seven-eighths of an inch clearance is allowed at each con- 
 nection between spindle and pinion or roll, respectively. For the top 
 pinion, a light coupling box is used in order that it may act as a safety for 
 the mill by breaking under excessive strain before any other part of the 
 mill is damaged. This box is twenty-two and one-half inches in width, 
 twenty-five and three-fourths inches in outside diameter, and two and one- 
 quarter inches in thickness at the thinnest point. The other boxes are 
 four inches thick. All coupling boxes are held in place by iron or wooden 
 stretcher blocks fastened in place by steel or leather straps. 
 
 Roll Housings: The roll housing on the engine side of the mill is set 
 with its center-line fourteen feet six and one-quarter inches from the center 
 line of the mill-end pinion housing, a cast iron separator and steel bolt 
 holding these two housings in line. This housing is cast steel but in other 
 respects is the same as the outside housing, which is made of cast iron. 
 Both are bolted to the mill shoes and stand twelve feet three inches high 
 above them; they are set with their center lines seven feet eleven inches 
 apart and are held in line by two cast iron separators and steel bolts, one 
 at the front and one at the back. Besides the mill rolls the housings also 
 support four feed rollers, two on each side of the rolls, sixteen inches in 
 diameter and five feet ten and one-eighth inches long. The windows of the 
 housings are three feet five and one-half inches wide, nine feet deep, and 
 begin two feet ten inches below the top of the housing. In the top of each 
 housing is left a hole for the housing nut, which is made of brass. Through 
 these nuts the housing screws for adjusting the top roll are inserted. The 
 nuts are larger at the bottom than at the top; they are twenty inches in 
 diameter at bottom, sixteen inches at top, and thirty-four inches high. 
 They are shrunk into the housings, and over them are fastened small caps, 
 twenty-seven inches high, on which the screw pinions rest. The housing 
 screws, the bottom ends of which press directly down on the screw brasses 
 in the cast iron breaker blocks on the rider bearing boxes of the top roll, 
 are made of .60% carbon open hearth steel, eight feet three inches long and ten 
 inches in diameter; the threads have a pitch of two inches. They must 
 allow a lift of twenty-five inches. These screws are provided with octagon 
 heads. Fitted about the heads are steel pinions which rest on the top of 
 the screw caps. The pinions have a pitch of two and one-quarter inches, 
 a pitch diameter of fifteen and eighty-two-hundredths inches, a face of 
 
TWO-HIGH BLOOMING MILL 371 
 
 eight inches, and twenty-two teeth. The pinions are operated through a 
 gear mounted on a spider which has a pitch diameter of eighty-three and 
 nine-hundredths inches, a pitch of two and one-fourth inches, a face of seven 
 inches, and one hundred sixteen teeth, and is, in turn, operated by means 
 of a pinion fastened to its shaft. This latter pinion has a pitch diameter 
 of twelve and fifty-four-hundredths inches, a pitch of three inches, 
 a face of ten inches, and thirteen teeth, and is operated by a rack with a 
 three inch pitch and a ten inch face. This rack is connected up to the 
 hydraulic cylinder located on the top of the pinion housing. Attached to 
 the rack is the finger, which, moving over a gauge, provides an indicator 
 for the size of the pass. The screws of the mill are required to lift twenty- 
 five inches, but a ten foot stroke of the rack will give the screws a vertical 
 movement of twenty-eight and one-half inches, the extra length of stroke 
 allowing for the wear of the roll necks and bearings. In rolling, this gauge 
 can be set to give correct readings on one pass only, and to gauge the other 
 passes it is necessary to add or subtract a certain quantity from the gauge 
 reading. A cast iron bridge is bolted to the top of the housing and 
 serves both to support the spider and to keep the housings properly 
 spaced. The housings are further supported and kept in line at the top by 
 two cast iron separators and two steel bolts. 
 
 Rolls: The top and bottom rolls in this mill are alike; there are two 
 in a set, and each roll weighs 26,000 pounds. A typical analysis of the rolls 
 gives the following results: .61% carbon, .75% manganese, .010% phos- 
 phorus, .030% sulphur. Their dimensions are: Total length, twelve feet 
 ten inches; length of body, six feet; diameter of wobbler, twenty and one- 
 half inches; diameter of necks, twenty-two inches; diameter of collars, 
 thirty-three and one-eighth inches. They have five passes, the widths and 
 diameters of which are 24"x31#", 12"x29>6", 8"x29%", 6"x29#", 
 4" x 29%". The rolls are ragged only in the twelve inch and eight inch 
 passes, and here the ragging is only one-sixteenth of an inch deep; all passes 
 are roughened sli ghtly with knurling wheels. These rolls are changed every 
 week and dressed in the roll shop. About three-sixteenths of an inch is 
 taken off each time they are dressed, and when the collars have been cut 
 down to thirty and one-half inches, the rolls are scrapped. Four sets are 
 kept on hand, and one set is used once in four weeks, giving a life of about 
 one year per set. The average tonnage is 82,650 tons for each set of rolls. 
 
 Roll Bearings: The bottom roll rests with each neck on a babbitt 
 lined cast steel bearing, two feet six inches wide, twenty-one and three- 
 fourths inches from front to back, and six and three-fourths inches thick at 
 the base; its bottom is made to fit the sill of the window, and its top is cut 
 out at a twelve inch radius with one and one-quarter inches of babbitt to 
 fit the neck of the bottom roll. Sheet steel shields are placed over the 
 necks of the bottom roll to keep scale from getting between the necks and 
 bearings. The top roll is carried in two cast steel carrier bearings, which 
 in turn are supported each by two three and one-half inch square steelyard 
 
372 THE ROLLING OF STEEL 
 
 rods. These rods are mounted in sockets hung from counterweighted arms 
 underneath the mill, the rods coming up through the housings and bottom 
 bearings on each side of the necks of the bottom roll. The carrier bearings 
 are three feet three inches wide, twenty-three and three-fourths inches from 
 front to back ancj are six inches thick at the base. They are flat on the 
 bottom and concave at the top with a twelve inch radius and babbitt metal 
 one and one-fourth inches thick. On the engine side of the inner bearings 
 are two lugs for receiving the pin to hold up the spindle carrier. The rider 
 bearing of the top roll is cast steel two feet two and one-half inches wide, 
 twenty-one and three-fourths inches from front to back, and two and one- 
 fourth inches thick, with one inch of babbitt metal. It is concave below 
 at a radius of twelve inches and beveled on top. The bearing box, also of 
 cast steel, is of the same dimensions as the bottom bearing, except that it 
 is seven inches thick. It is beveled underneath to receive the top bearing 
 and is flat on top. On it rests the cast steel breaker block, into which 
 the housing screw fits. The breaker blocks are protected by brasses, 
 which are placed in sockets on the tops of the blocks. 
 
 Hydraulic Shears: Immediately beyond the forty inch mill delivery 
 table, begins the No. 1 shear table, delivering to a hydraulic bloom shears. 
 This table is thirty-one feet long, and consists of fourteen cast steel rollers, 
 twelve inches in diameter and five feet eleven and one-quarter inches wide. 
 These are driven by a Westinghouse 30 h. p. 220 volt series wound D. C. 
 motor, controlled either at the bloom or the billet shears. At the end of 
 this table is an emergency shear. It is a vertical hydraulic shear, using water 
 at 500 pound pressure to the square inch; the plunger is forty-two inches in 
 diameter with a nineteen inch stroke.. The bottom shear knife is the 
 one actuated by the plunger; the knives are twenty-seven inches wide and 
 four inches thick. As this is an emergency shear, it is rarely used. 
 
 Steam Shears : No. 2 shear-table is immediately beyond the hydraulic 
 shears and has thirty-six driven rollers, and one idler, all similar to those 
 at No. 1 shear-table, except the last two, which are collared on one end. 
 The rollers are driven by a Westinghouse 50 h. p. 220 volt series wound D. C. 
 motor controlled at the steam shears. This table delivers the blooms and 
 slabs to the steam shears, the center of which is ninety-two feet ten and one- 
 half inches beyond the hydraulic shears. A bloom stamping machine is 
 located on this table midway between the two shears; it is of the idler wheel 
 type and is held in place hydraulically. The steam shears are driven by a 
 MacKintosh-Hemphill 18" x 20" simple vertical steam engine, the driving 
 shaft of which is meshed with the shears by a hydraulically operated clutch. 
 These shears are also vertical acting, the top knife blade being driven 
 down to meet the fixed lower one. The knives are twenty-seven and one- 
 half inches wide and three inches thick, and the top one has a ten and three- 
 fourths inch stroke. The steam shears are equipped with a gauge and 
 stopper for cutting a number of pieces the same length; the stopper can be 
 set to cut lengths from twelve inches to one hundred thirty-six inches, 
 
TWO-HIGH BLOOMING MILL 373 
 
 inclusive, in quarters of an inch; the piece to be cut is run through the shears 
 onto the rear table, which is sixteen feet long and consists of sixteen 
 hollow cast steel rollers, ten inches in diameter. It is driven through 
 universal joints by a Westinghouse 19 h. p. 220 volt series wound D. C. 
 motor and can be tilted at its receiving end hydraulically to inove down 
 with the shear knife. For pieces forty inches long or less, it is moved 
 nearer the shears to prevent the piece from falling into the pit for butt ends. 
 Beyond this table is the loading table for blooms and slabs; it is seventeen 
 feet long and has sixteen hollow cast steel rollers eight inches in diameter. 
 It is driven from a line shaft by the same type of motor as the shears rear 
 table. Halfway down this table on its inner side is a steam kicker with a 
 seven inch by four feet nine inch cylinder, which slides the bloom down 
 a chute to the buggies on the tracks below. A hydraulic stopper is located 
 at the end of this table. Crop ends or scrap can be run over the end of the 
 table to charging boxes below the end of it. Six feet six inches beyond 
 the end of this table begins the receiving table of a fourteen inch continuous 
 mill. 
 
 Manipulator: All reversing mills are provided with manipulators for 
 turning the ingot as desired between the passes, for moving the piece from 
 groove to groove and for straightening it as it enters the passes of the mill 
 when such straightening is necessary. They are located under the roll 
 table, and near the rolls on the entering side of the mill. They are of various 
 forms. The manipulator for this mill consists of two parallel sets of five 
 fingers each, and has both a vertical and horizontal movement. The frame 
 is beneath the table rolls and rests on a bottom frame which is supported 
 on four heads, connected to the arms of bell cranks. These cranks are 
 supported on a bed plate and are connected up by stretcher rods to a 
 hydraulic lifting cylinder, which has a stroke of fourteen and three-fourths 
 inches. This ratio of the length of the crank arms, however, increases this 
 lift to eighteen inches. On this frame are five rails to form a track 
 for the wheels of the upper frame which is moved horizontally by means 
 of a hydraulic cylinder. In the lowest position, the fingers are five inches 
 below the top level of the roll tables; in the highest, they extend thirteen 
 inches above it. 
 
 Design of the Rolls: All reversing blooming mill rolls are designed 
 with slight collars between the passes in order to control the spreading of 
 the material under the heavy reduction, as otherwise the material may 
 spread so far at the surface as to cause a protrusion, or fish tailing, of the 
 metal at the edges, which, becoming folded over, would cause laps. To 
 prevent the collars from cutting into the steel and thus forming laps all 
 the passes except the finishing are given a slight belly. A fillet at the base 
 of the collars serves to keep the corners of the piece well rounded. The 
 ragging on the rolls to increase the bite has already been referred to. As 
 to arrangement of the passes, different plans may be pursued, as may be 
 seen from a study of the accompanying drawings. The first two designs 
 
374 
 
 THE ROLLING OF STEEL 
 
 J 
 
 
 
 
 I - J 
 
 ] 10 r 
 
 
 i 
 
 
 , 
 
 00 s 
 
 N 1 
 
 
 
 
 
 nn 
 
BLOOMS 
 
 375 
 
 are for blooms, billets or slabs, whereas the third design can roll large 
 blooms only and billets only in connection with a roughing mill. 
 
 In the method shown in table 53 and fig. 60 the reduction is begun with the 
 ingot on edge, but when rolling soft steel, .08% to .22% carbon, the ingots are 
 sometimes rolled on the flat. This reduces the number of passes to seventeen, 
 the steel receiving only four passes in the first groove. In rolling blooms 
 of other sizes and slabs, about the same procedure is followed out, the steel 
 being given a sufficient number of passes to work it down to the required 
 size. In rolling some of the special steels, such as special drop forgings, 
 etc., it is often the practice to turn the steel after each pass in order to 
 avoid all danger of rolling in laps and seams. Sometimes where fairly sharp 
 corners are desired on the blooms they are given extra passes to hold the 
 edges up. This extra rolling is especially necessary where the blooms are 
 to be reheated in a continuous furnace, since if the corners are very rounded 
 the blooms, instead of sliding down, are liable to roll over the skids in the 
 bottom of the furnace. Furthermore, if the steel shows a tendency to 
 crack, the roller may nurse it along by taking lighter drafts. 
 
 Operation of Rolling: The sketch referred to above shows how an 
 18" x 21" ingot is broken down to a 6" x 4" bloom in nineteen passes. This 
 sketch, combined with the following table, gives about all the information 
 there is to give on this part of the work. 
 
 Table 53. The Rolling of an 18" x 21" Ingot. 
 
 No. OF 
 GROOVE 
 
 SIZE OF 
 GROOVE 
 ON ROLL 
 
 No. OF PASSES 
 AND MANIPULATION 
 
 18" x 21" INGOT 
 REDUCED TO 
 
 1 
 
 24" 
 
 2 
 
 18^" x 19" Bloom 
 
 
 
 Bloom turned 90 
 
 
 1 
 
 24" 
 
 4 
 
 12" x 19^" " 
 
 
 
 Bloom turned 90 
 
 
 2 
 
 12" 
 
 4 
 
 ll%" x 12^" 
 
 
 
 Bloom turned 90 
 
 
 2 
 
 12" 
 
 2 
 
 7M"xl2M" 
 
 
 
 Bloom turned 90 
 
 
 3 
 
 8' 
 
 2 
 
 7^" x 8M" \ 
 
 
 
 Bloom turned 90 
 
 - 
 
 3 
 
 8" 
 
 2 
 
 4" x 8" 
 
 
 
 Bloom turned 90 
 
 
 5 
 
 4" 
 
 1 
 
 6K"x 4^" " 
 
 
 
 Bloom turned 90 
 
 
 4 
 
 6" 
 
 1 
 
 5^"x 4^" 
 
 
 
 Bloom turned 90 
 
 
 4 
 
 6" 
 
 1 
 
 4" x 6" a 
 
 
 
 Finish 
 
 
 
 
 19 Passes 
 
 
376 
 
 THE ROLLING OF STEEL 
 
 
 
 *T 
 
 H 
 
 *-6 I 
 ^^ 
 
 KIT 
 
 IP 
 
BLOOMS 
 
 377 
 
 T 
 
 IT" 
 
 -T- 
 
 r 
 
 1 
 
 I 
 
 i 
 
 
 
 |M 
 
 
 % 
 
 cS 
 
 i 
 
 8 
 
 1 
 
 ( 
 
 
 i 
 
 i 
 
 
 1 
 
 i 
 
 
 V f^-L 
 
 ^ 
 
 
 I 
 
 
 
 c~!2J*-> 
 
 r- 8" 4 
 
 ----.24" 
 
 - I3|'/- 
 
 - 9|"H 
 
 s 
 
 
 
 
 Sf"- 
 
 FIG. 62. Another Design for 40-inch Two-High Reversing Blooming Mill Bolls. 
 
 Top and bottom rolls have same dimensions 
 FIG. 63. A 38-Inch Reversing Blooming Mill Designed to Roll a Fixed Size of Bloom. 
 
 SECTION IV. 
 
 EXAMPLES OF THREE-HIGH BLOOMING MILLS. 
 
 Plan of Study: Since a good idea of the relative dimensions of the 
 different parts of the blooming mill may be gained from the preceding 
 detailed description of the forty inch mill at Duquesne, such details, for 
 the sake of brevity, may now be omitted, and the description of the forty 
 inch three-high mill at Edgar Thomson be made more general with the idea 
 of' emphasizing the difference in construction and operation between the 
 two-high and the three-high blooming mills, only. 
 
378 THE ROLLING OF STEEL 
 
 The Engine and Connections: A tandem compound condensing 
 engine, size 50" x 78" x 60", furnishes the driving power for the mill. The 
 engine is housed in an engine room separate from the mill. It is provided 
 with a 75-ton flywheel, twenty-five feet in diameter, and runs at a speed of 
 54 r. p. m. This flywheel, supported between suitable bearings, is mounted 
 upon the driving shaft, which is connected, by means of a crab and coupling 
 box to the driving spindle. This spindle is nine feet eight and one-half 
 inches long, including ten inches at each end for the wobblers, and twenty- 
 one and three-fourths inches in diameter. It is supported at its center 
 by a stationary carrier bearing, and, extending through the wall of the 
 separate engine room, connects the driving shaft of the engine to the middle 
 pinion of the mill. 
 
 The Pinions and Spindles: The pinions, contained in three-high 
 housings similar to the roll-housings, are six feet four inches long over all, 
 and, when in place, measure forty inches from center to center of any two 
 adjacent ones. The lengths of the necks are twenty-one inches and their 
 diameters are twenty-two inches. These pinions are of the herring bone, 
 or helical toothed type. Unlike the reversing mill, where the use of a 
 vibrating spindle makes it necessary to set ^the pinions at some distance 
 from the mill, the three-high mill will be set with the pinions as close as 
 possible to the rolls; the roll spindles are, consequently, much shorter. 
 The spindles for this mill are four feet ten and three-fourths inches long 
 over all, and twenty-one and three-fourths inches in diameter. Each 
 spindle is supported at its center by means of a bearing mounted on a bar 
 that bridges the space between the inside roll housing and the opposite 
 pinion housing. Specially designed coupling boxes connect the spindles 
 with the pinions and the rolls. 
 
 The Roll Housings are of the open top type. At the top the window 
 of the housing is closed with a heavy cap, which is securely fastened to 
 the columns of the housing by means of heavy key bolts, the slotted ends of 
 which extend up through and above the ends of the cap. The screw down 
 passes downward through the center of the cap and rests on the top of the 
 upper bearing of the top roll, so that the pressure may be applied directly 
 over its center. The rigidity of the housings is increased by the use of 
 brace rods which extend from a height about the center of the top roll, 
 both fore and aft, to an anchorage provided by projections on the shoes. 
 The two housings are tied together by means of separators and bolts on 
 front and back attached just below the caps of the housings, so that they 
 are almost on a level with the upper side of the top roll. In width the 
 housings measure seven feet eight inches from center to center of the shoes, 
 and are approximately fourteen feet high from the lowest point in the base 
 to the top of the cap. 
 
 The Rolls are all of the same length and diameter of neck. The lengths 
 of the bodies are seventy-six inches, while the dimensions of the necks and 
 
THREE-HIGH BLOOMING MILL 
 
THE ROLLING OF STEEL 
 
 wobblers are the same as for the same parts of the pinions. As to the diam- 
 eters of the bodies, the three rolls are made different, the top roll being 
 the smallest and the bottom roll the largest and the middle roll of an inter- 
 mediate size. These diameters are such that the distance between the 
 centers of new rolls on new bearings is forty and fifteen-sixteenths inches 
 for the bottom and middle rolls and thirty-nine and eleven-sixteenths inches 
 for the middle and top rolls. This arrangement, the necessity for which 
 will be explained later, has the effect of throwing the rolls slightly out of 
 line with the pinions which measure forty inches from centers to centers. 
 This difference is distributed by centering the bottom roll five-sixteenths 
 inch below the center of its pinion and the top roll five-sixteenths inch 
 above the top pinion. The center of the middle roll is then five-eighths 
 inch above that of the middle pinion. The rolls are designed for seven 
 passes as shown on the accompanying sketch. The ingot, having been 
 reduced from 23%" x 23%" to 15^" x 18^" by four passes on a forty-eight 
 inch two-stand tandem bloomer, enters the first bottom pass of the forty- 
 inch mill on edge and is reduced in this and the six succeeding passes to a 
 9^" x 9J^" bloom. Hence, all the mills of the plant using blooms from 
 this mill are adjusted to take this size of bloom. It will be observed that 
 the edges of the collars are well rounded off to prevent the formation of 
 fins that might cause laps, and that the pitch line for the bottom passes 
 lies well below the clearance line of the rolls. The bottom and middle 
 rolls are made of steel, while the top roll is a sand roll. The greater strength 
 of the two lower rolls is required for the greater draught taken in the bottom 
 passes, which are edging passes. 
 
 Lifting Tables: The mill is provided with two lifting tables, each of 
 which is twenty-one feet seven and nine-sixteenths inches long from center 
 to center of the first and last rolls. Each table has a vertical motion only, 
 and is supported on four legs or vertical shafts, one at each corner, which 
 are connected to lever arms mounted on shafts with other arms for counter 
 weights and lifting. The torque of the counter weight just about equals 
 that produced by the table. The material is then raised and lowered by 
 a reversing electrical motor, which is provided with a magnetic brake for 
 automatically stopping the tables at the correct levels. By means of a 
 long lever arm, the two tables are connected and are raised and lowered 
 in unison. The one on the approach side of the mill is provided with 
 stationary vertical skid bars, or transfer fingers, between the table rolls, 
 which are so arranged that the act of lowering the table edges and trans- 
 fers the piece to the next bottom pass. The bloom from the forty-eight inch 
 mill is edged to enter the forty inch mill by means of single collar rolls. 
 Shears of the side cutting type, electrically operated, are provided fifty- 
 six feet from the roll table for cutting the piece into blooms of the desired 
 length after the required discard has been sheared off. 
 
THREE-HIGH BLOOMING MILL 381 
 
 Roll Design for Three=High Bloomers : The peculiarities, previously 
 pointed out, in the size, the arrangement, and the grooves of the rolls for 
 this mill are common to three-high bloomers, and represent the effort on 
 the part of the designers of the rolls to overcome certain difficulties inherent 
 in this type of mill. First, in order to avoid weakening the rolls by increas- 
 ing their length unduly, only a small number of passes, usually nine, are 
 available. Second, except at rail mills, which are the only mills in exist- 
 ence where any preliminary reduction of the ingots is made, this limited 
 number of passes means that very heavy drafts must be taken in order to 
 reduce the ingot to the more common bloom sizes. Third, in order to get 
 in the greatest possible number of passes on a set of rolls, the passes must 
 be placed one above the other, hence a groove in the middle roll must 
 serve for both an upper and lower pass. Fourth, the peripheral speed at 
 the base of the grooves in any two rolls forming a pass must be equal, or 
 nearly so, if the piece is to roll without curling when coming out of the 
 pass. * However, the pass diameter of the top roll for any pass may be a 
 little larger than the bottom, for then the piece will be held down but 
 may be prevented from curling down by the guards on the mill. On forty 
 inch mills this difference is about one-fourth of an inch and is determined 
 by practice. 
 
 An Example of Roll Design for Three-High Blooming Mill will 
 perhaps be the best answer to the question as to how all these conditions 
 are met. A specific problem and a method of solving it are hereby given: 
 
 Given: Ingot 21" x 23", Bloom 9" x 10", number passes 9, Size of Mill 42". 
 Required: To design rolls for the mill. 
 
 Solution: First: The draught on each pass is found. In finding 
 the draughts it is to be borne in mind that the draughts on the bottom passes , 
 being edging passes, should be heavier than on the top passes; that it is well 
 'to take the heaviest draughts on the first bottom passes while the steel 
 is hot and the piece is short, which will prevent great strains on the engine 
 as the momentum of the fly wheel will carry across a short length; that 
 the top passes are best made of equal draughts; and that little work can 
 be done on the finishing pass. The reduction in size of the ingot to the 
 bloom calls for twelve inches on one side and thirteen inches on the other, 
 or a total of twenty-five inches. Since the reductions on the bottom passes 
 are to be greater than those on the top, let this total draught be appor- 
 tioned to give ten inches on top passes and fifteen inches on bottom ones. 
 The draught on each top pass will then be two and one-half inches. The 
 draught on the bottom passes may be arbitrarily apportioned, but to accord 
 with the cautions stated above, they are determined by trial, and to give 
 the total reduction of fifteen inches they should, apparently, be apportioned 
 as follows: No. 1 pass, 3^"; No. 3 pass, 3%"; No. 5 pass, 3M"; No. 7 pass, 
 
THE ROLLING OF STEEL 
 
 3/4" > No. 9 pass, 1". The complete plan for working the ingot down to 
 
 size would then be as follows: 
 
 Size of the original ingot, 21" x 23". 
 
 No. 1 Pass, bottom, draught 3}/"; size of bloom produced, 21" x 193^". 
 
 No. 2 top 2Y 2 "\ " " " " 21"xl7". 
 
 Piece edged. 
 
 No. 3 Pass, bottom, draught 3%"; size of bloom produced, 17%" x 17". 
 No. 4 top, " 2^"; " 14%"xl7". 
 
 Piece edged. 
 
 No. 5 Pass, bottom draught 3^"; size of bloom produced, 14%" x 13%". 
 No. 6 top, " 2^"; " " 14%"xll". 
 
 Piece edged. 
 
 No. 7 Pass, bottom, draught 3%"', size of bloom produced, 11^" x 11". 
 No. 8 top, ,2^"; " " " 9" x 11". 
 
 Piece edged. 
 No. 9 Pass, bottom, draught 1"; size of bloom produced, 9" x 10". 
 
 Second: The most suitable pitches for the rolls are determined. 
 
 By pitch is meant the distance from center to center of a pair of rolls with- 
 out clearance, or the distance between any two of the pitch lines. These 
 are based on the size of the mill; the average pitch for the top and bottom 
 sets of passes are equal to its size, and should be such, for reasons already 
 noted, that each roll will be approximately one-fourth inch less in diameter 
 than the one above it. The pitch is, therefore, determined by trial as 
 follows: In this case the proper figures appear to be 
 
 " from center to center of top and middle roll. 
 
 bottom and middle roll. 
 (84"-H2=42"=the size of the mill.) 
 
 From these figures the working diameter of the passes are found as 
 follows: 
 
 43>i" pitch of bottom and middle roll. 
 
 19^" height of first pass, (See size of bloom for No. 1 pass.) 
 2 1 23%" pass diameter of first pass. 
 
 11%" ^"=11%" first pass radius of bottom roll, or first pass working 
 diameter=23^". 
 
 M"=llJir / first pass radius of middle roll, or first pass working 
 diameter=23M". 
 
 17" (size of No. 2 pass)=28K". 
 
 (pitch of top and middle roll) 28%"=12", first pass radius of 
 the top roll; or first pass working diameter of top roll=2 // xl2"=24. 
 As this diameter is a little larger (J") than that for the middle roll, 
 the pitches assigned above are assumed to be the proper ones. 
 
THREE-HIGH BLOOMING MILL 
 
 383 
 
384 THE ROLLING OF STEEL 
 
 Third: The size of each roll is determined. As a preliminary step 
 to finding the size of the rolls, the diameters of the middle and bottom 
 rolls may be assumed to be the same as their pitches, forty-three and one- 
 eighth inches. The pitch size for the middle and top roll is forty and 
 seven-eighths inches, and if from twice this pitch the diameter of the 
 middle roll is subtracted, the pitch diameter of the top roll is the result, 
 which in this case would be thirty-five and five-eighths inches. If, now, 
 the diameter of the working pass in the middle roll is subtracted from this 
 diameter of the roll, the result is twice the depth of the groove. 
 
 43^" 23%"=19^". 19%"-^2=9%" depth of first groove in middle roll. 
 Similarly, 38%" 24=14%". U%"-+-2=7%", depth of groove for the top 
 
 roll. The sum of these two quantities is seventeen inches which checks 
 with the size of second pass. But more of the piece lies in the middle 
 than in the top roll, so in order to get the same height of collar, eight and 
 one-half inches in each roll, it is necessary to increase the radius of the 
 top roll and decrease the radius of the middle by nineteen-sixteenths inches, 
 making their respective diameters forty-one inches and forty and three- 
 fourths inches. The diameter of the bottom roll would then be forty-five 
 and one-half inches (2 x 43^" 40%"). In order to get the proper clearance 
 between the rolls, which is assumed to be one inch, these diameters are 
 further reduced to forty inches, the diameter of top roll; thirty-nine and 
 three-fourths inches, the diameter of middle roll; and forty-four and one- 
 half inches, the diameter of bottom roll. These diameters give a much 
 deeper groove in the bottom roll than in the middle, which can be over 
 come by cutting down the collars on the bottom roll, which has the effect 
 merely of increasing the clearance. So to balance up the depths of these 
 grooves a clearance of two inches is allowed between these rolls, making 
 the final diameter of the bottom roll forty-two and one-half inches. The 
 finding of the depth of the remaining grooves is a simple manner. Thus, 
 for example, No. 4 pass is 14%" x 17". From the depth of the pass, fourteen 
 and three-fourths inches, the clearance of one inch is subtracted, leaving 
 thirteen and three-fourths inches. This depth is equally divided between 
 the top and middle rolls, making six and seven-eights inches in each. It 
 follows that the groove in the middle roll for No. 3 pass must be the same. 
 As the depth of this pass is seventeen and one-fourth inches, the groove in 
 the bottom roll must be eight and three-eighths inches, (17%" 6%" 
 2"8%"). The accompanying sketch shows a set of rolls designed 
 according to the explanation given above. To take care of variations due 
 to wear in the rolls and permit of their being dressed, thus increasing their 
 life, the entire set is made a little over-size in diameter of body, usually 
 about three-fourth of an inch. They are discarded when they have been 
 dressed down to the same amount undersize. 
 
V 
 
 SLABS 385 
 
 SECTION V. 
 
 THE ROLLING OF SLABS. 
 
 The Rolling of the Slab is the first step in the rolling of plates, just 
 as the bloom marks the first step in rolling the many shapes. Attention 
 has already been called to the rolling of slabs on. the reversing blooming 
 mill. For rolling narrow slabs, the blooming mill meets all the require- 
 ments, but the width of the slabs rolled on these mills is limited to the 
 maximum spread of the rolls on account of the necessity of edging the piece 
 near the last passes. In America, therefore, slabs are rolled for the most 
 part on the universal mill principle, in which the width of the slab is partly 
 controlled by means of vertical rolls which work on the edges of the slab. 
 The slabbing mills are not true universal mills, however, but double or 
 duplex mills, made up of one stand of rolls, similar to the blooming mills 
 but with plain instead of collared rolls, and one stand of vertical rolls near 
 to and in front of the horizontal rolls. Each mill is driven independently, 
 and both are reversing. By such an arrangement larger ingots may be 
 rolled than would be possible on the reversing blooming mill. Since the 
 piece is not edged under the horizontal rolls, ingots varying in thickness 
 and slabs of great width may be handled. The following sizes as to thick- 
 ness and widths of ingots are rolled by the thirty-two inch mill to be 
 described later, 23%" x 23%"; 26" x 40", 26" x 45", 26" x 48"; 26" x 53" 
 and 27" x 57". The thickness of the ingot is limited by the maximum 
 height to which the top horizontal roll may be lifted, while the width is 
 controlled by the spread of the vertical rolls. In preparation for the rolling, 
 the ingots are treated in soaking pits in the same manner as that already 
 described for the blooming mills. 
 
 The Thirty-two Inch Mill at Homestead as an Example of a 
 Slabbing Mill: This mill is an old mill and was originally designed to 
 roll armor plate. It is, therefore, somewhat larger and stronger than some 
 more recently constructed slabbing mills. However, the main features and 
 the principles of both the construction and operation are the same on this 
 mill as those of other slabbing mills. As noted above, the mill consists of 
 two separately driven stands of rolls the horizontal and vertical stands, 
 which are best described separately. 
 
 The Horizontal Mill : The 'rolls on this stand are four in number, 
 arranged one above the other on the plan of a four-high mill. Only the 
 two intermediate rolls actually come in contact with the ingot, however, 
 the topmost and bottommost rolls being used as reinforcing or stiffening 
 rolls to the two intermediate ones. All these rolls are nine feet two inches 
 long in the body, but the re-enforcing rolls are thirty-two inches in diameter, 
 while the intermediate ones are twenty-six inches in diameter. This 
 arrangement permits a more rapid reduction of the ingot and with less 
 
386 THE ROLLING OF STEEL 
 
 power than would be possible with only two rolls, which would have to be 
 of large diameters to give the great strength required. The smaller roll, 
 exposing little surface to the steel, sinks into the metal with less pressure 
 and is turned with less power. The four rolls are held in place by a cast 
 steel housing. The necks of the bottom re-enforcing roll rest on bearings 
 fitted into the bottom of the housing; this roll then supports the lower 
 intermediate roll, the contact being made the entire length of their bodies. 
 Lateral displacement of this lower intermediate roll is prevented by 
 babbitted side bearing boxes at either end. The two upper rolls are held 
 in two steel frames, one at each end, each of which is fitted with a brass 
 top bearing for the re-enforcing roll and a box fitted with bottom and side 
 bearings for the top intermediate roll. As these frames move up and down 
 with the adjustment of the top rolls, guide bars bolted to the outer edges 
 of the windows are provided to hold them in place, while liners inserted 
 between the frames and the sides of the windows prevent the wearing away 
 of the housings. The ends of two plunger rods rest against the bottom of 
 this frame while the rods extend to hydraulic cylinders which, located 
 beneath the mill and acting under a pressure of 600 Ibs. per sq. in., are 
 used for raising the top rolls. The rolls are lowered by means of screws 
 similar to those in the blooming mill, but in this case the power for the 
 screw down is obtained from a 60 h. p. motor mounted on a platform a 
 little above the top of the housing. The screws rest on breaker blocks 
 which serve as a safety to prevent the breaking of the rolls. The maximum 
 lift of the mill is nearly forty inches. For indicating the distance between 
 the rolls a gauge pole and disc are provided. The disc is mounted on top 
 of one of the screws with the marking pole adjacent to it. The circum- 
 ference of the disc is divided into 100 equal parts, while the pole is divided 
 into spaces of one inch each. These divisions on pole and disc are plainly 
 marked and permit the opening of the mill to be read to within 1-100 of 
 an inch. The Drive for the horizontal mill is connected to the intermediate 
 rolls, the re-enforcing rolls being friction driven. The motive power is 
 furnished by a 40" x 54" horizontal reversing engine, which is indirectly 
 connected to the leading or driving spindle of the mill. 
 
 The Vertical Mill is located about ten feet in front of the horizontal 
 mill, measuring from center to center of the rolls. Like the horizontal 
 mill the vertical mill has four rolls, two of which are re-enforcing; but these 
 rolls are much smaller than the horizontal ones, being only eighteen inches 
 in diameter and about forty-four inches in length, or a little longer than the 
 lift of the horizontal mill. The rolls are supported vertically in the housings 
 by means of bearing boxes at both the tops and bottoms. These boxes are 
 held in place by heavy rest bars which extend across the mill at top and 
 bottom and from housing to housing, between the windows of which they 
 are securely fastened. For adjusting the spread of the rolls inwardly two 
 screws, acting horizontally through the sides of the housings instead of 
 
THE SLABBING MILL 387 
 
 through the top as for horizontal rolls, are provided. They bear on a 
 frame that extends from the top to the bottom bearings, and are operated 
 by a 50 h. p. motor through a system of gears. For spreading the rolls apart 
 hydraulic jacks are used. In operating the mill, all four of the rolls are 
 driven. Starting with the engine, the connections are made as follows: 
 The engine, a 36" x 57" horizontal reversing steam engine, is mounted upon 
 a concrete foundation a little above the level of the bottom of the housings 
 and on the opposite side of the mill from that on which the engine for the 
 horizontal rolls is placed. The power is indirectly transmitted through 
 two gears, one above the other, to the driving shaft, which extends from 
 the upper gear to the farther side of the mill. To this shaft, the two outside 
 rolls are connected by means of bevelled crown and sleeve gears, while 
 a second set of gears connecting the inside and outside rolls in pairs furnish 
 the means by which the driving of the inside rolls is effected. The hori- 
 zontal rolls are run at full speed, while the speed of the vertical rolls is 
 controlled by the engineer to suit that of the horizontal mill. While the 
 maximum spread of the vertical rolls is about sixty-five inches, the widest 
 slabs rolled are only fifty-four inches wide, because this is the greatest 
 width the mill shears are built to cut. The excessive length of the hori- 
 zontal rolls (one hundred ten inches in the body, as previously given) is 
 explained by the fact that this mill was formerly used for rolling 
 armor plate. 
 
 Precautions to be Observed in Rolling Slabs: The essential part of 
 the rolling is the determination of the draughts to be taken on each pass 
 through the mill. This determination is made by the roller from the 
 dimensions of the ingot to be rolled, the slabs desired as given on the rolling 
 order sheets, the temperature of the ingot as it comes to the rolls and the 
 steam pressure available for the engines. The temperature controls the 
 draught in that the hotter the steel the less the pull on the mill and the 
 greater is the possible draught. Uniform temperature throughout the 
 ingot is also necessary to insure good rolling, as steel hotter on one side 
 than the other causes curling of the slabs, due to the fact that steel always 
 curls towards the cold side, because the elongation is less on that side. 
 To assist in rolling, ingots are fed to the rolls with the hot side up and 
 the small end first, thus affording a better grip by the rolls and preventing 
 or lessening the tendency for the slab to curl up when leaving the mill as 
 they do when the cold side is turned up. Slabs are often hotter on one 
 side than the other, which condition also causes curling, due to the greater 
 spreading or flowing of the steel on the hot side. Indirectly, the chemical 
 composition of the ingot regulates the possible draught; high carbon steel, 
 for example, cannot be heated as hot as common plate steel, hence longer 
 time and smaller draughts must be taken in the rolling. The total vertical 
 draught, which in all cases, except 27" x 57" ingots, amounts to about one 
 inch, is taken during the first few passes. From then on, the vertical rolls 
 
388 THE POLLING OF STEEL 
 
 are kept in contact with the steel at a pressure only sufficient to prevent 
 tearing of the edges which results when no pressure is applied on the sides 
 of the slab. With small ingots tearing of the steel is also caused when the 
 ingots are not hot enough for good rolling but still capable of being passed 
 through the rolls. In the case of large ingots that are cold, there is little 
 danger of injuring them, because there is not sufficient power to roll them, 
 as in the case of smaller ingots. Correct lining of the rolls is necessary to 
 make good slabs, since if the rolls are crossed or are higher on one end 
 than on the other the slabs curl. At the thirty-two inch mill the driving 
 end of the rolls is always kept slightly higher than the other end to allow 
 for the more rapid wearing of the bearings due to the extra weight on 
 that side of the mill. Above y%" the slabs will curl in passing through 
 the mill. The maximum draught, i. e., reduction in sectional area, that 
 can be taken by the horizontal rolls with the steel at a good rolling 
 temperature is approximately thirty square inches on the entering pass and 
 forty square inches on the return pass. The difference in reduction possible 
 on the entering and return passes is due to the fact that, on the return pass, 
 the vertical rolls aid in pulling the slab through the mill, while they cannot 
 effect any power by pushing the slab when going in the opposite direction. 
 For the entering pass thirty divided by the ingot width gives the approxi- 
 mate draught, or bite, and forty divided by the ingot width, the return 
 pass bite. The last pass taken is entering, and is a pass in which very little 
 pressure is used in order to straighten the plate, roll down the top ends, and 
 remove the convex surf ace due to the spring from the rolls. 
 
 Removal of Scale: During the rolling of an ingot the scale must be 
 removed from the surface to prevent the slab, and resulting plates, from 
 being pitted. The process employed in removing the scale depends upon the 
 kind of steel being rolled. For removing scale on low carbon steel salt and 
 water, the latter being sprayed on the slab at high pressure and the former 
 thrown on with scoops, are very effective. In case of high carbon steel the 
 scale sticks more firmly to the slab, and burlap sacks are used, as neces- 
 sary, in addition to salt and water. When nickel steel is rolled, coal is 
 used in place of salt, and burlap is also thrown under the rolls. Brush or 
 green twigs are often employed to serve the same purpose as burlap. The 
 actions of all these substances are somewhat similar. In each case the 
 substance is drawn under the rolls, which tend to bring it rapidly into 
 close contact with the hot metal. The material thus caught by the rolls 
 is gassified by the heat, and, in an effort to escape, the gases get beneath 
 the scale and carry it off with them. Coal and burlap, being less volatile 
 than salt or water, are carried a little farther beneath the rolls and give 
 a more violent action. Nickel scale is the most difficult of all to remove, 
 and if the first scale is melted off in the soaking pits and a second formed, 
 it is almost impossible to clean it off. All nickel bearing slabs are cleaned 
 first on one side, then turned over by cranes and cleaned on the other side. 
 
SLABS 389 
 
 Shearing Slabs at the Thirty=two Inch Mill: From the rolls the 
 slabs pass by means of motor driven roll tables to a hydraulic shear. Two 
 plungers are used to operate the knife, a small one on top for lifting the 
 blade and a large one on the bottom for pulling the knife down against the 
 steel, thus effecting the cutting. Two Wilson and Snyder pumps, an accumu- 
 lator and steam intensifiers comprise the operating equipment. The slabs 
 are cut to length by means of a scale of marks placed on a steel slab in front 
 and to one side of the shears. Graduations on this marker indicate the 
 distance from the shear-blade, so by running the slab out to any certain 
 mark the length of slab is indicated by the graduation. The roll table 
 approaching the shear-blade is also graduated in inches of distance from 
 the blade. By fixing the eye on any spot or mark on the slab at any distance 
 from the knife as shown by position on scale and watching this mark until 
 it is moved beneath the knife, the length of slab can be obtained. The 
 size (total weight) of the slab is determined by the dimensions and gauge 
 of the plate into which it is to be rolled. Since the width of the ingot 
 limits the width of the slab, planning the size of the slab starts with the 
 selection of an ingot of the proper size. Next, the thickness of the slab 
 must be determined, and then the length. It is evident that very careful 
 work is necessary in making up the mill schedule, if the steel is to be rolled 
 to best advantage. All this planning is done in the mill office, and the 
 shearman is generally given the lengths into which the slabs are to be cut, 
 though occasionally he may be ordered to cut to best advantage. The first 
 cut made on the slab is to remove the piped end. After the discard is 
 sheared off, a few slabs are cut, when the piece is turned around with a 
 manipulator, the bottom crops taken off, and the remainder cut into slabs. 
 From the rolling sheets, the shearman gets the length of slabs ordered and 
 the amount of discard that is to be taken before slabs can be cut. Cuts are 
 usually made up to the center of the slab before turning it around. By 
 turning the slab the shearman can tell how much steel will remain after 
 taking off the bottom crop and better decide as to how he shall cut the 
 remainder. Slabs are cut as ordered if possible, and if a piece is left over 
 that is too large, or heavy, for any slab ordered, it is marked as an "odd" 
 cut. For example, a 5000 pound slab is ordered, and, after the first slabs 
 have been cut, the remaining piece is 7000 pounds, or 2000 pounds over the 
 weight desired. Were this 7000 pounds to be used on a slab calling for only 
 5000 pounds, 2000 pounds would be scrapped in the plate mill. This system 
 is too expensive, so the slab is marked as an odd cut and placed on some 
 other order. The limits as to width and thickness of slab that may be 
 sheared on this shear are fifty-four inches and twenty inches, respectively. 
 The percentage of discard varies from about 15% on plain steel to 35% on 
 some special orders; the larger portion of the discard is taken from the top 
 of the ingot on account of the segregation and piping being mostly confined 
 to this section. All discarded steel is placed in open hearth charging 
 buggies and shipped to the open hearth, note being taken of the special 
 
390 THE ROLLING OF STEEL 
 
 alloy steel scrap, which is kept separate from the plain steel scrap. As the 
 slabs are cut, they are stamped with a serial number, beginning with one 
 from the first of the year. A recorder takes down the slabs made, size, 
 number, cut, etc., and enters it on the product side of the rolling order 
 sheet. The weight of ingot, weight of slabs made, and weight of scrap is 
 noted, and the information sent to the product department. From this 
 data the practice of the mill is figured. 
 
BILLETS AND OTHER PRODUCTS 391 
 
 CHAPTER VI. 
 
 THE ROLLING OF BILLETS AND OTHER SEMI-FINISHED 
 PRODUCTS. 
 
 SECTION I. 
 
 THE THREE-HIGH BILLET MILL. 
 
 General Features of Rolling Billets: A large percentage (over 50%) 
 of the steel produced is rolled into material of very small section. In order 
 to finish their product in one heat, the mills rolling such sections must 
 start with small billets. While many blooming mills of the reversing type 
 are able to roll billets as small as 4" x 4" or less, which is a size much too 
 large for the majority of the smaller mills, the inadvisability of employing 
 these large mills for rolling billets is at once evident, and accounts for the 
 existence of the billet mill. In order to reduce the cost of the billet, the 
 billet mill will be placed just after the blooming mill so as to effect the 
 reduction from ingot to billet on the original heat of the former. As to 
 the kinds of mills used for rolling billets, almost any mill of medium size 
 may be adapted to the work. Since the section is a very simple one and 
 so little in the way of accuracy as to form of section or of finish is necessary, 
 about the only requirements of the billet mill is that it be heavy enough 
 to handle fairly large blooms and speedy enough to reduce the piece to the 
 desired size before it becomes too cold. For the larger sized billets, a 
 single stand of three-high rolls placed after the bloomer serves very well, 
 but for small billets that are not intended for certain special purposes, like 
 forgings, for example, the continuous mill is the best mill for the purpose. 
 
 Example of Three-high Billet Mill The Twenty-eight Inch Mill 
 at Duquesne: As an example of the former type of mill the twenty-eight inch 
 mill at Duquesne will be described, because every device is employed to 
 increase the out-put, and it also is an example of how the mills are compelled 
 to adapt themselves to change in conditions. This mill was originally 
 designed as a rougher for a rail mill, but was rebuilt in 1907. The mill is 
 fed by a thirty-eight inch two-high bloomer which reduces an 18}/" x 20^" 
 ingot to a 7J4" x 8%" bloom. The twenty-eight inch mill at Clairton 
 placed after the forty-inch bloomer is very similar to the Duquesne twenty- 
 eight-inch mill. 
 
 Engine: The mill is direct driven by a Cooper Corlise tandem 
 compound horizontal condensing engine, 44" x 74" x 54", designed to 
 develop 2500 h. p. at 75 r. p. m. and 25% cut-off and to run at 80 r. p. m. 
 
392 THE ROLLING OF STEEL 
 
 at 120 to 125 pounds steam pressure. The engine is capable of developing 
 4500 to 5000 h. p. The ordinary speed is about 62 r. p. m. and the normal 
 load about 1200 h. p., the maximum being about 3500 i. h. p. per pass. The 
 steam consumption is about 300 pounds per ton of steel rolled. The exhaust 
 from the low pressure cylinder is taken to a central condensing plant 
 near the engine house; this plant is of the Weiss barometric type and 
 is equipped with the following apparatus: One 20" x 42" x 24" air pump 
 and two Wilson-Snyder 18" x 30" x 26" x 36" 8,000,000 gallon duplex com- 
 pound-plunger water pumps. The engine is controlled in the engine house 
 by a sixteen inch throttle valve and may be shut down quickly in an 
 emergency by a sixteen inch quick-closing valve just above the throttle. 
 
 Drive: The crank shaft is connected through a flexible coupling to 
 the spindle shaft. The cast steel coupling, five feet six inches in diameter 
 is keyed tight to the Crank shaft of the engine with a bronze half thrust 
 collar over the half coupling; a .80% carbon steel wearing plate is screwed 
 to the outboard bearing support of the engine and separates the collar from 
 it. Eight two and three-fourths inch bolts hold a second half coupling to 
 the first one. The former fits over a cast steel hub keyed to the engine 
 end of a cast steel spindle shaft, twenty inches in diameter. The engine 
 end of the spindle shaft is slightly curved to promote flexibility. The spindle 
 shaft is seventeen feet eight and one-half inches long and twenty inches in 
 diameter, and three feet seven and one-half inches from its end is the center 
 line of a 20" x 48" ring oil bearing, which supports the mill end of the spindle 
 shaft. The oil bearing is held in a cast iron yoke and base, mounted on a 
 cast iron hot plate. The bearing is lined with babbitt and is provided with 
 small oil grooves for lubricating. A tight crab, four feet in diameter, of 
 cast steel is keyed to the mill end of the spindle shaft; two and one-half 
 inch bolts hold a four foot cast steel loose crab to the tight crab. A 
 cast steel compound coupling fits over the mill end of the loose crab and 
 the adjacent wobbler of the leading spindle. The cast steel leading spindle 
 is three feet ten and one-half inches long and sixteen inches in diameter. 
 A plain cast steel coupling box joins the leading spindle to the middle of 
 the mill, fitting over the adjacent wobblers of each. 
 
 Pinions and Their Housings: The pinion housings for the twenty- 
 eight inch mill are steel castings bolted to the mill shoes; the housing 
 windows are seven feet three inches deep from the top to the sill and twenty- 
 six inches wide; one forged steel liner one inch thick is used on each sill, 
 and each window is faced on each side with a one inch forged steel liner. 
 These are all bolted to the housings. All these liners are forged steel of .40% 
 to .50% carbon. The pinions are steel castings of the helical tooth type. 
 They measure thirty-six inches in length of face, and have thirteen teeth 
 with a pitch diameter of twenty-nine inches. Their diameter at the shrouds 
 is twenty-seven inches, and the necks taper from seventeen and three- 
 eighths to seventeen inches in diameter; the wobblers are sixteen inches in 
 
THREE-HIGH BILLET MILL 393 
 
 diameter. The total length of the pinions is nine feet two inches. All six 
 pinion bearings are of the same pattern, being made of cast steel with three- 
 fourths inch babbitt and four narrow brass plates set 90 apart. The 
 bottom bearings rest flat on the housing sill liners, no beveling being 
 required; the bearings are twenty-three inches wide, twenty-eight inches 
 high, and twenty-three and one-half inches through. The cap for the 
 pinion housings is a solid steel casting fitting over both housings; it has 
 slots at its four corners for key bolts to hold down the pinions tightly beneath 
 it. Steel eye bolts are set in the caps, so that they can be lifted easily. 
 The housings are held in line also by steel separator rods on each side, 
 top and bottom. The bottom and top pinions are driven by the middle 
 pinion and all three are connected to their respective rolls by cast steel 
 coupling boxes and cast steel spindles, unsupported. 
 
 Housings and Roll Bearings: The roll housings are cast steel, closed 
 at the top with a cast steel cap; the housings are bolted to cast iron mill 
 shoes. The windows of the housings are nine feet two inches deep and two 
 feet nine inches wide. Three feet one and one-fourth inches above the sill 
 is a ledge on which rests the bearing for the middle roll; this roll is 
 held stationary, the others being adjusted to it. Bearings for this mill are 
 as follows: Bottom roll: two steel carrier bearings with babbitt and 
 brasses. Middle roll: same, and two rider bearings with similar babbitt 
 and brasses. Top roll: two forged steel babbitted saddles for carrying 
 the roll; two cast steel rider bearings with babbitt and brasses. The 
 middle roll is held down on its ledge by three and three-quarter inch rods, 
 which are in turn held down by two five inch set screws of one and three- 
 quarter inch pitch through the caps; the rods press on the rider bearings 
 of the middle roll. The screws are adjusted by means of wrenches which 
 fit over nuts at the top of the screws. The carrier bearing of the bottom 
 roll rests on a seat which is fastened to a seven-inch screw running up from 
 below the housing through its center; this screw turns in a charcoal iron 
 nut shrunk in the mill housing, and is regulated by a gear and pinion con- 
 nection from outside the housing. The gear is moved by a vertical rod 
 with a slotted wheel in the top; a hand lever is used to turn this wheel, 
 and thus the bottom roll is raised or lowered. The top roll is held up by 
 two counterweights through steelyard rods in each housing reaching up to 
 the top roll's carrier bearing; the roll is held down at each end by a single 
 seven-inch screw of one inch pitch reaching through the cap to the breaker 
 block on the rider bearing. The screw is adjusted like a bolt, with a short 
 wrench usually turned by a crane. 
 
 Rolls: In this mill the top and bottom rolls are similar and inter- 
 changeable, while the middle roll differs in that the barrel of the roll has 
 larger diameters than the bottom roll in those passes in which it acts as 
 the top roll, and smaller diameters where it acts as the bottom roll when 
 paired with the top roll. In passes Nos. 1, 3, 5 and 7 the diameters are 
 
394 
 
 THE ROLLING OF STEEL 
 
 If 
 
 FIQ. 66. Three-High Billet or Roughing Mill. 
 
THREE-HIGH BILLET MILL 395, 
 
 larger in the middle roll than in the top and bottom and in Nos. 2, 4 and 6 
 they are smaller. The rolls are thirty and three-eighths inches in diameter, 
 eleven feet four inches long over all and six feet four and one-half inches 
 long in the body; the passes are shown in the accompanying sketch. 
 One-sixteenth inch ragging is used. The rolls are made up in sets 
 of four, two middle rolls, one top, and one bottom roll making 
 up the set. The rolls are changed every two weeks, when a new 
 middle roll is inserted and the bottom and top interchanged, as 
 only passes Nos. 1, 3, 5 and 7 have been used in the bottom roll and passes 
 Nos. 2, 4 and 6 in the top roll; all seven passes have, of course, been used in 
 the middle roll. At the end of a four weeks' period, the entire set is returned 
 to the roll shop for dressing. About five sets are kept in stock. Recently 
 the diameter of the rolls have been increased, which permits more dressings 
 and gives longer life. Both adamite and sand cast iron rolls are used here, 
 as the reductions of the piece are small and no great strength is required. 
 Adamite rolls are annealed and are very hard. The sand cast rolls are 
 ordinary cast iron of the following composition, approximately: 1.87% 
 total carbon, 1.22% graphitic carbon, .65% combined carbon, .37% manga- 
 nese, .070% sulphur, .920% phosphorus, .70% silicon. The rolls weigh 
 18,000 to 19,000 pounds each. The passes are 7^", 7^", 6^", 6M", 5M", 
 5%" and 4" wide. The sketch, (Fig. 66) shows the relative dimensions. 
 When new, the rolls have a collar of thirty and three-eighths inches diam- 
 eter and three-fourths inch is taken off at each dressing of the cast iron 
 rolls and one-fourth inch for the adamite rolls; the rolls are scrapped when 
 the collars have been turned down to a diameter of twenty-eight and one- 
 fourth inches. 
 
 Guide Cages: Two inches above the center line of the bottom roll, 
 lugs are attached to front and rear of the roll housings to support guide 
 cages. These cages are cast steel frames for holding up the guides used 
 on all passes of this mill. They are bolted to the housings. The front 
 guide cage is six feet six and three-fourths inches long, reaching almost to 
 the face of the housing, and is about five feet high. It contains closed holes 
 in front of all passes using the bottom roll and in front of No. 6 pass of the 
 top roll; slots are provided in front of Nos. 2 and 4 passes. The rear guide 
 cage is practically the same as the front but has all open slots in front of 
 the top roll passes. Cast steel guides and side guards are bolted to these 
 guide cages. 
 
 Tables: The bloom from the thirty-eight inch mill comes from the 
 shear tables to the engine side of the roll table for the twenty-eight inch 
 mill. This table, together with the rear table, is of the lifting type and 
 is raised and lowered as a unit with the rear table. The front table contains 
 twelve cast steel rollers, each of which has five collars for turning the 
 billets. The size of these collars, beginning at the engine side, are: 16" x 
 2", 14" x 12K", 15" x 12^", 16" x 12^", and 16" x 1^". This arrange- 
 ment allows four grooves, 9^", 8J^", 7M" and 6" from end to end. The 
 
.396 THE ROLLING OF STEEL 
 
 diameters of the rollers at these grooves are 9", 10", 11" and 12", respect- 
 ively. The rollers are driven by a Crocker-Wheeler 75 h. p., 220 volt 
 series wound D. C. Motor. There are side guards on the edges of the table 
 and at the front end are side guards for putting the bloom from the thirty- 
 eight inch mill into the proper pass and for protecting the other grooves. 
 The table is thirty-four feet ten inches from center-line to center-line of 
 the end rollers and is about, six feet wide, inside. Coupled to the front of 
 the table at the last groove is an extension table consisting of four dead 
 rollers protected by side guards; it is fifteen feet long and fourteen inches 
 wide and is used as an extension for the bar when ready for the seventh 
 pass of the mill. Both front and rear tables are raised and lowered by 
 means of fourteen and twenty-one inch plungers operated by a hydraulic 
 cylinder; the hydraulic apparatus is located under one end of the front table 
 and is connected to each table by bell-cranks from a main shaft attached to 
 the cross-heads. The front table is equipped with a stationary manipulator 
 for advancing the bars from pass to pass; it consists of four sets of three 
 and one set of two cast steel fingers bolted to pedestals on the foundation 
 of the mill. The fingers are set between rollers Nos. 1 and 2, 4 and 5, 7 
 and 8, 9 and 10, and 11 and 12, in line with the wide collars; they are flat 
 cast steel plates mounted vertically and with their tops bent at an angle 
 giving a 45 slope in the direction it is desired to move the piece. The 
 fingers do not reach above the level of the pass when the table is elevated 
 and the bars run out on the collars of the rollers; as the table sinks, the bars 
 encounter the stationary fingers and slide down into the next groove. The 
 rear table, as mentioned, is operated through the same shaft as the front 
 table, but owing to the fact that it must raise the bars from the lower 
 roll to the middle one and advance them one pass, it has to travel through 
 an arc in rising to bring its grooves in line with the next passes. This is 
 done by causing the table to slide toward the next pass as it is raised by 
 the use of pull-over rods attached to pedestals on the proper side of the 
 bottom of the scale pit; when lowered, the table slides back into place 
 again. The table consists of twelve cast steel rollers, fourteen inches in 
 diameter, and six feet wide, set three feet two inches apart, making a table 
 thirty-seven feet long; the rollers are driven by a motor similar to the one 
 used on the front table. Rollers Nos. 2, 3, 4, 5, 6, 8, 10, and 12 have 19" x 
 4" collars on their ends for turning the piece, which should tumble off 
 them as it comes from the seventh pass. In addition there is a manipulator 
 in the first groove; this consists of five forged steel fingers two and one- 
 fourth inches wide mounted on rocker arms attached through a shaft to a 
 plunger in a cylinder pivoted to a support on the floor of the scale pit. The 
 upward motion of the table draws the fingers with it and, when the plunger 
 stops rising in the cylinder, causes them to turn the piece and advance it 
 for the second pass. This manipulator lies below the table when material 
 is delivered from the bottom roll and acts only to turn bars 90 from the 
 first pass to the second. The table is equipped with three heavy cast steel 
 
CONTINUOUS BILLET MILL 397 
 
 side guards between the four passes which the material uses in the bottom 
 roll. These reach back nine feet from the front to the table; there are also 
 light side-guards at each end of the rollers reaching the length of the table. 
 These tables make the operation of the mill practically automatic, and 
 make it possible to roll four pieces at the same time. 
 
 SECTION II. 
 
 THE CONTINUOUS BILLET MILL. 
 
 General Features of the Continuous Mill: The continuous mill, 
 often called a Morgan mill after the inventor, Chas. H. Morgan, consists 
 of a series of horizontal roll stands arranged one after the other, so that 
 the piece to be rolled enters the first stand and travels in a straight line 
 through the mill to the last stand where it issues as a finished bar, thus 
 making but one pass through each stand of rolls. In such a mill, where 
 the piece is being rolled in several different stands at the same time, it 
 is necessary that the surface speed of the different sets of rolls be so pro- 
 portioned that each set will travel at a speed as much greater than the 
 preceding one as the lengthening of the piece requires. With new rolls 
 and perfect adjustment to produce the proper reduction, this relation of 
 speed of the different stands is easily provided for by a system of driving 
 gears. To care for the wearing down of the rolls, the bottom roll is made 
 adjustable, and as a further precaution against little irregularities that 
 can't be overcome by adjustments, each set of rolls is purposely set to 
 run at a slightly greater speed than that required to conform to the speed 
 of the preceding set, so as to put the piece under tension at all times. For 
 turning the piece between passes twisting guides are employed. 
 
 Advantages and Disadvantages of Continuous Mills: High out-put 
 and low labor costs are the two chief advantages of this type of mill. In 
 addition, the mills roll the metal down very rapidly, thus giving less time 
 for oxidation and permitting more working in one heat, and yet the speed 
 of the roll is low, so that comparatively little power is required to run them. 
 Besides, the scrap losses are low, due to the fact that they can roll from 
 blooms of any length, which fact makes it unnecessary to cut the bloom 
 after leaving the bloomer, except to discard for pipe or other flaws that 
 occasionally occur. Finally, the rolls are so short as to be almost un- 
 breakable, and, therefore, very light rolls may be used for comparatively 
 heavy work with entire safety. As to the disadvantages, the great number 
 of rolls not only makes the first cost of the mill very high but adds im- 
 mensely to the cost of rolls for different sections. For the same reason, 
 much time is required for roll changes. Hence, the mill is best adapted 
 to roll large amounts of one section continuously. It is obvious that com- 
 plicated sections or those requiring great accuracy cannot be rolled on 
 
THE ROLLING OF STEEL 
 
 such a mill. These characteristics of the continuous mill, however, make 
 it particularly well suited for rolling billets, strips, such as hoop and cotton 
 ties, and skelp. They are also employed as roughing rolls for the various 
 combination mills. 
 
 Example of Continuous Billet Mill : As an example of the continuous 
 billet mill the fourteen inch number one mill at Duquesne has been selected, 
 because it is fed by the forty inch blooming mill, previously described. By 
 this combination the ingot is rolled down to a bloom approximately 6" x 4" 
 in the forty inch mill from which it is delivered on roll tables, after the 
 proper discard at the shears, to the continuous mill, where, without 
 reheating, the bloom is reduced to billets ranging in size from three and 
 one-quarter inches to one and three-eighth inches square. The mill con- 
 sists of ton stands of rolls, and is set in line with the bloomer. The distance 
 from the blooming mill shears to the first stand of rolls is eighty-four feet 
 eight inches. 
 
 Drive: This mill is driven by gears from a line shaft from an Allis 
 Chalmers horizontal vertical compound condensing Corliss valve 
 steam engine, size 44" x 78" x 60", with an indicated horse power of 3500. 
 This engine is opposite the shears and is set so its driving shaft extends in 
 a direction parallel to the mill line. The engine is designed to run at a 
 speed of 75 r. p. m. at a steam pressure of 130 Ibs. The maximum torque 
 the engine is designed to give at the roll circumference is 450,000 inch 
 pounds. The exhaust of this engine is taken to a central condensing plant. 
 The line shaft is coupled to the crank-shaft of the engine as follows: A 
 cast steel hub is forced on and held by keys to the end of the crank-shaft. 
 A phosphor-bronze thrust collar is bolted in halves over this joint. The 
 outer end of the hub, three feet ten inches in diameter, is bolted to a short 
 steel hub having wobblers twenty inches in diameter on its other end. A 
 cast iron coupling six feet ten inches long fits over this wobbler and that 
 of a similar but longer hub at its outer end. This hub is two feet seven 
 inches long, of cast steel, and its large end, three feet ten inches in diameter 
 is bolted to a short hub, twenty-one inches wide, which is keyed onto the 
 seventeen inch end of the line shaft. The line shaft of forged steel, is made 
 in two pieces, nine to thirteen inches diameter, and is seventy-two feet six 
 and three-fourths inches long. At ten points on the line shaft, beginning 
 at the engine end of the shaft are mitre gears respectively 4' 6", 4' 0", 3' 4^", 
 3' UK", 5' 2Y 2 ", 5' 5", 7' 0", 5' 5%" and 7' 11" apart; these mesh with 
 mitre gears keyed on cross over shafts that lead to their respective roll 
 stands. These gears are supported by bearing stands along the line shaft. 
 The crossover shafts drive the mill pinions, and give to each set of rolls, 
 beginning after No. 1 stand, a higher speed than that of the one preceding, 
 in order to take care of the increased length of the bar. The mill ends of 
 the crossover shafts are carried in bearings supported on pedestals; the 
 ends of the crossover shafts have cast iron half couplings keyed to them, 
 and these are bolted to other half couplings which are connected to the 
 
CONTINUOUS BILLET MILL 399 
 
 leading spindles by coupling boxes twelve and one-half inches long and 
 twelve inches in diameter. All spindles, pinions and coupling boxes on the 
 mill are cast steel. The pods on the spindles extend along their entire 
 length. The spindles, top and bottom, are all of the same dimensions: 
 two feet long, nine inches neck diameter, and nine and one-half inches body 
 diameter. The leading spindles are cast hollow so that they will break 
 under excessive strain before any other part of the mill, and are con- 
 nected by plain coupling boxes, each twelve and one-half inches long and 
 twelve inches in diameter, to the top pinions of the roll stands. 
 
 Pinions and Housings: The pinions are of the staggered tooth type 
 with three pods. They are four feet eight inches in total length, fourteen 
 and three-fourths inches wide across the face of the teeth, nine inches in 
 diameter at the wobblers, nine and one-half inches in diameter at the necks, 
 and fourteen and one-fourth inches in pitch diameter of the teeth, of which 
 there are fourteen. The pinion housings are cast iron, in one piece, and 
 are bolted to the pinion shoes, which run the length of the mill; each pair 
 is bolted together at the top as caps are not necessary. The windows 
 are cast to shape to receive the pinion bearing boxes, which are bolted to 
 them. The pinion bearing boxes are solid cast steel boxes, round in shape 
 and with lugs on the outer ends for bolting to the housings. They are 
 babbitted one-half inch deep; their dimensions are thirteen inches wide in 
 the windows and eleven and three-fourths inches long. The pinions are 
 joined to the rolls by solid spindles and coupling boxes already described. 
 
 Rolls and Housings: The roll housings are charcoal cast iron about 
 four and one-half feet high, bolted to the cast mill shoes; the mill shoes run 
 at right angles to the rolls through the length of the ten stands. The 
 housings have charcoal cast iron caps, one fitting over each. The caps are 
 notched at the four corners to receive the bolts to hold them fast to the 
 housings, and in the center of each side is bored a five inch hole for the 
 phosphor-bronze housing nuts, in which turn the housing screws; the housing 
 screws are of tool steel, twenty-four inches long, with one and one-half 
 threads per inch. Each housing has a window three feet seven and 
 three-fourths inches deep and thirteen inches wide, with a beveled sill at 
 the bottom and a ledge, twenty-one and one-quarter inches above the sill, 
 for resting the carrier bearing for the top roll. Each pair of housings is 
 held in line at the top by means of cast iron separators. The necessary 
 holes for set-pins and stud-bolts are drilled into the housings. The liners 
 used are the ordinary steel plates of varying thicknesses. 
 
 Adjustment of the Rolls: The method of adjusting rolls in mills of 
 this type is nearly always that of lining the bottom roll up or down to the 
 top roll. A cast steel screw box, therefore, is placed on the sill of each 
 housing and bolted to the housing; it is 16%" x 10" and is threaded with 
 nine half inch grooves, babbitted to prevent excessive wear. In each of 
 
400 THE ROLLING OF STEEL 
 
 these is placed a special screw bolt, left-hand thread for the outside housing 
 and right-hand thread for the inside housing. These bolts have short 
 square ends on the outer ends but longer squares on the inner ends; the 
 latter may be coupled together by a cast iron coupling. Above the screw 
 bolts and resting on them and the screw boxes are placed cast steel wedges 
 fourteen inches long and eight inches wide, with nine half inch grooves, 
 babbitted. The threads are, of course, the same as for the screw boxes. 
 On the wedges are rested the bottom bearings; these are steel castings. 
 On the bottom they are provided with a wedge which fits against the screw 
 wedge. The bearings have one inch of babbitt metal lining and two bronze 
 bearing pieces. No top bearing is necessary for the bottom roll, as there 
 is no upward pressure on it and it has no other piece to support, as the 
 carrier bearing for the top roll rests on the ledge mentioned in the preceding 
 paragraph. The breaker blocks are cast iron, the bottom ends of the set 
 screws resting directly on them. These screws are squared off above the 
 threads for adjustment by wrenches, and are provided with lock nuts. 
 
 Arrangement of Roll Stands and Guides: Owing to the fact that 
 the speed of travel of the bar and hence the speed of the rolls is greater 
 in each successive pass, the housings are placed closer and closer together 
 as the bar is reduced to avoid danger of buckling. In order from No. 1 
 stand, the center lines of the rolls are at the following intervals: 10' 0", 
 9' 0", 8' 0", V 0", 6' 6", 6' 0", 5' 6", 5' 6", 5' 6". For the purpose of obtain- 
 ing work on all sides of the bar and as the most convenient method of rolling, 
 the bar is twisted between every other pair of rolls, and for this reason 
 special guides have to be used. They are of cast steel made up especially 
 for these stands, so that they will give the bar just the proper twist or keep 
 it headed right to enter the next pair of the rolls. These guides are 
 set usually in cast steel guide boxes bolted to rest bars that are fastened 
 in ledges in the housings; where necessary, saddle bars are used to hold 
 down the guides and guide boxes. All guides are wedged tightly in place 
 with either steel or wooden wedges. The following is the arrangement of 
 the guides on this mill: No. 1 receiving guide is a combination straight 
 guide and crop shear bumper; all the rest of the receiving guides are straight; 
 but the delivery guides are alternated thus: No. 1, straight; No. 2, twisting; 
 No. 3, twisting; No. 4, straight; No. 5, twisting; No. 6, straight; No. 7, 
 twisting; No. 8, straight; No. 9, twisting; No. 10, straight. Where the 
 stands are far apart or the section is light, the bar is supported from below 
 by narrow plates reaching from one delivery guide to the next receiving 
 guide or else by light steel side guards. 
 
 The Rolls: The rolls for this mill are of the following dimensions: 
 Total length, four feet six inches; length of barrel, sixteen inches; diameter 
 of wobbler (3-pod), nine inches; diameter neck, ten inches; weight, 1500 to 
 1600 pounds. 
 
CONTINUOUS BILLET MILL 401 
 
 Table 54. Data Pertaining to Rolls for a 14" Continuous Billet Mill. 
 
 STAND COMPOSITION No. GROOVES BODY DIAMETER BEFORE TURNING 
 
 Top Bottom 
 
 1 Steel 1 13%" 13%" 
 
 2 1 13%" 13%" 
 
 3 " 2 13%" 
 
 4 Adamite 2 13^" 
 
 5 " 2 141^" 
 
 6 2 14M" 13%" 
 
 7 Chill Iron 4 13%" 13%" 
 
 8 " 4 13M" 13%" 
 
 9 4 14%" 14" 
 10 4 14%" 14%" 
 
 Six inches is allowed between the centers of the grooves when only two 
 grooves are cut, but only three and three-fourths inches is allowed in the 
 case of four-groove rolls. Following is a table of the speed of the rolls of 
 each successive stand and the observed delivery speed of the bar coming 
 out of it with the engine at normal speed of 75 r. p. m. 
 
 Table 55. Speed Ratios on Fourteen Inch Continuous Billet Mill. 
 
 STAND SPEED OF ROLLS DELIVERY SPEED OF BAR 
 
 REVOLUTIONS PER MINUTE FEET PER MINUTE 
 
 1 17 46.7 
 
 2 21.4 54.5 
 
 3 24.55 67.1 
 
 4 30.9 . 99.2 
 
 5 36.6 126.3 
 
 6 44.6 156. 
 
 7 57.6 188. 
 
 8 68.7 242. 
 
 9 89.1 326. 
 10 117.9 417. 
 
 Cropping Shears: Between the receiving table to the mill and the 
 first stand of rolls are hydraulic shears, pressure 450 Ibs. per square inch; 
 through these shears all blooms for the fourteen inch No. 1 mill pass. They 
 are capable of cutting blooms up to 7"xll" in size, but their usual work is 
 on 6" x 4" blooms. They are used to cut crops from the front end of the 
 bloom so it will enter No. 1 stand easily. When necessary they may be 
 used for shearing off pipes and bad pieces that have escaped discard at the 
 
402 
 
 THE ROLLING OF STEEL 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 N 
 
 o 1 Stand 
 
 
 
 
 H 
 
 
 
 SU*1 
 
 
 Steel 
 
 FIQ. 67. Rolls and Passes for 
 
CONTINUOUS BILLET MILL 
 
 403 
 
 the Continuous Billet Mill. 
 
404 THE ROLLING OF STEEL 
 
 blooming mill shears, and for severing the bloom in case of a cobble. They 
 are capable of a ten inch swing from the base, and at the end of their stroke 
 they strike the combined guide and bumper previously mentioned as the 
 No. 1 guide. They are thrown back into position automatically, when the 
 bloom is cut, by a heavy coil spring. The stroke of the knife blade is ten 
 inches. The shears are vertical acting with the top blade actuated by the 
 cylinder. No. 10 stand delivers the finished billet directly onto the receiv- 
 ing table for the steam flying shears; this table and its delivery table are 
 driven through bevel gears on a single line shaft; the line shaft is driven 
 by a jack shaft geared to a primary jack shaft which is in turn geared to 
 the crossover shaft for No. 10 stand. As various numbers of stands are 
 used for the various sizes of billets, the corresponding sizes of bars have 
 a different delivery speed and the shears receiving table must, therefore, 
 be driven proportionately, so that the billet maybe cut by the flying shears 
 without buckling and may be carried away, when cut, fast enough to keep 
 clear of the next billet. Accordingly, various sizes of gear wheels are 
 provided for the jack shaft to the table. The surface speed of the roll 
 table may be set at various rates to suit the delivery speed of the billet 
 by changing the gears on the jack shaft. 
 
 Flying Shears: The flying shears roll-table consists of eight cast steel 
 rollers, two, sixteen inches in diameter, dn the receiving side of the shears 
 and six, ten inches in diameter, on the delivery side. The rollers are notched 
 with a V-shaped groove so as to hold the bar, as it comes to the shears, 
 with one of its diagonals in the vertical, as it is in this position when it 
 leaves the last pass of the rolls and must be sheared in the same position. 
 The flying shears are placed with the center line of their knives twenty feet 
 beyond the center line of No. 10 stand. The shears are actuated by a 
 30" x 20" steam cylinder. The action of the shears is speeded up or slowed 
 down according to the delivery speed of the billet. Cutting under fifteen 
 foot lengths is not attempted for fear of not getting the shears back to 
 position in time to prevent buckling of the next billet. The knives on the 
 shears have a life of from three to sixty hours; and they have to be changed 
 for every size of billet. They have a half inch clearance above the square 
 being cut. The horizontal stroke of the shears is ten inches. 
 
 Hot Beds : The flying shears deliver upon a table 125 feet long, from 
 which steam operated rollers and pushers convey the bars to four hot beds 
 extending at right angles to the tables. All of these are controlled from 
 a pulpit in the yard. The rollers at the foot of No. 1 bed are skewed so as 
 to bring the billets against the first pusher and make them lie parallel 
 with it; all the other rollers are, as usual, set at right angles to the pieces. ' 
 At four points on the table are hydraulically operated stoppers for stopping 
 the bars at the hot bed desired or allowing them to pass to the horizontal 
 scrap bed beyond the last hot bed. The hot beds are sloped up at a slight 
 angle and are each thirty-one feet wide by fifty-three feet six inches 
 
CONTINUOUS BILLET MILL 
 
 405 
 
 W^^r- .^^JS 
 
 Fia. 68. 14" Continuous Mill 4" x 6" Blooms to 2" Billets. 
 Drawings from Actual Sections. 
 
406 THE ROLLING OF STEEL 
 
 long. They are built of rails, and the material is moved on each by a steam 
 pusher connected to a cable, driven through gears by two 8" x 10" vertical 
 twin simple 50 h. p. steam engines. Cold pushers are also cable connected 
 by gears and driven by similar engines, but of the horizontal type. They 
 convey the billets desired to the end of the bed and slide them over rail 
 ends from the beds into railroad cars just below the hot bed level. Alligator 
 scrap shears are provided at the end of the scrap bed, which is hand operated. 
 The accompanying prints are intended to show the forms of the 
 rolls, their kinds, shape of the various passes, and the different stages in 
 the reduction of the bloom to the billet. 
 
 SECTION III. 
 
 ROLLING OF SHEET BARS AND SKELP 
 
 Difficulties and Methods of Rolling Semi=Finished Flats: This 
 material may or may not be rolled from the original heat of the ingot. At 
 Duquesne, sheet bar, as well as billets and splice bars, is rolled on the 
 twenty-one inch mill from the original heat of the ingot, which, being first 
 reduced to a &W x 1Y" bloom on the thirty-eight inch mill, is passed to 
 the twenty-eight inch billet mill and on to the twenty-one inch mill without 
 reheating. At Edgar Thomson, for example, the 9}^" x 9J/6" bloom from 
 the three-high blooming mill is reheated, when the rolling is completed on 
 the No. 4 mill, which consists of a single train of three stands of three-high 
 rolls, or on one of the rail mills, usually the number one. However, the 
 method employed in reducing the material is the same, except as to details 
 of handling, which, of course, must be changed to suit the different mills. 
 Because the twenty-one inch mill at Duquesne represents a distinct type 
 of mill, this mill is selected as an example of a mill rolling sheet bar. The 
 problem to be overcome in rolling these flats lies in the difficulty of con- 
 trolling the width. In rolling blooms, billets and small slabs, the piece 
 is held to dimensions, not only by the shape of the grooves, but also by 
 edging the piece in certain of the passes. But in rolling sheet bar, the 
 thinness of the piece will not permit edging, after it leaves the roughers. 
 
 The Tongue and Groove Pass: For the purpose of controlling the 
 width and at the same time effecting a heavy reduction in the sectional 
 area, a form of closed box pass, called the tongue and groove pass, is used. 
 In this form of pass a groove, corresponding in width to the width of the 
 piece desired, is cut in one of the rolls which encloses one side and the 
 edges of the piece in rolling, while a tongue, cut on the opposite roll, fits 
 into the groove, thus closing the pass on the fourth side. The designing 
 of this pass presents some very interesting features. In order to insure a 
 proper delivery of the pieces from the rolls and provide for fitting the tongue 
 into the groove, the sides of the latter are cut at a slight angle to the bottom. 
 Owing to the heavy drafts taken, the metal is squeezed up into the clearance 
 between the tongue and the edges of the groove, thus forming a fin on each 
 
SHEET BAR 
 
 407 
 
 111 
 
 
 
 1 
 
 
 4"v4" Rillpt Go 
 
 fr 
 
 \ 
 
 _r^i 
 
 Sheet Bar 
 
 
 X V r-t 
 
 Bar 
 
 / 
 
 
 
 k_Z_J 
 
 
 
 K ? -> 
 
 
 No. 1 Stand-4' x4" Billet, Common Splice and Sheet Bar 
 
 D i 
 
 >1 J Dummy Pass 
 
 u 
 
 
 
 
 ^ 
 
 
 
 V-^,-V 
 
 \ 4"x4" to 14" Mill 
 
 L i 
 
 
 No. 2 Stand for Sheet Bar 
 
 No 3 Stand for Rolling Sheet Bar 
 
 >Vo. 4 Stand for Rolling Sheet Bar 
 
 No 5 Stand 
 
 Noa. 5 and 6 Stands for Rolling Sheet Bar 
 
 Fio. 69. Roll Stands for Boiling eight inch Sheet Bar. 
 
408 THE ROLLING OF STEEL 
 
 side of the piece, unless precautions are taken to prevent it. These fins 
 are prevented from forming by cutting the groove with fillets at the edges, 
 and arranging them so that the bevelled edges of the piece formed by the 
 fillets enter the succeeding pass opposite the openings formed by the clear- 
 ance between the rolls. In this way no fin is formed, because the spreading 
 of the material merely fills out the bevel of the fillet, leaving no excess 
 metal to be squeezed up between the rolls. To enter the first tongue and 
 groove roll the edges of the billet are well rounded off, which prevents 
 more than a very slight fin forming in this pass. Since the piece is to be 
 finished in plain rolls, no fillet is placed in the last tongue and groove pass. 
 The accompanying prints show the forms of these passes, and the different 
 steps in the reduction from the billet to sheet bar. 
 
 PM ** . -.*?" . 
 
 FIG. 70. Rolling Tongue and Groove for 8 inch Sheet Bar. 
 
 Sheet Bar is all approximately eight inches wide and varies in thickness 
 to give weights, per linear foot, from seven to forty-three pounds. The gauge 
 in inches is found by multiplying the weight per foot by .0372 in which factor 
 the weight of a cubic inch of steel is taken to be .28 pounds. After the mill is 
 once set for rolling sheet bar, the different weights of bar are obtained by 
 varying the distance between the rolls. As there is considerable difference 
 in temperature in different billets when rolled, as at this mill, from the 
 original heat of the ingot, it is difficult to hold the thickness constant at 
 the finishing stand, and, in order to keep the thickness uniform, a man is 
 stationed at this stand of rolls to adjust the screws up or down to suit the 
 temperature of the bar. As it is necessary to produce a very smooth surface 
 on sheet bar, on account of its being subsequently rolled into thin sheets, 
 chilled rolls are used in the finishing stands. For the same reason, water 
 and steam jets must be directed against both surfaces of the bar in order 
 to remove the scale. These jets are used both at the rolls and at the saws. 
 
 Example of a Mill Rolling Sheet Bar The Twenty=one Inch Mill 
 at Duquesne: As previously stated, this mill represents a distinct type. 
 The design aims to secure the advantages of the continuous mill and elimi- 
 nate the disadvantages. So, while it is practically continuous in action the 
 different stands of rolls are placed so far apart that the piece clears one 
 stand before it enters the next. As a tandem arrangement alone would 
 
SHEET BAR 409 
 
 spread the mill out over a too great length, the various stands of rolls are 
 usually arranged in trains that are in tandem. Such a layout requires long 
 roll tables for carrying the piece forward and suitable apparatuses for 
 transferring the piece transversely, such as lifting cradles, skids, diagonal 
 roll tables, and switch, or divided, tables. In this respect the twenty-one 
 inch mill at Duquesne is a good example. 
 
 The Layout for This Mill, as for all mills of this type, is somewhat 
 complicated. The mill consists of six stands of rolls arranged in two trains, 
 separately driven and of three stands each. The two trains are separated 
 by a distance of 119 feet. In each train the first and second stands next 
 to the engine are three-high, while the third, on the end of the train, is 
 two-high. For convenience the different stands are numbered in the order 
 in which the material passes through them. Observing this order, then, 
 stands Nos. 1, 4 and 5 compose the first train, while stands Nos. 2, 3 and 6 
 make up the second train. The first stand is located 105 feet beyond the 
 twenty-eight inch mill and is provided with a receiving table fifty-three 
 feet six inches long, equipped with switches for guiding the material from 
 the twenty-eight inch mill into the different passes in the first stand of 
 the twenty-one inch mill. These passes are three in number, all of different 
 sizes, one of which is employed as a finishing pass for billets and the other 
 two as working passes on material to be finished on the twenty-one inch 
 mill. 
 
 Arrangement of the Roll Tables: The delivery table for No. 1 stand 
 is provided with a center guard for diverting material to the billet table 
 that leads to the 4" x 4" billet shears, located beyond No. 2 stand. Further 
 along, by means of a switching device another division of the material 
 may be made, thus sending billets either through a dummy pass in N6. 2 
 stand to the fourteen inch No. 2 continuous mill, located about 100 feet 
 beyond, or to a working pass, when the material is to be finished at the 
 twenty-one inch mill. Since material must be cut into suitable lengths for 
 rolling on the twenty-one inch mill, a hydraulic shear is located 77 feet 
 from No. 1 stand and arranged to cut on the twenty-one inch mill-half of 
 the table only. The receiving table for the No. 2 stand begins at these 
 shears. It is provided with a stop which may be set for lengths from 
 twelve and one-half to thirty-eight feet. A manipulator for turning the 
 piece is also provided in this table. The delivery table for No. 2 stand is 
 65 feet long and has ten rollers fifteen inches in diameter and twenty inches 
 long; these rollers are separated by a side-guard from the rollers leading 
 to the fourteen inch mill No. 2. In connection with this table is a transfer 
 skid table for moving the piece to the receiving table for No. 3 stand. It 
 consists essentially of a frame of rails bolted together and hinged to the 
 table, onto which they are to deliver the steel. The frame, when in its 
 lowest position, lies below the roller tables, so that when the transfer is 
 raised, it picks up the steel, which slides down the rail skids onto the table. 
 
410 THE ROLLING OF STEEL 
 
 The skids are greased so that the steel will slide more easily. The frame 
 is raised and lowered by means of links keyed to a line shaft which is in 
 turn operated by a hydraulic cylinder. The transfer raises the bars twenty 
 inches from the top of the delivery table No. 2 to the top of the rollers 
 of the receiving table of No. 3 stand, the distance between the two tables 
 being eight feet three inches. As No. 3 stand is often used as a three-high 
 stand this table is of the tilting type, and operated by an hydraulic cylinder 
 placed near the stand. The delivery table of No. 3 stand is 109 feet long 
 and serves also as a receiving table for No. 4 stand. It is stationary and, in 
 order to receive the material when No. 3 stand is operated two-high as 
 well as three-high, it is inclined, extending from the top of the bottom 
 roll of No. 3 stand to the top of the middle roll in No. 4 stand. Collars 
 on its rolls serve to turn the piece between the stands, and its side guards 
 are adjustable so that they may be used to guide the piece into different 
 passes on No. 4 stand. The delivery table for No. 4 stand is 79 feet long, 
 and has guards on the side next to the engine only, in order that the piece 
 may be transferred by means of a skid table to the receiving table for 
 No. 5 stand. This transfer table is similar to that between No. 2 and No. 3 
 stands except that the piece here slides to a lower level, where it is stopped 
 by the guards on the receiving table for No. 5 stand. Connecting No. 5 
 and No. 6 stands is a stationary table provided with adjustable side guards 
 and vertical rollers for edging the piece as required. 
 
 Hot Saws and Shears: The delivery table of the last, or No. 6, stand 
 is about 104 feet long, and at its farther end are located two electrically 
 driven hot saws set thirty feet six inches apart. These saws, made of .80% 
 carbon steel, are forty-two inches in diameter and one-fourth inch thick. 
 This table feeds into a shear table, one hundred three feet nine inches long, 
 along which are situated seven electrically operated shears. These shears 
 are adjusted to cut at any lengths up to ninety-seven feet six inches, which 
 is the maximum distance between the first and last shears. They may be 
 made to cut in unison or separately, as desired. From the shear table the 
 piece may take a straight course to the sheet bar shears and bundling cradle, 
 or be diverted to the hot beds which are used for billets and splice bars. 
 Returning now to the billets from No. 1 stand, it was mentioned that 4" x 4" 
 billets could be diverted to a shear. This shear, of the duplex type, is 
 located several feet beyond No. 2 stand and is provided with a bisected 
 table of which each part leads to one of the two blades of the shears. A 
 gauge and automatic stopper, mounted on a gauge beam about two feet 
 above the delivery tables for the shears, can be set at one-quarter inch 
 intervals for any lengths from twenty-three and one-half inches to twenty 
 feet. From these shears, an elevated inclined roller conveyor carries the 
 short billets to eight bins, each of which has a capacity of 30 tons and is 
 located so as to empty directly into railroad cars by gravity. 
 
 Drive: Each train is direct driven by a William Tod Co. 34" x 58" 
 x 60" tandem compound horizontal condensing engine of 3500 h. p. The 
 
SHEET BAR MILL 411 
 
 usual speed of these engines is 67 r. p. m. but they can be run .as high as 
 82 r. p. m. Both trains are direct driven through a pinion shaft from the 
 crank-shaft of the engine. The connections between engine and mill and 
 between the stands of rolls are made in a manner exactly similar for both 
 trains, so that one description will suffice for both. To the end of the 
 engine crank shaft is keyed a half coupling of cast steel, and over it is bolted 
 a thrust collar. A second half coupling is bolted to the one next the engine. 
 A compound coupling box two feet two and one-half inches in diameter on 
 the engine end and nineteen and one-half inches on the mill end and nineteen 
 and one-half inches long for No. 1 train of rolls, and a similar box but two 
 feet two and one-half inches in diameter on the engine end and twenty-three 
 and one-half inches in diameter on the mill end and nineteen and one-half 
 inches long for No. 2 train of rolls fits over the wobbler of the second half- 
 coupling and the mill end wobbler of the leading spindle. The leading 
 spindle for No. 1 train is three feet four and three-fourths inches long, and 
 twelve inches in diameter, and has three pods on the engine and four on the 
 mill end. For No. 2 train the leading spindle is four feet one and one-half 
 inches long, sixteen inches in diameter on the engine end, where it has 
 three pods, and twelve inches in diameter on the mill end, which has four 
 pods; this spindle has twelve and one-half inches near the center turned 
 smooth to a fifteen inch diameter for a rider bearing. A coupling box 
 of cast steel, fifteen inches long and seventeen and one-half inches in diameter, 
 connects the leading spindle to the center pinion of the three in the pinion 
 housings. 
 
 Pinions and Housings: These pinions are plain steel castings, six 
 feet three inches long, with a pitch diameter of twenty-one inches, a neck 
 diameter of thirteen inches, a wobbler diameter of twelve inches, and a 
 width of twenty-four inches across the face of the teeth. There are eleven 
 teeth, of six inches pitch, cut in the helical manner. There are three pinions, 
 set one above the other in their proper bearings in cast steel pinion housings. 
 These housings stand about five feet eight inches above the mill floor and 
 are bolted to the mill shoes; their windows are twenty-two and one-half 
 inches wide and five feet eight inches deep, beveled at the bottom, and 
 are provided with forged steel side liners, five feet four and seven-sixteenths 
 inches long and two and three-fourths inches thick, and cast steel bottom 
 liners, 35" x 12" x 2". The housings are drilled for the necessary holes for 
 set pins, stud bolts, cap bolts, etc. One steel cast pinion housing cap covers 
 the housings; at its four corners are drilled four inch holes for the cap bolts 
 which are three feet five inches long, and are held in by a nut at the bottom 
 and a key at the top. In the middle are fastened two hooked lifting bolts, 
 for enabling the crane to get hold when the caps are to be removed. The 
 bearings are all of the solid type, of cast steel, and babbitted. Coupling 
 boxes, similar to the one connecting the leading spindle to the middle pinion, 
 connect the three pinions to their respective spindles. The spindles are 
 steel castings with wobblers of four pods each, extending from end to end; 
 
412 THE ROLLING OF STEEL 
 
 they are three feet six inches long and twelve inches in diameter. They 
 are supported on the mill end by another set of similar coupling boxes 
 which connect them to the rolls. 
 
 Rolls and Roll Housings : The rolls on this mill are twenty-six inches 
 long on the body for all stands except No. 3 which is thirty-six inches, 
 because, being a three-high stand, it contains a greater number of passes 
 than the two-high stands. The collars are usually twenty-two and one- 
 half inches in diameter; the body diameters range from fifteen to twenty- 
 seven inches; the diameter of the necks is thirteen inches, and of the wobblers 
 twelve inches. The total length is six feet ten inches for the rolls on No. 3 
 stand and six feet for the rest. The rolls range in weight from 3800 pounds 
 to 7000 pounds. The rolls are cut down each time they are dressed one- 
 eighth to three-eighths of an inch until a collar diameter of nineteen inches 
 is reached, when they are scrapped. The rolls are of the following materials 
 for the various products rolled: 
 
 No. 1 Stand Usually sand roll; rarely steel. 
 
 No. 2 Stand Sand Roll for billets and common splice bars; steel in 
 majority of cases for Duquesne and continuous rail 
 joint; always steel for sheet bar. 
 
 No. 3 Stand Sand Roll for billets and common splice bar; sometimes 
 the top roll for common splice bar is steel. Steel always 
 for sheet bar and nearly always for Duquesne and con- 
 tinuous rail joint; otherwise cast iron. 
 
 No. 4 Stand Sand Rolls always for everything. 
 
 Nos. 5 and 6 Stands Always sand rolls except for sheet bar; sheet 
 bar requires chilled iron rolls. 
 
 About three sets of billet rolls, about three sets of cast iron rolls for 
 sheet bar and five sets of chilled rolls, two sets of common splice bar rolls for 
 No. 6 stand and one set of other rolls are carried on hand; For rail joints rolls 
 are turned up as needed, and all splice bar rolls are ordered new when speci- 
 fications for a new section come in. The roll housings for the twenty-one 
 inch mill are similar for stands Nos. 1, 2, 3, and 4, which can be used as 
 three-high, and for Nos. 5 and 6 which are only two-high. In No. 1 train, 
 the rolls in No. 1 stand are connected by spindle and coupling boxes to those 
 in No. 4 stand and the bottom two of No. 4 are connected to No. 5 rolls. 
 No. 2 train is similarly connected; the top roll of Nos. 1 and 2 stands is a 
 dummy acting as a spindle and the bottom rolls in Nos. 1 and 4 stands 
 act the same way. The spindles, except those between No. 3 and No. 4 
 stands, are two feet nine inches long and twelve inches in diameter; the 
 latter are three feet six inches long and twelve inches in diameter. The 
 coupling boxes are all fifteen inches long and seventeen and one-half inches 
 in diameter. The roll housings are cast iron, held in line, top and bottom, 
 by cast iron separators. For Nos. 1, 2, 3, and 4 stands the housings stand 
 
DEFECTS IN BLOOMS AND BILLETS 413 
 
 five feet eight and one-half inches above the mill shoes and have twenty- 
 two and one-half inch windows; those for Nos. 5 and 6 stands rise three 
 feet eleven and three-fourths inches above the shoes and have twenty-two 
 and one-half inch windows. From each housing there are two cast iron 
 caps, held down by square key bolts, fitting through five inch square holes 
 in the caps. In the center of each cap is cast a hole for receiving a phosphor 
 bronze housing-nut, which is pressed into the cap and threaded for receiving 
 the housing screw. This screw, which is made of open hearth steel 'of 
 .36% to .40% carbon, is five and three-eighths inches in diameter at the base 
 and is threaded at a one inch pitch. On all stands but No. 6, a cast steel 
 rosette is fastened on top of the housing screws to provide means for turning 
 them. They thus hold the top rolls tightly down. On No. 6 stand a cast 
 steel disc is used instead of a rosette, and a long lever is attached 
 to it, which in turn is held in place by means of bolts through slots in the 
 outer edge of the disc. The bottom bearings for Nos. 1, 2, and 3 stands, 
 as well as all the riders and the carrier bearings are of cast steel. Brass 
 bearing pieces are used in some, but not in all, of the housings. The pre- 
 ceding prints show how the mill may be used for rolling billets as well as 
 sheet bar. The rolling of rail joints, to be described later, is gradually 
 being discontinued on this mill, the intention being to transfer this business 
 to Edgar Thomson Works. 
 
 SECTION IV. 
 
 SOME GENERAL PRECAUTIONS TO BE OBSERVED IN ROLLING 
 SEMI-FINISHED PRODUCTS. 
 
 Reasons for Studying Defects: A great many of the precautions 
 necessary to observe in rolling the semi-finished products have been 
 mentioned at various times in preceding pages. However, as failure to 
 observe the proper precautions in rolling gives rise to defects in the material 
 which may show up in the finished article and as the reader may be par- 
 ticularly interested in this phase of the business, it is thought that a list 
 of rolling defects and their causes may be found useful and interesting. 
 By giving the cause for each, it will be shown that many defects are unavoid- 
 able, and that even the most rigid inspection will not suffice to eliminate 
 some defects which are a common annoyance to the manufacturer and 
 consumer alike. 
 
 Rough Surface Due to Scale: One of the defects common to semi- 
 finished material and one that often shows up in the finished article is a 
 very rough or pitted surface. That this roughness most often is due to 
 the adherence of scale on the surface of the ingot during the rolling there 
 can be little doubt, because a careful examination of such defects will 
 generally reveal its presence in these pits. At first thought this defect is 
 likely to be attributed to a rolling of the scale into the surface, and the 
 remedy at once suggested is to clean the ingot of scale during rolling. But 
 
414 THE ROLLING OF STEEL 
 
 failure to remove the scale from the ingot will not always account for this 
 roughness. In such cases the blame for the defect is to be laid to the 
 presence of blow holes near the surface, which in the heating of the ingot 
 in the soaking pit become filled with molten oxides. The presence of the 
 oxide may be attributed to two causes, namely, to the oxidation of the 
 surface of the blow holes or to its introduction through small openings 
 which lead from the blow holes to the surface of the ingot. Thus, if the 
 ingot is subject to a temperature sufficiently high to fuse the oxides, the 
 oxide in the hole will melt, or the liquid oxide on the surface will flow through 
 these openings to the blow holes beneath and partially or completely fill 
 these small cavities. This oxide cannot be removed, and when the ingot 
 is rolled, it becomes so firmly embedded in the surface that even subsequent 
 pickling will not remove it. The only correction remaining for such defects, 
 then, is the very expensive one of chipping or grinding. Low carbon 
 chrome-nickel steel is very susceptible to this fault, and it is very difficult to 
 clean the scale from its surface. While this peculiarity of nickel steel may be 
 attributed to the same cause as that just cited for plain steels, there is 
 much evidence to show that scale pitting in this case is partly due to an 
 entirely different cause, namely, the reduction of the oxide of nickel by 
 metallic iron at the rolling temperature of this steel. Thus, as fast as this 
 alloying element is oxidized on the surface, its oxide is reduced by the free 
 iron beneath, the result being the formation of iron oxide under the surface 
 of the metal. This condition gives rise to an outer layer composed of 
 metallic alloy mingled with oxide, in which the oxide acts as a binder 
 between metal and surface scale. It can readily be seen that this layer 
 may vary in thickness, and the merging from all metal to all oxide is gradual, 
 resulting in what may be termed an interpenetration of metal and oxides, 
 which causes the scale to adhere most firmly to the surface. 
 
 Cobbling: The most frequent failure in rolling is cobbling. It occurs 
 at the bloomers as a turn down or twisting of the piece in the rolls, at the 
 rougher in the same way or by catching and buckling on the roll table, 
 and at the other mills as a roll table or mill accident. The piece may 
 catch on a table and buckle up and be prevented from coming through the 
 rolls; it may catch against a guide and buckle; or it may buckle against 
 the rolls, if delivered to them too fast. In such cases, practically all of the 
 piece has to be scrapped; the uninjured sections of partially cobbled blooms 
 can usually be finished and be made use of. 
 
 Laps: An over filling of a pass causes the steel to spread between the 
 collars of the* rolls and causes a lap; this is usually rolled down into the 
 surface, partially or altogether, if the steel is turned for the next pass, 
 and the place between the lap and the rest of the piece is left as a surface 
 crack, or seam. A lap may result from a crack in the rolls into which the 
 steel flows. 
 
 Collar Marks : Owing to overdraft or possibly defective heating, or, 
 in the blooming mill, to too infrequent turning of the piece, the steel will 
 overfill the groove, causing collar marks. Lack of alignment or proper 
 
DEFECTS IN BLOOMS AND BILLETS 415 
 
 adjustment of the rolls or any other incident that gives an unequal draught 
 will cause the collars to bite into the bloom and injure it. Collaring leaves 
 deep cuts that can seldom be rolled out, hence the injured portion of the 
 bloom is discarded at the shears. 
 
 Guide Marks: Guides, if they are too deep, out of line, or required 
 to do too heavy duty will score the surface of the steel, usually, in fine 
 lines, or else may tear its edges. 
 
 Ragging Marks: Ragging leaves protrusions on the surface of the 
 steel, and sometimes these are lapped over, showing in the finished steel 
 in irregular seams and cracks. On mills rolling steel that is to be rolled to 
 small section or a fine finish, only light ragging is used. 
 
 Off Size : If a bloom, slab, or billet is off size, it makes the weight 
 of the predetermined cut different from the actual cut and will prevent an 
 order from being accurately filled, or may even cause a deficit of steel to 
 apply on the order. On mills of the class in question it is difficult to work 
 to absolutely correct sizes, hence, a tolerance for weight as well as for 
 size should be given. 
 
 Unequal Draughts: These defects are due to the rolls being out of 
 parallel with each other, causing one side to be rolled light and the other 
 heavy. This condition may result also in a turning down of a lap in the 
 next pass in the case of a bloom, or a cobble in the case of a billet. 
 
 Seams: Seams may be caused by blow holes, by laps, or by tearing 
 of the steel due to causes which will be explained in a succeeding 
 paragraph. Slivers or scale, first rolled into the surface, and then torn 
 out, may leave cracks that will roll down to form seams. Seams may also 
 be caused by too much belly in the roll, or by not turning the piece often 
 enough during the rolling of the bloom, as illustrated in Fig. 71. Seams 
 are especially injurious in steels for forgings and for heat treatment. They 
 seldom fail to cause the steel to crack in quenching, particularly if the steel 
 is quenched in water. 
 
 Slivers : Slivers are due to defective teeming of the molten steel and 
 to a tearing of the corners of the steel in blooming, roughing, or finishing. 
 Tearing is attributed to many things, such as over oxidation in the open 
 hearth, burning, twisting in the rolls, and improperly adjusted guides. Soft 
 steels and high sulphur screw stock are especially subject to these defects. 
 
 Scabs : Scabs are found on steel if it is burned or if scale is rolled into it. 
 
 Shearing Defects: Failure to discard properly at the shears may 
 result in rejection of product. Aside from this neglect, other defects may 
 be produced by the shearing. Thus, a dull shear knife or one with too 
 great clearance will not make a clean cut and will leave a lip on the side 
 of the steel where its stroke ends. An unavoidable effect of shearing is 
 to produce what may be termed a mechanical or manufactured pipe. In 
 the shearing of billets and slabs, especially in the case of 4" x 4" billets and 
 larger, it frequently occurs that the sheared end shows a pulled-out con- 
 
416 
 
 THE ROLLING OF STEEL 
 
 FIG. 71. Good and Bad Practice in Rolling Blooms. 
 Top bloom shows effect of not turning the bloom often enough. 
 
DEFECTS IN BLOOMS AND BILLETS 417 
 
 dition. There are a number of things responsible for this condition on the 
 sheared ends. Thus, highly segregated steel will not shear uniformly and 
 often results in a pulled-out condition, and the same thing is almost sure to 
 happen in case the billet or bloom has a spongy center. The temperature 
 at which the billets are sheared plays a very important part, also. These 
 pulled-out cavities on the sheared end may have a depth of more than 
 one inch. It is readily seen what happens when such billets are re- 
 heated and rolled into small sizes. The pull-out is closed up and elongated 
 with the rolling, and when rolled into a small rod or any other smaller 
 shape the effect of this pull-out condition may extend far back into the 
 finished material. Very often this manufactured pipe is mistaken for a 
 genuine metallurgical pipe, since in most cases it is centrally located. The 
 injurious effect of a manufactured pipe on the physical properties of steel 
 is similar to that of a metallurgical pipe. 
 
 Splits or Cracks in Billets and Blooms: In breaking down ingots it 
 often happens that the metal does not yield properly to the draught, and 
 the surface structure is cracked or torn at a number of places, and some- 
 times to a depth of two or three inches. As the rolling is continued, these torn 
 surfaces are gradually closed, but not perfectly welded, and become much 
 elongated, so that it is not easy to detect them in the finished article, not 
 only because they are completely closed but because of the new scale, which, 
 forming after the rolling is finished, totally covers up all signs of their 
 presence. This feature makes their occurrence all the more serious, because, 
 though their dangerous character is recognized by the manufacturer, and 
 every attempt is made to eliminate them, the inability to detect them 
 often leads to their passing the inspection. These cracks are attributed 
 to many causes. In the first place, certain grades of steel, more particu- 
 larly those in which the carbon content lies between .18% and .22%, are 
 more susceptible to this defect than others. The sulphur and manganese 
 content also appears to affect the tendency of ingots to crack. Hence, 
 many steel men are inclined to lay most of the blame to chemical compo- 
 sition, while others hold that the fault lies in improper treatment in manu- 
 facture. It is a fact that steel not properly made may be red short, and 
 that the steel can be injured through too heavy draught and too much reduc- 
 tion without turning, or poorly designed passes in rolling cannot be denied. 
 It also appears that the heating of the ingot in the pits may exert an impor- 
 tant influence upon the rolling properties of the steel. 
 
 Inspection: Blooms, billets, and slabs are inspected on the mill 
 yard, and when defects are not very deep, they are chipped out with chipping 
 hammers, if so ordered, before the steel is shipped. All inspection is 
 directed to the elimination of the defects listed above, and rejection is 
 according to the strictness of the order in respect to this requirement. 
 Billets and sheet bars are hot bed inspected, and in addition to inspection 
 for surface defects sheet bar is also tested for exactness of weight. 
 
418 ROLLING FINISHED PRODUCTS 
 
 CHAPTER VII. 
 
 THE ROLLING OF THE HEAVIER FINISHED PRODUCTS- 
 PLATES. 
 
 SECTION I. 
 
 PREPARATION OF THE STEEL FOR ROLLING FINISHED PRODUCTS. 
 
 Reheating: While a few finished steel articles, such as plates, large 
 rails and heavy shapes, which on account of their large mass retain their 
 heat for a considerable length of time, may be rolled by rapid methods 
 directly from the ingot without reheating, the majority of articles are so 
 small that their temperature would fall far below the rolling range before 
 the great amount of reduction required could be accomplished. For all such 
 articles a reheating of the bloom, billet, or slab is a necessary step pre- 
 liminary to rolling. Needless to say, this reheating of the steel is a matter of 
 great importance and requires even more care than the heating of ingots. 
 Compared with hot ingots the nature of heating is very different, for here all 
 the heat is conducted toward the center from the surfaces exposed to it; 
 and since in practice it is well nigh impossible to expose all surfaces equally 
 to the heating medium, uneven heating is likely to occur, the result of which 
 is a variation in the dimensions of the finished section. Here, too, as with 
 ingots, the danger of burning or overheating is ever present. As the tem- 
 peratures attained are far above the critical range, the reheating tends to 
 undo the refining of the previous rolling. Since the extent of this obliter- 
 ation of the original structure is about in proportion to the temperature 
 above the critical attained it is desirable to keep the reheating temper- 
 ature as low as possible, and to finish the rolling as near the critical range 
 as practicable. However, some finished materials are so light that the 
 highest temperatures attainable without injury to the steel is barely 
 sufficient to complete *he rolling, and in addition the wear-and-tear on the 
 mills incident to rolling at the lower temperatures increases so rapidly as 
 to add very much to the expense of rolling. Again, the surface of the 
 metal is oxidized very rapidly in a flame or a hot atmosphere even slightly 
 oxidizing, and this oxidation results in the formation of an insulating coat of 
 scale that retards the heating. This sc ale may cause trouble in other ways, 
 also, because, at the temperature which it is often necessary to maintain in 
 the furnace in order to heat the steel to the proper temperature for rolling, 
 
REHEATING FOR ROLLING 419 
 
 it becomes pasty or quite fluid, and blooms or billets in contact in the furnace 
 are often cemented together by it. Besides, the surface of the piece, 
 being covered with this pasty scale, is liable to cause pieces of brick, sand, 
 or other foreign substances to adhere to it, and these, being rolled into 
 the steel, produce serious surface defects in the finished material, some of 
 which are known as scabs, brick spots, pitted surfaces, etc. The loss of 
 metal due to the formation of scale will sometimes amount to as much as 
 5% and is seldom under 2%. 
 
 Types of Reheating Furnaces: Reheating furnaces cannot as yet be 
 said to have reached a standard in design and construction. They are, 
 therefore, of various forms and types, and each furnace is constructed along 
 lines which were thought to be the best suited to the local conditions at 
 the time of its erection. Most of these furnaces are, however, of the 
 reverberatory type, and may be fired with either coal or gas for fuel, though 
 the gaseous fuels are much preferred for this purpose on account of the ease 
 with which the temperature may be regulated. In order to conserve the 
 heat as much as possible, they may be provided with waste heat boilers 
 or be constructed on the regenerative or recuperative principles. New 
 mills, however, the majority of which are being electrically driven* cannot 
 employ the waste heat boilers. The more modern furnaces are, then, 
 built either on the regenerative or the recuperative principle, in both of 
 which gaseous or liquid fuels or pulverized coal are used. 
 
 The Regenerative Reheating Furnace is similar in working principle 
 to an open hearth steel furnace. Gas and air ports at each end of the 
 furnace are connected by flues that lead to checker chambers, made of the 
 proper refractory materials for retaining the heat from waste gases and 
 giving up the same to ingoing air, or air and gas if producer gas is used, 
 the current being reversed at certain intervals of time. Unlike the open 
 hearth furnace, however, the checker work for these furnaces is usually 
 placed under the furnace, and, instead of the basin-like hearth, the reheating 
 furnace is provided with a floor practically on a level with the door sills. 
 A low bridge wall, which separates this floor from the up-and-down takes, 
 forms a kind of combustion chamber and prevents the flame from impinging 
 directly upon the steel. Above this combustion chamber the roof of the 
 furnace slopes downward toward the middle of the furnace and reverberates 
 the heat of the flame upon the floor. The steel is charged through lifting 
 doors in one side of the furnace, and it may be drawn either through these 
 same doors or through doors in the opposite side. Originally, it was the 
 general practice to line the bottom of these furnaces with refractory siliceous 
 sand, hence they are often called sand bottom furnaces. As the scale in 
 melting unites with this sand to form a slag of a too high per cent, of silica 
 to be used economically, these furnaces are now made up of magnesite or 
 cinder, which, melted into place, makes it possible to use the resulting 
 cindor in the open-hearth or blast furnace. These furnaces are used mainly 
 
420 
 
 ROLLING FINISHED PRODUCTS 
 
REHEATING FURNACES 421 
 
 for reheating heavy material, such as blooms, slabs, and the larger billets, 
 for which purpose they are very well adapted. 
 
 The Recuperative or "Continuous" Furnace works upon the principle 
 of counter-currents throughout. The combustion chamber is located at 
 one end of the furnace, where the heated steel is drawn, while the chamber 
 for recovery of the waste heat is located at the opposite end, which is 
 always nearest the stack and where the steel is charged. In one current, 
 the hot gases and flame from the combustion chamber are drawn by the 
 chimney draft over the floor, which is separated from the combustion 
 chamber by a bridge wall, and then downward through a series of spaced 
 iron pipes to the stack flue. In the other current the course of the steel 
 and the air for combustion run counter-current to the heat, the steel over 
 the floor of the furnace, the air through the enclosed space about the hot 
 pipes and a flue under the floor to the combustion chamber. The passage 
 of all may, therefore, be made continuous, hence the name, continuous 
 furnace. It will be observed that the billets move from the coldest part 
 of the furnace to the hottest part, hence they are heated very gradually, 
 reaching the rolling temperature just prior to drawing. The scale, there- 
 fore, does not melt, and no slag is formed in the continuous furnace if it is 
 fired with gas, oil or tar. When powdered coal is used for fuel, the silicious 
 ash unites with the scale to form an easily fused slag that collects at the 
 discharge end of the furnace. In order to push the billets through the 
 furnace suitable pushing devices must be provided, and to aid in this work 
 the floor of the furnace is sometimes inclined, sloping downward from the 
 charging to the drawing end. To prevent the tearing up of the floor, skids 
 for supporting the billets are provided, built into the bottom. These skids 
 are generally made of heavy pipe through which a stream of water flows to 
 keep them cool. An objectionable feature in the use of the skids is that 
 they cause cold spots in the steel where the billets rest upon them. To 
 overcome this defect the pipes are bent or off-set at the lower ends, or the 
 billets may be delivered from the skids to a section of the bottom lined 
 with magnesite. In this way the temperature of the cold spots is restored 
 to near that of the rest of the billet. 
 
 The Advantages of Continuous Reheating Furnaces are numerous. 
 In the first place, they are the best type of furnace to precede a continuous 
 mill. The use of complicated charging and drawing machines is avoided. 
 The heating, being confined to one end of the furnace, makes it easy to 
 regulate the temperature to suit the different grades of steel, and to heat 
 to the rolling temperature only those billets that are to be used at once. 
 Where the billets used are of constant length, the width of the furnace is 
 so proportioned to the length of billet that the entire bottom is covered 
 with the steel to be heated. Thus, there are no vacant areas on the bottom 
 to decrease the heating efficiency of the furnace. The accompanying 
 prints are intended to show the chief features in the modern construction 
 of these two types of furnaces. 
 
422 
 
 ROLLING FINISHED PRODUCTS 
 
 FIG 73 Continuous Heating Furnace. 
 
SHEARED PLATE 423 
 
 SECTION II. 
 
 THE ROLLING OF SHEARED PLATES. 
 
 Methods of Rolling Plates: As previously indicated, plates may bo 
 rolled either from ingots or from slabs, and on several different types of 
 mills. Thus, in England, and a few places in this country, Birmingham, 
 for example, plates are rolled on a two-high reversing mill consisting of a 
 train of two stands of plain rolls. In these mills, the stand nearer the engine 
 has both rolls driven and is used for roughing, while the second stand is 
 used' for finishing and has, therefore, only the bottom roll driven. In the 
 majority of these mills the rolls are run hot, that is, no attempt is made to 
 cool the rolls during the rolling. In America the practice for rolling plates 
 is entirely different. In all cases the rolls are kept cold by directing streams 
 or sprays of water upon them during the rolling, and two types of mill, 
 neither of which is like the English mills, are used. One of these, the 
 universal mill, has already been mentioned, and will be described more in 
 detail later, while the other, the invention of Mr. B. C. Lauth, of Pitts- 
 burgh, is a kind of three-high mill. In this mill the top and bottom rolls 
 are driven, are of the same size, and of large diameters, while the middle 
 roll is friction driven and, in diameter, is usually about two-thirds the size 
 of the other two rolls. The maximum size of the middle roll is determined 
 by the width of the housing windows, as the roll is removed by passing it 
 endways through this opening. The top roll can be raised and lowered in 
 the housing, and the middle roll, through suitable levers hydraulically or 
 electrically operated, can be brought into contact alternately with the top 
 and bottom rolls, which then play the part of re-enforcing rolls. Thus, in 
 making the bottom pass, the plate passes between this niiddle roll and the 
 bottom roll, while the top roll is used as a re-enforcing roll. On the return 
 pass, the middle roll is dropped down upon the bottom roll, and the piece, 
 having been raised to the proper level by a tilting table, passes between the 
 top and middle rolls. In either case, the amount of draught is controlled by 
 screw downs acting against the top roll in a manner somewhat like that 
 of the blooming mill. The advantage of this construction will become 
 apparent as this study advances. Plates rolled on this mill must be sheared 
 on all edges, hence they are called "sheared plates" to distinguish them 
 from universal mill plates which are sheared only on the ends to obtain 
 the lengths required. The size of the sheared plate mill is determined by 
 tha length of the bodies of its rolls, while universal mills are distinguished 
 by the maximum spread of the vertical rolls. 
 
 The One Hundred Forty Inch Mill at Homestead as ah Example of 
 a Sheared Plate Mill: In this mill the top and bottom rolls, which must 
 be carefully matched and of exactly the same diameter, are each approxi- 
 mately thirty-eight and three-fourths inches in diameter when new, and 
 
424 ROLLING FINISHED PRODUCTS 
 
 the middle roll twenty-two inches. All three are chilled rolls, the depth 
 of the chill being between one and one and one-half inches. The bottom 
 roll is held in place by bottom and side bearings of brass, which are fitted 
 into the bottom of the cast steel housings. For the top roll, which requires 
 both top and bottom as well as side bearings, riders for containing the 
 brasses are provided. This roll is supported from below by steel-yard rods 
 which extend from the bottom rider to the shorter arms of counterbalanced 
 levers in the pits beneath the mill. In this respect as well as in the method 
 of screwing down the top roll, the construction of the mill resembles the 
 forty-inch mill at Duquesne. For driving the screws, however, a 60 h. p. 
 motor is provided, instead of the hydraulic cylinder, and is connected to 
 the screws through a worm shaft and crown gear. For indicating the 
 draught on the mill a large drum or cylinder, about four feet in diameter 
 and with an altitude equal to the total lift of the mill, is mounted on the 
 top of one of the screws. The surface of this cylinder is divided vertically 
 into parallel spaces, the width of which equals the pitch of the screws, one 
 and one-quarter inches. The circumference of each circle separating a pair 
 of spaces is then divided by vertical lines into a number of equal parts. 
 By means of this arrangement and a stationary pointer, mounted on the 
 housing beside the cylinder and set to point at zero on the drum when all 
 the rolls are in contact and screwed down tight, the distance between the 
 rolls may be read off direct and with great accuracy. For holding the 
 middle roll in place, bearing boxes with side bearings which fit into chocks 
 placed on the side of the housing windows are provided. For keeping 
 this roll in line, liners are employed, and for raising and lowering it, a rest 
 bar built on the plan of a. swinging lever is fitted over each neck outside 
 of the housing and across the window. One end of each rest bar is supported 
 at an almost constant level by means of a turn buckle rod hung from the 
 top of the housing, while the opposite end is connected to the plunger of 
 a hydraulic cylinder which, located in the pit beneath the housings, furnishes 
 the power for raising and lowering the roll. In many cases this cylinder 
 is located on top of the housings, and in the most recently constructed mills, 
 electric motors are employed instead of the hydraulic cylinder. 
 
 The Drive and Connections : The top and bottom rolls are connected 
 to the pinions through coupling boxes and spindles similar to those in the 
 blooming mill. The spindles are eleven feet long, and both are supported at 
 their centers by suitable bearings. The saddle box for the vibrating 
 spindle is attached at one end to the pinion housing and at the other to 
 the roll bearing box, thus keeping the motion of the saddle and spindle 
 co-incident with that of the top roll. Since the middle roll of the mill is 
 friction driven, the middle pinion is used as a driving pinion only, and is 
 smaller than the top and bottom ones, the speed ratio being 11 to 19. The 
 pinions are of the helical toothed type and are held in cast steel housings. 
 A short spindle, four feet eleven inches long, connects the middle pinion to 
 the driving shaft of the engine on which is mounted the fly wheel. This 
 
SHEARED PLATE 425 
 
 mill is driven by a 42" x 66" x 60" tandem compound engine, capable of 
 giving 3500 h. p. at the speed of 64 r. p. m. Nearly all the new mills built 
 since 1916 are electrically driven. The new one hundred ten inch mill at 
 Homestead, otherwise known as the Liberty mill, is so driven. This motor 
 was built and installed by the .General Electric Co. It is designed to 
 develop 4000 h. p. and to give a speed of 82 r. p. m. on full load. It uses 
 alternating 3-phase current with a frequency of 25 cycles per second and a 
 pressure of 6600 volts. The installation is marked for its simplicity; the 
 peak loads are taken care of by means of a 55-ton fly wheel mounted on 
 the same shaft with the motor. 
 
 Difficulties in Rolling Sheared Plates: While the rolling of plate 
 may appear to the novice as one of the simplest of rolling operations, yet 
 there are problems connected with the rolling of wide plate that require 
 the combined skill and experience of the heater, the millwright, the roller, 
 and roll designer to overcome. If the slabs are not heated uniformly in 
 all parts, the plates will curl and buckle in rolling, while a similar effect 
 is produced if the rolls are even slightly out of alignment. The stretch 
 of the housings and the stoving up of the screw are minor considerations 
 in overcoming the difficulties of rolling exactly to gauge. The wearing 
 away of the rolls, which in actual operation, takes place faster in the middle 
 portions than at the ends, causes them to become hollow in a short time so 
 that the plate is thicker in the middle than at the edges. Since the pressure 
 for rolling must be applied at the ends of the rolls, this effect is increased 
 by the bending of the rolls. The opposite effect would be produced if the 
 rolls should become hot in the middle, the expansion causing an increase in 
 their diameters. This last complication is avoided by keeping the rolls cold 
 with water sprays above them. This water, running down upon the plate, 
 has a tendency to cool it faster, but this cooling is not as rapid as might 
 be expected, because the water 1 assumes the spheroidal state on striking 
 the very hot plate and glides off without being vaporized to any great 
 extent. At the one hundred forty inch mill, the spring and wear in the 
 rolls are provided for in the following manner : To remove the effects of wear 
 the top and bottom rolls are dressed do wn every Saturday, either in position 
 by attaching an electrically driven reduction gear to the driving pinion, 
 which virtually converts the mill into a lathe, or by removing them and 
 sending them to be lathe turned in the machine shop. - To neutralize the 
 spring in the rolls the middle roll is turned so that its diameter at the middle 
 is a little greater than that at the ends. If this swell or belly in the roll 
 were made to fit the top and bottom rolls it would almost represent an 
 arc of a very large circle, but as it is impossible to dress the roll in this 
 way, the lines are cut only approximately correct by tapering the ends and 
 leaving the central portion of the roll cylindrical in form. The amount of 
 the taper will vary with the hollowness of the mill, but is never less than 
 
426 ROLLING FINISHED PRODUCTS 
 
 one sixty-fourth nor greater than three sixty-fourths of an inch, thus making 
 the difference in diameter between the ends and the center vary from one- 
 thirty-second to three-thirty-seconds of an inch. The distance from the 
 end of the roll to which the taper extends may vary from forty-six to fifty- 
 six inches, thus making the central cylindrical portion twenty-eight to 
 forty-eight inches long, and depends upon the width of plate being rolled. 
 It is the practice, when advantageous, to roll the wide plates at the begin- 
 ning of the week, while the mill is full, and to roll the narrower plates at 
 the end of the week when the rolls have been worn down, and the hollow- 
 ness of the mill is more pronounced. Even with these changes the mill 
 will often become so hollow that it is necessary to roll the edges a little 
 below gauge in order to get the weight correct. Hollowness in the mill 
 tends to make the edges of the plate dovetail or buckle. During the week 
 the middle roll will be changed four to six times. 
 
 The Rolling Process on this mill, as on the English mill, may be looked 
 upon as being performed in two steps or stages, namely, a roughing or 
 breaking down stage and a finishing stage. In the breaking down of the 
 slab the most important feature of the rolling is the determination of the 
 draughts, the size of slab and the spring of the rolls being the controlling 
 factors ift this regard. With a heavy slab, that is, one six to ten inches 
 thick, a maximum draft, or bite, of about three-fourths inch is possible. 
 The amount of bite decreases as the slab thickness decreases and the width 
 increases. The greater the surface the less the possible draught on account 
 of the greater amount of work necessary in rolling. Following the first 
 few passes the draughts become smaller and smaller because of the increased 
 work required and also to allow material for the finishing. At least one- 
 fourth inch is allowed for the finishing passes, as this amount is required 
 to give sufficient material with which to remove the effect produced by 
 the spring of the rolls. Were the spring not removed, that is if the plates 
 were finished by a continuance of passes carrying the heavy draughts, the 
 middle of the resulting plates would be much heavier than the sides. By 
 decreasing the draughts the "spring" is removed, and the plate approaches 
 nearer the desired weight and gauge. Blind passes, that is, passes in which 
 no additional pressure is applied to the rolls, are also used in finishing for 
 the same reason. The heavier gauge plates cause less spring in the roll 
 and fewer finishing passes are necessary, while with light gauge plates, 
 especially on a full mill, it is necessary to start the finishing when about 
 one-half inch above the final gauge, since a bigger draught is necessary to 
 hold the plate and prevent buckling. Near the beginning of the rolling, 
 the slabs are passed through the mill transversely a few times to obtain 
 the desired width and are then rolled longitudinally. In gauging the width 
 six inches are allowed for shrinkage, shearing, etc., on plate not over eighty- 
 five inches wide, while seven to eight inches are allowed on plate from 
 
\ 
 SHEARED PLATE 427 
 
 ninety inches to the maximum of a hundred thirty-two inches. Extra 
 allowance on the wide plates is necessary to take care of the overlap, or 
 lamination, of the top and bottom sides due to the greater flowing of the 
 metal on these faces during rolling. Any large variation in the distance 
 between the rolls from end to end shows up in the plate at once, since under 
 such conditions it will curve to the side on which there is the less draught. 
 If the variation is small it may not show up in this way. The plates are 
 gauged four times daily at the edges and in the middle to determine the 
 hollowness and variation. Excessive hollowness is corrected by putting in 
 a new middle roll, while variations are overcome by inserting or removing 
 liners under the bearing boxes. During the rolling, scale is removed with 
 salt on common steels or with burlap and coal on nickel steel, as in the 
 rolling of slabs. 
 
 Cooling and Straightening: In order to keep the cooling of plate 
 uniform, the roll tables are preferably provided with collared, or disc rolls 
 instead of plain ones, which cause black streaks to appear across the plate. 
 Disc rollers in the tilting tables of the mill also make it possible to roll 
 very narrow slabs. These slabs are difficult to handle on plain rolls, because 
 they do not ride these rolls in a horizontal position, but tend to fall down 
 between them edgewise. This difficulty is overcome with the disc rollers, 
 for by alternating the discs the bearing surface from roll to roll is brought 
 nearer together than is possible with plain rolls, and there is no straight 
 line of separation between rollers. After the rolling is completed, the 
 plate is passed by roll tables to a Hilles & Jones cold roll straightener. 
 This straightener is constructed of two rows, one above the other, of small 
 rolls, five in the top and four in the bottom row. The centers of the rolls 
 in the top and bottom rows are alternated, so that the piece, on entering 
 between the rolls, is given a long bend or sweep which is then removed by 
 smaller sweeps until the piece, on passing out of the rolls is quite or nearly 
 straight. Two or more passes may be required to straighten some plates. 
 The best work is done on plates about one-half inch thick, as they still 
 retain enough heat when arriving at the cold rolls to be well and easily 
 straightened. The rolls are operated by motors, and the draught is set on 
 the mill by gears connected to a motor. From the rolling order sheets, the 
 cold roller obtains the gauge of the plate when finished and then sets the 
 cold rolls to take a plate of that gauge. 
 
 Laying=out and Stamping : After being straightened, the plates pass 
 to the cooling bed and then to the marking tables. Here the heat number, 
 slab number, customer's name, or mark, and plate dimensions are written 
 on the plate by the marker. The plate is now ready to be laid out by the 
 gauger. Laying out consists of drawing out in chalk the plate or plates 
 as they are to be sheared, it being remembered that the plate as rolled 
 from a slab may be made up so as to give material for two or more plates, 
 in which case the plate as rolled is called a combination. As the plates 
 
428 ROLLING FINISHED PRODUCTS 
 
 are still hot when gauged, allowance must be made for shrinkage, this 
 allowance amounting to one-fourth inch in width and length for each hundred 
 inches on plates up to one-fourth inch in thickness, and three-eighth inch 
 for each hundred inches of length or width on gauges over one-fourth inch. 
 This difference is necessary on account of the higher finishing temperature 
 of the thicker plates. In case plates come to the marking table that will 
 not make the plates ordered, on account of being too short or too narrow, 
 the material must either be applied on another order or put in stock. Any 
 plates with objectionable surface defects are rejected at the marking table 
 and stocked. Plates stocked at the table are treated in the same manner as 
 rejections, although they are not listed as such, because they have not been 
 finished at the time of stocking. If a plate has a snake or is pitted on one 
 end, this part may be sheared off and the remainder used on a different 
 order than that for which it was intended. Plates that are too narrow to 
 make the ordered plates are held until an order can be found calling for such a 
 gauge, width and length as can be cut from them. It is a duty of the stock 
 marker to replace such plates as are taken from the marking table. A 
 report is made on the rolling order sheet, by the marker, of the number 
 of plates made on an order, and the stock marker informs the office of all 
 plates made from stocked material. While plates are being laid out they 
 are stamped, giving heat number or slab number as is desired, and any 
 other stamp that may be called for on the order by the purchaser. Heat 
 number, slab number and size is also painted on the plates with white lead 
 after they are marked. Some orders require that heat numbers, etc., are 
 to be painted instead of being stamped, this being more generally the case 
 on very light gauge plates. The method used in laying out a plate can be 
 described from sketches, such as those shown in Fig. 74. 
 
 First, the width of the plate is taken at AB to determine the amount 
 of stock over that of the ordered width plus allowance for shrinkage. The 
 stock is then divided so as to give one-half to each side. Point C is set 
 leaving the distance BC as excess stock. Point O is now set along XY 
 in the same manner as point C. Line CO is drawn with a chalked string 
 or a straight edge. Widths OM EN CP and others, if needed, are 
 measured using line CO as a base. Line PM is then drawn by "the line 
 drawer." A right angled square is now used to draw line CP at right 
 angles to CO, thus squaring up the plate. The true length of the plate 
 is taken along CO, using C as a starting point, and OM is likewise 
 drawn at right angles to CO. 
 
 In case the plate is curved, the base lines must take the direction 
 indicated in Sketch II in order to avoid scrap, and in case the length of 
 the plate ordered is such as X'Y' the plate cannot be made on account of 
 the curve. This plate must be applied on orders that require lengths of 
 approximately XI and QR, with widths and gauge such as can be obtained 
 on the given plate. Whenever the number of plates ordered to the same 
 
SHEARED PLATES 
 
 dimensions will justify the expense, patterns of wood are made for laying 
 out. Sketch plates are always marked out to a templet, or pattern. 
 
 Test Pieces : In laying out the plate sufficient material must be given 
 to allow for the test pieces that are required. On sheared plates both 
 longitudinal and transverse test pieces are often taken. 
 
 Shearing: Three shears, one end and two side shears, are used on the 
 one hundred forty inch mill, and are so arranged that the plate does not 
 require turning to shear the sides. The ends are sheared first, and the 
 
 FIG. 74. Sketches Illustrating the Laying Out of Plates. 
 
 plate is then passed over castors to the side shears. This mill is also 
 equipped with a rotary shear for heads and other circular plates, an alligator 
 shear and a scrap shear. 
 
430 ROLLING FINISHED PRODUCTS 
 
 Shearing Tolerances: It is evident, even to the casual observer, that 
 conditions at the mills are such that shearing to exact dimensions is imposs- 
 ible. Observations made on any one mill will reveal many of these un- 
 favorable conditions and also show that it is impossible to remedy them. 
 The plates must be sheared in the order they are rolled, and, to keep up 
 with the mill, rapid working is required, a condition that makes it difficult 
 to lay out or shear accurately. Owing to variations in the thickness of the 
 plates and in the time required for rolling them, they leave the cooling 
 beds at widely varying temperatures. Since there is no way of knowing 
 accurately just what this temperature is at the time of laying out the plate, 
 proper allowance cannot be made for the shrinkage, the total amount of 
 which will also vary with the dimensions of the plate. Thus, while a plate 
 H m ch thick and 100 inches long may require an allowance of ^ inch 
 for length, a plate one inch thick and 200 inches long may require an 
 allowance of % inch. Finally, since no mechanical stops can be used on 
 plate shears, the plates must be adjusted to position under the shear knife 
 by eye-and-hand methods, which are not favorable to accurate work. Since 
 the conditions in the different mills, such as length and type of cooling bed, 
 kind of material rolled, etc., vary a great deal, it is practically impossible 
 to fix standard variations covering all kinds of plates that will be just to 
 the mills and the consumers alike. In justice to the former it should be 
 stated that every attempt is made to shear as near to the exact dimensions 
 ordered as the class of material would appear to call for and the mill 
 conditions will permit. 
 
 Inspection for Size: After the shearing, all plates are inspected for 
 size. If a plate does not measure up to the dimensions ordered or to within 
 the tolerances permitted by the order department, it is rejected and return- 
 ed to be applied on another order calling for the same grade of material. 
 
 Weighers : All plates are weighed separately, the weight and number 
 of plates made being recorded on a copy of the rolling order sheet given 
 to the weigher. 
 
 Checkers : The checker receives a copy of all rolling orders and checks 
 each item for size and pieces ordered. On the weigher's copy of the rolling 
 .order, he lists the estimated weight of the plate as ordered so as to give 
 the weigher the ordered weight, who, after taking the actual weight, can 
 determine at once whether the plate will meet the specifications as to 
 weight. The checker lists all plates ordered in the order book, and receives 
 a copy of all orders to be rolled on the mill, so as to avoid the making of 
 duplicates in case a plate is ordered twice. In case error is found in dimen- 
 sions of plates listed OL the rolling order, the checker informs the roller 
 and marker. The checker also lists all plates made in the order book, 
 thus keeping a record of the plates still on order. 
 
UNIVERSAL MILL PLATE 431 
 
 Slip Maker: The slip-maker's duty is to make a form giving the 
 following information: mill, slip number, customer's name, Carnegie order 
 number, sheet number of order, heat number, number pieces made, dimen- 
 sions, marks and actual weight. The dimensions, etc., are to be takenfrom 
 the plate, formerly painted on by the painter at the marking table. A 
 copy of the rolling order is at the disposal of the slipmaker to aid in identi- 
 fying plates, but data cannot be taken from the order without seeing the 
 marks on the plate, in as much as certain plates may, at the marking tables, 
 be applied on orders different from those they were originally intended for. 
 A new or separate slip is made out for each order number. The signature 
 of the slipmaker is required on each slip for identification. 
 
 Recorder : The duty of the recorder is to test the weights of all plates 
 made and note the turn on which they were rolled and sheared. The 
 record here taken is the final record of the product made and must be 
 taken very accurately. Special forms are used for recording, showing turn, 
 numbers, weight, descriptions of plates, etc. Special sheets are made for 
 alloy steels and small pieces that are inspected at the time of measuring 
 for size. 
 
 SECTION III. 
 
 UNIVERSAL MILL PLATES. 
 
 The Forty-eight Inch Mill at Homestead as an Example of Universal 
 Plate Mills: This mill consists of two horizontal rolls and four vertical 
 rolls, two on each side of the horizontal ones, all contained in the same 
 housings and driven by the same engine. This construction makes the mill 
 a very complicated piece of machinery, a detailed description of which 
 would be too lengthy to attempt here, so only such matters as are necessary 
 to an understanding of the working of the mill will be discussed. While the 
 working surface of the horizontal rolls is only four feet, which is the same 
 as the maximum spread of the vertical rolls, the total length of the rolls 
 is thirteen feet one inch, of which length twenty-four inches makes up the 
 wobblers, forty-five inches the necks, and forty inches, twenty inches on 
 each side, crosses the spaces in front of the vertical rolls, which must stand 
 between the housings. These rolls are connected to the engine and are driven 
 in the same way as the ordinary reversing mill. The total lift of the hori- 
 zontal rolls is twenty inches. The screw down on these rolls is the same 
 as that on the sheared plate mill, and the same form of graduated drum is 
 used to indicate the lift of the rolls, or the gauge of the plate. The vertical 
 rolls, whose centers are located three feet one inch from the centers of the 
 horizontal rolls, are seventeen and one-half inches in diameter at the body, 
 and the height of their rolling surface is two feet four and one-eighth inches. 
 They are provided with bearing boxes at both their tops and bottoms. A 
 five inch collar on the upper end of the lower neck rides on a side bearing 
 
432 ROLLING FINISHED PRODUCTS 
 
 in the bottom box which furnishes the vertical support for the roll. A 
 screw, bearing on a frame attached to the bearing boxes and actuated by 
 an electric motor, furnishes the means by which the pressure for rolling is 
 applied to these rolls; for spreading them hydraulic jacks are used. Large 
 discs, graduated on their circumferences and mounted on the screws, indicate 
 the spread of the vertical rolls. As already stated, these rolls are driven 
 through a system of gears by the same engine that drives the horizontal 
 rolls. Beginning with the engine, the power is transmitted to the horizontal 
 rolls in the usual manner for reversing mills, while the drive for the vertical 
 rolls is taken off the upper pinion. Upon the prolongation of the outside 
 bearing of this pinion, a second gear is keyed. This gear meshes with two 
 idlers, one on either side, which in turn mesh with gears mounted on the 
 ends of the two drive shafts for the vertical rolls. These shafts then extend 
 to and across the roll housings, where they are supported by suitable bear- 
 ings. On the section of these shafts included between the roll housings are 
 mounted four sliding miter gears which mesh into similar crown gears keyed 
 to the tops of the rolls. Through these gears the peripheral speed of the 
 vertical rolls is adjusted to equal the speed of the horizontal rolls, and 
 never more. Hence, the vertical rolls may be used on the piece only on 
 the entering side of the passes, because if the vertical rolls were used on 
 the delivery of the plate the greater speed of the piece due to the elongation 
 produced by the horizontal rolls would jam the material between the two 
 sets of rolls. The rolling of the piece on the entering side is preferable to 
 rolling on the delivery side, as then thin plates would tend to buckle or 
 bow up in the center on applying pressure from the vertical rolls. These 
 rolls cannot be brought closer together than twenty inches. Hence, the 
 mill has a range in width of plates from twenty to forty-six inches. 
 
 The Operation of Rolling: Rolling universal mill plates involves 
 most of the difficulties of rolling sheared plates, and in addition there are 
 several features, due to the vertical rolls, that are not peculiar to sheared 
 plate mills. Thus the piece must always enter the mill at right ang] es to the 
 horizontal rolls, as otherwise the action of the vertical rolls will cause the 
 plate to buckle or curl or jam between the rolls. As the plates rolled on 
 this mill are in very long lengths, the slightest variation in the spacing of the 
 horizontal rolls shows up as a decided camber in the plate as it runs out 
 on the table. To correct this defect, which is prone to occur on universal 
 mills, a spanner block is placed under the screw down on the roller's side 
 of the mill. By means of a spanner bar and a sledge hammer, this block 
 may be turned and the proper adjustment made on this end of the upper 
 horizontal roll to cause the plate to roll straight. As to the draught and 
 manipulation of the horizontal rolls, the plate is reduced in the same way 
 as on the sheared plate mill. On the vertical rolls the greatest draughts 
 are taken in the first few passes, the object being to reduce the piece to 
 the desired width as quickly as possible, after which the pressure on the 
 vertical rolls is just sufficient to hold the piece to width. This mill rolls 
 
UNIVERSAL MILL PLATE 433 
 
 many plates directly from the ingot. Ingots intended for this purpose are 
 rectangular in section, beijig from one to two inches wider at the smaller 
 end than the width of the plate desired. In beginning the rolling of these 
 slab ingots the rollers prefer to have the small end enter the mill first, as 
 in this way the suddenness of the pull on the mill is avoided and the draught 
 can be more easily adjusted, but many plates are rolled with the butt end 
 of the ingot entering first. The chief objection to rolling plates directly 
 from ingots is that the pipe and central line of segregation is rolled into 
 the plate, and in order to avoid it the scrapping of a large amount of 
 finished material is necessary. 
 
 Straightening. Marking and Shearing U. M. Plate: From the rolls, 
 the plate is carried on live roller tables to the two cooling beds, which 
 extend in opposite directions from both sides of the receiving table. Here 
 any curve or camber is removed from the plates by clamping them tightly 
 to a straight edge. The buckles thus produced on the edges of the plate 
 are then flattened out with wooden mallets. While the mill is provided 
 with a machine straightener, similar to the one employed at the one hundred 
 forty inch mill, it is seldom used, the general practice at universal plate 
 mills being to straighten the plates in the manner described above. Every 
 plate rolled on this mill has the name and letters "Carnegie, U. S. A." 
 rolled into it at intervening spaces of seven feet. While on the cooling beds 
 the plates are marked off for length, the heat number is stamped on, and 
 the slab number, size of plate, order number and the customer's name is 
 marked on with white paint. The plates then move on to the receiving 
 tables, which carry them to shears, where the plates are cut to length. A 
 large shear used for splitting plates is also provided. Unless otherwise 
 specified, two longitudinal tests for the physical laboratory are taken for 
 each order or on each heat of steel; one test is taken from the top of the 
 ingot, and the other from the bottom. The weighing, recording, and in- 
 spection of the plates are then conducted as for sheared plates. 
 
 Advantages of Universal Mill Plates: While the effect of the one 
 way rolling on Universal mill plates, as will be explained shortly, is such 
 as to require care and discrimination in their use, they, nevertheless, 
 possess certain advantages over sheared plates that make them more 
 desirable for some purposes. First, the possibility of producing plates of 
 great length with a rolled edge makes them available for many purposes, 
 such as girder construction, for which sheared plates are not suitable. 
 Second, the ability to roll to fairly exact widths reduces shearing costs to 
 a minimum. Third, the rolled edge eliminates all costs to the purchaser 
 for machining. As a fourth advantage the greater tonnages that these 
 mills are capable of producing may be cited, because this tends to keep the 
 first cost to the customer low. 
 
 Physical Properties of Plates : The effect of rolling on the physical 
 properties of steel is now generally recognized by the users of plates, and 
 
434 ROLLING FINISHED PRODUCTS 
 
 specifications are usually written accordingly. In a previous discussion of 
 this subject it was -made plain that the controlling factors during rolling 
 are the amount of work done and the temperature above the critical at 
 which the rolling is completed. To these there should now be added the 
 manner in which the rolling is performed. Attention has been called to 
 the different methods of rolling plates in the preceding description. It 
 will be recalled that the plate may be rolled from the slab with one reheating 
 or from the ingot direct without reheating. As to the difference in effect 
 produced by these two methods, there is little data on the subject, but 
 reasoning from the theoretical standpoint, there should be no difference. 
 The abandonment of the method of rolling from the ingot is probably due 
 to economic considerations rather than to any tendency of the method to 
 produce defective material. Again, in the rolling of the slab or ingot it 
 was pointed out that all the rolling may be in one direction only, or the 
 plate may be rolled both transversely and longitudinally. Here a marked 
 difference is observed to result from the two methods of rolling. For 
 example, if a slab or ingot be rolled in one direction only and a longitudinal 
 and transverse test piece be cut from the resulting plate, little difference 
 in tensile strength will be observed in pulling the two tests, but a marked 
 difference in ductility will be found. Thus, the longitudinal piece will give 
 from 4% to 7% greater elongation than the transverse piece, and 10% to 
 15% greater reduction in area. Concerning the amount of work and finish- 
 ing temperature, the ductility is affected in a somewhat erratic way, 
 while the tensile strength is increased by increased work and lower 
 finishing temperatures. Thus, a thin plate will show a very appreciable 
 increase in tensile strength over a thick one rolled from the same 
 slab or ingot, and to obtain the same strength in plates of 
 different thicknesses it is necessary to employ chemical control. The 
 following table is intended to show approximately the variations in the 
 carbon content, other metalloids being constant, that should be made to 
 produce plates of uniform strength when varying in thickness as indicated. 
 
 Table 57. Showing Variation of Carbon Content with the Thickness 
 of Plates to Give the Same Strength. 
 
 THICKNESS CARBON 
 
 OF PLATE REQUIRED 
 
 %" - 12% 
 
 Jie" 15% 
 
 W - - 17% 
 
 M"~- - 18% 
 
 1" 19% 
 
 1W - - --.20% 
 
 Inspection of Plates for size and weight is made by mill inspectors, 
 while inspection for surface and other defects are made either by mill or 
 customer's inspectors. Certain surface defects, as snakes and surface 
 
DEFECTS IN PLATES 
 
 435 
 
 marks, may do the plate no harm if they do not extend too deep into the 
 metal, and may be ground out with any suitable- device. For this purpose 
 a movable electrical grinding wheel is employed. The rigidness of 
 the inspection may be surmised from a glance at the inspector's list of causes 
 for rejection, which is here appended. Most of these reasons are self- 
 explanatory, while others have already been discussed, so that no further 
 explanation is required. 
 
 Table 58. Defects for Which Plates are Rejected. 
 
 DEFECTS ACQUIRED OR CAUSED FROM 
 
 THE 
 INGOT 
 
 IN 
 HEATING 
 
 IN 
 ROLLING 
 
 IN LAYING 
 OUT 
 
 IN 
 
 SHEARING 
 
 FROM MANY 
 SOURCES 
 
 Blister 
 
 Burnt 
 
 Split 
 
 Wrong 
 
 Scant 
 
 Snake 
 
 Pipe 
 
 Bricked 
 
 Slivers 
 
 Dimen- 
 
 Bad Edge 
 
 Seams 
 
 Slivers 
 
 Scabby 
 
 Dished 
 
 sions 
 
 Crop end 
 
 Test lost 
 
 Seams 
 
 Cinder spot 
 
 Pitted 
 
 Made 
 
 Test piece 
 
 Duplicate 
 
 
 
 Scored 
 
 wrong 
 
 cut 
 
 Broken 
 
 
 
 Buckled 
 
 
 Knifed 
 
 Ground 
 
 
 
 Over 
 
 
 
 too deep 
 
 
 
 weight 
 
 
 
 Cracked 
 
 
 
 Under 
 
 
 
 Tests are 
 
 
 
 weight 
 
 
 
 not within 
 
 
 
 Over and 
 
 
 
 specifica- 
 
 
 
 under 
 
 
 
 tion 
 
 
 
 gauge 
 
 
 
 
 
 
 Scale 
 
 
 
 
 
 
 Cambered 
 
 
 
 
 
 
 Laminated 
 
 
 
 
 
 
 Roll 
 
 
 
 
 
 
 marked 
 
 
 
 
 
 
 Bad edge 
 
 
 
 
 
 
 (Universal 
 
 
 
 
 
 
 Mill 
 
 
 
 
 
 
 Plates 
 
 
 
 
 
 
 only) 
 
 
 
 
436 THE ROLLING OF SECTIONS 
 
 CHAPTER VIII. 
 
 THE ROLLING OF LARGE SECTIONS. 
 
 SECTION I. 
 
 RAILROAD RAILS. 
 
 Development of Rail Manufacture: Dating from the invention of 
 the steam locomotive, the railroad rail represents one of the first sections 
 with which the rolling mill operators had to deal, as well as one of the 
 most difficult and certainly the most important. The importance of the 
 railroad as a factor in modern civilization and progress is recognized by 
 all, and that the rail is a most vital part in railroad operations is just 
 as evident. With the advancement in speed of travel and weight of 
 loads carried, more and more has been required of the rail, until to-day 
 no material is subjected to more severe punishment in service than the rail- 
 road rail. Exposed to the weather at all times, it is subjected, under 
 constantly varying conditions, to immense compression and bending stresses, 
 shocks, vibrations, friction and wear. The form of the rail,* then, should 
 be such as will give the greatest transverse strength, provide abundance 
 of metal for wear, present a wide base for fastening to the cross tie, and 
 still, for the sake of economy, be of the lightest section possible. Now, it 
 so happens, that the form that best meets all these requirements is the 
 section known as the American Tee Rail. It also happens that this section 
 was, in the early days of rolling mills, one of the most difficult sections 
 to roll, mainly on account of the wide flange. The history of rail develop- 
 ment as indicated in the sketches of Fig. 75 gives evidence of this fact. 
 
 Thus, the first real departure made from the original strap rail of 1808 
 was the chair rail of 1820. As the chair of this rail was expensive, an attempt 
 was made in the section of 1831 to roll a rail with a wide and relatively 
 heavy flange on the bottom to replace this chair. The difficulty of rolling 
 the flange led to the better balanced bull head of 1837, the U-shape of 1844 
 and the pear head rail of 1845. Then came the compound rail of 1856 and 
 the form of 1860, which is the U-shape of 1844 with the lower parts closed 
 in and welded to form the web. As neither of these forms proved service- 
 able, a demand for more metal in the head for wear forced a final return 
 in 1865-8 to the tee shape with wide thin flange. From this date the design 
 of rolls, quality of material, and lay-out of the mills has gradually been 
 improved until at the present time the American rail mills are not only 
 producing the largest tonnage of the world but also rails of the best possible 
 grade. 
 
RAILS 
 
 437 
 
 Methods of Rolling Rails: Rails were originally rolled on the pull- 
 over mill, and then on the reversing mill, which in England is the type 
 of mill still employed for this purpose. But in this country all rails are 
 rolled on the three-high mill, which formerly was usually made up of a single 
 train of three stands driven with one engine. With the increase in the size of 
 the section, which has almost doubled in weight within the last quarter* 
 century, and the growing demand for larger quantities and better quality 
 in the product, a more advantageous lay-out of the mills for handling this 
 
 1808 1820 
 
 Strap Rail Birkenshaw Chair Rail 
 19 Ibs. per yd. 25.8 Ibs. per yd. 
 
 1830 
 
 Clarence Chair Rail 
 33 Ibs. per yd. 
 
 1831 1837 
 
 Stevens', (the 1st. T-rail) Lock's (BullHead) 
 40.8 Ibs. per yd. Rail, 58 Ibs. per yd 
 
 1844 
 
 Evans' 40 Ib.U-Rail 
 
 First Rail Rolled in 
 
 United States. 
 
 1845 
 
 58 Ib. Pear Head 
 
 Rail, First T-rail 
 
 Rolled in U. S. 
 
 1856 
 
 Compound Type of 
 Rail, 60 Ibs. per yd. 
 
 1858 
 
 P. R. R. Standard 
 85 Ibs. per yd. 
 
 1883 1892 
 
 P. H. Dudley Design P. R. R. Standard 
 
 80.2 Ibs. per yd. 100 Ibs. per yd. 
 
 1910 
 
 C. R. R. of N. J. 
 135 Ibs. per yd. 
 
 1868 1874 
 
 Welch Design Chanute Design 
 67 Ibs. per yd. 60.3 Ibs. per yd. 
 First Bess, Rail 
 Rolled in U. S. 
 (1865) similar 
 
 to this 
 
 Fia. 75. Sketches of Rail Sections Illustrating the Evolution of the Railroad Rail 
 in America. 
 
 material became necessary. The more modern rail mills wilt, therefore, 
 consist of two or three trains, each separately driven and made up of one 
 or more stands of rolls, all so arranged that the rolling of the piece in any 
 one stand is complete before it is passed to the next succeeding one. With 
 this arrangement the output of the mill is greatly increased without much 
 increase in the speed of the rolls, because different pieces may be rolling 
 at different stages at the same time, and the turning of the piece between 
 passes is avoided. As to the heating of the steel, rails may, as previously 
 
438 THE ROLLING OF SECTIONS 
 
 stated, be rolled either from blooms that have been reheated after having 
 been rolled from the ingot, or on the original heat of the ingot. The latter 
 method, which was introduced a few years ago mainly to save the extra 
 cost of reheating, was until very recently looked upon with favor both by 
 the manufacturer and the consumer, who believed that the scheme would 
 have a beneficial effect upon the quality of rail produced, due to the fact 
 that the material was necessarily finished at a low temperature. Within 
 the last two years, however, sentiment appears to have taken a swing in 
 favor of reheating, because, as is claimed by the advocates of reheating, 
 the increased speed of rolling combined with the heavy draughts required 
 to complete the rolling on the original heat is liable to produce a condition 
 favorable to the formation of fractures. The effects of too rapid reduction 
 have already been discussed. For the same reason the temperature of th6 
 ingot is kept high, which fact increases the danger of overheating. In 
 addition, the difficulty of keeping the finishing temperature constant, pre- 
 sents a serious problem. On the other hand, by reheating the bloom the 
 two initial rolling temperatures may be lower and be kept more uniform, 
 and the shaping of the rail may progress more leisurely. As to the manner 
 of forming the section, there are two methods of rolling, known as the flat, 
 or slab-and-edging, and the diagonal, or angular, method. To impart even 
 a slight understanding of these methods requires a lengthy explanation; 
 but in the proper design of the various passes for the progressive forming 
 of the rail, as for any section, lies the crux of the rolling process. Therefore, 
 as the subject is one of great interest, an attempt is to be made to discuss 
 the matter in as brief and comprehensive a manner as possible under the 
 headings that follow. 
 
 How to Study Roll Design : The best way to explain roll design is 
 by an example, for it is as yet an art acquired mainly by experience. While 
 subject to natural laws, the scientific aspects of the subject have not been 
 fully developed, and the roll designer has few rules to learn. This con- 
 dition tends toward individuality in designing, with the result that it is 
 seldom two designers will be found to do the same thing in the same way. 
 To serve as such an example the flat method of rolling as carried out at 
 the Edgar Thomson Works will be described, because, of the two methods 
 of rolling, this is the older and the one more generally employed. Before 
 beginning with the example, however, some preliminary explanations are 
 required. 
 
 Precautions to be Observed in Designing the Rolls: Of course the 
 first consideration in roll designing is to produce a finished piece of the 
 correct size and form, and this must be done by spreading, bending and 
 directing the flow of the steel. The ease with which this forming is done 
 depends on the plasticity of the metal, which in turn is affected by the kind 
 of steel, whether open-hearth or Bessemer; the grade, whether high or low 
 carbon; and the temperature. With the speed of the rolls fixed, the tem- 
 perature confined to a very narrow range, and the kind and grade of steel 
 
RAILS 439 
 
 given, the only instrumentality remaining in the hands of the roll designer 
 is the size and shape of the passes, and in part of these, at least, the size 
 will be governed by the size of. the bloom. In designing the passes, a good 
 designer will endeavor to work the steel in such a manner that the quality 
 of the product will be benefited, and no defects will be developed. The 
 defects that require constant care are fins, laps, overfills and underfills 
 Laps may result from fins or a collaring of the piece in the rolls; overfills, 
 from worn rolls, bad or improper design; and underfills either from bad 
 design or incorrect adjustment of the rolls. 
 
 Stages of Reduction: The formation of the rail from the bloom may 
 be looked upon as taking place in three steps or stages. The first stage, 
 called the roughing, is merely one of preparation; in it a large amount of 
 work is done, but this work is expended mainly in reducing the size of the 
 section and elongating the piece. At the Edgar Thomson Works the piece 
 is reduced in seven to nine roughing passes. In the first four to six passes, 
 rolling by the slab-and-edging method, the section retains the rectangular 
 shape, while in the next three passes a little shaping of the flange is begun 
 to prepare the piece for the first finishers, in which there are five passes. 
 These passes are given names in order, indicative of the nature of the work 
 they are intended to perform, as follows: slabber, first former, second 
 former, third former, and the leader. The leader is the pass just previous 
 to the finishing pass, which is located in a separately driven stand of two- 
 high rolls. With this explanation of terms used, the different steps in the 
 design of the rolls and the rolling may now be traced. They are as follows: 
 
 The Section: No original designing of section is done by the roll 
 designer. The first requirement in the rolling of a new section is, then, that 
 the roll turner be supplied with a drawing or print of the section, which 
 must be accompanied with all the dimensions, preferably indicated on the 
 print. The weight of rail desired or expected should also be given. Here 
 the matter of dimensions is of extreme importance, for the designing of the 
 templets cannot be started until each and every dimension required is 
 given. These dimensions not only include linear measurements, such as 
 height of rail, width and thickness of parts, but radii of all curves, and 
 amount of slope on inclined surfaces expressed in degrees or percentages. 
 
 The Cold Templet: With all the necessary information before him, 
 the first step taken by the roll designer is to prepare a drawing for the 
 cold templet. This templet is made of brass and represents an exact section 
 of the rail when cold. This drawing is constructed on the axis of symmetry 
 of the rail, which is the vertical line drawn through the center of the head, 
 of the web, and of the flange. On this line the section of rail is symmetrically 
 constructed to the dimensions given on the drawing, all the dimensions 
 being made with extreme care and accuracy. All measurements are made 
 with micrometers or steel rules. If a rule is used a magnifying glass is 
 employed to take the readings. The construction lines are made as fine as 
 possible. Even the contraction and expansion of the drawing paper, due to 
 
440 
 
 THE ROLLING OF SECTIONS 
 
 the varying humidity of the atmosphere, is taken into consideration, and 
 the proper allowances are made. With this very accurate drawing com- 
 pleted, the area of the section is measured with a planimeter in order to 
 check up the weight of the section. If this weight should differ from that 
 given on the original print or drawing, the templet drawing is checked and, 
 if correct, the customer is notified that the actual weight will not be as 
 specified. No further work may then be done till this question of weight 
 is settled, when the brass templet will be made from the drawing. 
 
 PASS 
 
 HEAD 
 
 WEB 
 
 BASE 
 
 TOTAL 
 
 %RED. 
 
 SPREAD 
 
 FINISHING 
 
 14.4 
 
 5 4 
 
 12.1 
 
 31.9 
 
 5% 
 
 H" 
 
 LEADER 
 
 15.2 
 
 5.7 
 
 12.7 
 
 33.6 
 
 15% 
 
 &" 
 
 3RD FORM. 
 
 17.9 
 
 6.7 
 
 14.9 
 
 39.5 
 
 20% 
 
 A" 
 
 2nd FORM. 
 
 22.4 
 
 8.4 
 
 18.6 
 
 49.4 
 
 24% 
 
 &" 
 
 IST FORM. 
 
 29.5 
 
 11.1 
 
 24.3 
 
 64.9 
 
 22% 
 
 &" 
 
 SLABBER 
 
 37.8 
 
 14 2 
 
 31.2 
 
 83.2 
 
 , Reduction in Passes. 
 
 FIG. 76. Pass Template Drawing Slab-and-Edging 
 Method of Rolling Rails. 
 
 No. 
 
 Pass 
 
 % Red. 
 
 A 
 B 
 
 Finishing Pass 
 Leader 
 
 5. % 
 15. % 
 
 C 
 
 3rd. Forming 
 
 20. % 
 
 D 
 
 2nd. Forming 
 
 24. % 
 
 E 
 
 1st. Forming 
 
 22. % 
 
 F 
 
 Slabber 
 
 23. % 
 
 G 
 
 9th. Pass 
 
 24.5% 
 
 H 
 
 8th. 
 
 24.5 % 
 
 I 
 
 7th. 
 
 22. % 
 
 J 
 
 6th. 
 
 27. % 
 
 K 
 
 5th. 
 
 22.8% 
 
 L 
 
 4th. 
 
 17.7% 
 
 M 
 
 3rd. 
 
 28.8% 
 
 N 
 
 2nd. 
 
 17. % 
 
 
 
 1st. " 
 
 14.5% 
 
 P 
 
 Bloom 
 
 
ROLL DESIGN FOR RAILS 441 
 
 The Hot Templet: The next step, which is really the first step in 
 designing the roll passes, is the making of the hot templet. This templet 
 is exactly like the cold templet, but larger in size to allow for contraction, 
 as it represents the section of the rail at the finishing temperature of rolling. 
 The co-efficient of contraction, or exact amount to allow here, is determined 
 by experience. From this hot templet the various passes are designed 
 successively as the experience and judgment of the designer dictates. 
 
 The Pass Templet: In designing these templets, the designer con- 
 structs each in a drawing showing the different passes superimposed upon 
 each other as in the accompanying illustration. In actual practice these 
 drawings are constructed full size, but for convenience in printing this 
 photograph is three-eighths natural size. This illustration represents the 
 passes for a light rail rolled by the slab-and-edging method. In this method, 
 the axis of symmetry of the rail coincides with the pitch line and is parallel 
 to the train line of the rolls, as can be seen from the print. The more darkly 
 shaded area in the photograph represents the hot templet, with the pitch line 
 or axis of symmetry drawn through it, and from it the grooves in the finishing 
 pass are cut. From this pass the roll designer works back to the bloom. 
 As a preliminary step toward designing these passes, a table like that shown 
 attached to the photograph is prepared. In a vertical column, headed pass, 
 are placed the names of the various passes from slabber to finishing, while 
 the figures in the columns designated as head, web, and base represent 
 the sectional area of the different passes for these parts expressed in pounds 
 per yard, which is directly proportional to the area in square inches. This 
 table is prepared as follows: The areas of the different parts of the rail 
 are fixed by the hot templet. In this section the metal was proportioned 
 in the design so as to give 14.4 pounds per yard in the head, 5.4 pounds 
 per yard in the web and 12.1 pounds per yard in the flange, the total being 
 31.9 pounds per yard, which is heavier than the cold section by 1.4 pounds 
 per yard. This difference is due to the fact that the weight of the hot 
 section is calculated as if it were cold, a correction for difference in gravity 
 not being necessary for this purpose. Next will be put down under the 
 column headed per cent, reduction, the amount of reduction expressed in 
 per cent., which from experience and judgment, the designer thinks will 
 be best. In this case these amounts are as follows: In the finishing pass 
 one-thirty-second of an inch reduction on each side of the web is allowed 
 for the marking; in the leader, 15%; in the third former, 20%; in the second 
 former, 24%; in the first former, 22%; and in the slabber, 23%. From these 
 figures the areas of the different parts of each pass are calculated, and the 
 blanks in the table filled in as shown. Next, the amount to allow for spread 
 of the piece from one pass to the other is decided upon and placed in the 
 column headed spread. This allowance is expressed in fractions of an inch 
 and is measured and allowed for along the axis of symmetry as shown by 
 the positions of the rail heads in the photograph. With the reduction for 
 the various passes and their parts thus apportioned, the designer then pro- 
 
442 
 
 THE ROLLING OF SECTIONS 
 
 ceeds to draw in the passes, as designated in the photograph by the letters 
 a to p, and in doing so he keeps the following points constantly in mind. 
 First, is the danger of forming fins. As an aid in avoiding these defects the 
 
 FIG. 77. Showing Difference in Peripheral Speed of the Bolls at Web, Head, and 
 Flange. 
 
 piece is passed through the mill so that each side of the pass alternately enters 
 an open and closed side of the groove. How this is done on the three-high 
 mill without turning the piece can be seen from a study of Fig. 78. Even 
 
ROLL DESIGN FOR RAILS 443 
 
 with this arrangement fins would still be formed if the passes were not 
 properly designed. To avoid all danger of fins two tricks of design are 
 here resorted to. Thus, in the leader, or pass a, the corner of the head, 
 which is to come opposite the openings between the rolls in the finishing 
 pass, is well rounded off, so that the spread or flow of the metal will be 
 taken up in filling out this rounded corner and none will remain to be forced 
 into the clearance. For the same reason, that half of the flange on the 
 same side of the rail is left much shorter. It will be observed that this 
 provision is made in all the passes down to the slabber. From the photo- 
 graph it will be noticed that most of the work from the slabber is done 
 along the fishing of the rail, the metal being forced horizontally from the 
 central portion of the slab toward the head and flange and vertically into 
 the web. This is done by holding the working surface under the head 
 and on top of the flange at the angles shown. In designing this part of 
 these passes care must be taken to see that the line XY at the bottom 
 of pass d, for example, is not greater than X'Y' at the top of its preceding 
 pass c, as otherwise the piece would be collared when it entered d. Great 
 care is necessary in distributing the reduction of each part to prevent the 
 metal flowing away from parts where it is needed. Thus, for example, if 
 a too great reduction in the web takes place in one pass, it will produce a flow 
 of metal away from the head, causing the latter to be underfilled. The 
 cause for much of the trouble of this kind lies in the different diameters 
 of the pass, which causes a different roll speed for head, webandflange, 
 and hence different rates of elongation. If the elongation at one point due 
 to increased speed of the rolls, is not balanced by elongation produced through 
 compression at the point of less speed, the section will be imperfectly 
 formed, or cracks will result. The accompanying illustration (Fig. 77) will 
 help in understanding this point. In all reduction, it is a good plan to keep 
 the angle of bite well below the limiting angle of 30. In the first roughers 
 the difficulties of design are approximately the same as those of the three- 
 high bloomer, and no further explanation of these passes is required. 
 
 Preparation for the Rolling: After the passes have all been con- 
 structed properly in the drawing, a set of working templets, including both 
 male and female for the cold templet, is made from the drawings. The 
 working templets may be of steel. These templets may number from the 
 slabber down only, as the roil designer strives to keep the roughing passes 
 of such shapes and sizes that the same set may be used for a large number 
 of different sections of the same general design. When completed they go 
 to the tool shop, where they are used as patterns in making a set of tools t 
 for turning the rolls for the section. For the last six passes, that is, number- 
 ing from the slabber to the finishing, inclusive, twenty-two different tools 
 are required for each type and size of rail. After shaping these tools to a 
 little over size, they are tempered and then redressed to exact size before 
 they can be used to turn the rolls. When ready, templets and tools pass to 
 the roll shop, where the work of turning the rolls is done. Here the templets 
 
444 
 
 THE ROLLING OF SECTIONS 
 
 Fia. 78. Arrangement of Rolls and Passes for 
 
RAIL MILLS 
 
 445 
 
 Rolling Heavy Rails by the Slab-and-Edging Method. 
 
446 
 
 THE ROLLING OF SECTIONS 
 
 ire used by the roll turner in cutting the grooves, which must be of the 
 exact size and shape of the templets. At Edgar Thomson the roughing 
 rolls are adamite or sand rolls, more often the latter; the second rougher, 
 or former, are sand rolls; and the finishing are chilled rolls. 
 
 1st Houghing Stand. 
 
 1st Finishing Stand. 
 
 2nd Roughing Stand. 2nd Finishing Stand. 
 
 Fia. 79. Plan Showing Arrangement of Rolls for Diagonal Rolling of Rails from Billets 
 
RAIL MILLS 447 
 
 The Diagonal Method of rolling is represented on the accompanying 
 cut of a light rail mill where this method is employed altogether. It 
 differs from the slabbing method in that the shaping of the rail is begun 
 with the first pass in the roughers and, instead of first compressing the 
 bloom to a smaller size and then forming the section partly through com- 
 pression and partly by spreading, the process is one of compression from 
 beginning to end, as must be evident from observing the position of the 
 piece in passing through the rolls. From a quality standpoint the method 
 is thought to possess some advantage over the slabbing method by virtue 
 of the fact that a greater amount of work is done on the tops of the head 
 and flange, where it is needed. From the operator's standpoint it would 
 seem to have both advantages and disadvantages. As an instance of the 
 former, it is pointed out that the angular grooves make it easier to redress 
 the rolls and tend to give them a greater life, because the cuts to restore 
 the section need not be so deep. To restore a section of the flat grooving, 
 cuts as deep as one-half inch are usually required on smaller sections, while 
 as much as three-fourths inch is needed on the larger ones. The chief 
 disadvantage of using the method lies in the great side thrust of the rolls, 
 which is very undesirable and difficult to provide for. It also requires more 
 roll space than the flat method for the same number of passes. 
 
 The Mills: There are four rail mills at the Edgar Thomson plant, 
 but one of these, the oldest, is used exclusively for rolling sheet bar, billets, 
 etc. Of the mills used for rolling rails, No. 1 mill is the older. It consists 
 of three stands in tandem, each separately driven. The first roughing 
 stand is driven by a 40" x 78" x 60" horizontal vertical compound con- 
 densing engine, and the rolls are twenty-eight inches in diameter. The 
 second stand, driven by a 50" x 78" x 60" tandem compound condensing 
 engine, contains rolls twenty-seven and one-half inches in diameter. The 
 finishing stand, which is two-high, is made up of rolls twenty-five and one- 
 half inches in diameter, and is driven by a 32" x 56" x 48" tandem compound 
 condensing engine. This mill, originally built for a twenty-four inch mill, 
 is evidence of the rapid increase in the size of roll sections and the extremities 
 the mills are put to in order to meet the demands made upon them for 
 heavier rails. The mill now rolls rails from twenty-five pounds to one 
 hundred pounds per yard. The No. 2 mill was completed in the early part 
 of 1916 and is designed to roll rails up to one hundred fifty pounds per yard, 
 which weight, it is hoped, will meet all demands for some years to come. 
 Up to 1918 the largest section rolled was a hundred thirty pound rail. In 
 this mill all the rolls are thirty-two inches in diameter, when new. The 
 pitch of the pinions is twenty-nine inches, which adapts the mill to rolls 
 as small as twenty-eight inches in diameter. Like the No. 1 mill, the 
 No. 2 is made up of three trains in tandem, but the first roughing train 
 contains two stands in order to provide ample roll space for rolling large 
 sections by either the flat or diagonal methods. The bodies of the rolls 
 in the first three stands are sixtv-four inches in length, while in the 
 
448 THE ROLLING OF SECTIONS 
 
 finishing this dimension is reduced to forty inches. The first and second 
 trains are driven by 50" x 78" x 60" tandem compound engines. The fly 
 wheels on these engines weigh 100 tons each, and the speed of the engine 
 is about 60 r. p. m. The third train is driven by a 44" x 74" x 54" 
 horizontal vertical compound engine, which has a speed of 65 r. p. m. 
 The first or roughing stands on each of these two mills are served by lifting 
 tables, and the intermediate stands by tilting tables. The No. 3, an eighteen 
 inch mill, employs 19" x 42", 19" x 38" and 19" x 20" rolls, and rolls rails 
 from twelve pounds to forty pounds per yard. It consists of two trains 
 in tandem, each of two stands; number one and number three stands, both 
 three-high, make up the first train, while number two, a three-high stand, 
 and the two-high finishing stand are in the second train. Each train is 
 independently driven by a 1500 h. p. electric motor. Rails rolled on these 
 mills have distinguishing marks. Thus, heavy rails rolled on the No. 2 
 mill have the sign, 9 , rolled on the web, and light rails rolled on the No. I 
 mill are distinguished by the sign, , similarly located. 
 
 Rolling Heavy Rails: After the rolls have been properly turned 
 they are placed in the housing in their proper positions and carefully lined 
 up. A trial rolling on a short bloom will then be made, and during this 
 rolling the boss roller will watch the piece closely to see that it goes through 
 the mill all right. If the section is a new one, the roll turner and designer 
 will also be present to watch the trials. If the trial piece goes through 
 the mill without causing trouble, a section is sawed, cooled, and examined 
 by the roller and roll designer. In doing this, the piece thus rolled is gauged 
 by means of the male and female templet which the designer has furnished 
 the roller. If this section is found to be correct, the mill is then ready to 
 begin the rolling, which is really the simplest part of the process. The roller 
 watches the mill closely to see that everything is running right, and at 
 frequent intervals will gauge and examine samples of the rails. In addition, 
 he walks down to the cooling bed about every fifteen minutes to examine 
 the rails for any defects that may be caused in the rolling, such as collar 
 marks, underfills, roll marks, overfills, guide marks, cracks or seams. If 
 he finds a defect in rolling, he hastens to make the necessary adjustments 
 to correct the trouble. 
 
 Unavoidable Variations: One of the things that cannot be avoided 
 in the rolling is the wear of the rolls. While it occurs over the entire surface 
 of the groove the parts of the groove subject to fastest wear are those 
 which do the greatest work. Referring to the photograph of the drawing 
 for the pass templets, it may be seen where the greatest wear will take 
 place. This results in a decrease in the fishing of the rail as shown in the 
 following sketch. It is also a difficult matter to keep the base perfectly 
 flat, because the high collar in the finishing passes supporting this part of 
 the rail tends to wear away the edges faster than at the bottom of the groove. 
 
STEPS IN SHAPING RAILS 
 
 449 
 
 The slight overfill that results produces the defect known as the rocking 
 H This wearing is very rapid, and with the mill running steadily, one 
 
 base. 
 
 dressing of the rolls lasts but from twenty-four to thirty-six hours. 
 
 Overfills 
 
 Underfill 
 
 Decrease in the 
 Fishing due to 
 Worn Roll 
 
 Hocking Base 
 FIG. 80. Showing Defects in Rails Due to Wearing of the Rolls. 
 
 The Various Steps in Shaping of Rails: To trace the material from 
 the ingot, the work begins at the forty-eight inch mill previously mentioned, 
 where the very slow speed and relatively great reduction gives more of 
 the pressing and less of the stretching effect of rolling, and is intended to avoid 
 much of the danger of tearing or cracking the ingot. The large fillets used in 
 the grooves keep the corners of the ingot well rounded. This mill, reducing 
 the ingot from 233^" x 23^" to 15^" x 18^" leaves less work than usual 
 to be done on the three-high bloomer which produces a 9}^" x 9^" bloom. 
 From the bloomer the long bloom passes to the shears where the proper 
 discard, which is varied in different specifications, is made, and blooms of 
 different lengths are cut to suit the conditions. Large rails are rolled two 
 
450 THE ROLLING OF SECTIONS 
 
 lengths to the bloom, while lighter ones may run into three lengths to the 
 bloom. Leaving the shears, the blooms travel on roll tables to a distri- 
 buting point, where they are sent to No. 1 and No. 2 rail mill furnaces or 
 to No. 4 billet and sheet bar mill. The furnaces serving the two large 
 rail mills, which extend parallel to each other and are housed in the same 
 building, are arranged in one row at right angles to the mills and in such 
 a manner that the blooms may be charged on one side and drawn on the 
 other which is nearer the mills. From the rolls the piece passes on to the 
 saws, and then to the finishing and inspecting department. 
 
 Cutting: For cutting rails four high speed (1500 r. p. m.) toothed 
 circular saws are provided. The saws are mounted over the delivery table 
 on the free ends of tilting arms, whose axes are concentric with the drive 
 shafts. Belts then connect the saws with their driving shafts, which are 
 electrically propelled. On No. 1 mill all the saws are mounted on one shaft 
 which is driven by one motor, but in No. 2 mill each saw is mounted on a 
 separate carriage and is driven by an individual motor. The tilting arms 
 are electrically controlled so that all the saws may be made to cut simul- 
 taneously. These saws are adjustable to cut different lengths from thirty 
 to sixty feet, though thirty and thirty-three feet are standard lengths. In 
 cutting the rails, proper allowance must be made for shrinkage, which is 
 nearly three-sixteenths inch per foot, or about seven inches for a thirty- 
 three foot, hundred pound rail. The exact amount of the shrinkage depends 
 upon the temperature at which the rail is sawed, hence many railroads 
 specify the amount of shrinkage per rail, and in so doing fix the finishing 
 temperature of rolling. The allowance should be not less than one-fourth 
 inch over or under length specified. Since the rails do not always leave 
 the finishing rolls perfectly straight, it is not always possible to make a 
 square cut, and one-thirty-second inch off-square should be allowed. Rails 
 eighty-five pounds or over are rolled in double lengths, and on blooms on 
 which tests are taken, six to eight feet is allowed for physical test pieces, 
 which are cut from the ends of the piece. Smaller rails are rolled in triple 
 lengths. Great care is required in adjusting the height of the saw blocks 
 in order to avoid kinking or scratching the rail, and to secure a square cut. 
 From the saws, the rails pass under a stamping machine, which marks the 
 heat number and the position of the rail in the ingot, the latter being desig- 
 nated by letters beginning with A at the top of the ingot. At Edgar Thom- 
 son the A cut is discarded on all heavy rails. About sixty feet from the 
 stamper, is located the cambering machine which consists of a set of 
 horizontal rolls with a vertical roll on each side, all in one housing, and set 
 to bend the rail slightly so as to make the top surface of the rail convex 
 from end to end. A scale located near the end of the delivery table is used 
 for checking the weight of the rails as often as desired, before they are sent 
 to the cooling beds. The No. 1 and No. 2 mills, being arranged parallel 
 to each other, deliver their product, to what may be considered a single 
 large cooling bed, where the rails from both mills are slowly moved in one 
 direction, which is toward the finishing room. 
 
FINISHING RAILS 451 
 
 Recording: A complete record is kept of all the steel rolled in the 
 mills from the time it is made in the open hearth or Bessemer converter 
 until it is shipped. This work is done by the recorder, who traces the 
 material through the mills, and is so complete that it is easy to show, not 
 only the kind of steel from which each rail is rolled, but the heat, the number 
 of ingot, and its position in the ingot. At these mills the work of tracing 
 the material is much facilitated by the use of an electric signalling system. 
 
 Finishing and Inspection: The finishing room is arranged to the 
 best advantage possible for handling the rails. In it are located the 
 straightening and drilling machines (fourteen in number) in two rows parallel 
 to each other and to the two mills, and extending in the same direction 
 beginning at the mill side of the cooling beds. Between the row of 
 straighteners and the cooling beds is a system of roll tables, which carry 
 the rails from the beds and distribute them to the different straighteners. 
 Here the rails are marked with a stamp to indicate the individual work- 
 men responsible for this part of. the work. The straightening machines, 
 or gag presses, are provided with a bottom bed, on which the rail is 
 supported at two points from below, and a top block which moves up and 
 down between these two supporting lines with a fixed stroke of such length 
 that the block will not touch the rail by about two inches at its lowest point. 
 The block has a double face, each side of which is inclined toward the 
 center line, where the faces cut each other. This form, combined with the 
 different dimensions of the gag, a rectangular shaped block of steel which 
 is inserted between the face of the block and the rail to be straightened, 
 makes it possible to control the amount of bend that the rail receives and 
 to adapt the machine to the different dimensions of a rail. For the different 
 sizes of rails, the faces of the blocks are made adjustable by means of set- 
 screws and liners, while for different sections different gags must be used. 
 To straighten a rail, one workman is stationed in front of the machine and 
 another at the end. By sighting along the rail, the man at the end locates 
 the crooks in the rail and brings them under the blocks, while the man 
 before the machine, acting under directions from the workman at the end, 
 inserts the gag in such a way that the stroke of the machine will bend the 
 rail enough to straighten it, which requires that the rail be bent beyond 
 its elastic limit in order to give a permanent set. Next, the burrs made 
 by the saws on the ends of the rails are cut off with chisels, and smoothed 
 with a file. The rails are then given a preliminary inspection by an employee 
 of the Company, in order to avoid unnecessary work being done on rails that 
 may be rejected. As the inspection is completed, the rails are moved to the 
 drilling machines, which are arranged in pairs and so spaced that, when one 
 machine has completed the drilling on one end of the rail, it is moved under 
 the other machine which drills the holes in the opposite end. These machines 
 are each provided with three drilling spindles, the middle of which is fixed, 
 and may be made to drill from one to three holes at one time. The rails 
 are then moved sidewise through openings in the building to the inspection 
 
452 THE ROLLING OF SECTIONS 
 
 beds where they are walked, or inspected, both by a company's inspector 
 and by the customer's, if so specified. The outside inspection covers mainly 
 surface defects, such as seams, slivers, guide marks, and pits that may 
 have been overlooked by the inside inspector, and the bolt holes. The 
 rails, having been measured and gauged by the inside inspectors and rollers, 
 the dimensions are checked only at intervals by the outside inspector. 
 The location of the defects are marked with chalk. If these are located 
 near the end, that portion of the rail may be sawed off and the rail still 
 applied on the order as a short of first grade. Rails that fall below the 
 allowance as to length are also disposed of in the same way. If the defects 
 are many or near the center, the rail is either classed as a number two or 
 sent back to the mills to be rolled into a light rail. In both the inside and 
 outside inspection, the rails are walked twice, once with the base up, again 
 with the heads up. As the rails are accepted by the inspectors, -they are 
 counted, the number being checked up with the original order. Then 
 they are picked up by immense magnets attached to over head electric cranes 
 and placed in the cars where, after weighing, they are ready for shipment. 
 
 Light Rails : The rolling, finishing and inspection of light rails, as 
 well as the material used, are somewhat different from the same operations 
 for heavy rails. To begin with, the number three mill in which most of 
 these rails are rolled, is operated as a separate unit, practically independent 
 of the rest of the works, and with the exception of re-rollings from rejected 
 heavy rails, all its product is rolled from billets which are obtained from 
 other works. For this reason neither check analyses nor physical tests are 
 made on light rails, because, even if such tests were made, there would be 
 no way of indentifying the rail as having been made from the same steel from 
 which the tests were taken nor of knowing that the tests represented the 
 material in an order. The billets are heated in one of two continuous gas 
 fired furnaces and are rolled by the angular method on six passes in the mill. 
 Rerolled rails receive two additional passes in the first roughing stand, 
 which is provided with tilting tables for the purpose. Leaving the rolls, 
 the rails are sawed into lengths of thirty feet or under and are passed to the 
 cooling beds. When sufficiently cold they receive a preliminary inspection, 
 in which they are measured for length, and then passed through a roll 
 straightener to punching machines where both bolt holes and bond holes 
 are punched. Since the roll straighteners straighten the rail in one direction 
 only and usually fail to produce a perfectly straight rail, even in one 
 direction, gag presses are employed to complete the straightening. As for 
 heavy rails, light rails are subject to a second, though less rigid, inspection 
 after all work is completed. The mill is also equipped with cold saws, 
 but cold sawing is undesirable, for it works a great hardship upon the mill, 
 increasing the cost greatly, due to extra labor and scrap loss. For handling 
 the rails a large gantry crane of an improved form which travels on tracks 
 laid on the ground is provided. Light rails are weighed in cars before 
 shipment. 
 
RAIL JOINTS 
 
 453 
 
 Continuous Rail Joint. 
 
 Weber Rail Joint. 
 
 Duquesne Rail Joint. 
 
 100 Per Cent Rail Joint. 
 
 Duquesne and 100 Per Cent Joint. 
 
 Hatfield Rail Joint. 
 
 Wolhaupter Rail Joint. 
 
 Barschall Rail Joint. 
 
 Q & C Bonzano Rail Joint. 
 
 Abbott Rail Joint. 
 
 Williams Reinforced Rail Joint. Atlas Rail Joint 
 
 Fia. 81. Sketches Showing Different Kinds and Types of Rail Joints. 
 
454 
 
 THE ROLLING OF SECTIONS 
 
 Continuous Insulated Rail Joints 
 
 Keystone Insulated Rail Joint 
 
 Weber Insulated Rail Joint 
 
 Braddock Insulated Rail Joint 
 
 O'Brien Insulated Rail Joint 
 
 FIG. 82. Types of Insulating Rail Joints. 
 The heavy black lines represent insulating fiber. 
 
 SECTION II. 
 
 THE SHAPING OF RAIL JOINTS. 
 
 Rolling Rail Joints: Rail joints are made in so many different forms, 
 it is impossible to select any one that would serve as an example to illustrate 
 the problems involved in the rolling of the others. The accompanying 
 prints show the shapes of the different passes for each of three types of 
 rail joints, namely, the common splice bar, the Duquesne joint of the 
 depending flange type, and the continuous joint of the bed plate type. The 
 Braddock insulated joint is made up of two side plates and a bed plate, 
 both being comparatively simple to roll. In a general way, rail joints are 
 more or less difficult to roll, being subject to all the drawbacks of the rail 
 section and to many others in addition, due to their irregular section and 
 lack of symmetry. In the common splice bar, for example, the angles at 
 which the section is rolled are limited by danger of undercuts, and the 
 shape of the passes in which the piece is necessarily reduced are favorable 
 to the formation of laps and seams. In the Duquesne joint, these dangers 
 
RAIL JOINTS 
 
 455 
 
 FIG. 83. Passes for Rolling Common Splice Bar. 
 
456 
 
 . THE ROLLING OF SECTIONS 
 
 FIG. 84. Passes for Rolling Duquesne Splice Bar. 
 
RAIL JOINTS 
 
458 THE ROLLING OF SECTIONS 
 
 are multiplied, while the outstanding parts of the continuous section, by 
 striking the rolls first or being a trifle colder than the rest, often prevent 
 the piece from entering the pass rightly. For similar reasons, it is difficult 
 to make guides that will properly handle this material, and it is prone to 
 become cobbled or caught in the rolls of the tables. The base plate 
 part of the continuous joint must be rolled at the angle as shown on the 
 drawings. This part is later bent up hot at the splice bar shop to fit neatly 
 the flange of the rail. After being rolled, usually in pieces about ninety- 
 three feet long, the bars are hot sawed into three equal lengths, and sent 
 to the splice bar shop, which is located at the Edgar Thomson Works. 
 Here they are sheared to lengths required, punched, slotted and straightened 
 by methods shortly to be described. 
 
 Methods of Finishing Rail Joints: While the finishing of rail joints 
 bears no relation to the rolling of them and is a separate industry, yet for 
 the convenience of the reader it is best to complete the subject now, rather 
 than to postpone it for some other part of the book. There are four ways 
 by which rail joints may be worked: First, all the operations of shearing 
 to length, straightening, punching and notching may be performed upon 
 the cold pieces without heating in any way, when they are spoken of as 
 cold worked joints. Second, this cold working may be followed by an 
 annealing process to produce the cold worked and annealed bars. Third, 
 the bars may be heated, after shearing to length, and the work of punching, 
 etc., be done while they are hot, after which they are allowed to cool in 
 air. In this case they are called hot worked bars. Fourth, instead of 
 cooling the bars in the air after hot working they may be cooled by immers- 
 ing them in oil, when they are designated as hot worked and oil quenched 
 bars. It will be observed that all bars, no matter by what method they 
 are to be worked, are sheared cold, hot or cold sawing being too expensive 
 to be considered. Of course, these methods of treatment may be varied 
 somewhat, but it is doubtful if the additional benefits derived are com- 
 mensurate with the additional expense involved. 
 
 The Edgar Thomson Splice Bar Shop: The practice at this shop 
 coincides with that outlined above. For this reason the shop is made 
 up of four units designated as A, B, C, and D, of which all may be used for cold 
 working, but only A, B, and C are equipped with furnaces for hot working. 
 The furnace of unit B is used for heating continuous joints prior to bending the 
 depending flange. Units A and D each consist of a shear and two presses 
 for punching. A gag press for straightening such bars as need it is also 
 provided. Unit B, which is especially equipped for working continuous 
 joints, consists, in addition to' the furnace noted above, of one shear, two 
 punches, and a folding press, which not only folds the joint but also 
 straightens down the flange forming the bed plate, all in the one operation. 
 The Duquesne bar likewise requires a special tool for cutting out the excess 
 in the depending flange. For this reason, unit C consists of two punches, 
 a straightening press, and two shears. A continuous oil quenching tank 
 
FINISHING RAIL JOINTS 459 
 
 and two large annealing furnaces complete the main equipment of the shop 
 proper, while, in addition, a section of the shop is given to assembling 
 insulated joints. A machine shop for making the dies for punching and 
 shearing is housed in a building adjoining the working shop. 
 
 Cold Worked Bars : In this method the order of working is this : First , 
 the bars are sheared to length, inspected for straightness and flaws, 
 straightened, if necessary, punched and slotted. The dies on the shears 
 are, in all cases, made to conform exactly to the shape of the bars, so that 
 a nice clean cut is made without in any way deforming the bar. However, 
 in shearing the continuous joint, a slight bending of the outer edge at one 
 end is unavoidable. For a" similar reason, the bottom blocks, or dies, of the 
 punching and notching machine support the web while the punches descend 
 from ' above, pushing the material through conforming openings in the 
 blocks. All dies are made of the highest grade of special tool steels and 
 are kept in the best possible condition. In this way the hole is made as 
 smooth as it is possible to make it by punching. It is evident that it is a 
 great advantage to the shop to have the bolt holes on each bar alternately 
 round and oval rather than all oval or all round, for in the latter case only 
 every other bar could be punched on one arrangement of the dies, and the 
 bars would require careful matching. As to the effect of cold working, it 
 is quite clear that the bars of low carbon content, under .28% carbon, soon 
 recover from the effects of the working, but in the higher carbon bars these 
 effects are permanent and may do injury to the bar. On the entering side 
 of the punches the metal is compressed beyond its ultimate strength, while 
 the material on the opposite side is put under a tension, as may be observed 
 by an examination of any hole made by punching. One of the direct results 
 of this cold working is to increase the hardness of the metal about the hole, 
 but the worse effect, which applies only to high carbon bars, is found in 
 the very small cracks which extend into the metal along lines perpendicular 
 to the surface of the hole. For these reasons, cold working is the most 
 objectionable of all the methods cited. The difficulties, as well as the evil 
 effects of cold working, increase as the carbon content of the steel increases. 
 Hence, only low carbon bars (below .28% C.) should be cold worked. As 
 a rule the method is applied to the smaller angle bars and fish plates. 
 
 Cold Worked and Annealed Bars: The fine cracks due to cold working 
 cannot be eradicated by subsequent treatment, but the internal stresses 
 and strains are relieved by annealing and the bar as a whole is made more 
 ductile. Hence, some specifications, more particularly on angle bars and 
 insulated joints, will call for cold working to be followed by annealing. 
 The annealing furnaces provided at this shop are divided into two sections, 
 a heating chamber, which is fired with natural gas, and a cooling chamber. 
 The bars to be annealed are piled, crib fashion, upon steel supports that 
 rest on brick bottomed cars, which are pushed into the furnace. The 
 furnace is then heated to a temperature slightly above the critical point of 
 
460 'THE ROLLING OF SECTIONS 
 
 the steel. The proper temperature once reached, it is maintained until all 
 the steel is thoroughly heated, which generally takes about two and one- 
 half hours. The heat is then turned off and all doors in the furnace are 
 closed. The steel is now pushed into the cooling chamber, where it is 
 allowed to cool slowly. During this time the doors are closed tightly to 
 prevent, as much as possible, scale forming on the steel. The cooling 
 requires about two and one-half hours. After the steel is cold, the cars 
 are pushed out of the furnace, and the bars are loaded on cars for shipment. 
 
 Hot Worked Bars: In hot working common angle bars, the order of 
 procedure is as follows: shearing to length, heating, punching, notching, 
 straightening, if necessary, and cooling. Patented bars may require 
 additional operations. Thus, in working the Duquesne bar, the excess 
 flange is sheared off after the heating and just before the notching. In 
 hot punching any bar, in order to avoid spreading of the metal and con- 
 sequent distortion of the bar, it is necessary to employ a confining die, 
 that is, the cutting die must be enclosed in a die block or frame, the upper 
 surface of which, together with the die itself, conforms in shape to the 
 inside surface of the bar. Hot worked bars must, therefore, be punched 
 from the outside, or inward. Most cold worked bars may be punched from 
 the inside, or outward, as there is no danger of spreading the metal and 
 enclosed dies are not necessary. The straightening machines are presses 
 provided with a set of dies for each size of each section. One die conforms 
 to the size and shape of one side of the section and the second die to the other 
 side, and both are set in the press so that at the end of the stroke the space 
 between the dies is of the same shape as the bar and just equal to it in 
 thickness. As a rule the higher carbon angle bars, Duquesne and con- 
 tinuous joints are hot worked. In case the continuous joints are cold 
 worked, they must be heated before the flange is bent down. The furnaces 
 employed for hot working are of the continuous type. They are rectangular 
 in shape and wide enough to admit two rows of bars laid end to end. Natural , 
 gas is the fuel used. In order to obtain an even distribution of the heat, 
 the furnace is provided with four ports, and in addition four large burners 
 of the Bunsen type are located at the bottom along each side of the furnace 
 to supply heat under the bars. Recording pyrometers are employed, so the 
 exact temperature of the furnace may be ascertained at any time. After 
 being sheared to length, the cold bars are laid upon water cooled skid pipes 
 and pushed into the furnace from the rear by means of electricall} 7 operated 
 dogs. The length of the furnace and rate of charging is such that about 
 two hours are consumed in pushing each bar through the furnace, this 
 time being sufficient to bring the bar to the working temperature of about 
 800 to 830 C., which is a little higher than necessary for working, as the 
 temperature drops by the time the bar reaches the machine to 790 or 815. 
 The skids end near the front of the furnace, and the bars descend to a hearth, 
 whence they are removed with tongs through doors. Needless to say, the 
 bad effects of cold working are entirely avoided by hot working. 
 
STRUCTURAL SHAPES 461 
 
 Hot Worked and Oil Quenched: In this method, not only are the 
 evils of cold working avoided, but the strength and ductility, or toughness, 
 of the bar are much improved, also. The method, which is especially 
 applicable to high carbon angle bars and Duquesne joints, consists in hot 
 working the bars in the usual way and quenching them in oil before they 
 have cooled to a temperature below that of the critical range. The neces- 
 sity of completing the work before the temperature drops below this point 
 may require the bar to be heated to 30 or 40 C. higher than for ordinary 
 hot working, and permits no delay in the operation. For quenching the 
 bars, a continuous oil tank in close proximity to the presses is provided. 
 The tank is rectangular in shape and provided with a chain conveyor, which 
 slowly carries the bars through the oil, the speed of the chain and its 
 direction of travel being so regulated that the bars, upon entering at one 
 end of the tank, are carried down into the oil, across the tank, and up to 
 the opposite end and are cooled to 70 C. or less. The oil used is a special 
 grade of petroleum product that will not get viscous and has the most 
 favorable cooling properties. In order to keep it cooled to the proper 
 temperature, the oil from the quenching tank is pumped through a set of 
 water cooled pipes and into a large storage tank. The fresh oil for the 
 quenching tank is pumped from the bottom of this tank. By means of 
 this circulating system the temperature of the oil is kept, usually, at about 
 60 C. The temperature of the oil is taken at intervals of an hour or so 
 to make sure its temperature does not rise too high. 
 
 SECTION III. 
 
 STRUCTURAL AND OTHER SHAPES. 
 
 
 
 Plan of Study: It is needless to remark that a detailed description 
 of the rolling of each of the many sections included under this heading 
 would result in a very lengthy discussion, and it is doubtful if such a dis- 
 cussion would prove to be of much value in accomplishing the ends at which 
 this book aims. Besides, while the rolling of each shape or section presents 
 difficulties peculiar to itself alone, there are certain problems common to 
 all sections, and of these general features an example has already been 
 given in the description of the rolling of rails. Unlike rail mills, structural 
 shape mills vary much, both in type and equipment, and often the methods 
 for rolling a given section must be adapted to the mill conditions. In 
 general, the size of the section, for economic reasons, will determine the 
 size of the mill, and the different sizes of the same section will be rolled 
 on different mills. No one mill, therefore, can be selected as an example 
 of the rolling of even one of these shapes. For these reasons a brief and 
 more or less general discussion of the different roll designs for some of the 
 more common sections is all that will be attempted here, and it is hoped 
 the study will be found both interesting and profitable. 
 
462 
 
 THE ROLLING OF SECTIONS 
 
 Angles: Among the first shapes to be rolled was the angle. Three 
 methods of rolling this shape have been developed. In two of these methods 
 the forming of the angle is begun from a rectangular bloom, or if the bloom 
 is square, from a rectangular roughing pass in the mill. In what may be 
 termed the first method, the grooves are so designed that each successive 
 pass from the slab approaches the right angle of the finished bar, the piece 
 being gradually bent and reduced at the same time. In the second method, 
 called the butterfly method, the legs are kept flat until the leader and 
 finishing passes, when they are bent to form a right angle. In the third 
 method, the forming of the angle is begun from a rectangular bloom by 
 first working off one corner and then recessing it till the desired thickness 
 of the legs is obtained. 
 
 ...Billet. 
 
 FIG. 86. Methods of Rolling Angles. 
 
 The Three Methods Compared: The part of the angle that gives 
 the most trouble in rolling is the back and the apex, which must be square 
 and sharp. As shown in the accompanying sketches, this difficulty is 
 overcome in the first and second methods by reserving metal in the first 
 passes on both sides of the bar where the apex is to be formed. In the 
 last passes this excess metal is available to fill out what would be lacking 
 on account of the bending, which tends to draw the metal down from the 
 apex. In the third method the apex and back are perfect from the beginning. 
 As to the relative merits of the three methods, there is, of course, much 
 difference of opinion. However, there are two features about the butterfly 
 method that appears to a decided advantage when compared with either 
 of the other two. The ability to employ an edging pass so near the finishing 
 
STRUCTURAL SHAPES 
 
 463 
 
 to control the width makes it possible to reduce the piece rapidly in the 
 roughing passes and so cut down the total number of passes required to 
 form the section. Then, too, since nearly all the work on the section is 
 done in the flat passes, the difficulties encountered when deep grooves are 
 used in the rolls are entirely avoided. Usually angles are formed in from 
 nine to eleven passes. 
 
 Butterfly Method for Channels as first designed 
 
 FIG. 87. Methods of Rolling Channels. 
 
464 THE ROLLING OF SECTIONS 
 
 The Channel: Channels are rolled by two distinct methods known 
 as the butterfly and beam roughing methods. The butterfly method is 
 said to have originated in the year 1873 at the Upper Union Mills of Carnegie, 
 Klowman & Co., and is sometimes called the slabbing method. In being 
 formed by this method, the section resembles two angles being rolled side 
 by side in one set of grooves, and by the same butterfly method. In the 
 second method the bloom, in the rectangular form, is edged for the first 
 pass and is then worked down from each edge or face alternately by grooves 
 in the roughing passes until it much resembles a beam, and in reality it is 
 a beam in the rough. Succeeding passes, however, work off the flanges on 
 one side of the web. The function of these temporary flanges is to supply 
 metal for holding the height of the flanges on the opposite side, thus forming 
 a channel with full sharp edges and square sides. This method is said to 
 have an advantage over the butterfly method in that the roughing rolls 
 may be used either for beams or channels and a greater number of weights 
 may be taken from the same set of rolls. In rolling deep channels the 
 butterfly method would appear to overcome the difficulty of making the 
 flanges nicely, but more roll space is required than in the beam roughing 
 method, and since the great width of the section makes edging imprac- 
 ticable, it is more difficult to secure well formed edges on large channels. 
 In the beam method, the butterfly idea is used to some extent, and in order 
 to obtain the proper height and thickness of flange readily, the flanges are 
 rolled at an angle to the web and finally bent to right angles in the leader 
 and finishing pass in the same way, though to less degree, as in the butter- 
 fly method. For channels, about the same number of passes are required 
 as for angles. Very large channels are often rolled from shaped blooms, 
 as at the thirty-five inch mill at Homestead, which works in conjunction 
 with the forty inch blooming mill to roll large beams and channels. The 
 blooms for both these shapes have much the same form, the channels being 
 finished by the beam-rougher method in the thirty-five inch mill. 
 
 Beams, Ties, and Piling: Beams were, doubtless, first made by 
 riveting a plate and four angle bars together, then later by bolting or 
 riveting two channels together back to back. Up until 1895 they were 
 considered very difficult sections to roll, but the great demand for these 
 shapes during the succeeding years so stimulated thought in their manu- 
 facture that most of the former difficulties have been overcome, and standard 
 beams are now rolled with as little trouble as angles, channels, or rails. 
 Indeed, the rolling of these sections very much resembles that of rails, 
 for if the head of the rail be replaced by a flange the sections would be 
 practically the same. Like rails, there are two methods of rolling beams, 
 namely, the flat, or slab-and-edging, method and the diagonal. But un- 
 like rails, the diagonal method for beams is far superior to the older flat 
 method, because the oblique design of the passes makes it possible to secure a 
 much greater length of flange than was ever produced by the first method. 
 This advantage is forcibly illustrated in the steel tie section, the rolling of 
 
STRUCTURAL SHAPES 
 
 465 
 
 which was first successfully accomplished at Homestead. This section with 
 its very thin flange, which has almost no taper, would have been looked upon, 
 
 FIG. 88. Methods of Rolling Beams. 
 
 FIG. 89. Methods of Rolling Piling. 
 
 fifteen years ago as an impossibility from a rolling standpoint, and is con- 
 sidered one of the greatest achievements in roll design. U. S. steel piling 
 
466 THE ROLLING OF SECTIONS 
 
 is another section which much resembles the rail in rolling. Here, the ball 
 and web of the piling, as finished, is almost a duplicate of the head and web 
 of the rail, and the remainder is rolled as a flange up to the leading and 
 finishing passes, when the two halves or legs are bent down to form the 
 socket for the interlock. It is rolled either by the flat or diagonal methods. 
 
 Zees and Tees : Zees may be looked upon as double angles or channels 
 with reversed legs, or flanges, and the methods of rolling correspond to the 
 first two methods for angles. As to tees, it is doubtful if any other section 
 offers as little opportunity for variation. There is only one way by which 
 the tee can be rolled. In this method the shaping begins from a square bil- 
 let or bloom. One side of the square is retained to form the base, or table, 
 of the tee, while the edges of the opposite side are both recessed in the 
 first forming pass by collars on each side of a groove into which part of 
 the metal flows to start the stem. The piece is then edged for the next 
 pass, in which the stem is reduced between the flat surfaces of the rolls, 
 while the two parts of the table pass through grooves in the two rolls. 
 In the next pass the piece is turned with the stem up, which will pass 
 through an idle groove, while the table on each side of the stem will be 
 reduced between the plain surfaces of the rolls. This process is then 
 repeated, the piece being worked alternately on the stem and table, until 
 the section reaches the size desired in the finishing pass. Usually, in this 
 last pass the stem is in the groove of the lower roll and the table is 
 reduced between the rolls. In order to prevent the bar from following the 
 roll on the delivery side, this groove, as for all the idle grooves, must taper 
 slightly from the top and be large enough to give easy passage for the stem, 
 thus making it somewhat wider than the stem is thick. The reduction of 
 the table then results in the formation of a slight overfill at the base of the 
 stem. This bit of excess metal cannot be removed, because the stem was 
 necessarily finished in the preceding, or leading, pass. If the section were 
 finished by working on the stem, the same defect would be developed. 
 
 Finishing Sections: Most shape mills are provided with hot saws 
 located after the finishing pass and near the cooling beds. These saws are 
 intended mainly for cutting test pieces, but in some mills they are used 
 for cutting off crop ends or dividing mill lengths where the cooling beds 
 are too short to take full mill lengths. The test pieces are of two kinds, 
 namely, those for the roller, whose duty it is to seethat the section is rolled 
 to the correct dimensions and weight, and those for the physical laboratories. 
 With these exceptions, however, the full mill length is sent directly to the 
 cooling bed. Here, as in the case of rails, the cooling causes the shapes 
 to bend considerably, and cold straightening is necessary. For this purpose, 
 the piece is next passed through a cold roll straightening machine, and if 
 necessary through a gag press. The roll straightener is capable of 
 straightening in one direction only, so that if, through handling or other 
 cause, the piece has been twisted or bent laterally, it can be made straight, 
 
ROUNDS 
 
 467 
 
 in all directions, only by using a gag press after the roll straightener. From 
 the straighteners the material passes the cutting machines, which, in order 
 to keep up with the mill, must be either shears or cold saws. For this 
 reason, exact cutting to length is out of the question, and a liberal cutting 
 tolerance is always desired by the mill. As a rule, angles, zees and small 
 tees are sheared, while channels, beams and large tees are cut with cold 
 saws. During the cutting, any material that contains the more noticeable 
 defects is discarded and cut into scrap lengths, and after the cutting the 
 material is subjected to a very rigid inspection for the most minute surface 
 defects. 
 
 Rounds: In the steel business, rounds are often designated as hand 
 rounds or guide rounds, and these terms are indicative of the two methods 
 for rolling these shapes. In the first method, the bloom or billet is first 
 reduced in diamond shaped roughing passes to the form of a round cornered 
 square. Then, by means of tongs in the hands of the workmen, this square 
 is supported in the correct positions and passed through a single round- 
 forming, oval-shaped pass, or two similar round-forming passes, until it has 
 been worked into the round form, which can be accomplished by turning 
 
 Billet 
 
 Bounds 
 
 FIG. 90. Rolling a Half Inch Guide Round. 
 
 the piece through an angle of 90 after each pass through the rolls. From 
 three to five passes in the finishing stand are required to form the round 
 by this method of rolling. In the case of guide rounds the size of the billet 
 may be reduced in the roughing passes in a manner similar to that for hand 
 rounds, or by any other method that will give a round cornered square. 
 This square is then reduced in one or two passes to an oval, which is then 
 edged, and while supported in this position by a guide, it is put through 
 a round finishing pass. The compression in this pass shortens the long 
 axis, while the spreading of the metal lengthens the short axis of the oval 
 to equal the radii of the round. In a general way, large rounds, that is, 
 those over two inches in diameter, are rolled by hand, while small rounds, 
 
THE ROLLING OF SECTIONS 
 
 less than two inches in diameter, are rolled with guides, but there is a 
 narrow range from about one and three-fourths inches to two and one-half 
 inches where either method may be employed. Regarding the relative 
 merits of these two methods, hand rounds are by many preferred to guide 
 rounds where great accuracy and uniformity in diameter are required. 
 However, guide rounds are now rolled with a high degree of accuracy, and 
 since the uniformity and accuracy of the hand rounds depends on the skill 
 of the workmen, it is doubtful whether equal care and attention applied 
 to both methods would leave, on the average, much wherewith to choose 
 between them. 
 
 Cutting and Straightening Rounds: In order to keep up with the 
 mills, rounds are either hot sawed or cold sheared to convenient lengths, 
 and neither method is at all exact. Hence, in these lengths, either single 
 or multiples of those desired, proper allowance must be made for the exact 
 cutting, which is performed by special cutting machines after the round is 
 straightened. The straightening may be done either on the gag press or 
 on special straightening machines, called from the inventors the Brightman 
 and Abramsen straighteners. The Brightman consists of two rows, or sets, 
 of concave rolls mounted upon opposite sides of a revolving frame, so that, 
 with their axis of rotation at an angle to that of the frame, the concave 
 surfaces of opposite rolls bear on the round and grip it in such a manner as 
 to force the bar along longitudinally and at the same time bend it at the 
 crooked places enough to straighten it. In the other type of machine the 
 rolls are mounted on a stationary frame, while the piece itself is revolved 
 and forced through it. The straightening develops one serious defect, 
 which renders the bar unsuitable for some purposes. The scale in the spiral 
 path of the rolls is rolled into the surface, causing a slight pitting, which 
 can be removed only by machining. 
 
 Flats : Because they are the simplest, the flats were the first sections 
 to be rolled. The main problem involved in designing the rolls for flats is 
 the control of the width, which is done in two ways after the piece has left 
 the roughing rolls. The first method, called the flat and edging, consists 
 merely of rolling the piece on edge at intervals in deep grooves cut in the 
 roils. The other method of controlling the width lies in the use of the 
 tongue and groove passes as described under the rolling of sheet bars, and 
 is best suited to the rolling of thin material. In this method the last two 
 passes will be between plain rolls, while in the flat and edging method the 
 planisher will be an edging pass. It is in this pass that the three different 
 edges on flats are formed. Thus, if the bottom of the groove is flat, a 
 common swell or oval edge will be formed by the spreading of the material 
 in the center, which is hotter than the outside; but if the base of this groove 
 is made sufficiently concave, the edge of the bar will be round; if convex, a 
 square edge will be formed. In the eighteen inch mill at Clairton and also 
 in the fifteen inch mill at Lower Union City Mills there is used, next to the 
 finishing pass, a set of vertical rolls, which eliminates the difficulty of 
 rolling a wide thin flat in an ordinary groove. 
 
HEXAGONS AND DEFORMED BARS 469 
 
 Hexagons : There are two methods used for rolling hexagons. By one 
 method all six corners are formed in the rolls, three in the top and three in 
 the bottom. The clearance between the rolls in this case comes on opposite 
 flat surfaces of the bar. In this manner of rolling, the corners of the hexagon 
 cannot pinch out. Hexagons rolled thus are best suited for cold drawing 
 purposes as they will be free from pinches which draw out into laps. In this 
 method, the first pass in the strand, or first former, is a square which has 
 been broken down from the billet in the roughing and pony roughing stands. 
 The square is then put into the second strand, or former, and comes out 
 in the form of a six sided flat, the two widest sides of which are convex. 
 The bar is then edged into the leading or planishing pass where a reduction 
 of about 25% takes place, in the case of small hexagons. The bar coming 
 out of tins pass has six sides as before but has two of its corners formed 
 by the clearance of the rolls. The top and bottom sides are slightly concave 
 to allow for the spread as the bar is edged into the finishing. Here the 
 bar is given a light draft, in order to square it up, the reduction being only 
 8 to 10%. In the other method two of the corners of the finished bar are 
 formed at the clearance of the rolls, and hence care must be taken to keep 
 these corners from pinching out. The bar coming out of the planishing in 
 this method, is in the same position as the finished bar of the first method. 
 The bar is turned 90 and entered into the finishing so that its top and 
 bottom surface are flat or horizontal and two of its corners are in the 
 clearance between the rolls. In the one method but three passes are 
 required to form the finished bar from the square while the other method 
 requires four. 
 
 i 
 
 Deformed Bars: This term is meant to include any bar having an 
 irregular surface or a surface on which there are projections or depressions, 
 such as the various concrete reinforcing bars, clip iron, hame strap, etc. 
 Some of these bars are so complex as to excite the highest admiration and 
 astonishment from those not familiar with their manufacture. While they 
 require the greatest ingenuity on the part of the roll designer, they are, 
 however, produced with less difficulty than might be supposed. Briefly, 
 the secret of their formation consists in first working the metal down through 
 the ordinary passes to one of the common forms, such as a flat, square, 
 round, or oval, whichever is best for forming the bar desired, and then 
 putting it through one or two deforming grooves containing the necessary 
 recesses or elevations. It is here the chief trouble occurs. As the deforma- 
 tions must be made in the finishing pass, or the leader and finishing com- 
 bined, a heavy draught in these passes is necessary, and the metal is usually, 
 even with the fastest working, at a very low temperature for working, 
 These conditions put a heavy strain on the mill, and, owing to the lack 
 of plasticity in the metal, the projections will not fill out easily. In order 
 to avoid vibrations which cause the bar to slip in passing through the rolls, 
 thus causing the deformations to be made at irregular intervals, mills 
 rolling these sections are provided with separately driven finishing stands 
 
470 THE ROLLING OF STEEL 
 
 CHAPTER IX. 
 
 THE ROLLING OF STRIP AND MERCHANT MILL PRODUCTS 
 
 SECTION I. 
 
 STRIP, OR HOOP, MILLS AND THEIR PRODUCTS. 
 
 Meaning of the Word Hoop: As originally applied, the word hoop 
 meant that light narrow material which, cut into short lengths, was used 
 to bind casks, barrels, buckets, and the like, but as now employed the 
 word is a class name that stands for a large number of products. The 
 Carnegie Steel Company, for instance, uses the term to cover all materials 
 from 13 gauge to the thinnest material rolled on their mills, and from three- 
 eighths inch to eight and five-eighths inches in width. This range covers 
 material used for a great variety of purposes in addition to hoop, such as 
 skelp for tubes, blades for knives, and blanks for stamping hundreds of 
 hardware specialties, a class of material that should represent a little better 
 grade than ordinary hoop. In a way, this use of the word hoop is unfortu- 
 nate, especially as a more suitable class name is supplied by the term strip. 
 With strip for a class name, hoop would have retained its original meaning, 
 for which there is no substitute, and all danger of ambiguity would have 
 been avoided. 
 
 Hoop as a Rolling Specialty: Being, perhaps, the largest pro- 
 ducer of strip in the country, the Carnegie Steel Company's, mills furnish 
 the best example of the equipment and organization required to roll this 
 class of material. A specialty mill is not a mill that rolls a specialty 
 nor a variety of specialties, but one that has specialized in the rolling of 
 a single product. According to this definition, and using the term hoop 
 in its broad sense as outlined in the preceding paragraph, the hoop mills 
 of this company are best described as specialized specialty mills, because 
 they are laid down, not for the general purpose of rolling hoop or strip, but 
 in such a manner that each mill is designed and equipped to roll a certain 
 kind or grade of hoop. The advantage of such a system is at once evident, 
 making it possible for the mills to meet the many demands of the trade 
 most readily. Thus, they are able to give accuracy, where accuracy is 
 required; finish, where finish is desired; quantity, where quantity is the 
 main consideration; and all with a better quality to the customer and at 
 a greater saving to the producer than would be possible in any other way. 
 
 The Carnegie Hoop Mills: From what has just been said, it will 
 readily be surmised that the hoop mills of this company are many in number 
 and of various types, and no detailed description of all of them can be 
 
HOT STRIP OR HOOP 471 
 
 expected. This excuse will be better appreciated when it is known that 
 their hoop mills number no less than twenty-five, and among these at least 
 six types, or designs, of mill are represented. However, these mills present 
 certain features of a general nature, for the omission of which no valid 
 excuse can be found. In size their hoop mills range from seven to twelve 
 inches, the most common size being eight and ten inches. In some of the 
 mills the different stands of rolls will vary in size. Thus in the McCutcheon 
 No. 6 Mill, eleven inch rolls are employed in the roughing stands, nine 
 inch in the intermediates, and eight inch in the planisher and finisher. 
 The rolls may be of different sizes in the same stand, also, as will be explained 
 later. As to the materials of which the rolls are made, the roughing rolls 
 may be of steel and the intermediates of sand or adamite, but the finishing 
 rolls are always of the hardest chilled iron. The number of stands in these 
 milts vary from eight to twelve per mill, and their arrangement is such as 
 to produce the kind of hoop desired to best advantage, as will be more 
 clearly explained later. 
 
 Methods of Rolling Hoop : While hoop of the heavier and medium gauges 
 and narrower widths is successfully rolled by the flat and edging method, 
 it is readily understood that this method is not applicable to all sizes of 
 hoop. After the material leaves the roughing rolls there is but one way 
 remaining, then, by which it can be reduced, and that is by the tongue and 
 groove method already described. These tongue and groove passes will be 
 found to be three or four in number, though only two are used in some of 
 th6 mills, and to be followed by two or three stands of plain rolls. As to 
 the manner of breaking down in the roughing stands, there are two methods 
 of rolling one in which the billet is reduced by flat and edging passes and 
 another in which the roughing passes are ovals and squares. The latter 
 method gives a more rapid reduction, and will be employed usually where 
 only two roughing stands precede the tongue and groove rolls. 
 
 Precautions Required in Rolling Hoop: The chief difficulty 
 encountered in rolling this light strip lies in getting the hoop through the 
 rolls before it becomes too cold to roll, or in keeping the temperature 
 uniform at the finishing roll, which is necessary to produce a strip uniform 
 in gauge. Considering the rapidity with which the metal is cooled on 
 account of the thin section and the chilling effect of the cold rolls, the 
 accuracy as to gauge and width maintained by the hoop mills is little 
 short of astonishing. In this case the ingenuity of the engineer who builds 
 the mill and that of the roll turner who designs the rolls are both required, 
 in, addition to the skill of the rollers, to produce the desired result. By 
 reducing the number of passes to a minimum in the ways already described, 
 the roll designer has done his bit, but even then the last end of a long strip 
 will be cooled to a temperature so much lower than the first that it will 
 finish several thousandths of an inch thicker, unless the mill is adapted to 
 rolling it. These drawbacks are overcome by a combination of ingenious 
 
472 THE ROLLING OF STEEL 
 
 schemes. First, the mill is run at as high a speed as practicable. Second, 
 it is arranged and operated so that as little time as possible intervenes 
 between passes. Thus, in the more modern hoop mills, roughing rolls on 
 the continuous plan are used, and these are placed very close to the furnace, 
 whence the billet, being at once reduced in the two or three roughing passes, 
 passes over roll tables to the tongue and groove rolls, which are arranged 
 on the continuous plan or in train. If the latter plan is followed, mechanical 
 repeaters are employed for short lengths, or the piece is looped from pass 
 to pass by hand, by which means it may be rolling in two or more stands 
 at one time. In order to avoid having the loop, which increases with the 
 length of the piece, run far out on the floor, and thus become chilled, each 
 pass following the tongue and groove passes is made to run as much faster 
 as its predecessor as is necessary to take up the slack due to the elongation. 
 Other schemes to equalize the finishing temperature are also employed. 
 It is said that at some hoop mills the end of the billet which represents 
 the last end of the hoop to pass through the rolls is heated to a 
 higher temperature than the first end. In the cotton tie mill at Youngs- 
 town, where the strips are about 1800 feet in length, one end of the billet 
 will be in the furnace, while the other will be coiling some 200 feet away 
 at the other end of the mill as finished strip. Other factors affecting the 
 uniformity of gauge, or thickness, are the wear of the rolls and the bearings. 
 To overcome the wear of the former, which also affects the finish, the 
 path of the piece through the leading and finishing passes is moved over 
 to a new surface about every twenty minutes, or whenever these surfaces 
 become too rough. The vibration of the mill pinions produces waves 'or 
 slip marks on the surface, if the last stands are in train with the tongue 
 and groove rolls, hence the planishing and finishing rolls are generally 
 separately driven in mills rolling the best grades of hoop, and only one 
 of the rolls is driven, the other revolving by friction due to contact 
 with the driven roll. The planisher may. be in train with the tongue and 
 groove rolls, but the finishing stand is almost always separately driven. 
 At some of the mills where the finisher or planisher, or both, is in train 
 with the rest of the stands, the bottom roll is much larger in diameter 
 than the top roll or the rest of the mill, and is driven, while the top roll 
 revolves by friction. The larger diameter of the bottom roll increases the 
 speed of delivery of the stand. The finish on some hoop is of great im- 
 portance. To provide for this requirement scrapers are employed in front 
 of the chill, or finishing, rolls. These scrapers consist of two horizontal 
 bars spaced about eight inches apart and fixed parallel to and just in front 
 of the rolls, and of two other bars similarly arranged but fastened to the 
 two prongs of a fork that can be moved up and down by means of a lever. 
 In this way the movable bars can be lowered as desired into the spaces 
 between the fixed bars and the rolls. As soon as the rolls grip the piece 
 the scraper is brought into action, and the piece is bent sharply up and 
 down over its edges, thus cracking the scale and removing it at once. As 
 
HOT STRIP OR HOOP 473 
 
 the piece, after coming out of the finishing pass, is near or below the critical 
 temperature, no more scale is formed, and a smooth bluish surface results. 
 About five or six feet of hoop on the front end is not thus scraped, because 
 the great speed of the piece carries it through this distance before the 
 scraper can be brought into play. In order to eliminate this unfinished 
 end with as little waste as possible the shears are located, preferably, at 
 the mill end of the cooling bed, so that this end of the strip is the last 
 to be cut. The adjustment of the rolls is also an important matter in 
 rolling hoop. To illustrate this point, if one of the finishing rolls is enough 
 out of level to make a difference of only one-ten-thousandth of an inch in 
 the thickness of the hoop on its edges, it will bend toward the heavy side 
 in leaving the pass an amount equal to one foot in thirty feet of length. 
 As the stiffness of an ordinary hoop is only sufficient to push it, even on a 
 very smooth run out, a distance of about thirty feet, some means of cheaply 
 delivering the strip away from the last stand of rolls must be used. In 
 Carnegie mills, this delivery is accomplished in three ways, namely, by 
 hot coilers, by conveyor belts in the runouts, or by pneumatic runouts. 
 In the last named type of conveyor, air blown at high pressure through 
 holes in the bottom of the runout and at angles directed away from the 
 mill, lifts the strip, decreases the friction and helps to carry it along. 
 
 Finishing Hoop: Hoop may be finished in many ways. As to the 
 cutting of hoop, it is always cold sheared, and this may be done so as to 
 give either both ends square or one end square and one end round. For 
 this kind of cutting, a special die with a double edge, one round, the other 
 square, is used, and the shapes on the ends of the hoop are formed by 
 punching out a small part of the hoop. Ordinary hoop, not coiled, is cut 
 on alligator or small guillotine shears, but the cotton tie mill is equipped 
 with two shears of a special type, which consists of two revolving wheels 
 on each of which a shear knife is mounted. By regulating the relative 
 speeds of the wheels, these shear-blades are brought together in every so 
 many revolutions, so that as the strip is fed into the shear at a definite 
 speed, it is cut into approximately equal lengths. Anything near exact 
 cutting is impossible on this machine, due to slipping of the belts and the 
 play in the parts of the machine. Flaring and punching machines are 
 provided at some of the mills, and in this connection it is to be remembered 
 that the flare on a hoop is measured correctly by one-half the difference 
 between the largest and smallest diameters. All three methods of bundling, 
 in strips, in scrolls and in coils, are practiced. As to coils, some of the mills 
 can coil in multiple or single strips, while others can coil only in singles. 
 Hot coiling can be done only at certain mills, also. It is a matter of pride 
 with the men at the mills to turn out the finest product, and any order 
 that calls for extra quality and finish is sure to receive the attention it 
 deserves. On such material special tests, such as the acid test for scale 
 pits and bending-over tests for seams, are often made in addition to the 
 ordinary inspection for surface defects and gauging for thickness and width. 
 
474 
 
 THE ROLLING OF STEEL 
 
 SECTION II. 
 
 MERCHANT MILLS. 
 
 What the Merchant Mill Is: The first mills were what we of to-day 
 would call merchant mills, though the term was perhaps not then applied 
 to them. In the beginning of the industry, all mills doubtless rolled a 
 variety of simple sections such as rounds, squares, flats, etc., but as the 
 business grew and the demand for heavier material and for certain shapes 
 increased, mills designed to meet a given demand or to roll a certain kind 
 of product began to appear. Then it was that the smaller or older mills, 
 which continued to roll a variety of sections and often stocked material 
 that was retailed out later, were designated as merchant mills, in order to 
 distinguish them from the specialty mills and those whose product was 
 handled in large lots. Later on, especially in this country, the mill manner 
 of handling materials and doing business underwent a change, and the mills 
 were more or less divorced from the store house, so that these mills, while 
 they continue to roll a variety of sections, always roll to orders. Thus, 
 though they have lost all the characteristics, they still retain the name of 
 merchant mills. Therefore, in this country a merchant mill is any small 
 mill, say twenty-two inches and under, which regularly produces more than 
 one shape. 
 
 Kinds of Merchant Mills: In touring the various works, the visitor 
 is surprised and often not a little confused by the great number of and 
 seemingly meaningless terms applied to these mills. Thus, there is heard 
 
 the term "barmill" applied to two 
 mills of altogether different types. 
 The term ' 'guide mill' ' is apparently 
 used in the same way. Added to 
 these are such names as ''Morgan 
 mill," loop mill, shape mill, con- 
 tinuous mill, and semi-continuous 
 mill, hoop mill, Belgian mill, com- 
 bination mill, etc., and> such local 
 terms as the "iron mill," and the 
 "steel mill," the "electric mill," 
 or just "merchant mill," as at one 
 of our own plants where there is 
 but one merchant mill. While one 
 
 not familiar with the mills is inclined to think most of these names are 
 accidental localisms, many of them are really descriptive of the mills and 
 also form links in a systematic classification based mainly on construction 
 and design. These terms and the classification of the mills is best 
 explained from a historical viewpoint. For this reason it is desirable to 
 trace very briefly the evolution of the small mill. 
 
 Plan 
 
 Elevation 
 FIG. 91. First Merchant Rolling Mill. 
 
MERCHANT MILLS 
 
 475 
 
 Development of Merchant Mills: The first mill, which also stands 
 for the simplest kind of mill, consisted of a single stand of rolls driven in 
 one direction only, and for many years all bars or sections were rolled on 
 this simple mill. No guides were used at first, and the roller guided and 
 
 supported the bar between the rolls 
 by means of tongs. In order to 
 avoid the labor of pulling the bar 
 around the mill, the catcher re- 
 turned it by laying one end on the 
 top roll, which carried the piece 
 forward with little effort on the 
 catcher's part. To avoid this idle 
 pass, the idea of placing a third 
 roll above the second, so as to 
 work the bar as it passed in both 
 directions, was conceived and re- 
 sulted in a great economy in labor. 
 Up to this point the bars were com- 
 paratively short, sixteen to twenty 
 feet being the usual lengths. It 
 was next discovered that a great 
 saving in both labor and material 
 could be effected by making the 
 Pia. 92. Diagram of a Simple Type of bars longer, but to accomplish this 
 Three-High Merchant Mill. increase, a larger billet had to be 
 
 used. The increase in the size of the billet was secured by increasing its 
 length, which required wide heating furnaces, and by increasing the size of 
 its cross section, which called for a greater number of passes for reducing 
 than could be placed in one stand. To supply these extra passes, addi- 
 tional roll stands were needed. These stands were coupled together to 
 form the roll train of Figure 92. 
 
 The Guide Mill: The rolling of rounds, which have always been an 
 important product, led to the first use of guides. The method first used 
 was like the method for hand rounds previously referred to, but a greater 
 number of passes were used to finish the piece. Since the tonnage, 
 especially on small rounds, was held down by the time consumed in passing 
 the piece back and forth through the finishing groove, it was natural that 
 the mill man should seek some way of reducing the number of these passes. 
 This endeavor led to a really great discovery, namely, that a round could be 
 rolled in one pass from an oval, provided the oval was of the right dimensions 
 and was supported by metal guides. This success of the guide led to its 
 use for other shapes, also, and to its adoption in modified forms to nearly 
 all mills. Thus, another word, guide, came to be rather loosely used. In 
 general, a guide is any device used to support the piece in the correct 
 position during its passage through the groove. In order to perform this 
 
476 
 
 THE ROLLING OF STEEL 
 
 function, the guide must fit neatly against the roll or rolls, so one end must 
 be shaped to conform to the shape of the space in front of the rolls. 
 Entering guides are usually of the closed type. They are made in two parts. 
 
 Rougher 
 
 Heating 
 Furnace 
 
 Hot Bed 
 
 <& 
 
 Engine 
 
 FIG. 93. General Layout for an Eight Inch Guide Mill Belgian Type. 
 
 In each of these two parts a groove is cut parallel to its long axis, so that 
 when the two parts are fitted together the opening formed by the grooves 
 
MERCHANT MILLS 
 
 477 
 
 will be of the required shape of section to support the piece properly. The 
 guide proved to be of immense advantage in another way, in as much as it 
 permitted the rolling of very long lengths. Today, any mill designed to 
 roll sections that require the'use of guides may be called a guide mill. 
 
 The Belgian and Looping Mills: Though it was now possible to 
 roll in long lengths, a serious drawback was encountered in the old and 
 slow-going mill, for, if the piece were long and, especially, if the section 
 were small, the steel would get too cold to roll before the piece could be 
 finished. The remedy, of course, was found in greater speed, but here 
 trouble was again encountered because of the roughing rolls which refuse 
 to bite the billet if the speed is too great. Then there originated in Belgium 
 the scheme of setting up an independent roughing stand that could be driven 
 
 from the main drive shaft of the en- 
 /"*" ^ t gine at a lower speed than the finish- 
 
 f N \ i 1^ ing train, which was driven by power 
 
 transmitted by belt from a large 
 pulley on the drive shaft to a much 
 smaller one mounted on a short 
 shaft connected with the train 
 pinion. Up to this time the piece 
 was rolled throughout its length 
 in a given pass before it was 
 started into the next. Who the 
 man was, or what his degree, that 
 was responsible for the next ad- 
 vance in rolling, history does not 
 say, but doubtless it was some 
 
 FIG. 94. Early Type of Looping Mill. bold Belgian catcher who first con- 
 ceived the idea of catching the first end as it came through the rolls and 
 returning it immediately through the next pass, thus rolling the section in 
 two passes at once. Since the speed of the piece on the delivery side is 
 greater than that of the rolls, due to the elongation, the material overfed 
 and formed a loop. By this looping scheme, the capacity of the finishing 
 train was increased beyond that of the roughing stand. 
 
 The Semi-continuous or Combination Mill: Such was the extent 
 of the development until a few years prior to 1900, when two things com- 
 bined to force another advance in the merchant mill; one was the previous 
 development and success of the continuous mill as a semi-finishing mill; 
 the other was the necessity for economy due to labor troubles and 1 a severe 
 depression in the steel business. So, in order to decrease labor costs and 
 speed up the mill to the limit of the finishing train, the continuous rougher 
 was installed to replace the single three-high roughing stand of the Belgian 
 mill. By this inovation the mill force was decreased by about nine men, 
 while the output was increased 50%. The mill proper was then in a position 
 
478 
 
 THE ROLLING OF STEEL 
 
 png 
 
 anifcug 
 
 IM M M I 
 
 1 1 
 
 
 
 
 
 
 
 on 
 
 /^ 
 
 Fol 
 
 bfl L. _- 
 
 ^--^ 
 
 / 
 
 1 
 
 1 
 
 ! 
 \. 
 
 \ 
 x 
 
 i f i ! B r 
 
 Hi L! 
 . j|j 
 
 PI! 
 
 * 
 
 -Sftj- 
 
 
MERCHANT MILLS 479 
 
 to roll in unlimited lengths and thus effect another saving; but two things 
 combined to limit the length of the piece rolled, namely, the difficulty of 
 handling long pieces on the old style cooling bed and the problem of keeping 
 the bar uniformly hot for finishing. The first problem was solved by the 
 invention of the mechanical cooling bed. These beds consist of a runout, 
 in which are live rolls, for delivering the piece; a device for stopping and 
 transferring the material, which is delivered at speeds as high as twenty- 
 five feet per second, and at intervals of only one or two seconds between 
 pieces; a system of notched or fingered racks, by means of which the hot bars 
 are moved away from the runout; and a receiving table, also containing live 
 rolls, by means of which the bars are delivered to the shears. Examples 
 of these tables will be found at all modern mills. The problem of uniform 
 heat was then solved by speeding up the mill; by locating the roughing 
 rolls near a continuous furnace, as was pointed out in discussing the rolling 
 of hoops; and by arranging the drive of the mill so that the peripheral 
 speed of the rolls is increased in the successive passes to take up the slack 
 in the loops. Another device, by which the number of hands employed 
 about the mill was reduced, is the mechanical repeater. Many mills 
 equipped with these repeaters are entirely automatic in operation of the 
 rolls, and can roll material that is much too stiff to be turned or looped by 
 hand. 
 
 The Cross Country Mill : While these improvements were being made 
 in the looping mill, a mill of a different type altogether has been developed. 
 It is known as the cross-country mill, and is intended for rolling material 
 that does not lend itself well, on account of size or shape, to loop mill rolling. 
 These mills involve the continuous idea, but the stands are placed so far 
 apart that the piece must leave one set of rolls before entering the next. 
 For carrying the piece from stand to stand stationary roll tables are em- 
 ployed. To save space and avoid complicating the drive the stands may 
 be arranged on two or more parallel lines, and the direction of travel of 
 the piece will be reversed during the rolling. Three-high stands are placed 
 at the reversing points, so that, by means of mechanical repeaters and 
 diagonally placed tables, the feeding of the mill is made automatic through- 
 out. In the latest types of these mills, as for example the No. 7 mill at 
 Duquesne, which made its first trial rolling in July, 1917, the three-high 
 stands have been eliminated and two-high substituted, the piece being 
 moved over from the first roll line to the second then to the third by transfer 
 and skid tables. In general, the tendency is to return to the two-high 
 stand, because it is the simplest as well as the strongest and most rigid 
 in construction. 
 
 Future Development: From the preceding sketch, it would appear 
 that the merchant mill, so far as the mill is concerned, has been developed 
 almost to the highest stage of perfection. At present, the trend of improve- 
 ment is in two directions, one looking toward economy in power and the 
 
480 
 
 THE ROLLING OF STEEL 
 
 
 Heating Furnace 
 
 Heating Furnace 
 
MERCHANT MILLS 
 
 481 
 
 poq 8uijooo oj. 
 
482 THE ROLLING OF STEEL 
 
 other toward specialization. As to the first, the adoption of electricity for 
 driving the mills promises to effect economies. Much advancement has 
 been made since 1900 in the construction of electrical motors, which, through 
 supplemental equipment such as rotating resistors, balancing sets, etc., 
 have reached a very high state of efficiency, and a great majority of the 
 mills installed since 1916 are equipped with these motors. Specialization 
 is taking two courses. Thus, more and more specialty mills are being 
 erected, and other mills are being supplied with special equipment, such 
 as supplementary roll stands, which adapts them to the rolling of certain 
 sections yet leaves them free to roll other products, also. 
 
 SECTION III. 
 
 DESIGNING ROLLS AND MAKING UP SCHEDULES FOR MERCHANT MILLS. 
 
 Roll Designing for Merchant Mills : It does not require much imagin- 
 ation to see that the problems of the roll designer for merchant mills are 
 many and the most difficult to solve. That the art of designing rolls for 
 these small mills must have made wonderful progress is indicated by the 
 recent appearance of sections of the most intricate design, yet rolled with 
 minute accuracy. But the designing of passes for the sections is but one 
 of the problems confronting the roll designer, as a visit to the mills will 
 show. Among these should be mentioned the great number of designs, 
 which require the most systematic filing and recording of rolls, templets, 
 and tools. The ingenuity of the roll designer is taxed to the utmost to 
 keep the number of rolls at a minimum. So, the visitor in the mills will 
 find that numerous sizes, differing by only a few thousandths of an inch, 
 are placed in the same roll. The efforts of the roll designer along this 
 line are especially noticeable in the roughing stands. Here the process is 
 merely one of reduction, the piece leaving the roughers in the form of a 
 square or rectangle. To bring about this reduction one of five methods, 
 or combination of passes, may be employed. Thus, in continuous roughing, 
 the student will observe four methods in use, which may be designated as: 
 1. Diamond square; 2. Oval square; 3. Flat-and-edging; 4. All- 
 diamond-to-square. In three-high roughers, either the flat-and-edging or 
 the diamond-pass method is used. The following sketch is intended to 
 illustrate these methods. Nearly all the shaping, except in the case of 
 deformed bars already alluded to, is done in the stands, called the strands, 
 immediately following the rougher. The strands are sometimes preceded 
 by an auxiliary set of roughers called the pony roughers. The planisher 
 may be employed as a last forming pass only or, in addition, to give a finish 
 to the bar. The finishing pass is reserved exclusively for removing the 
 little irregularities of the previous rolling. 
 
 Economic Features of Roll Designing: It scarcely needs to be 
 pointed out that aside from the successful rolling of the various sections, 
 the chief incentive behind the efforts of the merchant mill roll designer 
 
MERCHANT MILL ROLL DESIGN 
 
484 THE ROLLING OF STEEL 
 
 is found in the necessity for economy. 
 The rolls, alone, as maybe surmised from 
 what has been said concerning their 
 manufacture, etc., are very expensive, 
 and a large number of rolls on hand 
 means a large investment in something 
 that is idle most of the time. Added 
 to this feature is the expense of roll 
 changes, during the time for which the 
 entire mill is closed down. In the elimi- ^_ 
 nation of roll changes much depends on ^y c2 
 the way the orders are scheduled. This 
 point should, therefore, always be con- 
 sidered in making up the mill schedules, 
 and is a problem to be solved in making 
 up promises of delivery for small lots. 
 As an example of how scheduling is 
 done, the system employed at the Upper 
 and Lower Union Mills at Youngstown 
 is explained. 
 
 Making Up Schedules The Order 
 in the Office: Upon receipt of the large 
 buyers' schedule, which usually comes 
 about the middle of the month for the 
 following month, the respective schedule 
 clerks pick out from the files the index 
 cards corresponding to the buyers' 
 schedule and file them in a desk file. At 
 the same time all those index cards with 
 promises on them for the next month are 
 taken out and put along with those 
 already in the desk file. In addition to 
 these the oldest index cards are also 
 worked in whenever it is possible . Before 
 these index cards are put in the desk 
 file, however, they are checked against 
 the book orders for mistakes, etc., and 
 the date of order. By this means it is 
 known that the item is either in the 
 desk file or out on the mill. The 
 index cards in the desk files thus represent 
 all the promised and scheduled material 
 for the next month. From these files 
 the schedule clerks make up rollings 
 for their respective mills. For this 
 
MERCHANT MILL PRODUCTS 
 
 485 
 
 joii oj, 
 
486 THE ROLLING OF STEEL 
 
 purpose index cards showing the different sizes of sections and grades 
 of steel on the schedule are also kept. To keep the scrap loss 
 on the mill to a minimum the longest exact cuts are put first, the exact short 
 lengths come next and after these are placed the common lengths. There- 
 fore, when shearing the long exact cuts, the short ends may be sheared into 
 the shorter lengths; and if a long piece comes at the end, it maybe sheared 
 to one of the long common length cuts. The amount of tonnage to a roll- 
 ing varies with the mill and the section being rolled. In making up a 
 rolling it must be watched that a car load of material is checked out, so 
 material will not have to be piled. If there is not enough material at 
 either Upper or Lower Mills to make up a car load for the customer the 
 products for both Upper and Lower for that customer are combined to 
 make up a car load, the shipping offices having previously decided at 
 which plant to start the car. Sometimes an order will be transferred from 
 one mill to the other so as to prevent this condition. In this case, the 
 order is transferred by making out two correction slips. One of the slips 
 goes to the one mill telling its foreman to cancel the order that is to be 
 transferred, and the other goes to the other mill telling its foreman to 
 reinstate the order mentioned. The correction slips are made out in 
 triplicate form, a copy of each remaining at the order department, and the 
 original and a copy of each going to the proper shipping office where the 
 original is returned and noted to the order department, while the copy is 
 kept on file at the shipping office. The schedule clerk in the order depart- 
 ment keeps a record of the orders he sends to the mill order clerk for each 
 mill and from time to time checks up with him regarding the amount of steel 
 on hand. After a rolling has been made up, the index cards are hectograph- 
 ed and three copies are made. These copies are known as mill order sheets. 
 
 The Order at the Mill Size of Billet or Bloom : The original index 
 cards, together with the copies, are sent to the mill order clerk, who gives 
 the index cards to the shipping office, keeps one copy of the mill order, 
 and sends the other two to the mill foreman. These should be sent over 
 to the mills at least twelve hours before they are to be put out on the mill, 
 so that the foreman will have time to figure out the weight billet required 
 and to order the steel from the mill stocker. Due to various influences, 
 however, the orders are often sent out only a few minutes before they are 
 put on the milL In figuring out the billet or bloom required for the order, 
 the foreman figures the length to which he can run the material on the 
 cooling bed and obtain multiple lengths of the cut ordered, taking into 
 consideration both the length of the hot bed and the mill practice. He 
 then multiplies the mill length that the bar is to be rolled on by the weight 
 per foot of the section, which gives him the weight billet or bloom needed, 
 not allowing for scrap or furnace loss. About 10% is added to most orders 
 to offset furnace and scrap loss. However, this factor changes with the 
 section and the cut. Common length cuts on large material requires some- 
 times as little as 5%, but on the other hand light material, such as crescents, 
 
MERCHANT MILL PRACTICE 487 
 
 on which the cut is exact and the inspection rigid, the amount added will 
 sometimes be as high as 25%. In figuring the weight billet required, the 
 width of the furnace must also be taken into account. Blooms for common 
 grades of steel are furnished by the Ohio Works in different weights, 
 increasing by fives from one hundred to three hundred pounds, while bil- 
 lets, unless otherwise requisitioned, are furnished in lengths of thirty feet. 
 After figuring the orders, the mill foreman makes out a steel order with 
 two carbon copies. The original is kept on file; one of the copies is given 
 to the stocker, who sees that the billets or blooms are cut to weight and 
 delivered to the heating furnaces; and the other is given to the steel yard 
 foreman, so he can check the same against his records. The orders are 
 then given to the shear foreman and the roller, the size of the section 
 being marked on the margin of the iatter's sheet in large clear figures in 
 order to eliminate any possible error in size. When the shipping office 
 receives the index card from the mill order clerk, the card is blue- 
 penciled showing date the order was received from the order department. 
 It is then returned to the mill order clerk who files it until he places the 
 order on the mill. The card is again given to the shipping office when the 
 order is placed on the mill, and, this time, the date the order is put on the 
 mill is marked with black pencil. A shipping clerk looks up the book 
 order and marks the index card with the information as to whether the 
 material is to be piled or loaded into a railroad car, which is dependent 
 upon how much material for that customer is being rolled. The marked 
 index card is then filed according to size and section. 
 
 SECTION IV. 
 
 ROLLING PRACTICE IN MERCHANT MILLS. 
 
 The Roller : While the roll designer is indispensible, and the roll shop 
 represents the heart of the rolling mill plant, much depends on the roller. 
 He must be a man with not a little ability, capable of exercising good judg- 
 ment at all times and possessing considerable practical knowledge. With 
 closer and closer rolling tolerances being demanded by the customer and 
 larger tonnages required, due to the rapid growth of the steel industry, 
 the roller has been placed in the difficult position of meeting rigid speci- 
 fications and breaking tonnage records at the same time. This achieve- 
 ment appears the more remarkable when it is recalled that all the training 
 the roller gets is along practical lines. His only school is the school of 
 experience in which a man is never graduated. It is manifestly impossible, 
 therefore, to explain satisfactorily the knowledge of the roller or the method 
 of his working. The best that can be done is to point out some of the 
 duties of the roller and mention a few of the precautions he must 
 observe. 
 
 Precautions in Rolling: In a word, it is the duty of the roller to see 
 that the rolling is .properly done and the material meets the specification 
 
488 THE ROLLING OF STEEL 
 
 as to size, shape and freedom from rolling defects. In building up rolls 
 in the housings care should be taken by the roller that they are plumb, 
 square and level. If the rolls are not plumb, i. e., if the line joining their 
 centers is not straight and perpendicular to a horizontal, the bar will not 
 deliver properly, and trouble with the guides will undoubtedly occur. This 
 condition is often caused by bearings wearing out or poor babbitting 
 originally and can only be remedied by changing bearings or, in some 
 instances, by the use of side liners or wedges. Should the rolls be out of 
 square, that is, if the center of the pass in the top roll is not directly above 
 the center of the pass in the lower roll, the bar will be out of square and 
 will twist as it issues from the rolls. In order to square up the rolls, set 
 screws which work against the bearings are provided on the sides of the 
 housings, and thus the rolls can be thrown either one way or the other as 
 the case may demand. The directions for correcting this fault are, in the 
 language of the mill, "Follow the twist," and the top roll is always thrown 
 over in the direction that the bar is twisting. In the event that the rolls 
 are not level, that is, perfectly parallel, more work will be done on one side 
 than on the other, with the result that the bar will not deliver straight 
 but will tend to curve around toward the side on which there is the lightest 
 draft, due to the other side being elongated the more. This condition may 
 be remedied either by the use of liners or by operating the screw down at 
 the proper side. Guides and guards play a most important part in rolling 
 mill practice, and the proper setting of these is one of the roller's most 
 important duties. His assistants may set the guides for the strand and 
 planishing rolls, but those for the finishing pass are always set by the 
 roller. Entering guides on the finishing passes are usually closed, the 
 inner end being so shaped that it will provide sufficient bearing to hold the 
 piece up in correct position while entering the pass. If the guides are not 
 set properly, the bar will not be formed rightly. Especially when rolling 
 rounds, the entering guides should be tight in order to hold the oval in a 
 vertical position, for a leaning to either one side or the other will produce 
 a high and a low shoulder on the finished round. The position of the 
 guides on the delivery side is also most important, for since the bar has a 
 tendency to follow the smaller roll diameter, the guide against which the 
 bar is thrown must be watched most carefully. If the bottom roll is the 
 smaller, then the bottom guide should not be placed too low, for the bar 
 coming out would have a tendency to follow the roll down for a short space 
 before striking the guide. This would cause an up and down kink, or a 
 buckle. A short guide has the same effect. 
 
 Rolling Defects: In addition to working for the proper size and finish 
 on the bar, the roller must watch for such surface defects as overfills or 
 pinches, underfills, buckles, slivers, seams, laps, firecracks, roll marks, etc. 
 Overfills or pinches must be watched especially in changing from Bessemer to 
 open hearth steel, as the latter has more of a tendency to spread than does 
 the former. When overfills occur, the amount of stock entering the pass 
 
MERCHANT MILL PRACTICE 489 
 
 that is producing the overfill must be reduced by adjusting the rolls in 
 preceding passes. Underfills are corrected by reversing these operations. 
 Buckles are sometimes caused by worn out pinions. Slivers can be 
 produced from many causes at the blooming mill or by the bar shearing 
 against a guide or collar of a roll at the finishing mills. The former 
 condition cannot be corrected by the merchant mill roller, but the latter 
 can be eliminated by a proper adjustment of the proper guide or by reducing 
 the stock in the bar which is shearing against the collar. Seams are defects 
 in the steel that cannot always be corrected by the finishing mill, as they 
 are usually formed in the bloom or the billet. A lap is caused by an overfill 
 or fin being formed and then being doubled over and rolled down in the 
 subsequent passes. This defect can be controlled by the roller by going 
 back to the stand at which the overfill was formed and reducing the stock. 
 Firecracks are caused by the rolls becoming overheated and cracking on 
 the surface. These cracks cause corresponding small elevations on the 
 surface of the bar, which in some instances condemn the material. When 
 this defect appears, the roller "moves over" and uses a "clean" pass. The 
 same procedure is followed when any other roll mark appears on the finished 
 bar. Roll marks occur at equal intervals along the bar and signify that 
 there is a piece out of the roll or that the roll is marking the bar with each 
 revolution in some other manner. 
 
 Two Different Finishes on bars are furnished at the Youngstown 
 plant, namely, common and special. The special finish has a smooth, highly 
 polished appearance and is produced by cleaning all scale from the bar at 
 the planishing stand and finishing at such a low temperature that no more 
 scale will form on the surface. 
 
 The Special Finish is produced on rounds by holding the square back 
 before entering the planishing, until a dark scale has formed and then bend- 
 ing the bar with a pair of special tongs as it enters the rolls. A stream of 
 water plays directly upon the bar as it enters both the planishing and finish- 
 ing passes, and scouring blocks covered with emery powder are used on the 
 finishing stand in order to keep the pass clean. Material requiring the 
 special finish is always rolled ahead of the common orders of like sizes, 
 so that a clean pass will be available. Flats are not scraped in order to 
 furnish the special finish, but are simply held back until their temperature 
 reaches the critical range before entering the planishing pass. On large 
 fiats, water is used on the bar as it issues from the planishing, so that the 
 scale thus broken up will be removed and not be rolled into the steel. 
 Another reason for using water on large sections is that they will be delivered 
 to the hot beds below the scale forming temperatures. Cooling is un- 
 necessary, however, on small sections, such as crescents, half ovals and 
 ovals, as these lose their heat so rapidly that even water is not used directly 
 on the bar. A scraper, located at the entering side of the finishing stand, 
 
490 THE ROLLING OF STEEL 
 
 is the only means used for producing the special finish upon these sections. 
 Due to the fact that scale adheres more tenaciously to open hearth than 
 it does to Bessemer steel, the latter takes a much better finish than the 
 former. Open hearth steel will become smooth but does not have the 
 highly polished appearance of Bessemer steel. It is the roller's experience 
 that Bessemer screw steel takes the best finish of any grade turned out at 
 the converting mills. This grade not only takes a smooth finish but some 
 times gives a mirror-like surface. It should be observed that the holding 
 back of the bar to produce this finish may so retard the rolling as to 
 decrease seriously the total output of the mill. 
 
 SECTION V. 
 
 SHEARING AND BUNDLING MERCHANT MILL PRODUCTS. 
 
 The Methods of Shearing and Bundling vary at the different plants 
 according to equipment, product, location, etc., and no description of value 
 yet general enough to be descriptive of all plants can be given. As the 
 greatest variety of sections are produced at the Youngstown Upper and 
 Lower Union Works, a brief outline of the methods and practices at this 
 plant may be found of value. 
 
 Duties of the Shear Foreman: The man who is responsible for 
 completing orders, i. e., for the shearing and the bundling on each mill, 
 is the shear foreman. His force is usually composed of a shearman, a 
 gauger, a push-up, a pull-up, and two or more bundlers. The first duty 
 of the shear foreman is to keep the different orders, heats and turns separate 
 on the cooling beds and to tag each item on the truck properly. The 
 different lots are kept separate on the trucks by means of bands. A load 
 sheet is made out for each truck, showing the material loaded on it. When 
 the truck is full, it is the duty of the shear foreman to notify the yard master 
 or dinkey engineer to pull the load to the warehouse, or shipping room. 
 It is also the duty of the shear foreman to set the gauge for shearing the 
 material, allowing a certain amount for contraction during the cooling 
 process. The amount allowed on the smaller mills is one-fourth inch over 
 or under for every five feet, but this amount varies with the size of the bar. 
 When material is to be bundled, the weight of the bundles is nearly always 
 specified, but when instructions are not given, it is the mill practice to 
 bundle material to weigh 100 to 150 pounds per bundle. By multiplying 
 the weight per foot of the section by the cut and dividing the product into 
 the weight of a bundle, the shear foreman determines how many bars to 
 put in a bundle; but in order to check himself up he weighs the first bundle 
 
MERCHANT MILL PRACTICE 491 
 
 of each new cut. It is the object of the shear foreman to put the same 
 number of bars in each bundle, as this not only makes all the bundles of 
 uniform weight but facilitates the recounting of a bar order by the men in 
 the warehouse. On some mills the head bundler counts the bars, while on 
 other mills tally boys are employed. In bundling material up to twelve 
 feet on domestic orders only two "Carnegie" bands are used. Up to twenty 
 feet three "Carnegie" bands are used, while four are used on all cuts twenty 
 feet and over. 
 
 Bundling Export Material : On export orders, the weight of a bundle is 
 always specified, the weight usually being 112 pounds, so that twenty bundles 
 make one gross ton. On lengths up to fourteen feet three bands, marked 
 "Carnegie, Made in U. S. A.," are used. From fourteen feet to twenty-two 
 feet four bands are used, and one additional is put on for every six feet above 
 twenty-two feet. On all export orders, special "export" tongs are used, 
 as with these tongs the material can be tied very securely. The following 
 rules apply to the handling of export orders: All orders must be complete. 
 All bundles, excepting the last, must contain the same number of bars. 
 All bundles are to be tightly tied with "export" tongs whenever possible, 
 and bands used are to be in accordance with instructions on orders. These 
 instructions may call for export bands, plain bands, or wire. All export 
 material must be loaded on separate trucks, except in very small orders, 
 or when the material is needed by the warehouse to complete an order 
 hurriedly. All trucks containing any export material must bear a red export 
 tag, and all excesses are to be loaded on trucks but kept distinctly separate, 
 tagged plainly as "Excess," and showing order number, customer, size, 
 etc. The name of the man who is responsible for tallying and bundling 
 is placed on the tag, and all orders are very plainly tagged, the tags being 
 so placed that there is little liability of their being torn, pulled off or made 
 illegible. On all trucks are placed load sheets which are put in oiled 
 envelopes so that weather conditions can not destroy, or blur, the writing. 
 
 Special Bundling: Some material, such as ovals, tees, etc., often 
 specify "Tie with wire," while other orders, especially for large material, 
 specify the material to be shipped loose. A good many orders, such as 
 hoops, flats, rounds, squares, and small nut steel, are often ordered coiled. 
 This is accomplished by means of coilers with adjustable pins so that coils 
 of different diameters can be furnished. The usual diameter, however, is 
 twenty-four inches. These coilers are located at one end of the hot bed 
 or at some other convenient point according to available space. All coilers 
 are electrically driven. 
 
 Handling the Material in the Warehouse: When the truck load of 
 material arrives at the warehouse, the pile-boy takes the load sheet into 
 the office and gets the corresponding index cards, because the load sheet 
 shows only the order number, size, cut, grade of steel and remarks. The 
 index cards show customers name and loading and shipping instructions. 
 
492 THE ROLLING OF STEEL 
 
 As each item is weighed off from the truck, it is entered on a weight sheet 
 by a weigh-man and checked off on both the index card and load sheet. 
 If the index card is marked "Car" a car is started for that customer, if 
 this has not already been done, and material is so loaded. If index card 
 is marked 'Tile" the item is piled and so noted on the weight sheet. One 
 weight sheet is made out for each truck load, and when this car has been 
 weighed off, the sheet is checked off on the car card, of which there is one 
 for each car, and returned to the office. Here the weight sheets are checked 
 against the order books. Memorandums are made out for the items that 
 have been piled so that these can be given out to the weighmen when a 
 car load has been started for the customer. When an item of an order is 
 under weight or short in number of pieces, the shortage is discovered by 
 a weekly survey of the order books by a clerk in the shipping office, who 
 makes out a reindex card for the shortage and notes the date made out 
 with blue pencil on both the reindex card and the book order. If, however, 
 there is a chance to get the item on the mill immediately, the book order 
 and the index card are black penciled. A rolling order is made out from 
 the reindex card and hectographed, after which the index card and three 
 mill order sheets are given to the mill order clerk. He returns the reindex 
 card to the shipping offices, and if he does not have the grade of steel in 
 stock, he has to make out a requisition for it, the procedure then being 
 the same as for ordinary orders. 
 
 Straightening: At the lower mill, straightening is done by the Labor 
 Department, machines for this purpose being located at three different 
 places. These machines are of the seven roll type, the rolls being built 
 up in a casting similar to roll housings. Four rolls are below and three 
 above, the bar passing between in grooves which are designed to fit each 
 separate section. On account of the vast amount of tire and window sash 
 sections rolled at the Upper Mills, the straightening is under the juris- 
 diction of a special department for that purpose. Angles and sash sections, 
 molding tees, and mud guard sections are straightened at the mill, there 
 being a straightening machine located at each mill where such material 
 is rolled. Round edge tire, however, is straightened in the tire house, 
 where two straightening machines of a new type are located. These are 
 more flexible than the old type and are more easily adjusted to the various 
 sizes. 
 
 Invoicing: After the cars have been loaded, the car cards are taken 
 into the shipping office, and from these, invoices are made out. The original 
 office copy of the invoice, which is made out in the shipping office, is sent 
 to the order department where it is carefully checked against the book 
 order, the number of bars or bundles and tonnage being entered on same. 
 The invoices are next given to a clerk who enters all detailed information 
 on a "recap" sheet. The credits for the various sections and sizes are also 
 entered under the proper headings on a "credit recap sheet." 
 
MERCHANT MILL PRACTICE 493 
 
 SECTION VI. 
 
 INSPECTION DEPARTMENT OF A MERCHANT MILL PLANT. 
 
 The Inspection Department makes all physical tests and keeps 
 records and samples of the various sections. One of the duties of this 
 department is to inspect and accept or reject all special steel before being 
 rolled into finished product. Check analysis is made of all steel requiring 
 the same, and a close inspection of all material when being rolled is provided 
 for. All special steel ordered from the semi-finishing mills must pass 
 inspection by this department. Approval or disapproval of material is 
 based upon chemical analysis and surface conditions. Only very low 
 limits for the various impurities are allowed on special steels, and the 
 inspection is very rigid. Consequently, any heats not falling within the 
 requirements must receive special attention. Much in the way of good 
 judgment is required to dispose of such heats satisfactorily. Usually, the 
 department will endeavor to consult the customer before permitting an off 
 grade heat to be rolled as originally planned. Certain orders require the 
 ladle analysis to be checked before being rolled into the finished bar. Orders 
 requesting check analysis are held in the yard until drillings from 
 the billets are analyzed. If this analysis shows that the composition of 
 the steel is as ordered, the steel is rolled on the order for which it was 
 originally intended. If the check analysis does not practically agree with 
 the ladle analysis, the steel is applied on a less particular order. Check 
 analysis may also be made on finished material. Some orders require that 
 the steel be inspected for surface defects before rolling. In such cases 
 each billet is carefully examined to detect any slivers, seams, checks, or 
 faulty shearing that may occur. Billets found to be defective are chipped 
 and put into condition for rolling if at all possible. When orders specify 
 physical requirements, it is then the duty of this department to supply such 
 chemical specifications as will' fulfill the physical requirements. 
 
 Another Function of this Department is that of mill inspection, 
 which is one of most importance. In this capacity the department acts 
 as a check upon the rolling. Mill inspection requires one man on each mill, 
 devoting his entire time to gauging and watching for faulty steel. Sections 
 not fulfilling the prescribed measurements are either held for further 
 inspection or thrown out as scrap. The rollers as well as the inspectors, 
 have the given dimensions and tolerances. The inspectors check the rollers 
 and inform them of any faults that the rollers themselves have not already 
 detected. In case an inspector does not accept steel as rolled, and the 
 roller continues to make the section, the inspector signals for the depart- 
 ment superintendent and lays the case before him. If the fault cannot be 
 remedied, and it is known the customer will not accept the steel as rolled, 
 the mill must go off the order. The defects watched for most closely 
 depend upon the section being rolled. Accurate size applies to all sections. 
 
494 MERCHANT MILL PRODUCTS 
 
 For the more complicated sections templets are furnished. Usually one 
 exact and one full templet is made. For gauging rounds, squares, flats, 
 etc., only a gauge and micrometer is used. Readings on the micrometer 
 are accurate to .001 inch, while on the gauge one-sixty-fourth inch is about 
 the most exact reading that can be determined. Other tools used for special 
 purposes are squares and steel tapes. The square is used to detect diamond- 
 ing in certain instances where each surface must be at right angles to the 
 adjacent one. The tape is used when inspecting clip sections, to determine 
 the regularity of the impressions that are rolled on the bar. 
 
 Surface Defects: The inspector is responsible for the detection of 
 surface defects. These may appear as buckles, kinks, overfills, underfills, 
 slivers, laps, seams, or burned steel. While the nature and causes of these 
 defects have already been more or less fully explained, the following resume" 
 of defects most likely to occur in merchant mill rolling is appended for 
 ready reference: 
 
 Buckles and Kinks: A bar, when delivered from the finishing rolls, 
 may be wavy, either up and down or sideways. The former is known as 
 a buckle, while the latter is a kink. These defects are more injurious to 
 some sections than others. However, all sections should be rolled as free 
 from buckles and kinks as possible. Crescents have a tendency to buckle, 
 consequently they must be watched closely. 
 
 Fins: If a bar has a fin or extra amount of metal at the sides where 
 the finishing rolls come together, the bar is said to be over-filled. Bars 
 rolled for cold drawing must be free from over-fills, for these draw into 
 laps. On the other hand, in order to get perfect corners on half ovals, a 
 well known file manufacturer requests a small overfill at the edges. 
 
 Underfills: When a bar is scant in certain dimensions or when it is 
 not completely filled oiit, the bar is said to be underfilled. This defect 
 sometimes appears on rounds and channels. 
 
 Slivers are loose pieces of steel rolled flat on a bar. They may be 
 present on the billet or be caused by faulty shearing, or incorrect entering 
 of the bar in a closed pass. Slivered steel is thrown out as scrap and seldom 
 held for further inspection. 
 
 Laps: If a bar is given a pass in the rolls after an overfill has been 
 produced, a lap usually results. This defect is especially liable to occur 
 with skelp, hoop and cotton-tie. Faulty ingots and poor rolling at the 
 semi-finishing mills also cause laps. 
 
 Seams: Steel must be inspected carefully for seams, a surface defect 
 always difficult to detect. A seam is a crevice in steel that is closed 
 up but not welded. Seams are caused by blow holes and cracks in the 
 ingot, as well as faulty methods of rolling. They render steel unfit for 
 hardening. 
 
INSPECTION 495 
 
 Burned Steel shows up in the finished bar in the form of rough, checked 
 edges. Burned steel is ruined, and the only alternative is to scrap it. 
 
 Roll Marks: Sometimes a roll is nicked, or a piece breaks off the 
 roll, resulting in periodic impressions along the bar. The defect is 
 corrected only by a new pass. Fire cracked rolls make similar impressions. 
 
 Finish: The finishing pass, if worn, does not give a smooth surface, 
 consequently when the surface of a bar is not up to the standard in finish 
 the pass must be changed. This defect is especially objectionable when 
 rolling cotton tie, hoop, skelp and sections requiring the special finish. 
 
 Pipe: The inspector must watch the sheared ends of bars for piping. 
 This defect, however, is usually detected at the billet shears before the 
 steel is charged. 
 
 Testing for Defects: Tests for detecting some of these defects are 
 employed by the inspectors. The tests most commonly employed for this 
 purpose are the upset, forging and pickle tests. The upset test consists of 
 heating a short sample piece in a forge or furnace and upsetting under a 
 hammer. This is a severe test and readily shows up seams and laps. For 
 forging tests, a bar about ten inches long is heated and forged along the 
 longitudinal axis of the bar, which is then nicked and broken. Piped steel 
 is readily detected by this treatment, as the forging opens up the pipe. 
 The upset test is more often used than the forging test, especially when 
 rolling forging steel. The pickle test consists of immersing for a few 
 moments short pieces of the material in dilute sulphuric acid. The acid 
 removes the scale from the bar, exposing to view and exaggating any surface 
 defects that may be covered by the scale. Tests pieces of hoop and file 
 steel are pickled. At the Youngstown Upper and Lower Union Mills all 
 forging tests must be turned in for the personal inspection of the depart- 
 ment superintendent, while the pickle tests are saved at the test house 
 for six months before being discarded. With hoop, the pickled samples are 
 held in a vise and the edges turned over with a hammer. This distortion 
 of the metal opens up any laps that may be present. On special section 
 the inspectors are required to save samples every half hour. These are 
 bundled together, properly tagged, and sent to the test house, where they 
 are saved for three months. 
 
 Other Duties of Inspectors: Besides inspecting material for surface 
 defects, the mill inspector must check the bundling requirements as well 
 as the length of cut. This practice reduces bundling and shearing errors 
 to a minimum. An hourly report with carbon copy is made by each mill 
 inspector. This report shows the variations in size of the bar for each 
 hour of the day. Notations are made of any steel held up and the cause 
 of the same. These reports must be delivered to the inspection office at the 
 end of each turn. Since the different sections are gauged differently, the 
 make-up of the reports will differ accordingly. 
 
496 MERCHANT MILL PRODUCTS 
 
 Manner of Gauging Different Sections: Rounds are gauged on four 
 diameters, distinguished at the mill as top-and-bottom, sides, and high and 
 low shoulder. By top-and-bottom of a round is meant those two surfaces 
 subject to compression in the finishing pass, and by sides is meant the 
 points opposite the clearance between the rolls in this pass. The shoulders 
 lie between these two diameters. The longer of the shoulder diameters 
 is called the high shoulder, the shorter the low shoulder. Three samples 
 are taken from each bar gauged, namely, front end, middle and last end. 
 These samples are taken at the shears as the original bar is being cut into 
 the lengths ordered. The top and bottom of a round may be distinguished 
 from the sides by the way the scale is broken along the sides. If an overfill 
 occurs, it shows also on the sides of the round. Usually, the ends of a mill 
 length of "a round are slightly overfilled. Flats, squares and nut steel are 
 gauged for width, thickness and diamonding, but only the variations in 
 width and thickness are reported. Cotton-tie, hoop, and skelp are reported 
 for width and gauge thickness. Special sections are usually gauged by 
 means of templets, but certain overall dimensions are generally given on the 
 report. A sketch is sometimes made and the important dimensions lettered. 
 The hourly variations for these dimensions are then inserted under the 
 proper heading. Clip-iron, box-strap and bit-mouth are peculiar sections 
 requiring special attention for gauging, and the various dimensions must 
 be watched to get the section uniform. Inspectors at the cotton-tie mil! 
 have special duties to perform. They must get from each buckle machine 
 every half hour a sample, which must be properly tagged and taken to 
 the department superintendent for personal inspection. They are required 
 to weigh ten bundles of cotton-tie every half hour and to post the weight 
 on a blackboard in plain view of the roller. Concrete bar is rolled to 
 weight, and inspectors must check the weights of the shear foreman. If 
 for any reason an inspector is not willing to take the responsibility of passing 
 slightly defective bars, the trucks-loaded with such steel are marked with 
 green tags, signifying, "hold for further inspection." This steel is examined 
 by the assistant chief inspector, who either scraps the entire lot or details 
 a specially instructed inspector to sort the good from the bad. Uses for 
 which steel is rolled, as well as customers' claims, offer guides as to what 
 defects a customer can accept . 
 
CIRCULAR SHAPES 497 
 
 CHAPTER X. 
 
 CIRCULAR SHAPES. 
 
 SECTION I. 
 
 SOME GENERAL FEATURES PERTAINING TO THE ROLLING OF CIRCULAR SHAPES. 
 
 The Rolling of Circular Shapes presents one of the most interesting 
 studies of the rolling mill industry, because it is the latest development 
 in rolling, and, though the idea of rolling wheels originated in Europe, it 
 is in America that the art has been most highly developed. The beginning 
 of this branch of the industry dates from the year 1903, when solid rolled 
 steel car wheels were first used under freight cars. The use of such wheels 
 resulted from the introduction in 1896 of all steel freight cars, which on 
 account of their increased weight and great carrying capacity required 
 a stronger and tougher wheel than any that had been made up to that time. 
 It was to meet this requirement that Mr. Charles T. Schoen, who was the 
 pioneer in the manufacture of all steel cars, perfected the mill, which now 
 bears his name, for rolling these wheels. Later on, Mr. Schoen's method 
 of preparing the steel, which will be explained later, was much improved 
 by the Carnegie Steel Company, who purchased this mill in 1908. Con- 
 sidered from the standpoint of circular shapes in general, the Schoen mill 
 has the one drawback of a very limited product. Being designed for one 
 particular purpose, it can roll only car wheels, or wheels of that class, and 
 of these it is limited to wheels between thirty and forty-two inches in 
 diameter. While wheels as small as twenty-eight inches in diameter have 
 been made on this mill, these smaller sizes are rolled with much difficulty, 
 due both to the form of the rolls and the manner of rolling. For forming 
 wheels less than thirty inches in diameter the Carnegie Steel Company has 
 found that the forging press gives the most satisfactory results. 
 
 Preparing the Blanks: The circular shapes all require a round blank 
 to start with. Mr. Schoen originally sheared his blanks from slabs with 
 a specially constructed punch-like shear, the further work being then com- 
 pleted in much the same manner as it is done today. But this method 
 had the serious fault of producing a wheel in which the line of segregation, 
 or pipe, if any were present in the slab, was located diametrically across 
 the wheel and terminated at both ends in the tread. From what has already 
 been said about pipes and segregated steel, it is easy to see how this location 
 of the segregated area might develop defects at these two opposite points. 
 As has already been intimated, the Carnegie Steel Company was responsible 
 for bringing about the correction of this fault, which is removed by locating 
 
498 THE ROLLING OF STEEL 
 
 the segregated line at the center and at right angles to the radii of the 
 wheel, where the faulty material may be punched out for the bore. It is 
 evident that the line of segregation may be so located in any one of three 
 ways, namely, by casting the blanks individually, by cutting the blanks 
 from round or hexagonal ingots, and by rolling the ingots into a round 
 bloom from which the blanks may be sheared or sawed. All these methods 
 are in use by the various manufacturers of wheels, but it would appear 
 that the second and the third method should produce the best wheel, because 
 more work is put on the steel. The third method is the one used by the 
 Carnegie Steel Company. The round blooms for the Schoen mills are 
 rolled at present on a twenty-eight inch bloomer at Homestead. Here a 
 22" x 22" ingot is slowly and carefully reduced in from twenty-one to thirty- 
 one passes to a round bloom, eleven or fifteen inches in diameter for forged 
 products, or fifteen inches in diameter for all wheels that are to be rolled 
 at Schoen. From the blooming mill, the bloom, is delivered to a patented 
 shear, known as the Slick shear, which is so located, in conjunction with the 
 delivery table and the manner of rolling, that first cuts are made from that 
 part of the bloom corresponding to the bottom or butt of the ingot. This 
 first cut, usually about 5% of the ingot, is just sufficient to square up the 
 butt end of the bloom and is always discarded. The remainder of the 
 bloom, excepting the discard for pipe, is then cut into lengths to give the 
 proper weight of metal required in the wheel, with an allowance of ten 
 pounds over or under weight, and, if requested, each cut is hand stamped 
 with a letter to indicate its position in the ingot, starting with A 
 for the first cut next to the discard at the top of the ingot. Cut A, 
 and often cut B, also, is used in making wheels for the use of the Steel Cor- 
 poration only. In any case the total discard, which may include both A 
 and B cuts, on wheels to the customer is never less than 25%, which amount 
 is sufficient to insure sound steel in the wheels. For marking the heat 
 number and weight of cut, the shear is provided with a stamp mounted 
 on the revolving clamp for the shear knife, so that each disc, or blank, is 
 plainly stamped with its heat number and weight. From the shears the 
 blanks are taken to a shipping yard, where they are carefully inspected for 
 surface defects, which are cut out by means of pneumatic chipping tools. 
 Such of the blanks as pass this inspection are then sent to the mills to be 
 worked into wheels. 
 
 SECTION II. 
 
 THE CARNEGIE SCHOEN METHOD FOR MANUFACTURING STEEL WHEELS. 
 
 The Carnegie Schoen Method : At the Schoen plant, which consists 
 of three separate units, the finished wheel is produced in several stages, 
 the number of which depend upon the kind of wheel, the unit in which it 
 is made, and the working conditions of the heating furnaces. Upon receipt 
 of the blanks at the plant, they are check weighed, and the heat number 
 of each blank as well as the letter indicating its position in the ingot are 
 
CIRCULAR SHAPES 499 
 
 all recorded in the form of a serial number. The blanks are then charged 
 in order into a reheating furnace, where they remain for about two and 
 one-half hours. In the older units, mills number one and two, the reheating 
 furnaces are all of the regenerative type and use producer gas; but in the 
 most recent mill, or number three, the first furnace is of the continuous 
 type, the bottom of which is inclined sufficiently to cause the blanks to 
 roll down from the feeding end as rapidly as they are removed at the drawing 
 end. This type of furnace heats the discs very slowly and gradually, 
 because it is intended merely to give the discs a preliminary heating, and 
 its temperature is therefore low. At the drawing end, the temperature is 
 maintained at about 800 C. After being subjected to this preliminary 
 heating, the discs are transferred to a regenerative furnace for the final 
 heating previous to forging. When the discs have reached the proper 
 temperature, they are withdrawn from the furnace and taken to the forging 
 presses where each is forged to a shape resembling that of the finished 
 wheel before dishing or coning. As to dimensions, this forged blank is 
 from four to five inches under size in diameter, some three-fourths inch 
 over size in that part of the web near the rim, somewhat oversize, also, 
 in depth of rim, a little oversize in width of rim, but of correct or slightly 
 full size in the hub and a small part of the web next to the hub. 
 
 Forging the Blanks First Method: The forging may be done in 
 one heat or two heats and on one press or on two different presses. When 
 two heats are used, two presses are usually employed for the forging. In 
 this method, the disc is cleaned of scale on its two ends and then placed 
 vertically in the first press, where it is first perfectly centered by two 
 arms which engage it from opposite sides of the press and so support it till 
 the top die has descended upon it. In this press the bottom die corresponds 
 to the outside face of the wheel while the top die is plain, but may be slightly 
 convex to cause a radial flow of the metal in taking the shape. The pressure, 
 applied in successive steps by means of accumulators and intensifier, starts 
 with about 700 pounds per square inch, then increases to 2500 pounds and, 
 finally, to as much as 4500 pounds per square inch, if needed. At the begin- 
 ning, the scale is cracked from the disc and falls into the bottom die, from 
 which it is blown by means of a steam jet, in order to avoid pitting the 
 surface of the blank. This pressing is, in itself, a severe test upon the 
 metal, and any flaws, such as seams or cracks, are sure to be exposed, 
 although they seldom occur. When such flaws do develop, the blank is 
 scrapped at this press. The perfect blanks are now placed in a second 
 reheating furnace, where their temperature is equalized and brought again 
 to the forging point. When these conditions have been attained, the blank 
 is cleaned of scale and placed in a second press, in which the top die con- 
 forms to the shape of the inside face of the wheel. Before applying the 
 pressure, a little fine coal is thrown on the bottom die and upon the blank 
 to prevent the dies sticking and to keep their surfaces smooth. After this 
 forging, the blank is placed in a third press, where the bore is punched. 
 
500 THE ROLLING OF STEEL 
 
 During the punching, the hub of the wheel is supported in neatly fitting 
 dies in order to avoid forcing this part of the wheel out of shape. 
 
 In the Second Method of Forging, the press is provided with the two 
 top dies mentioned in connection with the two presses used in the first 
 method. These dies are mounted upon a sliding frame in such a manner 
 that either may be brought at will beneath the piston of the press, thereby 
 dispensing with the first forging press and permitting the forging to be 
 accomplished in one operation when the conditions are favorable. Thus,. 
 if the blank is at a temperature sufficiently high and is evenly heated through- 
 out, the second top die, conforming to the inside of the wheel, is brought over 
 the blank, and the forging is completed in a single stage. If the conditions 
 are such as are likely to cause an uneven flow of the metal, which results 
 if the blanks are unevenly heated, the plain die is used first, then the inside 
 die is moved into position and the pressure applied, thus forming the blank 
 in two stages, but on a single heating. The bore is then immediately 
 punched as in the first method. After the forging, by whichever method 
 that may have been used, and the punching of the bore, the blank is placed 
 in a reheating furnace where it remains until it has reached the proper 
 temperature for rolling. 
 
 The Rolling Mill : The two older mills are very similar in every detail 
 of their construction, but in the number three mill, which made its first 
 trial rolling June 5th, 1917, a few changes looking toward an improvement 
 in construction over the older mills have been made. For this reason this 
 mill, rather than either of the older ones, will be described. This descrip- 
 tion is, however, made rather brief, for the mill, itself, is a somewhat 
 complicated piece of machinery, as the reader will surmise when he is told 
 that seven rolls play upon the wheel at one time during the rolling. These 
 rolls consist of one tread roll, two web rolls, and four (2 sets) rim rolls, 
 and are supported, together with all their bearings, pinions, or gears, 
 adjusting screws, levers, etc., in one pair of horizontal housings, which are 
 large steel castings and placed one above the other. The bottom housing 
 lies directly upon the mill foundation and forms the support for the rolls 
 and for the top housing some four feet above it. The housings are held 
 apart by suitable pillars or posts and are bound firmly together by means 
 of immense bolts. Between these housings the rolls are located; they may 
 be described as follows: The largest roll is the tread and flange roll. In 
 form it resembles a wheel, some thirty-three inches in diameter, and is so 
 located back near the central point of the housings that, during the rolling, 
 it revolves in the same vertical plane as the wheel and bears upon its tread 
 from the rear. Its face is somewhat wider than the rim of the wheel and 
 is grooved to correspond to the 'tread and flange. It is friction driven 
 and is carried on a slide bearing, so that, by means of a screw, connecting 
 the sliding box to a fixture at the rear of the housings and operated through 
 a worm shaft and gear by means of a 15 h. p. electric motor located on 
 
CIRCULAR SHAPES 
 
 501 
 
 FIG. 100. Drawing of Schoen Mill Showing Wheel with Web, Tread, and One Set 
 Rim Rolls in Position at End of the Rolling Operation. 
 
502 THE ROLLING OF STEEL 
 
 top of the upper housing, this roll may be moved backward or forward at 
 the will of the operator. Few operators, however, move this roll after the mill 
 is once adjusted to roll the wheels of a given size and type. On the opposite 
 sides of the tread roll are located the two web rolls. They are about three 
 feet in length, lie in a horizontal position and extend inward, so that their 
 center lines form angles of nearly 30 with the center line of the mill and 
 intersect at a point near the center of the wheel that is being rolled. On 
 their front ends they carry the rolling heads, or surfaces, which conform 
 to the shape of the wheel beneath the rim, while their rear ends are anchored 
 in rotating coupling boxes. Light steel spindles, some five feet in length 
 and provided with proper wobblers, connect these couplings to the two 
 bevel gears, one of which stands on each side of the mill at the rear. These 
 gears mesh into similar gears mounted on the driving shaft of a 750 h. p. 
 D. C. motor (500 volts, 1100 amperes, 130 r. p. m.), which, located at the 
 rear and on the center line of the mill, is used to drive these rolls. Just 
 back of the rolling heads, these rolls are supported in sliding bearings 
 which permit of their being spread as desired. The pressure for rolling 
 is transmitted to these bearings through radial levers, the long arms of 
 which are each attached above the housings to the same screw, so that the 
 same motion, but opposite in direction, and equal pressures are imparted to 
 the two rolls at the same time. This screw, which corresponds to the 
 adjusting screws on ordinary mills, is actuated by means of a 15 h. p. direct 
 current motor (220 volts, 64 amperes, 550 r. p. m.). By this means, the 
 power of the motor is multiplied many times and is capable, at its maximum, 
 of exerting such pressure on the web rolls as to stall the mill. As to the 
 relative altitude of these three rolls, they are so placed that their lines of 
 contact with the wheel in rolling and the axis of the wheel, itself, all lie in 
 the same horizontal plane. The four rim rolls, which are friction driven, 
 are located, one pair above and one pair below, the web rolls, so that all the 
 rolls lie within an arc of 180 of the circumference of the wheel being rolled. 
 These rolls are approximately twelve inches in length and nine inches in 
 diameter, and are so placed that the projected axes of rotation of the two 
 on either side of the wheel intersect at the axis of rotation of the wheel. 
 They are mounted upon sliding frames attached to the front of the mill 
 housings. These four frames are moved by horizontal screws connected 
 by vertical worm shafts and gears to a common shaft, which extends in 
 front of and beneath the housings and is operated by an electric motor set 
 some eight or ten feet to the right of the mill, measuring from the side of 
 the housings. In this way the spread of these rolls is made uniform. How- 
 ever, the bottom set of rim rolls, due to the manner of rolling, do nearly 
 all the work. An indicator, mounted on the upper horizontal screw attached 
 to the sliding frame on the right side of the mill, is in plain view of the 
 operator, who is able, by this means, to read the spread of the rolls and thus 
 control the width of the rim. These rolls may be so formed that they will 
 roll the sides of the rim at a slight angle to the vertical, so that these surfaces 
 
CIRCULAR SHAPES 503 
 
 will lie in parallel planes after the dishing, or coning, process. Two shelves 
 attached to the housings in front of the tread and web rolls and separated 
 by a space a little greater than the thickness of the wheel at the hub, gives 
 a support for the wheel, which is mounted on a loosely fitting mandrel 
 during the rolling. This mandrel is provided with removable bearings, 
 which rest, unattached, upon the shelves, thus leaving the wheel free to 
 move forward after the rolls have gripped it. 
 
 The Rolling Process: After the forged blank has attained the proper 
 temperature for rolling, it is removed from the furnace and carried to the 
 mill with a charging and drawing machine, where it is gripped beneath 
 the rim by tongs suspended by a small jib crane standing on the housing 
 above the rolls. The wheel, held vertically by the crane, is guided between 
 the two supporting shelves, and the mandrel is inserted through the punched 
 bore. The crane tongs are then released, and the wheel, resting on the 
 mandrel, is pushed back by hand against the tread roll and into the position 
 for rolling. The web and rim rolls are then brought to bear on the wheel, 
 the latter rather lightly at first. The large driving motor is started, and 
 the wheel is made to revolve by the action of the web rolls upon it. These 
 rolls, working upon both sides of the web and the under side of the rim, 
 force the metal back into the groove of the tread roll with considerable 
 pressure, until this part of the wheel has reached the dimensions for which 
 the mill is set, or the diameter desired, while the spread of the metal and 
 the width of the rim is controlled by pressure applied to the four rim rolls. 
 The diameter of the wheel is ascertained by means of a gauge, or caliper, 
 one end of which is attached to the tread roll housing.so that it is moved simul- 
 taneously with this roll, in the same direction and through the same space. 
 The other end of the caliper projects in front of the mill, and is provided 
 with a hinged arm or pointer, so that it may be raised out of the way for 
 inserting the blank or removing the wheel. The end of this pointer is 
 curved toward the mill at right angles to its length. At the beginning of 
 the rolling, the roller lowers this pointer to rest on the left hand shelf, in 
 which position its curved end extends toward the tread roll and is opposite 
 its line of contact with the wheel, the pointer having been adjusted so that 
 the distance between its point and the tread roll is equal to the diameter 
 of the wheel desired. With the rolling, the wheel increases in diameter 
 and moves forward on the loose sliding mandrel until a circle on the center 
 of its tread comes in contact with this pointer, when the roller stops the 
 mill and spreads the rolls for the release of the wheel. During the rolling, 
 jets of water are directed against the surfaces being rolled to remove the 
 scale and give a smooth finish to the wheel. In addition to the water, 
 a salt jet i.3 also directed against the tread. The actual rolling process 
 requires about one minute, so that the maximum capacity of the mill is 
 more than 500 wheels per day of twenty-four hours. However, on account 
 of the care exercised to assure a high quality of product, these mills are 
 operated at only 50 to 60% of their capacity. 
 
504 THE ROLLING OF STEEL 
 
 Effect of the Rolling : It will be observed that all the work of the rolling 
 is concentrated upon the outer part of the web and the rim, where the 
 additional refinement due to rolling is most needed. This refinement is 
 very marked, as is shown by Brinell tests on sections of the wheel and by 
 the visible difference in the structure of the metal between the hub and 
 rim. This effect is most marked on the tread, where the hardness of the 
 metal and closeness of grain can, no doubt, be considerably increased by 
 rolling at low temperature or by chilling the metal by using an increased 
 amount of water during the rolling. However, as such practice is likely 
 to cause spalling, it is not employed by the operators. As machining the 
 tread removes much of this super-refined metal, it would appear that the 
 wheel rolled to a finish should be far superior in wearing properties to the 
 machined wheel, on first run, at least. Evidence of this fact is Seen in the 
 increasing demand for rolled to finish wheels for passenger cars, even where 
 formerly only machined treads were used. The mill practice on rolled to 
 finish wheels is high, but a greater or less number of the wheels require 
 machining in order to eliminate slight surface defects or true up the dimen- 
 sions. 
 
 Punching Web Holes and Coning: After the rolling, the wheel is 
 taken on a buggy to a small press, where the web holes are punched, when 
 these are required. This press is fitted to punch either two or four holes, 
 one and three-fourths inches in diameter, and equally spaced on radii of 
 8;Mj, 9*4, 10, 11, or 12 inches, which are standard radii for all the different 
 sizes of wheels. From the punching press, or from the rolling mill, if web 
 holes are not required, the wheel is taken to the coning press, being hot 
 stamped in transit with the word Carnegie on the inner surface of the web. 
 This press is provided with dies which conform to the exact contour of the 
 finished wheel, the top die corresponding to the inside of the wheel. For 
 preserving the rotundity of the wheel, the bottom die is surrounded with 
 a series of tread blocks in the form of segments of a circle, while the top 
 die is similarly provided with a tapered ring to fit over these segments. 
 Thus, when the dies are brought together for coning, this ring slips over 
 the outside of the segments and forces them firmly against the tread while 
 the coning or dishing is being effected. As these blocks leave slight im- 
 pressions on the tread where adjoining blocks meet, the wheel is turned 
 through an arc of a few degrees and again subjected to the pressure, which 
 removes all but traces of these marks, except in occasional cases where 
 they are unusually deep. Upon removal from this press, the wheel is hot 
 stamped with a mill serial number, the heat number and the date, then 
 it is passed to the cooling bed. 
 
 Inspection of Carnegie Schoen Wheels is very rigid. When cold, the 
 wheel is rolled to the inspection platform for its initial inspection. This 
 inspection covers surface defects, location of the hub, rotundity of tread, 
 and the size, which is measured in Carnegie Standard tape sizes. These 
 
CIRCULAR SHAPES 
 
 505 
 
 A. The Blank. 
 
 B. Blank after First Forging. 
 
 C. Blank after Second Forging. 
 
 D. Wheel, after Punching, Rolling and Coning. 
 FIG. 101. Sketches Illustrating the Manufacture of Car Wheels by the Carnegie-Schoen Method. 
 
506 THE ROLLING OF STEEL 
 
 tapes are graduated in eighth's of an inch, beginning with seven feet for a 
 zero mark. The surface defects consist principally of over-fills, under-fills, 
 slivers, scale pits, and block marks, and as they are seldom deep, they may 
 be removed by machining. The tape size and all defects are plainly marked 
 on each wheel, the former with a stencil, the latter in chalk. After this 
 preliminary inspection, the wheels are machined as required to meet the 
 specifications or remove the defects. On rolled to finish wheels, the machin- 
 ing consists of rough boring and facing of the hub and cutting in the limit 
 of wear circle on the outside of the rim. The wheels are then rolled back 
 to the platforms for final inspection, which is even more rigid than the first. 
 In this inspection the wheels are tested for size, eccentricity and size of 
 bore, position and size of hub, thickness and height of flange, radius of 
 throat, thickness of rim, coning, rotundity, and soundness. After being 
 re-stenciled with tape size and marks requested by the customer, such wheels 
 as come within the allowable tolerances are mated and sent to the shipping 
 platform. 
 
 Heat Treating Car Wheels: Heat treating is a recent innovation in 
 the manufacture of car wheels, and may be said to be still in the experi- 
 mental stage. Owing to the irregular section of the wheel, quenching is a 
 difficult process, because, if the entire wheel is quenched, the uneven cooling 
 of the heavy and light parts set up stresses in the shape that result in the 
 destruction of the wheel. In order to overcome this tendency, the Carnegie 
 Steel Company's research department has developed a method whereby the 
 rim only is quenched, after which the wheel, before the hub and web have 
 cooled, is given a drawback at a suitable temperature under the lower 
 critical range. If the hub and web are cooled in air after the quenching of 
 the rim, the wheels show a dangerous tendency to crack by this method, also, 
 hence the quick draw back. From an economical point of view, it would 
 appear that the cheapest plan would be to quench the wheel on the rolling 
 heat, but on account of unavoidable variations in the finishing temperature 
 this treatment was found to be unsatisfactory. The wheels, therefore, are 
 allowed to become cold after rolling and coning, when they are reheated 
 above their critical range before quenching. This process is accomplished 
 in a rectangular tank provided with rollers grooved to conform to the 
 tread and flange and so placed in the tank that they support the wheel 
 in a vertical position transverse to the tank, which contains enough of the 
 quenching fluid to cover the rim only. With the rollers, which are mounted 
 on a shaft connected to a motor, revolving, the wheel, at the proper tem- 
 perature, is placed in position on the rollers, which immediately start the 
 wheel revolving, or spinning, also. The spinning is continued until the rim 
 becomes sufficiently cooled when it is withdrawn and immediately given a 
 draw back, as stated above. The process adds considerable to the cost of 
 the wheel, and though there are many wheels in service thus treated, and 
 apparently with promising results, sufficient time has not yet elapsed to 
 determine just to what extent the wheels are improved by the treatment. 
 
CIRCULAR SHAPES 507 
 
 The Forging of Circular Shapes : As previously indicated, only those 
 circular shapes which are more than thirty inches in diameter can be 
 finished by rolling on the Schoen mill. However, this plant, which 
 represents the circular shape department of the Homestead works, produces 
 a great number of smaller circular shapes by forging only. These smaller 
 shapes include such articles as wheels for low type street cars; double 
 flanged crane track wheels; automobile fly wheels; turbine discs; shaft 
 couplings; pipe flanges; pistons for locomotives; gear rings; and gear blanks 
 for automobiles, farm tractors, turbo generators, street cars, etc. In addi- 
 tion to these, a miscellaneous lot of circular shapes ranging in form from 
 the most intricate sections to plain discs, and in sizes from twenty-five 
 pounds to five hundred and even a thousand pounds are produced. For 
 forming all these shapes the same powerful presses are employed as are 
 used in preparing the rolled wheel blanks, and in general the methods of 
 forging are similar, with the exception that additional precautions in the 
 removal of scale before forging are observed. In this connection it should 
 be noted that forging will not produce the smooth finish obtained by rolling 
 on the Schoen mill, and all forged articles requiring a perfectly smooth 
 surface must be machined to finish. 
 
508 FORGING OF AXLES, SHAFTS, ETC. 
 
 CHAPTER XI. 
 
 FORGING OF AXLES, SHAFTS AND OTHER ROUND SHAPES. 
 
 SECTION I. 
 
 HOWARD AXLE WORKS AS AN EXAMPLE OF A FORGING SHOP. 
 
 The Plant and Its Equipment: Aside from the forging of armor 
 plate and other articles required by our government, small wheels and a 
 miscellaneous lot of shapes for its own use, the Carnegie Steel Company has 
 restricted its market activity in the forging line to the manufacture of 
 axles, shafts and similar heavy products. For these products the company 
 operates a plant especially designed for the work, known as the Howard 
 Axle Works, which may be taken here as an example of a modern forging 
 plant. The essential equipment of the plant includes three continuous coal 
 fired furnaces for heating the blooms, a twenty-four inch roughing mill of 
 two stands of rolls in tandem, ten 7000 pound and two 7500 pound double 
 acting steam forging hammers, three gag press straighteners, thirty double 
 cutting-off and centering machines, twenty-seven rough turning lathes, two 
 finishing lathes, one boring lathe, two hollow drill machines, and a complete 
 heat treating plant that will be described more in detail later. The forging 
 limits of the plant as to size is as follows: Maximum weight, 2500 pounds; 
 maximum length, ten feet; maximum diameter, twelve inches; minimum 
 diameter, three inches. As to arrangement, the layout of the plant provides 
 for the most economical handling of the materials. The blooms start in 
 at one end of the plant and continue in one direction, progressing step by 
 step through the various operations, until, upon arrival at the other end of 
 the plant, they are in a form ready for shipment. 
 
 Precautions to be Observed in the Manufacture of Axles: As the 
 
 failure of an axle in service usually results with serious consequences, great 
 responsibility rests upon the manufacturer at all times to see that each 
 and every axle is as nearly perfect as it is possible to make it. Before 
 describing the processes of manufacture, it may be well to point out 
 some of the things that may cause axles to fail, because the thing aimed 
 at in developing a method of manufacture is the elimination of as many 
 of the causes of failure as possible. There are many of these causes of 
 failure, according to some writers upon the subject, but the majority may 
 be traced to the following, which are to be looked upon as the chief sources 
 of danger: 1. Pipe; 2. Segregation; 3. Unequal or improper heating; 
 4. Slag inclusions; 5. Forging strains; 6. Incipient cracks. From this 
 list it is seen at a glance that some of these sources of danger are very 
 
INSPECTING AND HEATING BLOOMS 509 
 
 difficult to eliminate and that the making of a good axle must begin with 
 the making of the steel, itself. The other defects may be overcome by 
 proper attention to details during the processes of rolling and forging the 
 steel. Starting with the steel after it has been rolled into blooms, which 
 must correspond in dimensions and weight to the size of the axles it is 
 intended for, the various steps in the process of manufacture at these works 
 are as follows: 
 
 Inspection of the Blooms: Located at Homestead, Pa., the Howard 
 Works receives its steel from the Homestead Steel Works at Munhall. 
 Here, before the steel is shipped to the axle works, the blooms are subjected 
 to a rigid inspection. Those blooms that show any signs of pipe or 
 insufficient discard at the blooming mill shears are rejected. Surface 
 defects, such as seams, slivers and surface cracks, are carefully chipped 
 out, and those blooms in which the defects extend beyond certain deptns, 
 or occur on the part that corresponds to the wheel seat are also discarded. 
 Such blooms as pass the inspection are shipped to the axle works, where 
 they are stored under cover until needed. 
 
 Heating the Blooms: The proper heating of the blooms for forging 
 requires that they be uniformly heated throughout and be brought grad- 
 ually to the forging temperature, which should be kept as low as possible 
 and yet permit the work to be done. The advantages of a low finishing 
 temperature in securing maximum grain refinement is readily understood. 
 The importance of heating slowly is also realized, when it is pointed out 
 that rapid heating may cause the outside of the bloom, which is first to 
 rise in temperature, to expand away from the more slowly heating core and 
 thus cause ruptures that may not be welded up by the action of the hammers. 
 Slow heating gives the heat a chance to "soak" into the bloom, thus giving 
 that uniformity in temperature from center to surface so necessary to secure 
 a finished forging of the best quality. As to the proper temperature, the 
 manufacturer has always had to depend upon the eye and judgment of the 
 trained heater in the past, and must continue to do so to a great extent in 
 the future. The use of pyrometers does not replace this human element, 
 because the pyrometer records the temperature of the furnace and not of 
 the steel, particularly in the case of the continuous furnace. The rate of 
 heating is fixed by the type of furnace. At these works, therefore, the 
 continuous furnace is used because this type heats up the steel very grad- 
 ually. The bloom is placed in the furnace at the cold end and is slowly 
 pushed toward the hot end, so that it reaches a full forging temperature 
 only a short time before it is drawn from the furnace. 
 
 The Rolling and Forging Operation: Having been brought to the 
 proper temperature for forging, the blooms, within a certain range of sizes, 
 are pushed out of the hot end of the heating furnaces upon a conveyor, 
 which serves all three furnaces, and are carried by it to the rolling mill. 
 
510 FORGING OF AXLES, SHAFTS, ETC. 
 
 This mill consists of two stands of rolls in tandem, as previously stated. 
 Each stand is provided with four passes cut to take four different sizes of 
 square blooms. These sizes are 6*/, 7}/, 8 and 8^ inches. The passes 
 are shaped to round off the corners of the bloom, to secure which result 
 is the main object in the use of the mill. The reduction in cross sectional 
 area due to the rolling varies from 33/% to 5%. From the mill, a roll 
 table distributes the blooms to the hammers, which are arranged in two 
 rows, one on each side of the table. Adjustable deflecting rails built in 
 the side guards of the table serve to divert the blooms to small receiving 
 tables, of which there is one for each hammer, and leave them in positions 
 to be most conveniently grasped by the hammer tongs, which are suspended 
 from cranes. The tongs having been quickly clamped on, the bloom is 
 swung around between the forming dies of the hammer. These dies are 
 provided, when desired, with two or more grooves; one, the plain groove 
 used to do the greater part of the forging, is located directly under the 
 piston rod, while the other grooves, used to form special sections, such as 
 the journals, are placed beside the plain groove. The forging is begun at 
 the middle of the bloom, which is rapidly reduced by heavy blows of the 
 hammer, the forging progressing toward the free end of the bloom. Here, 
 by the special grooves in the die, the journal or other special section is 
 formed by a few strokes, when the piece is again placed in the plain groove, 
 and the forging is smoothed up and brought, by light strokes of the hammer, 
 to correct diameter, which is determined by caliper. The tup is then 
 brought to rest upon the axle, which is held between the dies while tho 
 tongs are released, and those on the opposite side of the hammer are made 
 fast to the finished end. The other end of the axle is then forged down 
 like the first, except that, in addition to diameter, the length is also fixed. 
 The crane is then swung around, and the axle is placed on the cooling bed, 
 where it is supported about three feet above the floor by two rails, which 
 arrangement allows it to be cooled uniformly by the air. The average 
 reduction in cross sectional area under the hammer is about 50%. Forgings 
 requiring blooms larger than eight and one-half inches are reduced entirely 
 by hammer. Two crews, each made up of a hammerman, who has charge 
 of the forging, and three helpers, and one hammer driver, are assigned to 
 each hammer. The crews work alternately, each crew completing one axle 
 at a time. 
 
 Advantages of the Method: Aside from the increased tonnage made 
 possible by the rapidity of the work, the method of forging employed at 
 Howard presents many advantages which bear directly upon the quality 
 of the product. The rolling mill, which accomplishes only a small fraction 
 of the total work done upon the axle, is a great help to the hammers. By 
 rounding off the corners of the bloom, it practically eliminates all danger 
 of forming hammer laps, and permits the forging to be accomplished in the 
 shortest time possible. Hence a low initial temperature can be used for 
 forging, and the work can be completed at a more uniform temperature. 
 
FINISHING PROCESSES 511 
 
 This uniformity of the finishing temperature is a very noticeable feature 
 at Howard. Thus, in observing closely various axles at different stages 
 of the forging operation the eye can detect little difference in temperature 
 between those axles on which the forging has just begun and those that 
 are being finished. That this rapid method of forging on one heat is far 
 superior to the old method of forging on two heats is apparent, because 
 it not only promotes greater uniformity in individual axles but eliminates, 
 to a far greater degree, the variation in different axles. 
 
 SECTION II. 
 
 FINISHING PROCESSES FOR FORGINGS. 
 
 Straightening: Except in the case of heat treated axles, and driving 
 and trailing axles, the next step after the forging is the straightening, which 
 is accomplished by means of gag presses. From the cooling beds the 
 forgings are carried forward by over-head cranes to similar beds in front 
 of the presses. Here each axle is inspected for straightness, and those 
 that require it are straightened. Heat treated axles are straightened after 
 being treated, but driving and trailing axles are too large to be straightened 
 by the gag press. 
 
 Cutting=off and Centering : After passing the inspection for straight- 
 ness, the forgings are moved forward by overhead cranes and distributed 
 to the cutting-off lathes. These lathes are double combination cutting-off 
 and centering machines, and are designed to work on both ends of the forging 
 at the same time. Upon being inserted in this machine, the forging is 
 grasped at the wheel seats by adjustable revolving centering clamps, which 
 
 hold it firmly to the central axis of ro- 
 tation, while two cutting tools, placed 
 one at each end and adjusted to the 
 correct length, are brought to bear 
 and cut off the excess metal at the 
 ends. In this cutting, a tolerance of 
 one-eighth inch over length and noth- 
 ing under is permitted. When these 
 tools have cut to within about one- 
 half inch of the center, they are run 
 back out of the way, the pieces of ex- 
 cess metal are detached with a sledge, 
 and with the forging still held by 
 the centering clamps, the revolving 
 centering tools are brought to bear 
 at each end. These tools are shaped 
 FIG. 102. Carnegie Standard Centering to cut a 60 degree cone-shaped cen- 
 tering hole, five-eighths inch in depth, 
 one and one-eighth inch in diameter at the top, and with a clearance hole 
 
512 FORGING OF AXLES, SHAFTS, ETC. 
 
 for points at the bottom one-half inch in depth and three-eighths inch in 
 diameter. When axles are ordered to be smooth forged only, the operation 
 of cutting off and centering completes the work done by the mill. On such 
 axles some excess stock is necessarily left on those parts that are to be 
 finished later. This allowance on car axles is generally one-half to three- 
 fourths inch over the finished diameters of the end collars, journals, and 
 dust guards, and one-fourth to three-eighths inch on wheelseats. 
 
 Rough Turning: On account of the saving that can be effected in 
 handling and transportation of excess weight, it is a decided advantage to 
 both the customer and the shop, especially to the latter, that all rough 
 turning be done before shipment is made, as it is only by rough turning 
 that certain flaws can be detected. Rough turned material falls into two 
 classes, known as "rough turned on journals and wheelseats,'* and "rough 
 turned all over." Axles of the first class are put into service with the 
 center portions between the wheelseats smooth forged to size. In the 
 case of axles rough turned all over, the center portions are forged slightly 
 over size to provide for the metal removed in turning to size. In the case 
 of car axles or other axles with a tapered body, this metal is removed at 
 the same time (or after) the journals and wheelseats are rough turned, 
 in a special lathe provided with two tools controlled by a former-bar whose 
 contour is the same as the middle portion of the axle. In finishing rough 
 turned axles, the wheelseats are finish-turned only, while the dust guards, 
 journals and collars are finish-turned and burnished, and in order to provide 
 the excess metal required for this work, these parts are rough turned one- 
 eighth inch over size on their diameters. 
 
 Hollow Boring: Owing to the many apparent advantages arising 
 therefrom, the practice of boring large axles and forgings longitudinally 
 through the center is being advocated more and more strongly. These 
 advantages are briefly discussed under the following headings: 
 
 1. Piping, it will be recalled, was given as one of the causes of failure. 
 While the Carnegie Steel Company, by a generous discard and close 
 inspection, aims to eliminate this defect, yet it is possible that some forms 
 of piping, notably compound pipes, may escape both the discard and the 
 inspection, and remain in the axle as a menace to safety. Hollow boring 
 gives the inspectors a chance to detect this hidden pipe. 
 
 2. Segregation was given as another source of failure. This defect 
 cannot be entirely overcome in the manufacture of steel, and inspection is 
 no safeguard against it. But as the area of greatest segregation lies about 
 the central axis, boring a hole of proper size longitudinally through the 
 center should, and does, remove the greater part of all the segregated 
 material from the axle. 
 
 3. Strength and Weight: The central portion of an axle removed 
 by boring is really a non-essential part so far as strength is concerned. 
 
HEAT TREATING FORCINGS 513 
 
 The transverse strength of rounds is proportional to the cubes of their 
 diameters. So, for example, if a three inch bore be made in a six inch axle, 
 the maximum loss in strength is but 12.5%; in an eight inch axle, but 5.25%. 
 These figures represent the loss in strength provided the center is as strong 
 as the outer portion, which condition is never true in axles or similar for gings, 
 so that the actual loss in strength in nearly every case would be much less 
 than these figures indicate. Again, the axle with the bored center may 
 actually be stronger than it would have been solid, provided it contained 
 much segregated material or the remnant of a pipe conditions that favor 
 the formation of internal cracks. Another factor concerns the relation 
 between the loss in strength and the loss in weight. In this connection 
 it will be observed that, whereas the strength varies as the cube of the 
 diameters, the weight varies as the squares. Referring to the example just 
 cited, ,and applying this law, the reader will find that while in the case 
 of the six and eight inch axles, the three inch boring gives a loss in strength 
 of 12.5% and 5.25%, respectively, the loss in weight is 25% for the first 
 and 14.3% for the second. 
 
 4. Hollow Boring and Heat Treating: As an aid in heat treating, 
 especially in quenching and tempering, or toughening, hollow boring is of 
 great importance. In heating, it permits the heat to be absorbed much 
 more rapidly, and in quenching, the heat is more rapidly removed than 
 in solid pieces, with the result that the structure is more uniform. Further- 
 more, contraction and expansion strains are largely overcome, and shrinkage 
 cavities in the center are avoided. The American Society for Testing 
 Materials specify that all forgings over seven inches in diameter that are 
 to be quenched shall be bored. The diameter of the hole bored should 
 equal or exceed 20% of the largest diameter of the forging exclusive of 
 collars or flanges. Howard Axle Works are equipped to bore holes either 
 two or three inches in diameter. 
 
 The Heat Treating Plant is housed in the same building with the 
 hammers and lathes and consists of two furnaces for heating with the 
 forgings in a horizontal position, one furnace for heating the material in 
 a vertical position, one water quenching tank, one oil quenching tank, and 
 all the necessary supplemental equipment for handling and testing the 
 material. 
 
 The Furnaces: The inside working space of the two furnaces of the 
 first type are each twenty-four feet in length and nine feet in width, and are 
 designed to heat uniformly to a height of about four feet above the bottom. 
 They are provided with removable bottoms of the car type, which much 
 facilitates the charging and drawing operations. This bottom is moved 
 into and out of the furnace by means of a toothed rack attached to the 
 bottom of the car and a stationary pinion actuated by an electric motor, the 
 car itself resting on rollers that move over a double track. The doors of the 
 furnace are of the vertically lifting type, and are hydraulically operated. 
 
514 FORGING OF AXLES, SHAFTS, ETC. 
 
 These features, together with the close proximity of the quenching tanks, 
 permit the drawing and quenching of a charge in the quickest possible time, 
 less than a minute being required to transfer a charge from the closed furnace 
 to either of the quenching tanks. The measures taken to secure uniform 
 heating are particularly noticeable in this furnace. The furnace is of the 
 reversing flame type, in which natural gas is employed as fuel, and is heated 
 by means of burners placed at space intervals of less than two feet along 
 each side, thus permitting the temperature at any point in the furnace to 
 be controlled to a nicety. At the top, the furnace is closed with a roof, 
 arched from side to side, while, inside, high bridge walls extend along in 
 front of the gas burners to prevent the flames from impinging upon the 
 charge. In order that the entire surface of the material may be exposed 
 to heat of the same intensity, the charge is supported at a height of about 
 eighteen inches above the floor of the car bottom by two steel rails that 
 extend the entire length of the car. These rails are spaced about four feet 
 apart and are supported by castings in the form of four-legged stools. The 
 floor of the car bottom is constructed of brick laid upon a bottom of steel 
 plates, and is of such thickness as to give ample insulation from the heat of 
 the furnace. The bottom is made to fit the furnace neatly, and the escape 
 of hot gases from the heating chamber is prevented by means of sand seals. 
 The construction of the furnace for heating the charge in a vertical position 
 is somewhat like that of a soaking pit. It has a capacity of about six axles 
 and sufficient head room for maximum lengths of ten feet. In operating 
 this type of furnace, the axles are loaded on a cast steel rack, which is 
 specially designed to support them in a vertical position, and are lowered 
 through the top into the furnace where they are maintained in a vertical 
 position throughout the heating operation. This furnace is seldom used, 
 as more satisfactory operating conditions are obtained by using the other 
 type. For taking temperatures the Siemens Water pyrometer is used 
 exclusively. 
 
 The Quenching Tanks: For use in connection with these heating 
 furnaces, the plant is equipped with one water quenching and one oil quench- 
 ing tank. These tanks are both placed as near as possible to the furnaces, 
 the water tank being directly in front of one of the furnaces of the horizontal- 
 heating type. This tank, approximately twenty-five feet long, twelve feet 
 wide and fourteen feet deep, is of the sub level type and is constructed of 
 concrete. When in use, the water level lies about two feet above the floor 
 of the shop. So, an ample volume of water is supplied for any charge it is 
 practicable to handle, and, in addition, provision is made whereby fresh 
 water may be introduced during the quenching operation at one corner of 
 the tank and the excess conducted away at the diagonally opposite corner, 
 both inlet and outlet being located near the top of the tank. Two beams 
 extending the full length of the tank and supported about two feet above 
 the bottom, prevent the charge from resting on the bottom when lowered 
 by the crane, thus securing more uniform cooling. The oil quenching tank 
 
HEAT TREATING FORGINGS 515 
 
 is some sixteen feet in length, nine feet in width, and ten feet in depth, 
 inside. It is made in two parts an oil container and a cooling jacket. 
 The container is made of steel plates and is set within the cooler, which is 
 constructed of concrete and is about twenty inches longer and wider than 
 the container, so that a space of about ten inches separates the walls of 
 the two vessels. This space is kept filled with cold water, which serves 
 to prevent the temperature of the oil rising too high during quenching, and 
 to cool it down rapidly after each charge. 
 
 The Testing Equipment includes all the latest devices for testing 
 materials. In the shop, two hollow drill machines for cutting out tests are 
 provided, and as all heat treated axles are given individual shock tests, 
 a drop testing machine for giving these proof tests is also located here. 
 Two drop testing machines, adapted for testing axles in accordance with 
 standard specifications are provided. Other physical tests are made in the 
 physical laboratory, which is equipped with one planing, one turning, one 
 pulling, one torsional, one bending and one Brinell machine, and all the 
 supplemental appliances for accurate testing. 
 
 Advantages of Heat Treating Axles: While the proper heat treating 
 of axles is accomplished with some difficulty on account of their size, and 
 is attended with great danger if improperly done, yet with proper equip- 
 ment, great care and good judgment, born of knowledge and experience on 
 the part of the operator, the dangers may be eliminated, and decided 
 advantages result therefrom. It is the only way in which the grain struc- 
 ture can be refined and made uniform, and in doing this all the evils due 
 to variations in the grain, which result from the heating and working of 
 the bloom, are overcome, as well as forging strains. But greatest of all 
 these advantages is the improvement in mechanical properties effected 
 through correct heat treatment. It offers the only positive means of 
 markedly increasing the strength and wearing properties of axles without 
 in any way increasing their weight a thing that is much desired under 
 modern conditions of traffic. 
 
516 CONSTITUTION OF STEEL 
 
 PART III. 
 
 THE CONSTITUTION HEAT TREATMENT AND COMPOSITION 
 
 OF STEEL. 
 
 Introductory: It is the desire in this part of this little book to center 
 the interest of the reader chiefly about the heat treatment of steel. So 
 much progress in the study of this subject has been made in recent years 
 that many are inclined to look upon it as something new. That remarkable 
 changes in the physical properties of a given steel can be brought about 
 through the agency of heat alone has been known for many years, but 
 until 1390 the subject had received very little attention from scientists. 
 Up to that time both the scientific knowledge about the subject and the 
 technical application of the art of heat treatment were very limited, being 
 confined for the most part to the making of tools and a few specialties. 
 The invention of the automobile, the aeroplane, and other machines, the 
 different parts of which are required to be light and at the same time suit- 
 able for the usages to which the parts are subjected, gave rise to demands 
 for steel of great strength combined with various other specific properties . 
 These demands directed the attention of investigators to heat treatment 
 because it was found that this was the only means of meeting these require- 
 ments. Practically all alloy steels must be heat treated in some way, and 
 few steels in their natural state will give their full value in service, so that 
 the various combinations of static and dynamic strength and wearing 
 qualities required can be obtained in their highest degree only by adjusting 
 and correlating both the chemical composition and the heat treatment. 
 Just as certain chemical components intensify one set of properties, and 
 others another set, so the heat treatment may be changed to develop 
 different qualities in a similar way. Thus, by combining the proper chemical 
 composition with the proper heat treatment, there results a product posses- 
 sing in the highest degree the properties most desired for the work the 
 steel is to do. So it is evident that the intelligent application of heat 
 treatment to secure the best results requires a thorough knowledge, on the 
 part of those supervising the work, of the composition of the steel and the 
 effect of the various elements that are to be found in all steels or that may 
 be added as alloys to produce the special steels. Again: Heat treatment 
 consists in heating and cooling steel under conditions that will produce 
 the desired change or changes in physical properties, and embraces the 
 three processes of annealing, hardening, and tempering, to which may be 
 
CONSTITUTION OF STEEL 517 
 
 added the special processes known as ' 'process annealing," "patenting," 
 "case hardening," etc. The remarkable changes in properties that may 
 be obtained, together with the phenomena observed during heating and 
 cooling, all connote vital changes that are brought about by the heat 
 treatment. As no change in composition of the metal takes place, the 
 cause for the phenomena must be sought in changes of arrangement or 
 condition of the constituents of the steel itself. Another pre-requisite, 
 then, to the study of heat treatment is the study of the structure and con- 
 stituents of steel, a thorough knowledge of which is essential to any 
 understanding of the subject, whatever. For this reason, the study of heat 
 treatment should be prefaced by a brief summary of the knowledge con- 
 cerning the structure and constitution of steel. 
 
 Before beginning this study, however, the student should understand 
 that part III of this book is intended merely as an introduction to the study 
 of metallography, heat treatment and composition of steel. Those who 
 desire a further knowledge of these valuable and fascinating subjects are 
 referred to such authorities as Albert Sauveur 1 , whose plan of developing 
 the subject is closely followed in this study; H. M. Howe 2 , whose iron 
 carbon diagrams are used herein; and D. K. Bullens 3 , whose practices in 
 heat treatment are frequently referred to. 
 
 iSee the Metallography and Heat Treatment of Iron and Steel, Published by 
 Sauveur and Boylston, Metallurgical Engineers, Cambridge, Mass. 
 
 2 See Iron, Steel and Other Alloys, and The Metallography of Iron and Steel. 
 Both published by McGraw-Hill Book Company, Inc., 239 West 39th Street, New 
 York. 
 
 See Steel and Its Heat Treatment, Published by John Wiley & Sons, Inc.. 
 New York City. _, 
 
518 CONSTITUENTS OF STEEL 
 
 CHAPTER I. 
 
 THE CONSTITUTION AND STRUCTURE OF PLAIN STEEL. 
 
 SECTION I. 
 
 STEEL AS AN ALLOY OF IRON AND CARBON. 
 
 The Constituents of Steel : Steel is not a single element or compound, 
 but a complex artificial product, composed of many elements held in the 
 solid mass as a mechanical mixture of alloys and chemical compounds with 
 the element iron. In ordinary steel, these elements are iron, carbon, 
 manganese, phosphorus, sulphur, silicon and oxygen, with traces of nitrogen, 
 hydrogen and other elements, such as aluminum, copper and arsenic. Of 
 these, all are to be considered as impurities except carbon, which is an 
 essential ingredient, and manganese, or other elements added for a definite 
 purpose. For the sake of simplicity and brevity only pure steel, consisting 
 of the two essential elements, iron and carbon, will be considered in this 
 chapter. Even in this case steel is found to be an aggregate made up of 
 mineral-like components, some of which are visible only with the aid of 
 the microscope after the surface of the specimen has been highly polished 
 and etched with dilute acids or other corrosive mixtures which affect the 
 various constituents in different ways. To the structure thus revealed by 
 the microscope the term micro-structure is given, to distinguish it from the 
 macroscopic structures, or those visible to the naked eye; and to the 
 different constituents mineralogical names have been applied. Thus, in 
 pure steels which have cooled slowly from a high temperature, three dis- 
 tinct constituents are recognized. They are called ferrite, pearlite, and 
 cementite, and in the different steels will vary in amount according to the 
 carbon content. 
 
 Ferrite is the term applied to pure iron, i. e., carbonless iron, when 
 it is considered as a microscopical constituent of steel. It is soft, ductile 
 and relatively weak, having a tensile strength of about 40,000 pounds 
 and an elongation of 40 per cent, in two inches. It has practically no 
 hardening power, a high electric conductivity, and can be magnetized. 
 It appears white in color after being etched with dilute alcoholic nitric or 
 picric acids. It is best seen in steels containing .10% to .30% carbon, 
 ' when it appears as a network surrounding bodies of pearlite, another con- 
 stituent of steel to be described shortly. 
 
 Cementite: As already stated, iron and carbon are the essential 
 elements in steel, and of these carbon may be termed the controlling element. 
 When steels are cooled slowly from high temperatures, from the fusion 
 
THE CONSTITUTION OF STEEL 
 
 519 
 
 1. Photomicrograph 
 showing a grain of pear- 
 lite surrounded wi th free 
 ferrite as a net work. 
 Specimen taken from an- 
 nsaled steel containing 
 carbon .48 per cent, and 
 manganese .54 per cent. 
 Magnified 750 diameters 
 
 Etched first in 5% 
 alcoholic picric acid for 
 eight seconds, then in 
 5% alcoholic nitric acid 
 for five seconds. White 
 area represents ferrite. 
 On account of a too rapid 
 rate of cooling, the pearl- 
 lite is not fully devel- 
 oped. Compare with 
 Figs. 48 and 113. 
 
 2. Photomicrograph 
 of a specimen of steel 
 containing carbon to the 
 extent of 1.50 percent. 
 The excess cementite is 
 here seen in the form of 
 spines, or needles. 
 
 Magnified 100 diam- 
 eters. Etched for eight 
 seconds in five percent, 
 alcoholic picric acid and 
 for two seconds in five 
 per cent, nitric acid, 
 making the free cement- 
 ite, which stands in 
 relief, brilliant white 
 in color. 
 
 Co, 
 
520 CONSTITUTION OF STEEL 
 
 point, for example, all the carbon is found combined with a definite amount 
 of iron in the form of a carbide of iron corresponding to the chemical formula 
 FesC. This compound consists of carbon, 6.67 per cent, and iron, 93.33 
 per cent., and it is known micrographically as cementite. Any excess iron 
 is practically free of carbon at atmospheric temperatures and remains as 
 ferrite in steels that have cooled slowly. Little is known about the 
 properties of cementite except that it is very hard and brittle. Indeed, 
 it is the hardest component of steel, and will scratch glass and feldspar 
 but not quartz. It is about two-thirds as magnetic as pure iron under an 
 exciting current. After polishing the surface of steel, it stands in relief, 
 and is brilliant white after etching with dilute hydrochloric or picric acids. 
 It occurs free in ordinary steels of more than .90% carbon, in which it 
 appears as a network or as spines and needles. It takes its name from 
 cement steel, made by the cementation process, which contains a great 
 deal of this carbide, FesC. 
 
 Pearlite: One of the most remarkable characteristics of cementite 
 and ferrite is their power of forming the conglomerate known as pearlite. 
 During the process of slowly cooling steel from higher temperatures, say 
 above 1000 C., it has been found that the cementite and ferrite always 
 form, at about 700 C., a mechanical mixture made up of definite amounts 
 of each and in the proportion of about seven parts ferrite to one part 
 cementite, so that the resultant conglomerate will contain approximately 
 .90% carbon. This constituent then consists of interstratified layers or bands 
 of ferrite and cementite, and is called pearlite on account of its resemblance 
 to mother of pearl. While pearlite commonly occurs in slowly cooled steels 
 in the lamellar formation, composed of alternate layers of ferrite and 
 cementite, it may under different rates of cooling and dependent on the 
 relative amounts of ferrite and cementite present, exist in other formations, 
 or phases, of which some authorities have recognized at least four, making 
 in all five modifications. Normal pearlite has a maximum tensile strength 
 of about 105,000 pounds, and an elongation of about 10% in two inches. It is 
 regarded as a separate and distinct constituent of steel because it forms 
 distinct masses or "grains," always contains this definite percentage of 
 carbon and is always formed at a definite temperature, or a range of 
 temperatures, to be more exact. 
 
 Manner of Freezing of Solutions and Alloys: In order to clarify 
 the explanation of the formation of pearlite, it is necessary to digress to 
 the extent of explaining some of the freezing laws of solutions. A study 
 of the freezing of solutions has shown that they fall into two classes, namely, 
 those in which the ingredients in solution in the liquid state remain in 
 solution in the solid state and those in which the state of solution is not 
 maintained in the solid state, that is, those in which the ingredients separate 
 on freezing. 
 
 An Example of the First Class of Solutions : One of the best examples 
 of the first kind of solution is a mixture of gold and silver. If quantities of 
 
FREEZING OF ALLOYS 
 
 521 
 
 these two metals be placed in a vessel and heated until they melt, a homo- 
 geneous mixture, or a liquid solution, results; and if this mixture be allowed 
 to cool to the solid state, it is still homogeneous, that is, it is a solid solution. 
 A study of many mixtures in which the proportions of gold to silver are varied 
 shows that freezing begins at a different temperature for each mixture. 
 Pure gold freezes at 1062 C. and pure silver at 961 C., and the freezing 
 points of the mixtures occur between these two points. Unlike the pure metals, 
 however, these mixtures do not solidify completely at a constant temperature, 
 but their freezing is prolonged through ranges of temperature. These facts, 
 definitely determined by experiment, may be represented by a diagram, or 
 curve, such as the following, in which the ordinates represent temperatures 
 and the abscissae the percentage of gold or silver or both. 
 
 Temperatures Degrees Centigrade 
 
 
 Liquid 
 
 Solution 
 
 
 
 Freezing Point 
 of Pure Gold. 
 
 Freezing Point 
 ot Pure Silver. 
 
 
 ; 
 
 Cl 
 
 
 
 "^ 
 
 ~**1 
 
 ^^ 
 
 ^ 
 
 
 
 s' 
 
 "f^^ 
 
 ^!^___ 
 
 ^^^ 
 
 
 
 
 
 
 
 Solid 
 
 Solution 
 
 
 
 %Gold 100 80 60 4 o 20 
 %bilver o 20 40 60 80 100 
 FIG. 104. Diagram of the Freezing of Liquid Gold-Silver Alloys. 
 
 To illustrate further, suppose sixty ounces of gold be mixed with forty 
 of silver, and the whole heated to a temperature of 1090 C. The locus of 
 this point would be at "1" in the region of the liquid state. It now this 
 molten mass be allowed to cool, crystals, each of which contains 60% gold 
 and 40% silver, begin to separate out at "f," about 1041 C., and continue 
 to do so until the point "s," about 990 C., where the last of the liquid 
 freezes, is reached. No further change takes place as the cooling proceeds, 
 and the solid mass is found to be homogeneous and of the same composition 
 as the liquid solution. Any other mixture of these two metals would give a 
 like result, except with respect to the freezing points, and the solid crystals 
 would be found to be made up of gold and silver in the same proportion 
 as they were in the liquid state. 
 
 Example of the Second Class of Solutions Salt and Water: It is 
 
 a well known fact that a solution of common table salt freezes at a lower 
 temperature than pure water. This lowering of the freezing point, or rather 
 the temperature at which freezing begins, varies with the proportion of 
 
522 
 
 THE CONSTITUTION OF STEEL 
 
 salt to water until this proportion has reached the definite limit of 23.5%, 
 when any further addition of salt causes the point at which freezing begins to 
 rise. This lowest temperature, at which the solution containing 23. 5% of salt 
 freezes, is 22 C. These facts are represented by the diagram of Fig. 105. 
 
 Two or three examples 
 will suffice to explain the 
 freezing of solutions contain- 
 ing varying amounts of salt, 
 and any other points about 
 the diagram that may not be 
 clear. Thus, suppose a solu- 
 tion containing 10% of salt is 
 at a temperature indicated by 
 "1". Although water freezes 
 at C., the temperature of 
 this solution must fall to the 
 40 point "f," about 10 C., 
 
 FIG. 105. Diagram Representing the Freezing before freezing begins. Here, 
 of Solutions of Salt; in Water. unlike the solution of gold 
 
 and silver, crystals of pure water begin to separate from the solution. The 
 separation of these crystals has the effect of increasing the percentage of salt 
 in the mother liquor, so that the separation of the water crystals continues 
 only so long as the temperature is being lowered. Furthermore, if the rate 
 of cooling has been uniform down to the point "f," a marked retardation 
 takes place here, because the heat of fusion of the water must be removed 
 before ice can be formed. With the removal of this heat, however, and that 
 necess ary to lower the temperature of the remaining solution, the separa- 
 tion of ice crystals continues, causing a concentration of salt in the mother 
 liquor that bears a definite relation to the temperature aa indicated by the 
 line M O. Finally, when a temperature of 22 C. is reached, the mother 
 liquor, which now contains 23.5% of salt, freezes as rapidly as the heat of 
 fusion is abstracted. When all this liquor has solidified, the temperature 
 of the solid mass will continue to fall uniformly in a manner similar to that 
 before freezing began. 
 
 If instead of the weak solution, a strong brine containing more than 
 23.5% of salt is substituted in the experiment just described, it is found 
 that, just as ice separated along the line M O., salt crystals separate out 
 along the line N O until the temperature 22 C. is reached and the mother 
 liquor contains 23.5% of salt. This liquor then freezes as described before. 
 When these facts were first observed, it was thought that the mother liquor 
 that is the last to freeze was a hydrate of sodium chloride of the formula 
 NaCl.lOH^O. and was called, therefore, the cryohydrate, cold hydrate, 
 meaning a hydrate that could exist in the solid state only at low temper- 
 atures. It has since been shown that these cryohydrates are not chemical 
 compounds, though they have a definite composition, but are mechanical 
 mixtures made up of crystallized salt and ice in intimate contact. 
 
FREEZING OF ALLOYS 
 
 523 
 
 Lead and Tin Solutions as Another Example of the Second Kind 
 of Freezing: Many of the fused alloys exhibit the same phenomena in 
 freezing that saline solutions do, showing that their constituent metals 
 form a solution when in the liquid state but that they are insoluble in one 
 another in the solid state. As an example of such an alloy, that of lead 
 and tin is most convenient for study. To cite a specific example, let a 
 mixture composed of 30% tin and 70% lead be heated to a temperature 
 of 350 C. As this temperature is above the fusion points of both lead and 
 tin, which melt at 327 C. and 232 C., respectively, it is sufficiently high 
 to insure that the mixture will be completely fused. Now, as this solution 
 cools down, no crystallization takes place until a temperature of about 
 270 C. is reached, when crystals of lead begin to separate out, making 
 the remaining solution poorer in lead but richer in tin in the same way as 
 the ratio of salt to water became greater during the freezing of the weak 
 saline solution. Likewise, as in the case of the salt solution, the separation 
 of the lead causes a retardation of the rate of cooling, showing that heat is 
 evolved thereby; and the freezing point of the mother liquor becomes lower, 
 so that no further separation of lead takes place until more heat is 
 abstracted. If the cooling be continued, however, the separation of the 
 lead will also continue, and if proper measures be taken, a number of loci 
 may be obtained of the cooling curves for alloys of different composition, 
 which, when plotted, give the curve M O as represented in the diagram 
 of Fig. 106. 
 
 Freezing point 
 of pure lead. 
 
 Freezing point 
 of pure tin. 
 
 Fia. 106. Diagram Illustrating the Freezing of Lead-Tin Alloys. 
 
 When the cooling and the accompanying separation of lead has reached 
 the point O., corresponding to a temperature of 180 C. and 31% of lead, 
 or 69% tin, in the mother liquor, the whole mass becomes solid, forming 
 a banded structure composed of minute crystals of lead and tin and cor- 
 responding to the cryohydrate of salt and water, but called, in the case 
 of alloys, eutectic alloy, which signifies easily melted alloy. To the right 
 of the point O, tin separates along N O like lead along M O. This manner 
 
524 
 
 CONSTITUTION OF STEEL 
 
 of freezing, where one metal separates alone, is known as selective freezing 
 to distinguish it from the kind of freezing illustrated by the gold-silver 
 alloys, which, since both metals separate together, is known as non=selec- 
 tive freezing. 
 
 The Iron=Carbon Eutectic: Coming now to a consideration of the 
 iron carbon alloys, the student finds that the freezing of alloys of these 
 two elements presents phenomena that are like those of both the gold- 
 silver and the lead-tin alloys. The freezing of these alloys is represented 
 by the following diagram, from which it is seen that the carbon content 
 
 IRON-CARBON SYSTEM 
 
 % CARBON 
 
 FIG. 107. Diagram Demonstrating the Freezing and Cooling of Iron Carbon Alloys. 
 After H. M. Howe. 
 
 of the eutectic alloy is about 4.30%. Therefore, when alloys that contain 
 more than this amount of carbon are cooled from temperatures above the 
 line M O N, carbon in the form of graphite separates along N O until 
 the point O is reached, when the eutectic solidifies. Naturally, the reader 
 would expect a similar separation of iron along the line M O; but it is 
 here that the freezing of the solution produces phenomena similar to the 
 freezing of gold-silver alloys, for it is found that, instead of pure iron separat- 
 
FREEZING OF IRON-CARBON ALLOYS 525 
 
 ing, a definite mixture, or alloy, containing approximately 2.0% carbon and 
 called primary austenite, separates from all mixtures in which the carbon 
 content is two per cent, or more. By drawing in the vertical line A D, 
 corresponding to about 2.0% carbon, this diagram is divided into two 
 parts. The part to the right of A D shows the freezing of the iron carbon 
 alloys to be like that of the lead-tin alloys, that is, selective, while that 
 part to the left of A D shows that the freezing of all iron-carbon alloys 
 whose carbon content is less than 2.0% is non-selective and analogous to 
 the freezing of the gold-silver alloy. For example, suppose an iron-carbon 
 alloy containing 1.0% carbon to be at a temperature of 1500 C. It is in the 
 liquid state and represents a homogeneous mixture of iron and carbon, or 
 a solution of carbon in iron. If now this solution is allowed to cool, crys- 
 tallization will begin when the temperature indicated by the corresponding 
 point "f" on the line M O is reached, and will continue up to the point 
 "s" on the line M P, when the solidification will have been completed. 
 Each crystal that separates contains 1.0% C., and when the point "s" is 
 passed, the mass represents a solid solution with a carbon content of 1.0%. 
 This solid solution is also known as primary austenite. Because of this 
 difference in the freezing of the iron carbon alloys between those contain- 
 ing more than 2.0% carbon and those containing less than 2.0% carbon, 
 the carbon content of 2.0% may be considered as the dividing line between 
 steel and pig iron; consequently, this study is concerned mainly in the 
 changes that occur in alloys whose composition is represented by the region 
 in the diagram that lies to the left of the line A D. 
 
 Formation of Pearlite and the Eutectoid : By studying the cooling 
 of the primary austenite through the region below M P, it is found that this 
 solid solution of carbon in iron undergoes changes similar in character to those 
 presented by the freezing of the liquid solution. These changes are repre- 
 sented by a secondary set of curves as shown in the part of the diagram to the 
 left of AD. This diagram indicates that a substance corresponding to the 
 eutectic of liquid alloys is formed and that it contains about .90% carbon, 
 but since it is formed from a solid solution, it is called the eutectoid, a 
 term that means "something of the nature of an eutectic." It will be 
 observed that as the primary austenite is cooled from M P, the line of 
 complete solidification, that any alloy with a carbon content greater than 
 .90% precipitates iron carbide, Fe 3 C, along the line P O', whereas those, 
 in which the carbon content is less than .90%, throw out of solution pure 
 iron or ferrite along M' O' until the eutectoid composition is reached. 
 The unchanged alloy, to which the term mother metal may be applied, 
 then undergoes a change wherein the precipitation of both the iron and 
 the iron carbide, FesC, is completed simultaneously, with the result that the 
 eutectoid thus formed consists of interstratified layers of ferrite and cement- 
 ite, commonly called pearlite, as previously explained. Hence, the term 
 eutectoid is often applied to pearlite, when it is desired to indicate the 
 manner of its formation and its structural characteristics. Since the metal 
 
526 CONSTITUTION OF STEEL 
 
 is in the solid form during these changes, it having reached its freezing 
 point 500 to 800 C. above the temperature of formation for pearlite, the 
 cause for these changes cannot be ascribed to a change in state. While 
 many theories have been advanced to explain this and other facts the 
 most satisfactory explanation is that iron exists in at least two, possibly 
 three, allotropic forms. Thus, below a temperature of 690 C. it exists in 
 a form designated as the alpha form, in which it has no power of dissolving 
 carbon or the carbide, cementite, whereas above this temperature it can 
 hold this constituent in solid solution. At these higher temperatures it is 
 designated as the gamma form, and the solid solution of carbon, or carbide, 
 in iron, is, micrographically, called austenite. Furthermore, the change 
 from austenite to pearlite is not instantaneous, and, as will be explained 
 later, several transition products may intervene, the complete series being 
 austenite, martensite, troostite, sorbite and pearlite. From what has been 
 said, it is evident that a steel that contains .90% carbon will, if cooled 
 slowly from any point above the critical temperature for its formation, 
 consist entirely of pearlite. Such steels are designated as eutectoid steels, 
 while those that contain less than .90% carbon are termed hypo=eutectoid 
 steels, and those in which the carbon content exceeds .90% are called 
 hyper=eutectoid steels. Other phenomena which accompany the cooling 
 of the primary austenite will be described in the next section. 
 
 Structural Composition of Slowly Cooled Steel: All steels that 
 have been cooled slowly from a temperature above that for the formation 
 of pearlite will contain it as a constituent. Thus, in the case of hypo- 
 eutectoid steels, all the carbon present will be found as pearlitic cementite, 
 the amount of pearlite being controlled by the amount of carbon present. 
 Any ferrite above that required by the cementite in the formation of pearlite 
 will be rejected as free, or excess, ferrite. In such steels, this excess ferrite 
 is in the form of a network surrounding small masses of the pearlite. In 
 the case of hyper-eutectoid steels, the amount of pearlite is again controlled 
 by the carbon, but in an indirect way. As all the carbon combines with 
 iron to forAi cementite, only a limited portion of ferrite remains for the 
 formation of pearlite. As this ferrite is not sufficient to interstratify with 
 all of the cementite, an excess of the latter remains. Like the rejected 
 ferrite, this excess cementite will also appear as a network about the masses 
 of pearlite. Thus, from the carbon content of a slowly cooled steel it is 
 possible to determine accurately the structural composition, or from the 
 relative proportions of pearlite and ferrite or cementite as revealed by 
 the microscope, the practiced metallographer can determine the approxi- 
 mate carbon content. 
 
 Effect of These Constituents Upon the Physical Properties: The 
 
 data on the physical properties of these constituents enables the student 
 to understand the remarkable effect of carbon upon the physical properties 
 of ordinary steel. In brief, the facts in their relation to the static strength 
 of slowly cooled steel are as follows: 1. Each constituent has the power 
 
THERMAL CRITICAL POINTS 
 
 527 
 
 to impart to the steel its own properties in proportion to the extent of its 
 presence. 2. Ferrite has the minimum tensile strength but maximum 
 ductility. 3. Pearlite has maximum tensile strength with low ductility. 
 4. Cementite has great hardness and brittleness with very little strength. 
 The effect of these constituents upon the properties of steel are plainly 
 shown in the accompanying diagram. 
 
 100 
 
 120,000 
 100,000 
 
 - 40 S 80,000 
 
 P. 
 
 GO 
 
 I 30 I 
 
 60,000 
 
 40 I 20 I 40,000 
 
 20 10 P 20,000 
 
 1.00 1.20 1.40 
 
 %CarbonO .20 40 .60 . 
 
 FIG. 108. Diagram Showing the Approximate Influence of Carbon Upon the 
 Strength and Ductility of Steel and its Relation to the Pearlite Content. 
 
 SECTION II. 
 
 THERMAL CRITICAL POINTS OF STEEL. 
 
 Nature of Critical Points or Ranges of Steel: The structural and 
 other changes in steel just described take place at temperatures known as 
 the thermal critical points, or critical ranges, because they are points in the 
 cooling or heating of the metal that are marked by the spontaneous evolution 
 or absorption of heat. The most marked of these is the range commonly 
 called the point of recalescence and point of decalescence. 
 
 Thermal Critical Point for Eutectoid Steel: For example, suppose 
 a piece of steel, containing .90% carbon and at a temperature of 1000 C., 
 be allowed to cool slowly. If, now, the rate of cooling be carefully ascer- 
 tained by means of a pyrometer, it will be found that the cooling proceeds 
 at first at a uniformly retarded rate, thus following the law for all cooling 
 bodies. But when a temperature of about 700 C. is reached, the uniformly 
 retarded cooling is momentarily arrested. A pyrometer will not only fail 
 to record any further decrease in temperature, but in most cases, when 
 the conditions are favorable, will show that the temperature of the cooling 
 
528 CONSTITUTION OF STEEL 
 
 mass actually rises. These facts show that heat is spontaneously generated 
 within the metallic body in amount sufficient to balance, or more than 
 balance, that lost through radiation and conduction. If the experiment is 
 performed in the dark, the steel will be observed to glow at this point due 
 to the heat evolved, and so the term recalescence has been applied to it. 
 Investigation has shown that the amount of heat given off in this case is 
 about 16 cal. per gram of pearlite. 
 
 Thermal Critical Points for Pure Iron: If instead of the eutectoid 
 steel, a piece of the purest iron obtainable be substituted in the experiment 
 just described, the cooling of this pure iron is found to be very unlike that 
 for the eutectoid steel. Thus, the metal will be found to cool at a uniformly 
 retarded rate till a temperature of 900 C. is reached, when a marked 
 increase in retardation occurs, showing that heat is being evolved, but 
 insufficient to cause an actual rise in temperature of the body of metal, 
 i. e., a recalescence. The cooling will then resume a normal rate until 
 the temperature of about 760 is reached, when a second evolution of heat 
 takes place, but not so pronounced as in the first instance. The metal then 
 cools normally to atmospheric temperatures. Thus, in pure iron there are 
 two evolutions of heat, i. e., two critical points, both of which take place 
 at a higher temperature than that noted for eutectoid steel and without 
 actual rise in temperature. Carbonless iron, therefore, has no point of 
 recalescence. 
 
 Thermal Critical Points of Low Carbon Steel: If the same experi- 
 ment be now performed with a steel containing even a small per cent, of 
 carbon, say .10%, the influence of this element is found to be very marked. 
 Three thermal retardations will be detected, the first, the most marked, 
 at about 850 C., the second near 760 C., and the third at the point of 
 recalescence, near 700 C. The last two are very faint. 
 
 Thermal Critical Points of Medium Carbon Steel: If the experi- 
 ment with low carbon steels be repeated with specimens containing higher 
 and higher percentages of carbon, the upper critical points observed in the 
 preceding experiment on .10% carbon steel and carbonless iron will be found 
 to be lower and lower as the percentage of carbon is increased, until, finally, 
 the determination of the rate of cooling of a steel containing .35% or .40% 
 carbon reveals only two critical points, the upper one at about 740 C. 
 and the other at the point of recalescence, 700 C. This fact means that 
 the carbon has caused the two upper points observed in pure iron and low 
 carbon steels to merge into one critical point. Furthermore, experiments 
 on steel containing a higher per cent, of carbon than .40 show that the two 
 lower critical points remaining also apparently merge into one on steels 
 containing .60% carbon and over. Theoretically, this apparent merging 
 should not take place till the steel is composed entirely of pearlite, that 
 is, when it has the eutectoid composition and contains .85% to .90% carbon. 
 The early merging is attributed to the difficulty of distinguishing by experi- 
 ment two critical points so close together. 
 
THERMAL CRITICAL POINTS 
 
 529 
 
 The Carbon=Iron Diagram for Steels and Methods of Notation: 
 
 These critical points or ranges are indicated graphically in the accom- 
 panying diagram of Fig. 109, which is seen to be the same a that used in 
 explaining the formation of pearlite. This diagram refers to the critical 
 
 1050 
 
 1000 ._ 
 
 960 
 
 % Carbon 
 
 %Cementite 0. 
 % Pearlite 0. 
 
 %Free Ferrite 100. 
 %FreeCementiteO. 
 
 1.80 
 
 24. 27. 
 
 87. 83.+ 
 
 0. 0. 
 
 13. 17. 
 
 Fia. 109. Diagram Showing Position of the Critical Ranges and the Relation of the 
 Carbon Content to that of Pearlite and Ferrite and Cementite. 
 
 points on cooling, which occur at temperatures somewhat lower than on 
 heating. All these critical ranges are denoted by the letter "A," followed 
 by either the small letter r, an abbreviation for the French word 
 "refroidissement," cooling, or the small letter c, which stands for 
 "chauffage," signifying heating. These signs, Ar and Ac, are further 
 modified by the numerals 1, 2, 3, indicating the point of recalescence, the 
 second, and the third points encountered on heating, respectively. Thus, 
 Acl means the first critical point passed upon heating the steel, and so on. 
 
530 CONSTITUTION OF STEEL 
 
 The Position of the Critical Ranges is affected in many ways. Atten- 
 tion has already been called to the difference between the ranges on heating 
 and cooling. ' In commercially pure carbon steels, Ar almost invariably 
 takes place between 690 C. and 720 C. and Ac 20 to 40 degrees higher. 
 It has been well established that this lagging of the point on cooling and 
 the point on heating behind the true point is a case of hysteresis, often 
 observed in physical phenomena. Evidence of the correctness of this 
 explanation is found in the fact that the slower the heating and cooling 
 the nearer the two points approach each other. The speed of cooling or 
 heating, then, is the first factor affecting the position of these points. A 
 second factor influencing the positions of Arj is found in the temperature 
 to which the steel is heated before cooling begins. The higher this tem- 
 perature and the longer it is held at the high temperature the lower the 
 position of Ar A will be; but this change in position of the critical point is 
 not pronounced and takes place very gradually and slowly. A third factor 
 is that of chemical composition. In general the presence of impurities in 
 the steel have a tendency to lower the position of Ac and Ar, and in some 
 cases this tendency is very decided. Thus, manganese lowers the position 
 of Ar some 25 C. to 50 C. for each per cent, of that element present in the 
 steel. Nickel and copper also lower the Ar range. In the case of nickel 
 and manganese the lowering is so pronounced that in a steel containing 
 13% Mn. and 25% Ni. no retardation is observed in cooling from a high 
 to atmospheric temperature, but appears on cooling in liquid air, which 
 indicates that the Art point has been lowered in this steel to below the 
 temperature of the air. In the ordinary steel of commercial quality, the 
 impurities are present in so small amounts that they can cause little vari- 
 ation in the position of the critical points. 
 
 
 
 Changes at the Thermal Critical Points: Besides the rise in tem- 
 perature or retardation of cooling already explained, careful investigation 
 has shown that other important changes take place in steels in passing 
 through these ranges. For convenience these changes and their effects 
 are summed up as follows: 
 
 I. Changes at A.^: The point AS, as already shown, applies to 
 carbonless iron and steels containing less than .35% carbon. The passing 
 of such steels through this point is accompanied by the following phenomena: 
 1. On cooling, the metal, which above Ar$ was contracting, undergoes a 
 sudden and marked expansion in volume on passing through Ar3. In linear 
 units the expansion amounts to about one one-thousandth of its length. This 
 dilation is then immediately followed by normal contraction, again. 2. Above 
 A 3 the metal has an electrical resistance about ten times greater than its 
 resistance at ordinary temperatures. At A.TS a sudden drop in this resist- 
 ance takes place, after which the decrease proceeds slowly at a uniform 
 rate until atmospheric temperature is reached. 3. A change in crystalline 
 
THERMAL CRITICAL POINTS 531 
 
 form of iron takes place at AS. Below this point iron crystallizes in the 
 cubic form, but this form changes on passing Acs to that of the octahedra. 
 4. The tensile strength of iron at the AS point has been shown to undergo 
 a distinct discontinuity, (see page 337). 5. The dissolving power of iron for 
 carbon is one of the most important changes which the properties of the metal 
 undergo in passing the Acs point. From what has been said in the preceding 
 section it may be surmised that below this point, or at least as long as 
 the iron is in the alpha form, it has no power of dissolving carbon, but 
 it gains this power immediately the Ac3 point has been passed. 6. An 
 abrupt structural change accompanies the passage of all low carbon steels 
 through this range. It is on reaching this point that the free ferrite begins 
 to be set free, which continues till the residual austenite is of the proper 
 composition for the formation of pearlite. 
 
 II. Changes at A 2 : As indicated in the iron-carbon diagram, A 2 
 occurs as a separate point only in carbonless iron and in steels containing 
 less than .35% carbon. Just as the retardation on cooling was found to 
 be faint at this point, so the changes in properties are not so numerous 
 nor so marked as at A$. Thus, there is no dilation, no structural change, 
 no change in crystalline form, and probably no change in the dissolving 
 power of iron for carbon on passing through the range A2. But the following 
 three changes in properties on passing A2 are to be specially noted. The 
 magnetic properties undergo a marked change on passing A 2 . Above 
 this point steel is non-magnetic, or para magnetic, but in passing through 
 Ar 2 it suddenly becomes strongly magnetic, and gradually becomes more 
 so as the cooling continues, finally gaining its full magnetism at a tem- 
 perature of about 500 C. A distinct discontinuity in both the tensile 
 strength and specific heat of iron at the A 2 point has been shown to take 
 place. 
 
 4 
 
 III. Changes at A 3 , 2 , which point results from themergingof Aaand A 2 , 
 are tfye same as those occurring in low carbon steels in passing through A$. 
 
 IV. Changes at Ait As previously explained, the point A! occurs 
 in all steels containing from a mere trace to .90 per cent, carbon. It cor- 
 responds to the transformation of residual austenite into pearlite. At this 
 point the following sudden changes in properties are noted: 1. A dilatation 
 takes place which increases with the carbon content, reaching a maximum 
 with .85% carbon. 2. Increased magnetism takes place for all steels on 
 cooling through this point. 3. A marked decrease in electrical resistance 
 is also noted on cooling through this range. 4. Below this point iron 
 loses entirely its power to dissolve carbon. 5. The important changes from 
 austenite to pearlite have already been shown to take place at AI, but it 
 is well to emphasize their significance here, for these changes give the key 
 to the rational treatment of steel. The spontaneous transformation of 
 austenite of eutectoid composition into pearlite, that is, a solid solution 
 
532 CONSTITUTION OF STEEL 
 
 into an aggregate of the eutectoid, makes possible the refining of steel by 
 heat treatment, because on being heated through its critical range, the 
 steel is changed from a coarse aggregate to a fine, almost amorphous, solid 
 solution. This fact is also the secret of hardening steel, as will be shown 
 later. 
 
 Causes of the Thermal Critical Points in Steel: In seeking a cause 
 for the existence of the thermal critical points in steel, all the phenomena 
 exhibited must be considered. Starting first, then, with the thermal 
 changes noted in the experiments previously described, it is well to note 
 that there are but three well known causes of spontaneous evolution of 
 heat in cooling bodies and of similar absorptions of heat on heating them. 
 Briefly, these causes are (1.) the formation or decomposition of chemical 
 compounds; (2.) changes of state, whether by solution or by the agency 
 of heat; and (3.) allotropic or polymorphic transformations, which are 
 always accompanied by either an absorption or evolution of heat when 
 the substance of the body passes from one allotropic condition to another. 
 As to the upper points, AS and A2, it has been shown that at these points 
 even carbonless iron either absorbs or evolves heat, depending upon whether 
 the metal is being heated or cooled while passing through the ranges. The 
 fact that the iron is pure, nothing being present with which it could combine, 
 and the fact that it is in- the solid state throughout the experiment, preclude 
 the possibility of either a chemical change or a change of state having 
 taken place to cause the thermal changes indicated. Only the explanation 
 founded on the basis of allotropy, therefore, remains. According to the 
 two thermal changes that occur, then, pure iron or ferrite exists in at least 
 three allotropic forms. Below the point A2 it is called alpha iron, between 
 A2 and AS it is known as beta iron, while above AS it is said to be in the 
 gamma form. While it is not desirable to undertake a discussion of this 
 theory here, it may be pointed out that all the changes in properties pre- 
 viously mentioned as accompanying these critical points are but additional 
 evidence of the correctness of this view. Since the influence of carbon is 
 to delay the separation of ferrite, it may be that beta iron is not formed 
 on cooling steels containing .35% carbon or more, for since the temperature 
 of the point Arg-2 in such steels is below that for the formation of beta 
 iron, it is probable that just as iodine passes directly from the solid to the 
 gaseous state by sublimation, or as amorphous sulphur at atmospheric 
 temperatures passes to rhombic sulphur, so gamma iron passes directly 
 to the alpha form. The fact that the point AI does not occur in carbonless 
 iron and only faintly in low carbon irons, while with increase of the carbon 
 it increases in intensity, is evidence that this point is due solely to the 
 presence of carbon. Unlike the points AS and A2, which are due to allo- 
 tropic forms of iron, AI is not due to any change in the carbon itself, but 
 merely to the formation of pearlite, which implies the crystallization or 
 falling out of solution of cementite, FesC, coupled with the complete 
 change of the ferrite from the gamma to the alpha state as well. Above 
 
CRYSTALLINE STRUCTURE 533 
 
 AI, the FesC, being in solution with the gamma iron and thoroughly diffused, 
 has the power of imparting its own properties, hardness and brittleness, 
 to the steel in a more pronounced way than when in the segregated form 
 in which it occurs below AI. Hence, formerly due to the theories then 
 advanced to explain the hardening effect of carbon, the term hardening 
 carbon was applied to it above AI, while below AI it was called cement 
 carbon. 
 
 SECTION III. 
 
 THE CRYSTALLINE STRUCTURE OF STEEL. 
 
 Crystals and Grains: That the crystalline structure of steel exerts 
 a deep influence upon its strength and ductility is a well known fact. Since 
 this structure lends itself to refinement through proper heat treatment, a 
 clear understanding of all the laws governing the crystallization of steel 
 is essential to the art of heat treating steel. When steel, like many other 
 substances, passes from the liquid to the solid state, the process of solidi- 
 fication is accompanied by crystallization, that is, the molecules of the 
 various ingredients arrange themselves so as to form small bodies having 
 regular geometrical outlines. Each of such bodies constitutes a crystal. 
 In the case of iron in the gamma form the crystals are octahedra, or small 
 eight sided bodies, but when the iron is in the alpha condition these crystals 
 are cubic in form. Crystals have the remarkable pr6perty, called cleavage, 
 of breaking most easily along certain planes usually parallel to the faces of 
 the crystal. Hence, these planes of easy rupture are called cleavage nrfanes. 
 The direction of the cleavage planes constitutes the orientation nrf the 
 crystal. Perfect crystals, called idiomorphic crystals, are formed only 
 when the conditions are favorable. Thus, with high fluidity, absence of 
 foreign particles, slow rate of cooling, and with the liquid at rest and 
 undisturbed, perfect crystals of large size may form. Because of the unfavor- 
 able conditions that usually prevail in its manufacture, the crystallization 
 of steel results in the formation of imperfect crystals with irregular forms 
 and smaller in size than perfect crystals. These imperfect crystals are, 
 scientifically, designated as allotrimorphic crystals, but the metallurgist 
 speaks of them simply as grains. 
 
 Crystallization of Steel : As an aid to understanding the crystalliza- 
 tion of steel it will, perhaps, be best to follow the crystallographic history 
 of a steel casting that is allowed to cool slowly from the casting temper- 
 ature to atmospheric temperature. For the present, let this steel be of 
 any carbon content. During the solidification period, what has been 
 termed the primary crystallization takes* place, which consists in the 
 formation of macroscopic tree-like bodies of austenite called dendrites. 
 Each of these dendrites is composed of small octahedra, which is repre- 
 sentative of the crystallographic form for austenite. Of these, Stead writes: 
 "The fine fir-tree crystallites, containing probably a fraction of the amount 
 
534 CONSTITUTION OF STEEL 
 
 of the carbon in the liquid steel, grow steadily forward from the cold surface 
 of the containing moulds. The crystallites develop branches in three 
 directions corresponding to the axes of the cube, and these branches throw 
 out similar branches themselves.. Eventually parts of the most fusible 
 portions are trapped between the branches and are the last to solidify. 
 When there is much phosphorus or some sulphur in the metal, they are 
 always present together with an excess of the carbon in the last residue 
 of metal that remains liquid, and although in cooling down, after the liquid 
 has solidified, the excess carbon diffuses out of it into the purer part, the 
 sulphides and phosphides do not, but remain fixed, and can generally be 
 detected in the solid metal." 
 
 After the solidification is complete a crystalline transformation, called 
 granulation, sets in and continues until the critical range is reached, where 
 all steels are found to be made up of grains, each grain having its own 
 orientation and being made up of small octahedra of crystalline matter. 
 The size of these grains varies with the rate of cooling, just as the rate of 
 cooling affects the size of crystal in any metal. In passing through the 
 critical range the structural changes are affected by the amount of carbon 
 present, and for this reason it is best to consider the three grades of 
 eutectoid, hypo-eutectoid, and hyper-eutectoid steels separately from this 
 point onward. 
 
 Crystallization of Eutectoid Steels: In the eutectoid steels the 
 growth of the grains will continue down to the point Ari, where the metal 
 will b I made up of grains of austenite containing ferrite and cementite in 
 proper proportions to form pearlite. Therefore, in passing through the 
 critical point, Ar , each austenite grain changes bodily into a grain of 
 pearlite. Hence, the coarse austenitic structure acquired by cooling from 
 a high temperature gives rise to a correspondingly coarse pearlite structure. 
 
 Crystallization of Hypo-Eutectoid Steel: As an example of the 
 genesis of the crystalline structure of steels of hypo-eutectoid composition 
 let a steel containing .60% carbon, corresponding to 72% pearlite and 28% 
 free ferrite, be selected. In such a steel the granulation will proceed till the 
 upper critical point Ar 3 - 2 is reached, where free ferrite begins to be rejected 
 and continues till the point Arj is reached, when the residual austenite, 
 being of the proper eutectoid composition, passes into pearlite as described 
 for eutectoid steels. An important point to be noted here is the fact that 
 this setting free of the excess ferrite is brought about through the rejection 
 of ferrite in excess of the eutectoid composition by each individual grain 
 of austenite, either to its boundaries or between its cleavage planes. 
 When each grain of austenite becomes a grain of pearlite, the ferrite pre- 
 viously rejected still remains as an envelope, thus forming the net-work 
 mentioned under "Ferrite." Therefore, the structure of cast hypo-eutectoid 
 steel is very coarse, for the following three reasons: 1. The slow and 
 
CRYSTALLINE STRUCTURE 535 
 
 undisturbed cooling promotes the formation of large austenite grains and 
 hence, later, of large pearlite grains. 2. The slow cooling between the 
 upper and lower critical points favors the rejection of a maximum amount 
 of free ferrite, which rejection makes for coarseness of structure. 3. The 
 slow cooling from the upper critical point to atmospheric temperature 
 promotes the crystallization of excess ferrite into large grains, especially 
 when this excess is large in amount, i. e., in very low carbon steel. Because 
 of its coarse structure, cast hypo-eutectoid steel is less tenacious and less 
 ductile than forged, rolled or properly annealed steel of similar composition. 
 
 Crystallization of Hyper=Eutectoid Steels: In the case of hyper- 
 eutectoid steels, the granulation proceeds to the point Acm, a temperature 
 that is indicated by the line Acm in Fig. 109, where excess cementite is 
 liberated like the excess ferrite in hypo-eutectoid steels. This excess 
 cementite is rejected either to the boundaries of the austenite grains or 
 between their cleavage planes, where, after the transformation of eutectoid 
 austenite into pearlite, it remains to form a network about the grains of 
 the latter. 
 
 The Effect of Work on Grain Size: The effect of work on the 
 mechanical properties of steel has been discussed in the second part of this 
 book. It remains to be pointed out here that the greatest benefits of working 
 as the steel cools to Ar t are due to a refinement in the grain size. The ex- 
 planation for this refinement is found in the fact that, as each grain is 
 distorted by the application of mechanical pressure, it endeavors to resume 
 its original form, but being hindered in this by the rigidity of the mass, 
 it breaks up into a number of smaller grains possessing the characteristic 
 form. If the steel be worked at temperatures below Ar , cold working, 
 no refinement r of grain takes place, as the great rigidity of the metal or 
 its lack of molecular energy prevents any readjustment of the grains at all. 
 Hence, cold worked steel will show a pronounced distortion of grain. 
 
 Crystalline Changes on Heating Steel: Since the preceding dis- 
 cussions have been concerned mainly with steel under conditions of cooling, 
 it may be profitable to review these changes from the standpoint of heating. 
 Starting, then, with a low carbon steel, containing, say, .20% carbon, it 
 will be found that, under normal conditions of manufacture and in its natural 
 state, this steel consists of approximately 24% pearlite and 76% free alpha 
 ferrite, the pearlite existing in small grains surrounded by the ferrite as a 
 network. Upon heating through the point Aci, the pearlite changes into 
 austenite, the iron of which is in the gamma form; but the free ferrite is 
 still in the alpha form. As the heating continues throughout the zone 
 bounded by Ac and Ac2 the austenite thus formed begins to absorb the 
 free ferrite. Upon passing through the Ac2 range the remaining alpha 
 ferrite changes into beta ferrite; and the steel as a whole will be found to 
 be hard and non-magnetic. As the heating progresses through the second 
 zone, lying between Ac2 and Acs, the beta ferrite, which is only a remnant 
 
536 CONSTITUTION OF STEEL 
 
 of the original alpha ferrite, is gradually absorbed, so that as the range 
 Acs is passed the whole of the steel passes into the condition of austenite, 
 or a solution of iron carbide in gamma iron. In a similar manner the changes 
 in the constituents of steels of any carbon content up to eutectoid steel 
 might be explained. In each case, however, it is to be noted that the 
 temperature at which the transformations are completed falls as the carbon 
 content is increased, being at its lowest when the eutectoid ratio has been 
 reached. In the case of hyper-eutectoid steels, the free cementite is 
 absorbed in a manner analogous to the absorption of free ferrite in hypo- 
 eutectoid steels. But the final solution of the cementite takes place at 
 a temperature range indicated by the line Acm, and much more slowly 
 than ferrite. This latter point, being a matter of great practical importance, 
 should be kept in mind. 
 
 Crystalline Refinement on Heating: Besides these structural 
 changes brought about by heating the steel through the various critical 
 ranges, there still remains a matter of extreme importance to be explained. 
 This matter refers to the crystalline, or grain, refinement observed when 
 a steel is heated through these ranges. Again assuming that the steel is 
 in a normal condition after manufacture in the usual manner, no change 
 on heating is observed to take place in the grain structure until the tem- 
 perature has reached that of the lower critical range, Ac^. At this tem- 
 perature, which marks the point where the original pearlite grains are 
 transformed into austenite grains, the maximum refinement, that is, the 
 smallest grain size possible, occurs. This refinement is to be expected 
 from the conditions of the formation of the austenite. Since the conditions 
 favorable for the formation of large grains require slow cooling from a high 
 temperature, the formation of the austenite at this low temperature permits 
 no growth of the grain structure at all. Hence, it is found to be almost 
 amorphous in respect to its grain structure. But it is to be especially 
 noted that, as the temperature is raised above this critical range, grain 
 growth begins, which fact results in a gradual coarsening of the grain of 
 the austenite as the temperature is progressively raised above this range. 
 It is also to be noted that this increase in grain size not only varies with 
 the temperature above the critical range, but also with the length of time 
 at which the steel is maintained at the high temperature. In eutectoid 
 steels, then, complete and maximum refinement of the grain takes place 
 immediately the point Ac is passed. But if the steel contains free ferrite 
 or free cementite, that is, if it is of hypo-eutectoid or hyper-eutectoid 
 grade, then the steel as a whole is not refined on passing Ac^ because the 
 excess ferrite or cementite remains unaltered. In all cases it is only when 
 all the constituents of the steel have passed into the state of a solid solution, 
 or austenite, that complete refinement can be obtained. To bring about 
 such a condition, it is necessary to heat such steels to a temperature a little 
 above that of their upper critical ranges as indicated on the iron carbon 
 diagram, on account of hysteresis previously discussed. 
 
CRYSTALLINE STRUCTURE 
 
 537 
 
 1. Heated to about 1300 C. 
 and quenched in water. 
 
 2. Heated considerably above 
 the critical range and quenched 
 in water. 
 
 3. Heated to just above the 
 critical range and quenched in 
 water. 
 
 4. Heated to just below the 
 critical range and quenched in 
 water. 
 
 5.' Steel as forged and cooled 
 in air to atmospheric temperatures. 
 
 FIG. 110. Natural Size Photographs Showing Effect of Heat Upon Grain Size of a 
 Rolled and Forged Steel, Carbon .75% (Metcalf's Experiment). 
 
 1. Steel as cast and cooled 
 naturally. 
 
 2. Heated to 927 C. 
 quenched in water. 
 
 and 
 
 FIG. 111. Natural Size Photographs Showing Effect of Heat Upon the Grain Size of 
 Cast Steel. Specimens, left to right, contain, .25% Carbon and .36% Carbon. 
 
 Practical Importance of Grain Structure: The proper refinement 
 of grain structure is of great practical importance. An illustration of the 
 way in which the facts pointed out above in connection with the effects 
 of mechanical work and heat upon the grain structure of steel may be prac- 
 
538 CONSTITUTION OF STEEL 
 
 tically applied is furnished by the welding of steel. If two steel bars are 
 welded together by scarfing the ends slightly and hammering lightly over 
 the weld only, as is the practice of most blacksmiths in welding iron bars, 
 it is found that, while the weld itself is strong, the welded bar will be weak 
 on each side of the weld. A bending test applied to such a weld generally 
 causes the bar to break a short distance from the weld, which fact is 
 responsible for the assertion, often made by some blacksmiths, that the 
 weld is stronger than the bar. A careful examination of the whole bar, 
 however, will usually show that the regions on each side of the weld are 
 the weakest points in the entire bar. Evidently this weakness has been 
 developed in the process of welding. The high temperature required in 
 welding increases the grain size of the ends to be welded for a considerable 
 distance along the bars. The subsequent hammering refines this large 
 grain in the weld itself, but not in the areas on each side of it. By changing 
 the manner of welding somewhat, the structure, of the welded bar can be 
 made almost uniform throughout, and these defective areas will not appear. 
 This result can be accomplished by making the weld in the following 
 manner, which is the usual practice in welding steel: The two ends to be 
 welded are first heated to a moderate forging temperature for a distance 
 of several inches back, the exact distance depending upon the size of the 
 bar; these ends are then stove up, or upset, that is, the heated regions 
 are shortened and thickened by hammering directly against the ends. 
 Next, the ends are scarfed, but instead of a short, blunt scarf sometimes 
 used, a well beveled scarf should be made. The scarfed ends are then 
 heated to a welding temperature; a flux of common river sand or a reliable 
 commercial welding compound is applied; the weld is made as usual; and 
 the thickened portion of the bar is forged down to a size conforming to the 
 remainder. This forging refines the grain which had previously been made 
 coarse by the heating, and restores the uniform structure of the bar. If 
 the bars have been stove up right in the beginning, the form of the weld 
 will be such as to require the greatest amount of forging where the grain 
 is the largest and will decrease to none where the steel was heated only to 
 the critical range. 
 
 Summary of Chapter I. The conditions and properties of the iron 
 carbon alloys and their constituents may be summed up about as follows: 
 Above the critical ranges, the iron is in the gamma form, and the carbon 
 is dissolved, thus imparting to the alloy, when the carbon is present to the 
 amount of about .30%, the power of hardening; the alloys are non-magnetic 
 and crystallize on cooling slowly, but mechanical working prevents the growth 
 of the crystals and reduces their size. Below the critical range, the metal 
 represents an aggregate of ferrite and cementite, FesC, and it possesses little 
 heardning power. Here the iron is in the alpha form, the alloys are magnetic, 
 no crystallization takes place, and mechanical working distorts the grain 
 structure. 
 
THE TREATING OF STEEL 539 
 
 CHAPTER II. 
 
 HEAT TREATING THEORY AND PRACTICE. 
 
 Introduction: While considerable time has now been spent in a 
 discussion of subjects that may appear to be purely theoretical in nature 
 and of little practical value, yet this course is justified, because the principles 
 explained form the basis of all heat treating processes, and a clear under- 
 standing of them is therefore essential. The practical application of these 
 principles will now be considered under the following headings, correspond- 
 ing to the three chief processes of heat treatment as explained in the 
 beginning. 
 
 SECTION I. 
 
 ANNEALING. 
 
 The Annealing Operation consists in (1) heating the steel to some 
 predetermined temperature, (2) keeping the temperature constant at the 
 predetermined point for a given length of time, and (3) cooling the steel 
 according to some predetermined course to atmospheric temperature. To 
 accomplish the desired result in the given steel to be treated requires that 
 all three of these steps in the annealing operation be very carefully planned 
 and as carefully carried out, for the success of the operation depends entirely 
 upon the proper correlation of the rate of heating, the temperature to which 
 the steel is heated, the time it is kept at the annealing temperature and 
 the fate of cooling. 
 
 Purpose of Annealing: Evidently, then, the annealing operation will 
 be modified to suit the end sought. In general, the purpose of annealing 
 may involve any one or all of the following aims: 1. To soften the steel 
 in order that it may meet certain physical requirements or be more easily 
 machined. 2. To relieve internal stresses and strains induced by forging, 
 rolling, or drawing, or by a non-uniform contraction in cooling. 3. To 
 remove coarseness of grain and thus secure a more desirable combination 
 of strength, elasticity and ductility for resisting the stresses to which it is 
 to be subjected in service. The treatment is generally applied (1) to hot 
 forged steel objects, because their grain structure is often more or less 
 heterogeneous and, owing to high finishing temperature, relatively coarse; 
 (2) to cold worked steel, such as sheets and cold drawn wire, which often 
 must be annealed in order to increase or restore its ductility; and (3) to steel 
 castings, which usually have so coarse a grain structure as to be very 
 deficient both in strength and ductility. 
 
540 THE TREATING OF STEEL 
 
 True Annealing and "Process" or "Works" Annealing: To accom- 
 plish the results sought as expressed above in aims (1) and (2), it is not 
 always necessary to heat the steel to the critical range. Thus, in the 
 "process" or "works" annealing employed in wire drawing, it is only 
 necessary to heat the steel, which contains less than .10% carbon, to about 
 550 C. in order to relieve the strained condition of the ferrite and restore 
 the ductility. The same is also true in the case of the "white annealing" 
 of cold rolled sheets. It is to be noted, however, that this treatment does 
 not develop the maximum softness, because the pearlite is not affected. But 
 as this constituent is present in so small amounts, its influence is scarcely 
 evident. This method is also sometimes applied to tool steels in order to 
 soften them for machining. All true or full annealing, however, requires 
 that the steel be heated to a temperature above that of its upper critical 
 range, and it is to this true annealing the following discussion is to be 
 confined. 
 
 Heating for True Annealing: The first step in the annealing 
 operation is to heat the steel past its critical range, for in so doing the 
 previous structure is completely obliterated and a new one, nearly amor- 
 phous, is born. As has been previously explained, this important change 
 is due to the passage of the steel structure from the state of an aggregate 
 of ferrite and cementite to a homogeneous solid solution. Should the steel 
 remain below the critical range, no structural change takes place, if the 
 case of strain relief noted above in cold worked steel be excepted. The 
 coarsening effect upon the grain size of steel, brought about by heating 
 above this range, has already been explained. The proper temperature, 
 then, for true annealing is one but slightly above the critical range of the 
 steel, and this temperature must be maintained uniformly as near the 
 range as possible during the time the steel remains at the annealing 
 temperature. 
 
 The following ranges of temperatures are recommended by the committee 
 on heat treatment of the American Society for Testing Materials. 
 
 TABLE 59. Annealing Temperatures as Recommended by the 
 American Society for Testing Materials. 
 
 Range of Carbon Content. Range of Annealing Temperature. 
 
 Less than 0.12 per cent. 875 to 925 degrees C. 
 
 0.12 to 0.25 per cent. 840 to 870 degrees C. 
 
 0.30 to 0.49 per cent. 815 to 840 degrees C. 
 
 0.50 to 1.00 per cent. 790 to 815 degrees C. 
 
 These temperatures are shown diagramatically in the accompanying 
 figure, together with recommendations by other authorities. 
 
ANNEALING 
 
 541 
 
 975 
 
 850 
 
 925 
 
 .7 
 
 .8 
 
 .1 .2 .3 .4 .5 .6 
 
 Per Cent. Carbon 
 
 Fia. 112. Annealing (and Hardening) Ranges Showing Approximately the Temper- 
 atures Recommended by Different Authorities. 
 
 Legend. 
 
 Sauveur (for treating f orgings) . 
 
 American Society for Testing Materials. 
 
 Bullens (for annealing and hardening). 
 
 Stead's Lower Curve (for refining and hardening). 
 Stead's Upper Curve (for annealing and normalizing). 
 
542 THE TREATING OF STEEL 
 
 In large bodies, the central portion will lag in temperature behind the 
 exterior, hence such objects should be heated very slowly, for very evident 
 reasons. The practice of raising the temperature of the furnace beyond 
 the proper annealing. temperature in order to drive the heat to the interior 
 of the piece is a great mistake, for then the temperature of the exterior 
 may be carried beyond the proper point with consequent evil results 
 attending. 
 
 Importance of Time in Heating for Annealing: The time the object 
 should remain at the annealing temperature is governed largely by its size. 
 Evidently, it .should be maintained at this temperature until it has become 
 uniformly heated throughout. The committee quoted above recommends 
 that an exposure of one hour is sufficient for pieces twelve inches thick. In 
 practice, however, it is often necessary to keep the object at the annealing 
 temperature for a much longer period than that indicated by the committee 
 or that which theoretically would appear sufficient. This is especially 
 true with plain steel in cases where the mechanical work upon the steel has 
 been severe, or where the steel has been improperly heated in working, and 
 in certain of the alloy steels. It has been shown by Bullens 1 "that the 
 greater the internal stress upon the steel the greater is the amount of lag, 
 or final release, of this stress behind the actual change of constituents. 
 That is, even though a totally new structure may be formed by the anneal- 
 ing temperature, there remains for a considerable length of time a tendency 
 of the new structure to return, upon slow cooling, to the stressed condition of 
 the original, even though the constituents themselves may be those born 
 at the new temperature. It is important, therefore, if a soft steel 
 free from all internal stresses and strains is desired, that a sufficient length 
 of time be allowed for the permanent elimination of these stresses and 
 strains, before cooling." To accomplish this result a period of time 
 extending over several hours, or even days, may be required. 
 
 Cooling: Having, thus, by proper rate of heating and length of time 
 of heating, obtained the steel in a state favorable to maximum refinement, 
 the next step is to cool it properly. As variations in the rate of cooling 
 produce very profound effects upon the physical properties of the metal, 
 this process is not as simple as it appears. The effects of cooling at different 
 rates and in different ways should, then, be carefully studied. The property 
 most noticeably affected by the cooling process is the hardness. As is 
 well known, this property in a given steel depends upon the rate of cooling 
 from above the critical range. Thus, by the most rapid cooling, it is possible 
 to develop the maximum hardness, or by the slowest cooling the greatest 
 softness, and by varying the rate of cooling any degree of hardness between 
 these extremes may be obtained. In searching for a reason for these changes 
 in properties, it is not surprising that investigators have found that 
 important structural changes accompany all cooling, and that these changes 
 vary, in the effects they produce, with the speed of the cooling. 
 
 iSteel and Its Heat Treatment, Second Edition, pp. 133 to 148. 
 
ANNEALING 543 
 
 Effect of Cooling on the Net Work: The first of these structural 
 changes that may be mentioned is the effect of different rates of slow cooling 
 upon the net work of ferrite described in the preceding section of this book. 
 Thus, in very slow cooling of hypo-eutectoid steels from the annealing 
 temperature a coalescence of the excess ferrite into large grains intermingling 
 with coarse pearlite grains results. If the steel be now cooled rather rapidly, 
 but still not so rapidly as to prevent the formation of pearlite, the excess 
 ferrite will be found to be of a fine grain structure and to form a fine network 
 about the pearlite. If, now, these two methods be combined, that is, if 
 the cooling be made to proceed rapidly through the upper part of the trans- 
 formation range then slowly to atmospheric temperature, the net work of 
 ferrite is fine, but the pearlite is better developed than in the second 
 case. 
 
 The Effect of Cooling Upon Pearlite now remains to be explained. 
 While the rate of cooling from below the critical range can have no effect 
 upon the pearlite, changing the rate of cooling while the steel is passing 
 through this range and the solid solution is being transformed into pearlite 
 will correspondingly change the arrangement of the ferrite and cementite 
 composing the pearlite, so that the same steel may be made to exhibit 
 widely differing physical properties. As mentioned before, the austenite 
 does not pass directly into the pearlitic condition on cooling through the 
 critical range, but makes the change by way of the three transition stages 
 known as martensite, troostite, and sorbite. Of these, only sorbite is 
 retained in the steel by annealing methods of cooling. In small sections 
 it is retained by air cooling through the lower critical range. By varying 
 the cooling through this range this change from sorbite to pearlite may 
 be controlled so as to produce five phases, or varieties, of pearlite having 
 different physical properties as follows: 
 
 1st Phase. True Sorbite with emulsified FesC. Very dark on etching. 
 Tensile strength, 150,000. Elongation, 10% in two inches. 
 
 2d Phase. Sorbitic pearlite with semi segregated FeaC. Dark on 
 etching. Tensile strength, 125,000. Elongation, 15% in two inches. 
 
 3d Phase. Finely laminated pearlite with FesC almost completely 
 segregated. Exhibits a play of gorgeous colours when lightly etched. 
 Tensile strength, 100,000. Elongation, 10% in two inches. 
 
 4th Phase. Fully laminated pearlite with completely segregated Fe^C. 
 Tensile strength, 85,000. Elongation, 8% in two inches. 
 
 5th Phase. Massive pearlite, consisting of coagulated FesC. and 
 ferrite. Tensile strength, 75,000. Elongation, 5% in two inches. 
 
 The second phase is the one sought in the process of patenting in wire 
 drawing. So, it is seen that, having obtained the greatest possible refine- 
 ment as to grain size and released all the internal stresses and strains in the 
 metal by heating to the proper temperature, it still remains to adjust the 
 physical properties by regulating the rate and the manner of cooling. 
 
544 
 
 THE TREATING OF STEEL 
 
 1. Sorbite. Cementite is emulsified. 
 Obtained in steels of low carbon content by 
 cooling rapidly to atmospheric temperatures. 
 
 2. Sorbitic Pearlite. Cementite is partly 
 segregated. Obtained by cooling rapidly through 
 the upper range only. 
 
 3. Pearlite. Cementite is largely segregated. 
 Obtained by moderately slow cooling. 
 
 4. Laminated Pearlite. Cementite is com- 
 pletely segregated. Obtained by slow cooling. 
 
 5. Massive Pearlite. Cementite and ferrite 
 are coagulated. Obtained by very slow cooling 
 to atmospheric temperatures. 
 
 FIG. 113. Microphotographs Showing Progressive Segregation of Cementite in the 
 Development of Pearlite from Sorbite. (White areas represent ferrite, black 
 areas, Cementite.) 
 
ANNEALING 546 
 
 Other Factors to consider in cooling are the carbon content and the 
 size of the object. In general, the lower the carbon the more rapid may 
 be the rate of cooling without affecting to any marked degree the softness 
 and ductility of the metal. For example, steels containing less than .15% 
 carbon may even be quenched in water, and those containing less than 
 .30% carbon, in oil, without markedly decreasing their ductility. In order 
 to secure the same rate of cooling in objects of different size, it is obviously 
 necessary to regulate the external conditions in accordance with the dimen- 
 sions of the objects treated. Thus, the cooling in air of a very fine wire 
 may be equivalent to quenching in oil or water an axle of the same carbon 
 content. 
 
 Methods of Cooling: In general, there are three methods of cooling, 
 namely, furnace cooling, insulated cooling and air cooling. Of these, furnace 
 cooling may be made the slowest, especially if the furnace is large and can 
 be effectually sealed from air draughts. This method gives maximum 
 softness and ductility. In other words, the tensile strength and elastic 
 limit will be at their lowest, while the elongation and reduction in area 
 will be at or near their maxima. Steel subjected to such treatment will 
 resist severe distortions. In what has been termed above as insulated 
 cooling, the object is removed from the furnace and covered with a blanket 
 of lime, sand, ashes, etc., or it may be placed in a brick or concrete lined 
 underground pit with a tight fitting cover, which in turn may be covered 
 with ashes or loose earth. In cases where large amounts of steel are placed 
 in a single pit, thia method may be slower even than furnace cooling. In 
 air cooling, the object is simply removed from the furnace and allowed to 
 cool in the air. Evidently, the rate of cooling by this method will be affected 
 by the size of the piece and the season of the year. In addition the physical 
 properties imparted will depend somewhat upon the carbon content. Hence, 
 the American Society for Testing Materials recommends that "Thick 
 objects with less than 0.50% of carbon may be cooled completely in air, 
 of course, protected from rain or snow. Objects with 0.50% of carbon or 
 more, and thin objects with from 0.30% to 0.50% of carbon, may be cooled 
 in air if their cooling is somewhat retarded, as for instance, by massing 
 them together, as happens in the case of rails.' 7 The effect of the more 
 rapid cooling in air is to increase the strength and elastic limit, but lower 
 the reduction and elongation. In order to hasten the cooling, articles of 
 low carbon content are sometimes immersed in water after they have 
 become black in color. This method is then called water annealing. 
 
 Combination Methods of Cooling: Besides the three general 
 methods of cooling described above, various combination methods have 
 been employed with great success. Three of these, as directed by Bullens, 
 are as follows: 
 
 1. "Heat to slightly over Acs, air cool to just over Ar 1; return to a 
 furnace which is held at that temperature (about 725 C.), heat until 
 
546 THE TREATING OF STEEL 
 
 uniform, and then cool slowly. The latter heating should not be any 
 longer than is possible. This method will tend to prevent the formation 
 of large amounts of free ferrite, but will affect the pearlite, as there will 
 be slow cooling through the Ar x range. 2. Heat to slightly over the Ac3 
 range, air cool to just under the Ar x range, return to a furnace and heat to 
 730 C. and slow cool. This method will effect a greater toughening, if 
 the temperature has not been prolonged too greatly at the second heating. 
 3. Heat to slightly above Acg air cool to below Ar^, return to a furnace 
 heated at a temperature slightly below Ar (660 to 670 C.), hold at this 
 temperature until uniformly heated, and slow cool (in lime or air). By 
 permitting the steel to air cool to a temperature below the lowest trans- 
 formation, advantage is taken of any 'hardening effect' or retardation in 
 the transformation of austenite into a conglomerate of pearlite and ferrite. 
 This effect will increase with the percentage of carbon and the smaller the 
 size of the piece. The reheating to a temperature below the lower critical 
 range, if not prolonged, will neither change the grain size nor allow of the 
 coalescing of the excess ferrite or of the individual constituents of the 
 pearlite, but will form a mass of irresolvable and intermixed pearlite and 
 ferrite known as 'sorbite.' At the same time, however, it will give the 
 maximum combination of large ductility, good strength and excellent 
 machining properties. This method is of particular value in the annealing 
 of tool steels, in which it has given most excellent results." 
 
 Double Annealing consists in heating the steel to a temperature con- 
 siderably over the Acs point, cooling rapidly to some point below the 
 lower transformation range, then immediately reheating to a point slightly 
 under or over Aci, and finally cooling slowly. This method is employed to 
 relieve the most severe strains, which do not respond readily to ordinary 
 annealing. The high first annealing temperature effaces the strains, while 
 the rapid cooling prevents their returning. As this cooling tends to harden 
 the metal, the second process is necessary to soften it and refine the grain, 
 coarsened by the first operation, as much as possible. The second heating, 
 of course, may be to a temperature just above Acs, when even better results 
 should be obtained, provided softness is the chief end sought. 
 
 Box Annealing: In many instances, especially with tool steel, it is 
 important that the surface be protected from oxidation, or decarbonization. 
 Some furnaces are now designed so that the object being heated may be 
 surrounded by a reducing atmosphere, and so oxidation is prevented. Where 
 such furnaces are not provided, it is the practice to pack the object in a 
 metal box, called an annealing box, with some refractory material, such as 
 sand, ground mica, etc., in the case of low carbon steel, or with some reducing 
 substances, as for example a mixture made up of a little charcoal with 
 ashes, burned bone, etc., in the case of higher carbon steels, like the tool 
 steels, for example. 
 
ANNEALING 547 
 
 Annealing Hyper=Eutectoid Steels: In annealing hyper-eutectoid 
 steels, as in the case of hypo-eutectoid steels, one or more of these objects 
 are aimed at; (1) release of strains, (2) softening in preparation for machin- 
 ing, and (3) change of structure. The first object may be accomplished 
 by a simple reheating at temperatures considerably below those of the 
 critical range. The second and third objects are more difficult to attain, 
 for the treatment administered will be governed by the amount of the excess 
 cementite and the form in which it exists. Thus, if the cementite is partly 
 diffused, that is, does not exist as a net work or as spines and needles, and 
 the grain size is small, conditions that may generally be expected in high 
 carbon tool steels, the forging hardness may be largely removed by anneal- 
 ing at a temperature slightly under Ac. or between 600 C and 700 C. The 
 steel should not be kept at this temperature any longer than is necessary to 
 heat it thoroughly and uniformly throughout, as prolonged heating may 
 cause the excess cementite to coagulate. Such treatment will release the 
 strains and soften the steel sufficiently for machining. On the other hand, 
 if the grain is coarse, making a complete change in structure desirable, it 
 will be necessary to heat to a temperature in excess of the Aci-2-s point, or 
 above 725 C. For steels with a carbon content approximating 0.90%, such 
 heating will bring about a complete change of structure and give the finest 
 grain-size obtainable through annealing. For steels with a carbon content 
 considerably in excess of the eutectoid ratio the annealing may be done at 
 similar temperatures, provided the excess cementite is more or less in 
 solution. If the cementite is not in solution and a maximum refinement is 
 desired, the steel may be oil quenched from a temperature somewhat over 
 the Aci-2-s range, and subsequently annealed at a temperature just below 
 that range, or it may be normalized and annealed. 
 
 Normalizing and Spheroidizing: These are two processes applied to 
 hyper-eutectoid steels in particular, though normalizing is often applied to 
 hypo-eutectoid steels also. If the free cementite in the former steel exists 
 as a network or as spines, which would make the steel difficult/ to machine, 
 annealing at the usual temperatures (Aci-2-s) Wl ^ not affect this cementite, 
 but will simply refine the ground mass. In order to eliminate this free 
 cementite, it is necessary first to normalize, that is, quench the steel from 
 a temperature above that of the Ac cm range. Usually, air cooling from 
 a temperature of, say, 960 C., or 1000 C., will not permit the cementite 
 to recoagulate. Lower carbon steels are heated to about the same tem- 
 perature, but quenching is never required. Hence, in practice, normalizing 
 usually consists in heating the steel to the temperatures mentioned and 
 cooling simply in air. The annealing may then be carried out at a tem- 
 perature of 745 C. or over, to secure the refining of the grain size and 
 complete softening of the steel. The heating for annealing should be just 
 as short as possible in order to prevent the separation of the excess cementite 
 again. The method given above may be modified for hyper-eutectoid steel 
 by annealing at a temperature slightly under the lower critical range instead 
 
548 THE TREATING OF STEEL 
 
 of over it. This method, however, is subject to the objection that the steel 
 will not be refined, but will possess a large grain size on account of the high 
 normalizing temperature. But on the other hand, the lower annealing tem- 
 perature entirely prevents the formation of free cementite either as spines 
 or as a network, and the excess cementite is thrown out, under these con- 
 ditions, as little nodules or "spheroids," if the reheating temperature has 
 been near the end of the lower critical range. Spheroidal cementite may 
 also be obtained by cooling very slowly through the end of the Art trans- 
 formation range. Spheroidizing is a great help in the machining of high- 
 carbon steels. 
 
 SECTION II. 
 
 HARDENING. 
 
 The Hardening Operation : The operation of hardening as applied 
 to steels containing a sufficient amount of carbon consists fundamentally 
 of the two operations of heating to a suitable temperature and suddenly, 
 or rapidly, cooling. The heating may be accomplished in a number of 
 ways, varying from costly and specially designed furnaces and baths heated 
 with gas or electricity to the simple forge fire of the blacksmith; but the 
 cooling is always brought about by plunging the steel into a suitable liquid, 
 a'process called quenching. Let the means be what they will, in properly 
 hardened steel the original structure, as it existed before the hardening 
 process, such as coarse grain size, network, etc., has entirely disappeared 
 and has been replaced by a new structure, totally different from that of the 
 unhardened steel. To understand thoroughly the hardening process a close 
 study of the two operations by which these changes are brought about 
 should be made. 
 
 Heating for Hardening: The structural changes that accompany the 
 heating of steel through its critical ranges have already been briefly 
 described. Graphically, these changes are represented in the central part of 
 the accompanying diagram (Fig. 114), depicting the heating and cooling of a 
 steel with a carbon content of about .90%. From this evidence it will be 
 seen that the function of the heating is to bring about the proper change in 
 structure so as to obtain (1) the formation of the hard constituents of the 
 steel and (2) the smallest grain size, or highest refinement of the crystalline 
 structure. From what has already been said, these structural changes can 
 be obtained only by heating the steel above its critical range. Any attempt 
 at hardening it at a temperature inferior to this range results in only a very 
 slight, if any, increase of the hardness. Again, the metal should not be 
 heated much above the top range, for then its grain structure is coarsened, as 
 has been previously explained, also, and no additional hardness is imparted. 
 Clearly, the best temperature to which the steel should be heated is one 
 just above the critical range. The proper temperature to which plain 
 carbon steels should be heated is the same as for the true annealing of the 
 same steels. What has been said about the rate of heating and the influence 
 
HARDENING 
 
 549 
 
 of size of section in annealing also applies to heating for hardening. 
 If any difference, additional emphasis should be placed on the uniformity of 
 heating. The rule for heating may be put thus: Heat slowly, uniformly, 
 and thoroughly, to the lowest temperature, and no higher, that will give 
 the desired results. To meet these requirements, the final heating of steel 
 for hardening is often, and commendably so, conducted in baths of molten 
 lead or of the chlorides of sodium, calcium, potassium or barium. 
 
 Effect of cooling at different 
 rates from above the critical 
 range. 
 
 Effect of rapid cooling from 
 different points relative to 
 the critical range. 
 
 Legend: P=Pearlite, A=Austenite, M=Martensite, T=Troostite, 
 S=Sorbite. 
 
 PIG. 114. Diagram Depicting the Different Methods by Which the Five Different 
 Structural Constituents of Eutectoid Steel May Be Obtained. (After Sauveur.) 
 
 Cooling for Hardening: Thus it is seen that heating for both anneal- 
 ing and hardening are very similar, and that the changes wrought in the 
 microscopic constituents of the steel are the same in both cases. The main 
 difference in the two operations is found in the rate of cooling through the 
 critical ranges, at least. For hardening, this cooling must be very rapid, 
 whereas in annealing it was characterized as slow. The transitions attend- 
 ing the transformation of austenite to pearlite on slowly cooling through 
 the critical ranges have been described. It will be recalled that this trans- 
 formation is not instantaneous, nor is it direct, but takes place by stages 
 through transitional structures called martensite, troostite, and sorbite, 
 the order on slow cooling being from austenite, to martensite to troostite, 
 to sorbite, to pearlite. On heating, this order is reversed. It now may 
 be explained that the secret of the hardening process is revealed by the 
 fact that rapid cooling through the critical range may prevent this trans- 
 formation in part or in whole, depending on the rate of the cooling. Thus, 
 by the most rapid cooling, the steel at atmospheric temperatures is found 
 to consist almost entirely of austenite, while a little slower cooling pro- 
 duces martensite; still slower, troostite; slow, sorbite; and very slow, 
 pearlite. Incidentally, it may be remarked that the constituents found in 
 the steel after treating also depends on the temperature within the critical 
 
550 
 
 THE TREATING OF STEEL 
 
 range at which the rapid cooling begins and the carbon content of the steel, 
 because these constituents are formed at different temperatures and the 
 presence of carbon retards the transformation. Since the properties of the 
 cooled steel are imparted to it by the constituent which predominates, a 
 study of the characteristics of these transitional constituents will, there- 
 fore, be of value. For the sake of brevity and convenience, this knowledge 
 is here put down in tabulated form. 
 
 Table 60. Data with Reference to the Constituents of Hardened Steel. 
 
 Name 
 
 Nature 
 
 Occurrence 
 
 Temperature of 
 Stability 
 
 Structure 
 
 Physical 
 Properties 
 
 Austenite 
 
 Solid solution of 
 
 Obtained in 1.50% 
 
 Normally in 
 
 Polyhedral 
 
 Varies with car- 
 
 
 FesC in gamma 
 
 C. steels when 
 
 region between 
 
 grains 
 
 bon content. 
 
 
 iron, Carbon 
 
 quenched in ice 
 
 A-l, 2, 3, and A 
 
 
 Very hard but 
 
 
 content from 
 
 water from 1050 
 
 cm. Becomes 
 
 
 softer than 
 
 
 trace to 2%. 
 
 C. Occurs in steel 
 
 pearlite on cool- 
 
 
 martensite. 
 
 
 
 containing 12% 
 
 ing slowly past 
 
 
 
 
 
 Mn. and 25% Ni. 
 
 AIT more rapid 
 
 
 
 
 
 even after slow 
 
 cooling forms 
 
 
 
 
 
 cooling. 
 
 martensite. 
 
 
 
 Martensite 
 
 Solution of FesC 
 
 Obtained easily by 
 
 Normally at 
 
 Fibers or flat 
 
 Varies with car- 
 
 
 in beta iron, 
 
 quenching small 
 
 slightly lower 
 
 plates lying par- 
 
 bon content. 
 
 
 Carbon content 
 
 bodies of hyper- 
 
 temperature 
 
 allel to three 
 
 Hardest con- 
 
 
 from trace to 
 
 eutectoid steel, in than austenite. 
 
 sides of a tri- 
 
 stituent of steel 
 
 
 1%. 
 
 cold water. More 
 
 Changes into 
 
 angle. 
 
 of eutectoid 
 
 
 
 difficult to obtain 
 
 troostite. 
 
 
 composition. A 
 
 
 
 in low carbon 
 
 
 
 little less hard 
 
 
 
 steels. Chief con- 
 
 
 
 than cementite. 
 
 
 
 stituent of hard- 
 
 
 
 
 
 
 ened carbon tool 
 
 
 
 
 
 
 steels. 
 
 
 
 
 Troostite 
 
 Mixture of FegC 
 
 Obtained by re- 
 
 Normally at 
 
 Slightly granular 
 
 Intermediate be- 
 
 
 in beta iron 
 
 heating marten- 
 
 lower temper- 
 
 with sorbite and 
 
 tween marten- 
 
 
 crystallized 
 
 sitic steel to 400 
 
 ature than mar- 
 
 martensite in- 
 
 site and sorbite. 
 
 
 FesC and crys- 
 
 C. or on cooling 
 
 tensite. 
 
 termingled and 
 
 Not so hard as 
 
 
 tallized alpha 
 
 relatively slowly 
 
 
 with f errite and 
 
 martensite but 
 
 
 iron. 
 
 through critical 
 
 
 cementite in 
 
 stronger and 
 
 
 
 range. Found in 
 
 
 hypo-and 
 
 more ductile. 
 
 
 
 center of large 
 
 
 hyper -eutectoid 
 
 
 
 
 objects quenched 
 
 
 steels. 
 
 
 
 
 in water. 
 
 
 
 
 Neither pearlite nor sorbite are, strictly speaking, constituents of 
 hardened steel, but the latter, on account of its position in the transforma- 
 tion scale, forms the connecting link between hardened and annealed steel, 
 hence may occur in both. The nature and properties of sorbite have already 
 been given, and it should here be recalled that it is the toughest constituent 
 of steel. The transition from austenite to pearlite is admirably illustrated 
 in the preceding diagram of figure 114. 
 
HARDENING 551 
 
 Cooling or Quenching Media: Since the rate of cooling controls 
 the hardening process, the selection of the proper quenching medium is a 
 matter of much importance. The withdrawal of heat, the only function 
 of a quenching liquid, from the metal immersed in it depends upon its 
 quantity, its specific heat, its conductivity, its viscosity, its volatility, its 
 latent heat of vaporization, and, to some extent, its initial temperature. 
 The quantity and specific heat of the liquid control the quantity of heat 
 the bath can absorb with a given rise in temperature. The viscosity affects 
 the flowing properties of the liquid, hence the convection of heat by it, and 
 therefore is a factor in the cooling properties of the liquid, for it is by con- 
 vection and conduction that the heat is carried away from the steel to 
 distant parts of the bath. The volatility indicates the temperature at 
 which the liquid will become a vapor and form bubbles of gas on the 
 surface of the steel, which tend to retard the cooling. Evidently, if the 
 latent heat, or heat of vaporization, is high the volatility may be relatively 
 low, for then much heat is required to change the liquid to the gaseous 
 state. Since the speed at which heat is transferred from one body to 
 another varies directly as the difference in their temperatures, it is evident 
 that the initial temperature of the quenching liquid must affect the rate 
 of cooling. Thus, water, one of the most efficient media for rapid cooling 
 in use commercially, has high specific and latent heats, and its viscosity 
 is very low, properties favoring its cooling power, while its volatility and 
 conductivity are both low, properties against it as a quenching agent. 
 Then just next to water in cooling power comes mercury, which has a lower 
 specific heat, a lower heat of vaporization and higher viscosity than water; 
 but it is less volatile and is a very much better conductor of heat. Thus, 
 while the one owes its cooling power to one set of properties, the other is 
 almost as efficient because of an altogether different set. A solution of 
 salt in water, brine, is a little more rapid than water, while water sprayed 
 under pressure upon the metal is a quicker cooling agent than either water 
 or brine. After water, the chief quenching media in use commercially are 
 the oils, all of which are much slower than water. An interesting experi- 
 ment reported by Messrs. Matthews and Staggl is intended to show the 
 quenching properties of the various liquids that are employed in com- 
 mercial hardening. In this experiment a suitable test piece of steel was 
 carefully heated to 1200 F. and quenched in 25 gallons of the medium 
 under examination, and the time required to cool the steel to 700 F. noted 
 with a stop watch, as well as the rise in temperature of the medium. In 
 each medium this operation was repeated successively until the medium 
 had either reached its boiling point or a temperature of 250 F. The results 
 of this experiment were then plotted somewhat as shown in the accom- 
 panying diagram. 
 
 Combination Methods of Quenching: Besides the straight water or 
 oil quenching and water spraying, many special quenching methods and 
 media have been tried. Many of the latter are fakes, but the three follow- 
 
 iSee "Factors in Hardening Tool Steel" by Matthews and Stagg. American 
 Society of Mechanical Engineers, 1915. 
 
552 
 
 THE TREATING OF STEEL 
 
 ing special methods of quenching have proved of great value. Thus, when 
 high tensile strength is required, yet on account of the size of the piece or 
 the chemical composition manganese too high, for example water 
 quenching is unwise, the bath of water may be covered with oil to an equal 
 depth, so that the piece upon being lowered into the bath is partly cooled 
 in this oil, which then forms a film over the surface that retards the cooling 
 by the water somewhat. This method is sometimes applied to large 
 forgings, such as axles. For small tools a thin film of oil on the water 
 suffices. Another method, used by some tool hardeners, consists of first 
 plunging the tool into water to remove a part of the heat, then into oil till 
 the cooling is complete. Information as to what is aimed at by this method 
 is not at hand, but it is evident that the method is not so severe as straight 
 water cooling. Where great toughness with little hardness is required, the 
 article may be plunged into and forcibly submerged in molten lead, as this 
 manner of quenching produces sorbite in steels under the eutectoid in 
 composition. 
 
 2 1 3 457 6 8 
 
 10 11 
 
 250 
 
 200 
 
 150 
 
 o 
 
 g 100 
 
 a so 
 
 8 9 
 
 BW 
 
 50 
 
 100 
 
 150 
 
 200 
 
 Time in Seconds Required to Cool Test Piece from 1200 F to 700 F. 
 Legend : 
 
 6 New Bleached Fish Oil. 
 7 New Cotton Seed Oil. 
 
 Cotton Seed. 
 
 B Brine. 
 W City Water. 
 
 1 New Fish Oil. 
 
 2 No. 2 Lard Oil. 
 
 3 Lard Oil in Use Two Years. 9 Mineral Tempering Oil. 
 
 4 Boiled Linseed Oil. 10 Dark Mineral Tempering Oil. 
 
 5 Raw Linseed Oil. 11 Very Viscous Tempering Oil. 
 
 NOTE. 10 and 11 are similar to cylinder oils. 
 
 FIG. 115. Diagram Illustrating Approximately the Quenching Power of Various 
 Liquids. (Data by Messrs. Matthews and Stagg). 
 
HARDENING 553 
 
 Manher of Quenching: Much skill is required on the part of the 
 operator in the quenching of steel to prevent cracking and warping. Both 
 defects are due to unequal or non-uniform cooling of the different parts of 
 the piece, and are more liable to occur, for obvious reasons, in bodies of 
 large size or of irregular section. It is to overcome this danger that large 
 axles are hollow bored before treatment. All large sections, if solid, must 
 be reheated immediately after hardening in order to relieve the internal 
 stresses and strains, else incipient fractures will result. Warping will 
 always occur in small sections, if the quenching is not uniform. As warping 
 more often results when the piece is plunged into the quenching bath at an 
 angle, it is always best to quench vertically, in the direction of greatest 
 length, whenever such procedure is possible. 
 
 Progressive Hardening: Progressive, or differential hardening, is 
 accomplished by quenching only a part of the object. In such a method the 
 heat is slowly withdrawn from the part furthest from the quenching liquid, 
 but more and more rapidly as the part quenched is approached, so that 
 the steel becomes progressively softer and tougher from the hardened part. 
 By withdrawal of the piece before cooling is complete, the heat in the 
 unquenched part may be made to temper the hardened portion. In this 
 method care is needed to avoid hardening, or quenching, rings, which form 
 if the piece is held in the quenching bath at a uniform depth. To avoid 
 them the piece should be raised and lowered during the quenching. This 
 method is employed in treating such tools as anvils, die blocks, edged 
 tools, pointed tools, etc. 
 
 Hardening Eutectoid Steels (C. .80 to 1.00%): Steel of eutectoid 
 composition possesses the maximum hardening power, that is, the difference 
 in hardness between the quenched and unquenched article of eutectoid 
 composition is greater than that in any other grade of the plain steels. So, 
 this statement does not mean that quenched eutecoid steels are the hardest 
 steels, for hyper-eutectoid steels may show much greater hardness both 
 before and after hardening than eutectoid steels, due to the presence of 
 free cementite or to more highly carbonized martensite, but their gain in 
 hardness on quenching is less. It is clear that eutectoid steels should be 
 hardened from a temperature just above Aci, that is, 750 to 800 C., for 
 at this temperature the carbon, being in solution and thoroughly diffused, 
 possesses its full hardening power, and the grain structure is at its finest. 
 For the maximum hardness, the metal should then be quenched as suddenly 
 as possible in water, by which treatment the austenite is changed into a 
 fine grained martensitic or troostito-martensitic structure, depending upon 
 the size of the article and other incidental conditions. In order to avoid 
 the danger of cracking, many operators will prefer to quench in oil, when 
 troostite may be the predominating constituent if the piece is of large size. 
 
 Hardening Hyper=Eutectoid Steels (1.00 to 1.50%): Steels that 
 contain more than .90% carbon are also hardened by heating just above 
 
554 THE TREATING OF STEEL 
 
 Acs-a-i, or in other words, at the same temperature as steel of %utectoid 
 composition, the reason for which is readily seen by a little reflection. To 
 cite an example, suppose the steel to be hardened has a carbon content of 
 1.30%. Such a steel in its natural state is composed approximately of 93% 
 pearlite and 7% free cementite. To cause both pearlite and free cementite 
 to change to austenite would require the steel to be heated above Ac cm> 
 about 950, but such a high temperature would result in a decided and un- 
 desirable coarsening of the grain size, which is avoided by heating only 
 above Ac3~2-i. Besides, quenching from this lower temperature would 
 give a harder steel than would be obtained in the first instance. For, 
 supposing the quenching is such as to produce martensite, in the first case, 
 the hardened steel would be composed of martensite only, whereas in the 
 second instance it would be made up of 93% martensite and 7% free 
 cementite, which being harder than martensite, would impart additional 
 hardness to the quenched steel. When for any cause it is desirable to avoid 
 the presence of any free cementite, the metal may be heated above Ac C m 
 and cooled in molten lead, then reheated to slightly above Aci, and quenched 
 as usual. The quenching in lead prevents the re-formation of free cementite 
 and is not severe enough to cause cracking or warping, while the reheating 
 to Aci accomplishes the grain refinement so much to be desired in these 
 steels. 
 
 Hardening Hypo=Eutectoid Steels (C .30 to .80%) : Steel containing 
 less than .30% carbon cannot be materially hardened by any of the ordinary 
 commercial methods of quenching on account of the separation of ferrite 
 from the solution, which takes place to some extent even with the most 
 rapid methods of cooling. For hardening hypo-eutectoid steels with a 
 higher carbon content, two methods may be employed. First, the metal 
 may be heated slightly above Aci and quenched, when only the pearlite of 
 the steel will be affected, it being changed into martensite or troostite 
 according to the rate of cooling, and the free ferrite will undergo no refine- 
 ment at all. Evidently, the second and better plan is to heat the metal 
 above Aca-s, when its entire bulk changes into hardenable austenite, which 
 on quenching rapidly may be converted into martensite or at least troostito- 
 martensite. While this martensite, because of its lower carbon content, is 
 not so hard as the martensite formed from the pearlite in the first method, 
 the steel as a whole will be harder and certainly more uniform and of a 
 much finer grain structure, because the original network of coarse ferrite 
 will have been absorbed and refined. 
 
 SECTION IIIo 
 
 THE TEMPERING OF HARDENED STEEL. 
 
 The Tempering Process : Tempering, sometimes spoken of as drawing- 
 back or simply as drawing, consists in reheating the metal after hardening 
 to some temperature below the critical ranges, and may have for its primary 
 
TEMPERING AND TOUGHENING 
 
 555 
 
 object (1) the regulation of the hardness and brittleness of the steel, (2) 
 the toughening of it, or (3) the release of the hardening strains. In 
 tempering, the release of the internal stresses and strains set up by the 
 hardening process is always aimed at, whatever may be the other results 
 sought, for the metal is incapable of giving its best service as long as these 
 or similar strains exist. This result is usually accomplished by the heating 
 in the tempering process, for even heating to the temperature of boiling 
 water will relieve these strains to some extent. This fact is often taken 
 advantage of to relieve strains when it is not desirable to soften the steel 
 to the extent that higher heating would involve. 
 
 Legend 
 =Pearlite. 
 
 I ir 
 
 A=Austenite, M=Martensite, T=Troostite, S=Sorbite, 
 
 I. 
 
 Slowly cooled. 
 (Annealed) 
 
 II. 
 
 Quickly cooled. 
 (Hardened) 
 
 III. Reheating Hardened 
 Steel (Tempering). 
 
 FIQ. 116. Diagram Depicting the Constituents Formed on Slowly Cooling and 
 Quickly Cooling Steel and on Reheating Hardened Steel. (After Sauveur.) 
 
 Nature and Theory of Tempering: According to the retention 
 theories of hardening, which are among the most plausible ones advanced 
 to account for the hardening and tempering of steel, hardened steel is in 
 a state of unstable equilibrium, or strain, and, therefore, is ever tending to 
 assume a more stable form or condition, which tendency implies a return of 
 the iron to the alpha form and of the carbon to the condition of cement 
 carbon, or segregated cementite. In hardened steel, this transition is 
 prevented by the rigidity of the metal, which is reduced by reheating. 
 The extent of this reduction of the rigidity being in proportion to the extent 
 of the reheating, the higher the temperature the greater will be the extent 
 of the transformation. Thus, supposing the hardening operation has 
 arrested the transition of austenite at the martensitic stage, this martensite 
 will begin, at a temperature some 150 C. above that of the atmosphere to 
 change into troostite, and the transformation will continue with the rising 
 temperature until the martensite has passed entirely into troostite, which 
 
556 THE TREATING OF STEEL 
 
 result is no sooner accomplished than the troostite in turn begins to change 
 into sorbite. When the temperature has been raised to a point near that 
 of the lower critical range, sorbite, to the exclusion of other constituents, 
 will predominate the structure. Any higher heating, then, carries the 
 transformation into the critical ranges, where the order of transition is 
 reversed, sorbite passing into troostite and then to martensite, which, as 
 the temperature, on rising, emerges from the range, becomes austenite. 
 The preceding diagram, Fig. 116, copied after Sauveur, will aid in under- 
 standing these changes when they take place under different conditions. 
 Evidently then, all heating for tempering is conducted below the critical 
 range. By properly adjusting the temperature the transition described 
 above may be arrested at any stage desired, and any combination 
 of physical properties of which the steel is capable may be obtained. It 
 is clear that the manner of cooling from the tempering temperature is 
 immaterial, though for the sake of speed or convenience quenching or air 
 cooling is the general practice in heat treating shops. 
 
 Methods of Determining Tempering Temperatures: The original 
 method of estimating tempering temperatures is by color. Thus, if a piece 
 of hardened steel is brightened or polished with a piece of emery, sand stone, 
 or other suitable means and is then slowly heated in contact with air, the 
 color of the brightened surface will, due to the formation of a surface film 
 of oxide, undergo a series of color changes, called temper colors, ranging from 
 faint yellow to blue, which will be characteristic of the different temper- 
 atures reached by continued application of heat. That these colors are 
 indicative of a known temperature or at least a definite condition of the 
 hardened steel is generally accepted; but it is evident that the method is 
 subject to the objection that differences in distinguishing different colors, 
 or shades of color, is bound to occur among different operators. Distinction 
 of these colors is also affected by different light conditions. The same 
 temper will not give the same color in a dimly lighted room as in a well 
 lighted one. While these shades are hard to describe, the color correspond- 
 ing to the same temperature often being differently described by different 
 individuals, the following table will give some idea of the colors corre- 
 sponding to the different temperature changes. 
 
 Table 61. Tempering Colors and Temperatures Corresponding 
 
 to Them. 
 
 Pale yellow '. 220 deg. C. Pale blue 297 deg. C. 
 
 Straw 230 " " Dark blue 316 " " 
 
 Golden yellow 243 " " Red in the dark 400 " * 
 
 Brown 255 " Red indirect sunlight. 525 " " 
 
 Brown dappled with purple . . 265 " " Red in sunlight 580 " " 
 
 Purple 277 Dark Red 700 " " 
 
 Bright blue 288 " " 
 
TEMPERING 557 
 
 That the color method has its limitations is now well established, and 
 so other methods are being developed on a more scientific basis. These 
 methods involve the use of sand baths or liquid baths, such as oil, 
 molten lead, or alloys, fused salts, etc., for heating the steel and the use 
 of pyrometers for controlling the temperature, and aim at the elimination 
 of the personal equation in the results obtained. Seeing that the tempering 
 action often takes place very rapidly and that a difference of 15 or 20 
 of temperature will often spell success or failure, such appliances would 
 appear to be a very necessary part of the equipment of the modern heat 
 treating shop. 
 
 Influence of Time in Tempering: From what has been said about 
 the transformations wrought in the tempering process, the reader might 
 infer that the time the steel is kept at the tempering temperature would 
 exert no influence. Lest such should be the case, occasion is taken to 
 explain that, contrary to this inference and, in fact, to the common belief, 
 maintaining the metal at the tempering temperature for a considerable 
 length of time will result in producing additional tempering. This fact is 
 evidenced by the change of the tempering colors at constant temperature. 
 Thus, it has been ascertained that the temper color, instead of remaining 
 the same at a given temperature, advances in the tempering color scale 
 as it would if the temperature were being raised, which phenomenon would 
 appear to indicate that the temper colors are indicative of the tempering 
 condition of the steel rather than of the temperature. For example, by 
 heating a steel to 277, where its temper color is purple, and keeping it 
 there till its color is bright blue, a temper corresponding to the temperature 
 288 is obtained. According to some investigators, however, the temper 
 will not follow the color to the end. They maintain that each temperature 
 has a maximum temper effect, which is reached quicker and quicker as 
 the temper temperature is raised. 
 
 Physical Properties Affected by Tempering: It is to be remembered 
 that besides the hardness, the other physical properties of the steel are 
 likewise affected by tempering. Thus, as the hardness and brittleness are 
 decreased, the tensile strength and elastic limit will follow the hardness 
 closely and be correspondingly decreased, while the ductility, i. e., elonga- 
 tion and reduction in area, will be increased, though not following in the 
 wake of the hardness with the same regularity as the tensile strength and 
 elastic limit. (See table 62, page 561) 
 
 Tempering the Steels of Different Structural Composition : Seeing 
 that the hardening process has developed a certain structure in the steel, 
 it may be well, in turning to the practical application of the principles 
 and theories of tempering the hardened steels, to consider this phase of 
 the subject from the standpoint of their structural composition. The 
 accompanying diagram is intended to depict the tempering of all the 
 hardened steels. 
 
558 
 
 THE TREATING OF STEEL 
 
 Tempering Austenitic Steels: As has already been shown, austenite 
 does not occur in steels hardened by any of the commercial methods. 
 Hence, a lengthy discussion of the tempering of austenitic steel is out of 
 place in this study. It may be pointed out, however, that instead of passing 
 
 Critical 
 Range 
 
 200 
 
 X II III IV V VJ, 
 
 Legend: A=Austenite, M=Martensite, T=Troostite, S=Sorbite. 
 Fie*. 117. Diagram Depicting the Tempering of Hardened Steel. (By Sauveur.) 
 
 into martensite then to troostite, as shown at I in the diagram, austenite 
 may on tempering, pass directly into troostite as indicated at II. The 
 diagram shows that austenite begins to be transformed at a very low tem- 
 perature, being completely converted into martensite at 200 C. or into 
 troostite at 400 C. 
 
 Tempering Martensitic Steels: If steel of high carbon content has 
 been fully hardened by quenching rapidly, as in water, it consists mainly 
 of martensite, if other conditions were at all favorable. This constituent 
 is more stable than austenite, and on tempering will begin to change to 
 troostite below 200 C. and this transformation is complete at about 
 400 C., as shown at III on the diagram. Recalling the properties of 
 troostite, it will be seen, then, that tempering between these two points 
 results in a material decrease of the hardness and brittleness accompanied 
 by a decrease also of tensile strength and elastic limit, while some increase 
 in the ductility will be noted. It is to be observed that just as martensite 
 predominates in the structure of hardened steels, and pearlite, of annealed 
 steels, so the presence of troostite indicates tempering, either as a separate 
 reheating process or by regulating the cooling in the hardening operation, 
 as for example, in oil quenching, which is often called oil tempering on this 
 account. 
 
 Tempering Troostitic Steels: Commercially hardened steels, 
 especially those quenched in oil, will sometimes show large proportions 
 of troostite. From the diagram it will be seen that to temper this steel 
 will require a temperature of at least 400, the temperature at which it 
 begins to be changed into sorbite. At 600 C. the transformation of 
 
TOUGHENING 559 
 
 troostitic sorbite is complete. Tempering here results in a marked tough- 
 ening of the steel with loss of much of the hardness. Hence, as explained 
 below, tempering between these temperatures is called toughening by 
 many of the heat treating experts. However, troostite steels produced 
 by quenching often contain considerable martensite. When such is the 
 cas,e the steel may be softened by reheating to about 400 C., when the 
 martensite will be destroyed. 
 
 Sorbite: As this constituent occurs at the top of the tempering 
 range, sorbitic steels cannot be tempered. Furthermore, as these steels 
 are not produced by the regular hardening methods, they are not properly 
 considered in connection with tempering. 
 
 SECTION IV. 
 
 THE TOUGHENING OF STEEL 
 
 Toughening is the term applied to certain treatments given usually to 
 steels of medium carbon content (C. .35% to .60%) in which strength and 
 toughness, rather than hardness and toughness, are the properties sought. 
 It is a treatment applied to railroad axles, piston rods, and other articles 
 subjected to fatigue, impact and dynamic stresses in service. As practiced 
 by the Carnegie Company, toughening consists in heating the steel to 
 temperatures varying from 775 C to 850C, depending upon the chemical 
 composition, quenching in oil or water, and then drawing back to such 
 high temperatures, 450 to 650 C, thai little, if any, of the hardness due to 
 the quenching remains. 
 
 Benefits of Toughening: Compared with annealing, toughening has 
 an advantage in that both the strength and ductility of the steel may be 
 increased to the limits of which the steel is capable. In annealing, strength 
 is sacrificed for ductility, but in toughening, the relation of these properties 
 may be nicely controlled. The effect of the quenching operation is to give 
 the greatest refinement of the grain and to develop the maximum strength 
 of which the steel is capable. The effect of the draw back is to relieve all 
 strains due to the quenching, and, without coarsening the grain, develop 
 the ductility, which will gradually increase, with a partial loss in strength, 
 of course, as the temperature of the draw back is raised. Between 500 and 
 600 C the draw back produces a steel composed entirely of sorbite, which 
 is the structure that gives the highest combination of strength and 
 ductility. The pearlite produced by the usual annealing methods is both 
 less strong and less ductile than sorbite. One feature of the toughening 
 operation is to increase the ratio of the elastic limit to the ultimate, 
 or tensile, strength. Thus, while in natural and annealed steels of toughening 
 grade this ratio is approximately 3:6, in properly toughened steel it is 
 about 4:6. In view of the fact that it is the elastic limit that actually 
 measures the working strength of the steel, this effect of toughening is 
 worthy of careful consideration. 
 
 Quenching for Toughening: As indicated above, either oil or water 
 is used as a quenching medium for toughening. Both media have their 
 
560 
 
 THE TREATING OF STEEL 
 
 Ladle 
 Analysis 
 
 Treatment 
 
 Mechanical 
 Properties 
 
 Structure 
 
 
 
 
 . 
 
 ^h^ 
 
 C. .49% 
 Mn. .66% 
 P. .020% 
 
 as 
 forged 
 
 Ultimate 
 Strength 
 96,370 Ibs. 
 
 Elastic 
 Limit 
 49,310 
 
 m 
 
 " 
 
 S. .026% 
 
 
 Elongation 
 
 in 2" 
 20.5 % 
 
 |9Bp^ 
 
 
 
 
 Reduction 
 of Area 
 
 34.7% 
 
 
 
 
 
 
 x 1 00. Pearlite Large 
 
 Grain Size. 
 
 C. .48% 
 
 Annealed 
 
 Ultimate 
 Strength 
 85,040 Ibs. 
 
 ^^s^ 
 
 ' 
 
 / 
 
 ^K 
 
 Mn. .54% 
 
 Heated 
 
 
 A 
 
 
 P. .020% 
 
 to 830 C 
 
 Elastic 
 Limit 
 44,920 Ibs. 
 
 si^P3t 
 
 iCjp 
 
 S. .036% 
 
 and cooled 
 in air 
 
 Elongation 
 in 2" 
 
 23% 
 
 Reduction 
 of Area 
 
 37.8% 
 
 '^fe 
 
 ffg/J 
 
 
 
 
 ^^^^-isJm.'JsZ^ 
 
 i***^ 
 
 
 
 
 x 100. Sorbitic Pearlite Grain Size Good. 
 
 C. .49% 
 
 Quenched 
 and 
 tempered 
 
 Ultimate 
 Strength 
 97,200 Ibs. 
 
 ^ fl 
 
 ,' >X >v 
 
 Mn. .66% 
 
 
 
 
 
 P. .020% 
 S. .026% 
 
 Heated to 
 
 825 C and 
 quenched 
 in water 
 
 Elastic 
 Limit 
 62,720 Ibs. 
 
 
 
 
 
 Elongation 
 
 in 2" 
 
 
 
 
 
 25% 
 
 
 
 
 Drawn back 
 at 585 C 
 
 Reduction 
 of Area 
 
 59.8% 
 
 ^^^HHpl 
 
 &^ 
 
 
 
 
 x 100. Sorbite Grain Size Excellent 
 
EFFECT OF HEAT TREATMENT 
 
 561 
 
 advantages and disadvantages. Water is the more rapid and drastic 
 medium, and on this account, is more liable to develop cracks in the steel, 
 hence some heat treaters recommend that only oil be used. On the other 
 hand, a deeper penetration of the effects of the quenching and greater 
 tensile strength and elastic limit are obtained with water than with oil, 
 thus making it easier to meet specifications calling for high tensile proper- 
 ties. Therefore, others will prefer water quenching under certain con- 
 ditions. In selecting the quenching medium, it is evident that much 
 depends upon the article, its shape, size, and the grade of steel it is made of, 
 and much upon the skill of the operator. All these features of toughening 
 are brought out in Fig. 118 and table 62. 
 
 Table 62 1 1 Illustrating the Effect of Various Heat Treatments upon 
 the Mechanical Properties of Medium Carbon Plain 
 Steels. 
 
 Chemical Composition; C. .38%, Mn. .55%, Si. .05%, P. .024%, 
 
 S. .050%. 
 Description of Pieces Treated; one inch rounds, 29 inches long; 
 
 14 pieces, all from same billet. 
 
 Description of Test Pieces; One test piece from each of the 14 
 pieces, turned to a diameter of -J/ inch, as in Fig. 47. 
 
 Heat Treatment 
 
 Physical Tests 
 
 Hardening and 
 Refining Deg. C 
 
 Anneal 
 and Draw, 
 Deg. C 
 
 Tensile 
 Strength 
 
 Elastic 
 Limit 
 
 Elongation 
 in 2"% 
 
 Reduction 
 in area % 
 
 Brinell 
 Number 
 
 Scleroe- 
 copeTest 
 
 As Rolled 
 
 
 85,000 
 
 50,000 
 
 30,0 
 
 48.9 
 
 163 
 
 25 
 
 Heated to 760 
 
 and' 
 
 
 
 
 
 
 
 cooled in f 
 
 urnace 
 
 74,000 
 
 42,500 
 
 32.0 
 
 54.7 
 
 134 
 
 23 
 
 Heated to 815 
 
 427 
 
 100,000 
 
 67,000 
 
 21.0 
 
 53.9 
 
 179 
 
 30 
 
 quenched in oil 
 
 482 
 
 98,000 
 
 66,000 
 
 23.5 
 
 52.8 
 
 170 
 
 29 
 
 
 
 538 
 
 90,000 
 
 59,000 
 
 26.5 
 
 54.7 
 
 170 
 
 29 
 
 
 
 593 
 
 89,000 
 
 58,000 
 
 26.5 
 
 63.5 
 
 170 
 
 29 
 
 
 649 
 
 75,000 
 
 53,000 
 
 33.5 
 
 64.7 
 
 156 
 
 27 
 
 
 704 
 
 71,000 
 
 51,000 
 
 34.0 
 
 59.3 
 
 137 
 
 25 
 
 Heated to 815 
 
 427 
 
 110,000 
 
 81,000 
 
 19.0 
 
 46.0 
 
 223 
 
 35 
 
 quenched in 
 
 482 
 
 103,000 
 
 71,000 
 
 22.5 
 
 54.7 
 
 192 
 
 33 
 
 water 
 
 538 
 
 95,000 
 
 68,500 
 
 23.5 
 
 61.6 
 
 187 
 
 32 
 
 
 593 
 
 89,000 
 
 63,000 
 
 28.5 
 
 63.0 
 
 179 
 
 29 
 
 
 649 
 
 82,000 
 
 57,500 
 
 30.5 
 
 65.4 
 
 156 
 
 26 
 
 
 704 
 
 73,000 
 
 51,000 
 
 34.0 
 
 59.8 
 
 143 
 
 25 
 
 
 
 
 
 
 
 
 
 !The data for this table, as well as that for tables 65, 68, and 69, were supplied 
 by Henry Wysor, of the Bethlehem Steel Company. 
 
562 TEE TREATING OF STEEL 
 
 SECTION V. 
 
 CASE HARDENING. 
 
 The Process of Carburizing Iron: It is a well known fact that if 
 a bar of wrought iron or soft steel be heated to a temperature close to or 
 above the critical range in contact with carbonaceous materials and within a 
 suitable receptacle from which air is excluded after the heating is started, 
 the bar of metal will absorb carbon, the amount so absorbed depending 
 upon the time the bar is kept in contact with the carbon, the temperature 
 maintained during the operation, the nature of the carbonaceous material, 
 and the initial composition of the bar itself. This characteristic of iron 
 with respect to carbon was first made use of in the manufacture of steel 
 from wrought iron by the cementation process, then for the surface car- 
 burizingof armor plate, and finally for case hardening, or surf ace carburizing, 
 smaller articles. Essentially, case hardening is but a special application 
 of the cementation process, in which the articles treated are but partially 
 carburized and the case extends but a short distance from the surface, 
 leaving the central portions of the articles unchanged in chemical com- 
 position. Thus, while the chief principle of carburizing iron has been 
 known and made use of for years, it is only within recent times that case 
 hardening has become a process of commercial importance. 
 
 Application of Case Hardening: The result sought, in most cases 
 where case hardening is employed, is the production of a hard, wear-resisting 
 surface upon a tough, ductile core. It is, therefore, applied to many tools, 
 to gears, to ball bearings and to various parts of automobiles, airplanes, 
 bicycles and the like in fact, wherever a combination of toughness and 
 lightness with a wear-resisting surface is desired. On account of the wide 
 application of the process and the fact that the art has not yet reached 
 the stage of fullest development, a wide variation in the methods of 
 applying the process is to be expected. This condition makes the subject 
 a difficult one to deal with briefly and at the same time satisfactorily. 
 In the following paragraphs, an attempt has been made to give a summary 
 of the facts as revealed by the work of many investigators who have pub- 
 lished or otherwise made known the results of their experiments and experi- 
 ence. Only general features are thus dealt with, because the working out 
 of details is largely a matter to be determined by experience. 
 
 The Two Periods of the Case Hardening Process: In order to 
 obtain the greatest benefits from case hardening, it is necessary that the 
 carburization be succeeded by proper heat treatment, or that the carburizing 
 process be considered as a part of a special heat treating process. The 
 chief factors that control the carburization have already been enumerated. 
 Since a relatively high temperature is employed in the carburizing process 
 and the cooling, at the end of the carburizing period, is usually slow, the 
 steel, as a whole, is in its softest condition, and has a large grain structure. 
 
CASE HARDENING 563 
 
 Therefore, the heat treating part of the process must combine a grain 
 refining operation for low carbon steel with a hardening and- grain refining 
 treatment for high carbon steels. 
 
 Kinds of Steel Suitable for Case Hardening: In general, the com- 
 position of the steel for case hardening is limited by the desire to eliminate 
 any elements that produce brittleness in the core, and also any that tend 
 to retard the absorption of carbon by the steel. The elements that may 
 be permitted and those that should be avoided in steel for case hardening 
 will, therefore, be easily recognized after a study of the two succeeding 
 chapters which are devoted to the effects of the elements upon the prop- 
 erties of steel. For convenience, however, a list of the elements with data 
 concerning their case hardening properties is given here. 
 
 Carbon: Since the tendency of carbon, especially when present in 
 any amount greater than .25 per cent., is to increase the brittleness, .30 
 per cent, is the limit to which the carbon in steel for case hardening may 
 rise. Therefore, the carbon content of steels for this purpose ranges from 
 .08 to .25 per cent. For ordinary purposes a carbon content of .08 to .15 
 per cent, is most satisfactory. However, as steel of this grade is difficult 
 to machine so as to give a smooth surface, and a fairly strong core is de- 
 sirable for some kinds of work, a carbon content of .15 to .25 per cent is 
 frequently specified. Needless to say, the higher carbon grade requires 
 more care in treatment than the low carbon grade. 
 
 Manganese: While manganese increases the ability of the steel to 
 absorb carbon, on account of its tendency to make the case brittle and 
 sensitive to shock, the per cent, of this element is generally kept below 
 .50, though steel with a manganese content of .70 per cent, is sometimes 
 employed. 
 
 Silicon: When the silicon content is raised to 2.0 per cent., the steel 
 refuses to absorb carbon, hence the steel should be kept as free of this 
 element as possible. The highest limit for silicon to give commercially 
 satisfactory results is about .30 per cent* 
 
 Phosphorus and Sulphur: The phosphorus and sulphur content of 
 the steel should be as low as possible, not over .05 per cent, for each of 
 these elements. 
 
 Nickel: Nickel strengthens and toughens steel, but retards the car- 
 burization of the metal. The rate of penetration is lowered in proportion 
 to the amount present, so that in a steel with a nickel content of 5 per cent., 
 the rate of penetration, using solid carburizing materials, is about half 
 that in the plain low carbon low manganese steels. But, offsetting this 
 disadvantage, nickel steels possess certain peculiarities of structure and 
 increased toughness, which make them desirable for carburizing purposes. 
 
 Vanadium: This element also lowers the rate of carbon penetration, 
 but since it is present in very small amounts, its action in this respect is 
 less pronounced than in the case of nickel. 
 
f.64 THE TREATING OF STEEL 
 
 Chromium: The low carbon chrome steels, especially those con- 
 taining about 0.5 per cent, chromium, are well adapted to case hardening, 
 for the chromium not only increases the rate of penetration and the con- 
 centration of the carbon in the case, but also materially reduces the grain 
 size. Furthermore, this amount of chromium does not harden the case 
 nor render it brittle beyond that which would be obtained by slightly 
 increasing the carbon content of the plain carbon steel. 
 
 The Carburizing Agent: Many investigations have been conducted 
 to determine just what the carburizing action is. Originally, it was held 
 that the carbon was absorbed directly at the surface of the metal and 
 there dissolved, the dissolved carbon being then disseminated towards the 
 interior. That dissolved carbon may move about, or diffuse, within the 
 metal is accepted, but it has been proved that carbon alone in contact with 
 iron has only a slight carburizing action and that, for commercial carburi- 
 zation, the presence of carbon bearing gases is necessary. By diffusing into 
 the steel where they may react with the iron, these gases act as carriers 
 of the carbon. The gases available for this purpose are carbon monoxide, 
 cyanogen, and gaseous hydrocarbons. Of these, carbon monoxide and 
 cyanogen bearing gases are the most effective, but the former gives a much 
 more uniform gradation of the carbon content from the exterior to the 
 interior of the case, and is to be preferred on that account. The reaction 
 by which carburization is effected with carbon monoxide is generally 
 assumed to be the following: 2 CO+3Fe=Fe 3 C+CO2. 
 
 Carburizing Materials: A great number of different carburizing 
 materials, consisting of gases, liquids and solids, have been tested by the 
 many investigators, and of these, solids are by far the most convenient for 
 the purpose as well as the most effective, when they are of the proper com- 
 position. In the use of solid carburizers, the chief essentials to success are 
 carbon in suitable form, a sufficiently high and properly regulated temper- 
 ature maintained for a proper length of time, and reasonable care as to the 
 details of preparing the carburizer and packing the articles therein. For 
 supplying the carbon, many different substances may be employed, such as 
 coke, wood charcoal, sugar charcoal, animal charcoal, charred bones, 
 charred leather, etc. Coke and wood charcoal are not as rapid as these 
 other forms of carbon. Care is required in using coke, bones, leather and 
 animal charcoal to guard against imparting phosphorus and sulphur to the 
 metal. The material selected should be ground or crushed to the proper 
 degree of fineness and to a fairly uniform size, after which it should be 
 sifted free of dust. 
 
 Packing and the Action of Charcoal Carburizer: Assuming that 
 charcoal is selected, the article or articles to be case hardened are packed 
 with the carburizer in a hardening pot or box of suitable size and shape. 
 The boxes may be of soft steel, wrought iron or cast iron. The walls should 
 be thin, about one-fourth inch in thickness, and of a size and shape that will 
 
CASE HARDENING 565 
 
 permit the rapid penetration of the heat, so that the lag in temperature of 
 the central part of the packed box behind the furnace temperature will be 
 as small as possible. The best shape is one that conforms to that of the 
 piece or pieces to be carburized. The article, or each of the articles, in a 
 charge should be placed so that it will be completely surrounded by a 
 layer of the carburizing material, about an inch in thickness. In case 
 it is desired to carburize only certain parts of the article, the parts 
 that are not to be carburized may be covered with asbestos cement, 
 or slaked lime or fire clay in the form of a paste. When the packing has 
 been completed, the open end of the box is closed with a neatly fitting 
 lid, which is pressed down firmly against the top layer of carburizing 
 material and fastened tightly in place, so that it will permit no displacement 
 of the pack or packing in handling. The small opening about the edge of 
 the lid is then luted with asbestos cement, clay, or a mixture of fire clay 
 and sand; then, the box is ready for charging. When the contents of the 
 box have reached a certain temperature in the furnace, the oxygen of the 
 air that fills the interstitial spaces of the packing reacts with the carbon, 
 giving carbon monoxide, which in turn reacts with iron to give iron carbide 
 and carbon dioxide gas, as previously described. The iron carbide, of 
 course, remains in the metal to form the case, while the carbon dioxide 
 is given off. When it comes in contact with the carburizing material, 
 CO gas is again generated, thus, CO2+C=2 CO. This CO then reacts 
 with iron as before. This cycle is made again and again, until the process 
 is stopped, or the iron becomes saturated with carbon. In the case of bones, 
 leather, and other animal or vegetable matter, other more complicated 
 reactions, due to cyanogen and hydrocarbon vapors given off by these 
 substances, occur in addition to the simple reactions resulting from the 
 carbon, as explained in connection with the use of charcoal. 
 
 Carburizing Mixtures and Compounds: These simple substances 
 may also be used as the base materials for various carburizing mixtures 
 designed to suit the conditions and the results desired. Thus, in the case 
 of thin cases, where it is desired to increase the speed or rate of penetration 
 and where the forming of a case of uniform thickness is essential, the follow- 
 ing mixtures have been recommended: 
 
 1 . Powdered wood charcoal with a little heavy hydrocarbon oil added. 
 
 2. Powdered wood charcoal, leather charcoal, and lampblack in the 
 proportion of 5, 2, and 3 parts, respectively. 
 
 3. Powdered wood charcoal, 7 parts, and animal charcoal, 3 parts. 
 
 4. Powdered wood charcoal, charred horn and animal charcoal, in 
 proportion of three parts of the first and two parts of each of the others. 
 
 The increased rate of carburization that may be obtained by the use 
 of these mixtures is due to the fact that they give off volatile hydrocarbons 
 and cyanogen compounds as well as carbon monoxide, and that these com- 
 pounds are capable of causing carburizing reactions independent of and in 
 addition to that involving carbon monoxide. 
 
566 THE TREATING OF STEEL 
 
 In addition to these, mixtures of wood charcoal with common salt or 
 with barium carbonate have been found very efficient and desirable car- 
 burizing materials. Just what part common salt may play in the process 
 is not known, but the action of barium carbonate is easily explained. At 
 the higher carburizing temperatures it is decomposed according to the 
 following reaction: BaCO3=BaO+CO2- The CC>2 thus generated is 
 immediately reduced by the hot carbon to CO gas, each volume of CO 2 
 giving two volumes of CO; thus, CO2+C=2 CO. The net effect of the 
 barium carbonate, then, is to increase materially the amount of the CO 
 available. By exposing the mixture, after use, to the air, the barium 
 oxide takes up CO2, forming barium carbonate again, so that with the ' 
 occasional addition of small amounts of charcoal the same mixture may be 
 used repeatedly. Another advantage secured in using the barium carbonate- 
 charcoal mixtures is that the danger of contaminating the steel with sulphur 
 is entirely avoided, as these materials may be obtained practically sulphur 
 free. The mixture that has been found to give the best results is one com- 
 posed of 40 parts of the carbonate to 60 parts of charcoal by weight. 
 
 When it is desired to obtain a thin case of high carbon content in a 
 very short interval of time, quick acting mixtures are used. The sub- 
 stances employed in these mixtures are wood charcoal, bituminous coal, 
 saw dust, charred leather, prussiate of potash, sal soda and common salt. 
 From these substances mixtures that will give various speeds of carburizing 
 may be made. For example, a mixture of 2 parts wood charcoal, 1 part 
 salt, and 3 parts saw dust is relatively slow in its action while a mixture 
 of 10 parts charred leather, 2 parts prussiate of potash and 10 parts saw 
 dust is characterized as very rapid. 
 
 Heating the Carburizing Pack: For heating up the charged carbur- 
 izing boxes, some form of gas fired muffle furnace is preferable. The 
 essential requirements of the furnace are that it must be capable of giving 
 a maximum temperature of at least 1000 C., any definite temperature 
 lower than 1000, and also be capable of maintaining these definite tem- 
 peratures uniformly throughout the heating chamber for periods of several 
 days at a time. In order to avoid the rapid oxidation and consequent 
 destruction of the carburizing boxes, a reducing atmosphere should be 
 maintained in the heating chamber, and furnaces constructed so as to effect 
 this result are most desirable. The furnace should be cold, or nearly so, 
 when the packed boxes are charged, and the heating, up to the carburizing 
 point, should be very gradual. The steel will thus have time to adjust 
 itself to the conditions; the pack will be uniformly heated throughout, so 
 that carburization will begin in all parts of the pack at the same time; 
 and the evolution and generation of gases, which begins at temperatures 
 slightly below 700 C., will not be too energetic. The temperature and the 
 length of time for carburizing depend on the depth and the carbon content 
 of the case desired, the carburizing material, and the character of the raw 
 
CASE HARDENING 567 
 
 iron or steel. In general, for a given set of materials, the higher the tem- 
 perature and the longer the time of carburizing, the greater will be the depth 
 of the carburized zone; and when solid carburizers are used, the same may 
 be said with respect to the maximum carbon content or carbon concen- 
 tration of the case. That the carburizing material may affect the speed of 
 the carburization has already been intimated in discussing carburizing 
 mixtures. A similar difference is also found in their action with respect to 
 the concentration of the carbon. Thus, while one carbunzer will give, for 
 example, a case with a surface hardness corresponding to .80% carbon at 
 870 C., 1.05% carbon at 900. C., etc., another will give only a .70% case 
 at 870 C. and a .90% case at 900 C. In treating ordinary carbon steel, 
 a temperature between 875 and 900 is considered best to avoid large grain 
 size and obtain the most satisfactory results. With this temperature 
 determined upon, the depth of the case and the concentration of the carbon 
 may be regulated by varying the time of carburizing and the composition 
 of the carburizing material employed. 
 
 Controlling the Temperature: In regulating the temperature of the 
 pack it should be kept in mind that the temperature of the furnace cannot 
 be relied upon to give the actual temperature of the interior of the pack. 
 The temperature of the latter, during the time it is being brought to heat, 
 will tend to lag behind that of the furnace, and after a temperature of 700 C. 
 is passed the chemical reactions within the pack itself may result in the 
 liberation or absorption of a quantity of heat sufficient to maintain its 
 temperature several degrees above or below that of the furnace. Evidently, 
 then, some means of ascertaining the temperature of the interior of the pack 
 is very desirable. For this purpose a pyrometer of the thermo-electric 
 type is admirably suited, because with this instrument the hot junction of 
 the thermo couple may be placed in the center of the pack as it is being 
 made up, or inserted through a small tube so placed. 
 
 Removal of the Articles from the Boxes After Carburizing: In 
 
 cases where it is desired to prevent the oxidation of the surface of the 
 articles treated, it is necessary either to permit them to cool nearly to 
 atmospheric temperature in the boxes or to quench them by emptying the 
 entire contents of the box, inverted and with its opening very close to the 
 surface, into the quenching liquid. Some materials are quenched from the 
 carburizing temperature for the purpose of hardening them, but in order to 
 refine the grain, which is coarse, due to the long period of exposure to a 
 relatively high temperature, and secure the greatest toughness combined 
 with greatest hardness, the carburized articles must be subjected to special 
 heat treating processes, in which case the articles may be removed from the 
 carburizing boxes at any convenient time and allowed to cool in the air 
 to atmospheric temperatures or at least to a temperature that gives a 
 black color. 
 
568 THE TREATING OF STEEL 
 
 Heat Treatment of Case Hardened Articles: The correct heat 
 treatment of case hardened articles involves a combination of methods 
 suitable to steels of different carbon content. Upon the core of low carbon 
 content there is superimposed a layer of high carbon steel, which may be 
 of hypo-eutectoid, eutectoid, or hyper-eutectoid composition, and the 
 treatments should be varied to correspond to these three different cases 
 and to the temperature at which the carburization was carried on. To 
 secure maximum refinement of grain in the core it is necessary to heat the 
 steel just above its Acg point, which for a .15% to .20% carbon core, is 
 a temperature near 900, and quench, preferably, in oil. As this temper- 
 ature is far above the Ac range of either a hypo-eutectoid or eutectoid 
 case, this treatment hardens the case but leaves its grain structure relatively 
 coarse. Therefore, the article should be reheated to a temperature slightly 
 above the Ac3-2- range of the case and again quenched in water or oil. 
 Finally, to prevent brittleness in the case and to remove strains, it is desir- 
 able to temper the steel at once by reheating to 200 or over, depending 
 upon the hardness it is desired the case shall retain. The temperature 
 mentioned would relieve strains but would reduce the hardness very little, 
 if any. Hyper-eutectoid cases require that the treatment described above 
 be modified to the extent that either the first reheating shall be above the 
 Ac cm range, or that the article be quenched from the carburizing temper- 
 ature, in order that the excess cementite may be retained in solution. The 
 further treatment may then be a repetition of that for hypo-eutectoid cases, 
 or merely a quenching from above the Ac3-2-i range (750 C.). This last 
 method leaves the core somewhat brittle, due to a large grain size, but 
 produces a surface of exceptional wear resisting properties. In any of these 
 cases, where the carburizing has been carried on at a high temperature 
 and has occupied a considerable period of time, double quenchings are 
 sometimes necessary to secure the best results. 
 
 Superficial Hardening: For the most superficial hardening and at 
 the same time the most rapid, such as is sometimes desirable for hardening 
 certain tools, cyanide of potassium or prussiate of potassium alone may be 
 used in either one of two ways. In one, the salt is melted and the article 
 to be hardened is brought to the quenching temperature by immersing it 
 in the fused salt, held at that temperature for a few minutes, the exact 
 time depending upon the amount or extent of the carburization desired, 
 and then quenched as for ordinary hardening, except that lime water should 
 be used to neutralize the poisonous cyanide. In the other method, the 
 article to be hardened is heated to the h ardening temperature and is then 
 sprinkled with the dry salt or plunged into a quantity of the dry salt. It 
 is then reheated to the hardening temperature and quenched, as in the 
 first method. Although often spoken of as such, this treatment is not a 
 true case hardening process. 
 
INFLUENCE OF ELEMENTS 569 
 
 CHAPTER III. 
 
 CONSTITUENT ELEMENTS OF COMMERCIAL CARBON STEEL 
 
 AND THEIR INFLUENCE UPON IJS 
 
 MECHANICAL PROPERTIES. 
 
 Introductory: Needless to say that a complete discussion of the 
 effects upon the properties of steel of all the elements that naturally may 
 be found in it or that may be added to it would be a very lengthy one, 
 indeed. Even a thorough study of the subject as limited by the title of 
 this chapter would involve an immense amount of labor on the part of the 
 writer and much time on the part of the reader to peruse it. The most 
 that is to be expected, therefore, in the following discourse is but a brief 
 summary of the opinions of the different authorities as presented in the 
 various text books, the trade papers, and the reports of conventions, and 
 some deductions and conclusions arrived at through personal experience. In 
 examining the information from these sources, the student is confronted 
 with much difference of opinion, which often results in much confusion of 
 thought. But a systematic search enables the student to arrive at the 
 conclusion that certain elements, like manganese, for example, are bene- 
 ficial; others, like oxygen, are harmful; some, like phosphorus and sulphur, 
 are of doubtful influence; while others may be beneficial or harmful, depend- 
 ing upon conditions. In this regard, it is important to note that opinion 
 at present is changing with respect to the influence of many of the elements. 
 This is particularly true of phosphorus and sulphur, both of which were 
 recently held to be injurious to steel under any conditions and at all times. 
 Now, however, these elements, far from being considered as foes to good 
 steel making, are, within certain limits, being looked upon as harmless to 
 the steel, and even as aids for certain purposes. With these things in 
 mind, an attempt has been made here to put down what appears to be the 
 truth concerning these elements as revealed after a study such as that 
 suggested above. 
 
 Properties of Iron: Since iron is the element that forms the base 
 material for the steel, the discussion of this subject is naturally begun 
 with a consideration of the properties of this element, though pure iron is 
 unknown commercially. As the physical and chemical properties of the 
 element will be found under the subjects of Physics and Chemistry and 
 the Heat Treatment of Carbon Steel, it is not necessary even to tabulate 
 them here. In this connection, special emphasis is to be laid upon the 
 strength and ductility of the element. Seeing that it is almost impossible 
 
570 INFLUENCE OF ELEMENTS 
 
 to obtain pure iron in sufficient quantity for testing, the determination of 
 these properties cannot be made directly. However, figures that appear 
 to be as near the true values as it is possible to get, have been assigned 
 for these properties by calculating from results of pulling tests upon the 
 purest forms of annealed or normalized commercial soft steels. After making 
 what would appear to be a proper allowance for the influence of the small 
 amounts of carbon that these steels contain, it has been established that 
 pure iron has an elastic limit of about 20000 pounds, a tensile strength, or 
 maximum stress, of 38000 to 40000 pounds, a reduction of area of 84%, and 
 an elongation, measured in 8 inches, of 51%. From these values it is seen 
 that pure iron is a very ductile substance, but weak as compared with steel. 
 
 Effect of Carbon : The influence of carbon upon iron is so character- 
 istic and beneficial that it is employed as the controlling element in regu- 
 lating the physical properties of all common steels. While this element is 
 capable of changing most of the physical properties of iron by uniting and 
 alloying with it, its most important influence is connected with the hard- 
 ness, strength, and ductility of the metal. Its effect upon these properties 
 may be varied in extent by heat treatment, as is fully explained in the 
 chapter on that subject. It is to be noted here, however, that, with respect 
 to its influence upon the strength and ductility of naturally cooled steel, 
 the average results obtained by four eminent investigators show that for 
 each 0.1% carbon added to steel up to .90%, these properties are affected 
 approximately as follows: 
 
 Yield point is raised 3987 pounds per sq. in. 
 
 Maximum stress is raised 9363 " " " 
 
 Elongation is reduced 4.33% 
 
 Reduction of area is reduced 7.27% 
 
 Above 1.00% in carbon content, the brittleness of steel increases so 
 rapidly, due to the presence of excess cementite, that its use is then limited 
 to articles, relatively few in number, requiring great hardness and little 
 toughness or ductility. Hence, the carbon content of commercial steel will 
 seldom exceed 1.10%. 
 
 Influence of Manganese: The chemical properties of manganese, 
 which impart to it the power of combining with the oxygen of ferrous-oxide 
 and of setting free the iron, make it invaluable as a cleansing, or deoxidizing 
 agent, and have been referred to, time and again, in describing the various 
 processes of making steel. It is here appropriate to consider the effect of 
 the manganese that remains in the steel after deoxidizing. Of this residual 
 manganese, it may be said that every one is agreed that its effects, when 
 present up to certain limits, varying with conditions and the use to which 
 the steel is to be put, are wholly beneficial. Aside from causing the steel 
 to roll and forge better, it is a well known fact that manganese adds some- 
 what to the tensile strength, this beneficial effect depending upon the 
 carbon content as well as that of the manganese. According to H. H. 
 
MANGANESE 571 
 
 Campbell 1 the tensile strength of untreated open hearth steel, containing 
 .30% manganese and over, rises for each increase of .01% in manganese 
 and with the carbon content as shown in the following table: 
 
 Table 63. The Effect of Manganese Upon the Tensile Strength of Steel. 
 
 Each increase of .01% Mn. above .30% or .40% raises the tensile strength 
 in: 
 
 CARBON CONTENT BASIC OPEN HEARTH ACID OPEN HEARTH, 
 
 (MN. ABOVE .30%) (MN. ABOVE .40%) 
 
 .05 110 Ibs. 
 
 .10 130 " 80 Ibs. 
 
 .15 150 " 120 
 
 .20 170 " 160 " 
 
 .25 190 " 200 " 
 
 .30 210 " 240 " 
 
 .35 230 " 280 " 
 
 .40 250 " 320 " 
 
 When the manganese content is less than .30%, this law of increase is 
 disturbed by other influences of an unknown character, which may even 
 cause a complete reversal of tendencies and the tensile strength to rise 
 when the content falls below .30%. Above 1.00%, manganese begins to 
 produce undue hardness and brittleness which becomes very marked as 
 the content reaches and passes 1.50%. Like the tenacity, these properties 
 are similarly affected by the relation of the manganese to the carbon content. 
 
 Influence of Manganese in Heat Treatment: Relative to heat 
 treatment, the effect of manganese upon the heat treating qualities of the 
 steel are not to be overlooked. In the case of ordinary open hearth steels, 
 this brittleness of the high manganese steels is associated with the tendency 
 of such steels to crack just before or during the quenching. Much care, 
 therefore, must be exercised in selecting steel for heat treatment to secure 
 the proper proportion of carbon and manganese, for which purpose the 
 following statements will be found to apply in a general way: 
 
 1. Steels containing 1.50% manganese cannot be quenched in water, 
 whatever their carbon content may be, but with the per cent, of carbon 
 no higher than .60 they may, depending on the design of the body and the 
 condition of the steel, be quenched in oil. 
 
 2. Steels containing 1.00% manganese and of low or medium carbon 
 content may be quenched in water, though the risk of cracking is still great. 
 
 3. A manganese content of .40% or less is required in high carbon 
 steels near the eutectoid (.90% C.) composition, when such steels are to 
 be hardened by quenching in water. 
 
 4. In hyper-eutectoid steels, such as high carbon tool steels, the 
 manganese content should not rise above .25%. 
 
 iSee Manufacture and Properties of Iron and Steel. Published by McGraw 
 Hill Book Co. 
 
572 INFLUENCE OF ELEMENTS 
 
 5. Each 0.1% of manganese lowers the critical range on heating by 
 about 3 C. 
 
 6. According to one authority, electric steel permits a higher content 
 of both the carbon and the manganese in heat treating than would be per- 
 missible with ordinary open hearth steel. 
 
 Influence of Manganese on Sulphur: Another great benefit to be 
 gained from the use of manganese is due to its ability to neutralize, or 
 offset, the evil effects of sulphur. Like oxygen, this element combines 
 with both iron and manganese to form sulphides, but in the presence of 
 both elements and at a high temperature it unites with the latter in pref- 
 erence to the former, thus producing manganese sulphide, MnS, which is 
 practically harmless in steel for reasons that will be explained shortly. 
 
 Influence of Sulphur: The effect of this element upon the tenacity 
 and ductility of steel, at least up to 0.1%, is so slight that it may be dis- 
 regarded. One investigator asserts that it accelerates corrosion of the 
 steel that contains it. Its most marked effects, however, are encountered 
 in hot working, i. e., rolling or forging, the steel, and they were formally 
 believed to be always evil ones. That in the form of ferrous sulphide, 
 FeS, it is capable of doing great harm in steel by causing redshortness is 
 conceded by all, but when neutralized with manganese in sufficient amount 
 it may be comparatively harmless, even when present to the extent of a 
 much higher content than one-tenth per cent. 
 
 Why Manganese Neutralizes the Effect of Sulphur: The only plaus- 
 ible explanation so far offered to account for the difference in the effect of 
 the two sulphides is that the iron sulphide forms films, or cell walls, about 
 the grains of the metal, and as this sulphide fuses at a red heat, these cell 
 walls, by becoming fluid, interrupt the continuity of the mass and so render 
 the steel hot short. Manganese sulphide, instead of forming envelopes 
 about the grains of the metal, collects, or segregates, into globules at tem- 
 peratures near that of the metal on solidifying, upon which the main body 
 of metal then contracts. Manganese sulphide has a much higher fusion 
 point than ferrous sulphide, hence does not melt at a rolling heat, but 
 becomes merely plastic like the rest of the metal. In this form, it is rolled 
 into fibers, which give to the steel, when present in sufficiently large quan- 
 tities, a fibrous structure similar to that of wrought iron. In order to 
 get the full benefit of the manganese, it is necessary that it should be present 
 in the steel to the extent of about three times the theoretical amount 
 required for the formation of the sulphide. Roughly, this means that the 
 per cent, of manganese should be five times that of the sulphur. 
 
 Uses for Sulphur in Steel: This fibrous structure of high sulphur 
 steel is made use of in the manufacture of free cutting steel, like screw 
 stock, for example, because the free cutting properties of this steel are 
 undoubtedly due to its fibrous structure. Thus, in this case, at least, 
 
SULPHUR AND PHOSPHORUS 573 
 
 sulphur is to be regarded as a friend rather than as a foe. In this con- 
 nection, it should be observed that experiments conducted during 1914 and 
 1915 in both this country and England tend to show that sulphur, when 
 accompanied with a sufficient amount of manganese, is not such an enemy 
 as it is sometimes supposed to be. Extensive investigations by our own 
 research department have shown that there is practically no difference in 
 the rolling, forging or welding qualities, nor in the physical properties, of 
 steels containing from .030% to .120% sulphur. It is interesting to consider 
 how the unfavorable attitude toward sulphur came about. Up to within 
 the present. decade, most of the steel produced in this country was made 
 by the acid Bessemer process in which the sulphur content would often 
 range from .070% to .100%. Yet there was no complaint about this steel, 
 and that it gave excellent service for nearly all purposes that steel is used 
 cannot be denied. But with the advent of basic steel, the notion became 
 prevalent, through academic discussions to explain why this steel should 
 be better than Bessemer, that even a small quantity of sulphur was harmful 
 to the steel; and consumers, also, naturally insisted on placing the limit for 
 sulphur at the lowest possible figure, under .040 per cent, or even under .030 
 per cent., in order to secure the better steel. Evidently, however, such an 
 attitude should be corrected now, for economic reasons, if no other. In 
 view of the fact that it is becoming increasingly difficult to keep the sulphur 
 content below .040%, it seems ridiculous to insist upon so low a limit, 
 when the evidence points so strongly to .100% as a limit that may be made 
 to serve as well, for many purposes, at least. As most basic steel made in 
 this country appears to tend naturally toward a sulphur content of from 
 .050% to .060%, even raising the limit to .080% would result in a great 
 saving. 
 
 Influence of Phosphorus: Phosphorus is another element that has 
 been painted a little blacker, perhaps, than it should. It has been 
 everywhere charged with producing cold shortness, or brittleness when 
 cold, but experiments and tests conducted by our research department, 
 during the first half of the year 1917, seem to indicate that up to .10% at 
 least, phosphorus does not produce brittleness in the metal to a degree that 
 is noticeably harmful. In these experiments, steel with phosphorus con- 
 tents ranging from .018% to .110% were subjected to severe cold bending, 
 stamping and pressing tests that steel is called upon to withstand in 
 shaping, with the result that the higher phosphorus steels stood up under 
 the tests as well as the low phosphorus grades, otherwise of identical com- 
 position. That it does increase the hardness and tensile strength of the 
 steel, causing at the same time a proportionate reduction in the ductility, 
 is well established as a fact. In this respect it is very similar to carbon. 
 Some authorities claim that it increases the tensile strength a little more 
 than carbon with a less reduction in the ductility; others say that its effect 
 is practically the same as carbon except that it increases the brittleness 
 a little more. Campbell claims that the tensile strength of basic steel is 
 
574 INFLUENCE OF ELEMENTS 
 
 increased 1000 pounds for each increase of .01% of phosphorus. It, also, 
 benefits the wearing properties of the steel in much the same way that 
 carbon does. In low carbon steels, it is used in many cases with entirely 
 beneficial results. Thus, it is useful in sheet bar, as it is claimed that it 
 prevents the sheets from sticking together in the pack during the rolling. 
 
 The Two Evils of Phosphorus: However, it is not to be inferred 
 that the indiscriminate use of high phosphorus steel is advocated, because 
 it has, according to Howe, Harbord and others, at least two evil tendencies 
 that make it a dangerous element in steels for certain purposes. Speaking 
 of these tendencies, Harbord states that, of all the impurities usually present 
 in steel, practical experience has established the fact that phosphorus is the 
 one that most prejudicially influences the physical properties of the metal 
 by producing brittleness under shock, and hence for practical commercial 
 purposes, phosphorus in steel should not exceed .080%. Again, Howe 
 maintains that while phosphorus sometimes affects iron but slightly, at 
 other times, under apparently similar conditions, it affects it profoundly. 
 In view of this fact, which may be called the treacherousness of phosphoretic 
 steel, it is difficult to define a limit for the maximum content of phosphorus 
 which can be safely allowed in steel, but reasoning that the lower this is, 
 the safer the material, many would insist upon a very low limit. That 
 this limit may be unreasonably low is illustrated in the case of structural 
 steel. Many users of this material refuse to accept any steel that contains 
 a higher percentage of this element than .04%. Yet a class of material 
 subjected to much more severe usage in service, namely, railroad rails made 
 by the Bessemer process, is permitted to contain as much^as .110% phos- 
 phorus. Furthermore, while structural material is subjected to static 
 stresses mainly, a class of stress that phosphoretic steel is most capable 
 of resisting, rails are required to withstand shocks and impacts, which 
 the evidence shows, high phosphorus steels should be least capable of 
 resisting. 
 
 Influence of Silicon: Apparently owing to the fact that all but 
 traces of silicon may be removed in any and all of the processes for manu- 
 facturing steel, the attention of investigators has not been so universally 
 directed to the effects of this element on steel as in the case of the other 
 impurities. Besides, whatever evidence may be collected will be found to 
 vary somewhat. Thus, while certain English investigators found that 
 steels containing as much as 2.00% silicon, a content much higher than 
 any employed in ordinary carbon steel, suffered a marked reduction in 
 ductility, others maintained that the ductility is not markedly affected up 
 to a content of .70%. All agreed that the -tensile strength is increased, 
 and some maintain that small percentages of silicon increased the resistance 
 of the steel to shock. In short, it is generally accepted by all practical 
 steel men that silicon up to .75% is beneficial, that it increases the yield 
 point and tensile strength but does not materially impair the ductility. 
 
SILICON AND OXYGEN 575 
 
 This statement is in accord with the experience of our own Company, who 
 found, for example, that in a certain specification calling for a tensile strength 
 of 80,000 pounds with an elongation of 20% in eight inches, requirements 
 that cannot be approached in plain untreated carbon steel, very satisfactory 
 results were obtained by the addition of .50% silicon. Like manganese, 
 silicon is a wonderful deoxidizer, or cleanser, of steel, and it is possible that 
 the improvement in the quality of steel, and of basic steel in particular, 
 which the addition of small quantities of the element produce, is due rather 
 to this property than to any influence the residual silicon in the steel may 
 have. Spring steel with the silicon ranging from .25% to .35% has greater 
 resiliency than steels of lower silicon content, and without increased brittle- 
 ness. In steel castings it is especially beneficial, as it tends to prevent 
 blow holes and thus promotes soundness. In steels intended to be case 
 hardened, silicon is an objectionable element, as it retards the absorption 
 of carbon. Therefore, in such steels the silicon content should be low, as 
 the retarding effect begins at about .04 per cent. In sheet bar, silicon is 
 like phosphorus in that it tends to prevent the sheets in a pack from 
 sticking together. For this purpose .06 per cent, is sufficient. 
 
 The Influence of Oxygen in steel has been thoroughly discussed and 
 emphasized in connection with the various methods for manufacturing steel. 
 It may, however, again be pointed out that its effects are all evil ones, 
 causing both red shortness and cold shortness in steel, and that when 
 present even in so small amounts as .03% it shows a marked tendency to 
 produce brittleness under shock. The amount of oxygen steel is capable 
 of retaining is small. That retained even by over-blown Bessemer steel 
 without deoxidizing is less than .15%. 
 
 Combined Effect of the Elements on Tensile Strength of Steel: 
 
 These elements, iron, carbon, manganese, sulphur, phosphorus, silicon and 
 oxygen, are the ones found in all commercial grades of steel. Having 
 discussed their effects separately, it may now be advantageous to consider 
 their combined effect upon the metal. Of these elements only oxygen is 
 to be looked upon as being always an enemy. The influence of manganese 
 in steels that are not to be heat treated is always good, as is also that of 
 silicon in small amounts. The tensile strength is raised by carbon, man- 
 ganese, phosphorus, and silicon, while the ductility is decreased by carbon, 
 manganese, and phosphorus. According to H. H. Campbell the influence 
 of these elements varies in the different kinds of steels; and for the two 
 kinds of open hearth steels in their natural state, that is, without any heat 
 treatment, he sums up the combined effect of carbon, phosphorus and 
 manganese on the tensile strength to be approximately as given in the 
 following formulas: 
 
 First Method (of Least Squares) : 
 
 A. Acid Steel 38600+1210 C+890 P+R=Ultimate Strength. 
 
 B. Basic Steel 37430+950 C+85 Mn+1050 P+R=Ultimate Strength. 
 
576 INFLUENCE OF ELEMENTS 
 
 Second Method (by Plotting) : 
 
 C. Acid Steel 40000+1000 C+1000 P+X Mn+R=Ultimate Strength. 
 
 D. Basic Steel 41500+770 C+1000 P+Y Mn+R=Ultimate Strength. 
 
 In these formulas 38600, 37430, 40000 and 41500 represents the initial 
 strength of pure iron; C, P, Mn, stand for carbon, phosphorus and manganese 
 expressed in hundredths of one per cent., respectively; X and Y represent 
 variables changing with the carbon content as given under the heading, 
 Influence of Manganese; and R is a factor that depends on the finishing 
 temperature, and may be either plus or minus. 
 
 For low carbon plain basic steel, such as that used for plates and 
 structural shapes, rolled at the ordinary temperature for hot rolling, the 
 following simple formula is used by many inspectors: 
 
 T (Ultimate Strength)=39000+950 C+1050 P+85 Mn. 
 The symbols in this formula have the same significance as the same 
 symbols in Campbell's formulas. 
 
 The Influence of Copper upon the mechanical properties of steel, 
 when present in small amounts, say up to .50%, is not very pronounced. 
 In terms of tenths of one per cent., the effect of copper as determined by 
 several different investigators is about as follows: The yield point is 
 increased 1800 pounds in steels of low carbon, and 720 pounds in those of 
 medium carbon content; the maximum stress, or ultimate strength, is 
 increased 1200 pounds for low carbon, and 600 pounds for medium; the 
 elongation is decreased .75% for low carbon, and .25% for medium; the 
 reduction of area is decreased .45% for low carbon, and .50% for medium. 
 From these results it is to be decided that the effect of small percentages 
 of copper is slight, and what effect it has is beneficial. This declaration 
 agrees, also, with the verdict of the American Society for Testing Materials. 
 For years copper was looked upon as being very injurious to the steel, it 
 being charged with making the steel red-short and unweldable. However, 
 as early as 1899 A. L. Colby made an extensive series of investigations to 
 determine what really were the effects of small percentages of copper upon 
 the physical properties of the steel. Briefly, these investigations and the 
 results obtained were as follows: A steel shaft 15 inches in diameter by 
 fourteen feet long, corresponding in composition with the propeller shafts 
 adopted by the U. S. Navy Board, but containing also .565% of copper, 
 was forged without difficulty. Test specimens were doubled flat in the 
 cold without showing cracks or flaws, and the tensile strength and ductility 
 were well up to requirements of the Navy. In another series of tests the 
 material, containing .553% copper, was forged into a gun-tube, and satisfied 
 all the requirements for the U. S. Navy for a 6 inch gun. Mild steel in the 
 form of ship-plates, containing .573% copper, passed all the tests required, 
 except a quarter inch plate which was rolled too cold. The bending and 
 quenching tests of the bars cut longitudinally were also satisfactory, but 
 
COPPER, TIN, ARSENIC 577 
 
 some, bent transversely to the direction of rolling, developed cracks. The 
 material could be successfully welded, only one of the specimens tested 
 breaking at the weld, and even then the breaking load was 61,630 pounds 
 per square inch. Flanged cold, the material gave excellent results, and 
 though most severely tested, developed neither defects nor flaws. Other 
 investigations were directed to merchant bars, rails, and nickel steel all 
 containing copper, and in no case was there any evidence of red-shortness, 
 although the copper ranged from .089% to .486%. Colby's conclusions 
 were that a good steel may contain as much as 1% of copper without suffer- 
 ing, provided that the sulphur content is not also high, in which case the 
 metal is likely to crack in rolling. 
 
 Even small amounts of copper in steel causes the latter to resist cor- 
 rosion by acids much better than steels that do not contain it. The 
 research department of the American Sheet and Tin Plate Company has 
 shown that .15% to .25% of copper in steel sheets of heavy gauge practically 
 preserves them from general corrosion, and that the resistance to corrosion 
 begins to be manifested by the steel with the copper content as low as 
 .03%. While copper compounds occur in many iron ores, only traces, if 
 any, are to be found in the Lake Superior Ores. Hence steels made from 
 these ores are practically copper free, except in cases where it is added to 
 produce the non-corroding steels. Occasionally, however, whether intro- 
 duced through accident or from the ore, steels will be found to contain 
 copper to the small amount of .01% to .02%. 
 
 Influence of Tin: While this element is not found in any of the iron 
 ores, the use of detinned scrap may result in its introduction into the steel 
 during the process of manufacture. Hence, the effects of small quantities 
 of tin in steel are not to be overlooked, but unfortunately this matter does 
 not appear to have been very thoroughly investigated. What work has 
 been done shows that tin forms an alloy or a compound with iron, which 
 has the property of making the steel very hard at rolling temperature. 
 Thus, at one steel works it was impossible to roll a heat of steel into which 
 there, had accidently been introduced tin to the extent of .75%. Tin in 
 steel increases the yield point and the ultimate strength of the metal, but 
 to a less degree than carbon or phosphorus. So far as they have gone, 
 investigations appear to indicate that .05% tin in steel would have little 
 influence upon its mechanical or physical properties, but that larger quan- 
 tities must be avoided. 
 
 Influence of Arsenic: This element does not occur in any of the iron 
 ores from the Lake Superior region, and is, therefore, never found in steels 
 made from these ores. When present, however, in small amounts, unless 
 special precautions are taken in making an analysis of the steel, it is reported 
 as phosphorus. Small amounts of arsenic do not affect the physical prop- 
 erties of steel; above .20% its effect is similar to that of phosphorus, causing 
 cold shortness. 
 
578 ALLOY STEELS 
 
 CHAPTER IV. 
 
 ALLOY STEELS. 
 
 SECTION I. 
 
 INTRODUCTORY. 
 
 Definitions: So many different elements may occur naturally in 
 steel, or be added to it, in such varying amounts with corresponding vari- 
 ations in effects, that it is a difficult matter to determine just what con- 
 stitutes an alloy steel even from the standpoint of chemical composition 
 alone, When it is further considered that the different methods of manu- 
 facture also exert their influence, and that certain elements may be added 
 or allowed to remain for widely different reasons, the difficulty of wording 
 concisely an adequate definition becomes more apparent. The definition 
 adopted by the International Association for Testing Materials is as follows: 
 "Alloy steel is steel which owes its distinctive properties chiefly to some 
 element or elements other than carbon, or jointly to such other elements 
 and carbon. Some of the alloy steels necessarily contain an important 
 percentage of carbon, even as much as 1.25%. There is no agreement as 
 to where the line between alloy steel and carbon steel shall be drawn." 
 In this connection it is well to note that elements other than carbon are 
 always to be desired in steel of commercial grade, at least. Such elements 
 may be added or permitted to remain for three distinct reasons, namely, 
 (1) to correct or prevent defects that otherwise would be liable to occur 
 in the final product; (2) to impart to the steel some distinctive property 
 or to improve materially its natural properties; (3) to form alloys for the 
 purpose of experimentation and investigation. The addition of silicon and 
 manganese to steel illustrates the point it is desired to explain. In ordinary 
 practice small amounts of these elements are added to deoxidize the steel, 
 and incidentally the small amounts that remain in the metal may improve 
 its properties. Large amounts of these elements, 1.50% to 3.50% in the 
 case of silicon, and 11% to 14% in the case of manganese, may be added 
 to impart properties to the steel that are distinctive and useful. Other 
 proportions may be used, of course, which result in imparting properties 
 that, while they are distinctive, are not useful, and so these iron alloys 
 have only a scientific value. With these facts in mind, we agree with 
 Henry D. Hibbard of the Bureau of Mines 1 who suggested the following 
 definitions: 
 
 iSee Manufacture and Uses of Alloy Steels. Bureau of Mines Bulletin 100. 
 
ALLOY STEELS 579 
 
 "Simple steel, which is often called carbon steel (or plain steel), 
 consists chiefly of iron, carbon, and manganese. Other elements are always 
 present, but are not essential to the formation of the steel, and the content 
 of carbon or manganese, or both, may be very small." 
 
 "Alloy steel is steel that contains one or more elements other than 
 carbon in sufficient proportion to modify or improve substantially and 
 positively some of its useful properties." These steels, since they contain 
 a special element, are sometimes called special steels. 
 
 "Alloy-treated steel is simple steel to which one or more alloying 
 elements have been added for curative purposes, but in which the excess 
 of the element or elements is not enough to make it an alloy steel." 
 
 All the alloy steels manufactured by The Carnegie Steel Co. are made 
 either in the open hearth or the electric furnace, and the alloying elements 
 chiefly employed are nickel, chromium and vanadium. While a large tonnage 
 of steel, containing slightly higher percentages of copper, silicon or manganese 
 than that prescribed by ordinary practice for carbon steels, is made, these 
 elements in such small quantities are not to be considered as alloys but 
 rather as curative or intensifying elements. A definition that would agree 
 with the customs of this company, then, would appear to be the following: 
 An alloy steel is steel, made by the open-hearth or the electric process, 
 which contains, in addition to carbon, some element or elements added 
 with the object of modifying and substantially improving its mechanical 
 properties in such a way as to make it more suitable for the purpose for 
 which it is intended. This definition does not include the addition of 
 copper for obtaining non-corrosive steel, a chemical property, nor the 
 addition of small amounts of phosphorus to basic steel for sheet bar, nor 
 the sulphur in screw stock, nor manganese and silicon added for curative 
 purposes. 
 
 Carnegie Types and Grades: To illustrate the importance of the 
 elements mentioned above as being the ones chiefly employed in alloy 
 steels, the types and grades of commercial alloy steels manufactured by 
 the open-hearth process and fixed in 1919 as standards by the Carnegie 
 Steel Company may be cited. They are as shown in table 63: 
 
 These types and grades are subject to change, of course, and except 
 from the standpoint of tonnage represent only a small part of the whole 
 field of alloy steels. They are, however, representative of the steels made 
 from the three alloying elements that, up to the present time, have been 
 found to be of greatest value commercially, namely, nickel, chromium, 
 and vanadium. Since it is necessary to limit the scope of this chapter in 
 some way, it seems appropriate to confine it to a discussion of the influence 
 of these elements upon steel and of the general characteristics of the steels 
 represented by the types given in this table. 
 
580 ALLOY STEELS 
 
 Table 64. Carnegie Standard Open Hearth Alloy Steels. 
 Nickel Steel. Low Nickel-Chrome Steel. 
 
 Carbon 10 to .50% Carbon 15 to .45% 
 
 Manganese 50 to .80 Manganese 50 to .80 
 
 Phosphorus not over. . .04 Phosphorus not over. . .04 
 
 Sulphur not over 045 Sulphur not over 045 
 
 Nickel 3.25 to 3.75 Nickel 1.00 to 1.50 
 
 Chromium 45 to .75 
 
 Chrome Steel. Chrome- Vanadium Steel. 
 
 Carbon 15 to .70 Carbon 15 to .55 
 
 Manganese 25 to .50 Manganese 50 to .80 
 
 Phosphorus not over. . .04 Phosphorus not over. . .04 
 
 Sulphur not over 045 Sulphur not over 04 
 
 Chromium 60 to .90 Silicon not over 20 
 
 Chromium 80 to 1.10 
 
 Vanadium not under .. .15 
 
 Chrome-Vanadium Spring Steel. Special Low Chrome Spring Steel. 
 
 Carbon.45to.55,.50to.60, .55to.65% Carbon 80 to .95% 
 
 Manganese 80 to 1.00 Manganese 30 to .50 
 
 Phosphorus not over 04 Phosphorus not over. .04 
 
 Sulphur not over 05 Sulphur not over 05 
 
 Silicon not over 20 Silicon not over 20 
 
 Chromium 1.00 to 1.25 Chromium 20 to .40 
 
 Vanadium not under 15 
 
 SECTION II. 
 
 NICKEL STEEL. 
 
 Manufacture of Simple Nickel Steel: Nickel steel, said to have 
 been used for the first time in 1888, may be made by any of the various 
 processes for the manufacture of steel, but the greater portion is produced 
 by the open-hearth process. At the steel-melting temperature nickel is 
 chemically negative to iron, which is capable of reducing its oxides and 
 preventing its oxidation, even when the bath is a highly oxidizing one. 
 Nickel may, therefore, be added to the bath at any time practically withoui 
 any loss or waste, but its addition is usually made just long enough before 
 tapping to enable it to become properly diffused. For the same reason 
 nickel cannot deoxidize iron, neither will it decompose carbon monoxide 
 nor hold other gases in solution, though it is said to prevent, or hinder in 
 a measure, the segregation of carbon and the other metalloids. It is not 
 used, then, for a curative agent, but only for its beneficial effect upon the 
 physical properties of the steel, for which purpose it is preeminently a 
 strength giving element. 
 
NICKEL STEELS 581 
 
 The Different Nickel Steels and Their General Characteristics: 
 
 The nickel content of the useful nickel steels varies from 2% to 46%, which 
 is a wider range than that covered by any other alloying element. Below 
 2% the benefits derived from its addition alone to steel are very slight 
 and are not worth the extra cost. The great bulk of simple nickel steel, 
 containing from two to four per cent, nickel, is used for structural purposes, 
 such as bridges, gun forgings, machine parts, engines, large dynamos, steel 
 rolls, and various parts of automobiles, because of the superior mechanical 
 properties imparted by the metal when added in these amounts. Thus, 
 for each 1% of nickel added above 2% and up to 4.00% an increase of 
 approximately 6000 pounds in the tensile strength of this steel over the 
 carbon steel is noted, while only a slight decrease, if any, in the ductility 
 occurs, and all this improvement is secured without any heat treatment 
 whatever. The best results, costs and benefits considered, appear to be 
 obtained when the nickel content is between 3 and 4 per cent., the content 
 aimed at for structural purposes being 3.25% to 3.50%. This steel also 
 resists rusting and abrasion better than the plain carbon steels. Nickel 
 steel of this grade lends itself well to heat treatment, and may also be used 
 for case hardening, the only objection to its use for this purpose being the 
 slight tendency of the nickel to retard the rate of penetration of the carbon. 
 When the nickel content is raised above 5%, the metal becomes very hard, 
 is difficult to work either hot or cold, and is rolled only by taking the greatest 
 care. It is in demand where great resistance to shock is a prime quality, 
 such as shield plates for protecting the ammunition of field artillery and 
 the men serving the guns from rifle fire. Up to 8%, nickel increases the 
 hardness of the steel to which it is added, but leaves the metal still amenable 
 to heat treating. Steels containing 10% or more of nickel cannot be 
 hardened by quenching, but become softer after being subjected to this 
 heat treating operation. In 1914 a new alloy steel, containing 13% nickel 
 and .55% carbon, was discovered by Arnold and Read. It is so hard that 
 it cannot be machined or drilled, has a yield point of 134000 pounds, a 
 tensile strength of 195000 pounds, and an elongation of 12% in two inches. 
 Before this discovery, 15% nickel steel, tensile strength about 170000 pounds, 
 was thought to be the strongest one of the series. This steel has been employed 
 occasionally for shafting. Nickel steel containing 22% nickel is used when 
 resistance to rusting is the prime consideration. Thus, it was employed 
 in the valve stems of the salt water fire protective system installed by the 
 city of New York, and in similarly exposed parts of the pumps used in the 
 drainage system for the city of New Orleans. It is also said to be suitable 
 for the spark poles in spark plugs for internal combustion engines. 24% 
 to 32% nickel steel is used for electrical resistance, such as those employed 
 in irons, toasters, and other household heaters. Nickel steels with a nickel 
 content of about 24% are non-magnetic. 36% nickel steel is characterized 
 by an extremely low coefficient of expansion, hence is used in balance wheels 
 of watches, the pendulums of clocks, etc., in order to dispense with com- 
 
582 ALLOY STEELS 
 
 pensation. It is known as invar. Finally, 46% nickel steel, containing 
 only .15% carbon, is known as platinite, because it has about the same 
 coefficient of expansion as platinum and glass. Hence, it is employed in 
 the lead wires of incandescent lamp bulbs, where formally platinum was 
 held to be indispensable. Later, a 38% nickel steel wire, Coated with 
 copper, was found to give better satisfaction than platinite. 
 
 Reasons for These Peculiarities of the Nickel Steels: A study of 
 the explanation offered to account for the peculiar influence of nickel upon 
 steel is both interesting and instructive. Referring again to the 3.5% nickel 
 steel, it is to be noted that nickel primarily influences the strength of the 
 steel, and, to a less degree, the ductility. These facts are explained when 
 the solubility of nickel in iron is considered. Thus, when nickel is added 
 to steel, say of hypo-eutectoid composition, it dissolves in the iron to form 
 an iron-nickel alloy. When this steel is cooled through the critical range, 
 it is this alloy that replaces both the free ferrite and the pearlitic ferrite 
 of the carbon steel. Naturally, a change in the physical properties due to 
 this fact alone are to be expected. But it is in the influence of this alloy 
 upon the formation of pearlite that the reason for the great increase in 
 tensile strength of nickel steel is found. The separation of the cementite 
 from the iron-nickel-carbon solution does not take place as readily as from 
 a plain iron-carbon solution, hence the pearlite areas are larger and less 
 clearly defined than in plain carbon steels. In other words, just as carbon 
 and rapid cooling are obstructing agents to the transformation from 
 austenite to pearlite, so also does the nickel act in a similar manner. As 
 long as the nickel content is very low, not over 2%, this influence shows 
 itself only in a slight change in the physical properties as noted. These 
 changes become more marked with increase of nickel, as is to be expected, 
 but the quick change in heat treating properties at 8% or 10% and at about 
 25% are not thus accounted for. A little reflection, however, shows all 
 these characteristics to be due to the same cause, namely, the retarding 
 action of the nickel upon the transformation ranges. One writer represents 
 this influence of nickel upon the critical points in heating as follows: 
 
 .01% Nickel lowers the Ac 3 range .235 C. 
 
 .01% " " " Ac 3-2 range .180 C. 
 
 .01% " " " Ac 2 range .087 C. 
 
 .01% " " " Ac range .103 C. 
 
 No figures are given for the Ar points, but other authorities have 
 established that the Ar^ point is about 80 C. below the Aci point. In 
 addition to and in connection with these facts, the effect of nickel upon 
 the eutectoid ratio should also be considered. The statement that nickel 
 interferes with the free formation of pearlite has already been made, and 
 it now remains to be pointed out that nickel, up to about 8%, reduces the 
 eutectoid ratio below that for straight carbon steels. According to 
 Bullens the eutectoid for a steel containing 3% nickel is reached when 
 the carbon content is .75% and for one containing 7% nickel this value 
 
NICKEL STEELS 
 
 583 
 
 falls to .60% carbon. All these facts, as they relate to the 3% nickel steel, 
 which is the one we are most interested in, have been assembled and are 
 represented in the accompanying diagram copied after Bullens, who was 
 the first to represent the effects of nickel in this way. 
 
 9.-0 
 
 .9 
 
 Fia. 119. Diagram Showing effect of Nickel Upon the Critical Ranges. 
 Compared with Carbon Steel. (After Bullens, Steel and its Heat Treatment. 
 
584 ALLOY STEELS 
 
 The preceding diagram is intended to depict the general effects of nickel 
 upon the transformation ranges, which become lower and lower as the 
 nickel content is increased, and the eutectoid ratio which decreases with 
 increase of nickel. The diagram shows the position of the Ac and Ar points 
 for carbon steel and also for steel containing 3% nickel. Thus: 
 
 Solid line indicates the position of the ranges on cooling carbon steel. 
 Dotted line indicates the position of the ranges on heating carbon steel. 
 Dot and dash line indicates the position of the ranges on heating 3% nickel 
 
 steel. 
 Dash and dash line indicates approximately the position of the ranges on 
 
 cooling 3% nickel steel. Due to a number of factors, the Arranges 
 
 are subject to considerable variation. 
 
 When the nickel content has been increased to 25%, these ranges are 
 found to lie in a position that is entirely below atmospheric temperatures. 
 
 Structural Changes Due to Nickel : From the preceding data a simple 
 calculation will show that as the nickel, or nickel and carbon contents, are 
 increased, the transformation ranges are progressively lowered until they 
 reach atmospheric temperatures. This fact forms a basis for the classi- 
 fication of the nickel steels, which are divided into the following three 
 divisions: 
 
 1. Pearlitic-Nickel Steels are those in which the nickel and carbon 
 contents are such that, when slowly cooled from a high temperature, they 
 will consist in whole or in pare of pearlite. In these steels the nickel ranges 
 from to 10% and follows inversely the percentage of carbon, which theo- 
 retically ranges from 0. to 1.60%. 
 
 2. Martensitic-Nickel Steels: In these steels the nickel and carbon 
 contents are high enough to lower the critical ranges to such a degree that 
 only a partial transformation from austenite to pearlite occurs even on 
 slow cooling. In these steels the nickel contents range from 10% to 25% 
 with the carbon varying as above. 
 
 3. Austenitic=Nickel Steels: Above 25% the influence of the nickel 
 is so great that the transformation range is lowered to atmospheric tem- 
 peratures, and the steel is always austenitic regardless of the carbon content. 
 As previously pointed out, only the pearlitic steels containing about 3.50% 
 nickel are of real importance commercially. 
 
 The Constitutional Theory of Ternary Steels: In causing these 
 structural changes the action of nickel is in accord with that of all the 
 alloying elements. Briefly stated, the theory is that, upon the introduction 
 of a third element into a given carbon steel, the steel remains at first 
 pearlitic in structure, but as the content of the special element is increased 
 the steel becomes martensitic, then austenitic or cementitic, depending 
 upon the chemical action and alloying powers of the special element with 
 
NICKEL STEEL 
 
 585 
 
 respect to carbon and iron; and also that by keeping the amount of the 
 special element constant, the same transformations may be effected by 
 raising the carbon content. This statement, in so far as it relates to nickel, 
 steels is expressed diagrammatically in the accompanying figure. 
 
 FIG. 120. 
 
 Per Cent. Carbon 
 Constitutional Diagram for the Nickel Steels. 
 
 This diagram -shows that with a very low carbon content, say about 
 .05%, the steel remains in the pearlitic condition until the nickel content 
 reaches 10%, when it will be found to be in the martensitic condition. With 
 a nickel content of 30% the same steel would be austenitic. But with a 
 carbon content of about .80%, for example, the steel becomes martensitic 
 when the nickel content exceeds 6%, and austenitic when it reaches 16% 
 or 17%. Diagrams like the preceding are useful in illustrating the effect 
 of the different alloying elements and will frequently be made use of in 
 the discussions to follow. 
 
 Heat Treating Pearlitic Nickel Steels: From what has been said, it 
 should be apparent that the heat treating of nickel steel, to secure the 
 desired results, is an art that requires much experience and knowledge. 
 Hence, it is only desirable to indicate what should be the proper treatment 
 for this steel, and although the heating and cooling of this steel presents 
 some phenomena quite distinctive from carbon steels, it is considered that 
 this object has already been attained. However, a few remarks as to how 
 the low nickel steels are benefited by heat treatment may not be out of 
 
586 
 
 ALLOY STEELS 
 
 place. A heat treated nickel steel has a lower reduction and elongation 
 than a correspondingly heat treated steel without nickel, but the increase 
 in strength is much greater. Thus, for the same strength, the nickel steel 
 is much tougher, and on this account nickel is much to be preferred to carbon 
 for increasing tensile strength. The tensile strength and elastic limit are 
 both affected by the temperature of the drawback, being decreased as this 
 temperature is raised, but the reduction in area and elongation are not 
 so correspondingly and gradually increased as in the plain steels. In this 
 connection, a study of tables 62 and 65 will be found of value. 
 
 Table 65: Illustrating the Effect of Various Heat Treatments upon 
 the Mechanical Properties of Three Per Cent Nickel Steels. 
 
 Chemical Composition: C. .37%, Mn. .61%, Si. .22%, P. .022%, 
 S. .034%, Ni. 3.27%. 
 
 Description of Pieces Treated: One inch rounds, 25 inches long; 
 14 pieces, all from same billet. 
 
 Description of Test Pieces: One test piece from each of the 14 
 pieces, turned to a diameter of ^ inch, as in Fig. 47. 
 
 HEAT TREATMENT 
 
 PHYSICAL TEST 
 
 Anneal and 
 Hardening and Draw 
 Refining, Deg. C Deg . 
 
 Tensile 
 Strength 
 
 Elastic 
 Limit 
 
 Elongation 
 
 in 2 ' % 
 
 Reduction 
 m area % 
 
 Brinell 
 Number 
 
 Scleros- 
 cope Test 
 
 As Rolled 
 
 
 113,000 
 
 69,000 
 
 22.5 
 
 47.7 
 
 229 
 
 32 
 
 Heated to 76 
 
 and 
 
 
 
 
 
 
 
 cooled in F 
 
 urnace 
 
 99,000 
 
 60,000 
 
 26.1 
 
 48.3 
 
 183 
 
 26 
 
 Heated to 
 
 427 
 
 168,000 
 
 150,000 
 
 13.5 
 
 47.2 
 
 331 
 
 48 
 
 815 quench- 
 ed in oil 
 
 482 
 538 
 
 148,000 
 133,000 
 
 131,000 
 118,000 
 
 18.0 
 21.0 
 
 48.3 
 57.3 
 
 285 
 262 
 
 44 
 40 
 
 
 593 
 
 110,000 
 
 90,000 
 
 28.0 
 
 64.7 
 
 217 
 
 33 
 
 
 649 
 
 104,000 
 
 70,000 
 
 28.0 
 
 63.0 
 
 207 
 
 27 
 
 
 704 
 
 95,000 
 
 62,500 
 
 28.5 
 
 51.7 
 
 179 
 
 26 
 
 Heated to 
 
 427 
 
 185,000 
 
 165,000 
 
 13.0 
 
 51.1 
 
 352 
 
 50 
 
 815 quench- 
 
 482 
 
 150,000 
 
 136,000 
 
 17.5 
 
 47.2 
 
 311 
 
 44 
 
 ed in water 
 
 538 
 
 148,000 
 
 135,000 
 
 16.0 
 
 54.7 
 
 293 
 
 44 
 
 
 593 
 
 133,500 
 
 115,000 
 
 18.5 
 
 56.5 
 
 269 
 
 37 
 
 
 649 
 
 108,000 
 
 80,000 
 
 27.5 
 
 62.8 
 
 223 
 
 32 
 
 
 704 
 
 98,000 
 
 67,000 
 
 28.0 
 
 58.6 
 
 179 
 
 29 
 
 
CHROME STEEL 587 
 
 SECTION III. 
 
 CHROME STEEL. 
 
 The Manufacture of Simple Chromium Steels is carried on by the 
 open hearth, the electric, or the crucible process. At the temperature 
 of molten steel, chromium is capable of reducing iron oxides, hence it is 
 oxidized to a great extent in the open hearth, especially during the melting 
 and boiling of the charge. All simple chrome steel is made by the addition 
 of ferro-chromium to the charge. When the steel is made in crucibles, the 
 ferro-chromium is added with the original charge; if in the electric furnace, 
 this addition may be made at any time; but if made in the open hearth 
 furnace, the ferro-chromium is added to the steel just long enough before 
 casting for the alloy to be melted and become well mixed through the 
 charge, as otherwise great waste of the chromium results. Chromium, 
 however, is not oxidized readily enough to be of any value as a curative 
 or a deoxidizing agent, and is used only for its effect as an alloying element. 
 Simple chromium steels are worked by the same methods and in the same 
 way as carbon steels, but unlike nickel steels, they are seldom, if ever, used 
 in their natural state, heat treatment being necessary to develop the bene- 
 ficial effects of the chromium, which is most active in responding to this 
 treatment. Simple chromium steel was one of the first alloy steels to be 
 made. 
 
 Influence of Chromium: This element is preeminently a hardening 
 agent in steel. Unlike nickel, which merely dissolves in iron, chromium 
 forms a carbide. In steel, therefore, at least a part of the chromium will 
 be in this form; but it never reacts with the carbon to the exclusion of iron, 
 arid in steel this carbide may exist either as iron-chromium carbide, 
 xFe3C-yCr 3 C2 or as a solution of FesC and Cr 3 C2. Thus, while nickel 
 is found in the ferrite of the steel, chromium is associated with the cementite, 
 and imparts what might be termed a mineral hardness to the steel. But 
 the great hardness of the chrome steels is due, also, to another cause. 
 This iron-chromium cementite is not as readily dissolved or diffused as 
 the ordinary cementite on heating the steel through the critical range, 
 nor does it segregate, or separate to form pearlite, as readily on cooling. 
 This fact accounts,*for the peculiar changes in the critical ranges of the 
 steel that the addition of small amounts of this element brings about, for 
 while it tends to raise the Ac range it also lowers the Ar range. Thus, 
 in quenching it helps to prevent the transformation of the austenite, and 
 so adds much to the hardness in this way. From these statements it is 
 to be inferred that chromium may have no hardening power when not in 
 the presence of carbon. Such a conclusion agrees with the facts, as it has 
 been shown by Harbord that very low carbon chrome steels have prac- 
 tically no hardening power. It is now, also, easily understood why the 
 degree of hardness" of the steel is dependent, within certain limits, upon 
 
588 
 
 ALLOY STEELS 
 
 the carbon content as well as upon the chromium. An exceedingly fine 
 grain structure is characteristic of heat treated chrome steels, which fine- 
 ness of grain confers the valuable property of toughness. Thus, the net 
 result from the influence of this element is to increase the tensile strength 
 and elastic limit, without a noticeable loss in the ductility. One 
 investigator has found chromium to be very efficient in retarding corrosion 
 of the steel by neutral media, such as sea water, and, therefore, recommends 
 its use in ship plates. 
 
 The Microscopic Constituents of the Chrome Steels : The influence 
 of chromium and carbon in determining the constitution of the steel is 
 shown in the accompanying diagram. 
 
 Cementitic Structure 
 
 (Double, Iron, Chromium Carbide 
 
 in Matrix of Martensite) 
 
 1.20 
 
 1.40 
 
 1.00 
 
 Per. Cent. Carbon 
 FIG. 121. Constitutional Diagram for Chromium Steels. 
 
 From this diagram it is seen that, when the chromium content exceeds 
 7% in steels of low carbon content or about 5% in steels of high carbon 
 content, the steel is composed only of martensite, hence is very hard and 
 strong, but is lacking in ductility and is inclined to be brittle. It can be 
 neither hardened nor softened by heat treatment. It will be noted that 
 unlike nickel, increasing the chromium content beyond a certain limit in 
 a steel with a given carbon content fails to produce the austenitic condition, 
 but gives a new structure made up of grains of the double carbide embedded 
 in martensite. Between these two areas is a narrow range in which the 
 carbide grains are somewhat less numerous than in the cementite region 
 proper. This range marks the gradual transition from the martensitic to 
 the cementitic condition with the gradual increase in the chromium and 
 the carbon content. Like the martensitic condition, steel of cementitic 
 composition is not affected by heat treatment. For obvious reasons, then, 
 to produce steels of greatest usefulness, the chromium content will be 
 restricted to that required to give the pearlitic condition only. 
 
CHROME STEELS 589 
 
 Uses of the Simple Chrome Steels: These steels are used 
 wherever extreme hardness is desired. Thus, they have long been used 
 for stamp shoes and dies for crushing hard ores, like some of the gold and 
 silver ores. Another use is for five-ply plates for safes, where their great 
 hardness is valued on account of the resistance they offer to the drilling 
 tools used by burglars. Rolls for cold rolling metals are made of steel 
 containing about .9% carbon and 2% chromium, while several thousand 
 tons of steel containing about 1.30% carbon and .5% chromium are used 
 annually for files. It is often used in steel for various special purposes, 
 as for example the steel known by the name of "Crucia," which is nothing 
 more than a good grade of spring steel to which has been added from .20% 
 to .40% chromium. The Carnegie Steel Company manufactures this steel 
 as a part of their regular product. A type for axes and hammers, which 
 contains .60% to .70% carbon and .60% to .90% chromium; another for 
 chains, containing .25% to .33% carbon and .65% to .95% chromium; and 
 a third for track bolts with .25% to .40% carbon and .60% to .90% chromium 
 are also manufactured by this company. But the most important use for 
 these steels is in the balls and rolls for bearings. For this purpose they are 
 employed in low carbons for case hardening and in high carbons for heat 
 treating, i. e., quenching and tempering. Of the tonnage furnished by 
 Carnegie Steel Company for this purpose, that for case hardening is made 
 in the open hearth, while the high carbon material is produced in the electric 
 furnace. 
 
 Heat Treatment of Simple Chrome Steel: To cite an example of the 
 high grade of chrome steel: One large maker of bearings uses steel con- 
 taining carbon, 1.10%; chromium, 1.40%; manganese, 0.35%; sulphur, 
 0.025%; and phosphorus, 0.025%. Sizes smaller than one-half inch diam- 
 eter are heat-treated by being quenched in water from 774 C. and then 
 drawn to 190 C. for half an hour. For larger balls, the quenching tem- 
 perature is 802 C. The second heating does not produce even an oxide 
 color, but is enough to relieve in some degree the internal stresses due 
 to the irregular cooling of quenching, so that the balls are less liable to 
 crack spontaneously or to be broken in use. The strength of a good, well- 
 treated ball is prodigious; a ball three-fourths of an inch in diameter, tested 
 by the three-ball method, sustained a load of 52,000 pounds. On the small 
 area of contact the intensity of the pressure amounts to over one million 
 pounds per square inch. The Society of Automobile Engineers recommends 
 less chromium than that given above, or 1 to 1.2 per cent. The critical 
 ranges for these steels containing .90% carbon, vary about as follows: 
 
 For a chromium content of .50%, from 720 C. to 745 C. 
 For a chromium content of 1.50%, from 760 C. to 785 C. 
 
 As indicating what may be expected by varying the treatment of the 
 steels of this grade, the following will serve as an illustration: 
 
590 
 
 ALLOY STEELS 
 
 Table 66. Physical Properties of a Heat Treated Chrome Steel. 
 
 Material: 1 inch rounds. 
 
 Analysis: C., .64%; Mn., .27%; Si., .18%; Cr., 1.01%. 
 
 Treatment: Heated to 870-871 C. Quenched in oil, and tempered as 
 indicated. 
 
 TEMPERING 
 TEMPER- 
 ATURES 
 
 TENSILE 
 
 STRENGTH 
 
 ELASTIC 
 LIMIT 
 
 ELONGA- 
 TION IN 
 2 INCHES 
 
 REDUCTION 
 IN AREA 
 
 BRINNEL 
 HARD- 
 NESS 
 
 400 C. 
 500 C. 
 600 C. 
 
 228,000 
 212,500 
 186,500 
 
 170,500 
 155,500 
 128,000 
 
 5.2% 
 8.4% 
 10.3% 
 
 13.7% 
 19.8% 
 
 22.2% 
 
 478 
 445 
 389 
 
 SECTION IV. 
 
 CHROME NICKEL STEELS. 
 
 Influence of Chromium and Nickel When Combined: Having 
 considered the effects of chromium and nickel when added separately to 
 the steel, the student is interested in knowing what their combined influence 
 may be. In the woids of Bullens, an impartial judge of high standing as 
 an authority upon the subject of alloy steels, especially from a practical 
 standpoint: "The chrome-nickel steels probably represent the best all- 
 round alloy steels in commercial use for general purposes. Chrome-nickel 
 steels of suitable composition appear to have combined in them the beneficial 
 effects of both the chrome and nickel, but without the disadvantages which 
 are inherent in the use of either one separately. Moreover, the presence of 
 both chrome and nickel seems to intensify certain physical characteristics. 
 To the increased ductility and toughness conferred by nickel on the ferrite 
 there is added the mineral hardness given to the cementite and pearlite 
 by the chrome, but with a greater resultant effect. Again, while the 
 addition of nickel alone serves to diminish the susceptibility to brittleness 
 in the steel upon prolonged heating or sudden cooling in comparison with 
 the corresponding straight carbon steels and, on the other hand, the 
 presence of chrome alone tends to the opposite effect, a suitable combination 
 of the two alloying elements tends to neutralize the harmful effects and 
 also to magnify the good points. This is not only brought out in the static 
 strength and ductility, but also in the dynamic strength or fatigue 
 resistance." 
 
 Types of Chrome=Nickel Steel : According to the testimony of some 
 heat treating experts there appears to be a certain ratio of chrome to nickel 
 which gives the most efficient combination of the physical properties. 
 Thus, if the nickel and chromium are present in the right proportions, the 
 lesser susceptibility of the nickel to brittleness, for example, will so modify 
 the greater tendency to brittleness which is given by chrome alone, that 
 a better steel is obtained than when this ratio is not observed. This ratio 
 is said to be about 2^ parts of nickel to one part of chromium. Further- 
 more, it is claimed that if the chromium content greatly exceeds this relation 
 
CHROME-NICKEL STEEL 
 
 591 
 
 to nickel, the temperature limits are so narrowed that the successful treat- 
 ment of the steel is made very difficult. .Thus, in the three standard types 
 of these steels, known as low nickel-chrome steel containing about 1.5% 
 nickel, medium nickel-chrome steel with about 2.50% nickel, and high 
 nickel-chrome steel, in which the nickel content rises to that of the simple 
 nickel steels with 3.5% nickel, the chromium content* should be approxi- 
 mately .60%, 1.00% and 1.50%, respectively. 
 
 Mayari Steel, termed a natural chrome-nickel steel, is made from 
 certain ores found at Mayari, Cuba. These ores contain enough nickel 
 and chromium to give a pig iron with 1.3% to 1.5% nickel and 2.5% to 3.0% 
 chromium. When this iron is converted into steel, for which purpose the 
 open hearth or the duplex, Bessemer-open hearth, processes are employed, 
 practically all the nickel remains in the steel, but a large part of the chrom- 
 ium is wasted. The steel is thus a species of low-nickel-chrome, containing 
 roughly from 1.00% to 1.50% nickel and from .20% to .70% chromium. 
 This steel is undoubtedly of excellent quality for certain purposes, but 
 where the highest quality of chrome-nickel steel is required, most authori- 
 ties agree that the synthetic alloy steels are superior. 
 
 Uses of Chrome=Nickel Steels: Low chrome nickel steel is the type 
 more commonly employed because of its low price and the greater ease by 
 which it is machined. In static properties it is nearly equal to the higher 
 nickel grades, but in resistance to dynamic stresses, shocks, etc., the latter 
 are superior. Besides, since the purpose to which it is adapted is 
 about the same as 3.5% simple nickel steel, low-nickel chrome is being 
 substituted for this steel, also on account of the price. All three grades 
 are used in automobiles, the carbon content being varied about as shown 
 in the following table: 
 
 TABLE 67: Grades and Composition of Nickel-Chrome Steels. 
 
 GRADE 
 
 CARBON 
 
 % 
 
 MN. 
 MAX. % 
 
 Si. 
 % 
 
 S. 
 MAX. % 
 
 P. 
 MAX. % 
 
 Ni. 
 
 % 
 
 CR. 
 
 % 
 
 Low 
 
 0.20 to 0.55 
 
 0.70 
 
 Low 
 
 0.050 
 
 0.04 
 
 1.25 
 
 0.60 
 
 Medium. . 
 
 .20 to .55 
 
 .70 
 
 Low 
 
 .050 
 
 .04 
 
 1.75 
 
 1.10 
 
 High.... 
 
 .20 to .55 
 
 .70 
 
 Low 
 
 .050 
 
 .04 
 
 3.50 
 
 1.50 
 
 An important use of these steels is in armor plate. Thick armor is 
 face hardened by a carbonizing process, but the body has the original 
 composition, which is approximately as follows: C., .33%; Mn., .32%; 
 S., .03%; P., .014%; Si., .06%; Ni., 4.00%; and Cr., 2.00%. Medium armor, 
 three to five inches in thickness, is not face hardened, but is given high 
 properties throughout by the proper heat treatment. The composition of 
 
592 ALLOY STEELS 
 
 this armor is approximately, C., .30%; Mn., .34%; S., .03%; P., .03%; 
 S ., .13%: Ni., 3.66%, and Cr., 1.45%. The nickel chromium steels are 
 used in the manufacture of most armor piercing projectiles, also. 
 
 Heat Treatment of Chrome=Nickle Steels: The heat treatment of 
 these steels is about the same in kind and method as that for simple nickel 
 and chrome steels, and is varied to suit the kind of material and the purpose 
 for which the steel is to be used. To give a general idea of the proper 
 treatment for these steels, the recommendations of Bullens may be cited: 
 
 I. "For forgings: 
 
 a. Quench in oil from about 175 to 200 F. (97 to 110C.) over 
 the critical range. 
 
 b. Quench in oil from about 50F (28C) over the critical range. 
 
 c. Anneal at about 75F (42C) under the critical range and 
 machine. 
 
 d. Quench in the proper medium from about 50 F (28 C) over the 
 range. 
 
 e. Draw the temper to suit the work in hand." 
 
 II. "For shafts and other structural parts in which the desired physical 
 
 properties may be obtained by a drawing temperature of about 
 900 F (500 C.) or over, and which will leave the steel in a 
 machinable condition, treatment I. may be modified at (c) as thus 
 noted, and no further treatment will be required. 
 
 a. Quench in oil from about 175 to 200 F (97 to 110C) over 
 the critical range. 
 
 b. Quench in oil from about 50F (28C) over the critical range. 
 
 c. Draw at 900F (482C) or more, as the work may require. 
 Machine." 
 
 III. "The full treatment as given under (I) may be modified, if desired 
 
 to the following, for parts to be drawn below 900 or 1000 F. 
 (482 or 540 C.) 
 
 a. Quench in oil from about 175 to 200 (97 to 110C) over the 
 critical range. 
 
 b. Reheat to about 25 to 50 F. (14 to 28C) over the critical 
 
 range and cool slowly. Machine. 
 
 c. Quench in oil from about 50 (28C) over the critical range. 
 
 d. Draw to the temperature required by the work." 
 
ALLOY STEELS 
 
 593 
 
 The data supplied in tables 68 and 69 will serve as a basis for comparing 
 the mechanical properties of the chrome-nickel steels, both in their natural 
 and heat-treated states 
 
 Table 68 : Illustrating the Effect of Various Heat Treatments upon the 
 Mechanical Properties of Low=Chrome=Nickel Steels. 
 
 Chemical Composition: C. .39%, Mn. .52%, Si. .18%, P. .017%, 
 S. .042%, Ni. 1.18%, Cr. .58%. 
 
 Description of Pieces Treated; One inch rounds, 25 inches long; 
 14 pieces, all from same billet. 
 
 Description of Test Pieces; One test piece from each of the 14 
 pieces, turned to a diameter of % inch, as in Fig. 47. 
 
 HEAT TREATMENT 
 
 PHYSICAL TESTS 
 
 Hardening and 
 Refining Deg. C 
 
 Anneal and 
 Draw, 
 Deg. C 
 
 Tensile 
 Strength 
 
 Elastic 
 Limit 
 
 Elongation 
 in 2"% 
 
 Reduction 
 in area% 
 
 Brinell 
 Number 
 
 Scleros- 
 copeTest 
 
 As Rolled 
 
 
 99,000 
 
 59,000 
 
 25.5 
 
 54.4 
 
 208 
 
 29 
 
 Heated to 
 
 760 and 
 
 89,000 
 
 54,000 
 
 30.0 
 
 56.5 
 
 170 
 
 25 
 
 cooled in F 
 
 urnace 
 
 
 
 
 
 
 
 Heated to 
 
 427 
 
 139,000 
 
 114,000 
 
 15.5 
 
 51.4 
 
 331 
 
 42 
 
 845 quench- 
 ed in oil 
 
 482 
 538 
 
 130,000 
 113,000 
 
 113,000 
 88,000 
 
 15.0 
 17.5 
 
 53.9 
 59.8 
 
 321 
 255 
 
 44 
 34 
 
 
 593 
 
 113,000 
 
 86,000 
 
 21.0 
 
 62.8 
 
 229 
 
 32 
 
 
 649 
 
 108,000 
 
 81,000 
 
 25.0 
 
 64.7 
 
 229 
 
 31 
 
 
 704 
 
 97,000 
 
 70,000 
 
 27.0 
 
 67.7 
 
 207 
 
 29 
 
 Heated to 
 
 427 
 
 174,000 
 
 158,000 
 
 14.0 
 
 47.2 
 
 363 
 
 46 
 
 845 quench- 
 
 482 
 
 150,000 
 
 135,000 
 
 16.5 
 
 45.7 
 
 221 
 
 44 
 
 ed in water 
 
 538 
 
 140,000 
 
 125,000 
 
 20.0 
 
 55.2 
 
 285 
 
 44 
 
 
 593 
 
 124,000 
 
 108,000 
 
 22.0 
 
 60.8 
 
 255 
 
 40 
 
 
 649 
 
 110,000 
 
 91,000 
 
 26.0 
 
 66.0 
 
 229 
 
 33 
 
 
 704 
 
 90,000 
 
 70,000 
 
 34.0 
 
 69.2 
 
 187 
 
 25 
 
 
 
 
 
 
 
 
 
594 
 
 ALLOY STEELS 
 
 Table 69 : Illustrating the Effect of Various Heat Treatments upon the 
 Mechanical Properties of High=Chrome=Nickel Steel. 
 
 Chemical Composition; C. .34%, Mn. .64%, Si. .20%, P. .010%, 
 S. .036%, Ni. 3.65%, Cr. 1.24%. 
 
 Description of Pieces Treated; One inch rounds, 25 inches long; 
 14 pieces, all from same billet. 
 
 Description of Test Pieces; One test piece from each of the* 14 
 pieces, turned to a diameter of % inch, as in Fig. 47. 
 
 HEAT TREATMENT 
 
 PHYSICAL TESTS 
 
 Hardening and 
 Refining Deg. C 
 
 Anneal and 
 Draw, 
 Deg. C. 
 
 Tensile 
 Strength 
 
 Elastic 
 Limit 
 
 Elongation 
 in 2"% 
 
 Reduction 
 in area % 
 
 Brinell 
 Number 
 
 Scleros- 
 cope Test 
 
 As Rolled 
 
 
 172,500 
 
 138,000 
 
 11.5 
 
 26.5 
 
 444 
 
 57 
 
 Heated to 76 
 
 and 
 
 125,000 
 
 78,000 
 
 18.0 
 
 47.2 
 
 255 
 
 38 
 
 cooled in F 
 
 urnace 
 
 
 
 
 
 
 
 Heated to 
 
 427 
 
 190,000 
 
 166,000 
 
 12.5 
 
 50.0 
 
 415 
 
 55 
 
 845 quench- 
 ed in oil 
 
 482 
 538 
 
 170,000 
 142,000 
 
 154,000 
 126,000 
 
 15.0 
 20.0 
 
 44.3 
 
 57.8 
 
 341 
 321 
 
 41 
 43 
 
 
 593 
 
 127,500 
 
 115,000 
 
 22.0 
 
 62.3 
 
 285 
 
 36 
 
 
 649 
 
 117,000 
 
 99,000 
 
 23.0 
 
 64.7 
 
 255 
 
 37 
 
 
 704 
 
 115.000 
 
 69,000 
 
 24.0 
 
 56.0 
 
 255 
 
 35 
 
 Heated to 
 
 427 
 
 164,000 
 
 152,000 
 
 17.5 
 
 56.0 
 
 375 
 
 51 
 
 845 quench- 
 
 482 
 
 160,000 
 
 147,500 
 
 15.5 
 
 53.9 
 
 341 
 
 48 
 
 ed in water 
 
 538 
 
 145,000 
 
 130,000 
 
 21.0 
 
 58.6 
 
 321 
 
 43 
 
 
 593 
 
 124,000 
 
 100,000 
 
 22.5 
 
 62.8 
 
 269 
 
 40 
 
 
 649 
 
 120,000 
 
 99,000 
 
 24.0 
 
 63.7 
 
 248 
 
 36 
 
 
 704 
 
 120,000 
 
 69,000 
 
 17.0 
 
 50.0 
 
 269 
 
 36 
 
VANADIUM STEEL 
 
 595 
 
 SECTION V. 
 
 VANADIUM STEELS. 
 
 Simple Vanadium Steels do not at present have the standing they 
 formerly had, and the only reason for mentioning them here is to enable 
 the reader to get an idea of the effects of vanadium alone, so that he can 
 better understand the reasons for chrome vanadium steel, to be considered 
 later. Since 1917 the Carnegie Steel Company has manufactured but 
 one grade of vanadium steel, known as type "F," which is used as a flux 
 in oxyacetylene, oxyhydrogen, and electric welding. 
 
 Influence of Vanadium: Unlike nickel and chromium, vanadium is 
 an intense deoxidizing agent, being capable of carrying the cleansing of the 
 steel beyond the point obtainable by manganese, or even silicon and 
 aluminum. Thus, this element performs the double function of a curative 
 and an alloying element. It is added to the steel, in the form of ferro- 
 vanadium, containing about 35% of the element, at the time of tapping, if 
 the steel is being made by the open hearth process, and preferably after 
 the other ladle additions to avoid undue wastage. Of the vanadium that 
 is retained by the steel, only very small amounts, from .15% to .25%, are 
 required to affect the physical properties of the steel. Like most of the 
 alloying elements, it tends to give a finer and denser structure than that 
 of carbon steel, and for the most part its action is similar to other alloying 
 elements. But in some respects, at least, it presents characteristic 
 phenomena. Thus, there are good reasons to believe that, when present 
 even in the small amounts noted above, it is both dissolved in the ferrite, 
 like nickel, and exists as a carbide or a double carbide in the cementite, 
 like chromium. It has a powerful influence upon the transformation ranges, 
 as is seen from the fact that while steels containing .2% carbon and .7% 
 vanadium, or .8% carbon and .5% vanadium, are pearlitic, any increase 
 of the vanadium content over these limits renders the steel cementitic as 
 shown in the following diagram: 
 
 Cement itt 
 
 1.00 1.20 
 
 1.40 
 
 1.60 
 
 Per Cent. Carbon 
 FIG. 122. Constitutional Diagram for Vanadium Steels. 
 
 As revealed by the static physical tests, the benefits from vanadium 
 are somewhat similar to those for the pearlitic nickel steels, that is, it 
 gives a combination of high elastic limit and ductility. With proper heat 
 treatment, it is also said to resist shock, alternate stresses, wear 
 and fatigue. 
 
596 
 
 ALLOY STEELS 
 
 SECTION VI. 
 
 CHROME- VANADIUM STEELS. 
 
 Effect of Combining Chromium and Vanadium: In order that the 
 influence of vanadium as indicated above may be realized to the fullest, 
 the presence of another element as an intensifier is required. Just as 
 chromium intensifies the influence of nickel, so does it also stimulate 
 vanadium, but to a much greater degree, it is said, than with the former. 
 Hence, though various combinations have been tried, such as vanadium- 
 nickel, chrome-vanadium-nickel, etc., the tendency at present points to 
 the general adoption of the one combination, chrome-vanadium. 
 
 Properties and Uses of Chrome- Vanadium Steels : The hot working 
 of these steels presents no difficulties, the steel behaving in the press and 
 rolls much like the higher carbon plain steels. In physical properties, they 
 are similar to chrome-nickel steel except that their contraction of area 
 for a given elastic limit is a little greater. They are also said to be more 
 easily machined than chrome nickel steel and are more free from surface 
 defects, such as scale-pits and seams. While some enthusiasts maintain 
 that it is the steel best adapted to resist shock and fatigue, others hold 
 that the chrome-nickel steel answers all requirements just as well. Perhaps 
 the truth of the matter is that, while both steels are available for most 
 purposes, there are limited fields in which one may excel the other and in 
 which each has its own sphere of usefulness. Most of the chrome-vanadium 
 steel made by the Carnegie Steel Company is used for driving axles and 
 other forgings for locomotives, automobile springs and axles, compressed 
 air flasks, torpedo tubes, and gun forgings. They are nearly always heat 
 treated before being put into service, but in some automobiles the frames, 
 and even part of the forgings and shafts, are made of the steel in its natural 
 state. The composition and properties of three grades of this steel in the 
 untreated condition are given herewith. 
 
 Table 70. The Composition and Mechanical Properties of 
 Untreated Chrome=Vanadium Steels. 
 
 COMPOSITION 
 
 Grade 
 No. 
 
 C. 
 
 % 
 
 Mn. 
 
 % 
 
 Si. 
 
 % 
 
 s. 
 % 
 
 P. 
 
 % 
 
 V. 
 
 % 
 
 Cr. 
 
 % 
 
 Tensile 
 Strength, 
 Pounds 
 
 Elastic 
 Limit, 
 Pounds 
 
 Contra- 
 ction of 
 Area, % 
 
 Elonga- 
 tion in 2 
 Inches% 
 
 1 
 2 
 
 3 
 
 0.57 
 .46 
 .18 
 
 0.84 
 .48 
 .32 
 
 0.27 
 .20 
 .18 
 
 0.04 
 .03 
 .03 
 
 0.01 
 .01 
 .01 
 
 0.31 
 .14 
 .20 
 
 1.36 
 1.17 
 .74 
 
 142,000 
 125,000 
 65,000 
 
 114,000 
 95,000 
 47,200 
 
 42 
 55 
 62 
 
 14 
 20 
 23 
 
 TENSILE PROPERTIES 
 
 Some idea of the general results obtained from heat treating the chrome 
 vanadium steels and of the methods employed may be gained from the 
 following table: 
 
CHROME-VANADIUM STEEL 
 
 597 
 
 Table 71. Physical Properties of Treated Chrome= Vanadium Steels. 
 Tests made on Small Rolled Sections 
 
 CHEMICAL COMPOSITION 
 
 HEAT 
 TREATMENT 
 
 MINIMUM PHYSICAL RESULTS AFTER 
 TREATMENT 
 
 Carbon 
 PerCent. 
 
 Manga- 
 nese, 
 PerCent. 
 
 Chrome- 
 ium, 
 PerCent 
 
 Vana- 
 dium, 
 PerCent 
 
 Water 
 Quenched, 
 Deg. C. 
 
 Draw 
 Deg. C. 
 
 Tensile 
 Strength, 
 Lbs. per 
 Sq.In. 
 
 Elastic Limit, 
 Lbs. per 
 So. In. 
 
 M 
 
 Reduction 
 of Area, 
 Per Cent. 
 
 *.19 
 
 .49 
 
 .96 
 
 .23 
 
 850. 
 
 749 
 
 101,400 
 
 91,200 
 
 27 
 
 69 
 
 
 
 
 
 
 712 
 
 109,500 
 
 98,800 
 
 24 
 
 66 
 
 
 
 
 
 
 675 
 
 121,200 
 
 113,100 
 
 22 
 
 66 
 
 
 
 
 
 
 635 
 
 129,600 
 
 121,200 
 
 21 
 
 60 
 
 
 
 
 
 
 596 
 
 136,000 
 
 125,700 
 
 20 
 
 60 
 
 
 
 
 
 
 574 
 
 146,900 
 
 . 137,200 
 
 20 
 
 58 
 
 *.25 
 
 .30 
 
 .85 
 
 .14 
 
 850. 
 
 748 
 
 88,640 
 
 78,960 
 
 29 
 
 77 
 
 
 
 
 
 
 697 
 
 98,520 
 
 92,200 
 
 26 
 
 72 
 
 
 
 
 
 
 657 
 
 118,700 
 
 114,100 
 
 24 
 
 72 
 
 
 
 
 
 
 607 
 
 121,800 
 
 118,200 
 
 22 
 
 69 
 
 ... 
 
 
 
 
 
 574 
 
 132,400 
 
 125,800 
 
 20 
 
 64 
 
 f.38 
 
 .41 
 
 .97 
 
 .24 
 
 815 
 
 704 ' 
 
 100,000 
 
 88,000 
 
 27 
 
 65 
 
 
 
 
 
 
 649 
 
 118,000 
 
 101,000 
 
 23 
 
 60 
 
 
 
 
 
 
 593 
 
 130,000 
 
 108,000 
 
 18 
 
 52 
 
 
 
 
 
 
 538 
 
 130,000 
 
 108,000 
 
 16 
 
 50 
 
 
 
 
 
 
 482 
 
 157,500 
 
 142,000 
 
 15 
 
 47 
 
 
 
 
 
 .... 
 
 427 
 
 177,500 
 
 155,000 
 
 11 
 
 36 
 
 
 
 
 
 Oil 
 
 
 
 
 
 
 
 
 
 
 Quenched 
 
 
 
 
 
 
 
 
 
 
 815 
 
 700 
 
 95,000 
 
 75,000 
 
 30 
 
 65 
 
 
 
 
 
 
 650 
 
 117,000 
 
 98,000 
 
 20 
 
 57 
 
 
 
 
 
 
 595 
 
 125,000 
 
 105,000 
 
 18 
 
 56 
 
 
 
 
 
 
 540 
 
 133,000 
 
 113,000 
 
 18 
 
 50 
 
 
 
 
 
 .... 
 
 480 
 
 143,000 
 
 116,000 
 
 16 
 
 48 
 
 
 
 
 
 
 
 430 
 
 149,000 
 
 120,000 
 
 15 
 
 49 
 
 
 
 
 
 Annealed 
 
 760 
 
 88,500 
 
 58,000 
 
 30 
 
 59 
 
 * 
 
 
 
 
 As Rolled 
 
 
 127,500 
 
 96,500 
 
 20 
 
 48 
 
 *Silicon .05%. Phosphorus under .035%. Sulphur under .04%. 
 fSilicon .25%. Phosphorus .021%. Sulphur .041%. Tests on 1" rounds. 
 
WORD INDEX 
 
 Abrasion tests 35 
 
 Abramsen straightening machine 468 
 
 Absolute temperature 7 
 
 Absorption 5 
 
 Acetanilide 109 
 
 Acetylene ' 69 
 
 Acid 8-15 
 
 Acid anhydride 15 
 
 Acid Bessemer iron 129 
 
 " ore 40 
 
 Acid Bessemer process 172 to 194 
 
 Acid Bessemer steel 187 
 
 Acid Flux 118 
 
 Acid lining in converter 184 
 
 " in open hearth 200 
 
 Acid refractory 29 
 
 Acid slag 123 
 
 Adamite rolls v 323 
 
 Adhesion 4 
 
 Afterblow 200 
 
 Air chamber in open hearth 216 
 
 Air cooling 211-542 
 
 Airport 213 
 
 Alkalies 39 
 
 Alligator Shears 473 
 
 Allotrimorphic crystals 533 
 
 Allotropy 23-528-532 
 
 Alloy 8 
 
 Alloy steel 579 
 
 Alloy steel ingots 362 
 
 Alloy treated steel 579 
 
 Alphairon 532 
 
 Alternating current 251 
 
 Alternator 246-247-253 
 
 Alumina 26-39 
 
 39-41-160 
 
 118 
 
 Aluminum 1 1-26-166 
 
 Ammonia 25-116 
 
 Ammonia gas 100-116 
 
 Ammonia liquor 99-116 
 
 Ammonium sulphate 116 
 
 Amorphous carbon 23 
 
 Ampere 245-262 
 
 Angles 462 
 
 Angular fracture 306 
 
 Angular method of rolling 446 
 
 Anhydride 15-16 
 
 Aniline 109 
 
 Aniline hydrochloride 109 
 
 " salt.. 109 
 
 in ores 
 
 Annealing steel 
 
 box 
 
 " oven (furnace) . 
 
 temperatures . . . 
 Anthracia oils 
 
 539 
 
 546 
 
 459-546 
 
 540-541 
 
 115 
 
 Anthracite coal 66-78-190-228 
 
 Anthranilic acid 113 
 
 Antifriction metal (see bearing metals) ... 331 
 
 Antimony 12 
 
 Antipyrin 115 
 
 Arc, electric 264 
 
 Arc, furnace 261-265 
 
 Archean era 79 
 
 Armor plate 591 
 
 Arsenic 12 
 
 " in blast furnace 166 
 
 Ash in coal 75-76-80 
 
 " in coke 85-160 
 
 Aspirin 115 
 
 Atomic weights 11-12 
 
 Atoms 10 
 
 Austenite 526 
 
 Austenitic nickel steel 584 
 
 Available base 118 
 
 Avogadro's hypothesis 13 
 
 Axles, manufacture of 508 to 515 
 
 Azo benzene ... 109 
 
 Ball stuff 184 
 
 Banking 159 
 
 Bank ovens 87 
 
 Barium 11 
 
 Barium carbonate in case hardening 565 
 
 Bar mill 474 
 
 Base, chemical 8 
 
 Base of rail 449 
 
 Basic anhydride 15 
 
 " Bessemer process 172 
 
 " flux 118 
 
 " open hearth process 198 
 
 " ore 40 
 
 " pig iron 129 
 
 " process 175 
 
 " refractories 30 
 
 " slag 123 
 
 Battery of coke ovens 86 
 
 Bauxite 31-32 
 
 " brick 32 
 
 Beams 465 
 
 Bearing metal 331 
 
600 
 
 WORD INDEX 
 
 Beehive coke 86 to 90 
 
 " coke oven 86 
 
 " process 86 
 
 Belgian mill 477 
 
 " oven 90 
 
 Bell and hopper 139 
 
 Bellypipe 134 
 
 Bending test 308 
 
 Benzaldehyde 112 
 
 ...69-105 to 110 
 
 Benzene series 
 
 sulphonic acid 110 
 
 Benzenyl trichloride 112 
 
 Benzidine 108-109 
 
 Benzine 69 
 
 Benzol 101 to 110 
 
 " manufacture of 100 to 105 
 
 " gradesof 107 
 
 " usesof 107-109 
 
 Benzoic acid 112 
 
 Bertrand-Thiel process 201 
 
 Bessemer converter, construction of 183 
 
 Henry 176 
 
 " ore 40 
 
 " iron 129 
 
 " steel 190 
 
 steel process 175 
 
 " (steel) converter 183 to 186 
 
 reactions in 193 to 196 
 
 Beta iron 532 
 
 Billet 365 
 
 " mill 391 to 406 
 
 Binary compounds 18 
 
 Binder in coal 80 
 
 Binding material for refractories 29-30-31 
 
 Birmingham ore district 42 
 
 Bituminous coal 66-78-80 
 
 Blanks for wheels 497-505 
 
 Blast for Bessemer converters 180 
 
 " for blast furnace 150 
 
 " for cupola 179 
 
 " furnace, construction of 131 
 
 equipment of 130 
 
 foundation of 131 
 
 " gas 66-71 
 
 Bleeder 141 
 
 Blind pass 426 
 
 Blister 435 
 
 Block of ovens 87 
 
 Bloom 365 
 
 Blooming mill 366 to 384 
 
 Blow 187 
 
 Blower 188 
 
 Blowing engine 150-180 
 
 Blowholes... 346 
 
 Blow in 
 
 Blow off valve 
 
 Blowing in burden . . . 
 
 Blow pipe 
 
 Body of rolls 
 
 Boil in a converter . . . 
 ' "open hearth. 
 
 Bone coal 
 
 Boron 
 
 Bosh... 
 
 152 
 
 143-156 
 
 153 
 
 134 
 
 322 
 
 188 
 
 221 
 
 80 
 
 12 
 
 135 
 
 135 
 
 " brick 137 
 
 " plates 136 
 
 " plate boxes 136 
 
 Bott 134 
 
 Bottom blown converter 183 
 
 " casting 229-324 
 
 " of blast furnace 132 
 
 " of converter 185 
 
 of open hearth 219 
 
 " pouring 228 
 
 " stuff 185 
 
 Box annealing 546 
 
 Box pass 323 
 
 Boyles' law (see volume of a gas) 2 
 
 Braddock insulated rail joints 454 
 
 Braddock works (see Edgar Thomson) .... 177-183 
 
 Brasses 331 
 
 Breaking load 304 
 
 Breakout in open hearth 230 
 
 Breeze, coke 90 
 
 Brick, silica 29 
 
 " clay 30 
 
 " hearth and bosh 137-138 
 
 " in-wall 137-138 
 
 " top 137-138 
 
 Bridge wall 211 
 
 Brightman straightener 468 
 
 Brine for hardening 551-552 
 
 Brinell tests 310-311 
 
 " hardness number 310-311 
 
 Briquettes (or Briquets) 66-220-242 
 
 British thermal unit 7-60 
 
 Bronze 331 
 
 Brookville coal bed 79 
 
 Brown coal 77 
 
 Brown ore 36-37 
 
 B.T.UorB. t. u 7-60 
 
 Buckles in plates 426-435 
 
 " in bars 494 
 
 Bucket hoist 140 
 
 Buggy for Bessemer plant 182 
 
 ' for open hearth 207 
 
 Bulkhead 211 
 
 Bullens 517 
 
 Bull head rail 437 
 
WORD INDEX 
 
 601 
 
 Bundling merchant bars 
 " for export 
 Burdening blast furnace 
 Burden on a blast furnace 
 Burned steel 
 Bustle pipe 
 Butane 
 Butterfly method of rolling 
 By-pass 
 By-product coke 
 gas (Cokeoven Gas) 
 plant 
 process 
 advantages of 
 
 491 
 491 
 159 
 159 
 ... 495 
 ... 135 
 70 
 .. 462-463 
 150 
 85-90 to 98 
 . . . 66-71 
 98 
 90 
 . . 91 
 
 Carbon monoxide in steel 
 " steel (see plain steel) 
 " in pig iron 
 " combined 
 " graphitic 
 " steel 
 " steel for case hardening.. . 
 Carbonizing agent 
 box 
 materials 
 mixtures 
 " pack 
 Carbonless iron 
 Car dumper 
 
 269-518 
 579 
 127 
 127 
 127 
 570 
 563 
 564 
 564 
 564 
 565 
 566 
 518 
 151 
 ...497 to 506 
 504 
 567 
 562-568 
 172 
 152 
 312 
 429 
 475 
 77 
 533 
 
 Carnegie Schoen wheels 
 tape size 
 
 Calcination of dolomite 
 " limestone 
 " magnesite 
 Calcined dolomite 
 magnesite 
 Calcining plant , 
 Calcite (see limestone) 
 Calcium 
 carbide in electric furnace slag 
 Calcium carbonate 
 fhiroide (seeFluorspar) 120-223- 
 oxide 
 silicates 
 " sulphate 
 sulphide 
 
 25 
 25 
 ... 25-231 
 32 
 32 
 202 
 36 
 . ... 11-25 
 . . .122-228 
 . ... 25-166 
 268-270-282 
 25 
 123 
 237 
 . . 270-287 
 503 
 
 Case 
 " hardening 
 Casting iron 
 machine 
 " steel 
 Castors 
 Catcher 
 
 Cellulose 
 
 Cement carbon 
 
 Cementation process 
 Cementite in pig iron 
 " steel 
 Cenozoic era 
 Centering for axles 
 Centigrade 
 Channels 
 
 173-562 
 127 
 518 
 79 
 511 
 6-7 
 463-464 
 66 
 324 
 
 Caliper 
 
 
 
 Charcoal 
 
 
 . . . 60-61 
 60-63 
 
 " iron 
 
 " intensity 
 
 Charging blast furnace 
 boxes 
 machine 
 Chemical calculations 
 
 156 
 207 
 206 
 19 
 
 Calorimeter 
 Camber in rails 
 Campbell furnace 
 H. H 
 
 . . . . 62-63 
 . . . . 450 
 . . . . 200 
 570 
 201 
 
 
 8-9 
 
 compound 
 
 8 
 
 Carbolic acid 
 
 115 
 
 equations 
 formulas 
 nomenclature 
 " radicals 
 " reactions, laws of 
 
 13 
 13 
 18 
 14 
 17 
 
 Carbon 
 
 23 
 
 ' compounds 
 ' dioxide in Bessemer process . . . 
 " in blast furnace. . . . 
 " in O. H. process 
 " in producer gas 
 in blast furnace 
 iron diagram 
 ' monoxide 
 in Bessemer process 
 " blast furnace. . . . 
 " " case hardening... 
 " electric furnace. . 
 " 0. H. process. .. . 
 " " " producer eas 
 
 24 
 194 
 . . . . 163-167 
 238 
 72 
 . ... 163 
 ....524-529 
 22 
 .... 194 
 167 
 564 
 . . . . 286 
 . . . . 238 
 72 
 
 " symbols 
 Chemistry 
 Chill depth of 
 
 9 
 3 
 325 
 
 " test 
 
 155 
 
 Chilled hearth 
 
 158 
 324-326 
 324 
 
 " roll 
 
 Chimneying 
 
 158 
 
 
 143 
 
 
 11 
 
 Chocks. . . 
 
 ...322-331 
 
602 
 
 WORD INDEX 
 
 Chromite 26-31-32-231 
 
 Chrome brick v 31 
 
 " nickel steel 590 
 
 " ore 231 
 
 " steel 587 
 
 " vanadium steel 595 
 
 Chromium 26 
 
 in blast furnace 166 
 
 " steel for case hardening 564 
 
 Cinder 120 
 
 " cooler 134 
 
 " notch ' 134 
 
 1 pitman 218 
 
 Circular mil 257 
 
 Circular shapes 497 to 507 
 
 Clairtoncoke plant , 93 
 
 Clarion coal bed 79 
 
 Clay 29 
 
 " fire 29 
 
 " flint 29-138 
 
 " plastic 29-138 
 
 " brick 30 
 
 Cleaning plant for blast furnace gas 146 
 
 Cleavage 533 
 
 planes. 533 
 
 Clinton iron ore 79 
 
 Closed pass 323 
 
 Coal, kinds of 66 
 
 " gas 66-71 
 
 " tar 66-67 
 
 Cobalt 11 
 
 Cobble 414 
 
 Cogging mill 365 
 
 Cohesion 4 
 
 Coil, of hoop 473 
 
 Coke 66-80-85-160 
 
 " breeze 90-98 
 
 " oven Gas 66-71 
 
 Coking process 88 
 
 Colby 263-576 
 
 Cold bend tests 308 
 
 " blast 150 
 
 " short 233 
 
 ' templet 439 
 
 " blast valve 143 
 
 " working 313-459 
 
 Collar marks 414 
 
 Collars on axles 512 
 
 " rolls 323 
 
 Combination mill 330-477 
 
 Combination plate 427 
 
 Combined carbon 127 
 
 water 15-41 
 
 Combining weights 9-10 
 
 Combustion 60 
 
 Compression tests 33 
 
 Concentric converter 183 
 
 Conduction of electricity 244-256 
 
 " heat 59 
 
 Conductors of electricity 245 
 
 Confining die 460 
 
 Congo red 114 
 
 Coning of wheels 504 
 
 Constitutional diagram 585-588-595 
 
 Continuous coil 473 
 
 " furnace 421 
 
 mill 397 to 405 
 
 Contraction of area (see reduction of area) 307 
 
 of ingots 343 
 
 tests 34 
 
 Convection of heat 59 
 
 Converter 183-186 
 
 " reactions 193-197 
 
 Cooling bed 479 
 
 for annealing 543 
 
 " hardening 551 
 
 " tempering 557 
 
 Cope i. 324 
 
 Copper.' 12 
 
 ' bearing steel 576 
 
 " effects on steel 576 
 
 " in blast furnace 166 
 
 " in bearing metal 331 
 
 Corrosion of steel 577 
 
 Cort, Henry 318 
 
 Cotton seed oil 553 
 
 Cotton tie 472-495 
 
 Coulomb 245 
 
 Coupling box 318-322-330-334 
 
 Covington drawing machine 90 
 
 Cracks in billets and blooms 417 
 
 " " ingots 349 
 
 Cresole 115 
 
 Critical points 527 
 
 " range 527 
 
 Cross country mill 330-479 
 
 Crucible of blast furnace 132 
 
 " steel 174 
 
 Crude still 102 
 
 Cryohydrate 522 
 
 Crystals 533 
 
 Crystallization 5-348 
 
 of steel 533 
 
 water of 15 
 
 Cup and cone 139 
 
 Cupola 179 
 
 " charge 178-204 
 
 Cupped fracture 306 
 
 Cutting axles 511 
 
 " rails 450 
 
 " shapes 466-468 
 
 Cuyuna ore range 47 
 
WORD INDEX 
 
 603 
 
 Dalton 
 
 Debenzolating tower .... 
 
 Decalescence 
 
 Decane 
 
 Deep seated blow holes. . 
 
 10 
 100 
 527 
 
 70 
 346 
 
 Defects in semi-finished material 413 to 417 
 
 Definite proportions, law of 9 
 
 Deformed bars 469 
 
 Deliquescent substance 16 
 
 Delta connections 255 
 
 Dendrite 533 
 
 Density 4-34 
 
 Deoxidation of Bessemer steel 190-197 
 
 in electric furnace 269-286 
 
 Dephosphorizing slag 270 
 
 Destructive distillation 85 
 
 Desulphurizing 269-286 
 
 Detinned scrap 577 
 
 Diagonal method of rolling 447-465 
 
 Diamond, form of carbon 23 
 
 pass 323 
 
 Di-basicacid 19 
 
 Die for punching rail joints 459 
 
 Differential hardening 553 
 
 Diffusion 5 
 
 Dilation of steel 530 
 
 Dimethylaniline 112 
 
 Diphenylamine 112 
 
 Direct current 253 
 
 ' process, Siemens 199 
 
 Discard 343-498-509 
 
 Divided circuit 258 
 
 Dolomite 26-31-32-119-230 
 
 Double acting hammer 315 
 
 annealing 546 
 
 " bell and hopper 139 
 
 Down comer 141 
 
 " take 141 
 
 Drag-over mill ^. 330 
 
 Draught 339-381 
 
 Drawing coke 89 
 
 Dressing rolls 329 
 
 Drill tests for ore 47 
 
 (See drill exploration) 
 Drop bottom 179 
 
 " forging 318 
 
 " tests 309-515 
 
 Dry bases of analysis ' 41 
 
 Dry blast 150 
 
 Dry chemistry 15 
 
 Drying Bessemer bottoms 186 
 
 blast furnace 152 
 
 " O.H. furnace 218 
 
 Ductility ... 5 
 
 Duquesne Works 368 
 
 rail joint 453-455 
 
 gas cleaning plant 147 
 
 Duplex process 293 to 297 
 
 " slag 297 
 
 Dust catcher 146 
 
 Dynamic stress 301 
 
 Dynamo 250 
 
 Dyne 243 
 
 Eccentric converter 180 
 
 Edgar Thomson mills 447 
 
 splice bar shop 458 
 
 works 180 
 
 Edging pass 462-468-471 
 
 Efflorescence 15 
 
 Effusion 5 
 
 Elastic deformation 304 
 
 " limit 303-307 
 
 Elasticity 4 
 
 Electric arc 264 
 
 " furnaces 265 
 
 current 246 to 256 
 
 furnace slag 208 
 
 furnaces 261 to 267 
 
 " heating 262 to 267 
 
 pyrometers ^ 64-65 
 
 " steel, properties of 289 to 291 
 
 " steel, uses of 291 
 
 Electrical resistance pyrometer 64 
 
 Electrodes 277-279 
 
 Electrolysis 14-261 
 
 Electrolytes 8 
 
 Electro magnetic induction 249-250 
 
 Electromotive force 246 
 
 Electron (theory) 13 
 
 Element, chemical 
 
 Elongation, per cent, of 301-307 
 
 Empirical formula 68-69 
 
 Endothermic reaction 8-16 
 
 Energy 5-243 
 
 ' kinds of 
 
 " laws of 6 
 
 Erg 244 
 
 Ethane 70 
 
 Ether 
 
 Ethylbenzene 105-106 
 
 Ethylene 69 
 
 Eutectic alloy 348-523 
 
 Eutectoid 525 
 
 " steel 525-527 
 
 Exothermic reaction 8-16 
 
 Expansion tests 
 
 Exploration for iron ore 47 
 
 Explosion doors 141 
 
 Extension * 
 
604 
 
 WORD INDEX 
 
 Fahrenheit scale 
 
 Falling weight test (see drop test) . 
 
 Fatigue stress 
 
 Feldspar 
 
 Ferric oxide 
 
 Ferrite. . . 
 
 174 
 7 
 
 309 
 
 301 
 29-36 
 22-36 
 
 518 
 
 " in pig iron ...................... 127-524 
 
 " "steel .................... ..... 518-524 
 
 Ferro chromium ....................... 228 
 
 " manganese ................... 129-190-228 
 
 " phosphorus ...................... 228 
 
 " silicon ....................... 129-190-228 
 
 " vanadium ....................... 228 
 
 Ferroso-ferric oxide ..................... 22-36 
 
 Ferrous and ferric compounds ........... 27 
 
 " compounds .................... 27 
 
 oxide in open hearth ............ 234 
 
 products ...................... 169 
 
 " sulphate ....................... 27 
 
 Fibrous structure ...................... 170 
 
 Fillet ................................. 408 
 
 Fin .................................. 336-494 
 
 Finish on bars ......................... 489-495 
 
 Finished products ...................... 418-515 
 
 Finisher (see finishing stand) ............ 337 
 
 Finishing period in 0. H. process ........ 270 
 
 " rails ......................... 450-452 
 
 rolls ......................... 337 
 
 splice bars .................... 452 
 
 stand ........................ 337 
 
 steel ......................... 190-226 
 
 " temperature .................. 313-450 
 
 Firebrick ............................. 32 
 
 " clay .............................. 32 
 
 " clay brick ......................... 32 
 
 " cracks on rolls ..................... 495 
 
 " damp ............................ 69 
 
 Firestone ............................. 185 
 
 First helper ........................... 218 
 
 Fish oil for hardening .................. 552 
 
 Fixed carbon .......................... 76-85 
 
 Fixed converter ........................ 183 
 
 Flame in converter ..................... 188 
 
 " " open hearth ................... 220 
 
 Flare on hoop ......................... 473 
 
 Flats, finish of ......................... 489 
 
 " rolling of ......................... 468 
 
 Floors in open hearth ................... 209 
 
 Flue dust ............................. 38-146 
 
 Fluid compression ...................... 
 
 Fireclay .............................. 29 
 
 Fluor spar ..................... 120-223-268-270 
 
 Fluorine. . . 12 
 
 Flushing cinder (see tapping slag) 122 
 
 tar 
 
 Flux 
 
 " in the electric furnaces 
 
 Flying shear 
 
 Foot-pound 
 
 Force 
 
 Forge iron 
 
 Forging 
 
 axles 
 
 " hammer 
 
 " press ' 
 
 wheel blanks 
 
 Former bar 
 
 Foundry iron 
 
 Four pass stove 
 
 Fracture tests at blast furnace . 
 " open hearth.. 
 
 Fractures of steel 
 
 Free cementite . . . 
 
 .... 98-99 
 117 
 
 .... 271 
 .... 404 
 244 
 243 
 129 
 
 ....316-318 
 ... 508-510 
 316 
 .... 317 
 
 495 
 499 
 512 
 129 
 142 
 155 
 .... 224 
 
 305-537 
 
 520-526 
 
 " ferrite 520-526 
 
 Freezing of alloys 521 to 526 
 
 Frick Coke Company 86 
 
 " Furnace 264 
 
 Frothing slag (see run off) 222 
 
 Fuels 58 to 116 
 
 " for open hearth 203 
 
 Fuel oil 69 
 
 Full annealing 540 
 
 Fume from converter 193 
 
 Furnace cooling 459-545 
 
 lines 137 
 
 Fusion tests 33 
 
 Gag 451 
 
 Gag press 451 
 
 Gallium 11 
 
 Gamma iron 532 
 
 Gangue . 39 
 
 Canister 29 
 
 " brick 29-32 
 
 Gas 2 
 
 " holes (see blow holes) 346 
 
 " port... 211 
 
 " producer 71 to 75 
 
 " volume, laws of 2 
 
 " washer 148 
 
 Gaseous fuels 70 to 75 
 
 Gasoline 69 
 
 Gayley dry blast 150 
 
 Generalized formula 68-69 
 
 Generator, electric 250 
 
 Girod furnace 266 
 
 Glucinum ... 1 1 
 
WORD INDEX 
 
 605 
 
 Gob 
 
 86 
 
 Heat of vaporization 
 
 59 
 
 Goethite 
 
 37 
 
 " reactions in blast furnace 
 
 163-164 
 
 Gogebic iron range 
 
 43 
 
 " " " converter 
 
 194 
 
 Gold 
 
 11 
 
 " treating, axles .- 
 
 513-515 
 
 Gold-silver alloys 
 
 521 
 135 
 
 " car wheels 
 " " rail joints 
 
 506 
 459 to 461 
 
 
 323 
 
 ' ' treatment 
 
 539 to 568 
 
 
 55 
 
 " units 
 
 7 
 
 Grains 
 Gram 
 
 533-538 
 4 
 
 Heater 
 Heating furnace 
 
 336-362 
 ...419 to 422 
 
 " calorie (see small calorie) 
 
 7 
 
 " " cinder 
 
 419-421 
 
 " molecular volume 
 
 13-21 
 
 
 102 
 
 " weight 
 
 21-61 
 
 
 448 
 
 Granulated slag 
 
 152 
 
 Helix 
 
 247 
 
 Granulating pit. 
 
 152 
 
 Helical Pinions 
 
 334 
 
 Granulation of steel . . 
 
 534 
 
 Hematite 
 
 36 
 
 Graphite ... . 
 
 ....23-31-32 
 
 Heptane .... 
 
 . ... 70 
 
 Graphitic carbon 
 Gravitation, law of 
 
 127 
 4 
 
 Heroult furnace 
 Hexagon, rolling of 
 
 266-275 
 469 
 
 Gray iron 
 
 127 
 
 Hexane 
 
 70 
 
 
 264 
 
 High carbon steel 
 
 226 
 
 Grizzley bar 
 
 98 
 
 Hiorth furnace 
 
 264 
 
 Gronwall furnace 
 
 266 
 
 Hoist for blast furnace 
 
 14D 
 
 Grooved roll 
 
 323 
 
 Holley, Alexander 
 
 177 
 
 Grooves in rolls 
 
 323 
 
 Hollow boring for axles 
 
 513 
 
 Guard 
 
 334 
 
 Homogeneous substance 
 
 8 
 
 Guide 
 
 334-475 
 
 Hooke's law 
 
 4 
 
 " rounds 
 
 467 
 
 Hoop 
 
 470 
 
 " marks 
 
 415 
 
 
 473 
 
 " mill \ 
 
 . 330-475 
 
 
 470 
 
 Gun brick 
 
 94 
 
 
 470 to 473 
 
 " for blast furnace 
 
 155 
 
 
 139 
 
 Half cup fracture *. 
 
 306 
 
 
 244 
 
 Hammer forging 
 
 316 
 
 Hot blast 
 
 150 
 
 " steam 
 
 316 
 
 
 135 
 
 
 317 
 
 
 143 
 
 Hand guide mill 
 
 475 
 
 Hot iron 
 
 127-171 
 
 " mill 
 
 330-474 
 
 
 221 
 
 " round 
 
 467 
 
 
 313 
 
 Harden furnace 
 
 264 
 
 
 450-466-468 
 
 HarHfning P.arbnn 
 
 533 
 
 
 176 
 
 " process of 
 
 548 
 
 
 158 
 
 Hardness 
 
 5 
 
 " templet 
 
 441 
 
 
 311 
 
 " top mould 
 
 346 
 
 " test 
 
 309 
 
 " worked splice bars 
 
 ... 460 
 
 Head of rails 
 
 440 
 
 " working 
 
 ...313 to 321 
 
 " " water 
 
 245 
 
 
 2-332-368-378 
 
 Hearth brick 
 
 137 
 
 
 ....333-370 
 
 " jacket 
 
 132 
 
 " top . . 
 
 333-370 
 
 " of blast furnace 
 
 " 132 
 
 Howard Axle Works . . 
 
 ..508 to 515 
 
 " of open hearth 
 
 210 
 
 
 73 
 
 " and bosh brick 
 
 137-138 
 
 
 222 
 
 Heat 
 " cracks in rolls 
 
 6 
 495 
 
 Hydraulic press (see forging press) . 
 
 318 
 245 
 
 " of formation 
 
 60 
 
 
 
 245 
 
 Heat of fusion 
 
 59 
 
 Hydrocarbon . . . 
 
 24 
 
GOG 
 
 WORD INDEX 
 
 Hydrochloric acid 21 
 
 Hydrogen 12-22 
 
 Hydroquinol 109 
 
 Hydroxide : 19 
 
 Hygroscopic 16 
 
 Hyper-eutectoid steel 526 
 
 Hypochlorous acid 18 
 
 Hypo-eutectoid steel 526 
 
 Hysteresis 530 
 
 Idiomorphic crystals 533 
 
 Ignition point (see kindling temperature) . 70 
 
 Impact stress 301 
 
 " test 34-309 
 
 Impedance 259 
 
 Impenetrability 4 
 
 Impurities in steel 518 
 
 Incipient crack 508 
 
 Indigo 113 
 
 Inductance 259 
 
 Induction furnace 263 
 
 Inertia 3 
 
 Ingot 342 
 
 Ingot defects. . . '. 342 to 351 
 
 " mould 182-205 
 
 Ingotism 348 
 
 Inorganic chemistry 8 to 27 
 
 Inspection 417-430-451-493-498-504-509 
 
 department 493 
 
 of merchant mill products . . . 493 to 495 
 
 "plates 430 
 
 "rails 451 
 
 " semi-finished products 417 
 
 "wheels 504 
 
 Insulator 245 
 
 Invar 582 
 
 Inwall 136-137 
 
 " brick 137-138 
 
 Iodine 12 
 
 Ions 14 
 
 Iron 12-27-125 
 
 " action in the converter 194 
 
 " carbonate 27-36-37 
 
 " early history of 125 
 
 1 notch 132 
 
 " ore 36 
 
 " oxides 36 
 
 " oxides in open hearth 232 
 
 " pyrites 36 
 
 " silicates 36 
 
 " sulphates 27 
 
 " sulphides 36-39 
 
 Irregular fracture 306 
 
 Isomeric compounds 105 
 
 Jacket 
 
 Jones, W. R.. 
 
 132 
 
 177 
 
 Joule 244-245 
 
 Journal 512 
 
 Jump roll : a plain roll wi th collars on each 
 end. Used in rolling flats. 
 
 Kaolin (see clay) 29 
 
 Keeper, one in charge of a blast furnace . . ' 
 
 Keller furnace ' 266 
 
 Kelly, Wm 176 
 
 Kerosene 69 
 
 Kidney ore 37 
 
 Kilogram 4 
 
 " meter 244 
 
 Kilometer 4 
 
 Kilowatt 244 
 
 hour 244-262 
 
 Kindling temperature of gases 70 
 
 Kinks 494 
 
 Kish, graphite-like substance given off by 
 
 pig iron 
 
 Kjellin furnace 263 
 
 Koppers by-product coke oven 93 
 
 Laboratory tests on fuels 62 
 
 " refractories 33 
 
 Ladle, steel 204 
 
 " additions 190-226 
 
 " reactions 191 
 
 " test (see sampling) 192-229 
 
 Lag (see hysteresis). 530 
 
 Lake Superior ore 79 
 
 " " " district 42 
 
 Lamination : 435 
 
 Lap 336-494 
 
 Lard oil in quenching 552 
 
 Large calorie 7 
 
 Larry 86-97 
 
 Latent heat 59 
 
 Lauth.B. C ..... 423 
 
 " mill 423 
 
 Law of constancy of nature 15 
 
 " definite proportions 9 
 
 " "ebulition 59 
 
 " evaporation 59 
 
 " "fusion 59 
 
 " " heat exchange 58 
 
 ' " mass action 4 
 
 ' " multiple proportions 10 
 
 Lead 11 
 
 " bath, quenching ' 552 
 
 " hardening (sse quenching in lead) 552 
 
 " tempering 557 
 
 Lead-tin alloy 523 
 
 Leading spindle 333 
 
 Leg pipe 135 
 
 Lenz's law 250 
 
 Leveling in coke oven 88 
 
WORK INDEX 
 
 607 
 
 Liberty mill 
 Liftin" table 
 
 425 
 335-380 
 
 Malleability . 
 
 5 
 
 Manganese 
 in converter 
 in steel 
 
 ..11-27-128-176-235 
 194 
 570 
 
 Light oil 
 
 101-102-115 
 
 ' rails 
 Lignite 
 Lime 
 
 452 
 66-77 
 30-39 
 223 
 
 " sulphide 
 
 98fi 
 
 in steel for case hardening 563 
 Manipulator 873 
 
 " boil 
 
 " cooling 
 
 545 
 
 Mantle of blast furnace 
 
 136 
 
 Limestone 
 
 25 
 
 Marquette ore range 
 Marsh gas 
 Martensite 
 
 43 
 69 
 526-549-550-553-558 
 584 
 
 " analysis of 
 
 120-160 
 
 " as a flux 
 
 119 
 
 in blast furnace 
 in open hearth 
 Limiting angle of rolling 
 Limnite 
 Limonite 
 
 160-170 
 223-239 
 340 
 37 
 36-37 
 
 Martensitic nickel stel 
 
 Martin Brothers 
 
 201 
 
 " process 
 
 201 
 
 Mass 
 " action, law of 
 Massive pearlite 
 
 4 
 17 
 544 
 
 Liner 
 
 332 
 
 Lining of blast furnace 
 
 137 
 
 Matrix 
 
 585 
 
 " converter 
 " electric furnace 
 Linseed oil, quenching 
 Liquid 
 Lithium 
 Locus of a point 
 
 184 
 275 
 551 
 2 
 11 
 521 
 
 Matter... 
 
 2 
 
 " classes of 
 
 2 
 
 ' conservation of 
 ' sciences of 
 " Btatesof 
 Matthews and Stagg 
 
 2 
 3 
 2 
 551 
 
 Lodestone (see magnetite) 
 
 36 
 
 90 
 
 Maximum load 
 
 304 
 
 " stress 
 
 304 
 
 test piece 
 
 434 
 
 Mayari steel 
 
 591 
 
 Looping mill 
 
 330-477 
 
 Mechanical mixture 
 properties 
 Medium carbon steel 
 Melter 
 
 ..- 8 
 ....299-301-303-304 
 226 
 218 
 
 Lorry (see larry) 
 
 97 
 245 
 226 
 79 
 79 
 
 Loss in head . . . 
 
 
 Lower Freeport coal bed 
 
 Menominee ore range 
 
 43 
 
 Merchant bar ... . 
 
 173 
 
 
 182 
 
 mill 
 Mercury 
 
 474-482 
 
 
 38 
 
 11 
 
 Lute 
 
 97 
 
 aa quenching agent . 
 Metadihydroxylbenzene .... 
 Metal 
 
 551 
 110 
 9 
 
 Machine cast pig 
 Macroscopic structure 
 Macro-structure 
 Magascopic (see macroscopic) 
 Magnesia 
 Magnesian limestone (see dolomite) 
 
 152 
 518 
 518 
 518 
 . . 25-30-39-166 
 26-31-119-231 
 25-32-228-231 
 11-25 
 247 
 247 
 247 
 248 
 27-36-37 
 248 
 
 Metalloid 
 
 9 
 
 Metallurgy 
 
 2 
 
 Metasilicic acid 
 
 123 
 
 Metaxylene 
 Metcalf's Experiment 
 Meter 
 
 ...105-106 
 537 
 4 
 
 
 Methane 
 
 69-70 
 
 Magnet 
 
 Metric system 
 
 4 
 
 Magnetic fields 
 flux 
 induction 
 iron ore (see magnetite) . 
 
 Micro structure 
 
 518 
 
 Microscopic structure 
 Middle Kittanning 
 Mil 
 
 518 
 79 
 
 256 
 
 " foot 
 
 257 
 
 
 
 247 
 
 Mill, classes of 
 
 365 
 
 
 246 
 
 " shoe 
 
 322-332 
 
 Magnetite 
 
 27-36 
 
 Milling pit 
 system of mining ore 
 Mineral 
 
 52 
 50-52 
 
 
 218 
 
 
 172 
 
 36 
 
 " cast iron... 
 
 172 
 
 " oil in quenching 
 
 551 
 
608 
 
 WORD INDEX 
 
 Missabe ore range 46 
 
 Mixer 181-204 
 
 Modolus of elasticity 301-307 
 
 Molds for ingots 182-205 
 
 Molecule 2 
 
 " of elements 13 
 
 Molecular weight 19 
 
 Molten metal (see hot metal) 221 
 
 Molybdenum 11 
 
 Monell charge 219 
 
 " process 201 
 
 Monkey for blast furnace 134 
 
 " cooler 134 
 
 Morgan, C. H 397 
 
 " mill 397 
 
 Mothballs 113 
 
 Motor Benzol 107 
 
 Mother liquor 100-521 
 
 " metal 525 
 
 Moulds for ingots 182-205 
 
 Muck bar 173 
 
 Multiple proportions 9 
 
 Muriatic acid (see hydrochloric acid) 21 
 
 Mushet.R 176 
 
 Naphtha 69-101-105-111 
 
 Naphthalene 101-113-114 
 
 Naphthionic acid 114 
 
 Naphthol 114 
 
 Nascent state 13 
 
 Natural gas 66-71 
 
 Neck of a roll 322 
 
 Necking of test piece 305 
 
 Neutral flux 120 
 
 refractories 31 
 
 " substance (see salt) 8 
 
 Nickel 11-228 
 
 " steel 580 
 
 " in steel for case hardening 563 
 
 Nickel-Chrome steel 590 
 
 Nigger heads in open hearth 224 
 
 Ninety per cent, benzol 103-107 
 
 Nitric acid 25 
 
 Nitro benzene 109 
 
 Nitrogen 12-25 
 
 in blast furnace 165 
 
 Nitronaphthalene 114 
 
 Nitrotoluene 112 
 
 Nodulizing (see spheroidizing) 547 
 
 Nomenclature 18 
 
 Non Bessemer ore 40 
 
 " coking coal 80 
 
 " electrolyte 8-9 
 
 " metal 9 
 
 Normalize 547 
 
 Nose of converter ... 183 
 
 Occlusion see absorption) 5 
 
 Octane 70 
 
 Offtakes 140 
 
 Ohm 245-257-262 
 
 Ohm's law 256 
 
 Oil and tar burner 68 
 
 " hardened (see oil quenched) 551-552-553 
 
 " scrubbers 101 
 
 " temper (see also oil quenched) 551-558 
 
 " of vitrol (see HaSOi) 23 
 
 One level type of open hearth 209 
 
 Open hearth 209 
 
 " process 198-242 
 
 " slag 122-234-242 
 
 " steel, making of 198-242 
 
 " pass 323 
 
 " pit 48-50 
 
 " " mining 50 
 
 " top blast furnace 139 
 
 Optical pyrometer 65 
 
 Ore 36 
 
 " analysis of 160 
 
 " boil 222 
 
 " bridge 151 
 
 " composition of 36-37 
 
 " down 224 
 
 " grading of 55 
 
 " transportation of 56 
 
 " valuation of 38 
 
 Orientation 533 
 
 Orthosilicic acid 123 
 
 Ortho xylene 105-106 
 
 Oval groove 467 
 
 Overblown steel 575 
 
 Overburden 53 
 
 Oxidation 16 
 
 Oxide 16 
 
 Oxygen 12-22 
 
 '* in blast furnace 162 
 
 " in the electric furnace 268-269 
 
 575 
 
 Palaeozoic era 79 
 
 P and A tar extractor 99 
 
 Parabenzene sulphonic acid 110 
 
 Paraffin wax 69 
 
 Paraxylene 105-106 
 
 Pass in rolling 323-337 
 
 Pass-over mill (see pull-over mill) 330 
 
 Pass templet 441 
 
 Pearlite 127-520-525 
 
 " in pig iron 127-524 
 
 " in steel 519-523-526 
 
 Pearlitic chrome steel 588 
 
 nickel steel 584 
 
 Peat... . 66-77 
 
WORD INDEX 
 
 609 
 
 Pen stock 135 
 
 Pentane 70 
 
 Permanent magnet 247 
 
 set 303 
 
 Petroleum 66-68 
 
 Phase of electric current 253 
 
 " " pearlite 543 
 
 Phenylhydracine 108 
 
 Phenol 110 
 
 Phosphoric acid 25 
 
 Phosphoretic steel 574 
 
 Phosphorus 12-25 
 
 " in blast furnace 165 
 
 " " electric furnace 268 
 
 " open hearth 237 
 
 " ore 40 
 
 " pig iron 129 
 
 "steel 573 
 
 Phthalic acid 114 
 
 Phthalimide 113 
 
 Physical change 8 
 
 properties 3 
 
 " of plates 433 
 
 testing of steel 300 
 
 Picric acid 110 
 
 Pickling test 495 
 
 Pig and ore process 199 
 
 " scrap process 201 
 
 " casting machine 152 
 
 " iron, composition of 127-174-228 
 
 " nickel .\ 228 
 
 Pigging up 224 
 
 Piling, rolling of 465 
 
 Pillaring 158 
 
 Pinions 322-333-369-378 
 
 " housings 333 
 
 Pipe 343 to 346-495 
 
 " in axles 508-512 
 
 Pit annealing v 545 
 
 " casting 324 
 
 Pitch for rolls '. 382 
 
 " line 328 
 
 Pittsburgh coal bed 79 
 
 Plain steel 579 
 
 Planishing pass 337 
 
 rolls 337 
 
 Plastic clay 29 
 
 ' deformation 305 
 
 Plasticity 5 
 
 Plate mills 423-435 
 
 Plates, kindsof 423 
 
 " rolling of 423-425-428 
 
 Platinite 582 
 
 Platinum 11 
 
 Pneumatic process 175 
 
 Poling 192 
 
 Polyphase currents, electric 253 
 
 Pony roughing stand 337 
 
 Porosity 
 
 4 
 
 Potash (K 2 0) 
 
 20 
 
 Potassium 
 
 .... 12-166 
 
 Potential 
 
 .... 243 
 
 Pouring (see teeming) 
 
 .... 192-228 
 
 Powdered coal 
 
 81 
 
 Power 
 
 .... 244 
 
 " factor 
 
 .... 259 
 
 Preliminary test 
 
 .... 287 
 
 Press, forging 
 
 317 
 
 Pressing (see forging) 
 
 317 
 
 Primary austenite 
 
 525 
 
 Prismatic sulphur 
 
 23 
 
 Producer 
 
 71 
 
 " gas 
 
 .... 66-71 
 
 Progressive distillation analysis 
 
 .... 75-76 
 
 " hardening 
 
 .... 553 
 
 Propane 
 
 70 
 
 Properties of matter 
 
 3 
 
 Prospecting for ore 
 
 47 
 
 Proximate analysis 
 
 .... 75-76 
 
 Puddle bar (see muck bar) 
 
 173 
 
 " iron (see wrought iron) 
 
 .... 173 
 
 Pulling test 
 
 .... 301 
 
 Pull-over mill 
 
 330 
 
 Pulpit 
 
 179 
 
 Pulverized coal (see powdered coal) . . 
 
 . . . .81-82-83 
 
 " burner 
 
 84 
 
 Punching splice bars 
 
 .... 459 
 
 Purification processes 
 
 175 
 
 Pyro-chemical process 
 
 128 
 
 Pyrometers 
 
 64 
 
 Quartz 
 
 .... 24-36 
 
 
 
 Quenching 
 
 548-551 
 
 bath 
 
 552 
 
 " media 
 
 552 
 
 " tank 
 
 514 
 
 Quicklime (see calcium oxide) 
 
 25 
 
 Quick silver (see mercury) 
 
 11 
 
 Rabble 
 
 230 
 
 Radiation 
 
 59 
 
 " pyrometer 
 
 65 
 
 Radical 
 
 14 
 
 Ragging 
 
 340 
 
 marks 
 
 415 
 
 Rail, evolution of 
 
 437 
 
 " joints, rolling of 
 
 454 
 
 " " treatment of 
 
 459 
 
 " " types of 
 
 .....453-454 
 
 " mills 
 
 ..442 to 449 
 
 " rolling of 
 
 ..448 to 452 
 
 " steel, Bessemer 
 
 190 
 
610 
 
 WORD INDEX 
 
 Raisers 53 
 
 Raw materials (see basic materials) 58 
 
 Reactions, chemical 13 
 
 " in blast furnace 163-171 
 
 Reaumur 7 
 
 Recalescence 527 
 
 Recarbonizing (see recarburizing) 191-226 
 
 Recarburizing, Bessemer steel 191 
 
 open hearth steel 226 
 
 Recuperative furnace 421-422 
 
 principle 63 
 
 Red ore 36 
 
 " short (see hot short) 176 
 
 Redstone coal bed 79 
 
 Reduction 16 
 
 of area 301-307 
 
 Refractories 28 
 
 Regenerative chambers 212-217 
 
 furnace 419 
 
 principle 63 
 
 Reheating furnace 419 to 422 
 
 Repeater 330-479 
 
 Repeating mill 330 
 
 Rephosphorization 226-298 
 
 Rerun benzol : 103-104 
 
 " toluol 104 
 
 Resorcinol 110 
 
 Residual manganese 191 
 
 Resiliency 575 
 
 Resistance, electrical 256 
 
 " furnace 263 
 
 " pyrometer 64 
 
 Resting period 302 
 
 Retardation 527 
 
 Retention theory 555 
 
 Retort coke (sse by-product coke) 85 to 98 
 
 oven (330 by-product coke) 93 
 
 Reverberatory furnace 420 
 
 Reversing mill 330-366-367-368-369 
 
 Rhombic sulphur 23 
 
 Roasting (see calcining) 25 
 
 Roll bearings 331-371 
 
 " design for billet mills 394-402 
 
 " " blooming mills 373-377-379-383 
 
 " " " merchant mills 482 
 
 " " rails 438-447 
 
 " " general remarks 327 
 
 " how to study 438 
 
 " marks 495 
 
 " size of 327-384 
 
 " table 335 
 
 " tables 322 
 
 Rolling defects 413-494 
 
 history of 318 
 
 " principles of 319-321 
 
 Rolls 322 to 331 
 
 " for billet mills 393-400-402 
 
 " " blooming mills 371-377-379-383 
 
 Roof of open hearth 211 
 
 Rotary shears 429 
 
 Rough surface, cause of 415 
 
 turning axles 512 
 
 Roughing passes 444-450-483-484 
 
 stand 477 
 
 rolls 337-423-444-416 
 
 Rounds, kinds 467 
 
 rolling of 467 
 
 " straightening of 468 
 
 Run off slag 222 
 
 Run-of-mine coal 86 
 
 Runners 141-155 
 
 Running stopper 205 
 
 Saccharin 112 
 
 Sack's mill 331 
 
 Salamander 132 
 
 Salt 8 
 
 Salt and ice 522 
 
 Salicylic acid 115 
 
 Sampling pig iron 155 
 
 " Bteel 192-229 
 
 Sand 32 
 
 " as a flux 118 
 
 " as a refractory 29 
 
 " bottom furnace 419 
 
 " roUV 323 
 
 " seal 514 
 
 " test 155 
 
 Sandstone (a sedimentary rock) 
 
 Saturated solution 522 
 
 Sauveur 517 
 
 Scabs 349-415 
 
 Scaffolding 158 
 
 Scale .,..' 233 
 
 Scarf 538 
 
 Schedule for merchant mills 482-485 
 
 Schoen, Chas. E 497 
 
 " mill 331-500 
 
 Scleroscope 309 
 
 Scramming 50-53 
 
 Scrap and pig process 201-219 
 
 " in Bessemer process 188 
 
 " in open hearth process 201-219 
 
 Screw down 333 
 
 " steel 190 
 
 " stock 190-572 
 
 Scull (see skull) 225 
 
 Seams, cause of 346-415 
 
 " in bars 494 
 
 in blooms, billets, etc 415 
 
 " in rails 452 
 
 Second helper 218 
 
WORD INDEX 
 
 611 
 
 Second strand 337-485 
 
 Section mill (see shape mill) 461 
 
 Segercone 33 
 
 Segregated cementite 544 
 
 Segregation 348 
 
 in axles 508-512 
 
 Selective freezing 524 
 
 Selenium 12 
 
 Self-fluxing ore 117 
 
 " inductance 259 
 
 Semi-continuous mill 477 
 
 " finished products 364 to 414 
 
 Sensible heat 58 
 
 Sesquioxide base 124 
 
 Seventy- two hour coke 87 
 
 Sewickley coal bed 79 
 
 Shaft of blast furnace (see stack) 136 
 
 Shakedown 225 
 
 Shape mill 461 
 
 Shaped bloom 463 
 
 Shear foreman 490 
 
 " steel 174 
 
 Sheared plate mill 423 
 
 Shearing defects 415 
 
 " forces ! 301 
 
 ...429-433 
 
 301 
 
 Shears for blooming mill 372 
 
 Sheet bar 190-408 
 
 "mill 408-413 
 
 "rollingof 406to413 
 
 Shoe of a mill 322-332 
 
 Shore scleroscope 309 
 
 Shrinkage of steel in ingot 343 
 
 Side blown converter 183 
 
 Side shears 429 
 
 Siderite 37 
 
 Siemens, William. 
 
 Siemen's direct process 199 
 
 " furnace 199 
 
 " pig and ore process 201 
 
 pyrometer (see water pyrometer) 64 
 
 steel (see open hearth steel) 198 
 
 Siemens-Martin process 201 
 
 Silica 24-39 
 
 " brick 29 
 
 " in Bessemer process 
 
 " " blast furnace 165 
 
 " " open hearth process 234 
 
 "sand 32 
 
 Silicate 121-123 
 
 Silicon, acids of . . jf 
 
 " in converter 
 
 " " open hearth process 
 
 " "pig iron 24-127-129-171 
 
 Silicon in steel for case hardening 
 
 Silico-spiegel. . 
 
 Silver 
 
 Simple steel 
 
 Sine curve 
 
 Single coil of hoop 
 
 5C3 
 
 129 
 12 
 579 
 252 
 473 
 
 ' level open hearth (see one-level) . . . 209 
 " phase current .................... 252-3 
 
 Sinkhead ............................. 206-324 
 
 Sintering .............................. 219 
 
 Skelp ................................. 190 
 
 Skew'back ............................ 211 
 
 " brick .............. .............. 211 
 
 " channels ........................ 211 
 
 Skewed rolls ........................... 404 
 
 Skimmer .............................. 279 
 
 Skin of ingot .......................... 343 
 
 Skip hoist ............................. 140 
 
 Skull ................................. 225 
 
 Slab .................................. 365 
 
 " rolling of ....................... 385 to 390 
 
 Slab-and-edging method of rolling flats ... 468 
 
 " " " rails 438 to 449 
 
 Slabbing mill ....................... 385 to 388 
 
 Slaked or Slacked lime (CaO+H20) ........ 
 
 Slag ............................... 15-120-152 
 
 " in Bessemer process ................ 122 
 
 " "blastfurnace .................... 121 
 
 " " electric process .................. 122-288 
 
 " functions of ........................ 120 
 
 " hole .............................. 210 
 
 " notch (see cinder notch) ............ 134 
 
 " in open hearth process ........... 122-234-242 
 
 " pockets ........................... 211 
 
 Slagging tests ......................... 34 
 
 Sleevebrick ........................... 205 
 
 Slicing, system of mining ............... 53 
 
 Slips ................................. 157 
 
 Slivers ................................ 41M94 
 
 Sloppy heat ........................... 
 
 Slurry ................. ............... 138 
 
 Small calorie .......................... 7 
 
 Smelt ................................ H7 
 
 Snake ................................ 435 
 
 Snort valve ........................... 150 
 
 Soaking ingots ........................ 359-362 
 
 pi t ........................... 342-352 
 
 " "efficiency of ................. 362 
 
 Soda (same as sodium oxide) ................ 
 
 (also sodium carbonate) .................. 
 
 Sodium ............................... H-166 
 
 Soft steel (see low carbon steel) .......... 226 
 
 Solid ................................. 
 
 " solution .......................... 519-521 
 
612 
 
 WORD INDEX 
 
 Solute, dissolving substance 
 
 Solution 8 
 
 Solvent, substance dissolved '. 
 
 naphtha 113 
 
 Sorbite 543-559 
 
 Sorbitic steel 543 
 
 Spalling test 35 
 
 Spanner bar 432 
 
 41 block 432 
 
 Spar (see fluor spar) 120 
 
 Spawling tests (see epalling tests) 35 
 
 Special steel 579 
 
 Specific gravity ' 5 
 
 " heat 58 
 
 " pyrometer 64 
 
 " resistance 257 
 
 Speed in rolling 339 
 
 Spheroidize 547 
 
 Spiegel 129-190-228 
 
 " cupolas 180-204 
 
 Spindle 322-334-369-378 
 
 Splasher 132 
 
 Splice bar 453-461 
 
 " "analysis of 190 
 
 " finishing of 458 
 
 " rolling of 454 
 
 Spoon 225 
 
 Spout of open hearth 210 
 
 Spread in rolling 319-328 
 
 Spring heat (spring steel heat) 
 
 Square mil 257 
 
 Stack of blast furnace 136 
 
 " " open hearth furnace 217 
 
 Standofrolls 322 
 
 Standard ferro 129 
 
 Standards (see housings) 322 
 
 Star connections 255 
 
 Stassano furnace 265 
 
 Static stress 301 
 
 Stationary furnace (one not movable) 
 
 Stead 533 
 
 Steam shovel mining 48-50 
 
 Steel 173 
 
 " ladle 204 
 
 " rolls 323-326 
 
 " spout 210 
 
 " tie, rolling of 465 
 
 Sticker 224 
 
 Stock distributor 140 
 
 " house 151 
 
 " indicator 140 
 
 " line 133 
 
 " pile (see ore pile) 131-150-151 
 
 " yard 131-207 
 
 Stone (see limestone) 25 
 
 Stool... 182 
 
 Stopper 182-205 
 
 Stove burners 143 
 
 " linings 145 
 
 " valves 143 
 
 Stoves, kinds of 142 
 
 Straightening axles 511 
 
 bars 492 
 
 machine 427-468-492 
 
 plates 427-433 
 
 press 451-511 
 
 rails 451 
 
 rolls 427 
 
 rounds 468 
 
 shapes 466 
 
 Strain 5 
 
 Strand 337 
 
 Stratified structure 519 
 
 Straw oil 101 
 
 Stress 5 
 
 " kinds of 301 
 
 Stretch (see elongation) 301-307 
 
 Strip 470 
 
 Stripper 181-205 
 
 Stripping 50-51 
 
 Structural formula 68-69 
 
 shapes, rolling of 461 to 469 
 
 steel, testing of 301 
 
 Subcutaneous blow holes (see blow holes 
 
 near the skin) 346 
 
 Substance : 2 
 
 S-ulphanilic acid 112 
 
 Sulphates (see salts of sulphuric acid) .... 27-116 
 
 Sulphides 18 
 
 Sulphur 12-23 
 
 in Bessemer process 175 
 
 " blast furnace 165-171 
 
 "open hearth 236 
 
 " pig iron 
 
 Sulphuric acid 
 
 Sulphurous acid 
 
 Superficial hardening . . 
 Surface defects . . . 
 
 128 
 
 23 
 
 19 
 
 568 
 
 .413-435-451-494-504 
 
 Surface hardening (see case hardening) ... 562 
 
 Sweat 221 
 
 Swedish Bessemer process 176 
 
 " iron 176 
 
 Sweep 324 
 
 Symbols, chemical 9 
 
 Synthesis 16 
 
 Table, of a tee shape 466 
 
 Tables, of a mill V. 335 
 
 Talbot furnace 294 
 
 process 201 
 
 Tandem mill (see Belgian mill) 477 
 
WORD INDEX 
 
 613 
 
 Tap a heat (see tapping furnace) 154-225 
 
 Tap hole, blast furnace 132 
 
 " " open hearth furnace 210 
 
 Tape size 504 
 
 Tapping, blast furnace 154 
 
 " electric furnace 281 
 
 hole 132-210 
 
 " open hearth furnace 225 
 
 rod 225 
 
 slag (see basic slag) 122 
 
 Tar 99-114 
 
 " as a fuel 67 
 
 " burner 68 
 
 " extractor 99 
 
 " refmementof 114 
 
 ' uses of 115 
 
 Teaming (same as teeming) 192-228 
 
 Teerail 436 
 
 " rolling of 466 
 
 Teeming, Bessemer steel 192 
 
 electric steel 281 
 
 " ladle 182 
 
 open hearth steel 228 
 
 Temper colors 556 
 
 Temperature 6 
 
 " tests in open hearth 225 
 
 Tempering 554 
 
 austenic steel 558 
 
 " martensitic steel 558 
 
 troostitic steel 558 
 
 Templet 439-440-441-442 
 
 Temporary magnet 248 
 
 Tenacity (see tensile strength) 301 
 
 Tensile strength 301-307 
 
 Tensional stresses 301 
 
 Terminology 18-19 
 
 Ternary compounds 18 
 
 " steels 580-584-587-595 
 
 Test Mould. .; 229 
 
 Test piece 302-309 
 
 1 224-225 
 
 515 
 
 ...303-309 
 33 
 
 .299 to 311 
 149 
 58 
 
 .527 to 533 
 64 
 64 
 6-7 
 ...175-177 
 
 spoon. 
 
 Testing axles 
 
 machine 
 
 refractories 
 
 " steel 
 
 Theiscn's cleaner 
 
 Thermal capacity 
 
 critical points 
 
 Thermo- electric couple 
 
 pyrometer .... 
 
 Thermometer 
 
 Thomas-Gilchrist process . 
 
 Three-high blooming mill 377 to 384 
 
 " mill 330 
 
 " pass stove 142 
 
 " phase current 253-255 
 
 Ties, rolling of 464 
 
 Tilt hammer 316 
 
 Tilting converter 183 
 
 " furnace 200-294 
 
 " table 335 
 
 Tin II 
 
 " insteel 577 
 
 Titania (Ti02) 166 
 
 Titanium 12-165 
 
 T.N.T Ill 
 
 Tolerance, reasons for!90-226-329-348-428-430-467 
 
 Toluene Ill 
 
 sulphonic acid 112 
 
 Toluidin 112 
 
 Toluol 101-105-106-111-112-113 
 
 Tongue and groove pass 406 
 
 Top brick 137-138 
 
 " of blast furnace 139-141 
 
 Torsional stresses 301 
 
 test 308 
 
 Total carbon 127 
 
 Toughening 559 
 
 Trade heat 219 
 
 Trainof rolls 477 
 
 Transformation range (see critical range) 527 
 
 Transformer 260 
 
 Transporting ores 56 
 
 Tri-basic acid 19 
 
 Trinitrotoluol (T. N. T.) Ill 
 
 Triplex process 297 
 
 Troostite. 550-556 
 
 True annealing 540 
 
 Trunnel head 87 
 
 Try hole 140 
 
 Tub(seeladle) 204 
 
 Tungsten 12 
 
 Turgite 37 
 
 Turning rolls 329 
 
 Tuyere 134 
 
 " brick 134 
 
 " cooler 134 
 
 " cap 135 
 
 " stock 135 
 
 Tweer (see Tuyere) 134 
 
 Twere (see tuyere) 134 
 
 Twisted guide (see twisting guide) 400 
 
 Two Level Open Hearth 209 
 
 Two-high mill 330 
 
 Two-pass stove 142 
 
 Two phase current 253 
 
 Twyere (see tuyere) 134 
 
 Ultimate analysis 75-76 
 
 strength 304-307 
 
 Uncombined c arbon (see graphite) 1 27 
 
 Underfill 494 
 
614 
 
 WORD INDEX 
 
 Universal mill 
 
 . .330-431-433 
 
 Watering coke 
 
 89-97 
 
 " plates 
 
 432 
 
 Watt 
 
 245-262 
 
 Up-and-down takes 
 
 211 
 
 Watt-hour 
 
 262 
 
 Upper Freeport coal bed 
 
 79 
 
 Waynesburg coal bed 
 
 79 
 
 " Kittanning coal bed 
 
 79 
 
 Web holes 
 
 504 
 
 Upset test 
 
 495 
 
 " of a rail 
 
 440-442-449 
 
 Uranium 
 
 ... 11 
 
 Welding steel 
 
 538 
 
 Valence 
 
 10 
 
 Well of open hearth 
 Wellman furnace 
 
 212 
 200 
 
 Vanadium 
 
 12 
 
 Wenstrom mill 
 
 331 
 
 " in case hardening 
 
 563 
 
 Wet chemistry 
 
 15 
 
 steel 
 
 595 
 
 Wheel blank. 
 
 497-504 
 
 Vaporization heat of 
 
 59 
 
 " seat 
 
 512 
 
 " "in hardening 
 
 552 
 
 
 540 
 
 V-connection 
 Vermilion ore range 
 Vibrating spindle 
 Viscosity of quenching liquid 
 Volatile matter of coal 
 Volt 
 
 225 
 46 
 334-370 
 ,552 
 76-98 
 245-262 
 
 " metal 
 Wicket 
 Wind box 
 Window of housings 
 Wobbler 
 Wood 
 
 331 
 211 
 186 
 332 
 322 
 66-76 
 
 Voltaic cell 
 
 246 
 
 Work 
 
 243 
 
 Wabbler (see wobbler) 
 
 322 
 
 " effects of on steel 
 Working period in open hearth . . . 
 
 312-535 
 223 
 
 Wall of blast furnace 
 Walls of open hearth 
 
 136 
 
 210 
 
 Works annealing 
 Wrought iron. . . 
 
 540 
 170-171-173 
 
 Wash heat 
 
 219 
 
 
 
 " oil 
 
 101 
 
 Xanthosiderite 
 
 37 
 
 Washing open hearth furnace 
 
 219 
 
 Xylene 
 
 105 
 
 Waste heat 
 
 87 
 
 Xylol ; 
 
 105 
 
 Water, composition and formula. . . 
 4 ' annealing 
 
 13 
 545 
 
 Yield point 
 
 303 
 
 
 64 
 
 Y connections 
 
 255 
 
 
 551 
 
 Yellow brass 
 
 331 
 
 " gas 
 hardening (see quenching) . . 
 quenching 
 " seal 
 separator 
 " trough... 
 
 71 
 551 
 551 
 73 
 149 
 139 
 
 Youngs modulus 
 
 Zee's (or Z's) 
 Zinc 
 " in blast furnace 
 Zone of fusion . . . 
 
 307 
 
 466 
 11 
 166 
 ...167-169 
 
14 DAY USE 
 
 :ETURN TO DESK FROM WHICH BORROWED 
 
 LOAN DEPT. 
 
 This book is due on the last date stamped below, or 
 on the date to which renewed. 
 i books are subject to immediate recall 
 
 . 
 
 ~~ 
 
 LD 21A-50m-12,'60 
 (B6221slO)476B 
 
 m General Library 
 
 University of California 
 
 Berkeley 
 
YB 53337 
 
 425274 
 
 UNIVERSITY OF CALIFORNIA LIBRARY