FRANKLIN INSTITUTE LIBRARY 
 
 PHILADELPHIA 
 
 tmmm 
 
PORTLAND CEMENT. 
 
 A MONOGRAPH. 
 
 BY 
 
 CHARLES D. JAMESON, 
 
 -v?iiM:V':>i/s:)c.:c;*4 : 
 I *. .* * *. : *■•; I *. I . . 1 1 •! . ! . 
 
 PROFESSOR O!' i'.NGrNK'cRIXr., STAI'l'. 1* X 1 **■' RSri'V OI' |f)\VA. 
 
 ( 
 
 IOWA CITY: 
 Kepublican Printing Company. 
 1 SOo. 
 
a? ACS 
 
 TP 
 
 Copyright, 
 By Charles D. Jameson, 
 
 :• , . .1895- . . .• 
 
HON. D. N. RICHARDSON, 
 
 DAVEXPf)RT, IOWA: 
 
 DEAR SIR: 
 
 I BEG TO DEDICATE THIS MONOGRAPH TO YOU IN 
 MEMORY OF YOUR MiVNY KINDNESSES TO ME DURING 
 THE EIGHT YEARS OF MY PROFESSORSHIP IN THE STATE 
 UNIVERSITY OF IOWA, AND TO SHOW MY vVPPRECIATION 
 OF THE FACT THAT, AS A MEMBER OF THE BOARD OF 
 REGENTS OF THE STATE UNIVERSITY OF IOWA FOR 
 EIGHTEEN YEARS, YOU HAVE DONE MORE THAN ANY 
 OTHER ONE MAN TO RAISE THE UNIVERSITY TO ITS 
 PRESENT HIGH LEVEL. 
 
 RESPECTFULLY YOURS, 
 
 CHARLES D. JAMESON, 
 
 PROFESSOR OF ENGINEERING, 
 
 STATE UNIVERSITY OF IOWA. 
 
 Iowa City, 
 
 May 5, iSgs- 
 
 3 07 2.^ 
 
PREFACE. 
 
 The following monograph is the outgrowth of a short 
 course of lectures delivered b}- me each year to the Junior 
 Engineering students of the State University of Iowa, upon 
 Limes, Mortars and Cements. Owing to the great increase 
 in the amount of Portland cement used and the great value it 
 possesses as a building material, this monograph treats only 
 of the making, testing and using of Portland cement. 
 
 The term "Portland Cement" as here used means an 
 artificial cement made by mixing in certain known propor- 
 tions, clay and chalk containing silica, alumina, iron, and 
 carbonate of hme, and burning this mixture to^the point of 
 incipient vitrifaction and then reducing this burned product 
 to an impalpable powder. 
 
 The term "Portland Cement" primaril}^ means an artificial 
 mixture. The term " Natural Portland " has very much the 
 same meaning as natural artificial would have. 
 
 I wish to acknowledge my indebtedness to Dr. L. W. 
 Andrews, Ph. D., Professor of Chemistry, State University 
 of Iowa, for his kindness in writing the chapter upon the 
 Chemical Processes, etc., page 91, and where quotations 
 have been made from existing publications, credit has been 
 given in a foot note. 
 
 My endeavor in writing this monograph has been to place 
 before the engineering professsion the following essential 
 points : 
 
vi PREFACE. 
 
 JP^irst. — In the making of Portland cement. The selection 
 of the raw materials, their proper treatment by the different 
 methods in general use. The brnming of this material with 
 the types of kilns used. The reduction of the clinker to 
 cement powder and its proper storage. 
 
 Second. — In the testing of Portland cement. The require- 
 ments as set forth in the different standard specifications. 
 Methods to be followed in the testing, and the various 
 mechanical devices used in making the tests. 
 
 Third. — In the using of Portland cement. The compar- 
 ative value of different cements. The uses of Portland 
 cement as a material of construction. Proper methods of 
 manipulation. Estimates of quantities and cost. 
 
 More space has been devoted to the testing and using of 
 cement than to its making; for the reason, that, upon these 
 two points every civil engineer should be thoroughly informed 
 while a knowledge of the details of cement making is, from 
 necessity, confined to a limited class of cement experts. 
 
 I venture hope that this book may prove of value not only 
 to students in engineering, but to the engineering profession 
 at large. 
 
 Charles D. Jameson. 
 
 Iowa City, 
 May gill, iSgs- 
 
CONTENTS. 
 
 CHAPTER I. 
 
 GENERAL CONSIDERATIONS. 
 
 Pages 
 
 Importance of a Correct Knowledge of the Proper- 
 ties of Limes and Cements. — Constituents of Lime. 
 — Hydraulic Lime and Cement. — The burning of 
 Lime. — Lime Kiln. — Slaking. — Hydraulic Lime. 
 — Cement. — Natural Cement. — Portland Cement. i-io 
 
 CHAPTER II. 
 
 HISTORICAL DATA. 
 
 Antiquity of the Use of Cement. — Marcus Vitruvius 
 Pollio on Lime. — Pozzolana. — Cement Experi- 
 ments of John Smeaton. — Early Patents on Port- 
 land Cement. — Cement Making in German}-, 
 France, and the United States 
 
 CHAPTER III. 
 
 MANUFACTURE OF PORTLAND CEMENT. 
 
 Selection of Raw Material. — Deposits in the United 
 States. — Location of Works. — Design of Works. 
 — Reduction of Raw Material. — Wet Process. — 
 Wash Mills. — Dry Process. — Semi-Wet. — Mixing- 
 Machines. — Pug Mill. — Burning Cement. — Dome 
 Kiln. — Hoffman Ring Kiln. — Bock Tunnel Kiln. 
 — Other Patent Kilns. — Grinding. — Sifting. — 
 Storing. — Packing 16-41 
 
viii 
 
 CONTENTS. 
 
 CHAPTER IV. 
 
 TESTING CEMENTS 
 
 Pages 
 
 Tensile Strength. — Crushing. — Fineness. — Form of 
 Briquette. — Making Briquettes. — Molds. — Bri- 
 quette Machines.—Bohme Hammer.— Jameson Bri- 
 quette Machine. — Clips.— Mixing Mortar. — Sand. 
 — Pans and Slabs. — Tanks. — Increased Tempera- 
 ture. — Setting. — Vicat's Needle Test. — German 
 Needle Test.— S. U. I. Needle Test.— Adhesion. 
 — Cross -Strain. — Blowing. — Warm Tests. — 
 Abrasion. — Association of German Cement Man- 
 ufacturers 42-68 
 
 CHAPTER V. 
 
 .\BSTRACTS FROM AMERICAN, GERMAN, FRENCH, 
 AND AUSTRIAN SPECIFICATIONS FOR 
 PORTLAND CEMENT. 
 
 Report of the Committee of the American Society 
 of Civil Engineers. — German Specifications for 
 Standard Portland Cement Tests.— French Speci- 
 fications for Portland Cement. — Austrian Specifi- 
 cations for Fineness and Strength of Cement. — 
 English Specifications for Portland Cement . . 69-90 
 
 CHAPTER VI. 
 
 THE CHEMICAL PROCESSES CONCERNED IN THE 
 HARDENING OF HYDRAULIC CEMENTS, 
 By Liiuncelot Andrews, Ph.D., Professor of Cheinistrj', 
 St.ite I'niversity of Iowa. 
 
 I. Classification.— II. The Cause of the Setting 
 of Cements in General, and Conditions Affecting 
 It. — III. Composition and Characteristics of Port- 
 land Cement. — IV. Chemical Processes Occur- 
 ring during the Hardening of Portland Cement . 91-98 
 
CONTENTS. 
 
 ix 
 
 CHAPTER VII. 
 
 CEMENT TESTING MACHINES AND STONE CRUSHERS. 
 
 Pages 
 
 Riehle Brothers' looo-pound Cement Tester. — Tin- 
 ius Olsen Cement Tester.— Fairbanks Cement 
 Tester.— Farrel Marsden Stone Crusher.— Re- 
 volving Screen.— Gates Rock and Ore Breaker. 
 —Tables of Capacities, Requisite Power, Cost, etc. 99-106 
 
 CHAPTER VIII. 
 
 THE USE OF PORTLAND CEMENT. 
 
 Lime Mortar.— Lime and Cement Mortar. — Sand. 
 Water. Mixing. — Cement as Plaster. — Con- 
 crete.— Materials.— Size of Aggregates.— Reduc- 
 tion of Aggregates. — Screening Gravel. — Hand 
 Broken. — Mixing Concrete. — Platforms. — Ma- 
 chine Mixing.— Depositing Concrete —Depositing 
 under Water. — Pavements and Sidewalks. — Ma- 
 terials. Description of Methods. — Cement Curb- 
 ing. —Quantities.— Repairing Masonry. — Artifi- 
 cial Stone. — Monolithic Concrete Structures. — 
 Silicating Process.— Victoria Stone.— Sylvester's 
 Process of Repelling Moisture from Exterior Walls. 
 
 McMurtie Stone. — Frear Stone. — Ransom 
 
 Stone.— Sorel Stone.— Estimate of Quantities.— 
 
 Lime and Cement Mortar and Concrete . . . 107-140 
 
 CHAPTER IX. 
 
 CEMENT TESTS. 
 
 Engineering Department, State University of Iowa. 
 —No. I. Standard Tests of Natural and Pordand 
 Cements, by Crellin, Howe, and Remley, class '90; 
 diagrams, etc., by Remley.— No. 2. Comparative 
 Effect of Fresh and Salt Water in Cement Mortar, 
 by M. 1. Powers, class '9[.— No. 3. Results Ob- 
 tained by Subjecting the same Brands of Cements 
 to both the Hot and Cold Water Tests, by Frank 
 Woolston, class '93 H^-i^S 
 
PORTLAND CEMENT. 
 A MONOGRAPH. 
 
 CHAPTER L 
 GENERAL CONSIDERATIONS. 
 
 Among the most important of building materials may be 
 placed lime and cement. There is very little "Engineering 
 Construction" into which they do not enter to a greater or 
 less extent, and their ultimate value depends more upon the 
 manner in which they are used than is the case with most 
 building materials. 
 
 Take wood, stone, or iron, for example, provided a good 
 quality has been selected, nothing the constructor can do will 
 materially affect this quality. The material may be used inju- 
 diciously, and uneconomically, but the ultimate strength re- 
 mains the same. A square inch of steel will stand about the 
 same tensile strain under all circumstances. With limes and 
 cemenis, however, this is not the case. No matter how per- 
 fect the original quality of the cement mav be, it may become 
 absolutely worthless and a source of danger, if not properly 
 handle and applied after being received. 
 
 This fact makes it of the utmost importance that, at least, 
 all engineers should be familiar with the manufacture, use, 
 and methods of testing limes and cements. They should know : 
 
2 
 
 PORTLAND CEMENT. 
 
 First. — The best and most rapid methods of testing cements, 
 in order to ascertain the quality and characteristics of the 
 cement received. 
 
 Second. — Knowing the quality of the cements to be used 
 and the results to be attained, they should be familiar with 
 the most approved methods of manipulation in order that the 
 desired results may be attained, and attained at a minimum 
 cost. 
 
 Third. — They should be familiar with the maximum results 
 that may be obtained with an}^ given type of cement. 
 
 A thorough understanding of these three points is necessar}' 
 in order to make the general use of cement safe and at the 
 same time economical. The engineer should be able to 
 recognize a good cement when he gets it, and if it is not 
 good, to indicate the probable reason of its failure, whether in 
 the raw material, the method of making, or the treatment it 
 received after the makincif. 
 
 Both lime and cement, when used for building purposes, 
 are mixed with a certain amount of water and used in a more 
 or less plastic condition. While in this plastic condition the}' 
 are placed in the w^ork in whatever position or form is required 
 and then this mixture hardens with more or less rapidity. 
 
 This hardening is called scttiiig; and it is this property of 
 setting under different conditions that forms one of the radical 
 differences between limes and cements. 
 
 Lime Mortar will only set when exposed to the action of 
 the air. and therefore can only be used in la3'ers so thin that 
 the air can penetrate to all parts of it. All parts of this lime 
 mortar must not onh' be accessible to the air. but to insure 
 setting the air must be dry. 
 
 Where lime mortar has been used in cellars that are damp 
 and in the plastering of houses exposed tc the damp sea winds 
 its absorption of moisture has been so great that it never has 
 become thoroughly set. and is always more or less dagiip and 
 soft. With cement and cement mortar this is not the case. 
 
 A mixture of cement and water, properly made, will not 
 only set in the open air. but will set when immersed in water 
 
LIME AND CEMENT. 
 
 3 
 
 or when in a vacuum. That is, contact with the air is not 
 necessary, in order that the process of setting may take 
 place. In fact, not only is contact with the air not necessary 
 for the setting of cement, but in order that the maximum 
 results may be reached, all cement mortar should be kept 
 either wet or immersed until it has become thoroughly set. 
 
 This may be considered as one of the most important rules 
 governing the use of cement, viz: 
 
 The quality of any cement work is very materially improved 
 by keeping it wet during the process of setting. Some idea 
 can be formed of the amount of the improvement due to 
 keeping the work wet, by a study of the diagrams in Chapter 
 IX. From this fact, that contact with the air is not necessary 
 for the proper setting of cement, it is evident that there is 
 almost no limit to the mass of cement mortar or concrete that 
 can be used. No matter how massive the structure may be, 
 and no matter in what thickness the cement mortar or concrete 
 may have been used, if the proper materials have been 
 properly manipulated, this mass will set thoroughly through- 
 out. Therefore while lime can only be used in a dry location 
 and in thin layers, cement can be used in any location and in 
 any quantities. 
 
 Lime mortar under the best of circumstances has very little 
 strength, either of adhesion or cohesion, while the best Port- 
 land cements properly used attain a strength superior to that 
 of any of the building stones, with the exception of some 
 granites, quartzites, and trap. 
 
 The chemical changes and reactions that take place during 
 the process of setting, in either lime or cement, are not, 
 apparently, well understood, as much conflict of opinion exists 
 among chemists upon this question. In Chapter VL by 
 Launcelot Andrews, Ph. D., F. C. S., will be found the results 
 of the most recent chemical researches upon the question. 
 
 The general differences between limes and cements, from 
 an engineering standpoint may, therefore, be taken as lying 
 in the fact that limes will set only in contact with dry air 
 while for the setting of cement, not only is the presence of 
 
4 PORTLAND OEMENT. 
 
 dry air not necessary but the best results obtain when the 
 cement is kept wet or immersed in water. 
 
 There can be no sharp line drawn between limes and 
 cements, although there is no difficulty in distinguishing at sight 
 between pure lime and good cements. 
 
 The ordinary lime of commerce consists of the calcined 
 carbonate of lime in a state of greater or less purity. 
 
 The constituents of cement are carbonate of Ume, sihca, and 
 alumina with iron, with a few other ingredients of more or 
 less importance. These three, carbonate of lime, silica, and 
 alumina, with iron, however, are the most important, and are 
 always present in varying proportions. It is the relative 
 proportions in which these constituents are mixed that make 
 the resulting cement more or less hydraulic, that is, the power 
 of setting under water, as the hydrauHcity of the resulting 
 compound varies as the percentage of the ingredients vary. 
 We have cementing compounds from pure lime at one end of 
 the list to Portland cement at the other. They can be 
 divided into the following three classes: Lime, Hydraulic 
 Lime, and Cement. 
 
 Lime consists of practically pure carbonate of lime with 
 less than lo % of impurities. 
 
 Hydraulic Lime has mixed with the hme from lo to 25 % 
 of silica, alumina and iron. 
 
 Cement contains: 
 
 Lime, - - - 55 to 65 % 
 Silica, - - - 18 to 24 ^ 
 Alumina and iron, - 8 to 14 % 
 
 These usually amount to 94 or 96 % of the whole. The 
 balance may be made up of magnesia, alkalies, and sulphuric 
 anhydride. These last are present in minute quantities, and 
 although they undoubtedly have some influence upon the 
 qualities of the cement, still this effect is verv slight. 
 
 Lime. — Lime is made by the simple calcination of more or 
 less pure carbonate of lime. This is found as limestone in all 
 parts of the world. The calcination is done usually in a kiln 
 
LIME BURNING. 
 
 5 
 
 of a form of construction shown in Fig. i. The fuel most 
 commonly used is wood, but either soft coal or coke may be 
 used. The method of charging the kiln is as follows: A 
 rough open arch is built of Umestone above the bottom of the 
 kiln. Upon this is placed a layer of fuel, then a layer of 
 Hmestone, a layer of fuel, and so on, alternately, to the top 
 of the kiln. The layers of fuel grow less as the top is 
 approached. The fire is started at the bottom, and the 
 temperature gradually increases. As the fuel is consumed 
 
 the limestone drops to- 
 wards the bottom, and 
 more fuel and limestone 
 is added at the top. As 
 rapidly as the Hmestone 
 becomes sufficiently 
 burned it is removed 
 from the bottom of the 
 kiln. It comes from the 
 kiln in hard, white, rock- 
 like pieces. 
 
 One peculiar- 
 ity of the freshly 
 burned lime is 
 the great avid- 
 ity it has for wa- 
 ter, and, when 
 it is exposed 
 to moisture, the 
 ^reat amount of 
 carbon it will 
 absorb. 
 
 Lime that has 
 not been expos- 
 ed to moisture, 
 and is in more 
 or less the same 
 condition in 
 
 Fig. I. 
 
6 
 
 PORTLAND CEMENT. 
 
 which it came from the kihi, is called J^fic/c/uue. When this 
 lime has been exposed to moisture in any shape, and allowed 
 to absorb as much as possible, it is called Slaked Lime. 
 
 All lime, before being used, must be slaked. This can be 
 done by Drowning, Sprinkling, or exposure to the air. 
 
 Drowning. — The quicklime is spread out in a water-tight 
 box and water added until it is completely covered. The 
 entire amount of water needed should be put in at once. 
 When the water is added the temperature rises, the mas? 
 effervesces, the quicklime increases rapidly in bulk, slowdy 
 disintegrates, and tinall}- falls to pieces in a fine white powder, 
 soluble in water. The impurities are separated from the lime. 
 If additional water is added after the process of slaking has 
 commenced the temperature is lowered and the slaking is not 
 done as thoroughly as it otherwise would be. After the 
 water has been added it is a good plan to cover the box so as 
 to retain the heat as much as possible. 
 
 The increase in bulk due to slaking is 200 or 300 9^ . The 
 lime, after the slaking is complete, is run off through an 
 opening in one end of the box. It is about the consistenc}' of 
 very thick cream. The opening is covered with a grating 
 or netting to prevent the passage of hard lumps, etc. The 
 slaked lime is run either into another box or into an excavation 
 in the ground, this second box or excavation being man}' times 
 laro-er than the slakin"" box. The slaked lime soon becomes 
 a stiff paste, and should be covered with sand or boards. It 
 should not be used for mortar for a number of da3's, usually 
 about 10, until it has become thoroughly cool. 
 
 Slaking by Sprinkling". — The quick lime is spread out in a 
 la3-er6or 8 inches in thickness, and thoroughly sprinkled with 
 water. It slowly disintegrates and falls into powder. There 
 is no great increase in the temperature, and no effervescence 
 takes place. One drawback to this method is the space and 
 time required, which are both much greater than is required 
 for JJroiL'j/i/iq, and there is no reliable data to show that the 
 lime thus slaked is in any way improved. 
 
 Air Slaking. — The lime is spread in layers 4 to 6 inches 
 
HYDRxVULIC LIME. 
 
 7 
 
 thick, and exposed to the air. It must be turned a number 
 of times in order to insure thorough exposure to the air. The 
 time and space required are both very great and the gain, if 
 any, small Thoroughly slaked lime paste can be put up in 
 air-tight casks and kept without deteriorating for almost any 
 leno:th of time. 
 
 HYDRAULIC LLME. 
 
 Hydraulic Lime is made by the simple calcination of lime- 
 stone that contains anywhere from lo to 25 per cent, of the 
 requisite impurities. The temperature required for calcination 
 is but slightly higher than that needed for the burning of quick- 
 lime. The material must be slaked the same as quicklime 
 before it can be used, and is reduced to powder in this way. 
 No grinding machinery for reduction is used. The slaking 
 is much slower than is the case with quicklime and as the 
 proportion of impurities increase it becomes slower and slower, 
 until at last a point is reached where the resulting substance 
 passes from hydraulic lime to natural cement. The reduction 
 must be done by grinding and the hydraulicity becomes a 
 prominent characteristic. 
 
 Cement.— Cement as used in an engineering sense means 
 such a combination of lime, silica, alumina and iron, that when 
 properly calcined, reduced to pow^der, and gauged with a 
 proper amount of water has the property of setting under 
 water and in places where it is not exposed to the action of the 
 air. It also has the property of setting when in contact with 
 the air. 
 
 For good results to obtain, the proportion of the requisite 
 constituents must be within certain narrow and well-defined 
 limits. These proportion have already been given. 
 
 The cements used in building construction can be divided 
 into two general classes. 
 
 NATURAL CEMENT. 
 ARTIFICL\L OR PORTLAND CEMENTS. 
 In what follows the term Portland cement means always 
 an artificial cement as distinguished from natural cement. 
 
8 
 
 PORTLAND CEMENT. 
 
 ^ Natural Cements.—In many parts of this country and 
 Europe, there have been found immense deposits of impure 
 hmestone, that contain with more or less accuracy the necessary 
 constituents for the making of cement. These constituents 
 have been mixed by nature and for cement making must be 
 used in the proportions found. 
 
 The actual proportions do not usually conform with that 
 required for the best cement, but the extreme cheapness at 
 which this cement can be put upon the market renders it a 
 most valuable addition to the materials of construction. The 
 natural cement does not in any way compete with the Portlands, 
 but it has done much to raise the standard of masonry 
 construction, from the fact, that its small cost allows of its use 
 in many places where lime was used before. 
 
 The difference between the natural and Portland cements 
 as to the raw materials used is this : 
 
 The desirable constituents in each are the same. 
 In making natural cements, some impure Hmestone that 
 contains as nearly as may be the correct proportions of lime, 
 silica, and alumina is used, and the value of the resulting 
 cement depends upon the correctness with which nature has 
 mixed these ingredients. It is found good, bad and indifferent. 
 
 With the raw material for Portland cement, however, nothing 
 is left to chance. It is known, within certain narrow limits, 
 what the constituents should be, and in what proportions they 
 should be present. This being known, such materials are 
 used as contain these constituents in a more or less pure 
 state, and then these comparatively pure raw materials are 
 mechanically mixed in the correct proportions. 
 
 The mere fact that the raw materials are nearly perfect 
 does not insure good cement, as the best of raw material may 
 be rendered useless by improper burning or grinding. But, 
 on the other hand, no good cement is possible unless the raw 
 materials are good. Of course any mechanical mixture of 
 lime, sihca, alumina, etc., within the hmits named, will give 
 good Portland cement if properly burned and ground. But 
 in selecting raw material there is one other, most important, 
 question that must be considered, viz., that of cost. 
 
PORTLAND CEMENT. ^ 
 
 In order to make a perfect mechanical mixture the materials 
 must be reduced to an impalpable powder. The harder the 
 materials, the more expensive this is; consequently, in selecting 
 the raw material, the question of the cost of reduction must 
 be considered. 
 
 The advantages of Portland cement over natural cement are 
 two, viz. : 
 
 ist. The Portland cement is much better -per sc. The 
 best natural cement never attains the hardness nor has the 
 strength or durability of the most ordinary Portland. 
 
 2nd. Where proper care is used, Portland cement of any 
 one brand possesses a uniformity of quality that can never be 
 attained in the making of natural cement. 
 
 Examine almost any stone quarry, and the impossibility of 
 obtaining a uniform quaHty of stone in any quantity wdll be 
 seen at once. The quality of the stone varies in different parts 
 of the quarry and in different layers of the stone, no tw^o 
 layers containing the same chemical constituents. As the 
 stone is used in the condition in which it comes from the 
 quarry, it will be seen that there will be an unavoidable 
 variation in the quaHty of the resulting cements. 
 
 With Portland cement this is different. The raw materials 
 are practically pure, and after experiments have given the 
 proportions of mixing and the subsequent methods of treat- 
 ment, there is no excuse for any irregularity in the results. 
 
 This uniformity in results is the one great point to be worked 
 for in cement making. It can only be accompHshed by the 
 exercise of the greatest care in the selection and treatment 
 of the raw materials. In the process of making there are 
 some radical differences between natural and Portland cements. 
 In the calcination the natural cements require a temperature 
 but little above that required for lime burning, while the 
 Portland cements require a temperature just short of that 
 required for vitrification. The mixing, grinding, etc., all 
 increase the cost of the Portland until at last the finished 
 product brings about $3.00 per barrel on the market, wdiile 
 the natural cement sells for 50 cents. 
 
10 
 
 PORTLAND CEMENT. 
 
 True economy in the choice of cements consists in using 
 the one best adapted for the work in hand. When the work 
 is such as to justify the increased expense on account of 
 required durability or strength, then the best Portlands should 
 be used. But on less important work or masonry of a cheaper 
 character, the natural cements should be used. Nothing has 
 done more to improve the character of all masonr}' work during 
 the last twenty-hve 3'ears, than the cheapness and excellency 
 of these liffht-burned natural cements. 
 
CHAPTER II. 
 
 HISTORICAL DATA. 
 
 The use of some cementing substance for building purposes 
 runs back into the darkness of pre-historic times. We have 
 the record of no age of the world in which some form of 
 cement has not been used. At the very earliest periods of 
 histor\', not only was cement in some form used, but the 
 proper methods of manipulation were well understood. One 
 of the earliest w^riters upon engineering construction, whose 
 writings are now available, w^as Marcus Vitruvius PoUio who, 
 as an architect, engineer, and author, worked under the patron- 
 age of the Roman Emperor Augustus. No better idea can 
 be obtained of the advanced knowledge upon limes and 
 cements at that age of the world than by making some 
 abstracts from the Works of Vitruvius. We wash to call par- 
 ticular attention to the line of reasoning b}' which the action 
 of lime, etc., are accounted for. 
 
 " Having treated of the different sorts of sand, we proceed 
 ■'•to an explanation of the nature of lime — wdiich is burned 
 '•from either white stone or flint. That which is of a close 
 ■'•and hard texture is better for building walls, and that which 
 ■'■is more porous is better for plastering when slaked for 
 "making mortar, if pit sand be used, three parts of sand to one 
 ■'•of Hme. If river or sea sand, then two parts of sand to one 
 "of lime *If to river or sea sand, potsherds 
 
 ■*The gain in quality due to mixing finely ground harned clay ^\ iti"i lime 
 appears to have been well known. 
 
12 
 
 PORTLAND CEMENT. 
 
 '•ground and passed through a sieve, in the proportion of one- 
 • third part, be added the mortar will be better for use. The 
 "cause of the mass becoming solid, when sand and water are 
 '•added to the lime appears to be, that stones, like other 
 '•bodies, are a compound of elements: those which contain a 
 '•large quantity of air being soft, those which have a greater 
 "proportion of water being tough, of earth hard, of fire 
 "brittle. For stones which, when burnt, would make excel- 
 "lent lime, if pounded and mixed with sand without burning, 
 '•would neither bind the work, nor set hard; but havino- 
 "passed through the kiln, and having lost the property of their 
 " former tenacity by the action of intense heat, their adhesive- 
 '•ness being exhausted, the pores are left open and inactive. 
 '•The moisture and air which were in the body of the stone, 
 "having therefore been extracted and exhausted, the heat 
 " being partially retained, when the substance is immersed in 
 " water before the heat can be dissipated, it acquires strength 
 " by the water rushing into all its pores, effervesces, and at 
 "last the' heat is excluded. Hence limestone, previous to its 
 " burning, is much heavier than it is after it has passed 
 "through the kiln; for, though equal in bulk, it is known, by 
 '■the abstraction of the moisture it previously contained, to 
 " lose one-third of its weight by the process. The pores of 
 '• limestone, being thus opened, it more easily takes up the 
 "sand mixed with it, and adheres thereto; and hence in dry- 
 "ing, binds the stones together, by which sound work is 
 " obtained." 
 
 The use of Pozzolana was well understood and its cement- 
 ing qualities w^re accounted for in the same manner as that 
 first given for limes. Although we have no written records, 
 still we know that lime was used by the Egyptians, thousands 
 of years before the Christian Era, and we have very fair 
 evidence that the)', also, understood the mixing of clay with 
 the lime and thus making a crude form of Portland cement. 
 
 But leaving the ages of antiquarian uncertainty and coming 
 down to the times of modern civilization, we find that as late 
 as 1757 nothing reliable was known upon the manufacture 
 
HISTORICAL DATA. 
 
 13 
 
 and use of hydraulic cements. Up to that time the purer 
 the limestone the better the lime for constructive purposes. 
 Neither hydraulic lime nor cement was known. In i757 
 Smeaton, the engineer in charge of the rebuilding of the 
 Eddystone Lighthouse, commenced his experiments upon the 
 various obtainable materials for making mortars. The structure 
 was of such vast importance and the strength required so 
 great that Smeaton, not satisfied with the action of the mortars 
 in general use, commenced experiments to determine the 
 constituents necessary in a cement that would set under water 
 and under salt water.* 
 
 Smeaton was the first to break down the tradition that the 
 purest and hardest limestone was the best, at least for hydraulic 
 purposes, and the first to prove that a proper mixture of 
 carbonate of lime and clay was what gave the best results. 
 He was the first to discover that this calcined mixture of clay 
 and carbonate of lime was the real cause of hj^draulicity. 
 The Eddystone Lighthouse stands to-day, not only as a guide 
 to "ships that pass in the night," but also as a monument to 
 mark the starting point in all that we know concerning 
 hydraulic cements. 
 
 From 1757 until about 1824 very little advance was made 
 in cement making and no advance was made in our knowledge 
 of cements beyond that left by Smeaton. A patent dated 
 December 15th, 1824, was granted to John Aspdin, of Leeds, 
 bricklayer, for the manufacture of Portland cement. Edgar 
 Dobbs, of Southwark, was granted a patent for a Portland 
 cement mixture in 1810. Maurice St. Leger, of Camberwell, 
 was granted a similar patent in May, 1818. These last two 
 patents expressly state in their specifications that the material 
 was not to be subjected to sufficient heat for vitrification. All 
 experimenters in Europe appeared satisfied with producing 
 hydraulic lime, with the exception of Aspdin. He soon 
 discovered the advantages of incipient vitrification and produced 
 Portland cement. C ement works were slowly established, 
 
 *The student should by all means read Smiles' Life of John Smeaton. 
 
14 PORTLAND CEMENT. 
 
 but owing to the uncertainty in the character of the product^ 
 and the great fight against the new cement, by the well- 
 established makers of Roman cement* the industry languished 
 and barely lived for many years. 
 
 The manufacture of Portland cement in Germany was 
 commenced in 1852, near Stettin. In 1855 the first Stettin 
 Portland cement works were erected; in 1877 there were, in 
 Germany, thirty large cement works. The German works, 
 at first, copied the English methods, but now their methods 
 are widely different. In 1892, in Germany, there were sixty- 
 two large Portland cement factories and the production for 
 that year was ten million six hundred thousand barrels. 
 Besides these there are also ninety-six smaller factories, the 
 production of which is not given. -j- 
 
 The largest Portland cement works in France are at 
 Boulogne. The product equals in character either the Ger- 
 man or the English. 
 
 The first Portland cement works in this country were 
 established in 1875. manufacture has been carried on to 
 
 the greatest extent near AUentown and Egypt, Pennsylvania. 
 There are factories at Belief ontaine, Ohio; South Bend, Indi- 
 ana; Warner's, New York; and the latest and one of the best 
 at Yankton, South Dakota. 
 
 Saylor's Portland cement, made at AUentown, has the best 
 established reputation and has been much used. The Buckeye, 
 of Bellefontaine, Ohio, has given as high results in laboratory 
 tests as any cement tested by the author. The Western 
 Portland cement, of Yankton, South Dakota, has only been 
 on the market a few years, but thus far it has given results, 
 both in the laboratory and in actual work, that have never 
 been excelled by either German or English Portland cements. 
 In the United States by far the greater part of the hydraulic 
 cement used is of the light-burned natural cement. The 
 Rosendale type, made in Ulster county. New York, supplies 
 
 * Roman cement is similar'to our light-burned natural cement. 
 •[From Trans. Am. Soc. C. E., Vol. XXX., No. i, Gary on Cement. 
 
PRODUCTION. 
 
 15 
 
 nearly one-half the demand for this cement. Louisville, Ken- 
 tucky, Utica, lUinois, and Milwaukee, Wisconsin, are the cen- 
 ters of production of this type of cement in the west. 
 
 In 1889 there was imported into this country 650,000 
 barrels of Portland cement and 150,000 barrels were made 
 here. The same year 4,200,000 barrels of natural cement 
 was made in this country. 
 
CHAPTER III. 
 
 MANUFACTURE OF PORTLAND CEMENT. 
 
 One of the first things to be decided upon in establishing a 
 Portland cement manufactory is the selection of the raw 
 materials necessary. The constituents necessary in the cement 
 making materials are lime, silica, alumina, and iron, and for 
 the making of the cement the additional item of fuel. 
 
 The necessar}' carbonate of lime can be found in the shape 
 of limestone in almost any locality. But as this must be 
 reduced to a fine powder during the first stage of cement 
 making, the ordinary limestone is too hard to make this 
 possible with the necessary economy. Therefore the softest 
 and purest limestones are usually the only ones used. This 
 is found in the shape of chalk and in all degrees of hardness 
 and purity. The softer it is the more cheaply it can be 
 worked, and the purer it is the more uniformity there is in 
 the resulting cement. There are immense deposits of this 
 material in this country, and consequentl}^ no necessity as yet 
 of using any of the harder and more expensive varieties. 
 
 Those deposits that have been worked with good results 
 are: One running south through New York state, and 
 visible at numerous points through Pennsylvania and Virginia. 
 One most excellent deposit, that is worked at Bellefontaine, 
 Ohio, from which the Buckeye Portland cement is made. 
 One that is worked at South Bend, Indiana; and in the middle 
 west, a deposit of almost pure chalk, that, starting north of 
 the Canada line, runs down the valle}' of the Missouri on the 
 western border of Iowa, and keeping south, can be traced 
 
RAW MATERIALS. 
 
 into Louisiana and Texas. From that point to about half 
 way to the City of Mexico the author has no knowledge 
 of this deposit, but, beginning at a point about one hundred 
 miles north of Leon, Mexico, there is a good supply of chalk 
 to the Isthmus of Panama. On the Pacific coast there is an 
 almost unlimited amount of good material, that as yet is 
 very httle worked. The Missouri belt of chalk is worked at 
 Yankton, South Dakota, and no better cement making materi- 
 als can be found than at this point, both from a chemical and 
 economical stand-point. 
 
 The carbonate of lime, then, should be the softest and 
 purest that can be found, other things being equal. 
 
 The silica, alumina and iron are usually found in the shape of 
 clay. The clay is usually dark blue and to a certain extent 
 indurated. It is usually of such a softness that it may be 
 removed with a pick, but of such a hardness that the use of 
 some low grade explosive is often economical. About sixty 
 per cent, of the clay should be silica. But it must be remem- 
 bered in selecting clay that this silica must be in a state of 
 chemical combination and not in the shape of sand. This is a 
 point that has sometimes been neglected with deleterious 
 results. The chemist in analyzing clay for cement making 
 must remember that the sand in the clay must not enter into 
 his estimate as to the amount of silica available for cement 
 making, and clay containing as little sand as possible should 
 be used. 
 
 If in the preparation of the raw materials the treatment of 
 tlie clay was such that any sand present was reduced to an 
 impalpable powder, then the sand might act in the required 
 way as silica, but as no such methods are used, the sand 
 becomes a distinct foreign element that is detrimental to the 
 manufacture of good cement. Fortunately for cement makers 
 in this country, the chalk and clay necessary for cement 
 making are often or usually found in contiguous layers. Such 
 an arrangement does much to lessen the cost of production. 
 
 The cost of the fuel that is to be used for the burning of 
 the cement is the one other item of raw material that must be 
 
i8 
 
 PORTLAND CEMENT. 
 
 considered. Therefore as to the quality of the cement, the 
 chalk and clay must be very carefull}' tested, and not only 
 tested in the beginning as to their general suitability, but 
 tested da\' by day as the work goes on. Each separate 
 layer in the quarries should be tested and the correct propor- 
 tions to be used determined. 
 
 Given suitable cla}' and chalk, with correct subsequent 
 treatment, and the cement will be good and uniform. The 
 cost of producing the cement will depend upon the character 
 of the clay and chalk and the cost of the fuel. 
 
 The value of the cement to the makers will be the price at 
 which it can be sold and this will depend upon the cOst of 
 transportation to an available market. These various questions- 
 will decide within certain limits, the location of the works. 
 
 As to the relative financial importance of the raw material 
 and fuel. The carbonate of lime is usually about three times, 
 that of the clay by w^eight. The amount of fuel necessar\- 
 runs from lO to 20 per cent, of the weight of the raw material. 
 The resvilting cement is about 60 per cent, of the weight of 
 the chalk and clay. The weight of the fuel entireh' disappears^ 
 during the process of manufacture. 
 
 From this data we can make a fair estimate of the best 
 location for the works, when the raw materials are separated.. 
 
 Cheap transportation to market is necessary for the ultimate 
 success of the enterprise and whenever possible the works 
 should be so located as to make water transportation available. 
 
 There is one more point to be considered in connection with 
 the reduction of the raw material, viz.: the material used 
 should ahvays be that, that can be reduced to a powder with 
 the least expense — provided the other qualities are equaL 
 Chalk and clay usually fulfill this condition of cheap reduction 
 and are therefore most generally used. 15ut when such 
 material cannot be obtained and harder working material must 
 be used, such as crystalline limestone or slate, it is still possible 
 to reduce such material to powder by means of modern grind- 
 ing machiner\\ The increased cost due to this is not so great 
 but that a large profit remains when the product sells at the 
 ordinary market rates. 
 
DESIGN OF WORKS. 19 
 
 In the early days of the manufacture of Portland cement, 
 some of the makers in England were compelled to use a lime- 
 stone in the place of chalk, and in order to use this they 
 resorted to what is known as the Double Calcination Process, 
 That is, the limestone was first burned to lime, the lime 
 slaked, and then this paste or powder mixed with the cla}' arrd 
 the whole burned to a clinker. This process is very little 
 practiced now, owing to the great improvements in the modern 
 machinery of reduction. Whatever process is used the result 
 should be that the raw materials are reduced to exceedingly 
 fine powder and thorouglily mixed. 
 
 After the location of the works has been decided upon, the 
 next thing is the designing of the works themselves. This, 
 should be done by an engineer who is perfectly familiar with 
 the entire process of cement making. The w^iole should be 
 carefully studied out and reduced to paper before an}- con- 
 struction is begun. As much of , the process of manufacture 
 as possible should be done by gra\'ity. If the raw material is 
 so located as to permit, the works should be located below the 
 level of the quarries. 
 
 The quarries are on the highest level. The preparation of 
 the raw material next. Dry ing floors, plates, and kilns next. 
 Reduction of clinker next. Storage, packing, and shipping* 
 on the lowest level. It is rare that such an ideal arrangement 
 as this can be made, but such an arrangement should be 
 borne in mind, and other things being equal, the works made 
 to conform to it as nearly as possible. 
 
 •Reduction of Raw Material. — The method to be employed 
 in the reduction of Razo ^TaleriaL for the making of Portland 
 cement, depends entireh' upon the character of the material 
 used. 
 
 The object of any method used is, however, the reduction 
 of the Raiv Material to an impalpable powder, and the 
 intimate mechanical mixture of these powders in the correct 
 proportions. The processes in general use can be divided 
 into two classes. 
 
 The Wet Process and the Dry Process. — In nearly every 
 
20 
 
 PORTLAND CEMENT. 
 
 case, however, some modification or combination of these 
 processes is used. 
 
 THE "WET PROCESS." 
 
 The raw materials, to which this process is suited, consist 
 of some form of clay and chalk of such a character as to be 
 nearly soluble in water. The material is taken from the 
 quarries and by means of crushers or grinders reduced to a 
 coarse powder. 
 
 in some cases where the material is easily acted upon by 
 water, it is not passed through any crusher or grinder, but 
 simply used in the form in which it comes from the quarry. 
 Whatever preparatory method may be used, however, the 
 material when in proper condition is put into water. 
 
 As this is the point where the tw^o materials clay and chalk 
 are mixed, much care must be exercised. 
 
 A careful chemical examination of each material must be 
 made, the correct proportions decided upon, and then care 
 must be taken that these proportions are used. The materials, 
 in the correct proportions, are then put into some variety of 
 wash mill or mixing machine, with a great quantity of water. 
 
 In some cases each ingredient is put into the mill separatel}' 
 and the entire mixing done there. While in there the dry raw 
 materials are more or less mixed before any water is added. 
 
 A simple method for partial mixing is to spread the material 
 in layers upon a floor, the two materials alternating and the 
 thickness of the layers being proportional to the relative 
 amount of each material needed. There are six or eight, of 
 these layers in a pile. The material is removed from the pile 
 to the mixing machines by means of barrows, and the man in 
 loading the barrows, cuts through all the layers with his 
 spade. In this way the material becomes quite thoroughlv 
 mixed before the mixing machine is reached. 
 
 The V/ash Mill. — T he construction of the wash mill will be 
 understood by an examination of Figs. 2 and 3.* There is a 
 
 *These figures and the 
 ^Manufacture and Uses," 
 
 descriptions are talven from " Portland 
 hy Henry Reed, pp. 20S and 20<j. 
 
 Cement : 
 
 its 
 
WASH MILLS. 
 
 21 
 
 great variety of these mills. The earliest form is shown in 
 Fig. 2. There is a circular trough of masonry about 20 feet 
 in diameter and 6 or 7 feet deep. In the center of this is a 
 masonry pier, which has on it a fixed sole-plate and socket,. 
 
 Fig. 2. 
 
 that receives the toe of a vertical shaft. From this shaft is 
 supported a circular rim of wood that has a rack on the under 
 side. This rack is driven by a pinion, as shown in Fig. 2. 
 There are a number of arms extending from the center to 
 
 Fig. 3. 
 
22 
 
 PORTLAND CKMKNT. 
 
 this wooden wheel, and from these arms are suspended, by 
 chains, harrows. The cement making material is put into 
 the trough with a great surplus of water, and a rotary motion 
 given to the wooden wheel. The harrows are thus dragged 
 through the material and water, and the material thoroughly 
 amalgamated. 
 
 A more modern form is shown in Fig. 3, where a bar is 
 substituted for the horizontal wheel. This bar being driven by 
 gearing at the center, long knives, bolted firmly to this bar, 
 are substituted for the harrows. 
 
 A mill of the size iriven is driven at about 20 revolutions 
 per minute. The correct speed in any case, however, must 
 be determined by the character of the material and the rela- 
 tive amount of water used. 
 
 In some manufactories they make a practice of double- 
 working the material. That is, it is thoroughly worked in 
 one mill, and then transferred to another, more water added, 
 and the amalgamation completed. In modern practice this is 
 not usually necessary. The process of working the material 
 in the wash mill is as follows: In the first place, only such 
 material can be used in this process as will become thoroughly 
 disintegrated hy water and can be held in suspension by it in 
 a finely divided state. The material is reduced by crushers 
 and triturators to the necessar}' degree of fineness and put 
 into the wash mill in the correct proportions. I'his correctness 
 of proportions is one of great importance, and one to which too 
 much attention cannot be gi\'en. Without this correctness the 
 resultin(»: cement will be inferior or useless. Sufficient water 
 is admitted to the trough to hold the entire mass in suspension 
 with ease. For the proper mixing of the materials too much 
 water cannot be used. The only limit is the amount of water 
 supply and the length of time necessary for get^-ing rid of the 
 water in '-backs," or settling reservoirs. For economy, the 
 least amount of water that will thoroughl}- accomplish the 
 disintegration and amalgamation should be used, and this can 
 only be decided upon by experiment. Different materials 
 require different amounts of water. 
 
REMOVAL OF SAND. 
 
 23 
 
 Wlien the water has been admitted, the wash mill is started 
 and the whole mass thoroughly agitated. 
 
 The raw material must become so finely divided that it is 
 held in suspension and almost in solution by the water. The 
 amount of water is increased and an overflow^ opened. This 
 overflow allows the liquid to escape into a sluice. It flows 
 along this sluice until it empties into the - backs." 
 
 In order to guard as much as possible against the introduction 
 of sand into the ultimate mixture, the following arran^IJement 
 is usually made : 
 
 At different points in the sluices or chutes, there are 
 depressions or cavities, and in these cavities the movement of 
 the liquid is much less rapid than in the chute. The sand 
 having a much greater specific gra^■ity than the finely divided 
 •clay and chalk, settles to the bottom of these cavities and is 
 retained there. In this manner a portion of any sand present 
 is prevented from passing into the ••backs." 
 
 The best method however, to prevent sand from being 
 present in the cement mixture is to use a clay that contains as 
 little sand in it as possible. No matter what precautions are 
 taken to prevent it. some sand will get into the •• backs" if 
 there is anv in the cla}' used, and the more sand in the cement 
 mixture the poorer will be the quality of the cement produced. 
 
 The principle upon which the wash mill is worked is, that 
 the materials are reduced to such a finely divided state, in such 
 an abundance of water, that they are practically in solution 
 and when the}- have been sutflciently agitated to bring about 
 this result, the liquid is run off into '•backs" or settling 
 reservoirs and the cement material allowed to settle. 
 
 There are some disadvantages connected with this process. 
 The clay and chalk being of different specific gravities there 
 is a possible danger that however well they may have been 
 mixed in the wash mill, they may to some extent become 
 separated in passing through the sluices. Even if they reach 
 the "backs" in the correct proportions, there is danger that 
 they may settle unevenly and in layers. 
 
 The only way in which this can be guarded against is a 
 
\ 
 
 24 PORTLAND CPLMENT. 
 
 careful testing of the material from time to time, as the work 
 goes on. The '"backs" require a great deal of floor space 
 and in some localities would either be the cause of great 
 expense or prohibit the use of the "Wet Process" in its 
 entirety. 
 
 When such is the case, however, there are modifications of 
 this process that eliminate many of its disadvantages, without 
 destroying its utility. 
 
 Another drawback to the continued use of the simple •• Wet 
 Process" is the length of time required before the raw material 
 is ready for the kilns. Often two or three months are required 
 to get rid of the water in the backs. The material must 
 settle and the water be carefully drawn off from the top. 
 
 When the material has finally settled and as much water 
 has been drawn off as possible, the remainder of the water is 
 removed by evaporation, until the '-slurry" has stiffened to a 
 degree that it can be handled well with a shovel. It is then 
 broken up in pieces and taken to the drying floors, where it 
 is thoroughly dried by artificial heat, usually the escaping 
 heat from the kilns. All of this takes many weeks and neces- 
 sitates the carrying of an immense amount of raw material. 
 
 In the "Dry Process" this loss of time is eliminated and in 
 some works the amount of prepared raw material on hand is 
 never more than one day's supply. This, of course, is 
 attained at an increase in the cost of preparation. Modifica- 
 tions and combinations of the "Wet" and '-Dry" systems are 
 used in order to modify the evils of each. 
 
 THE "DRV PROCESS." 
 
 In 1852 the Germans began to turn their attention to the 
 manufacture of the " Patent English Portland Cement." They 
 had no raw materials available, similar to the chalk and clay. 
 But there was an abundance of suitable limestone and highly 
 indurated shale. The " Wet Process," or reduction by water, 
 was entirely out of the question, and this fact led to the 
 inauguration of the so-called " Dry Process." The raw ma- 
 terials are reduced by purely mechanical means. Taken 
 
••DRY PROCESS." 25-. 
 
 from the quarries, they are passed through ordinary stone 
 crushers and reduced to the size of macadam, 2 or 2)4 inches, 
 as a maximum. This crushed material is then passed through 
 another form of crusher, called a triturator, that reduces it to 
 the size of coffee beans. It is then reduced to a flour bv 
 means of some form of grinding machine. These grinding- 
 machines are of three general t3'pes, the vertical runner, the 
 ball pulverizers and the horizontal millstones. 
 
 Up to the point of being ground the two materials are 
 usually reduced separatel}'. At this point they can be eco- 
 nomically mixed in the correct proportions and ground to- 
 gether. In this manner an accurate and thorough mixing of 
 the two ingredients is accomplished. 
 
 The dry powder is now slightly dampened and run through 
 a brick machine. The bricks thus formed are dried thor- 
 oughly by artificial heat, and are ready for burning. But a 
 few hours' time is necessary to dry the bricks, and but 
 comparatively little storage space is required. 
 
 The material having been made into bricks under consider- 
 able pressure, the mass is compact and dense, and burns with 
 much less uniformity than the dried slurry from the backs.'' 
 
 The raw material is taken from the quarry, crushed, ground, 
 made into bricks, dried, burned, ground to cement, and sold, 
 in much less time than is occupied b}- the material in the 
 ••backs" by the Wet Process." 
 
 All of this, however, is accomplished at a great increase in 
 cost, and there are man}/ points in this process that need 
 careful watching. 
 
 In order that the result may be of the best, most accurate 
 and thorough grinding is necessary. This is the point at 
 which the extra expense is incurred. It is an absolute neces- 
 sity that the materials should be reduced to impalpability and 
 unless it is done in the grinding it is not done at all. With 
 modern machinery and the experience of the last fifty 3-ears 
 there are but very few^ cement works that use either the 
 ''Dry" or '-Wet Process," but usually some modification 
 suited to the special character of the raw material. 
 
26 
 
 PORTLAND C;EMENT. 
 
 Semi-Wet. — In the preparation of the raw material l)y 
 means of the wash mill the following modification of the 
 "Wet Process" is often used. 
 
 The material in the mill is mixed with much less water 
 than in the --wet." The mixing is done as thoroughly as 
 possible and then the slurry removed from the mill, not by an 
 overflow, but by conduit near the bottom. This conduit 
 conducts the slurr\- to two millstones between which it is 
 thoroughlv ground. The advantage of this process is that 
 there is much less water in the prepared slurry and consequently 
 less time is required for drying. 
 
 Another modification that is used in this countr}' is as 
 follows : 
 
 The material is quarried "and ci-ushed to a coarse powder. 
 It is then thrown into a large mixing machine with a great 
 surplus of water. The machine is started and the contents 
 tlioroughl}' amalgamated. The whole mass is then lifted to a 
 higher level by pumps and run off through sluices to the 
 drying floor. 
 
 Mixing Machine. — "l~he mixing machine consists of a boiler 
 plate cylinder about 15 feet in diameter and 15 feet in length. 
 Inside of this is a rotating mixer driven by steam or water 
 power. Cement works of any size will need a number of 
 these machines. The slurry is run out over the drying floors 
 in layers about 4 inches thick. When sufficiently hardened 
 it is cut into bricks about i foot square. This is done either 
 by a man using a specially formed spade or b}^ means of a 
 harrow with cutting blades, that is drawn across the slurry in 
 two directions, at right angles with each other. When these 
 bricks have sufficiently dried they are turned up on edge and 
 finally taken to the drying rooms and the drying completed 
 by artificial heat. They are then either placed in the kiln for 
 burning or stored away for future use. 
 
 Pug-Mill. — The material may be reduced and mixed with 
 just sufficient water to make stiff mud. It is then driven 
 through a pug-mill, which completes the mixing, and is cut 
 into bricks as it issues from the pug-mill. These bricks 
 require but little time to dry. 
 
CEMENT BURNING. 
 
 27 
 
 These are only a few of the most important methods used 
 for the preparation of the raw material. All the methods in 
 actual use are some modification or combination of the "Wet" 
 and "Dry Process." The method that should be used at any 
 one locality must depend upon the character of the material, 
 its resistance to reduction, the water supply, the available 
 ground space, and the question of the value of time. 
 
 These questions can only be answered correctly by an 
 expert in cement making, and such should he employed in 
 designing the works. The result that must be attained, in 
 order to produce good cement, is that the raw material shall 
 be reduced to minute fineness and that the mechanical mixture 
 ■of the different raw materials used must be practically perfect. 
 
 Burning Cement. — The raw material must be properly 
 reduced, proportioned, and intimately mived. When this is 
 done it is made into bricks or blocks, dried, and is then ready 
 for the burning. 
 
 Upon the well doing of this depends the quality of the 
 cement, as well as upon an\- of the previous steps in cement 
 making, but no more. Unless the preparation of tlie raw^ 
 material has been properly attended to, no amount of correct 
 burning can make good cement, and, further, no amount of 
 good burning can make good cement unless the treatment of 
 •the cement after burning is properly done. 
 
 The accurate performance of each of the three stages is 
 necessary for the production of good cement. 
 
 There are three well-defined periods in cement burning: 
 
 h'irst, the drixdng off of all the \vater that remains in the 
 slurry. 
 
 Second, the driving off of the carbonic acid. 
 Third, the fusing together of the silica, lime, alumina, and 
 iron. 
 
 These different periods occur in the order given here, and 
 •each requires a different temperature for its proper accoiu- 
 plishment. 
 
 Before the making of Portland cement, all lime and cement 
 burners had lived in constant dread of over-burning their 
 
28 
 
 PORTLAND CEMENT. 
 
 material, and even after a good idea had been formed of the 
 constituents necessary for Portland cement, still it was many 
 years before any one realized what an increase in temperature 
 was needed to give the best result in the cement. 
 
 For 'the making of the best possible cement from given 
 materials, the materials must be burned just as near to the 
 point of actual vitrifaction as is possible without reaching it. 
 If this point is passed, the cement is over-burned, and the 
 amount it deteriorates on account of this depends upon how 
 much it is over-burnecl. 
 
 The particles that are perfectly vitrified have no more 
 cementitious properties than so much powdered glass. 
 
 There is this trouble about over-burned cement. Its color 
 is good, possibly a trifle lighter, and its weight is increased, 
 therefore it cannot always be distinguished by the e3'e from 
 cement that is very good. For many years an increase in 
 \N'eight was looked upon as an improvement in the quality of 
 the cement. Up to a certain point this was true, but beyond 
 this point it was not. The more cement is burned, the more 
 it w^eighs, and as distinguished from the light-burned anv 
 increase in weight marks an improvement in the quality, but 
 when the point of vitrifaction is passed, the weight still 
 increases, while the quality deteriorates. Therefore, in any 
 specificatio,ns where the weight is to be specified, care must 
 be taken that the required weight is not reached by over- 
 burn in <»•. 
 
 The color of Portland cement is a light bluish gray, and it 
 can be readily distinguished from the ^^arious colors of the 
 light-burned natural cements. 
 
 The color of the different natural cements runs from a light 
 yellow through all shades of yellow and chocolate browns to 
 almost black. The color depending upon the material from 
 which it is made. 
 
 Kilns. — There are two types of kiln in more or less general 
 use and a great number of patent kilns that are not suited for 
 general use. A few of these are suited for use under special 
 conditions, and many, as far as the author can find out, are of 
 no use whatever. 
 
DOME KILN. 
 
 29 
 
 Dome Kiln. — The earliest type of kiln used for cement 
 burning was what is known as the dome kiln. The general 
 form is shown in Fig. 4. This form of kiln is still in more 
 general use than any other. 
 
 The relative height and width of these kilns have been 
 changed and rechanged many times during the last sixty 
 years of Portland cement burning and possibly some of the 
 details changed, but in all of its essential features it remains 
 unchanged. The process of using this kihi is as follows: 
 
 ^rhe kiln being cold, 
 kindling and light fuel 
 are put in on the grate 
 bars. On top of this a 
 heavy layer of coal, and 
 then a layer of slurry, a 
 layer of coal, a layer of 
 slurry, etc., until a suffi- 
 cient amount of material 
 is in the kiln. The fire 
 is started at the bottom 
 and burns up- 
 ward, slowly, 
 gradually heat- 
 infj; the whole 
 mass. The kiln 
 warms up, and 
 becomes hot. 
 The lirst action 
 of the heat is to 
 drive any re- 
 maining moist- 
 ure fro m the 
 slurry. A mod- 
 erate tempera- 
 ture only is 
 needed for this. 
 As more fuel 
 
30 
 
 PORTLiVND CEMENU'. 
 
 ignites and the tire approaches the slurry, tlie temperature 
 increases until it begins to act upon tlie carbonic acid. During 
 tlie first stage there is a visible vapor being driven off. This, 
 ceases when the mass has become thorougiilv heated to a 
 good red heat, and the carbonic acid begins to be driven off. 
 The amount of carbonic acid that must he driven off is enor- 
 mous, amounting often to 30 or 35 per cent, of tlie Aveight of 
 the cement material. When this has been accomplished and 
 the whole mass raised to a white heat, the third and last stage 
 begins, that is the fusing together of the cement making 
 materials. At tliis point is reached the uncertain Lv of cement 
 burning. 
 
 The product may be under-burned, over-burned or correcth' 
 burned, and which of these results is attained depends upon 
 the proportion ment of fuel and slurry. The relative amounts 
 of each must be decided upon before the tiring begins, and 
 when once decided upon cannot be in any way changed. As 
 the fuel is consumed the clinker sinks to the bottom and is 
 removed, fresh fuel and slurry being added at the top. In 
 this wa\' the kihi can be made more or less continuous. 
 Under other circumstances the kiln is charged and the entire 
 mass burned, then the kiln allowed to cool and the clinker 
 -removed. The disadvantages connected with the kiln are as 
 follows : 
 
 The relative amount of fuel must be decided upon before 
 the burning commences and once the kiln is started the 
 amount of fuel cannot be in any way changed. The relative 
 amount of fuel necessary depends upon the character of the 
 fuel, the character of the slurry, the shape and height of the 
 kiln, the temperature, the humidity of the atmosphere, and 
 the direction and force of the wind. As will be seen many 
 of these conditions are entirely be3'ond the control of the 
 cement maker and are conditions that are liable to change at 
 any time during the burning. 
 
 There is an enormous w^aste of fuel that goes off with the 
 carbonic acid. 
 
 jNIany attempts have been made to utilize this wasted heat 
 
HOFFMAN RING KILN. 
 
 31 
 
 by heatinu; with it, the drying rooms, etc. Some of these 
 appHcations have been more or less successful, and some have 
 not. The one thing that should never be lost sight of in all 
 of these schemes for utilizing this waste heat, is that the main 
 object of the kiln and the fuel it consumes is to burn the 
 cement, and not to heat the dr3^ing floors, and care must be. 
 taken that, in thus attempting to utilize the waste heat, the 
 work of the kiln as a kiln is not lowered. 
 
 The clinker as it comes from the kiln is sorted; the thor- 
 oughlv well-burned taken to the crusher and grinders, and 
 the under-burned put to one side to be reburned. This 
 re-burning is sometimes done in the same kiln, but at some 
 works they have smaller kilns for this purpose. None but 
 properly burned clinker should be passed on to the grinders. 
 
 We have thus draw n attention to some of the salient fea- 
 tures of the dome kiln and some of the principal disadvan- 
 tages connected with its use. and these disadvantages become 
 more emphasized when it is used as an intermittent kiln, that 
 is, when charged cold, heated up, and then allowed to cool off 
 after each charge is burned. When this is done, there is the 
 immense loss of heat in reheating the kiln each time, together 
 with the unavoidable loss of time. 
 
 In addition to all of this, there is the expense of repairing 
 the lining of the kiln after each charge, the heating and 
 cooling causing the lining to crack and break. 
 
 The Hoffman Ring Kiln. — Soon after the manufacture of 
 Portland cement began in Germany the manifold disadvan- 
 tages attending the use of the dome kiln became evident, 
 and an effort was made at once to substitute some more 
 economical and improved t\'pe. This soon led to the inven- 
 tion of the Hoffman Ring Kiln, shown in section Fig. 5 and 
 in plan Fig. 6. The essential features of this type of kiln, as 
 can be seen from the ligures, are as follows: 
 
 There is one large central chinmey, and around this is built 
 a series of compartments. These compartments are connected 
 directly with a smoke chamber by means of flues. This smoke 
 chamber is annular in shape, and surrounds the base of the 
 
HOFFMAN RINCx KILN. , 33 
 
 chimney. The draft through these flues can be regulated by 
 means of valvular dampers that are worked from the outside. 
 The outside compartments are separated from each other by 
 means of sliding doors or partitions that may be opened or 
 shut at will. Each of these compartments has an opening on 
 the outside, through which the cement material is introduced, 
 and through which the process of burning can be watched. 
 The fuel is put in through openings at the top. The arrange- 
 ments are perfect for the admission of the requisite amount of 
 fresh air, in order to complete the combustion. 
 
 The kiln shown, Figs. 5 and 6, is a '-'down draft" kiln. 
 That is, the fire is at the top and the heat is drawn down 
 through the cement material and passes out through the 
 flue at the bottom. In some of these "ring" kilns the fire is 
 on the outside and near the bottom, and the opening of the 
 flue near the top. This, however, is a mere detail. In what- 
 ever part of the compartment the fire and fuel are placed, the 
 object is the same, namely: that the heat shall be distributed 
 with uniformity through the entire compartment, and that the 
 combustion shall be as nearly perfect as possible. 
 
 The method of operating the kiln is as follows: the mate- 
 rial to be burned is placed in the compartments. The fire 
 in one compartment is started, the heat carried directly to 
 the chimney. The compartment begins to warm up. All 
 the moisture is driven from the cement making material. 
 The heat increases, and the carbonic acid is acted upon. 
 Now the heat, instead of being carried directly to the chim- 
 ney, is turned to one side and passed through one, two, or 
 three adjacent chambers. 'It warms up these chambers and 
 drives off the moisture. In this way all the heat is utilized. 
 When more heat is required in these adjacent chambers a fire 
 in each one of them is started. These fires serve to complete 
 the cement burning and the waste heat is passed along 
 through other compartments in order to be utilized. As the 
 material in any compartment becomes thoroughly burned, the 
 fire is allowed to die out, and cool air is admitted in order to 
 hasten the cooling of the material. This cement material, in 
 
34 . PORTLAND CEMENT. 
 
 the condition in which it comes from the kihi, is called clinker. 
 As soon as the material has become sufficiently cool it is 
 removed from the kiln and fresh material put in. The 
 chamber has not been allowed to cool sufficiently to cause 
 much injury to the lining by contraction. When the fresh 
 material is in place, the heat from some adjacent chamber is 
 turned through it, eventually its own fire is started, and so 
 the process is made continuous. 
 
 The material to be burned is preferably in the shape of 
 pressed bricks rather than that of merely dried slurry. The 
 bricks having been subjected to a certain amount of pressure, 
 in the making, are firm, and retain their form better, during 
 the burning, than dried slurry. There is, possibh', a slight 
 increase in the amount of fuel required to complete the 
 burning, but this is more than compensated for by the 
 absence of any danger of dusting." 
 
 The advantages of using bricks instead of dried slurr}^ are 
 the same when a dome kiln is used as with a Hoffman kiln. 
 
 By the term -'dusting" is meant the disintegration or falling 
 apart of the cement material during the process of burning. 
 When this happens in a dome kiln the draft is stopped and 
 further burning becomes impossible. 
 
 In placing the bricks in a Hoffman kiln, they should be so 
 piled as to allow of the heat being drawn through e\'ery part 
 of the chamber. 
 
 The advantages attending the use of the Hoffman Ring- 
 Kiln, or kilns of a similar t\'pe, are as follows: 
 
 The saving in the amount of fuel is from lo to 20 per cent, 
 owing to the possibility of the continued use of the heat 
 from one chamber to another. Also from the fact that the 
 arrangement of Hues is such that the combustion is much 
 more perfect than is ever possible witli a dome kiln, and also 
 for the reason that the kiln is never cooled down below a 
 certain point and thus the amount of fuel that is required to 
 warm up the dome kiln is saved, after the ring kiln is once 
 started. In the Hoffman the fire is entireh' under control. 
 The burning of the cement material can be watched and the 
 
RING KILNS. 2S 
 
 amount of heat acting upon it increased or decreased at will. 
 From this fact there exists no excuse for not having the 
 cement burned to exactly the right point. 
 
 There need be no uncertainty and no rule-of-thumb as to 
 the amount of fuel needed. Each stage of the burning is 
 under full control the whole time. Nothing is left to chance. 
 There are some drawbacks attending the use of the continuous 
 ring kiln. In order to make it an economical kiln it is 
 necessary to keep it running as a continuous kiln. That is, 
 keep it running for a season or a considerable length of time. 
 When not run as a continuous kiln, but in an intermittent 
 manner, it is b}' far more expensive than the dome kiln. 
 There is another very serious difficulty that is met with in 
 the continuous running of these kilns. The finding of some 
 suitable material for the lining of the cement chambers. This 
 difficulty arises, not so much from the fact that the temperature 
 to which the lining is exposed is so exceedinglv high, but that 
 the lime in the cement material combines with the silica and 
 alumina in the tire-brick lining and causes them to flux and 
 mix with the slurry. For this reason the intermittent kilns 
 are still much used, notwithstanding the ver}- serious defects 
 that exist in them. The heat to which the lining of these 
 intermittent kilns is subjected is not so great as that in a con- 
 tinuous kiln. 
 
 There have been a Lnxat number of kilns desio-ned of the 
 continuous ring type. But in order to secure a patent upon 
 any of these designs it was necessary to depart more or less 
 from the design of the original Hoffman kiln, and, almost 
 without exception, it has been found that the excellency of 
 the kilns is diminished just exactly as the original details are 
 departed from. There are a number of special types of kilns 
 designed for cement burning that deserve short descriptions. 
 None of these are in general use, and probably even the best 
 of them would not be found economical if used upon a large 
 scale. Still the first to be described possesses some positive 
 merits in itself, while the last two are at least ingenious in 
 their conception. 
 
36 PORTLAND CEMENT. 
 
 The Bock Tunnel Kiln. — This form of kiln has been used 
 with more less success in places where it was necessary to 
 burn a certain limited amount of cement and of making the 
 first cost of the plant as little as possible. 
 
 The general design of the kiln will be understood by an 
 examination of Figs. 7, 8, 9. The tunnel is built of brick and 
 lined with fire-brick, with a 2-inch air space between the 
 lining and the outside. The cross-section of the tunnel may 
 vary to suit the work that it is expected to do. Where only 
 a moderate temperature is required, such as is needed for the 
 burning of common brick and lime, a square section with a 
 flat arch for a roof, will answer the purpose. Where the 
 product is to be Portland cement or paving brick it will be 
 found better to give a certain amount of batter to the sides, 
 with a deeper arch for the top. 
 
 The chimney is built at one end of the tunnel, and usually 
 at the feeding end. The material to be burned is run in on 
 trucks that are fastened together in such a manner as to make 
 them practically one long truck. Each of these trucks is 
 loaded with about 500 brick or their equivalent in cement 
 material. The fire boxes are on top in the kilns built for burn- 
 ing lime and ordinary brick. But when cement is to be burned 
 the boxes on top have been found not to be sufficient, and 
 additional fire boxes with gratings have been introduced on 
 each side. The tunnel is practically divided into two sepa- 
 rate compartments, the one above the other. The manner in 
 which this is accomplished is shown in Fig. 8. The platforms 
 of the trucks project until they nearly reach the sides of the 
 tunnel. Directly under this platform and running the full 
 length of the tunnel on each side, is an iron trough, Fig. 9. 
 This trough is filled with sand. From the edge of the 
 platforms is an iron lip that projects downward into this sand. 
 By this arrangement all communication between the upper 
 and lower part of the tunnel is cut off. The track and 
 running gear of the trucks are in the lower portion of the 
 tunnel, and the fire, with the material to be burned, in the 
 upper portion. 
 
38 PORTLAND CEMENT. 
 
 Fresh air is admitted to the lower portion at the chimney 
 end. This air passes along under the trucks and assists in 
 keeping everything relatively cool. At the other end of the 
 tunnel it passes around and up over the end of the truck, and 
 returns to the chimney through the material that is being- 
 burned. In this manner the proper amount of air is furnished 
 for combustion. By means of proper dampers, etc., the 
 amount of air admitted is regulated to that required. The 
 material being burned may be watched during the whole 
 process, and the amount of fuel increased or decreased, as 
 necessar}'. 
 
 The trucks are constructed principally of wrought-iron, the 
 platform being covered with fire-brick, laid in fire-clay. The 
 running gear is lubricated with a preparation of graphite. 
 
 The trucks after being loaded with the raw material are 
 forced into the tunnel by means of a hydraulic press or an 
 endless screw worked by steam. 
 
 The capacitv of the kiln is almost without limit, as it 
 depends upon the length of the tunnel or upon the number of 
 them. As yet these kilns have not been used in this country, 
 so far as the author is aware, but in France and German}^ they 
 have been used to a considerable extent in the burning of 
 brick and to a slight extent in the burning of cement. 
 
 The cost of one of these kilns of a capacity capable of 
 burning 8,000 bricks per day is about 2.000 to 2,500 dollars 
 and the number of brick required for construction about 
 140,000. The price would include track and trucks. 
 
 In the kiln just described the kiln is a tunnel and the raw 
 material is loaded on trucks and placed in it. There is 
 another type of tunnel kiln, as it might be described. In 
 this case the tunnel is a wrought-iron tube or cylinder that 
 has a slight inclination with the horizontal. 
 
 This cylinder is made to revolve upon its longitudinal axis. 
 It is set in a fire box and the whole apparatus so designed 
 that the cylinder may be heated to a white heat, or almost 
 any other temperature required, and made to revolve at any 
 required speed. The raw^ material is reduced to the requisite 
 
PATENT KILNS. 39 
 
 degree of fineness, and the different ingredients then most 
 thoroughly and intimately mixed. This dry powder is 
 introduced into the upper end of the white hot cylinder. 
 The cylinder is slowly revolved, and the fine dust is slowly 
 moved down and along the cylinder until it comes out of the 
 lower end burned. The speed with which it passes through 
 the cylinder is determined by the inclination of the cylinder, 
 the speed with which it revolves, and the amount of burned 
 cement powder that can be turned out per hour depends 
 upon the temperature at which the cylinder is held. This 
 type of kiln has some interesting features. It has never been 
 used in this country, but the principle upon which it is based 
 might well repay a careful investigation. 
 
 There is still another type of experimental kiln that, 
 although it has never been used on a large scale as far as the 
 author knows, still it may not be without merit. In the last 
 described kiln we had a metal cylinder surrounded by fire, 
 the cement making material passing through the cylinder. 
 In the type now under consideration, the tube is heated to a 
 white heat by the burning of gas on the inside. The ra\^ 
 material is tempered with water. The heated cylinder is fixed 
 horizontally in a trough, in such a manner that it can be 
 revolved. The trough is filled with the slurry, to a point just 
 beyond where it touches the cylinder. 
 
 As the heated cylinder revolves a thin coating of slurry 
 adheres to it and at some point of the revolution this thin 
 coating is removed by a scraper. The temperature of the 
 cylinder is supposed to be sufficient to thoroughly burn the 
 .slurry into clinker before it is scraped off. 
 
 Grinding. — After the cement has been thoroughly burned^ 
 the clinker is removed from the kiln and sorted. Only the 
 well-burned clinker is taken to be ground. That portion that 
 is under-burned is placed by itself to be reburned, and the 
 over-burned is rejected. As soon as the sorting has been 
 completed, the well-burned clinker is removed to be ground 
 to powder. The clinker is first passed through some type of 
 crusher, similar to those described in Chapter VII. In this 
 
40 
 
 PORTLAND CEMENT. 
 
 manner it is reduced to the size of coffee grains, and is then 
 reduced to powder by means of millstones. Nothing yet has 
 been devised that will do the work as well and as cheaply as 
 properly dressed millstones. 
 
 The necessity of reducing the clinker to flour and of doing 
 this as cheaply as possible has led to the designing of a 
 countless variety of mills. The two types upon which the 
 most work has been done are what are known as edge runners, 
 and the ball mills. Of these two the ball mills are giving the 
 most satisfaction. 
 
 There is this peculiarity in regard to the grinding of 
 cement. The mere fact that it is fine does not satisfj^ the 
 necessary conditions. It must be flour}-. Two cements mav 
 have the requisite degree of fineness and still one of them be 
 all grit and the other all flour. The more floury the cement 
 is, the better will be the results obtained with it. As yet 
 there has been no grinding machine invented that for an 
 equal amount of grinding will produce such a large per cent- 
 age of flour as the ordinary millstones. 
 
 In regard to the amount of power consumed in grinding 
 cement by each of the above principles, the following table is 
 taken from "Faija on Portland Cement," Trans. Am. Soc. C. 
 E., October, 1893, p. 59: 
 
 Millstones, 30 to 32 I. H. P. per ton per hour. 
 
 Ball mills, 16 to 18 I. H. P. " - 
 
 Edge runner, 12 to I. H. P. " " " " 
 
 In each case the cement was ground to about a fineness of 
 5 % residue on 2,500-mesh sieve. 
 
 Sifting. — In order to insure a uniformity of fineness, some 
 makers have all the cement sifted as it comes from the stones. 
 The residue on the sieves is thrown back and regroimd. 
 
 The sifting of the cement is in itself a most excellent idea. 
 But care must be exercised, that the workmen do not make 
 this subsequent sifting an excuse for careless grinding, and 
 thus necessitate the regrinding of a large per cent, of the 
 output. 
 
STORING AND PACKING. 41 
 
 Storing. — After the cement has been ground it should be 
 stored in suitable store-rooms for a number of da3-s at least. 
 
 Very little Portland cement is at its best immediately after 
 being ground, and under the remarks upon the testing of 
 cements we have noted some of the peculiarities of the 
 cement in this respect. The cement should be stored in 
 perfectly dry and well ventilated store-houses. It should be 
 spread out in layers of 6 to 10 inches deep. These la3'ers 
 can be turned when desired and thus the beneficial effect of 
 storage hastened to some extent. 
 
 Most cement makers do not seem to realize the great 
 advantages that would obtain by the having of large and 
 suitable store- houses. Not only would the quality of the 
 cement be improved but, with large store-room space, the 
 works could run even when the market was temporarily dull. 
 
 Packing", — When the cement has become thoroughly cool 
 or is to be forwarded to the market, it is packed in barrels 
 or sacks. The barrels are lined with water-proof paper, in 
 order to protect the cement from moisture. When the cement 
 is to be used at once, it is usually packed in sacks. These 
 sacks are either of paper or burlap. The paper is the cheaper- 
 and is also preferable, as better preserving the cement. The 
 weight of a barrel of Portland cement is about 400 pounds,, 
 gross. The sacks are of such a size that they hold % or 
 as much cement as a barrel. 
 
CHAPTER IV. 
 
 TESTING CEMENTS. 
 
 In canying on any work of construction in which cement 
 forms an important factor, it is of the utmost importance that 
 the engineer should know exactly what class of cement he is 
 using and what degree of excellency he may expect to obtain. 
 
 In order to obtain this information with a reasonable degree 
 of accuracy some method of testing the cement is necessary. 
 What these methods shall be is still a mooted question among 
 the users of cements, and one upon which a great diversity of 
 opinion obtains. 
 
 T he following are the usual characteristics for which cement 
 is tested, viz. : 
 
 Tensile Strength, Coiupressive Strength, J^ineness, Setting 
 Cross Strain, Ad/iesion, Cracking, and Checking. 
 
 Tensile Strength. — The most usual method of testing the 
 quality of cement is by means of tension. The neat cement, 
 or any desired proportions of cement and sand, is mixed with 
 sufficient water to make a wQvy stiff paste. This paste is 
 molded into briquettes, and after a certain lapse of time 
 these briquettes are pulled apart, and the number of pounds' 
 strain necessar}- to do this is recorded. The shape of the 
 briquette and the details of manipulation are fully described 
 later. 
 
 This is the most usual test for cement. 
 
 In actual work cement is never used in such a manner as to 
 
CEMENT TESTING. 43 
 
 subject it to a tensile strain, and the reason that it is subjected 
 to tensile strain in testing is simply one of convenience. 
 
 Tension can be applied to a briquette with more facility 
 and uniformity than compression. Any irregularity in the 
 making of the briquette can be more nearly equalized in the 
 testing machine. 
 
 The apparatus necessary for these tests can be very easily 
 improvised at a small cost. They occupy little space, and can 
 be readily moved from place. 
 
 The objections to tests by crushing are: 
 
 ist. With any reasonable amount of care in the preparation 
 of the specimens and the carr3ang on the tests, the results are 
 most variable and uncertain. 
 
 In order to make these results of value, much care and 
 much skill is necessary in the preparation of the specimens 
 to be tested. The size of all the specimens to be tested must 
 be exactly the same. It is not enough that we know the 
 dimensions of the different specimens. In order that correct 
 comparative results may obtain, these specimens must all be 
 of the same dimensions. There is very little known as to the 
 actual effect of increasing the size of a specimen to be subjected 
 to compression. Take a specimen one inch square and two 
 inches long, it will take a certain number of pounds to crush it. 
 
 Take a specimen of the same material, four inches long 
 and two inches square, that is containing four square inches 
 of crushing area. It will be found that the required crushing- 
 force will be much greater than four times the amount required 
 to crush the first specimen, or the resistance to crushing per 
 square inch is greater in the larger specimen than in the 
 smaller. A more unequal result ma}- be obtained if the length 
 of the specimen in proportion to its sectional area is varied. 
 From this can be seen the necessity of absolute uniformit}' in 
 the size of the specimens to be tested. 
 
 This requirement does not exist to such a degree in tension 
 tests. 
 
 Another great drawback to compression tests is the fact 
 that in order to make the results of value the ends of the 
 
44 PORTLAND CEMENT. 
 
 specimens must be parallel and at right angles to the sides. 
 In order to insure this, it is necessary that each specimen to be- 
 tested should be ground with special apparatus and great care. 
 
 In order to get any satisfactory results, even when all the 
 above conditions have been complied with, it is necessary to 
 make the specimens of such a size that the ordinary cement 
 testing machines are not powerful enough to crush them, and 
 a larger and much more expensive machine is needed. 
 
 P"or these reasons and others, can be seen the objections of 
 requiring the compressive test, in any series of standard tests, 
 and this notwithstanding the fact that cement is usviall}^ used 
 in compression. 
 
 Fineness. — Cement is usually tested to ascertain the degree 
 of fineness to which it is ground. The quality of any cement 
 is improved by fine grinding. The degree to which this 
 should be carried is a mere question of cost. 
 
 The finer the cement is ground, the better it is in quality, but 
 also the greater the cost of manufacture, until at last a point 
 is reached where any greater increase in fineness costs more 
 than it miproves the quality of the cement. The grinding 
 should stop just before this point is reached. 
 
 It is perhaps needless to say that in most cements the 
 grinding stops much before this point is reached. The test 
 for fineness is to pass the cement through a sieve of 10,000 
 meshes to the inch. That is 100 wires to the lineal inch. 
 The wires are what are known as No. 40 Stub's Gauge. 
 
 The amount of cement to be passed through the sieve is 
 carefully weighed and then that portion weighed that is left 
 on the sieve. Thus the so-called per centage of fineness is 
 obtained. A cement that will all pass through a sieve of 625 
 meshes per square inch, and only leaves 4 or 5 % on a sieve 
 of 2,500 meshes per square inch, is, for practical purposes, 
 fine enough. 
 
 The cement tliat is to be made into briquettes and pats 
 must not be sifted, unless for special reasons, but must be 
 used exactly as it comes from the barrel. In case good 
 results are not obtained by the tests made with the cement as 
 
BRIQUETTES. 45 
 
 taken from the barrels, then some of the cement should be 
 sifted, and briquettes made of that portion that passes the sieve. 
 A comparison of the results obtained with the cement sifted 
 and vinsifted will show how much the quality of the cement 
 would be improved by liner grinding. 
 Form of Briquette. — Fig. 10 shows the form of the bri- 
 quette that was recom- 
 
 if T'^'tI^ mended by the American 
 
 Societ\' of Civil Engineers 
 and has since become the 
 standard form in this coun- 
 try. The weakest sec- 
 tion, as will be seen, is at 
 the center. The briquette 
 
 -\- _ jr~^= IJ is one inch thick and at 
 
 V the smallest section is just 
 
 ^' one inch across, giving a 
 
 breaking area of one square inch. This breaking section is 
 much smaller than that used in England or on the Continent. 
 It was adopted by the American Societ}- of Civil Engineers 
 because of the fact that it was found to be sufficiently large to 
 give good results, while its smallness rendered it much easier 
 to make the briquettes of a uniform density throughout, with 
 a freedom from air bubbles. 
 
 The Making of Briquettes. — Briquettes for testing cement 
 are made in two w^ays, by hand and by machine. When 
 made by hand, the mortar is made of the right consist«-:ncy 
 and pressed into molds by means of a small trowel. The 
 briquette is left in the mold until sufficiently set to allow of 
 being handled. When only a few briquettes are to be made, 
 and all of these made by one person, the hand molds answer 
 the purpose. There is always this disadvantage attending 
 their use: the mortar being placed in the molds by hand 
 and .pressed down with a trowel, it is almost impossible to 
 give the same amount of pressure to each briquette, and the 
 personal equation of the maker becomes a very prominent 
 feature in the briquettes, and when these briquettes are made 
 
46 
 
 PORTLAND CEMENT. 
 
 b\- two or more persons satisfactory results are almost im- 
 possible. 
 
 Let two or more persons make a number of briquettes 
 from the same lot of cement, with the same proportions of 
 water and in apparently the same manner, and when these 
 briquettes are broken there will be a marked variation in the 
 results obtained from each lot. This variation is often so 
 marked as to materially vitiate the result. 
 
 Molds. — The form of the molds used for hand -made 
 briquettes is shown in Figs, ii, 12, 13. The material is 
 usually gun-metal or some other alloy of copper that is suffi- 
 
 FlG. II. 
 
 Fig. 12. 
 
 Fig. 14. 
 
 Fig. I T.. 
 
 cientl}' hard and does not become 
 rusted by exposure to moisture. 
 The molds are made in two 
 pieces in order to facilitate the 
 removal of the briquette. When 
 in use the two parts are held to- 
 gether by means of a spring, as. 
 shown in Fig. 14, or by clamp, as 
 shown in Fig. 15. 
 
 Briquette Machines.^ — The evils 
 attending the use of hand-made 
 
THE BOHME HAMMER. 47- 
 
 briquettes have been already mentioned, and numerous at- 
 tempts have been made to so modify these methods as to 
 
 insure a greater uniform- 
 ity in results obtained from 
 briquettes made from the 
 same cement. All the 
 work done for this pur- 
 pose has been along the 
 line of obtaining uniform- 
 ity in the pressure ap- 
 plied to the cement dur- 
 ing the making of the 
 briquette. 
 
 The Bbhme Hammer. 
 — In Germany the stand- 
 ard briquettes are made 
 with the "Bohme Hammer apparatus," Fig. 16.* 
 
 The apparatus consists of a tilt hammer with automatic 
 action. 
 
 "The hammer is driven by a cam wheel of 10 cams actuated 
 by simple gearing. The wa-ought-iron handle of the hammer 
 "is let into the cross-head which carries the axle of the ham- 
 "mer, and keyed to this cross-head and to the cap, so that 
 "it may be readil}' replaced if worn. The steel hammer 
 "weighing 41^2 pound, is similarly fastened to the cap. As 
 "soon as the intended number of blows has been delivered, 
 "the mechanism is automatically checked, the proper setting 
 "have been made for this purpose before beginning the work. 
 
 " The number of blows required in the German Standard 
 "tests is 150. The forms to receive the mortar consist of a 
 "lower and upper case held together by springs. The lower 
 "case for compression specimens consists of two angle-irons 
 "held on a plane plate by a grinding strip and a screw acting 
 "on the latter. Upward m.otion is prevented by two wedge- 
 
 *The following description is 
 1893. "Gary on Testing of Por 
 
 taken from the 
 tland Cement." 
 
 Trans. 
 
 Am. Soc. C. E., Oct.^ 
 
Fig. i6. 
 
PORTLx\ND CEMENT. 
 
 "shaped surfaces. The lower case and half the upper one is 
 "filled with the mortar to be tested and a plate laid upon its 
 "surface. On this plate the blows are deUvered. It is of 
 "vital importance that the apparatus rest on a firm non-elastic 
 "foundation, preferably it should be placed and fastened on a 
 " pier of masonr3\" 
 
 The Jameson Briquette Machine. — The description of 
 the machine is as follows: The main portion of the machine 
 consists of the cylinder A, Figs. 17 and 18, which is flanged 
 at the lower end, this flange corresponding in shape and size 
 to the upper part of the base as seen in Fig. 19. The cyUnder 
 
 l"lG. 19. 
 
 is bolted to the base by four bolts, each boh provided with a 
 filler that holds the lower face of the cylinder i inch above 
 the base plate. Both of these faces are accurately planed. 
 
 It is between these two planed faces that the molding plate 
 Z, Fig. 20, swings, the fillers on the bolts acting as stops. 
 
 The cylinder A is made in two parts, bolted together. The 
 bore is the shape and size of the briquette and is shown in 
 plan in Fig. 19. 
 
JAMESON BRIQUETTE MACHINE. 51 
 
 In this bore works a solid plunger of the shape of the bore. 
 The length of this plunger is of suflicient length to cover the 
 feed hole, when at its lowest point. The plunger is connected 
 
 to the lever B by the connect- 
 ing rod C. The fulcrum of 
 the lever B is at D. 
 
 About the point Fig. 19^ 
 the plate L swings in such a 
 manner, that when at either 
 extreme one of the openings is 
 directly beneath the bore of 
 the cylinder, while the other 
 is directly over the extractor 
 Fig 19. 
 
 When the extractor is low- 
 ered ita top is a little below the 
 lower side of the molding plate. 
 These extractors /'"and F are 
 of the shape of the opening in 
 the molding plate, and ore rais- 
 FiG. 20. ed by the levers G G. The 
 
 object of the extractors is to force the briquettes from the 
 mold. On the outside of the cylinder is the hopper /, Fig. 17. 
 
 The method of operation is as follow^s: The piston is raised 
 until it is above the feed hole, and the cement or mortar in 
 the hopper is forced into the cylinder. 
 
 The molding plate is pushed against one of the stops so 
 as to bring one of the openings under the cj'linder bore. 
 
 The lever B is forced down, causing the plunger to force 
 the cement or mortar into the opening in the molding plate. 
 The molding plate is then swung against the other stop. 
 This movement cuts off the briquette and places it directl}' 
 over the extractor. The other opening in the molding plate 
 is directly under the cylinder bore. The extractor is raised 
 by the lever G, the briquette forced out and removed. The 
 extractor is lowered, the main plunger forced down again, the 
 molding plate swung and another briquette made. The 
 
52 PORTLAND CEMENT. 
 
 cylinder will hold sufficient material for three briquettes. It 
 should then be filled again. 
 
 The machine is best operated by two men, one to feed and 
 operate the long lever, and the other to swing the molding 
 plate and remove the briquettes. 
 
 In regard to the capacity of the machine: three students 
 have made three thousand briquettes in ten hours. 
 
 As this machine was an experiment and the strictest econ- 
 omy as to cost was necessary, it is in some respects crude, 
 and could be made much more convenient for working. 
 Probably the first fault that will be noticed is the fact that 
 there is no method of regulating the amount of pressure used 
 in making the briquette or rather that in no way can one be 
 sure that an equal pressure is exerted on each of the briquettes. 
 Experiments were made to test the necessity of this, and as 
 no decrease in the per centage of variation in the briquettes 
 
 was found, such an exact reg- 
 ulation was not deemed nec- 
 essary. This probably arises 
 from the fact that the actual 
 pressure is so great under all 
 circumstances that the actual 
 variation forms but a small 
 per centage of it, not suffi- 
 cient to' vary the results. 
 
 In experimenting with a 
 fixed pressure, the plunger 
 was lowered until it rested 
 upon the top of the cement 
 in the cylinder. Then a 
 weight was dropped a fixed distance upon it. 
 
 Clips. — The shape of the clips is shown in Figs. 21 and 
 22. In the early testing of cements the briquette was placed 
 in these clips in direct contact with the metal. In time, how- 
 ever, this was found to be objectionable from the fact that the 
 briquettes were not always of such a form that the clips came 
 to a full bearing. The consequence was that the clips, only 
 
CLIPS 
 
 53 
 
 Fig. 22. 
 
 bearing on one side, brought a cross-strain upon the briquette, 
 and it often broke at a point much below the 
 normal strain or else broke at some point other 
 than the minimum section. 
 
 This fact led in a short time to the cushioning 
 of the briquette by putting pieces of blotting 
 paper between it and the clips. In this way 
 much of the evil was done away with, but it was 
 more or less a nuisance to use the blotting paper 
 and there was an irregularity of practice among 
 different investigators. 
 
 Sheets of rubber were tried with marked suc- 
 cess, and this led at once to the making of the clips 
 of the form shown in Fig. 23, where that portion of 
 the clif that comes in contact with the briquette 
 is covered with a small piece of rubber tubing. 
 The arrangement is such that this tubing can be replaced 
 when necessary, and much more uniform results have been 
 obtained since the introduction of the rubber 
 buffer. 
 
 In placing briquettes ^in the testing machine 
 care should be exercised that everything comes 
 to a full and even bearing. Nothing should be 
 hurried, and a certain amount of skill is neces- 
 sary. The strain should be applied very slowly, 
 about 500 pounds per minute. When the bri- 
 quette has broken, a note should be made of the 
 number of pounds' strain at which it broke, and 
 remarks as to the character of the fracture, — 
 whether it occurred at the minimum section or 
 not, and whether it is a square, even fracture. 
 
 If not at the minimum section, or if irregular, 
 in shape, a note should be made of any apparent 
 fault in the briquette that might account for the irregularity of 
 breaking. Even with the greatest possible care being used 
 in the making of the briquettes and in their subsequent 
 manipulation there will be found a great per centage of 
 
54 PORTLAND CEMENT. 
 
 variation in the results obtained. An examination of the tests 
 made in the Testing Laboratories of the State University of 
 Iowa will clearly show this. 
 
 Taking a good imported Portland cement and making the 
 briquettes by hand, it was found at the end of twenty-eight 
 days that there would be a variation in the breaking strength 
 of about ten per cent. When the briquettes were made in 
 large quantities in Jameson's Briquette Machine the variation 
 was found to be reduced to about four per cent., and upon 
 selecting one hundred apparently perfect briquettes from a 
 lot of five hundred, the per centage of variation fell below 
 two per cent. These experiments also show the great advan- 
 tage that obtains with machine-made briquettes. 
 
 The author has never had the opporLuaity of using the 
 Bohme Hammer apparatus, and therefore can give no com- 
 parative results. 
 
 In order to insure a greater uniformity in hand-made bri- 
 quettes it is the usual custom in England to mold the 
 briquette in the following manner: the description is taken 
 from ''A Manual of Lime and Cement," by A. H. Heath: 
 
 *'The mixed cement is to be lightly, placed in the molds, and 
 '•then is to be pressed for five minutes under a load of ten 
 "pounds placed on top of the soft cement projecting above 
 '•the mold. The loading block to be shaped to the mold, with 
 " inch clearance, and is to be placed on the cement sym- 
 '• metrically; after loading, the surplus cement to be cut off 
 "with a trowel or knife, and the briquette smoothed level 
 "with the top of the mold." But with the exercise of all 
 possible care the hand-made briquettes do not in any way 
 compare with those made by machine. The advantage of 
 the machine is two fold. First the great pressure to which 
 the briquettes are subjected (almost six hundred pounds) 
 makes them of a nearly uniform density, and allows of their 
 being taken at once from the machine and placed on a slab. 
 
 Second, in the rapidity with which the}' can be made. This 
 rapidity is not only an advantage in itself as a saving of time, 
 but has also this great advantage, that it allows of the mixing 
 
MIXING MORTAR. 55 
 
 •of a much o-reater quantity of mortar at -once, and thus 
 making a greater number of briquettes from mortar that is 
 uniform throughout. 
 
 In making briquettes by hand it is usual to mix sufficient 
 mortar to make five briquettes. If more mortar than this is 
 made at one time and the cement is in any way quick setting, 
 there is danger of an incipient set and the necessity of 
 reworking the mortar. The evils attending such an opera- 
 tion are pointed out in the chapter upon the using of cement. 
 In making briquettes b}^ machine sufficient mortar for eiglity 
 briquettes can be mixed at once, and the time required to 
 make this number by machine is less than that required to 
 make five by hand. 
 
 Mixing- the Mortar. — Until within a few years the bri- 
 quettes for testing cement were made solely of neat cement. 
 
 The American Society of Civil Engineers, however, in 1885 
 adopted the suggestion of their committee that in testing 
 cement the briquettes should be made of neat cement. Then, 
 for natural cements, i part cement and i part sand; and for 
 Portland cements, i part cement and 3 parts sand. 
 
 The proportions are by weight, and the per centage of 
 water used is in proportion to the combined weight of cement 
 and sand. The actual amount of water used depends to some 
 extent upon the quality of both the sand and cement, and 
 also the proportions in which they are used. As little water 
 as possible should be used, just sufficient to give the mass the 
 consistency of damp sand. There is very little danger of 
 using too little water. An approximate amount is 20 to 25 9^, 
 for neat cement, 15 % for one part of sand, and 10 to 12 % 
 for three parts of sand. The actual amount used should be 
 carefully noted. 
 
 The temperature of both the water and the laboratory 
 should be constant, between 60° and 70" Fah. 
 
 The mixing of the mortar for testing should be done upon 
 some non-absorbent surface. If only a few briquettes are to 
 be made, say five or ten, a heavy glass slab answers the 
 purpose. When a larger amount, sufficient for use in a 
 
56 
 
 PORTLAND CEMENT. 
 
 briquette machine, the author has found a small porcelain- 
 lined sink the most convenient. 
 
 The cement and sand are most thoroughly mixed dry, and 
 then the proper amount of water added gradually. The 
 mixing must be most thoroughly done until the whole mass is 
 of a uniform dampness and consistency. This mixing must 
 also be carried on with rapidity. 
 
 As soon as the mixing of the entire mass has been com- 
 pleted the work of molding the briquettes should begin at 
 once. 
 
 • Sand. — There are two kinds of sand that are used in 
 cement tests. Which should be used depends upon the 
 object for which the tests are being made. If this object is 
 to determine the value of the cement for any particular piece 
 of work, the sand used should be the sand that is to be used 
 in the work, and it is the mortar that is to be tested, not the 
 cement alone. 
 
 In case, however, the results should be unsatisfactory, and 
 any doubt exist as to the quality of the sand, then the cement 
 should be tested with "standard" sand before being rejected. 
 In laboratory tests where a long series of experiments are 
 being carried on, the results of which are to be compared 
 with those obtained by other experimenters, it is necessary 
 that some standard sand should be used, sand that can be 
 readily duplicated if desired. For this purpose, crushed 
 quartz of a given definite size, such as is used for the making 
 of sand-paper, should be used. This sand can be procured 
 of any desired size, is clean, sharp and in every way most 
 excellent. Its use renders possible the carrying on of com- 
 parative tests at any number of places. 
 
 Pans or Slabs. — If the briquettes are made by hand in 
 molds, they must be left in the mold sufficiently long to 
 become hard, usually not less than two hours. If machine 
 made, they are delivered direct from the machine, in a condi- 
 tion to be handled. These fresh briquettes should be placed 
 upon some non-absorbent substance. For a small number,, 
 sheets of glass are very convenient; for greater numbers,, 
 
MANIPULATION OF BRIQUETTES. 
 
 shallow pans of galvanized iron are good. These pans, slabs,, 
 etc., are sold by all dealers in testing laboratory outfits. 
 When the briquettes are placed on these glass slabs or in the 
 pans, they are covered with a damp cloth and allowed to 
 remain there for twenty-four hours. 
 
 The temperature of the testing room should not be above 
 70° nor below 60°, and of course the briquettes should not 
 be left in any position where they will be exposed to the 
 sun or any violent change of temperature. After twenty-four 
 hours under a damp cloth, a portion of the briquettes should 
 be immersed in water. 
 
 Tanks. — The galvanized iron pans for sale by the dealers 
 are about two inches deep, and can be had to hold one 
 hundred briquettes, the briquettes being on edge. 
 
 Where a briquette machine is used, as in a University 
 Laboratory, and a great number of briquettes made, the 
 author has found the following arrangement of tanks very 
 satisfactory. The tanks are made of wood of any length 
 that will suit the size of the tank room. They are about four 
 inches deep on the inside, and three or four inches wider than 
 the galvanized iron pans. There is an inlet for the water at 
 one end and an overflow at the other. As many of these 
 tanks can be made as is necessary, and they can be placed, the 
 one over the other, with sufficient space between. The pans 
 containing the briquettes can be set in them without in any 
 way disturbing the briquettes, and the water turned on. Each 
 pan is numbered, and thus the different briquettes are entirely 
 distinct. 
 
 Increased Temperature. — In mixing the mortar for bri- 
 quettes careful note must be made as to any increase in 
 temperature. This increase in temperature indicates the pres- 
 ence of free lime. In the best Portland cements the increase 
 of temperature upon the addition of water is very slight, but 
 in some cases is sufficient to be noted by the hand. With the 
 light-burned natural cements there is a very marked rise in 
 the temperature, and in working any such cement in the 
 laboratory, rubber gloves should be used. The amount of the 
 
i8 PORTLAND CEMENT. 
 
 increase of temperature is, to a certain extent, proportional to 
 the amount of free lime in the cement, and various attempts 
 have been made to use this fact as a test for the actual per 
 centage of free hme present. This method, however, of 
 determining the free lime has not been found practical, owing 
 to the fact that, to make such observations of any value, it is 
 necessary to use more care and refinement in conducting the 
 experiments than is practically possible in a cement laboratory. 
 
 Setting. — The test for setting and hardness that is usually 
 used is what is called the needle test. This test was first used, 
 to any extent, in this country by Gen. Q. A. Gilmore, and is 
 thus described in his "Practical Treatise on Limes, Hydraulic 
 Cements and Mortars" pages 30, 31: — ''Their rclaih'c hard- 
 ness''' (referring to the setting of different cements) : --This 
 ••was measured by the penetration of a steel point or needle, 
 ••impelled by the impact of a falling body. The needle, which 
 "is slightly conical, or tapering towards the point, is truncated 
 "at right angles to the axis, so as to give a diameter at the 
 "lower end of of an inch. It protrudes from a socket in 
 "the lower extremity of a vertical rod or spindle to which it is 
 "firmly secured by means of a thumb-screw. To the upper 
 "extremity of the spindle is attached a diagonal scale of steel, 
 "accurately graduated to tenths, hundredths and thousandths 
 "of an inch, and provided with a horizontal index firmly fixed 
 "to the frame-work of the instrument. 
 
 "The absolute penetration of the needle is obtained by 
 "taking the difference between the index readings before and 
 "after impact. The falling body is a hollow metal C3'linder, 
 '•weighing one pound, of which the exterior diameter is about 
 "equal to the length. 
 
 " This cyhnder, in its descent, passes freely over the spindle, 
 "and strikes upon the shoulder attached just above the screw. 
 "The mortar used to determine the hardness was that of the 
 "broken prisms, and the penetrations w^ere taken the same 
 "day, generally but a few hours after they had been broken. 
 " As these fragments were two inches square in cross-section, 
 ^•and seldom less than two and one-half inches long, they 
 
KK!':i)Ll'] TESTS. 5p 
 
 •-admit of several trials with the needle. An averaae of not 
 *'less than four penetrations, and sometimes more, at each 
 "end of the prisms, was taken on all occasions, except when 
 "the fragment split open at a lower number, which was some- 
 " times the case." 
 
 Gen. Gilmore also used a ^^V inch wire with flat end, loaded 
 ■with lib., and a ^i^- inch wire loaded with one pound. These 
 were used on cakes of neat cements 2 or 3 inches in diameter, 
 -L inch thick in the center and ] inch thick at the edges. 
 One cake was left in the air and one cake immersed in water. 
 The time at W'hich the loaded wires cease to penetrate the 
 pat is to be noted. 
 
 In England a Portland cement is considered to be quick 
 setting when it will bear the -^V inch needle loaded with 4 ozs. 
 in ten minutes after mixing with water, and to be slow 
 setting if the time occupied is 30 minutes or upwards to 5 
 
 Fig. 24. 
 
 hours. When not sufficiently hard to bear the needle after 5 
 hours, the cement may be rejected. These tests are made 
 with neat cement. 
 
 Vicat's needle test, Fig. 24, is used at the Boulogne Harbor 
 Works; io| ozs. (300 grammes) upon a needle with a cross- 
 section of 0.00155 square inches (i square millimeter). At 
 
<5o 
 
 PORTLAND CEMENT. 
 
 Boulogne a slow setting cement was specified; and one that 
 would support this needle in less than half an hour was rejected. 
 
 German Needle Test. — The standard German needle test 
 is a needle that has 0.0315 square inch sectional area, weighted 
 with 10 ozs. This is placed vertically upon the top of the 
 cement pat to he tested. This pat is i| inches thihk and may 
 be inclosed in a metal ring or mold 3.15 inch interior diameter. 
 
 The cement is said to be completely set when the needle will 
 make no impression upon the top, and setting is said to have 
 begun when the needle fails to pass through the entire cake. 
 
 The author has tried these various methods of testing the 
 hardness and the setting properties of cements and has found 
 them all more or less unsatisfactory. In the first place, there 
 is an absence of uniformity of methods used by the different 
 experimenters, and this lack of uniformity renders any com- 
 parison of the results obtained of very little value. 
 
 Another source of error or lack of uniformity results from 
 the fact that too much depends vipon the personal equation of 
 the investigator. No two persons w-ill obtain the same result 
 on the same cement with any of these needle tests. 
 
 In order to make the tests of the greatest value they must 
 be made to a great extent mechanically, and the results 
 obtained and recorded automatically. The results obtained 
 by any of the methods heretofore described are only obtained 
 at stated intervals, and are merely isolated results. 
 
 There is one point in the setting of cement that is not 
 touched upon in these methods and that is the rate of setting.. 
 We get merely two points in a curve, — the beginning and the 
 end. We know when the cement begins to show an appre- 
 ciable set, and we know the point at which the increase of 
 hardness becomes so slow that the cement is said to be set.. 
 Between these two points we have nothing to show the shape 
 of the curve. In any tests that have been made, that show 
 rate of setting from time to time, it will be found that different 
 cements set in an entirely different manner. 
 
 The author has used a machine in his own laboratories in 
 the summer of 1894, for the examination of the setting quali- 
 
S. U. I. NEEDLE TEST. 
 
 6i 
 
 ties of cements, that has given most satisfactory results, and, 
 •SO far as he knows, is of original design. 
 
 A small drill such as is used by dentists, with a bur on the 
 end -i~ of an inch in diameter, takes the place of the needle. 
 The shank of this drill is about i| inches long and smaller 
 than the burr. 
 
 The drill is clamped to a vertical spindle. This spindle has 
 a screw thread cut on it and passes through a metal plate, 
 where it fits into a corresponding thread. This plate is fixed 
 in a horizontal position and the spindle is raised or lowered 
 when turned in the plate. 
 
 The spindle is connected with clock-work in such a manner 
 that a slow rotary motion can be given to it. This clock-work 
 is driven by a light spring and the spindle, gearing, etc., must be 
 kept clean and well oiled in order to insure uniformity of work. 
 
 The upper end of the spindle passes above the clock-work, 
 to a bell crank the shorter arm of which is fastened to the 
 spindle. The longer arm carries a pencil that records an}^ 
 movement on a chronograph, the chronograph being operated 
 T3y separate clock-work. 
 
 The method of using the machine is as follows: a pat of 
 neat cement in a mold is placed in the machine below the 
 drill. The drill is lowered until it just touches the cement, 
 then the clock-work of the machine and of the chronograph 
 are both started. 
 
 The drill rotates and works its way down into the cement, 
 the distance that it penetrates being recorded on the chrono- 
 graph, with the time. As the cement begins to set, the amount 
 •of resistance encountered by the drill becomes greater, the 
 Tate of rotation less, and consequently the rate of penetration 
 less, until at last a point is reached when the cement has 
 became so hard that the drill can make no impression upon 
 it and the machine stops. We then have written upon the 
 chronograph an exact autobiography of the rate and manner 
 of setting of the cement. Nothing has been left to the skill 
 or lack of skill in the investigatoi- and the results are contin- 
 uously recorded. 
 
62 
 
 PORTLAND CEMENT. 
 
 Fig. 
 
 25- 
 
 Adhesion. — The best comparative results in the tests for 
 adhesion can be obtained by using two pieces of coarsely 
 ground glass about 4 inches wide, 8 inches long and i or il 
 inches thick. If this is used in a number of experiments, 
 carried on at different times and in different places it insures 
 a uniformity of material upon which to work. 
 
 The mortar should be made in the required proportions, 
 using less water than would be used in ordinary masonr}' work. 
 
 The mortar is then spread upon one piece of glass and the 
 other piece placed at right angles with the first as shown in 
 Fig, 25 and pressed down until the mortar 
 joint is I of an inch thick. 
 
 The sample is then allowed to stand 24 
 hours in the air under a damp cloth and 
 then immersed in water. They are pulled 
 apart at the end of 7 or 28 days. 
 
 When it is wished to make the tests for 
 any particular piece of work and the results 
 are not to be compared with results obtained in other places. 
 The material (brick or stone) that is to be used in the work 
 should be used in the tests. When brick or stone are used 
 in tests they should be wet in exactly the same manner as 
 they are to be in the work. 
 
 A few tests made in this wa}' will not only give the adhesive 
 value of the mortar, but by varying the treatment of the brick 
 on stone used, very interesting results can be obtained in 
 regard to the value of thoroughly wetting the material before 
 the mortar is applied. 
 
 Cross-Strain. — The specimen to be subjected to cross- 
 strain can be of almost any size. About two 
 inches square and eight inches long is con- 
 venient. The making of the specimen and 
 its subsequent treatment are governed by the 
 rules that have been given for the making 
 of briquettes. When the specimen is to be 
 broken it should rest upon two supports a known distance 
 apart. With an eight inch specimen these supports are usually 
 six inches apart. 
 
 XL 
 
 Fig. 26. 
 
BLOWING. 
 
 63 
 
 The form of the support should be similar to that shown in 
 Fig. 26. The knife edge is slightly rounded so as not to cut 
 into the specimen. The load is brought to bear upon the 
 specimen midway between the supports. 
 
 Blowing". — There is one most important feature that all 
 commercially good cements must have, and that is stability of 
 form. It is absolutely necessary for the future durability of 
 the work that the cement used should neither shrink nor 
 swell after the process of setting has once commenced. 
 
 It is one of the peculiarities of Portland cement that when 
 its constituents have been improperly mixed or the process of 
 manufacture improperly carried on, it will have a tendency to 
 expand, crack and disintegrate after the setting has com- 
 menced. This expanding, cracking and disintegrating is 
 called '-blowing," and such cement is said to be "blowej^" 
 cement. There are a number of methods used in testinfj for 
 this feature. Some of the cement is mixed with a small 
 amount of water and then pressed firmly into a straight glass 
 lamp chimney. If any dangerous amount of expansion takes 
 place the chimney will be broken. By this method also can be 
 examined any shrinkage of the cement. This is done by 
 pouring some colored liquid into the chimney after the cement 
 has thoroughly set, and an idea of the amount of shrinkage 
 can be obtained by the amount of liquid that runs down the 
 inside. In Germany at some of the cement works great care 
 has been taken to investigate with accuracy the actual amount 
 of expansion that takes place. For this purpose Bauschinger's 
 caliper apparatus has been used. With this apparatus the 
 expansion can be measured with an accuracy of . ^j^^j ,j of an inch. 
 
 This expansion may not be perceptible until two or three 
 days after the setting has begun, or it may become perceptible 
 almost immediately. The quicker setting the cement is, the 
 more quickly the blowing will commence. 
 
 One of the best methods of testing for blowing is as follows: 
 viz.: Gauge the cement with only sufficient water to give it 
 the consistency of damp sand; make a few ounces of it into a 
 pat about 3 inches in diameter, or J inch thick at the 
 
.•64 PORTLAND CEMENT. 
 
 ' edges and 4- inch in the center. Make this pat upon a plate 
 of glass or some non-absorbent material. The pat should be 
 left on the glass under a damp cloth for 24 hours, and then 
 immersed in water. If the pat, at the end of three days in 
 water, shows no signs of cracking or disintegrating at the 
 edges, it can be considered safe. In examining the pat for 
 cracks the fine hair-like cracks found on the surface, that cross 
 and recross each other are not due to blowing, but are merely 
 the result of slight changes in temperature. They do not 
 •denote a poor cement. The cracks due to blowing are wedge- 
 shaped, radiating from the center, and usually accompanied 
 by a certain amount of disintegration. 
 
 In order to hasten this test, and thereby save much time, 
 the following method is used and recommended by Mr. Henry 
 Faija, M. Inst. C. E. The apparatus consists of a vessel of 
 . such desi^rn that water in it can be maintained at a constant 
 temperature of 110° to 115° Fah. 
 
 The vessel is covered and under the cover but above the 
 water level is a rack upon which the pat can be placed. The 
 pat is made and placed on the rack where it is subjected to the 
 moist atmosphere above the warm water, the temperature 
 of the moist air being a little over 100° Fah. The pat is kept 
 on the rack about .6 or 8 hours and then immersed in the 
 warm water for 16 or 18 hours. If at the end of that time it 
 remains firm, showing no signs of disintegration and still 
 adheres to the glass, it can be considered safe. 
 
 In case however, that it does disintegrate, it does not 
 necessaril}' indicate that either the constituents or method of 
 making are at fault. The cement may simply be too fresh. 
 Cement when tested about twenty-four hours after being 
 ground will often "blow" but if well stored will in a short 
 time become good cement. When the cement being tested 
 is fresh and "blows" there is the possibilit}^ that it may be 
 good cement and in order not to unjustly condemn it, a 
 sufficient quantity should be spread out in a thin layer and 
 then tested again at the end of three or four days. 
 
 Another peculiarity of some Portland cement is that when 
 
< 
 
 ABRASION. 65 
 
 taken from the mill and gauged with water it will be found 
 to be moderately slow setting. Gauge the same cement in 
 24 or 30 hours and it will set almost before it can be mixed. 
 Store in a cool dry place and in a few days it will become 
 more slow setting. 
 
 The older Portland xement is the more slow setting it is, 
 and usually the greater the ultimate strength reached b}^ the 
 mortar. This of course is only true when the cement has 
 been stored in a dry, cool place. 
 
 In regard to slow setting cement. Tradition attributes the 
 property of slow setting to the presence of an excessive 
 amount of lime. This is partly true. That is a cement that 
 has an excessive amount of lime in it, is usually slow setting, 
 but not invariably so. There are however, other factors that 
 affect the setting of cement even more than the presence of 
 an excessive amount of lime. 
 
 Take a cement made with the correct proportions of lime, 
 silica, alumina and iron, and this cement may be either slow 
 or rapid setting, depending, first, upon its age as before noted 
 and, second, upon the degree of calcination to which it has 
 been subjected. 
 
 An under-burned cement will set quicker than an over- 
 burned one. This is true until a point of absolute vitrifaction 
 is reached, when the resulting product will never set. 
 
 Abrasion. — There is one more test to which cement is 
 sometimes subjected, and that is abrasion. Cement that is to 
 be used for sidewalks, floors and artificial stone paving blocks 
 should be subjected to this test. Not only should the neat 
 cement be tested but also the mixture of cement and sand that 
 is to be used in the actual work. 
 
 The resistance of a cement walk to wear depends not only 
 upon the character of the cement used but also the hardness 
 and toughness of the aggregate. 
 
 First the aggregate and then cement, or vice versa, is 
 worn aw^ay, and if the cement is not hard but still has suffi- 
 cient cementing qualities to hold the aggregate in place, the 
 walk may still be durable. 
 
66 
 
 PORTLAND CEMENT. 
 
 The method of making the test for abrasion is the same as 
 that employed with paving bricks, etc. A cube of the neat 
 cement or cement and sand is made. This is allowed to 
 harden under water for not less than seven da3-s. The cube 
 is then taken from the water and thoroughly dried. After 
 drying, it is brushed off with a stiff brush and carefully 
 weighed. 
 
 TJie-gTinding machines are of two kinds. The first, which 
 has been much used in Berlin : a horizontal cast-iron disc that 
 rotates about 22 rotations per minute. The cube, i cm. in 
 size, is held in a clamp and pressed down by a simple lever, 
 weighted with 56 lbs.; 308 grains Napus Quartz No 3 is put 
 on the plate at the start and the same amount at the end of 
 the fifteenth revolution. After 30 revolutions, the cube is 
 taken out and weighed again, and the percentage of loss noted. 
 
 The author has found the following apparatus most satis- 
 factory. The cube used was 3 inches; a coarse emery wheel 
 4 inches wide and 15 inches in diameter. The wheel was set 
 vertically and given 100 revolutions per minute. The cube 
 was loaded by a weight of 10 lbs. at the end of lever 3 feet 
 long. The cube was subjected to 200 revolutions. 
 
 In concluding these remarks upon the testing of Portland 
 cement we wish to say that nothing has done more to improve 
 the quality of Portland cement than a rigid system of tests, and 
 an absolute enforcement of the specifications in regard to tests. 
 
 Cement makers and users soon discovered what degree of 
 excejlf^n^^^ v;p^ pr>s^il>lf>, \yith, fn^'p^^. r^w ,n-'a*^^^':ials. They also 
 soon discovered what were the causes that resulted ni an 
 inferior cement. 
 
 When all of this became known, there was no further excuse 
 for putting poor cement upon the market. Much poor 
 Portland cement was still, however, sold and there existed 
 much irregularit}' in the quality of the output, at different 
 times, from the same works. All of this was merely 'the 
 result of carelessness on the part of the cement makers and 
 this carelessness continued because the users of cements were 
 either ignorant of what they should have received for their 
 
GERMAN GUARANTY. 
 
 67 
 
 money, or because they did not hold the makers down to 
 strict tests and specifications. 
 
 The marked improvement in the output of cement manu- 
 factories began about 1863 or '65, and in Germany. England 
 had been the home of Portland cement since its invention and 
 commanded the markets of the world. In order to open up a 
 place for German cement it was necessary to make some 
 improvement in the product. This was not an affair of much 
 difficulty as the English makers had grown careless. This 
 superiority of production, however, was not in itself sutficient. 
 
 The German makers must not only make a Portland cement 
 better than the English article, but they must make the users 
 of cement appreciate the fact that their cement was the best. 
 
 They accomplished this b}^ simply guaranteeing their ce- 
 ment. That is, they guaranteed that their cement should 
 give not less than certain specified results in certain tests. 
 The two tests were fineness and tensile strength. The test 
 for fineness was a residue of not less than 3 % on a 2,500- 
 mesh sieve, and in tensile strength of not less than 500 lbs. 
 per square inch at the end of 7 days. The briquette being of 
 neat cement. 
 
 This guaranty was entirely optional on the part of the 
 makers, and was done as an advertisement. 
 
 From year to year, however, the effect has been to improve 
 the quality of the cement. 
 
 In order to advance with as much rapidity as possible along 
 the lines of improvement, the Association of German Cement 
 Manufacturers was founded in 1877. They have set forth a 
 set of specifications for the uniform Deliverv and Testing of 
 Portland Cement, and these specifications are now general!}^ 
 accepted. 
 
 They have also defined what material shall be sold under 
 the name of Portland cement, and any member who mixes 
 any foreign material with the raw material before burning, or 
 during burning, or with the cement after burning, and sells 
 the same under the name of Portland cement shall be expelled 
 from the Association. The result, of the work of this Associ- 
 
6S 
 
 PORTLAND CEMENT. 
 
 ation, has been that to-day the factories which belong to it 
 turn out cements that never fall below a certain standard, and 
 this standard is sufficiently high to make any of the cement 
 ;safe for all ordinary uses. It is this uniformity of product, not 
 only in the product of one factory at different times, but in the 
 product of a number of factories at all times, that places the 
 German Portland cements far in advance of any others. 
 
CHAPTER V. 
 
 ABSTRACTS FROM AMERICAN, GERMAN, FRENCH, 
 AND AUSTRIAN SPECIFICATIONS FOR 
 PORTLAND CEMENT. 
 
 American Society of Civil Engineers — Report of 
 THE Committee on a Uniform System 
 FOR Tests of Cement. 
 
 TESTS OF CEMENT. 
 
 The testing of cement is not so simple a process as it is; 
 sometimes thought to be. No small degree of experience is 
 necessary before one can manipulate the materials so as to 
 obtain even approximately accurate results. 
 
 The first tests of inexperienced, though intelligent and care- 
 ful, persons, are usually very contradictory and inaccurate, and 
 no amount of experience can eliminate the variations intro- 
 duced by the personal equations of the most conscientious 
 observers. Many things, apparently of minor importance, 
 exert such a marked influence upon the results, that it is onl}^ 
 by the greatest care in eveiy particular, aided by experience 
 and intelligence, that trustworthy tests can be made. 
 
 The test for tensile strength on a sectional area of one 
 square inch is recommended, because, all things considered, it 
 seems best for general use. For the smaU briquette there is 
 less danger of air bubbles, the amount of material to be 
 handled is smaller, and the machine for breaking may be 
 lighter and less costly. 
 
i-O PORTLAND CEMENT: 
 
 The tensile test, if properly made, is a good, though not a 
 perfect, indication of the value of a cement. The time requi- 
 site for making this test, whether applied to either the natural 
 or the Portland cements, is considerable (at least seven days, 
 if a reasonably reliable indication is to be obtained), and as 
 work is usually carried on is frequently impracticable. For 
 this reason, short time tests are allowable in cases of necessity 
 though the most that can be done in such testing is to 
 determine if the brand of cement is of its average quality. It 
 is beUeved, however, that if a neat cement stands the one day 
 tensile test, and the tests for checking and for fineness, its 
 safety for use will be sufficiently indicated in the case of a 
 brand of good reputation: for, it being proved to be of 
 average quahty, it is fair to suppose that its subsequent condi- 
 tion will be what former experiments, to which it owes its 
 reputation, indicate that it should be. It cannot be said that 
 a new and untried cement will by the same tests be proved to 
 be satisfactory; only a series of tests for a considerable period, 
 and with a full dose of sand, will show the full value of any 
 cement; and it would be safer to use a trustworthy brand 
 without applying any tests whatever than to accept a new 
 article which had been tested only as neat cement and for but 
 one day only. 
 
 The test for compressive strength is a very valuable one in 
 point of fact, but the appliances for crushing are usually 
 somewhat cumbersome and expensive, so much so that it 
 seems undesirable that both tests should be embodied in a 
 uniform method proposed for general adoption. Where great 
 interests are at stake, however, and large contracts for cement 
 depend on the decision of an engineer as to quality, both tests 
 should be used if the requisite appliances for making them are 
 within reach. After the tensile strength has been obtained, 
 the ends of the broken briquettes, reduced to one-inch cubes 
 by grinding and rubbing, should be used to obtain the com- 
 pressive strength. 
 
 The adhesive test being in a large measure variable and 
 uncertain, and therefore untrustworthy, is not recommended. 
 
CEMENT SPECIFICATIONS. 
 FINENESS. 
 
 71 
 
 The strength of a cement depends greatly upon the fineness 
 to which it is ground, especially when mixed with a large 
 dose of sand. It is, therefore, recommended that the tests be 
 made with cement that has passed through a No. 100 sieve 
 (10,000 meshes to the square inch) made of No. 40 wire. 
 Stub's Wire Gauge. The results thus obtained will indicate 
 the grade which the cement can attain, under the condition 
 that it is finely ground, but it does not show whether or not a 
 given cement offered for sale shall be accepted and used. The 
 determination of this question requires that the tests should 
 also be applied to the cement as found in the market. Its 
 quality may be so high that it will stand the tests even if 
 very coarse and granular, and on the other hand, it may be so 
 low that no amount of pulverization can redeem it. In other 
 words, fineness is no sure indication of the value of a cement, 
 although all cements are improved bv fine grinding. Cement 
 of the better grades is now usually ground so fine that only 
 from 5 to 10 % is rejected by a sieve of 2,500 meshes per 
 square inch, and it has been made so fine that only from 3 to 
 10 % is rejected by a sieve of 32,000 meshes per square inch. 
 The finer the cement, if otherwise good, the larger dose of 
 sand it will take, and the greater its value. 
 
 CHECKING OR CRACKING. 
 
 The test for checking or cracking is an important one, and 
 though simple should never be omitted. It is as follows: 
 make two cakes of neat cement, 2 or 3 inches in diameter, 
 about I inch thick, with thin edges. ^ Note the time in 
 minutes that these cakes, when mixed with water to the 
 consistency of a stiff plastic mortar, take to set hard enough 
 to stand the wire test recommended by Gen. Gilmore, -^L-inch 
 diameter wire loaded with J- of a pound, and J^-inch loaded 
 with I pound. One of these cakes, when hard enough, should 
 be put in water and examined from day to day to see if it 
 becomes contorted, or if cracks show themselves at the edges, 
 such, contortions or cracks indicating that the cement is unfit 
 
72 
 
 PORTLAND CEMENT. 
 
 for use at that time. In some cases the tendency to crack, if 
 caused by the presence of too much unslaked Hme, will 
 disappear with age. The remaining cake should be kept in 
 the air and its color observed, which, for a good cement, 
 should be uniform; the Portland cements being of a bluish- 
 gray throughout, yellowish blotches indicating a poor quality; 
 and the natural cements being light or dark, according to the 
 character of the rock of which they are made. The color of 
 the cements when left in the air indicates the qualit}" much 
 better than when they are put in water. 
 
 TESTS RECOMMENDED. 
 
 It is recommended that tests for h3'draulic cement be con- 
 fined to methods for determining fineness, liability to checking- 
 or cracking, and tensile strength; and for tiie latter, for tests 
 of seven days and upward, that a mixture of one part of 
 cement to one part of sand for natural cements, and three 
 parts of sand for Portland cements, be used, in addition to 
 trials of the neat cement. The quantities- used in the mixture 
 should be determined by weight. 
 
 The tests should be applied to the cements as offered for 
 sale. If satisfactory results are obtained with a full dose of 
 sand, the trials need go no further. If not, the coarser 
 particles should first be excluded by using a No. lOO sieve, in 
 order to determine approximately the grade the cement would 
 take if ground fine; for fineness is always attainable, while 
 inherent merit ma}^ not be. 
 
 Note. — Your committee thinks it useful to insert b ere a table showing the 
 average minimum and maximum tensile strength per square inch which 
 some good cements have attained when tested under the conditions speci- 
 fied elsewhere in this report. Within the limits given in the following- 
 table, the value of a cement varies closely with the tensile strength when 
 tested with the full dose of sand: 
 
 American JVaiural Cement, neat: 
 
 One day; one hour, or until set, in air, the rest of the 24 
 hours in water, from 40 pounds to 80 pounds. 
 
 One week; one day in air, 6 days in water, from 60 pounds, 
 to 100 pounds. 
 
CEMENT SPECIFICATIONS. 73, 
 
 One month (28 daj'-s) ; one day in air, 27 days in water,, 
 from 100 pounds to 150 pounds. 
 
 One year; one day in air, the remainder in water, from 300- 
 pounds to 400 pounds. 
 
 American and .Forei'^n Portland Cements, neat: 
 
 One da}^; one hour, or until set, in air, the rest of the 24 
 hours in water, from 100 pounds to 140 pounds. 
 
 One week; one day in air, 6 days in water, from 250 
 pounds to 550 pounds. 
 
 One month (28 days); one day in air, 27 days in water, 
 from 350 pounds to 700 pounds. 
 
 One year; one day in air, the remainder in water, from 450 
 pounds to 800 pounds. 
 
 American Natural Cements, one -part of Cement to 
 one part of Sand: 
 
 One week; one day in air, 6 days in water, from 30 pounds- 
 to 50 pounds. 
 
 One month (28 days); one day in air, 27 days in water,, 
 from 50 pounds to 80 pounds. 
 
 One year; one day in air, the remainder in water, from 200 
 pounds to 300 pounds. 
 
 American and Foreign Portland Cements, one part of 
 Cement to three parts of Sand: 
 
 One week; one day in air, 6 days in water, from 80 
 pounds to 125 pounds. 
 
 One month (28 days); one day in air, 27 days in water, 
 from 100 pounds to 200 pounds. 
 
 One year; one day in air, the remainder in water, from 200 
 pounds to 350 pounds. 
 
 Standards of minimum fineness and tensile strength for 
 Portland cement, as given below, have been adopted in some 
 foreign countries. 
 
PORTLAND CEMENT. 
 
 /;/ Germany, by Berlin Society of Architects, Society of Manu- 
 facturers of Bricks, Lime, and Cement, Society 
 of Contractors, and Society of German 
 Cement Makers : 
 
 Standard of 1877. — Fineness, not more than 25 % to be left 
 on sieve of 5,806 meshes per square inch. ' 
 
 Tensile strength, i part cement, 3 parts sand, i day in air, 
 27 days in water, 113.78 pounds per square inch. 
 
 Standard of 1878. — Fineness, not more than 20 % to be left 
 on the sieve, as above. 
 
 Tensile strength, same mixture and time as above, 142.23 
 pounds per square inch. 
 
 In Austria, by Austrian Association of Engineers and 
 Architects : 
 
 Standard of 1878. — Fineness same as German of 1878. 
 
 Tensile strength, same mixture as above, 7 days, i day in 
 air, 6 days in water, 113.78 pounds per square inch. 
 
 Twent3-eight days, i day in air, 27 days in water, 170.68 
 pounds per square inch. 
 
 In Austria a standard for the minimum fineness and tensile 
 strength of Roman cement was established and generally 
 accepted, as follows: 
 
 Standard of 1878. — Fineness, same as Portland. 
 Tensile strength (i part of cement, 3 parts of sand) for — 
 Quick setting (taking 15 minutes or less to set) : 
 Seven days, i day in air, 6 days in water, 23 pounds per 
 square inch. 
 
 Twenty-eight days, i da}^ in air, 27 da3-s in water, 56.9 
 pounds per square inch. 
 
 Slow setting cement (taking more than 15 minutes to set) : 
 
 Seven days, one day in air, 6 days in water, 42.6 pounds 
 per square inch. 
 
 Twenty-eight days, one day in air, 27 days in water, 85.3 
 pounds per square inch. 
 
CEMENT SPECIE^ICATIONS. 75 
 
 The Roman cements correspond to those classified in this 
 report under the head of Natural Cements. 
 
 Standards have been adopted also in Sweden and Russia. 
 
 MIXING, ETC. 
 
 The proportions of cement, sand, and water should be 
 carefully determined by weight, the sand and cement mixed 
 dry, and all the water added at once. The mixing must be 
 rapid and thorough, and the mortar, which should be stiff 
 and plastic, should be firmly pressed with a ti-owel, without 
 ramming, and struck off level; the molds in each instance, 
 while being charged and manipulated, to be laid directly on 
 glass, slate, or some other non-absorbent material. The 
 molding must be completed before incipient setting begins. 
 As soon as the briquettes are hard enough to bear it, they 
 should be taken from the molds and kept covered with a 
 damp cloth until they are immersed. For the sake of unir 
 formity, the briquettes, both of neat cement and those con- 
 taining sand, should be immersed in water at the end of 
 twenty-four hours, except in the case of one day tests. 
 
 Ordinary, fresh, clean water, having a temperature between 
 60" and 70^ Fah. should be used for water of mixture and 
 immersion of samples. 
 
 The proportion of water required varies with the fineness, 
 age, or other conditions of the cement, and the temperature 
 of the air, but is approximately as follows : for briquettes of 
 neat cement, Portland, about 25 ; natural, about 30 %. For 
 briquettes of one part cement, one part sand, about 15 % of 
 total weight of sand and cement. For briquettes of one part 
 cement, three parts sand, about 12 % of total weight of sand 
 and cement. The object is to produce the plasticity of rather 
 stiff plasterer's mortar. 
 
 An average of five briquettes may be made for each test, 
 only those breaking at the smallest section to be taken. The 
 briquettes should alwa3^s be put in the testing machine and 
 broken immediately after being taken out of the water, and 
 the temperature of the briquettes and of the testing room 
 should be constant between 60'' and 70" Fah. 
 
y5 PORTLAND CEMENT. 
 
 The Stress should be appHed to each briquette at a uniform 
 rate of about 400 pounds per minute, starting each time at o. 
 With a weak mixture one-half the speed is recommended. 
 
 WEIGHT. 
 
 The relation of the weight of cement to its tensile strength 
 is an uncertain one. In practical work, if used alone, it is of 
 little value as a test, while in connection with the other tests 
 recommended it is unnecessary, except when the relative bulk 
 of equal weights of cements is desired. 
 
 We recommend that the cubic foot be substituted for the 
 bushel as the standard unit, whenever it is thought best to 
 use this test. 
 
 SETTING. 
 
 The rapidity with which a cement sets or loses its plasticit}^ 
 furnishes no indication of its ultimate strength. It simply 
 shows its initial hydraulic activity. 
 
 For purposes of nomenclature, the various cements may be 
 'divided arbitrarily into two classes, namely; quick setting, or 
 those that set in less than -|- an hour; and slow setting, or those 
 i-equiring .V an hour or more to set. The cement must be 
 adapted to the work required, as no one cement is equally 
 good for all purposes. For submarine work a quick setting 
 cement is often imperatively demanded, and no other will 
 answer, while for work above the water-Hne less hydraulic 
 activity will usually be preferred. Each individual case 
 demands special treatment. The slow setting natural cements 
 should not become warm while setting, but the quick setting 
 one may, to a moderate extent, within the degree producing 
 cracks. Cracks in Portland cement indicate too much car- 
 bonate of lime, and in the Vicat cements too much lime in the 
 original mixture. 
 
 SAMPLING. 
 
 There is no uniformity of practice among engineers as to 
 the sampling of the cement to be tested, some testing every 
 tenth barrel, others every fifth, and others still every barrel 
 delivered. Usually, where cement has a good reputation, and 
 
CEMENT SPECIFICATIONS. 
 
 77 
 
 is used in large masses, such as concrete in heavy foundations, 
 or in the backing or hearting of thick walls, the testing of 
 every fifth barrel seems to be sufficient; but in very important 
 work, where the strength of each barrel may in a great 
 measure determine the strength of that portion of the work 
 where it is used, or in the thin walls of sewers, etc., etc., every 
 barrel should be tested, one briquette being made from it. 
 
 In selecting cement for experimental purposes, take the 
 samples from the interior of the original packages, at sufficient 
 depth to insure a fair exponent of the quality, and store the 
 ♦ same in tightly closed receptacles impervious to light or 
 dampness until required for manipulation, when each sample 
 of cement should be so thoroughly mixed, by sifting or other- 
 wise, that it shall be uniform in character throughout its mass. 
 
 SIEVES. 
 
 For ascertaining the fineness of cement, it will be convenient 
 to use three sieves, viz. : 
 
 No. 50 (2,500 meshes to the square inch), wire to be of 
 No. 35 Stub's Wire Gauge. 
 
 No. 74 (5,476 meshes to the square inch), wire to be of 
 No. 37 Stub's Wire Gauge. 
 
 No. 100 (10,000 meshes to the square inch), wire to be of 
 No. 40 Stub's Wire Gauge. 
 
 The object is to determine b}' weight the percentage of 
 each sample that is rejected by these sieves, with a view not 
 only of furnishing the means of comparison between tests 
 made of different cements by different observers, but indicating 
 to the manufacturer the capacity of his cement for improve- 
 ment in a direction always and easily within his reach. As 
 already suggested in another connection, the tests for tensile 
 strength should be applied to the cement as offered in the 
 market, as well as to that portion of it which passess the No. 
 100 sieve. 
 
 For sand, two sieves are recommended, viz. : 
 No. 20 (400 meshes to the square inch), wire to be of No. 
 28 Stub's Wire Gauge. 
 
78 PORTLAND CEMENT. 
 
 No. 30 (900 meshes to the square inch), wire to be of No. 
 31 Stub's Wire Gauge. 
 
 These sieves can be furnished in sets, as follows, an arrange- 
 ment having been made wdth a manufacturer of such articles, 
 by which he agrees to furnish them of the best quality of 
 brass wire cloth, set in metal frames, the cloth to be as true 
 to count as it is possible to make it, and the wire to be of the 
 required gauge. Each set wall be enclosed in a box, the sieves 
 being nested. 
 
 Set A, three cement sieves, to cost $4.80: 
 
 No. lOO, . . . 7 in. diameter. 
 
 No. 74, . . . 61 
 
 No. 50, . . . 6 " " 
 
 Set B, two sand sieves, to cost $4.00: 
 
 No. 30, ... 8 in. diameter. 
 No. 20, . . . 7I- " " 
 
 STANDARD SAND. 
 
 The question of a standard sand seems one of great impor- 
 tance, for it has been found that sands looking ahke and sifted 
 through the same sieves give results varying within rather 
 wdde limits. 
 
 The material that seems hkely to give the best results is 
 the crushed quartz used in the manufacture of sand paper. It 
 is a commercial product, made in large quantities and of 
 standard grades, and can be furnished of a fairly uniform 
 quality. It is clean and sharp, and although the present price 
 is somewhat excessive (3 cents per pound), it is believed that 
 it can be furnished in quantity for about $5.00 per barrel of 
 300 pounds. As it would be used for tests only, for purposes 
 of comparison with the local sands, and with tests of different 
 cements, not much of it would be required. The price of the 
 German standard sand is about $1.25 per 112 pounds, but the 
 article being washed river sand is probably inferior to crushed 
 quartz. Crushed granite can be furnished at a somewhat less 
 rate than quartz, and crushed trap for about the same as 
 granite, but no satisfactory estimate has been obtained for 
 
CEMENT SPECIFICATIONS. 
 
 either of these. The use of crushed quartz is recommended 
 by your committee, the degree of fineness to be such that it 
 will all pass a No. 20 sieve and be caught on a No. 30 sieve. 
 Of the regular grade, from 15 to 37 % of crushed quartz No. 
 3 passes a No. 30 sieve, and none of it passes a No. 50 sieve. 
 As at present furnished, it would need resifting to bring it to 
 the standard size; but if there were sufficient demand to 
 warrant it, it could undoubtedly be furnished of the size of 
 grain required at little, if any, extra expense. 
 
 A bed of uniform, clean sand of the proper size of grain 
 has not been found, and it is believed that to wash, dry, and 
 sift any of the available sands, would so greatly increase its 
 cost that the product would not be much cheaper than the 
 crushed quartz, and would be much inferior to it in sharpness 
 and uniform hardness of particles. 
 
 MOLDS. 
 
 The molds furnished are usually of iron or brass, the price 
 of the former being $2, and of the latter $3 each. Wooden 
 molds, if well oiled to prevent their absorbing water, answer 
 a good purpose for temporary use, but speedily become unfit 
 for accurate work. A cheap, durable, accurate, and non- 
 corrodible mold is much to be desired. 
 
 CLIPS. 
 
 For using the clips recommended in the preliminary report 
 it was found in some instances that the specimens were 
 broken at one of the points where they were held. This was 
 undoubtedly caused by the insufficient surface of the clip, 
 which, forming a blunt point, forced out the material. Where 
 the specimens were sufficiently soft to allow this point to be 
 imbedded, they broke at the smallest section, but when hard 
 enough to resist such imbedding, they showed a wedge- 
 shaped fracture at the clips. To remedy this, the point 
 should be slightly flattened so as to allow of more metal 
 surface in contact with the briquette. Cfips made in this way 
 have been used, and good results obtained. 
 
 To adapt the one-inch clips of the Riehle machine, only a.. 
 
So 
 
 PORTLAND CEMENT. 
 
 slight amount of work is necessary; tlie ends being rounded, 
 will admit the proposed new form of briquette, and yet not 
 prevent the use of the old one, thus allowing comparative 
 tests of the two forms to be made without changing the clips. 
 
 There should be a strengthening rib upon the outside of 
 the clips to prevent them from bending or breaking when the 
 specimens are very strong. 
 
 The clips should be hung on pivots so as to avoid as much 
 as possible cross strain upon the briquettes. 
 
 MACHINES. 
 
 No special machine has been recommended, as those in 
 common use are of good form for accurate work, if properly 
 used, though in some cases they are needlessly strong and 
 expensive. Machines with spring balances are to be avoided 
 as more liable to error than others. 
 
 It is by no means certain that there exists any great 
 difference in well made machines of the standard forms given. 
 
 AMOUNT OF MATERIAL. 
 
 The amount of material needed for making five briquettes 
 of the standard size recommended is: for the neat cements, 
 about one and two-thirds pounds; and for those with sand, in 
 the proportion of three parts of sand to one of cement, about 
 one and one-quarter pounds of sand and six and two-thirds 
 ounces of cement. 
 
 German Specifications for Standard Portland 
 Cement Tests. 
 
 Definition. Portland cement is a product resulting from the 
 vitrifaction of a thorough mixture of material, whose principal 
 component parts are Hme and alumina, and the grinding of 
 the vitrified material to a fine powder. 
 
 I. Packing and Weight: As a rule Portland cement is 
 to be packed in standard barrels of i8o kilo. (397 lbs.), gross 
 weight, and about 170 kilo. (374 lbs.), net weight, and in 
 ihalf standard barrels of 90 kilo. (198 lbs.), gross weight, and 
 iibout 83 kilo. (133 lbs.), net weight. The gross weight is 
 
CEMENT SPECIFICATIONS. 
 
 8i 
 
 to be marked on the barrels. If the cement is called for in 
 bags or barrels of other weight, the gross weight of the same 
 must be clearly marked upon these packages. Losses and 
 variations in weight of the single packages up to 2 % of the 
 same will be allowed. Barrels and sacks, in addition to the 
 weight shall show in legible writing the name and trade mark 
 of the manufacturer. 
 
 2. Time of Setting: Slow or quick setting cement may 
 be called for according to the use for which the cement is to 
 be put. Cements which do not set in less than 2 hours, are 
 to be considered slow setting cements. 
 
 3. Constancy of Volume: Portland cement shall be of 
 constant volume. As a preliminary test, admitting of forming 
 a rapid opinion, the heating test is recommended. The 
 decisive test shall be that a paste of neat cement made on a 
 glass plate protected against drying and placed under water 
 after 24 hours, shall not show after the lapse of a longer period 
 of time any blowing cracks, or change of shape. 
 
 4. Fineness of Grinding: Portland cement shall be so 
 finely ground that a batch of the same shall not leave a residue 
 of more than 10 % upon a sieve of 900 meshes per square 
 centimeter (5,806 meshes per sq. in.). The thickness of the 
 wire of the sieve shall equal half the space between the wires. 
 For test 100 g. (3I- oz.) of cement shall be used. 
 
 5. Tests of Strength: The cohesive power of Portland 
 cement shall be determined by the testing of a mixture of 
 cement and sand. The tests shall be both tensile and com- 
 pressive, made according to a uniform method, with test 
 pieces cf the same form and cross section, and with the same 
 apparatus. At the same time a determination of the strength 
 of the neat cement is to be recommended. 
 
 6. Tensile and Compressive Strength: Good slow setting 
 cement, in the proportion of three parts by weight of standard 
 sand to one part of cement shall have when tested, after 28 
 days' hardening (i in air and 27 in water), a minimum tensile 
 strength of at least 16 kilo. q. c. m. (16 kilogrammes per sq. 
 centimeter) (227 lbs. per sq. in.). The compressive strength 
 shall be at least 160 kilo. q. c. m. (2,270 lbs. per sq. m.). 
 
82 
 
 PORTLAND CEMENT. 
 
 Cement which shows a higher tensile or compressive 
 strength admits in many cases of a greater addition of sand, 
 and from this point of view, as well as on account of its greater 
 strength for the same amount of sand, is entitled to a corre- 
 spondingly higher price. 
 
 For slow setting cements the strength after 20 days is less 
 in general than the one above specified, therefore, in giving 
 the results of tests, the time of setting shall also be given. 
 
 The tests shall be made in the following manner : 
 
 To determine the time of setting cement, a slow setting 
 neat cement shall be mixed 3 minutes, and a quick setting 
 neat cement i minute with water to a stiff paste. A cake 
 about 1.5 c. m. (0.59 in.) thick, with thin edges, shall be 
 formed of this paste on a plate of glass. The consistency of 
 the cement paste for this cake shall be such, that when 
 brought with a trowel on the plate, the paste will only begin 
 to run toward the edges of the same after the paste has been 
 repeatedly jarred. As a rule 27 % to 30 % water will suffice 
 to give the necessary consistency to the paste. As soon as 
 the cake is sufficiently hardened, so that it will resist a slight 
 pressure of the finger nail, the cement is to be considered as. 
 having set. 
 
 For the exact determination of the time of setting, and for 
 determining the beginning of the time of setting, which latter 
 is of importance in the case of quick setting cements, since 
 they must be worked up before they begin to set, a standard 
 needle 300 g. (10 oz.) in weight, and i sq. mm. ( .00155 sq.. 
 in.) in cross section, is used. A metal ring 4 cm. ( 1-575 i"-) 
 in height and 8 cm. (3.15 in.) clear diameter (inside diameter) 
 is placed on a glass plate, filled with cement paste of the 
 above consistency and brought under the needle. The moment 
 at which the needle is no longer capable of completely pen- 
 etrating the cement cakes is considered the beginning of the 
 time of setting. The time elapsing between this and the 
 moment when the standard needle no longer leaves an appre- 
 ciable impression on the hardened cake is considered the time 
 of setting. 
 
CEMENT SPECIFICATIONS. 83 
 
 For making the heat test (3)3 stiff paste of neat cement 
 and water is made, and from this cakes 8 cm. (3.15 in.) to 
 10 cm. (3-94 in-) in diameter and i cm. (.394 in.) thick 
 are formed on a smooth, impermeable plate, covered with 
 blotting paper. Two of these cakes, which are to be pro- 
 tected against drying, in order to prevent drying cracks, are 
 placed, after the lapse of twenty-four hours, or at least only 
 after they have set, with their smooth surfaces on a metal 
 plate and exposed, for at least one hour, to a temperature of 
 from 110° C. to 120° C. (230° to 248" F.) until no more 
 water escapes. For this purpose the drying closets in use in 
 chemical laboratories may be utilized. If, after this treatment, 
 theVakes show no edge cracks, the cement is to be consid- 
 ered in general of constant volume. If such cracks do appear, 
 the cement is not to be condemned, but the results of the 
 decisive test with the cakes hardening on glass plates under 
 water must be waited for. It must, however, be noticed that 
 the heat test does not admit of a final conclusion as to the 
 constancy of volame of those cements which contain more 
 than 3 % of calcium sulphate (gypsum) or other sulphur 
 combinations. 
 
 For making the final test, the cake made for the purpose of 
 determining the time of setting, for slow setting cements, is 
 placed under water after the lapse of twenty-four hours, but, at 
 all events, not until after it is set. For quick setting cements 
 this can be done after a shorter period. The cakes, especially 
 those of slow setting cements, must be protected against 
 draughts and sunshine until their final setting. This is best 
 accomplished by keeping them in a covered box lined with 
 zinc, or under wet cloths. In this manner the formation of 
 heat cracks is avoided, which are generally formed in the 
 center of the cake and may be taken by an inexperienced 
 person for cracks formed by blowing. 
 
 In order to obtain concordant results in the tests, sand of 
 uniform size of grain and uniform quality must be used. This 
 standard sand is obtained by washing and drying the purest 
 quartz sand obtainable, sifting the same through a sieve with 
 
PORTLAND CEMENT. 
 
 60 meshes per square centimeter (387 per sq. in.), thereby 
 separating the coarsest particles, and by removing from the 
 sand so obtained, by means of a sieve of 120 meshes per 
 square centimeter (774 per sq. in.), the finest particles. The 
 diameter for the wires of the sieves shall be 0.38 mm. and 
 0.32 mm. (.015 in. and .013 in.) respectively. Since not all 
 quartz sand, even under the same method of treatment, gives 
 the same resulting strengths in the mortars, one must know 
 w^iether the standard sand at one's disposal gives concordant 
 results with the standard sand furnished by the German 
 Society of Cement Manufacturers and also used at the Royal 
 Testing Station at Berlin (Charlottenburg). 
 
 For each test, in order to obtain correct average results, 
 at least 6 test pieces are to be made. Tensile test pieces can 
 be made made either by hand or by machinery. 
 
 HAND WORK. 
 
 On a metal or thick glass plate 5 sheets of blotting paper 
 soaked in water are laid, and on these are placed 5 molds 
 wetted with water; 250 grammes (8.75 oz.) of cement and 
 750 grammes (26.25 oz.) of standard sand are weighed and 
 thoroughly mixed dry in a vessel. Then 100 cubic centimeters 
 (100 g. or 35 oz.) of fresh water are added, and the whole 
 mass thoroughly mixed for 5 minutes. With the mortar so 
 obtained the molds are at once filled, with one filKng, so high 
 as to be rounded on top, the mortar being well pressed in. 
 By means of an iron trowel 5 to 8 centimeters (1.96 in. to 3.14 
 in.) wide, 35 centimeters (13.79 in.) long, and weighing about 
 250 grammes, (8.75 oz.) the projecting mortar is pounded 
 first gently and from the side, then harder into the molds until 
 the mortar grows elastic, and water flushes to the surface. 
 A pounding of at least one minute is absolutely essential. An 
 additional filling and pounding in of the mortar is not admis- 
 sible, since the test pieces of the same cement shall have the 
 same densities at the different testing stations. The mass 
 projecting over the mold is now cut off with a knife, and the 
 surface smoothed. The mold is carefully taken off and the 
 test piece placed in a box lined with zinc, which is to be 
 
CEMENT SPECIFICATIONS. 85 
 
 provided with a cover, to prevent a non-uniform drying of 
 the test pieces at different temperatures. Twenty-four horirs 
 after being made, the test pieces are placed under water, and" 
 care has to be taken that they remain under w^ater during 
 the whole period of hardening. 
 
 MACHINE WORK. 
 
 After the mold, provided with a guide mold, has been 
 clamped, by means of set screws, on the bed-plate of the 
 pounding machine, for each test, 180 grammes (6.3 oz.) of 
 the mortar, made as above, are placed in the mold and the iron 
 follower is set in. B3' means of Bohme's hammer apparatus, 
 with a hammer weighing 2 kilogrammes, (4.4 lbs.), 150 blows 
 are struck on the follower. 
 
 After the guide mold and follower have been removed, the 
 test piece is scraped off, smoothed, taken with the mold from 
 the bed-plate and for the rest treated as for the hand work. 
 By accurately following the directions given above, hand and 
 machine work give well concording results. In all cases of 
 doubt the machine work is to be decisive. 
 
 COMPRESSIVE TESTS. 
 
 In order to obtain concordant values in compression tests at 
 different stations, machine making is necessary. Four hundred 
 grammes (14 oz.) of neat cement, and 1,200 grammes (42 
 oz.) dry standard sand are thoroughly mixed dry in a vessel, 
 and 160 cubic centimeters (5.6 oz.) of water are added thereto, 
 and then the mortar is thoroughly mixed for 5 minutes. Of 
 this mortar 850 grammes (30 oz.) are placed in the cubic 
 molds, provided with guide mold, and the mold is then screwed 
 on the bed-plate under the pounding machine. The iron fol- 
 lower is placed in the form and by means of Bohme's trip 
 hammer, 150 blows are struck, by a hammer weighing 2 
 kilogrammes (4.4 lbs.). 
 
 After removing the guide mold and follower, the test piece 
 is smoothed off, with the mold from the bed-plate, and for the 
 rest treated as for hand work, as given above. 
 
86 
 
 PORTLAND CEMENT. 
 
 MAKING TEST PIECES OF NEAT CEMENT. 
 The inside of the molds is slightly oiled, and the same are 
 placed on a metal or glass plate without blotting paper. One 
 thousand grammes (35 ozs.) of cement are weighed out, two 
 hundred grammes (7 ozs.) of water are added, and the whole 
 mass thoroughly mixed for five minutes (best with pestle). 
 The forms are well filled (rounded), and then proceed as for 
 hand work, as given above. The molds can only be taken 
 off after the cement has sufficiently hardened. Since, by the 
 pounding in of the neat cement, test pieces of uniform consist- 
 ency are to be obtained, for finely ground or quick setting 
 cements, the amount of water must be correspondingly in- 
 creased. The volume of water used is always to be stated 
 in giving the strength obtained. 
 
 TREATMENT OF TEST PIECES AT TIME OF TESTING. 
 
 All specimens are to be tested directly after their removal 
 from the water. Since the time of testing is of influence on 
 the result in tensile tests, the increase of load shall be one 
 hundred grammes (3.5 oz.) per second. The mean of the 
 four best results shall be considered the final tensile strength. 
 In testing compression pieces, the pressure is always to be 
 exerted on two side faces of the cube, but not on the bottom 
 or top. The mean of the four highest tests shall be consid- 
 ered as the final compressive strength. 
 
 Abstracts from French Specifications for 
 Portland Cement. 
 
 chemical analysis. 
 The cement must not contain more than i % of sulphuric 
 acids or sulphides in determinable proportion. Cements con- 
 taining more than 4 % of ferric oxide, or in which the ratio of 
 the combined silica and alumina to the lime is less than 0.44, 
 are to be regarded as doubtful. 
 
 MIXING THE MORTAR. 
 In mixing the mortar for testing, sea water is specified, and 
 
CEMENT SPECIFICATIONS. 87 
 
 both air and water are to be maintained at a temperature of 
 15° to 18" C. (59° to 64.4° F.) during the continuance of the 
 experiments. The quantity of water is ascertained by a 
 prehminary experiment, and the four following tests are given 
 to . serve as an indication whether the proportion of water 
 added is correct : 
 
 1. The consistence of the mortar should not change if it 
 be gauged for an additional period of three minutes after the 
 initial five minutes. 
 
 2. A small quantity of the mortar dropped from the trowel 
 upon the marble slab from a height of about 0.50 metre 
 (1.64 ft.j should leave the trowel clean, and retain its form 
 approximately without cracking. 
 
 3. A small quantity of the mortar worked gently in the 
 hands should be easily molded into a ball, on the surface of 
 which w^ater should appear. When this ball is dropped from 
 a height of 0.50 m.etre (1.64 ft.) it should retain a rounded 
 shape without cracking. 
 
 4. If a slightly smaller quantity of water be used, the 
 mortar should be crumbly, and crack when dropped upon the 
 slab. On the other hand, the addition of a further quantity of 
 water — i to 2 '^0 of the weight of the cement — would soften 
 the mortar, rendering it more adhesive, and preventing it 
 from retaining its form when allowed to fall upon the slab. 
 It is recommended to commence with a rather smaller quan- 
 tity of water than may be ultimately required, and then to 
 make fresh mixings with a slight additional quantity of w^ater. 
 
 The mortar is to be mixed with a trowel for five minutes 
 upon a marble slab. 
 
 STRENGTH. 
 
 The form of briquette and method of molding are the same 
 as required by the German specifications; the breaking sec- 
 tion is 5 square centimetres (0.775 sq. in.). Six briquettes 
 are broken after an interval of 7 days, six after 28 days, and 
 the remaining six after 84 days. The mean of the three 
 highest figures of each series of tests is taken as the tensile 
 
88 
 
 PORTLAND CEMENT. 
 
 Strength of the cement under examination. The minimum 
 strength specified for the neat cement in 7 days is 20 kilo- 
 grammes per square centimetre (284.5 lbs. per sq. in.); in 
 28 da3''s, 35 kilogrammes per square centimetre (49. 8 lbs. per 
 sq. in.); and at least 45 kilogrammes per square centimetre 
 (640 lbs. per sq. in.) in 84 days. If, however, the strength 
 in 28 days is not more than 5 kilogrammes per square centi- 
 metre (71.12 lbs. per sq. in.) in excess of that at 7 days, then 
 it must be at least 55 kilogrammes per square centimetre 
 (782.27 lbs. per sq. in.) in 28 days, and in any case where 
 this strength is not attained at 28 days it must be exceeded in 
 84 days 
 
 Tests of cement mixed with sand are also specified. The 
 standard sand is produced by crushing quartzite obtained 
 from quarries near Cherbourg, and sifting it through sieves 
 of 64 and 144 meshes per square centimetre (413 and 929 
 meshes per sq. in.). That which remains between these two 
 sieves is washed and dried, and constitutes the standard sand. 
 Three hundred and seventy-five grammes (13.25 oz.) of this 
 sand is mixed v/ith 125 grammes (4.41 oz.) of cement, and 
 water is added in the proportion of 12 parts by weight to 100 
 parts of sand and cement combined. The sand and cement 
 are first carefully mixed in a basin or capsule, then the whole 
 of the sea water is added at once, and the mixture stirred with 
 a spatula for 5 minutes. At the expiration of 7 days the 
 strength of the sand-cement briquettes should be at least 8 
 kilogrammes per square centimetre (113.78 lbs. per sq. in.),, 
 and in 28 days 15 kilogrammes per square centimetre (213.35 
 lbs. per sq. in.). In 28 days the strength should exceed that 
 at 7 da3/s by 2 kilogrammes per square centimetre (28.45 lbs. 
 per sq. in.). In 84 days the strength must be greater than at 
 28 days, and at least 18 kilogrammes per square centimetre 
 (256 lbs. per sq. in.). The 84-day tests are only considered 
 indispensable for those cements which may have stood the 
 two previous tests; but if, while the cement is in store, the 
 84-day tests should be unsatisfactory, it may be rejected. 
 
CEMENT SPECIFICATIONS. 89 
 FINENESS OF GRINDING. 
 The degree of fineness to which the cement must be ground 
 is not specified, it being considered that very fine grinding 
 increases the strength chiefly during the duration of the tests, 
 and that the subsequent increase of strength is less with fine 
 than with coarse cement. 
 
 , TIME OF SETTING. 
 
 Essentially the same as the German specifications. Any 
 cement commencing to set in less than 30 minutes, or failing 
 to commence to set within 3 hours, is to be rejected; and the 
 final set must have taken place within 12 hours. In each case 
 the time is reckoned from the moment the water is poured 
 upon the cement. 
 
 Austrian Specifications for Fineness and 
 Strength of Cement. 
 
 FOR PORTLAND. 
 
 Fineness, not more than 20 % to be left on sieve of 5,806 
 meshes per square inch. Tensile strength (i part cement 
 and 3 parts sand), i day in air and 6 in water, 113.78 lbs. per 
 sq. in.; i day in air and 27 in water, 170.68 lbs. per sq. in. 
 
 FOR ROMAN. 
 
 Fineness, same as for PorUand. Tensile strength (i part 
 cement and 3 parts sand), for quick setting cements (taking 
 15 minutes, or less, to set), i day in air and 6 days in water, 
 23 lbs. per sq. in.; i day in air and 27 in water, 56.9 lbs. per 
 sq. in. For slow setting cements (taking more than 15 min- 
 utes to set), I day in air and 6 days in water, 42.6 lbs. per 
 sq. in.; i day in air and 27 in water, 85.3 lbs. per sq. in. 
 
 English Specifications for Portland Cement. 
 
 The following is a summary of the specifications used by 
 Mr. Henry Faija, an accepted English authority: 
 
 Fineness is to be such that the cement will all pass through 
 
■po PORTLAND CEMENT. 
 
 a sieve having 625 holes (25') to the square inch, and leave 
 only 10 % residue when sifted through a sieve having 2,500 
 holes (50') to the square inch. 
 
 EXPANSION OR CONTRACTION. 
 
 A pat made and submitted to moist heat and warm water 
 at a temperature of about 100" F. shall show^ no sign of 
 blowing in 24 hours. 
 
 TENSILE STRENGTH. 
 
 Briquettes of slow setting Portland, which have been gauged, 
 treated, and tested in the prescribed manner, to carry an aver- 
 age tensile strain, without fracture, of at least 176 lbs. per sq. 
 in. at the expiration of 3 days from guaging; and those tested 
 at the expiration of 7 days to show an increase of at least 50 % 
 over the strength of those at 3 days, but to carry a minimum 
 of 350 lbs. per sq. in. 
 
 For quick setting Portland, at least 176 lbs. per sq. in. at 3 
 days, and an increase at 7 days of 20 to 25 but a minimum 
 of 400 lbs. per sq. in. Very high tensile strengths at early 
 dates generally indicate a cement verging on an unsound one. 
 
CHAPTER VI. 
 
 THE CHEMICAL PROCESSES CONCERNED IN THE 
 HARDENING OF HYDRAULIC CEMENTS. 
 
 By LAUNCELOT ANDREWS, Ph.D. 
 
 L 
 
 CLASSIFICATION. 
 
 Under the general term "cement" the engineer under- 
 stands certain mixtures which possess the property in common 
 of hardening to a strong mass when brought into contact with 
 water. We may conveniently divide the practically important 
 cements into three groups: first, those containing silicates as 
 an essential constituent, such as Portland cement and the 
 hvdrauhc limes; second, those consisting chiefly of neutral 
 salts capable of combining with water of crystallization to 
 form a hard mass, such as plaster of Paris, either used alone 
 or together with alum, etc.; third, substances which form 
 basic salts when treated with water, such as mixtures of 
 magnesia with magnesium chloride, or of plaster of Paris with 
 lime; fourth, mortar, or mixtures of lime with substances such 
 as sand, upon which it has no action. In this last class, the 
 absorption of carbonic acid from the air plays an important 
 part in the final hardening. 
 
 IL 
 
 THE CAUSE OF THE "SETTING" OF CEMENTS IN GENERAL 
 AND CONDITIONS AFFECTING IT.* 
 
 The setting of a cement is, in general, a complex process. 
 
 *The following section of this article is witli minor changes reprinted 
 from a paper by Andrews and Spanutius published in Vol. I, No. 2, p. 41 
 of the Transit, and now out of print. 
 
92 PORTLAND CEMENT. 
 
 partly chemical in its nature and in part mechanical. Broadly 
 stated, the chemical changes which occm- may be said rather 
 to afford opportunity' for the mechanical changes which result 
 in hardening, than themselves to cause the 'hardening. The 
 chemical changes are therefore susceptible of wide variation 
 without materially influencing the result. 
 
 In all cases a dry cement consists of grains with interspaces 
 between them and in contact at but few points. When treated 
 with water, a portion of these grains dissolves and is then 
 redeposited, in combination with water, in a crystalline form, 
 between the grains, binding them together at numerous points 
 of contact. 
 
 The solidity of the resulting mass depends ;_;fr5/!", on the area 
 of contact between the grains; second, on the original volume 
 occupied by the cement, that is to say, upon its density; third, 
 upon the increase of volume which results from the combi- 
 nation with water ; fourth, upon the inherent tenacit}^ of the 
 material binding the grains together, the latter varying with 
 its chemical composition. 
 
 The area of contact between the grains will be increased 
 by pressure, which tends to bring the grains closer together, 
 and will also be increased by the increase of volume which 
 the cement undergoes when combining with water. In an 
 ideal cement, this increase of volume, would be just equal to 
 the volume of the empty spaces existing between the grains 
 while the cement was dr^^ Any increase of volume beyond 
 this point will simply tend to tear apart the partially solidified 
 mass, causing a crumbling of the cement. K similar effect 
 will be produced if, in consequence of a too rapid combination 
 with water, the temperature should rise sufficiently high to 
 produce steam. 
 
 That the crumbling which calcined lime undergoes on being 
 slaked is simply a result of the mechanically disintegrating 
 action of the evolved steam, may be shown b}' submitting a 
 piece of burnt lime to the action of dry steam, carefully 
 avoiding all condensation of water to the liquid state. 
 
 Under these conditions the lime will rapidly slake, without 
 
CHEMICAL PROCESSES. 93 
 
 cracking or crumbling, forming a stony mass of calcium hy- 
 drate. The tearing action of the steam may likewise be coun- 
 teracted by outside pressure; thus, if a steel tube perforated 
 with a number of fine holes be completely filled with burnt lime 
 and both ends closed by steel plugs and the entire arrangement 
 immersed in water, the contents will gradually become con- 
 verted to a rocky cylinder of calcium hydrate. Since, in 
 practice, it is impossible to adopt either of these devices, there 
 is no resource left but to so adjust the composition of the 
 cement as to cause its combination with water to take place 
 so slowly that the heat will escape nearly as fast as produced; 
 an injurious rise of temperature being in this way prevented. 
 Incidentally a further advantage accrues from retarding the 
 act of h3^dration. The processes of crystallization, to which, 
 as we have seen, the hardening of all cements is directly due, 
 takes place but slowly. If, therefore, the hydration takes place 
 rapidly, the crystallization will be imperfect and many of the 
 individual crystals will be too minute to bridge over the 
 distance between contiguous grains of cement. Under these 
 circumstances incomplete adhesion will follow and a weak 
 or friable cement will be the result. Now it is evident that a 
 cement consisting of coarse grains should combine with water 
 more slowly than one consisting of fine grains, in order to 
 give a product of equal tenacity, because in the former case 
 the interstices are larger and the crystal formation must be 
 larger and more complete in order to bridge over the inter- 
 spaces. As a practical consequence, if a very rapidly setting 
 cement is needed for any particular purpose, it must be very 
 finely ground. 
 
 In some cements, of which plaster of Paris may be taken 
 as the type, water simply combines with some constituent of 
 the cement, converting it into a more bulky and crystalline 
 hydrate. In others, various chemical reactions occur in 
 addition to the mere combination with water, giving rise to 
 new compounds. Thus in ordinary mortar, carbonic acid is 
 taken up from the air, forming calcium carbonate, to which is 
 due the great solidity and hardness of very old mortar. 
 
94 PORTLAND CEMENT. 
 
 The hardening of Portland cement is dependent upon the 
 combination of water with certain substances produced from 
 the constituents of cement by the action of the water and is 
 according!}^ a complex process. 
 
 III. 
 
 COMPOSITION AND CHARACTERISTICS OF 
 PORTLAND CEMENT. 
 
 Portland cement is a product formed by sintering* together 
 materials containing only clay and lime, and finely pulveriz- 
 ing. It is allowable to add not more than 2 per cent, of 
 plaster of Paris, or of some similar substance, for the pur- 
 pose of rendering the setting of the cement slower. Be- 
 yond this, all additions or substitutions are to be regarded as 
 adulterations.^ 
 
 An ideal cement of this class should possess the following 
 composition: 
 
 Lime, . . . 62.2 % 
 Silica, . . . 28.2 
 x\lumina, . . . 9.6 
 
 100.0 
 
 But up to about a third of the alumina may be replaced by 
 ferric oxide, which would correspond to the composition : 
 
 Lime, . . . 61.7 
 
 Silica, . . . 27.4 
 
 Alumina, . . . 7.5 
 
 Ferric oxide, . . 3.4 
 
 The following table shows how nearly the actual Portland 
 cements approach in composition the above ideal. The max- 
 ima and minima given are derived from analyses of nine rep- 
 resentative English and German manufacturers. 
 
 *That is, heating to the 
 •[Definition adopted by 
 Manufacturers. 
 
 temperature of incipient fusion, 
 the Association of German Portland 
 
 Cement 
 
ADULTERATION. 
 
 
 MAXIMUM 
 
 MINIMUM 
 
 Lime, 
 
 62 
 
 57 
 
 Silica, 
 
 24 
 
 19 
 
 Alumina, 
 
 8.8 
 
 5-2 
 
 Ferric oxide, 
 
 5-0 
 
 2.0 
 
 Magnesia, 
 
 3-5 
 
 0.3 
 
 Potash, 
 
 I.O 
 
 0.4 
 
 Soda, 
 
 0.8 
 
 0.4 
 
 Sulphuric acid, 
 
 I.O 
 
 
 Sand, etc.. 
 
 2.9 
 
 I.O 
 
 Carbonic acid, 
 
 1.9 
 
 0.0 
 
 Water, 
 
 1-5 
 
 0.0 
 
 Portland cement should not have a lower specific gravity 
 than 3.00 in the case of long stored samples, nor lower than 
 3.12 when freshly ignited. The best qualities exhibit a 
 density notably higher than this, viz.: 3.14 to 3.15. 
 
 It should show a loss of weight upon ignition to a bright 
 red heat, of from 0.34 per cent, to 2.6 per cent.,* a greater 
 loss showing imperfect manufacture or too great age. 
 
 One gram upon shaking with 100 c. c. water should impart to 
 the latter not more than 4 per cent, (on the weight of the 
 cement) of lime. The total magnesia should not exceed 3 
 per cent. 
 
 Adulteration with blast furnace slag may be detected in the 
 following simple way : if a pure Portland cement be shaken 
 up with water and then acidified with hydrochloric acid and 
 allowed to stand, it will soon settle, leaving the water clear. 
 If, however, slag be present, sulphur will separate under 
 the circumstances mentioned and will impart a milkiness or 
 opalescence to the water which will remain for many hours. -j- 
 
 A, perhaps, more positive indication of the same sophistica- 
 tion is afforded in the course of chemical analysis by the 
 reducing action of ferrous compounds always present in slag. 
 Fresenius has shown that i gram of an adulterated cement 
 should not be capable of reducing more than .0028 grams of 
 
 *Post, Chemisch-Technische 
 "I'Fresenius. 
 
 Analyse. 
 
p6 PORTLAND CEMENT. 
 
 potassium permanganate, whereas a slag cement has on an 
 average a ten times more powerful reducing action. 
 
 The presence of more than 66 per cent, of lime is likely to 
 cause the cement containing it to crumble or crack during 
 the setting. 
 
 This action is due to the increase of volume which lime 
 undergoes in slaking and consequent disruption of the net- 
 work of crystals the formation of which is the controlling- 
 factor in the setting. Nevertheless, the limit given is valid 
 only for cements of otherwise normal composition. The 
 injurious action of an excessively high percentage of lime 
 may be to a certain extent counteracted by a high percentage 
 of silica or also by calcining the cement at a higher tempera- 
 ture than usual. In either case the setting will be slower and 
 the injurious action of the lime less apparent. 
 
 On the other hand, magnesia enhances the tendency to swell 
 and crumble, hence the 3 per cent, limit given above. 
 
 The influence of the form and size of the cement grains 
 upon the hardening qualities is of the greatest importance, as 
 must be evident from the considerations advanced in the 
 previous section. 
 
 All grains so large as not to pass a sieve of 75 meshes to 
 the linear inch are to be regarded as inert or wholly passive 
 constituents. 
 
 They should not constitute more than 20 per cent, of the total 
 weight. The hardness and tenacity of the product increases 
 steadily with increasing fineness up to the extreme limits which 
 have been investigated, that is of grains so fine that all will 
 pass a sieve of 175 meshes to the linear inch. 
 
 The form of the grain depends upon the temperature at 
 which the mass has been sintered and upon the composition. 
 It is, therefore, difficult to separate the influence of this factc>r 
 from that of others. 
 
 It may, however, be said that flattened, slaty-grained 
 cements, such as are produced by thorough sintering at a high 
 temperature, are better than those having rounded grains, 
 which are formed at lower telnperatures, and the question may 
 
CHEMICAL PROCESSES. C)7 
 
 be left open whether the superiority of the first kind is due 
 directly to the form of grain or to the more perfect burning, 
 with the probabilities in favor of a concurrence of both causes. 
 
 IV. 
 
 CHEMICAL PROCESSES OCCURRING DURING THE 
 HARDENING OF PORTLAND CEMENT. 
 
 Le Chatelier has discovered by examining thin sections of 
 burnt Portland cement with the polarizing microscope, the 
 presence of calcium aluminate, CagAl^Og and of a calcium 
 ferrialuminate, Ca3(Al Fe)^O^j^ and of calcium metasilicate, 
 Ca Si O3. Calcium orthosilicate, Ca^Si O^, and the aluminate, 
 Ca^AlgOg, are also present (Landrin and others). 
 
 When the cement is mixed with water, the calcium orthosili- 
 cate and the calcium aluminate combine with water to form a 
 hydrated calcium silicate, H^Ca Si and (H O Ca 0)3A1, 
 hydrated calcium aluminate, which latter at once begins to 
 crystalHze out in the form of felted needles which extend in 
 every direction and are the first cause of the setting of the 
 cement. 
 
 SCa^Si O^+CagAl OgH-6H,0-- 
 2(H OCaO)3Al+3H3Ca Si O,. 
 
 At the same time a much slower process begins which 
 consists in a gradual crystallizing of the hydrated calcium sili- 
 cate and in the hydration of that portion of the silicates of lime 
 mentioned above which does not react with the aluminate. 
 To this tardy change is due the secondary hardening of the 
 cement, a process which takes place best under water. Lastly, 
 carbonic acid is absorbed from the air, converting any excess 
 of hme into carbonate and still further contributing to the 
 hardness of the product and to its insolubility in water. The 
 amount of carbonate of lime formed in this way in old cements 
 sometimes exceeds 12 per cent.* 
 
 The temperature of the water with which the cement is 
 m.ade up has a great influence upon the rapidity of the harden- 
 
 * Feichtinger, Dingl Journal, 152, pp. 40, 108. 
 
PORTLAND CEMENT. 
 
 ing and hence upon the tenacity finally attained. The chemical 
 reactions taking place in this process are, like all others, greatly 
 accelerated by elevation of temperature. This diminishes the 
 strength of the product because crystallization is essentially 
 slow, and if the crystals of hydrated lime aluminate do not have 
 time to form completely, their interlacing network, upon which 
 the toughness of the mass so greatly depends, wih not be 
 properly developed. The addition of gypsum to the cement 
 renders the setting slow by reducing the solubility of the 
 calcium aluminate but, for the same reason, it induces the for- 
 mation of smaller and, therefore, less efficient crystals and 
 hence acts injuriously when present in more than certain very 
 small quantities. 
 
CHAPTER VII. 
 
 Cement Testing Machines and Stone Crushers. 
 
 CEMENT TESTING MACHINES. 
 
 The different types of cement testing machines in general 
 use are shown in Figs. 27, 28, 29, 30. 
 
 The object of the machines is to bring a steadily increasing 
 
 Fig. 27. RiEiiLE Brothers' 1,000 pouisD Cement Tester. 
 
I03 
 
 PORTLAND CEMENT. 
 
 strain upon the specimen up to the point of rupture, and in 
 addition to this, the machine must be so designed that the 
 amount of strain, to which the specimen is subjected, may be 
 measured in pounds or some unit of weight. In the machines 
 shown in Figs. 2 7 and 28, the strain is applied to the specimen 
 
 Fig. 28. TiNius Olsen Cement Tester. 
 
 by means of a screw worked by a worm gear. I'he amount 
 of strain is measured by a weight sliding upon a bar similar 
 to the long arm of a steetyard. This bar being kept balanced 
 during the entire test, the amount of strain upon the specimen 
 at any time is indicated by the position of the sliding weight. 
 
CEMENT TESTING MACHINES. 
 
 lOI 
 
 The position of the specimen and the arrangement of clips 
 can be seen in the figures. Figs. 29 and 30 show a different 
 type of machine. In . machines of this type the strain is 
 applied to the specimen, not by means of a screw and worm 
 gear, but by means of a gradually increasing weight at the 
 
 1 I 29. FAiRBAN.is' Cement Tesj ^r. 
 
I02 
 
 PORTLAND CEMENT. 
 
 end of the long lever. In Fig. 29, the pan 7^ hangs at the end 
 of the long lever D. The hopper B is filled with fine shot. 
 The briquette is put in place and the levers balanced. A 
 valve in the hopper is opened and the shot runs through the 
 outlet / into the pan F". The rapidity with which the shot 
 runs into the pan can be regulated. The weight in the pan 
 increases and with it the strain on the specimen, until the 
 point of rupture is reached. The specimen breaks, the arm 
 D falls and shuts off the shot. The pan ^is then removed, 
 hung on the hook below E and weighed by means of the 
 sliding weight R on the lever D. From this may be obtained 
 at once the amount of strain under which the specimen broke. 
 Fig. 30 shows the outline of a similar type of machine. 
 
 Fiu. 30. 
 
 In regard to the relative merits of the different machines 
 illustrated, each investigator must decide for himselfj Most 
 accurate and satisfactory results ma}- be obtained with either 
 type. Any of these direct lever machines are to be preferred 
 to spring balances. 
 
 There are a number of hydraulic machines used in Europe 
 
CEMENT TESTING MACHINES. 
 
 103 
 
 that give good results, but they are more expensive than the 
 lever machines, and possess no advantages over them, within 
 the limit of power required. The capacity of the ordinary 
 cement testing machine is 1,000 pounds. Full descriptions 
 
 Fig. 31. Sectional View of Farret. Marsdex Stone Cri'.sher. 
 
 Fig. 32. Farrei, Marsden Stone Crlsher on Trucks. 
 
I04 PORTLAND CEMENT. 
 
 Fig. 23. r\\RRFL Marsden Revolving Screen for Broken Stone. 
 
 of these machines, with details, cost, etc., may be obtained 
 from the manufacturers. They also furnish complete outfits 
 for ceaient testing laboratories. 
 
 STONE CRUSHERS. 
 
 The t\'pe of stone crusher that .has been in most general 
 use for ths last thirty years is what is known as the Blake or 
 Marsden. Fig. 31 shows a sectional view of this type as now 
 made by the Farrel Marsden Foundr}^ Company, Ansonia, 
 Connecl-icut. 
 
 It is a cjmparatively cheap machine for the amount of 
 work it can do, and as the parts are few and strong, the 
 expense for repairs is almost nothing. Where the material to 
 be crushed is strongly laminated the crushing is not done in 
 as satisfactory a manner as with the revolving t3'^pe. Fig. 32 
 shows the crusher mounted on wheels, to facilitate its move- 
 ment from one place to another. 
 
 Fig. 33 shows the revolving screen that is furnished when it 
 
STONE CRUSHERS. 
 
 is desired to grade the size of the material. The size to 
 which the material is crushed may be regulated by changing 
 the pieces marked 7, 7, for similar pieces of greater or less 
 length, and by raising or lowering the wedge 10. 
 
 The following table in regard to size and capacity of the 
 Farrel Marsden crusher is taken from their published cata- 
 logue. The experience of the author has shown that the 
 most economical results are obtained when the engines used 
 are 15 to 25 % more powerful than those given in the table. 
 
 FAREL MARSDEN CRUSHER DIMENSIONS, CAPACITIES, ETC.: 
 
 
 Receiving 
 Capacity 
 
 Approximate 
 product of 2-in. 
 stone per liour 
 
 Approximate 
 weight 
 
 Horse- 
 power 
 
 Prices 
 
 3 
 
 10x4 in. 
 
 3 cubic yds. 
 
 4,900 lbs. 
 
 6 
 
 $ 275.00 
 
 4 
 
 iox7 " 
 
 5 " " 
 
 7,800 " 
 
 12 
 
 500.00 
 
 5 
 
 15x9 " 
 
 8 
 
 14,500 " 
 
 15 
 
 750.00 
 
 6 
 
 15X10 " 
 
 9 - 
 
 15.000 " 
 
 15 
 
 800.00 
 
 8 
 
 20x10 " 
 
 10 " 
 
 1 7.03J 
 
 20 
 
 1050.00 
 
 SEdTioNAL Perspective View 
 
 Fig. 34. Gates' Rock and Ore Breaker — Sectional View. 
 
io6 
 
 PORTLAND CEMENl'. 
 
 The Gates rock and ore breaker, Fig. 34 shows a sectional 
 view and Fig- 35 a general ^•ie^v. The machine has given 
 ■satisfaction and great numbers are in use. 
 
 gates' Rf)eK AND ORE BREAKERS — n 1 M EX S I ON S, PRICES, ETC. 
 
 ir. 
 
 Uimeiisions 
 of each reciv- 
 ing opening 
 abont. 
 
 X. is 
 
 ou n 
 'u 9 
 
 5 8-^^ . 
 
 — M M u CS 3 
 
 a ' '7/ '.S. 0 
 >,•'< a u '-I 
 •= 1 "^-r S ^ 
 
 u , bc - 
 
 p. 3 -Q = C 
 
 "c . 
 
 M 
 
 Size engine rec- 
 ommended to 
 drive breaker 
 elevators and 
 screen. 
 
 Indicated 
 Horse Power. 
 
 Prices. 
 
 INCHES. 
 
 rOUND.S. 
 
 I.NCHKS. 
 
 GO 
 
 2X 4 
 
 500 
 
 
 13 
 
 I to 1^/2 
 
 $ 125 
 
 O 
 
 
 3.300 
 
 2 to 4 
 
 28 
 
 4 " 5 
 
 400 
 
 I 
 
 5X12 
 
 5/300 
 
 4 " « 
 
 42 >4 
 
 8 " 10 
 
 600 
 
 2 
 
 6X14 
 
 7,800 
 
 6 " 12 
 
 4634 
 
 12 " 15 
 
 800 
 
 3 
 
 7X15 
 
 13,800 
 
 10 " 20 
 
 54/2 
 
 20 " 25 
 
 1,200 
 
 4 
 
 8X18 
 
 21,500 
 
 15 " 30 
 
 79?^ 
 
 25 •' 30 
 
 1,900 
 
 
 10X20 
 
 27,000 
 
 25 " 40 
 
 88 
 
 30 " 40 
 
 2,500 
 
 6 
 
 11X24 
 
 40,500 
 
 30 " 60 
 
 103 
 
 40 " 60 
 
 3.500 
 
 VA 
 
 14X30 
 
 65,800 
 
 75 "■ 125 
 
 120 
 
 75 " 125 
 
 6,000 
 
 8 
 
 18X42 
 
 89,000 
 
 100 to 150 
 
 132 
 
 100 " 150 
 
 7,000 
 
 Fig. 36. Gates' KocK and Ure Breaker — (jEXeral V'imw, 
 
CHAPTER VIII. 
 
 THE USE OF PORTLAND CEMENT. 
 
 In engineering construction Portland cement is used in 
 mortar, in concrete, and in the making of artificial sto)ie. It is 
 used either neat or mixed with a certain amount of sand. 
 
 Mortar. — From an engineering stand-point mortar may be 
 defined as a building material made by mixing lime, cement, 
 or plaster of Paris, either separately or in any combination, 
 with a certain amount of water, and to this mixture may be 
 added or not certain proportions of sand. The use of mortar 
 is to bind together the stone or bnck of which the masonry 
 is constructed. 
 
 Lime Mortar. — In the making of lime mortar the use of 
 sand is indispensable, if good results are looked for. The 
 reason of this is that the lime shrinks so much in setting, that 
 if lime and water alone were used, the mortar would detach 
 itself entirely from the adjacent stones or brick, and no bond 
 would result. By the introduction of a large proportion of 
 sand the percentage of this shrinkage is very much decreased. 
 The sand, being a perfectly inert mass, has no tendency to 
 either expand or contract. Another advantage attending the 
 use of sand in the lime mortar is its cheapness. The bulk of 
 mortar is much increased by the addition of sand, and this 
 increase gained at a less cost than if Hme alone had been used. 
 
 Lime mortar is the cheapest form of mortar that can be 
 made. This is due to the small first cost of the lime, and then 
 to the fact that the lime expands to three or four times its 
 original bulk in slaking. 
 
 The utility of lime mortar is limited, owing to the very 
 
io8 
 
 PORTLAND CEMENT. 
 
 slight strength that it attains under the most favorable cir- 
 cumstances, and from the fact that it cannot be used to 
 advantage in damp places. Its principal use is in the shape of 
 plaster and in laying up ordinary brick walls that are to be 
 subjected to no extraordinary strain, and are in such a location 
 as to be comparatively dry. 
 
 The use of lime mortar in ordinary foundations, cellar 
 walls, retaining walls, and in all masonry subjected to great 
 strain or moisture has been superseded by the use of some 
 form of cement mortar. Also in the construction of any 
 masonry in which the thickness is such as to e'xclude all 
 possibility of the air penetrating the interior. Under all 
 ordinary circumstances the mortar is made of some brand of 
 natural cement and the use of Portland cement is reserved for 
 the more important points. There has been much discussion 
 as to which cement was the most economical for use, natural 
 cement or Portland cement. Excepting in a very few cases, 
 the subject does not appear to us as admitting of any discus- 
 sion. An examination of the diagrams in Chapter IX. will 
 show at once the great superiority in strength that Portland 
 cement possesses and the question turns upon the amount of 
 strength and certain other qualities required in the mortar. 
 
 The effect of sand in reducing the strength of mortar is 
 relatively about the same in each cement. The practice of 
 adding an exorbitant amount of sand to Portland cement, in 
 order to reduce the cost of the mortar, is bad. The resulting 
 mortar is porous, affected by the weather, and liable to 
 disintegrate. Of course, if suflicient sand is added to Port- 
 land cement, the cost of the mortar may be reduced to that of 
 natural cement mortar, and possibly the Portland cement 
 mortar may still be the stronger in any laboratory tests. 
 But it will not be as durable. When the strength required is 
 no more than can be reached by mortar made of natural 
 cement and an ordinary amount of sand, the natural cement 
 should be used; and when more strength is required, the 
 Portland cement should be used. What has just been said 
 applies merely to the ultimate strength of the mortar. There 
 
INCIPIENT SET. IO9 
 
 is one other point, however, that must be frequently consid- 
 ered, viz. : the setting properties of the cement. 
 
 The Portland cements are usually what are termed " slow 
 setting," while the light-burned natural cements are "quick 
 setting." This property may in some cases settle the question 
 as to w^hich cement should be used. It should be remem- 
 bered, however, that there is a great variation in the setting 
 qualities of the different Portland cements, and also in the 
 same brands at different ages. Usually, the older a Portland 
 cement is, , the slower setting it is, provided it has been 
 properly stored. 
 
 Many of the best Portland cement factories can furnish, 
 upon demand, either a slow or qtiick setting cement that will 
 meet the standard specifications for strength. 
 
 On the other hand, the natural cements may be made slow 
 setting by the addition of a little lime paste to the mortar. 
 
 The following proportions make a good slow setting mortar: 
 
 1 part lime. 
 
 2 parts natural cement. 
 9 " sand. 
 
 By the addition of a small amount of plaster of Paris to 
 either natural or Portland cement mortar the setting is very 
 much hastened, and the ultimate strength of the mortar not 
 affected. The amount of plaster of Paris used should not 
 exceed 3 There are very few conditions under which a 
 
 quick setting cement is absolutely necessar}^, and the require- 
 ments as to ultimate strength should usually decide the brand 
 of cement to be used. When a slow setting cement may be 
 used it possesses many advantages. The mason has more 
 time in which to do his work, and therefore it is better done. 
 In the use of the quick setting cements, the mortar can only 
 be mixed in small quantities, and even then there is danger 
 that it will set before being used, and thus have to be thrown 
 away. 
 
 There is one thing that must be carefully guarded against 
 in all work with cement. There seems to be a general 
 tradition, believed by many masons, that the mortar made 
 
no PORTLAND CEMENT. 
 
 from certain cements is improved by allowing it to partially 
 set, and then regauging it with more water and reworking it. 
 Nothing could be more injurious to the mortar. The amount 
 of injury that results from this depends entirely upon to what 
 degree the mortar has been allowed to set before it is rew^orked. 
 
 The setting of the cement is caused by the formation of 
 needle-like crystals, shooting out in every direction from each 
 grain. These crystals bind the whole mass together, and the 
 beginning of the formation of these crystals is the beginning 
 of the so-called setting. If the cement is reworked after the 
 formation of these crystals commences, the crystals afe broken, 
 the process of crystallization interfered with, and the ultimate 
 utility of the mortar very much reduced. To the mason, in 
 the handling of the mortar, there is some advantage in this 
 reworking of it. A mortar made of a quick setting cement 
 must be handled quickly and mixed in small quantities. It 
 does not work "smoothly," but is apt to be harsh and brash. 
 B}^ allowing this incipient set to take place and then reworking 
 it, the mortar becomes " smooth working " and slower setting. 
 The reason why it is apparently slower setting is evident. A 
 portion of its setting qualities has been destroyed, and if this 
 so-called incipient set is continued a sufficient length of time, 
 and then the mortar reworked, a mortar may at last be 
 obtained that will have no setting properties at ail. 
 
 The regauging of mortar should never be allowed under 
 any circumstances. If the mortar sets too quickly and works 
 '•brash," the evil can be remedied by the introduction of a 
 small amount of lime, and . the quality of the mortar not 
 materially affected. 
 
 Sand. — The strength of the mortar depends not only upon 
 the strength of the cement, but also upon the quality of the 
 sand and the relative amount of water used. The sand should 
 be clean and sharp. River sand is often water worn and the 
 grains rounded; such sand does not make as good mortar as 
 sand having angular grains. Bank sand is the best when it 
 is not mixed with earth or cla}^ If the grains are covered 
 with a deposit of earth or clay, then the cement does not come 
 
SAND. 
 
 Ill 
 
 in direct contact with them and the resuUing mortar is inferior 
 in quahty. The requisite qualities of good sand have been 
 well understood from the earliest times. The following 
 specitications for sand were written by Marcus Vitruvius 
 Pollio for the benefit of the Roman Augustus, about the year 
 25, B. C, and may be regarded as a standard to-day. 
 
 ''In buildings of rubble work it is of the first importance 
 '•that the sand be fit for mixing with the lime, and unalloyed 
 
 with earth The best .... is that 
 
 which, when rubbed between the fingers, yields a grating 
 ••sound. That, also, which is earthy, and does not possess 
 "the roughness above named, is fit for the purpose, if it 
 "merely leave a stain or any particles of earth on a white 
 •'garment, which can be easily brushed away. If there be 
 "no sand pits where it can be dug, river sand or sifted gravel 
 " must be used. Even sea sand may be had recourse to, 
 '•but it dries very slowly; and walls wherein it is used must 
 •' not be much loaded, unless carried up in small portions at 
 "a time. It is not, however, fit for those walls that are to 
 
 "receive vaulting If sand have been dug a 
 
 "long time, and exposed to the sun, the moon, and the rain, 
 
 "it loses binding quality and becomes earthy "* 
 
 In regard to Vitruvius' specitications for lime and cement, 
 they would hardly pass for standard to-day. Although con- 
 sidering the*material with which he had to do, it is surprising 
 how well he understood the results that would obtain. The 
 course of reasoning which he follows in accounting for these 
 results is most ingenious and plausible, although totally 
 incorrect. It will well repay any engineer or architect to 
 read this treatise, and many of our architects of to-day would 
 do well to follow some of the rules he laid down, althovigh 
 they may smile at the course of reasoning he followed, in 
 proving his conclusions. 
 
 But Vitruvius aside, the sand should be sharp, angular and 
 clean. The coarseness of the grain that is desirable depends 
 upon the character of the work to be done. But where 
 
 From Gwilt's Vitruvius. Priestly and Weale, London, 1826. 
 
112 
 
 PORTLAND CEMENT. 
 
 merely strength is required a moderately coarse grain, with a 
 mixture of finer grain, gives the best results. When a fine 
 finish is required the grain should be smaller and possibly 
 less sand used. 
 
 In testing the mortar for any particular work, it should be 
 tested with the sand that is to be used, and then tested with 
 the standard sand, crushed quartz or glass. By this method 
 of treatment, not only will the value of the cement be obtained, 
 but also the relative value of the sand that is to be used. 
 
 Water. — The amount of water that is used has a great 
 influence upon the quality of the mortar. Within certain 
 limits the less water the better the mortar. Just sufficient 
 should be used to render the mixture as stiff as can be worked 
 with advantage. When more than this amount is used it has 
 a weakening effect upon the mortar. 
 
 When there is is a superfluity of water, it simply dries out in 
 time and leaves the mortar porous and liable to disintegration. 
 
 Mixing. — In mixing mortar the sand and cement used 
 should be thoroughly mixed di'y and then the proper amount 
 of water added. This mixing is usually done by hand and is 
 done in a mortar box. The two materials are spread out in 
 separate layers and then thoroughly mixed by being turned 
 with shovels or hoes. After the water is added the whole 
 mass is thoroughly mixed and used without loss of time. The 
 stone or brick with which the mortar is to be used must be 
 thoroughly wet before the mortar is applied. In case this is 
 not done, the stone or brick will draw the water from the 
 mortar before setting begins, and the result will be either a 
 very inferior mortar or a mere dust of sand and cement 
 powder that has no binding qualities. 
 
 The one thing to be guarded against in the use of cement, 
 is the superfluous use of water. There is very little danger 
 of using too little. 
 
 Masons, almost without exception, injure the quality of their 
 mortar by the excessive use of water. This is more particu- 
 larl}' the case in the making of concrete, and additional atten- 
 tion will be called to it under that head. 
 
CONCRETE. 
 
 113 
 
 Cement as Plaster. — Cement is not generally used for 
 interior plaster work, but within the last few years a number of 
 patent cements have been put on the market for this purpose. 
 
 Nearly all of these cements possess decided advantages 
 over the lime mortar and the difference in expense is very , 
 slight. The advantages resulting from their use are, the 
 greater ultmiate hardness and durability; the smooth finish that 
 can be given that enables them to be cleaned readily with soap 
 and water; the decrease in time needed to apply them and 
 the rapidity with which they set and dry. Rooms plastered 
 with them can be occupied with safety in much less time than 
 those in which lime mortar has been used. Their cost has 
 been so much reduced that there is no reason why their use 
 should not become general, and they should certainly be used in 
 the interiors of hospitals, where perfect cleanliness is desirable. 
 
 The composition of these patent cements is usually some 
 combination of a light-burned natural cement and plaster of 
 Paris. In some cases an over-clayed Portland cement has 
 been used and in others a mixture of plaster of Paris and burnt 
 clay has given most satisfactory results. 
 
 CONCRETE. 
 
 x\mong the most important uses to which cement is put, is 
 the making of concrete. It is in this line of work that by far 
 the larger portion of Portland cement is used. 
 
 Concrete may be defined, from an engineering stand-point, 
 as a mixture, of sand, broken stones, broken brick, gravel, 
 or any such material, held together by some cementing sub- 
 stance such as lime, natural cement, Portland cement or 
 some patent cement, mixed with the proper amount of water. 
 
 The materials comprising concrete, may be divided into 
 two classes, the material to be held together and the material 
 holding it together. The broken stone, brick or gravel is 
 called the a'^gregate, the cementing material is called the 
 matrix. 
 
 Materials. The materials to be used in the making of 
 concrete depend upon the object for which the concrete is to 
 
114 PORTLAND CEMENT. 
 
 be used, the available material and the juitifiable cost. The 
 question of available money has too often entered into the 
 question of concrete construction in this country. The result 
 has been that much poor concrete has been used 'and in most 
 cases has proven a failure. Concrete should not be used 
 unless there is sufficient money to guarantee good, suitable 
 materials and good work. 
 
 Good concrete and suitable materials do not of necessity 
 mean the use of the best Portland cement. What is meant, 
 is the use of such a cement and such an aggregate, that the 
 resulting concrete shall attain a strength many times beyond 
 that, that will be required by any future load. Unless this 
 can be done, it is better to use some substitute. In regard 
 to the quality of the cement that should be used, that will 
 depend upon the purpose for which the concrete is to be iised. 
 and will be treated of under tlie different, types of concrete. 
 The aggregate must possess the following qualities, to a 
 greater or less extent: 
 
 Hardness, Toughness, Durability, 
 
 Angularity. Cleanliness. 
 
 The ultimate quality of the concrete depends primarily 
 upon the qualities of the raw material used. The aggregate 
 may be some material such as stone or brick broken to a suit- 
 able size or it may be gravel. The necessity of hardness, 
 toughness and durability in the aggregate is self evident and 
 needs no explanation. 
 
 x\ngularity of fracture or structure improves the qualitv of 
 the concrete as it makes the binding together of the material 
 much tirmei- and stronger. For this reason broken stone is in 
 many ways preferable to gravel, provided the quality of the 
 materials is equal. The gravel being composed of stones 
 worn more or less smooth and rounded by water, does not 
 offer the hold for the cement that is offered by broken stone.* 
 
 *In an abstract from a recent Go\ ernment Report it has been claimed 
 that tlie bond between ceinent mortar and rounded pebbles was stronger 
 than between the same mortar and broken stone. The broken stone must 
 ha\e been of a \ery inferior quality or some llnkno^vn element have 
 entered into the makiny- or testing of the concrete. — Aithor. 
 
AGGREGATES. 
 
 Whatever material is used for an aggregate there is one 
 condition that is absolutely indispensable, if good results are 
 obtained and that is cleanliness. The aggregate must be 
 clean, and to insure this, it must be freshly excavated 'if 
 gravel, or freshly broken if stone. Any material that has 
 been exposed for a year or more will deteriorate in value as 
 an aggregate. By cleanliness is meant that, it must be free 
 from dust, dirt, and earthy matter and from any mould or veg- 
 etable grovvtii. If the particles of aggregate are covered 
 with dust or dirt of any kind, the cement is thereby pre- 
 vented from coming in contact with it and the resulting bond 
 is not perfect. Under the head of mixing concrete this sub- 
 ject wall be again touched upon. 
 
 Size of Aggregates. — Within certain limits the size of the 
 aggregate depends upon the future use of the concrete. For 
 the purposes of foundations and such general work, where 
 the requirement is strength and not artistic effect, the size 
 miy vary, but should not exceed, in broken stone, a piece 
 that will pass through an inch and a half ring. With this 
 size as a maximum, the pieces may grade down to the size of 
 a coffee bean. The graduation in the size of the aggregate 
 lessens the amount of cement mortar necessary-. 
 
 Reducing the Aggregates. — When broken stone or other 
 similar material is used, it is usually broken in a stone cntslicr. 
 The stone crushers used are of the same t3'pe as those already 
 described in Chapter VII. The material as it comes from the 
 crusher must be entirely freed from the coating of fine dust 
 with which it is covered. This can easily be done by turning- 
 water on it. 
 
 Screening. — The broken stone should be screened, as it 
 coaies from the crusher. The object of the screening, is not 
 to reduce the material to uniform size, but simply to keep the 
 size of the stone between a maximum and a minimum limit. 
 The finer particles, those smaller than coffee beans, should not 
 be used, and all of the dust should be carefully screened or 
 washed out. As a maximum limit no piece larger than an 
 inch and a half in any direction should be used. 
 
ii6 
 
 PORTLAND CEMEJ^T. 
 
 It is not desirable that the aggregate should be of uniform 
 size. When of different sizes they pack better and require 
 less mortar to make good concrete. The amount of mortar 
 needed will depend upon the voids between the particles com- 
 posing the aggregate. It should be sufficient to entirely fill 
 these voids and to give a thorough and complete coating to 
 each particle. In regard to what particular stones make the 
 best aggregate, there is no stone that excels the best quali- 
 ties of granite or trap. These two varieties, together with 
 deposits similar to the Sioux Fails Quartzite, possess all of the 
 requisite qualities for the making of good concrete. The 
 sandstones are usually much too soft and are not usually con 
 sidered suitable. Some deposits of limestone will make fair 
 concrete. The hardest and toughest varieties. None of it 
 however can be considered first-class and should only be used 
 when no other material is available, and then used with much 
 care. Limestone does not crush well. There is a tendency 
 to powder and also to break into long thin slabs. There is a 
 loss of from 15 to 25% in the stone from these causes and this 
 should be remembered in making any estimate as to quanti- 
 ties needed, or as to probable cost. The breaking of the stone 
 into thin slabs wall be obviated by the use of a rotary stone 
 crusher, but the amount of dust and waste will also be very 
 much increased. The fact that limestone is so generally dis- 
 tributed and so cheaply quarried, is the only excuse for its 
 very general use in the making of concrete. 
 
 Gravel. — Good, clean gravel, when obtainable, makes a 
 cheap and durable aggregate. The lack of angularit}^ is the 
 one bad quality it u'sually has. The material of which it is 
 composed is usually hard and durable, and if the quality of the 
 mortar used is sufticientl}^ rich, the resulting concrete is good. 
 
 The gravel should be carefully screened and all earthy or 
 loamy material removed. It is not necessary to remove any 
 clean, sharp sand that may be mixed with the gravel. This 
 sand can be used with the gravel and less sand used with the 
 cement in making the mortar. 
 
 Hand Broken. — When broken stone is used as an aggre- 
 
STONE CRUSHING. 
 
 gate, it is often broken by hand in the place of being broken 
 by machinery. 
 
 This is always the case where a limited quantity of the 
 material is needed, unless it can be purchased already broken 
 from some central supply station. Stone broken by hand 
 possesses some advantages over machine broken stone, The 
 stone can be broken to any required size and this size runs 
 with greater uniformit}'. There is almost a total absence of 
 dust and much less waste in the stone. Hand broken stone 
 does not require screening. 
 
 Where machine broken stone is not available from a central 
 supply, and the quantity needed not large, it should alwa3-s be 
 hand broken. The great cost of machine broken stone is the 
 lirst cost of the plant. 
 
 x\fter the plant is once in place, and provided there is suffi- 
 cient material to keep the machine running all the time, it costs 
 less than one-fourth as much to break by machine as it does by 
 hand. But if the arrangements and needs are such that the 
 machine only runs at irregular intervals, the machine crushing 
 may be even more expensive than that done by hand. 
 
 As to the cost of the plant, that depends entirely upon the 
 type of crusher used and the capacity of the plant. (See 
 tables in Chapter VII.) 
 
 The crusher should always be placed in or near the quarrv- 
 There is a loss of from 12 to 25% in the crushing, and of 
 course this loss should be moved as short a distance as possible. 
 
 In buying stone, the better method is to buy it b}- weight. 
 This is more satisfactory to both parties. When stone is 
 bought by the cubic yard, there is always a question as to its 
 measurement and also as to the manner in which it is piled. 
 The weight of a cubic 3'ard of stone must be determined and 
 when this is once done there can be no dispute as to the 
 amount that has been purchased. 
 
 One must remember that i cu. yd. of solid stone will make 
 about 1.8 cu. yds. of broken stone. The stone loosely piled 
 and broken to a maximum of 2 inch cubes. 
 
 This will decrease about 10% when well rammed in place. 
 
ii8 
 
 PORTLAN D f J- AIEKT. 
 
 Mixing- Concrete. — The proportions of the different ingre- 
 dients that compose tlie concrete depend vipon the character 
 of the ingredients, the character of the concrete required, etc., 
 and are treated of elsewhere. But, whatever the ingredients 
 may be and in vvliatever proportions they may be used, tlie 
 quality of the resulting concrete can be very materially affected 
 by the methods used in mixing, and the subsequent manner of 
 manipulation. Where concrete is to be used in comparatively 
 small quantities, it is mixed by hand and this should be done 
 ia thi; following manner: The mixing is usually done in a 
 rough wooden mixing box, not more than lo or 12 inches deep 
 and 4 feet wide. The length of the box depends upon the 
 amount of material that is to be handled at once and for the 
 working of two men a box about 14 feet long is convenient. 
 
 The box should be approximately water tight. 
 
 The correct proportions of sand and cement are put into 
 the box and ilioroiigJdy mixed dr\'. This mixture is then 
 drawn to the sides and ends of the box and the proper 
 amount of water is poured into the center. The whole mass 
 is then mixed by either hoes or shovels. The amount of water 
 used should be just sufiicient to cause the mixture to have the 
 appearance of wet sand. The important point to be guarded 
 against is the excessive use of water. There is comparatively 
 slight danger of using too little. 
 
 As soon as the sand, cement and water have been thoroughly 
 mixed, the broken stone or gravel is added. 
 
 Care should be taken that the aggregate used is clean. 
 All dust, dirt or earthy material should be carefully washed 
 out and the aggregate should be wet before being put into the 
 mortar. The wetting should be done in such a manner that 
 any surplus water not absorbed by the aggregate may drain 
 awa\'. The aggregate in this condition, is then added to the 
 mortar and the whole thoroughly mixed, until every particle 
 of the aggregate is covered with a coating of the mortar. 
 
 The concrete is then ready to be put in place and should be 
 used at once. 
 
 Platforms. — In using mixing boxes, a certain amount of 
 
MACHINE MIXING. up 
 
 time or expense is lost in lifting the material over the side of 
 the box. This has led in some cases to the substitution of a 
 platform, upon which the mixing is done. 
 
 These mixing platforms have been used with success in 
 cases where the concrete was to be used as a foundation for 
 pavements or for the subway for cable roads, etc. The width 
 of a platform should be such that a man standing at one side 
 can work be3'ond the center with a shovel. The length 
 depends upon the amount of concrete to be mixed at one time. 
 
 When the platforms are only 8 or lo feet long they either 
 ma}' rest directly upon the ground or have wooden shoes under 
 them, similar to those under a stone drag. When longer, they 
 have been built with small, broad iron trucks. 
 
 The method of manipulation is as follows: 
 
 The platform is dragged into place. The concrete mate- 
 rial has already been deposited at that point in approxi- 
 mately the correct proportions. A barrel of cement is placed 
 on the platform and both ends knocked out. It is then ended 
 from one end of the platform to the other and the cement 
 dropping out each time the barrel is turned, is thus dis- 
 tributed over the platform. A number of empty barrels 
 without heads are then placed upon one end of the platform 
 and filled with sand and gravel. These barrels are ended 
 from one end to the other and the aggregate thus distributed. 
 This dr\' material is then turned rapidly with shovels, three 
 times, then drawn to each side and the proper amount of 
 water added. The water is usually in a barrel or tank 
 mounted upon w^heels. As soon as the water is added the 
 whole mass is turned and mixed some three or four times and 
 then shoveled at once into place. The platform is dragged 
 ahead to the next point and the work goes on. 
 
 Machine Mixing-. — When the amount of concrete to be 
 mixed in one place becomes considerable, the mixing is 
 usually done by machiner}-. There are a great number of 
 these concrete mixing machines upon the market, but we will 
 treat only of the different t3'pes, and not of the merits of the 
 individual machine. 
 
I20 
 
 PORTLAND CEMENT. 
 
 The type of machine to be used depends to a certain 
 extent upon the character of concrete to be made. That is, 
 whether it is to be made of coarse broken stone or finely 
 broken stone and sand. One of the simplest forms of con- 
 crete mixer is that shown in Plate II. It consists of a cubical 
 wooden or iron box, this box is fixed to a horizontal shaft, 
 the shaft being fastened to two of the diagonall}^ opposite 
 corners. By means of gearing the shaft is rotated and with 
 it the box. There is a proper opening in the box, that can be 
 closed tightly. The concrete material, cement, sand, gravel 
 or broken stone, and water, is placed in the box in the correct 
 proportions and the opening closed. The box is revolved a 
 number of times — usually about five — then stopped and the 
 thoroughly mixed material is run out into a barrow, box, cart, 
 or any other receptacle. The box being hung by the corners, 
 the material is thrown from side to side and from end to end 
 in each revolution. The mixing by means of this machine is 
 best done when the aggregate has a maximum size of a 
 hen's egg; when sand and fineh^ ground aggregate, alone are 
 used, the mixing is not as well done, as the material does not 
 fall as freely from one side of the box to the other. Care 
 
MACHINE MIXERS. 
 
 121 
 
 must be taken not to rotate the box too rapidly, but to allow 
 sufficient time for the material to fall from one side to the 
 other. Fig". 36 shows a more elaborate mixing machine upon 
 this same principle. 
 
 In the type of machine just described, the different mate- 
 rials for making the concrete are all put in the tumbler box 
 or cylinder and all mixed at once. There is a type of machine 
 
 Fig. 37. 
 
 shown in Fig. 37, in which the sand, cement and water are 
 thoroughl}^ mixed and then the broken stone or gravel added. 
 The sand, cement and water are put in the cement hopper. 
 In the bottom of this hopper is a large endless screw that 
 runs in the tube. The revolving of this screw thoroughly 
 mixes the mortar and forces it out of the upper end of the 
 tube into the chute, down which it runs into the mixing cylin- 
 der. The broken stone or gravel is placed in the box below 
 the tube, carried up by means of the chain and bucket eleva- 
 tor, and thrown into the chute, from which it passes with the 
 mortar into the mixing cylinder and comes out at the lower 
 end thoroughly mixed. The capacity of this machine is about 
 65 cubic yards per day. 
 
 Machines similar to this are made that have two delivery 
 chutes at the back for the discharge of the concrete. Such 
 machines are used for the depositing of concrete in two 
 parallel trenches, when it is used as a foundation for street 
 railway tracks. In types of mixing machines, such as just 
 described, the mixing is done by some form of a revolving' 
 
122 
 
 PORTLAND CEMENT. 
 
 endless screw or revolving beaters and they can onl}- be used 
 with advantage when the aggregate is small. 
 
 Fig. 38 shows a very sim- 
 ple form of mixing machine 
 that may be run by hand. A 
 small one of this tvpe has 
 been found ver}' useful in 
 laboratory experiments. 
 
 The qualit}' of the cement 
 used depends upon the re- 
 quired character of the con- 
 crete. For foundations for 
 building, an}' first-class nat- 
 ural cement will answer the 
 purpose. 
 
 For very important and 
 expensive foundations, par- 
 ticularly those under water, 
 the best Portland should be 
 used. By the best Portland, 
 is meant, not only a Portland 
 that will comply with the test 
 specifications, but a brand of cement that has been tested to 
 such an extent that the engineer may have no doubt as to its 
 jLiiiformity of character. This quality of uniformity of produc- 
 tion will'do more to sell a cement than any other. 
 
 For foundations for pavments, the best natural cements 
 answer the purpose. 
 
 For lining cisterns and reservoirs, some natural cements are 
 suitable, and a good Portland with about 20 lime makes a 
 most excellent working mortar. 
 
 For all side-walk work, only the best Portland should be 
 used. In an}' work where the concrete will be subjected to 
 attrition or wear, only the best Portland should be used. 
 
 In the laying of cement walks, attempts have been made to 
 use a base of natural cement concrete with a covering or 
 wearing surface of Portland cement. So far this has not been 
 
conc\ui:tk. 123 
 
 successful. The union between the two kinds of concrete has 
 not been perfect and the result has been that, in a short time, 
 the top was scaled off. 
 
 For the manufacture of artilicial stone, only the best Port- 
 land should be used, unless some special patent cement, made 
 for that purpose, is used. 
 
 There are a number of these patent processes in vogue, 
 that appear to give good results. As the author has not experi- 
 mented upon them, and is not familiar with their composition, 
 he is unable to give any opinion as to their relative merits. 
 
 Only a few of the more general uses of cement have been 
 mentioned and what has been said in regard to the use of nat- 
 ural or Portland cements must be taken in the form of sugges- 
 tions rather than a statement of absolute facts. 
 
 Sand. — The sand should be clean and sharp. 
 
 Aggregates clean, angular and of graduated size up to the 
 maximum. Hard and tough. 
 
 Proportions. — The proportions, in which the ingredients 
 should be used, depend upon so many points, such as the qual- 
 ity of the constituents and requirements in the concrete, that 
 they will not be taken up here. There are a number of speci- 
 fications given, for various types of work that show w^hat 
 proportions are used. 
 
 Water. — The amount of water used is almost universally 
 too great. The concrete when read}' to be put in place should 
 •be merely a damp, incoherent mass. 
 
 There is very little danger of using too little water. 
 
 Mixing". — Under ordinar}' circumstances the sand and 
 'Cement should be mixed dry, and then sufficient water added. 
 The stone or gravel, having been thoroughly washed and wet 
 should then be added and the whole mass worked and mixed. 
 
 Depositing Concrete. — Until within a comparatively few 
 years there was much error existing as to the best methods of 
 depositing concrete in place. For some unaccountable reason 
 most of the older specifications specify that the concrete shall 
 not be tamped after being put in place, and in order that it may 
 be, to a certain extent, solidified, it was specified that it should 
 
124 PORTLAND CEMENT. 
 
 be dumped in place from a certain height. Nothing could 
 have been more erroneous than such ideas. The concrete 
 should not be allowed to fall a considerable height. Such a 
 fall tends to separate the larger pieces from the smaller, and 
 thus results in a lack of uniformity in the concrete. The con- 
 crete should be put in place with care, not allowed to fall more 
 than 3 or 5 feet. 
 
 It should then be spread evenly and thoroughly tamped. 
 The tamping should be done quickly, but should be done thor- 
 oughly. The concrete should not be laid in layers more than 
 6 or 8 inches in thickness. That is in layers not thicker than 
 can be well tamped. Succeeding layers should be put in place 
 before the previous ones have become set. 
 
 When the work is, from necessit}-, stopped for a length of 
 time sufficient for the cement to become set, it should be left 
 in the form of steps, and just before the next deposits of 
 concrete are made, it is a good plan to cover the surface of the 
 old concrete with a coating of thin grout, made of neat cement 
 and an abundance of water, in order to make a perfect union 
 between the deposits. 
 
 It is often specified that the tamping shall be continued until 
 the water rises to the surface. If no more than the proper 
 amount of water has been used, the tamping will have been 
 sufficient long before any of this water will have appeared on 
 the surface. 
 
 Depositing- Concrete Under Water. — When concrete is 
 to be deposited under water, some special device is necessar}-. 
 
 The concrete cannot be thrown into the wati^r and allowed 
 to sink by its own weiglit, for the reason that all the cement 
 and mortar would thus be washed from the aggregate. 
 Various devices are used. 
 
 The concrete is often placed in large paper sacks and low- 
 ered in place in this condition. The action of the water and 
 the weight of the concrete, soon breaks the sacks and allows 
 the concrete to unite. 
 
 Where the depth of water is not too great, a wooden tube 
 or box can be used. The length of the box being greater 
 
DEPOSITING CONCRETE. 
 
 125 
 
 than the depth of water. The box is water tight, and the 
 lower end fitted with a cover that can be opened from above. 
 A certain amount of concrete is put into the lower end of the 
 box and the box then lowered into the water until the lower 
 end is near the place where the concrete is to be deposited. 
 The cover is then opened and the concrete slides into place. 
 
 There are numerous devices in the shape of buckets that 
 are used. The essential features of all of these buckets for 
 
 Fig. 39. 
 
 depositing- concrete, are that the form should be such that when 
 the bucket is opened, all of the contents ma}^ fall out and also 
 that the device for opening shall be such that it can be worked 
 from above. Fig. 39 shows one of these types used. Figs. 
 
 40 and 41 show a 
 device for a heavy 
 canvas bag open 
 at both ends. The 
 lower end is drawn 
 together by a rope 
 fastened in the 
 manner shown in 
 
 Fig. 41. The bag is filled with concrete and lowered into 
 place. Then the sling rope is pulled, the sHng comes off, the 
 lower end opens and the concrete shdes mto place. 
 
 Fig. 41. 
 
 Fig. 42. 
 
126 
 
 PORTLAND CEMENT. 
 
 Pavements and Sidewalks. — The use of Portland cement 
 for the paving of streets has never been general. But there 
 are some few localities, such as in parts of London, where 
 it has been used with great success, as far as the wearing 
 qualities are concerned. 
 
 In order that concrete may serve as a pavement for general 
 traffic, only the very best materials of every kind must be used 
 and this necessity renders the cost so great as to be prohibi- 
 tory excepting in rare instances. The manufacture of paving 
 brick has reached such a degree of excellency and the product 
 is sold at such a reasonable price that, probably, the day for 
 cement pavement has passed. 
 
 In any of the Portland cement pavements in London the 
 concrete was not made in place, but molded into blocks which 
 after hardening were put in place as artificial stone blocks. 
 As a foundation for a street pavement, concrete is used in 
 enormous quantities. For this purpose the light-burned natural 
 cements answer and thus the cost of the concrete is very much 
 reduced. 
 
 The concrete for this purpose usually consists of the follow- 
 ing proportions. 
 
 1 part natural cement. 
 
 2 to 3 parts sand, coarse and gravelly. 
 5 " 7 " broken stone. 
 
 This cement concrete makes an ideal street covering when 
 a suitable wearing surface is put over it. It presents a smooth, 
 even surface and may be given any form or shape required. 
 It is impervious to water and thus fulfills one of the most 
 important requirements of street covering. The method of 
 applying it is as follows: 
 
 The sub-grade is given the cross-section of the finished 
 surface and rolled smooth and even. 
 
 The concrete is made and deposited in layers of the required 
 thickness, usually from 6 to 9 inches, depending upon the 
 nature of the sub-soil below and the character of the traffic 
 above. The concrete is given the required contour by the 
 use of a template. It is then thoroughly tamped, and the top 
 
CEMENT WALKS. 
 
 smoothed off. The concrete can be made at the sides of the 
 streets and moved into place by barrows, or it can be made 
 on platforms (see page ii8) and put in place by shovels. 
 
 The concrete is usually allowed to set for three days, then a 
 layer of sand from i to 2 inches in thickness is evenly spread 
 over it, and the wearing surface, in the shape of granite blocks, 
 bricks, or wooden blocks are put in place. 
 
 In order to obtain the best results with any given concrete, 
 it is necessary that it should be kept damp and protected from 
 the heat of the sun. 
 
 For this purpose the sand is usually thrown on a few hours 
 after the concrete is in place, and the sand kept damp for 
 several days. 
 
 Such a foundation as this, makes it a rather expensive mat- 
 ter to tear up the street for gas and water pipes, but with 
 proper care the concrete can be replaced in a manner that in 
 no way causes deterioration to the pavement. 
 
 Before a foundation such as this is laid on any street, the 
 property owners should receive ample notification of the fact 
 and also of what amount they may be fined, in addition to the 
 actual cost of the repairs, in case they are the cause of the 
 necessity of cutting the pavement. 
 
 Sidewalks. — For the purpose of constructing walks for foot 
 travel alone, cement concrete has no superior. 
 
 Materials. — None but the best materials should be used. 
 Portland cement, clean, sharp sand, and gravel or broken stone. 
 If broken stone is used it should be broken somewhat finer 
 than for macadam, as the walks are not made more than four 
 inches deep. The walks are made of two classes of concrete, 
 3i .inches of ordinary concrete consisting of cement, sand, 
 gravel or broken stone. On top of this is a finishing coat of 
 I an inch or less, made of a rich cement mortar alone. Many 
 attempts have been made to use the natural cement for the 
 base and the expensive Portland cement for only the finishing 
 coat. 
 
 Such experiments have as yet resulted in failure, the top 
 coat invariably separating from the base and breakincr off. 
 
J 28 PORTLAND CEMENT. 
 
 This is probably due to the fact that, the Portland cement 
 mortar was put upon the base before the base was set and the 
 difference in the amount of contraction of the two concretes, 
 while setting, prevented a union being formed. With our 
 present knowledge of the making of cement walks, only the 
 best Portland cement should be used throughout. 
 
 Foundation. — In localities subject to low temperatures much 
 care should be given to the foundation below the concrete. 
 The earth is excavated to a depth of 8 or lo inches below the 
 required surface of the walk and from 4 to 6 inches of sand or 
 furnace cinders spread and well tamped. 
 
 Frames. The necessary frames are usually made of 2x4 
 
 inch studding. Suppose the walk is to be 6 feet wide. On 
 one side of the walk is set a piece of 2 X 4 on edge, the top 
 being just the height of the finished walk. This is held in 
 place by means of stakes driven on each side, those on the 
 outside being much the stronger. Down the center of the 
 walk is placed another frame, at such a distance from the out- 
 side frame that the clear distance between them is one-half the 
 width of the walk. Two cross pieces are put in at such a 
 distance apart that one batch of concrete will fill the interven- 
 ing space. For framing around curves, thin ceiling stuff may 
 be used, or sheet iron. 
 
 Mixing the Concrete. — The concrete is mixed by hand, 
 and in the same manner as described on page 118. The work 
 is done in two boxes, one large one for the heavy layer, and 
 a smaller one for the top coating. The lower layer should be 
 mixed with as litde water as possible. The top coating has 
 much more water, and should be of such a consistency as to 
 work well under a trowel. 
 
 Manipulation.— The frames are filled with the coarse con- 
 crete until it comes up even with the top. It is smoothed off 
 with a straight edge, resting on the two frames and moved 
 back and forth. It is then tamped with small, fiat, iron 
 tampers, the tampers having a smooth face 4 inches square 
 and a wooden handle. The concrete is compressed suffi- 
 ciently by the tamping to give the right depth for the top 
 
CEMENT WALKS. 1 29 
 
 coating. After the tamping the surface is quickly smoothed 
 off with a trowel. Before the top coating is put on this 
 bottom coat is cut into blocks of any desired size, usuall}^ 
 .about 3 feet square. This cutting is done with a trowel, 
 worked along a straight edge, the trowel being forced en- 
 tirely through the concrete. The surface being smoothed off, 
 the top coating is applied. This is spread quickly and evenly 
 with a wooden float and brought flush with the top of the 
 frames. It is struck perfectly true with a straight edge, and 
 then troweled down smoothly. The troweler works with a 
 smoothing trowel in each hand and puts considerable pressure 
 upon them. After the top coat has been smoothed the 
 character of the desired surface decides the next step. If a 
 perfectly smooth surface is desired, some neat cement is 
 sifted over the top, and the whole surface polished down. 
 When a rougher surface is desired, the top is smoothed down 
 and then some coarse dean sand sifted over it. The troweler 
 goes lightly over the surface to partially imbed the sand. 
 Another method is to indent the surface. A small brass 
 roller is used, the surface of which is covered with pointed 
 studs that, in passing, indent the surface. After the surface 
 has been finished it is marked off in Hues to correspond with 
 the cuttings through the concrete. The marking is done by 
 means of a chalk line stretched between the ends of the 
 desired mark. This line is then pressed into the soft mortar 
 with a small trowel, and being removed, leaves the mark. 
 After the surface has been finished the walk is left to harden. 
 For several days the surface should be kept wet. This is 
 particularly necessary in dry, hot weather, in order to avoid 
 the appearance of sun cracks. These so-called sun cracks are 
 fine, hair-like cracks that may often be seen upon the surface 
 of finished cement work. They do not in any way impair the 
 durability of the work, and are due to the too quick drying of 
 the surface. 
 
 As the work of laying the concrete advances, the small 
 stakes on the inside of the frames are removed. 
 
 In a few days the frames are removed and are used over 
 
130 
 
 PORTLAND CEMENT. 
 
 and over again. When the other side of the walk is put in, 
 the concrete ah-eady in place serves as a frame on that side. 
 
 Cement Curbing. — The frames for cement curbing should 
 be made of two-inch plank dressed on one side. The width 
 of the frame should equal the depth of the curbing. The planks 
 are held together by cleats nailed on the rough side. These 
 cleats should be from 3 to 4 inches wide and at one end of 
 each section the cleat should project half its width beyond the 
 ends of the plank. The ends of the plank should be cut off 
 true and square, and when the sections are fitted together, the 
 end of one section will come inside this projecting cleat and 
 thus be held in line. The frames are held in place by means 
 of heavy stakes driven on the outside and pieces of board 
 sawed to the desired thickness of the curbing, on the inside^ 
 holding the frames apart. 
 
 The concrete for curbing is made of cement and sand, no 
 gravel or broken stone being used, unless the curbing is of 
 extraordinary thickness. This concrete is mixed about the con- 
 sistency of damp sand, filled in between the frames and thor- 
 oughly tamped. The top is finished off with a trowel, if 
 square, or to any desired cross-section by means of a template. 
 The frames are not removed until the concrete has become 
 thoroughly set. The back frame is removed first and the earth 
 tamped back in place. 
 
 The number of men in a gang for making cement walks is 
 four. One troweler and three helpers. One of these helpers 
 assists on the frames and tamps the concrete. The other two 
 mix the mortar and concrete. 
 
 Outfit. — Two tampers, i axe, i sledge hammer, i saw, i 
 hand hammer, 2 hoes, 2 shovels. 2 wheelbarrows, i large 
 mixing box 14 ft. long, 3 ft. wide and i ft. deep. 
 
 One small mortar box, 2 or 3 wooden buckets, i straight 
 edge, I pointing trowel, 2 floats, 2 smoothing trowels, i chalk 
 line, I sieve; gravel screen if necessary. 
 
 In straight work, when no time is lost setting the frames, 
 one troweler should keep three helpers busy and should lay 
 and finish 500 square feet per day. 
 
REPAIRING MASONRY, 
 
 For a gang of four men, the following amounts of material 
 will make a convenient amount of concrete and mortar to 
 handle and an amount that can be handled in such a time as to 
 obviate any danger of an incipient set. 
 
 The proportions given are correct for thoroughl}^ good 
 cement walks and have been proved by the author. 
 
 Concrete Base. — One and one-half bbls. Portland cement, 
 I cu. yd. river gravel and coarse sand. 
 
 Top Coating. — One-eighth cu. yd. sharp clean sand, J bbl. 
 Portland cement. This amount will make between 90 and 
 100 square feet of cement walk 4 inches thick. 
 
 Making in Molds. — Slabs for cement walks and curbing 
 may be made in molds under cover and stored away for future 
 use. They are then put in place and handled in the same 
 manner as sandstone. When made in molds the slabs are 
 not more than 2 feet square usually. Any desired color can 
 be given to the work by mixing this color with the mortar. 
 
 Repairing Masonry. — Owing to the immense amount of 
 masonry construction that has been done in this country during 
 the last fifty years, there have been many instances where an 
 inferior quality of building stone has been used in structures 
 of considerable importance. In some cases the stone did not 
 possess good "weathering" qualities and soon began to disin- 
 tegrate. This type of masonry has been very frequently 
 repaired by the use of Portland cement mortar. The exposed 
 face of the masonry being covered with a coating of mortar, 
 put on in the same manner as plaster. 
 
 The surface to be so repaired should be thoroughly cleaned, 
 all dust, dirt and vegetable growth removed, and then thor- 
 oughly wet, before the mortar is applied. 
 
 For good work the mortar should be proportioned i part 
 cement and not more than 3 parts clean, sharp sand. Suf- 
 ficient water should be used to give it the consistency of lime- 
 mortar used for plastering. It can be applied with a trowel 
 and then worked down and smoothed. 
 
 Another method of applying the mortar is as follows: 
 some of the details of the method have been patented. 
 
132 PORTLAND CEMENT. 
 
 The surface is cleaned and wet, all of the joints are thor- 
 oughly scraped out and all loose mortar or stones removed. 
 
 A shield of plank is put up at any distance from the masonry, 
 so that the distance between it and the masonry shall be the 
 thickness required for the cement coating. A water tight 
 joint is made between the shield and the masonry to be 
 repaired. 
 
 The cement mortar is made very thin by the excessive use 
 of water. This form of mortar is called " Grout." This 
 semi-liquid mass is then forced into the space back of the 
 shield and into all the interstices of the masonry by means of a 
 force pump. 
 
 In this manner, arches, piers, retaining walls, etc., have been 
 repaired in a most satisfactory manner.* 
 
 Artificial Stone. — In the different uses of cement thus far 
 spoken of, the concrete has been put in place and allowed to 
 harden there. There is, however, another most important 
 method of using concrete and that is molding it into blocks of 
 any required shape allowing it to harden in the molds and 
 then using it in the form of artificial stone, in the same manner 
 as any other variety of stone. Artificial stone blocks of great 
 size have been used in large quantities at different ports along 
 the Mediterranean for the purpose of building break- waters 
 and sea-walls. There was no good building stone obtainable 
 within a reasonable distance and there were all the necessary 
 constitutents for good concrete in abundance. Lime, sand 
 and clay. 
 
 These blocks were usually what is known as " Coignet's 
 Bt^ton Agglomere" and the ingredients were more or less, 
 hydraulic lime and fine sand. The jetties of the Suez Canal 
 are built of Beton Agglomdre composed of hydraulic lime of 
 Theil and the exceedingly fine sand of the desert. The blocks 
 weighed about 20 tons and were allowed to harden some two or 
 three months in the air before they were immersed in the sea. 
 
 *For a full description of this process, see Engineering News. 
 
MONOLITHIC STRUCTURES. 
 
 The result has been satisfactory although no engineer 
 would take the chances of failure by using only hydraulic 
 lime when by the addition of a little Portland cement the 
 concrete would have been so vastly improved and all possible 
 danger done away with. This Beton has been used in great 
 quantities all over France for the building of sewers, aqueducts 
 and buildings of all sorts and sizes. 
 
 In the manufacture of Beton, it will be noticed that no broken 
 stone or gravel is used. Some type of cement or lime and 
 sharp sand. In France at least its use has been attended with 
 marked success. One of the largest works in which it has 
 been used in monolithic form' is the Vanne Aqueduct, that 
 supplies Paris with water, 37 miles in length. There are 2 
 or 3 miles of arches, some of them are 50 feet high, 8 or 10 
 arch bridges 75 to 100 feet in span. The pipe, itself, is 6i- 
 feet interior diameter, 9 inches thick on top and 12 inches on 
 the sides near the water level. It appears to be impermeable 
 to water. Nearly 40 miles of sewers, in Paris have been 
 built of the same material and have been satisfactory. 
 
 In this country, no monolithic sewers or aqueducts of cement 
 concrete have been constructed, for the reason that iron, stone- 
 ware pipe and brick are better suited to the purpose at a 
 much less cost. The locks on the Hennepin canal are mono- 
 lithic structues of Portland cement concrete. The materials 
 used are imported Portland cement, river gravel and sand, 
 and broken stone. Plate I. shows the stone crushing plant. 
 Plate II. shows the concrete mixing machine, which consists of 
 a box hung at diagonal corners and capable of being revolved. 
 Plates III. and IV. give different views of the locks, more or 
 less completed. A good idea of the massive structure of the 
 work can be obtained from these plates. Heavy timber frames 
 are erected and the concrete filled in between, in layers and 
 tamped. The whole mass is kept wet during the process of 
 settino" and in order to insure this for the interior of the mass 
 vertical holes were left in the concrete and these were kept 
 filled with water until such a time, that none of it was absorbed 
 by the concrete. 
 
134 PORTLAND CEMENT. 
 
 Probably the most striking examples that we have of mono- 
 lithic concrete structures are some of the large hotels in Florida 
 and parts of California. 
 
 Plate V. is an exterior view of the Museum Building of the 
 Leland Stanford Jr. University, and Plate VI. shows an interior 
 view of the same. This is a monolithic concrete structure and 
 was built by Ransom, Smith & Co., of San Francisco, California. 
 
 Plate VII. shows a small bridge built upon the University 
 grounds and is a good example of what is possible in the direc- 
 tion of imitating ornamental stone work. 
 
 Portland cement concrete, if allowed a sufficient time to 
 harden, is capable of offering great resistance to wear of any 
 .kmd and has great durabilty in resisting the action of the 
 elements. But a number of months, at least, are necessary 
 before this great resisting power becomes developed. The 
 fact that such a length of time is necessary before Portland 
 cement concrete possesses the resisting power of good building- 
 stone, has been a very serious drawback to its general use 
 as artificial stone for finer ornamental and architectural pur- 
 poses, unless "the artificial stone was carefully stored under 
 cover for a considerable period of time the finer details would 
 become effaced upon exposure. The better the ultimate 
 qualit}'- of the concrete the slower it is in hardening, and con- 
 sequently the longer time it must be stored before it can be 
 used W'ith safety. When used in comparatively large masses 
 or in such a form as not to render fine delicate tracery neces- 
 sary, the small amount of disintegration that takes place is 
 of no importance. The facility and ease with which concrete 
 may be molded into any required shape, makes it a most 
 desirable material for many constructive purposes. 
 
 These two facts soon led to experiments being made to 
 obviate this defect, and also to produce some solution that 
 could be applied not only to artificial stone but also natural 
 building stones that would render them impervious to w^ater 
 and to the action of vitiated atmospheres. 
 
 The application of a liquid glass has been found to be among 
 the most beneficial. 
 
SILICATING PROCESSES. 
 
 An analysis of the compound used is about : 
 
 Silicic acid 23.21 
 
 Soda, 8.90 
 
 Potash, 2.52 
 
 Water, 65.37 
 
 100.00 
 
 This compound, when sufticiently diluted with water and 
 •applied to the surface of stone, either natural or artificial, 
 forms a thin surface of glass over it, that presents a resistance 
 to the elements, that would not be obtained by the concrete 
 alone for many months and never by the natural stone. 
 
 The most successful application of this silicating process 
 has been made in England in the manufature of paving blocks. 
 The following description of which is taken from Henry Reid 
 on the Manufacture and Use of Portland Cement, page 383. 
 
 Victoria Stone. Aggregate. The best granite, broken 
 finely and carefully washed to remove dust and any claye\^ 
 material. Only the very best Portland cement is used. This 
 is a sine qua non. In the molding room the materials are 
 mixed in varying proportions, to suit the future requirements 
 •of the blocks. The slabs are made with both sides alike, so 
 that when one side becomes worn it may be turned over. 
 After being molded the slabs are kept at least seven days 
 before being dipped in the siHca tanks. It is allowed to stay 
 in the baths some ten days and then stored away for use. 
 
 Concrete made of the best materials and with the greatest 
 care has considerable power of absorption, and it is this 
 ■quality that is taken advantage of in this process. 
 
 * This silicating process would only be allowable in concrete 
 made of Portland or some extremely hydraulic cement, for 
 the reason that this double silicate of soda and potash which 
 is formed completely excludes the air from any communication 
 
 * See "A Practical Treatise on Natural and Artificial Concrete," by 
 Henry Reid, page 154. There is quite a detailed article in this book upon 
 artificial stone and the silicating process. 
 
136 
 
 PORTLAND CEMENT. 
 
 with the interior of the mass, and if the concrete were made 
 of lime as a matrix the interior never would become hard. 
 
 There is another compound that in solution has been found 
 to render masonry impervious to water, Sylvester's Process 
 for Repelling Moisture from External Walls." This process 
 consists of the application of two washes to the exterior face 
 of the masonry. 
 
 No. I — Castile soap, -| lbs. 
 
 Water, i gal. 
 
 No. 2 — Alum, 1- lb. 
 
 Water, 4 gals. 
 
 Apply No. I as near the temperature of boiling as possi- 
 ble with a stiff brush and rub until the masonry is covered 
 with a lather. Permit to dry 24 hours, and then appl}- 
 solution No. 2. In 24 hours repeat this; and on ordinary- 
 brick masonry about three applications are usually sufficient.* 
 
 The properties of this solution have led to the patenting of 
 the so-called McMurtie Stone. This stone is made of Portland 
 cement concrete, in the pores of which are formed the com- 
 pounds of alumina and the acids resulting from the double' 
 decomposition of alum and soap. These compounds are 
 insoluble in water, and are not acted on by the carbonic acid 
 in the air. The early strength of the stone is increased, and 
 the ultimate strength not at all diminished. 
 
 Frear Stone. — Composed of clean, sharp sand and Port- 
 land cement, to which is added some gum shellac. 
 
 Proportions — 
 
 I part cement. 
 i\ parts sand. 
 
 I ounce shellac to i cubic foot of stone. 
 
 Rammed in molds and stored for some months. Used some 
 in Chicago, but not always with satisfactory results. 
 
 Ransom Stone. — Crushed and washed granite, sand, or 
 gravel. 
 
 *See Baker's Masonry 
 
 Construction, p. 
 
ESTIMATES OF QUANTITIES. 
 
 METHOD OF MANUFACTURE. 
 
 One gallon silicate of soda mixed with i bushel of sand or 
 other aggregate, rammed into molds; immersed, under pres- 
 sure, in hot solution of chloride of calcium; thoroughly washed 
 in cold water. Result most excellent when good materials 
 are used. 
 
 Sorel Stone. — Cement, oxychloride of magnesium; formed 
 by adding solution of chloride of magnesium to the oxide of 
 magnesium. The proper strengths and proportions being used 
 a cement of great hydraulic energ}' is the result. 
 
 Used with properly powdered stone of good quality, it 
 makes the hardest and strongest artificial stone yet produced^ 
 and is in every way equal to the natural stone that furnishes 
 the powder. It is used principally for emery wheels, imitation 
 marble, soapstone stoves, etc. 
 
 ESTIMATES OF QUANTITIES. 
 
 Lime. — ^One barrel unslaked lime weighs about 230 pounds. 
 One barrel unslaked lime will make about 2 \ barrels of stiff 
 lime paste, equal to 0.3 cu. yd. 
 
 One barrel of lime paste and three barrels of sand will make 
 three barrels of mortar, equal to 0.4 cu. yd. One barrel of 
 unslaked Hme will make 6.75 barrels of good i to 3 mortar,, 
 equal to 0.95 cu. yd. 
 
 Cement. — Portland weighs 400 pounds per barrel gross and 
 about 375 pounds net. 
 
 Rosendale 300 pounds net per barrel. 
 
 Milwaukee and Utica about 260 pounds net. 
 
 Portland as packed will measure about 1.2 barrels loose. 
 
 Rosendale about 1.25 to 1.40. 
 
 Milwaukee and Utica about i.i. 
 
 One cubic foot dry cement shaken down but not pressed 
 will make about 0.63 cu. ft. stiff paste, when mixed with 25 to 
 30% of water. 
 
 One barrel of Rosendale will make about 3.75 cu. ft. stiff 
 paste, or about 80 pounds Rosendale will make one cu. ft. stiff 
 paste. 
 
138 
 
 PORTLAND CEMENT. 
 
 Volume for volume Portland will make about the same 
 •nmount of paste as the naturals That is lOJ pounds of Port- 
 land will make one cu. foot of paste. 
 
 In mixing mortar in large quantities the units of measure- 
 ment are usually a packed barrel of cement and a loose barrel 
 of sand. 
 
 CEMENT AND SAND REQUIRED FOR ONE CUBIC YARD 
 OF MORTAR. 
 
 O 
 
 Portland in barrels, 7-i4 
 
 Utica, 6.42 
 
 Sand cu. 3'ds., . . o 
 
 PROPORTIOX OF .SAXD TO CEMEXT. 
 
 123456 
 
 4.16 2.85 2.00 1.70 1.25 i.iS 
 
 3.73 2.57 1.80 1.53 1. 13 1.06 
 
 0,67 0.84 0.94 0.98 0.99 1. 00 
 
 The cement is given in barrels packed, the sand in cubic 
 vards. 
 
 AMOUNT OF MORTAR REQUIRED FOR A CUBIC YARD OF MASONRY. 
 
 DESCRIPTION OF MASONRY. 
 
 Concrete; broken stone; no gravel in screenings 
 
 Rough rubble. 
 
 Rough jointed rubble 
 
 Squared stone masonry 
 
 Ashlar i2"-2o" courses | " to -| " joints 
 
 " 2o"-3o" i" to I" 
 
 largest blocks and courses 
 
 Brickwork; |" to 4'' joints 
 
 " -■' " to i " " . 
 
 a J " u * 
 
 8 
 
 VOL. OF 
 TAR CU. 
 
 MIN. I 
 
 0.50 
 
 0.33 
 0.25 
 0.15 
 
 0.07 
 0.05 
 0.03 
 
 0.35 
 0.25 
 
 O.IO 
 
 MOR- 
 YDS. 
 MAX. 
 
 0.55 
 0.40 
 0.30 
 0.20 
 0.08 
 0.06 
 0.04 
 0.40 
 0.30 
 0.15 
 
 Trautwine's Engineer's Pocket Book, p. 679: 
 For I cu. yd", of concrete of broken stone and sand, without 
 voids, I cu. yd. broken stone with 0.5 bulk voids, requiring 
 0.5 cu. yds. of sand and 0.25 cu. yds. of cement. Abstracted 
 from Gillmore on Limes, Cements, etc., p. 321. 
 
 *Up to this point this da/a for estimates is talven from Baker's Masonry 
 Constrviction, p. 87, by permission of the author. 
 
ESTIMATES OP CONCRETE. 1 39 
 
 CONCRETE NO. i. 
 
 I bbl. German Portland cement, ) , . , 
 
 - 5.4 bbls. concrete mortar. 
 51- bbls. loose sand, ) 
 
 6 bbls. gravel, ^ 12I bbls. mixed and shaken down, con- 
 ■9 bbls. broken stone, j' taining 26}^ % voids. 
 
 The above makes 50 cu. ft. of rammed cement. 
 
 CONCRETE NO. 2. 
 
 I bbl. German Portland cement, ) , , , 
 
 V 5.7 bbls. concrete mortar. 
 -6 bbls. damp, loose sand, ) 
 
 S bbls. gravel, / ^ . -j 
 
 ^ ^ , - contams 30 % voids. 
 
 9 bbls. broken stone, ) 
 
 The above makes 50 cu. ft. rammed concrete. 
 
 CONCRETE NO. 3. 
 
 I bbl. French Portland cement, ) 
 
 I bbl. slaked, ground Hme powder, - 7 bbls. mortar. 
 
 7 bbls. loose sand, ) 
 
 13 bbls. gravel, | 2 2|- bbls. mixed together and shaken 
 
 13 bbls. broken stone, \ down with 24 % voids. 
 
 The above makes 86 1- cu. ft. of rammed concrete. The 
 strength is good, crushing at the end of two months at 300 
 pounds per square inch. 
 
 CONCRETE NO. 5- 
 I bbl. French Portland cement, \ 
 
 i| bbls. slaked, ground hme, I 7-9 bbls. of concrete mortar. 
 8 bbls. loose sand, ) 
 
 16 bbls. gravel, \ 28 bbls. mixed together and shaken 
 
 16 bbls. broken stone, ( down with 24 % of voids. 
 
 One batch of the above makes 105 cu. ft. rammed concrete, 
 suitable for ordinary uses. 
 
140 PORTLAND CEMENT. 
 
 CONCRETE NO. 5. 
 I bbl. Rosendale cement, ] 
 
 3 bbls. damp, loose sand, -3.27 bbls. of concrete mortar. 
 
 5 bbls. broken stone, 
 
 The above makes 21.75 ^u. ft. of rammed concrete. 
 
 CONCRETE NO. 6. 
 
 I bbl. Portland cement, ^ 
 
 I bbl. unslaked lime, [ 
 
 •r^ 1 A r IO-37 bbls. of concrete mortar. 
 
 10 bbls. loose sand, l ^' 
 
 ' 1 
 
 16 bbls. broken stone, J 
 
 The above makes 69I- cu. ft. of concrete rammed in place. 
 
\Ti: VI. Interior of Museum Building, Leland Stanford 
 University. Palo Alto, California. 
 
 .Altixdi.riHic Construction' undick Ransom's Concrktk and Twisted Iron Patents. 
 Ransom, Smith & Co., loi Sansome St., San Francisco, Cal. 
 
CHAPTER IX. 
 CEMENT TESTS. 
 
 Engineering Department, State UxNiversity of Iowa. 
 
 The Cement Tests, of which this chapter is a report, were 
 made in the Cement Laboratories of the Engineering Depart- 
 ment of the State University of Iowa, Iowa City. 
 
 The entire work was done by various engineering students 
 and under the personal supervision and direction of the author. 
 The different tests are here numbered, merely for conveni- 
 ence. The numbers indicate nothing however as to the order 
 in which the work was done. The results of each test were 
 wa"itten by the different students and what follows are abstracts 
 from these various reports: 
 
 Cement Test No. i. 
 
 The tests were made by E. W. Crellen, J. H. Howe, and 
 Hubert Remley, class of '90. Mr. Remley completed the 
 tests and wrote the report. (See The Transit, Vol. I., No. 2.) 
 
 The points studied in these tests were as follows: 
 
 («) The difference in strength of briquettes of the same 
 age and same cement, due to being allowed to harden in dry 
 air and to hardening under water. 
 
 The rate of increase in strength in the different cements 
 due to time, under the two above mentioned conditions. 
 
 (c) The actual breaking strength per square inch of the 
 various cements tested at different ages. 
 
 ((i) The differences that exist in the behavior of the hght- 
 burned natural cements and the Portland cements. 
 
 Temperature. — The temperature of the laboratory was 
 between 60° and 70° Fah. and the temperature of the water 
 was from 5° to 10° lower. 
 
142 PORTLAND CEMENT. 
 
 Immersion Tanks. — The tanks were so arranged that the 
 water was constantly changing. 
 
 Briquettes. — The size and shape of the briquette was the 
 same as that shown in Fig. 10, page 45. 
 
 Molds. — For hand made briquettes the molds used were 
 similar to those shown in Fig. 11, page 46. 
 
 Briquette Machine. — The Jameson Briquette Machine was. 
 used, Figs. 17, t8, 19, 20; pp. 49-52. 
 
 Testing Machine. — The testing machine, used for break- 
 ing the briquettes was Reihle Bros, i.ooo' pound cement tester,. 
 Fig. 27, page 99. 
 
 Making Briquettes. — When the briquettes were to be 
 made by hand, the cement and water were mixed upon a large 
 glass slab. When the briquette machine was to be used, the 
 mixing was done in a small iron sink. 
 
 Cement. — All of the cement tested was furnished, free of 
 cost, by the makers. The makers were informed that the 
 cement was to be used for a series of laboratory tests. 
 
 Treatment of Briquettes. — All briquettes after being made 
 were placed upon some non-absorbent surface, such as glass, 
 and covered with a wet cloth. At the end of twenty-four 
 hours the cloth was removed and a portion of the briquettes 
 were immersed and the remainder, allowed to harden in the air. 
 
 Age of Briquettes. — The age of the briquettes given in the 
 following tables, and shown in the diagrams, dates from the 
 time of the removal of the wet cloth, or tv/enty-four hours after 
 the briquettes were made. 
 
 Breaking Briquettes. — Those briquettes that hardened un- 
 der water were taken from the water and immediately broken. 
 
 The clips, on the testing machine were, as those shown in 
 Fig. 23, page 53, with rubber buffers. 
 
 Sand. — All sand used was a sharp, white, flint-glass sand 
 from near Aurora, 111. The size used was what passed through 
 a No. 20 sieve and was caught on a No. 30 sieve. 
 
 PERCENTAGE OF FINENESS OF CEiMENT.* 
 
 The percentage of fineness should be found for each cement 
 
 *What follows is taken directly from Mr. Remley's report, 
 Vol. I., No. 2. 
 
 , — The Transit 
 
CEMENT TESTS. ' 143 
 
 since the degree of fineness to which it is ground, affects the 
 strength to a considerable extent. A quantity of the cement 
 is carefully weighed, and sifted through a sieve having 2,500 
 meshes to the square inch. It must be carefully shaken until 
 no more will pass through, and this can be shown by sifting 
 over white paper. The residue is then weighed and the per- 
 centage of fineness is found for this cement with this sieve. 
 What passes is then sifted through a sieve having 10,000 
 meshes to the square inch, the same care being taken as before, 
 to pass all that will pass. The amount remaining in the sieve 
 is then weighed and the weight added to the weight of the 
 residue remaining in the coarser sieve and subtracted from 
 the weight of the whole. This s/ion/d give the weight of what 
 passes both sieves, but some of the cement is lost in the pro- 
 cess, passing off into the air in the shape of dust, so that the 
 residue only which is heavy and coarse, and not liable to be 
 blown away, should be taken into account in determining the 
 percentage of fineness, the total amount being divided into the 
 difference of the whole and residue remaining. The sieves 
 used in these experiments were of Richie's manufacture. 
 
 RESULTS OF EXPERIMENTS WITH CONCLUSIONS. 
 
 The following gives the brands and corresponding numbers 
 used : 
 
 TA.BLE I. 
 
 No. 
 
 Hra.n'u 
 
 Class'. 
 
 Address. 
 
 No. 
 
 I. 
 
 
 
 Natural. 
 
 Milwaukee, Wis. 
 
 No. 
 
 
 K. 13. & S. 
 
 
 Portland. 
 
 London Eng. 
 
 No. 
 
 3' 
 
 Louisville .... 
 
 
 Natural. 
 
 Louisville, Kj. 
 
 No. 
 
 4- 
 
 Empire (Ivond'n 
 
 Brand) 
 
 Portland. 
 
 \Varners, N. Y. 
 
 No. 
 
 S- 
 
 (jibb's (Diam'd 
 
 Brand) 
 
 Portland. 
 
 Cray's Essex, Ehl;-. 
 
 No. 
 
 6. 
 
 Rosendale 
 
 
 Natural. 
 
 New York State. 
 
 No. 
 
 7- 
 
 F. 0. Norton 
 
 
 Natural. 
 
 New York State. 
 
 No. 
 
 8. 
 
 
 
 Portland. 
 
 London, Eng. 
 
 No. 
 
 9- 
 
 
 
 Portland. 
 
 Germany. 
 
 No. 
 
 10. 
 
 
 
 Portland. 
 
 Belief oniaine, O. 
 
 No. 
 
 II. 
 
 
 
 Portland. 
 
 Antwerp, Belgium. 
 
 No. 
 
 13- 
 
 Millen's Patent 
 
 
 Portland. 
 
 Warners, N. Y. 
 
 No. 
 
 14. 
 
 Utica Black Ball 
 
 
 Natural. 
 
 La Salle, 111. 
 
 No. 
 
 5.S- 
 
 Hoffman's Rosendale.. 
 
 Natural. 
 
 Kingston, N. Y. 
 
 No. 
 
 16. 
 
 South Bend Cement Co. 
 
 Portland. 
 
 South Bend, Ind. 
 
 No. 
 
 17- 
 
 
 
 Portland. 
 
 Warners, N. Y. 
 
* PORTLAND CEMENT. 
 
 WATER Hy\RDENED BRIQUETTES. 
 
 Table 2 gives the results obtained by the breaking of about 
 2,200 hand made briquettes after the expiration of six months 
 time. Excluding those of No. 10, of which only a few were 
 broken, each of the figures is the average strain withstood b}' 
 about 24 water hardened specimens. 
 
 TABLE IL 
 
 
 
 Unsifted Cement. 
 
 
 
 Sifted Cement. 
 
 
 Number. 
 
 
 
 
 
 
 
 
 
 
 
 OF 
 
 Cement. 
 
 Neat. 
 
 I part 
 
 2 parts 
 
 3 parts 
 
 4 parts 
 
 Neat. 
 
 I part 
 
 2 parts 
 
 3 parts 
 
 4 parts 
 
 
 
 sand. 
 
 sand. 
 
 sand. 
 
 sand, 
 
 
 sand. 
 
 sand. 
 
 sand. 
 
 sand. 
 
 No. I. 
 
 333 
 
 222 
 
 186 
 
 
 134 
 
 316 
 
 229 
 
 207 
 
 177 
 
 155 
 
 No. 2. 
 
 623 
 
 443 
 
 279 
 
 189 
 
 I bo 
 
 553 
 
 474 
 
 342 
 
 272 
 
 206 
 
 No. 3. 
 
 325 
 
 253 
 
 156 
 
 143 
 
 lOI 
 
 256 
 
 228 
 
 
 150 
 
 96 
 
 No. 4. 
 
 5i« 
 
 335 
 
 276 
 
 214 
 
 163 
 
 450 
 
 420 
 
 336 
 
 269 
 
 242 
 
 No. 5. 
 
 609 
 
 442 
 
 296 
 
 238 
 
 175 
 
 539 
 
 476 
 
 360 
 
 281 
 
 228 
 
 No. 6. 
 
 362 
 
 237 
 
 178 
 
 132 
 
 121 
 
 334 
 
 256 
 
 204 
 
 152 
 
 131 
 
 No. 7. 
 
 350 
 
 265 
 
 221 
 
 166 
 
 145 
 
 316 
 
 259 
 
 218 
 
 189 
 
 
 No. 8. 
 
 576 
 
 373 
 
 273 
 
 204 
 
 165 
 
 518 
 
 455 
 
 344 
 
 250 
 
 210 
 
 No. 9. 
 
 553 
 
 447 
 
 363 
 
 285 
 
 230 
 
 559 
 
 529 
 
 394 
 
 313 
 
 235 
 
 No. 10. 
 
 669 
 
 
 
 
 
 628 
 
 409 
 
 240 
 
 215 
 
 130 
 
 As before stated, all proportions of sand and cement were 
 taken by weight. This will account for the difference of 
 these results from those usually obtained. The advantage of 
 this method of proportioning is that the proportion can be 
 more exactly made, but since in practice all is done by volume, 
 it may be better to discard this method. 
 
 A series of rough experiments show that the average 
 w^eight of a cubic centimeter of artificial cement dry and 
 pressed down was 136 milligrams, while that of natural ce- 
 ment was only 114 under the same conditions, and that of 
 sand was 156. The ratio of weight to weight is 6 to 5, or in 
 equal weights of both cements the volumes would be in the 
 ratio of -| to -I . 
 
 Taking the weights of unit volumes of natural and artificial 
 cements and sand as 114, 136 and 156 respectively, as given 
 above, the following table was computed, showing the per 
 cent, of volume of cement in briquettes of different propor- 
 tions of sand and cement: 
 
CEMENT TESTS. 
 TABLE III. 
 
 CLASS. 
 
 KEAT. 
 
 I PART 
 
 2 I'ARTS 
 
 3 PARTS 
 
 4 PARTS 
 
 S PARTS 
 
 
 SAND. 
 
 SAND. 
 
 SAND. 
 
 SAND. 
 
 SAND. 
 
 Natural 
 
 100 
 
 57-7 
 
 40.6 
 
 3I-.S 
 
 25-5 
 
 21.4 
 
 Portland .... 
 
 100 
 
 53-4 
 
 
 27.6 
 
 22.3 
 
 
 It will be readily seen from this table that in briquettes of 
 the same proportion of sand and cement, by weight, that there 
 is more of the natural cement by volume than there is of the 
 Portland cement. 
 
 The apparently small difference between the breaking strain 
 of four part sand briquettes of Portland and natural cements 
 will be explained by the difference in weight, as compared 
 with the volume of the two cements. Table III. shows that 
 in a five part sand natural cement briquette there is very 
 nearly the same amount of cement by volume as in a four 
 part sand Portland cement briquette, so that in order to com- 
 pare the relative strength for the same volume of cement we 
 must compare either the neat of the Portland with that of the 
 natural, or the four parts sand Portland with the live parts 
 sand natural. 
 
 Since the different cements of the same class are of differ- 
 ent weight, the same difficulty is involved, and the only exact 
 comparison can be made with the neat cement. It is also on 
 account of the difference of weight that the strength of a 
 heavier cement appears to decrease more rapidly with the 
 addition of sand. For example, the strain withstood by 4"*" is 
 163, while that withstood by 7^*" is 145, but of No. 4 there is 
 only 23 per cent, by volume, while of No. 7 there is 26 per 
 cent, of cement in the briquettes. 
 
 Table II. gives higher breaking strains than are usually 
 given, and this is accounted for by Table III., where it will be 
 seen that the proportion of cement by volume in a briquette 
 proportioned by weight is greater than in one proportioned by 
 volume. Thus, in a four part sand natural cement briquette, 
 the percentage by weight of cement is 20, while by volume it 
 is 25.5, or it must be compared with a three part sand 
 briquette in other tables where the volume is the basis of 
 
146 PORTLAND CEMENT. 
 
 proportioning. As a general rule, the strength of natural 
 cement, when used with four parts sand, can be compared 
 with Portland cement with three parts sand by weight, the 
 difference by volume being slight. 
 
 Taking all things into consideration, together with the fact 
 that the test is not made on a given weight, but on a given 
 volume, it would appear better to take the proportions of 
 sand and cement by volume. 
 
 According to the Trautwine, the average cost of Portland 
 cement is $2.63 per barrel of about 400 pounds gross, or 380 
 pounds net, or 7 cents per pound, while that of the natural 
 cement is $1.21 per barrel of 300 pounds net, or 4 cents per 
 pound. 
 
 Table IV. gives the average breaking strains of the natural 
 and Portland cements. No. 10 excepted, taken from Table II. 
 
 TABLE IV. 
 
 Class. 
 
 Unsifted. 
 
 Sifted. 
 
 Neat. 
 
 I part 
 sand. 
 
 sparts 
 sand. 
 
 Sparts 
 aand. 
 
 4parts 
 sand. 
 
 Neat. 
 
 I part 
 sand. 
 
 2parts 
 sand. 
 
 3parts 
 sand. 
 
 4 parts 
 sand. 
 
 
 
 244 
 408 
 
 185 
 297 
 
 226 
 
 125 
 183 
 
 305 
 524 
 
 243 
 471 
 
 199 
 
 355 
 
 167 
 277 
 
 144 
 224 
 
 The natural cement being only 83.8 per cent, as heavy as 
 the Portland, the comparative cost of a given bulk of the two 
 is in the ratio of .7 to .838 x .4 or .7 to .335. 
 
 In order to attain the same strength with a natural cement 
 as with a Portland, it is necessarv to use a larger amount. 
 Now the question is whether the same strength can be attained 
 for the same cost. The amount that may be economically 
 spent for natural cement to arrive at the same total strength 
 may be found by comparing the strength and cost of unit 
 volumes. In the case of neat unsifted cement the following 
 proportion is had: 567 : .7 : 1342 :x, or the strength of the 
 Portland is to the cost as the strength of the natural is to the 
 amount that could be economically spent, or x equals .415, 
 while the actual comparative cost is .335. From this it 
 appears that it would be justifiable to spend 23.9 per cent. 
 
CEMENT TESTS. 1 47 
 
 more than it actually costs, or .4956 cents per pound to get 
 the same total strength as from the Portland. In the following 
 table the first column shows the proportions of sand and 
 unsifted cement; the second, the amount actually spent for a 
 unit volume of the natural; the third, the comparative amount 
 that can be economically spent to arrive at the same strength 
 as with Portland; the fourth, the percentage of increase over 
 the cost, and the fifth, the number of cents per pound allow- 
 able to spend. 
 
 TABLE VI. 
 
 I 
 
 
 
 4 
 
 5 
 
 Neat. 
 
 •335 
 
 •415 
 
 23^9 
 
 .495 
 
 I part sand. 
 
 •335 
 
 •389 
 
 16. 1 
 
 .464 
 
 2 parts sand. 
 
 •335 
 
 •391 
 
 16.7 
 
 .467 
 
 3 parts sand. 
 
 •335 
 
 .407 
 
 21-5 
 
 .486 
 
 4 parts sand. 
 
 •335 
 
 .418 
 
 24.2 
 
 ■497 
 
 From this it would appear that when the proportion of 
 sand and cement is one to one, that either the comparative 
 cost of Portland is the least or that of the natural is the most, 
 but from other tables it will be seen that the latter is the case. 
 
 Table VII. shows the strain a given bulk of cement will 
 stand when united with different proportions of sand by 
 weight. It was compiled from Tables II. and III,, by divid- 
 ing the strain actually withstood, in Table II., by the per- 
 centage of cement in Table III., for corresponding groups of 
 specimens. 
 
 TABLE Vn. 
 
 CLA.SS. 
 
 Unsifted. 
 
 Sifted. 
 
 Neat. 
 
 I part 
 sand. 
 
 2pai-ts 
 sand. 
 
 3parts 
 sand. 
 
 4parts 
 sand. 
 
 Neat. 
 
 I part 
 sand. 
 
 2parts 
 sand. 
 
 3parts 
 sand. 
 
 4parts 
 sand. 
 
 Natural 
 
 342 
 576 
 
 423 
 764 
 
 456 
 S16 
 
 476 
 819 
 
 490 
 820 
 
 305 
 524 
 
 422 
 881 
 
 490 
 
 975 
 
 530 
 1004 
 
 565 
 1005 
 
 Ter cent, of strength of 
 natural to Portland . . 
 
 59^2 
 
 55-3 
 
 55-9 
 
 58.1 
 
 59^7 
 
 58.2 
 
 47^9 
 
 50.2 
 
 52.8 
 
 56.2 
 
 This table shows that after the addition of one part sand 
 
1^8 PORTLAND CEMENT. 
 
 the rate of total increase of strength is comparatively small, 
 being less in the Portland than in the natural. The percentage 
 of breaking strain of the natural as compared with the Portland 
 is least when the proportion of sand and cement is one to one, 
 continually increasing both in the case of the sifted and the 
 unsifted with the addition of sand until with four parts sand it 
 is nearly the same as with the neat cements. The rate of 
 increase of both the natural and Portland cement continually 
 decreases with the addition of sand, arriving, in the case of the 
 Portland, at nearly the minimum when the proportion of sand 
 and cement is two to one with the unsifted cement, and three 
 to one with the sifted. In both the unsifted and the sifted the 
 strength of the Portland increases more rapidly at first with 
 the addition of sand than the natural, while at about the 
 proportion of four to one of sand and cement the percentage 
 of strength of natural to Portland is almost equal to that of the 
 neat cements. 
 
 The total strength of a given bulk of cement, as indicated in 
 the table is greater with the sifted cement upon the addition of 
 sand because of the larger amount of active material in the 
 cement, the residue caught on the sieves being perfectly inert 
 and acting as so much sand. The fact that a portion of the 
 unsifted cement is inert, does not, however, explain the fact 
 that the percentage of strength of natural to Portland when 
 sifted, is less than when unsifted. Table VII. also shows that, 
 taking the sifted neat as composed almost wholly of active 
 particles, that the strength at first increases rapidly with the 
 amount of sand or inactive material added, overcoming the 
 tendency to an increased strength due to a larger amount of 
 active material, up to and even above the point where the grit 
 is from zero to 25 per cent, of the active material, or in other 
 words, that the unsifted cement having a percentage of fine- 
 ness of from zero to 25 and consequently less active material, 
 is stronrfer than the same bulk of sifted cement with zero to 
 25 per cent, more active material. The fact that the natural 
 cement contains, on an average, 2.5 per cent more grit, 
 explains its higher percentage of strength as compared with 
 
CEMENT TESTS. I^p 
 
 the Portland in the unsifted specimens, because, taking the 
 sifted cement as active, and since the strength increases with 
 the amount of grit within Hmits more rapidly than with the 
 amount of active material added, the strength of the former 
 increases more than that of the latter from one to two percent. 
 
 In using sand and cement, it is necessary to have the cement 
 hnely ground in order that each individual grain of sand may 
 be covered with a layer of cement to cause a firm bond of 
 union. Fine grinding produces a lighter cement, reduces the 
 percentage of fineness, increases the amount of active material 
 for a given weight, and allows a more perfect intermixing of 
 sand and cement. Experiments go to show that as a rule, the 
 heavier the cement, the degree of fineness being the same, the 
 stronger it is, so that either the degree of fineness and the 
 weight per bushel, the weight per bushel and the strength, or 
 the degree of fineness and the strength may be specified, and 
 the cement filling the specifications will be satisfactory in most 
 cases. Buyers should insist on fine grinding as more active 
 material is purchased for the money and with this, insist on a 
 strength with neat sifted cement. The results show that the 
 strength of specimens of neat cement of the same brand dimin- 
 ishes with the percentage of fineness. 
 
 PERCENTAGE OF FINENESS. 
 
 Number. 
 
 50 Sieve. 
 
 103 Sieve. 
 
 No. I 
 
 95-75 
 
 95-1 
 
 84.6 
 
 S9.9 
 
 90.5 
 
 86.55 
 
 92.95 
 
 91-55 
 
 98.71 
 
 99-25 
 
 98.4 
 
 87. 
 
 93-4 
 
 95-15 
 
 87.65 
 84-5 
 
 77-4 
 77.2 
 
 81.8 
 
 78.3 
 87.1 
 78.8 
 93-2 
 95-7 
 88.6 
 
 77-3 
 
 84- 3 
 
 85- 85 
 
 No. 2 
 
 No. 3 
 
 No. 4 
 
 No. s 
 
 No. 6 
 
 No. 7 
 
 No. 8 
 
 No. 9 
 
 No. lo 
 
 No. 13 
 
 No. 14 
 
 No. IS 
 
 No. 16 
 
 Examination of Table II. shows that the strength, as a rule, 
 decreases in the following order: n, s, Is, In, 2s, 2n, 3s, 3n, 4s, 
 4n; and all conclusions drawn from those followino- Table II. 
 
I^O PORTLAND CEMENT. 
 
 will apply to Table II. and to the individual results. Individual 
 specimens show great variations in strength and cannot be 
 trusted to give good results when taken alone, but rather the 
 mean of many specimens must be taken. 
 
 An examination of the books showing the breaking strains 
 of the individual specimens shows that many do not break in 
 the minimum section, and that, as a general rule, those that 
 broke in a larger section withstood a strain greater in propor- 
 tion to the increase of section. Some cements make a point 
 of breaking in other than the minimum section. At sight it 
 might be supposed that on account of this fact the strain 
 recorded was too great, but rather the strain recorded is too 
 small, because the specimen did not break at the smallest, and 
 hence the weakest, portion, but at a larger section because of 
 the pressure of the clips. Nearly 25 per cent, break along a 
 line connecting the points of contact of the clips, some cements 
 being more apt to break here than others. As an example, 
 out of 148 machine made briquettes of No. 11 103 broke 
 along this line, while of 120 of No. 16 only 28 broke in other 
 than the minimum section, or 70 per cent, of No. 11 to less 
 than 22 per cent, of No. 16. Both were neat, unsifted Portland 
 cements. 
 
 Diagrams. — The results of the tests are shown in the 
 diagrams. There is a separate diagram for each brand of 
 cement tested, the breaking strain of the briquettes hardened 
 in air as in the original, being indicated by a dotted curve. 
 The marked dots also belong to the dry cement. The brand, 
 number, place of manufacture, class and percentage of fineness 
 is also indicated on the diagram. The breaking strains of 
 some of the individual specimens are indicated also. That of 
 specimens hardened in water are shown by open circles of a 
 larger diameter than of the black " average dots." The break- 
 ing strains of the air hardened specimens are shown by the 
 circles with a cross within them. Only the individual speci- 
 mens showing a greater strain than the average are thus 
 indicated, and, of course, a circle, which in some cases appears 
 as a dot, when below a line, does not belong to that line. The 
 
CEMENT TESTS. 1^1 
 
 dots occur only at i week, 4 weeks, near 13 and 26 weeks 
 and nowhere else, and this will help to distinguish dots and 
 circles. The circles are of slightly larger diameter also. 
 The greatest difficulty is with the diagrams of Nos. 6 and 15. 
 The large black dots averaging about 75 pounds above the 
 dotted line in No. 6 are supposed to be crossed circles. All 
 in No. 15 should be crossed. One diagram, as marked, shows 
 the average breaking strain of the Portland and natural cements 
 tested, hardened in water. A good idea of the effect of time 
 on the different brands of cement will be obtained from the 
 careful study of the diagrams. Perhaps the first thing that 
 strikes the observer is the irregularity of the curves, but a 
 general curve can be easily traced, the cut balancing the fill, 
 thus giving a regular curve. Since specimens broken on the 
 same day vary greatly, and since the increase of strength due 
 to one week's time is not a great deal, irregularity is to be 
 expected, but as is seen in the diagram showing the average 
 of all of each class, the irregularity is done away with largely, 
 because of the greater number of specimens entering into the 
 averages. 
 
 These experiments are of especial value in showing the 
 effect of time as well as the strength. 
 
 TABLE IX. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0 
 
 rt 
 
 0 
 
 . u 
 
 
 
 
 
 
 
 
 
 CO 
 
 d 
 
 
 
 
 vd 
 
 
 
 c n 4-; 
 
 
 No. 
 
 No. 
 
 No. 
 
 c 
 ?5 
 
 No. 
 
 No. 
 
 No. 
 
 0 
 Z 
 
 No. 
 
 No. 
 
 0 
 
 'A 
 
 No. 
 
 0 
 
 M 
 
 rt 
 
 c S 5 
 
 (U u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I week. 
 
 84 
 
 82 
 
 260 
 
 362 
 
 134 
 
 102 
 
 410 
 
 459 
 
 516 
 
 398 
 
 III 
 
 307 
 
 387 
 
 103 
 
 
 
 I week to i mo. 
 
 41 
 
 33 
 
 130 
 
 102 
 
 27 
 
 68 
 
 76 
 
 148 
 
 126 
 
 114 
 
 J 20 
 
 135 
 
 119 
 
 57 
 
 30-7 
 67.2 
 
 55^3 
 
 I week to 3 mo. 
 
 174 
 
 78 
 
 251 
 
 274 
 
 251 
 
 285 
 
 249 
 
 354 
 
 192 
 
 282 
 
 237 
 
 221 
 
 261 
 
 204 
 
 189. 
 
 I week to 6 mo. 
 
 262 
 
 282 
 
 353 
 
 341 
 
 305 
 
 339 
 
 
 386 
 
 265 
 
 371 
 
 245 
 
 
 355 
 
 252 
 
 91.7 
 
 2<(4.6 
 
 I mo. to 3 mo. 
 
 133 
 
 45 
 
 121 
 
 172 
 
 224 
 
 217 
 
 173 
 
 206 
 
 66 
 
 168 
 
 117 
 
 86 
 
 142 
 
 147 
 
 28.1 
 
 CI.9 
 
 I mo. to 6 mo. 
 
 221 
 
 249 
 
 223 
 
 239 
 
 278 
 
 271 
 
 
 238 
 
 139 
 
 257 
 
 125 
 
 
 236 
 
 195 
 48 
 
 46.6 
 
 121. 9 
 
 3 mo. to 6 mo. 
 
 88 
 
 204 
 
 102 
 
 67 
 
 54 
 
 54 
 
 
 32 
 
 7? 
 
 89 
 
 8 
 
 
 94 
 
 14-5 
 
 15.6 
 
 The gain in pounds in any cement from any one of the four 
 dates to any other may be seen. The " average " columns were 
 taken from the averages of the strains of natural and Portland 
 cements and the per cent, columns give the percentage of the 
 increase of strength between the times indicated, being calcu- 
 
152 
 
 PORTLAND CEMENT 
 
 lated by combining the absolute strength with the increase of 
 strength indicated in the "average" columns. The first line 
 gives the actual breaking strain in pounds at the expiration of 
 one week. From this it is seen that the natural cements, 
 Nos. I and 3, are considerably below the average at this age, 
 and that below three months, the increase is less, but from 
 three to six months, the increase is more rapid than that of 
 the average, being very apparent in the case of No. 3. Nos. 
 6, 7, and 14 are above the average, and the curves lie very 
 near together. Of the Portland cements. No. 11 is noticeable 
 on account of the rapidity with which it attains its strength, 
 reaching 516 pounds in seven days. Of course, the increase 
 after this spurt is less than that of the average. No. 10 
 appears to increase but little after three months, but attains a 
 high strength before that time, the strength being way above 
 the average. No. 13 with an initial strength at the end of 
 the first week a little above the average increases at a rate 
 considerably above it. Nos. 4, 5, and 16 are below the aver- 
 age, both in the point of initial and increase of strength. 
 It will be noticed that all the natural cements and all of the 
 Portland show nearly the same increase of strength, the Port- 
 land increasing more rapidly than the natural. I'he actual 
 difference between the strength of Portland and natural is at 
 the end of — 
 
 One week. 
 One month, . 
 Three months, 
 Six months, . 
 Average, 
 
 284 pounds. 
 
 346 " 
 
 341 " 
 387 " 
 
 337 " 
 
 After one month the increase of strength is very nearly the 
 same for the average of both natural and Portland, and for the 
 individual brands entering into the averages, so that a month 
 test would give a very nearly correct idea of the comparative 
 strength of the different brands up to six months. Thus when 
 a set of specimens is broken after one month, the breaking 
 strain for three and six months may be very approximately 
 
CEMENT TESTS. 
 
 determined b}' adding tiie mean of the increase of average 
 strength given, or 145 pounds and 215 pounds respectively to 
 the breaking strain at one month. The table showing the 
 per cent, of increase is apt to give the impression that the 
 natural increases more rapidly than the Portland, but when it 
 is remembered that the absolute strength of the natural at six 
 months is about one-half that of the Portland, it will be readily 
 seen that although the per cent, of increase is greater the 
 actual increase is not. All experiments carried on elsewhere 
 go to show that the Portland increases in strength after six 
 months and more rapidly than the natural. 
 
 The percentage of strength of natural to Portland is at — 
 
 Air Hardened Briquettes. — The dotted broken line on the 
 diagrams shows the increase of strength with time in the same 
 manner as the full line shows the strength of the "wet" 
 specimens. 
 
 Perhaps the first thing that engages the attention is the 
 irregularity of the breaking strains. Some cements attain a 
 certain strength in a very short time and weaken with age, 
 while others act more in the manner of the water hardened 
 specimens. The dry briquettes harden more rapidly than the 
 wet ones, but ultimately attain much les3 strength. Natural 
 cement briquettes, hardened in air, reach a greater strength 
 than those hardened in water, and usually hold it for about 
 three months, but after that time the latter is the stronger. 
 
 For the first three or four days after the immersion of the 
 briquettes the Portland cement hardened in air seems to hold 
 the supremacy in the point of strength, but it soon loses it. 
 
 Table XII. gives the average breaking strains of the natural 
 and Portland cements of the groups of five and ten mentioned 
 in the scheme, at — 
 
 TABLE X. 
 
 Three months. 
 
 Six months, ... 
 
 Hand made, six months 
 
 One week. 
 One month. 
 
 26.6 % 
 31-6 % 
 47-4 I0 
 47-3 % 
 59-3 % 
 
154 PORTLAND CEMENT. 
 
 TABLE XIL 
 
 Time. 
 
 Natural. 
 
 Portland. 
 
 In air. 
 
 In water. 
 
 In air. 
 
 In water. 
 
 I week. 
 I month. 
 3 months. 
 6 months. 
 
 1 66 
 
 275 
 330 
 314 
 
 103 
 160 
 307 
 
 355 
 
 313 
 44S 
 
 463 
 536 
 
 3S7 
 506 
 648 
 742 
 
 From this table it will be seen that the natural cement 
 hardened in air seems to lose its strength after three months. 
 The drop in strength from 340 to 314 is partially due to the 
 fact that at the age of three months No. 3 (Diagram No. 3) 
 possessed an abnormal strength, but since there was no fault 
 to find with the specimens broken, the results obtained from 
 their breaking could not be thrown out just because of their 
 strength. The diagram of No. 3 will also show the actual 
 difference between the breaking strains of these five and of 
 those composing the curve. But the fault does not lie wholly 
 in No. 3, for Nos. 14 and 15 of the natural cements also lose 
 a strength once attained. Although the average strength of 
 the Portland cement increases with age, Portland cements Nos. 
 4, 5, and 13 lose strength with age. The following table 
 corresponds to Table IX. for the water hardened specimens, 
 and shows the increase of strength of specimens hardened in 
 air from — 
 
 TABLE Xin. 
 
 Time. 
 
 0 
 IS 
 
 0 
 % 
 
 0 
 
 No. 6. 
 
 No. 7. 
 
 No. 8. 
 
 No. 13. 
 
 No. 14. 
 
 0 
 
 2; 
 
 c 
 2 
 
 0 
 2; 
 
 Average Port. ' 
 
 Average Nat. j 
 
 Per Cent, of 
 increase Port. 
 
 Per Cent, of 
 increase 
 Nat. 
 
 I week. 
 
 120 
 
 209 
 
 265 
 
 120 
 
 204 
 
 384 
 
 280 
 
 164 
 
 223 
 
 381 
 
 358 
 
 313 
 
 166 
 
 
 
 I week to i month. 
 
 123 
 
 177 
 
 269 
 
 90 
 
 116 
 
 24 
 
 221 
 
 24 
 
 200 
 
 II 
 
 93 
 
 135 
 
 109 
 
 43-1 
 
 65-7 
 
 I week to 3 months. 
 
 353 
 
 81 
 
 171 
 
 135 
 
 126 
 
 182 
 
 140 
 
 194 
 
 43 
 
 221 
 
 
 150 
 
 170 
 
 47-9 
 
 102.4 
 
 I week to 6 months. 
 
 97 
 
 340 
 
 243 
 
 241 
 
 161 
 
 208 
 
 216 
 
 151 
 
 
 
 
 223 
 
 148 
 
 71.2 
 
 89.1 
 
 I month to 3 months. 
 
 230 
 
 96 
 
 98 
 
 45 
 
 ID 
 
 158 
 
 81 
 
 170 
 
 157 
 
 210 
 
 
 15 
 
 61 
 
 
 22.2 
 
 I month to 6 months. 
 
 26 
 
 163 
 
 26 
 
 151 
 
 45 
 
 184 
 
 5 
 
 127 
 
 
 
 
 88 
 
 39 
 
 19.6 
 
 14.2 
 
 3 months to 6 months. 
 
 256 
 
 259 
 
 72 
 
 106 
 
 35 
 
 26 
 
 76 
 
 43 
 
 
 
 
 73 
 
 22 
 
 15-8 
 
 6-5 
 
 The percentage of increase for both natural and Portland 
 is greater for the period, one week to one month, in the air 
 
CEMENT TESTS. 
 
 hardened specimens than in the water hardened, the percent- 
 age of increase of the water hardened being considerably 
 more after one month. The greater part of the ultimate 
 strength of the air hardened cement is gained during the first 
 month, although the Portland cement does gain about 20 per 
 cent, more strength during the following five months. The 
 actual difference between the strength of natural and Portland 
 cements hardened in air is at — 
 
 One week, 147 
 
 One month, 173 
 
 Three months, 127 
 
 Six months, 222 
 
 Average, 167 
 
 Since the average difference between the strength of natural 
 and Portland cements hardened in air is only 167 pounds w^iile 
 that between those hardened in water is 337 pounds, it is seen 
 that Portland cement has not the qualities for hardening in air 
 that makes it so valuable as a water hardening cement, the 
 strength of the natural cement hardened in air and water 
 being very nearly the same. The percentage of strength of 
 natural to Portland is at — 
 
 TABLE XIV. 
 
 One week, 53- % 
 
 One month 61.4 % 
 
 Three months, . . . . 72-6 % 
 
 Six months, 58.6 % 
 
 These percentages compared with those in Table X. also 
 show that the difference of strength between the natural and 
 Portland cements is less in the air hardened for the first three 
 months, being nearly the same at six months. 
 
 As in the specimens hardened in water, the month's test 
 may be considered as given a fairly good comparative test, 
 but since the breaking strains of the individual specimens as 
 well as those of the average are so irregular, tests of air 
 hardened cements are not of much practical value. The 
 
156 
 
 PORTLAND CEMENT. 
 
 range of the breaking strains of specimens of the same cement 
 broken on the same day is very often over 300 pounds. The 
 following table gives the percentage of water, temperature of 
 the mixing room, and the remarks on the setting as taken 
 from the headings in the record book. 
 
 TABLE XV. 
 
 No. 
 
 No. 
 No. 
 No. 
 No. 
 No. 
 No. 
 No. 
 No. 10 
 No. II 
 No. 13 
 No. 14 
 No. 15 
 No. 16 
 
 No. 17 
 
 Braxd. 
 
 Milwaukee 
 
 Louisville 
 
 Empire (London). . . . 
 
 Gibb's Portland 
 
 Rosendale 
 
 F. O. Norton's 
 
 Burham's Portland . . 
 
 Buckeye 
 
 Josson & Co.'s Port. . 
 
 Millen's Patent 
 
 Utica Black Ball 
 
 Hoffman's Rosendale. 
 South Bend Cem. Co. 
 
 Empire 30 
 
 Pr. ct. of 
 Ware 
 
 31.8 
 26.8 
 
 21.4 
 
 25 
 
 30 
 
 22.5 
 
 26.25 
 
 23-125 
 
 21-5 
 
 33-75 
 
 32-5 
 
 27-4S 
 
 Temp. 
 
 70° 
 62° 
 63° 
 65° 
 69° 
 67° 
 70° 
 
 68° 
 73' 
 65° 
 69° 
 
 67° 
 
 63° 
 
 79° 
 
 Remarks. 
 
 Quick setting for Portland- Cement. 
 
 Tern, rises slightly on addition of 
 water. Quick setting for Portland. 
 
 Tern, rises slightly on addition of 
 water. 
 
 The time required for setting varies with the cement, the 
 Portland as a rule, requiring more time. A good cement 
 should not heat on the addition of water nor should it swell 
 in the least. 
 
 The amount of water used was such that when the pressure 
 was applied, no water was forced out of the machine, but still 
 large enough to moisten perfectly every particle of cement. 
 The advantage of recording the percentage of water is that 
 when mixing another batch, it is only neccessary to observe 
 what percentage was used before to get the proper amount. 
 
 The temperature of the tank and mixing room was very 
 nearly constant, ranging only from 62° in mid winter to 79° 
 in the hottest part of July. The temperature of the water 
 was about eight degrees below that of the rooms. 
 
 The natural cements may have almost any color from the 
 very light straw colored Utica through the brown Louisville, 
 to chocolate Rosendale. The Portland cements are usually a 
 grayish blue or green, but never chocolate colored. 
 
CEMENT TESTS. 
 
 The following gives the strength of both hand and machine 
 made briquettes of neat unsifted cement each broken at the 
 age of six months : 
 
 Ce.ment. 
 
 H.^ND Made. 
 
 Machine. 
 
 No. 
 
 
 
 333 
 325 
 .S18 
 609 
 362 
 3.=io 
 
 346 
 364 
 613 
 
 703 
 439 
 441 
 
 No. 
 
 
 
 No. 
 
 
 
 No. 
 
 
 
 No. 
 
 6. 
 
 
 No. 
 
 
 
 
 
 
 The above are only numbers that were tested in both cases 
 at this age. 
 
 It will be noticed that the machine made briquettes stand a 
 greater strain than the hand made. This is due, among 
 other things, to the increase of pressure in forcing the cement 
 in the molds, and to the small amount of water used in the 
 construction of machine made briquettes. It might be reasoned 
 that because of the greater strength of machine made bri- 
 quettes that the comparative test is not a fair one because the 
 strength of the Portland cement is increased more than that of 
 the natural. This, however, is a false deduction. The per- 
 centage of increase is nearly the same on an average. One 
 reason that the difference in strength between machine and 
 hand made briquettes of Portland cement is greater than that 
 of the natural cements may be due to the fact that the natural 
 cement, setting more rapidly, did not receive all the patting 
 that the Portland cement did — a temptation that the students 
 found it impossible to resist. The Portland cement was 
 removed from the mold before the initial set was as perfect as 
 was the case with the natural cements, and experiments car- 
 ried on at the Lake Tunnel Office in Chicago, show that the 
 length of time the briquette is allowed to stand in the mold 
 affects the breaking strain, increasing with the degree of initial 
 set before removing. Machine made briquettes being forced 
 into the mold under considerable pressure are removed 
 immediately, which it is impossible to do with hand made, 
 and none of the setting takes place in the mold, and all are 
 treated aHke. Hubert Remley, '90. 
 
CEMENT TESTS. 
 
i6o 
 
 PORTLAND CEMENT. 
 
CEMENT 
 
 TESTS. 
 
i66 
 
 PORTLAND CEMENT. 
 
CEMENT TESTS. 
 
 167 
 
SALT IN CEMENT MORTAR. 
 
 169 
 
 No. 2. 
 
 Abstract from the Graduating Thesis of M. I. 
 Powers, '91, Engineering Department, State 
 University of Iowa. 
 
 FRESH AND Sx\LT WATER IN CEMENT MORTAR. 
 
 The variance in the results of recent experiments and the 
 conclusions and results announced in text-books and earher 
 engineering papers as regards the relative effect of fresh and 
 salt water upon the strength of cement mortar led the writer 
 to make tests upon the subject the subject of a thesis. The 
 tests were begun in February, and were Hmited to thirteen 
 weeks. Since acquiring the results and making the diagrams 
 it would be a source of much satisfaction to be able to 
 continue the tests longer, which is impossible, however, all of 
 the allotted time having been fully taken up. 
 
 In the following tests, seven brands of cement were used 
 (see following tables), three of which were Portland and four 
 natural cements. Five hundred briquettes were made of each 
 different brand of cement as follows: two hundred with fresh 
 water, two hundred with sah water (3 per cent, salt), and one 
 hundred with 10 per cent, solution of salt water. Fifteen 
 briquettes of each kind were broken at the interval of every 
 week from the time they were made for a period of thirteen 
 weeks, excepting those made with a 10 per cent, solution, 
 which lasted but seven weeks. In making the briquettes, the 
 cement was first carefully weighed in small batches, from 
 fifteen to twenty pounds, depending upon the rapidity with 
 which the cement set, and placed in a large galvanized iron 
 mixing basin. The water was also carefully weighed and 
 added, the minimum amount being used that would thor- 
 oughly mix the cement. The exact percentage of water used 
 may be seen in the following tables. The cement was thor- 
 oughly mixed by hand. Experiments were made looking 
 toward a machine mixer, but none were found to give as good 
 satisfaction as hand mixing. Experiments with a pug machine, 
 on the sam- principle as that described by Mr. A. Bent 
 Russell, C. E., in the Engineering News of January 3, 1891, 
 
170 
 
 PORTLAND CEMENT. 
 
 and used by the St. Louis waterworks extension with alleged 
 good results, showed the cement to be rolled up in small 
 balls or lumps after shaking. 
 
 Satisfactory results may be obtained with a small quantity 
 of cement, sufficient for one briquette, but where the briquettes 
 are machine made, and a considerable quantit}- of cement 
 required at each mixing, the results are not as good as could 
 be desired. After mixing, the cement was molded into 
 briquettes by the machine designed last year by Professor 
 Jameson for that purpose, and described in The Transit, 
 Number 2, and the Engineering News of February 7, 1891. 
 This machine gives a pressure of 150 to 175 pounds per 
 square inch, which aids in reducing the amount of water to a 
 minimum. When made, the briquettes were placed in a 
 galvanized iron pan, and covered with a damp cloth for 
 twenty-four hours, and at the end of this time immersed in 
 water in similar pans; those made with fresh water immersed 
 in fresh water, and those made with 3 and 10 per cent, 
 solutions of salt water immersed in 3 and 10 per cent, solu- 
 tions respectively. The temperature of the laboratory and 
 tank rooms was kept at a constant temperature from 60° 
 to 65". 
 
 The average breaking strain of the fifteen briquettes broken 
 each week was taken as the breaking strain of that cement, 
 for that week, and the tables and diagrams made from these 
 strains. They were tested with a Rieh'e Bros, standard 
 cement testing machine, with the addition of rubber buffer 
 clips, which gave the breakage of fully 90 per cent, of the 
 briquettes at the minimum section. 
 
 TABLE NO. I. 
 
 No. 
 
 Brand. 
 
 Kind. 
 
 Per. Ct. 
 Water. 
 
 No. Biiq. 
 per lb. 
 
 Address. 
 
 I. 
 
 Utica Black Ball. 
 
 Natural. 
 
 22.3 
 
 3-4 
 
 La Salle, 111. 
 
 2. 
 
 Gibbs' Portland 
 
 Portland. 
 
 15.6 
 
 3-8 
 
 Grays Essex Eng. 
 
 3- 
 
 Mil vvaukee . . 
 
 Natural. 
 
 20. s 
 
 4-S 
 
 VI il vvaukee, W is 
 
 4- 
 
 Buckeye Portland. . 
 
 Portland. 
 
 16.6 
 
 3-6 
 
 Be lefontaine, O. 
 
 6. 
 
 Utica J. Clark 
 
 Natural. 
 
 24.1 
 
 49 
 
 Utica, N. Y. 
 
 Hoffman Rosendale. . . . 
 
 Natural. 
 
 20.0 
 
 4.1 
 
 Kingston, N. Y. 
 
 7- 
 
 South Bend Cement. . . . 
 
 Portland. 
 
 19.2 
 
 37 
 
 South Bend, Lid. 
 
SALT IN CEMENT MORTAR. 
 
 171 
 
 Table No. 1. shows the name and character of the cement, 
 the percentage of water used in mixing, the number of 
 briquettes made to each pound of cement, and the addresses 
 of the several cement firms. The diagrams show the average 
 breaking results for each week, for each kind of cement with 
 fresh and salt water solutions, and from these it will be seen 
 that salt water increases the strength of cements considerably 
 at first, but does not seem to continue the strength, at least 
 not with the same percentage of increase as at first, as the 
 cement mortar grows older. By looking at the diagrams, it 
 will be noticed that the line representing the averages of salt 
 water breakages, drops below the fresh water line in some of 
 the cements during the last two or three weeks. Here it 
 would be interesting to continue the tests further in order to 
 see if this drop is a permanent one, or whether it is due 
 simply to some irregularity in the cement, which latter is rather 
 improbable. One significant fact is obtained from the tables 
 and diagrams, namely, that the considerable percentage of 
 increase in strength shown at first by the use of salt water 
 does not continue the same with increasing age. 
 
 Briquettes mixed with a 10 per cent, solution of salt water 
 were stronger in every case at first with the natural cements 
 than those made with a 3 per cent, solution, while the Portland 
 cements, with one exception, were weaker. From this and 
 the diagrams showing the averages of Portland and natural 
 cements with salt and fresh water, it will be seen that the 
 effect of salt water upon Portland and natural cements is 
 noticeably different. 
 
 In the Engineering News, of December 20th, 1890, is a 
 statement of tests which were made by Mr. John Gartland, 
 at Governor's Island, New York Harbor. These tests were 
 made for the purpose of comparing the relative strengths of 
 cement mortar when mixed with fresh and with salt water. 
 From the results of his tests the conclusion is stated that the 
 permanent gain in strength in Rosendale cements due to the 
 use of salt water was about 20 per cent., and in the case of 
 Portland cement not less than 10 per cent, permanent gain; 
 
172 
 
 PORTLAND CEMENT 
 
 2,531 briquettes were broken and fifteen different brands of 
 cement were used. In the tests which form the subject of 
 this paper, seven different brands of cement were used and 
 3.500 briquettes broken, thus giving a much larger number of 
 tests for averages than were used in the New York Harbor 
 tests. In the former tests, the gain at first in both natural and 
 Pordand cements, taking the averages of the several brands 
 of both, was about 30 per cent., but in the third month in both 
 and in the second month with the natural cements, the gain 
 becomes a minus quantity and the cements become weaker 
 than those mixed with fresh water. Here too, the results, as 
 shown by table III., are sufficiently uniform to show that the 
 variation can scarcely be credited to irregularities in the 
 cements in testing. 
 
 There can be no doubt that cement mixed with sea water 
 gains considerably in strength during the first few weeks, but 
 that it does not hold out, is clearly proved by these results 
 from the Cement Testing Laboratories of the University. 
 TABLE NO. III. 
 
 Natu i\L Cements. 
 
 
 Average 
 
 Average 
 
 Per cent. Average 
 
 u 
 
 Fresh Water 
 
 Salt Water 
 
 Gain in Strength 
 
 >: 
 
 Tests. 
 
 3 Per Cent. 
 
 of Salt over Fresh. 
 
 I 
 
 86.7 
 
 103.9 
 
 20 
 
 2 
 
 104.0 
 
 135-4 
 
 3^ 
 
 3 
 
 1 18.5 
 
 152-7 
 
 28 
 
 4 
 
 143.0 
 
 133-I 
 
 7 
 
 .S 
 
 154-5 
 
 I 70. 1 
 
 10 
 
 6 
 
 1S0.2 
 
 195-3 
 
 8 
 
 7 
 
 199.7 
 
 189.2 
 
 5 
 
 8 
 
 200.9 
 
 207.3 
 
 3 
 
 9 
 
 221.8 
 
 2i3-4 
 
 3 
 
 10 
 
 2335 
 
 216.7 
 
 8 
 
 1 1 
 
 262.7 
 
 247.9 
 
 
 12 
 
 271.2 
 
 247.8 
 
 9 i 
 
 13 
 
 279-5 
 
 260.3 
 
 7 
 
 PORTLAND Cements. 
 
 Average | Average of i Per. cent. Average 
 Fresh Water: Salt WaterjGain in Strength 
 Tests. 3 Per Cent, of Salt over Fresh. 
 
 331-9 
 350.4 
 376.7 
 403.1 
 492.4 
 
 443-5 
 508.0 
 
 513-0 
 
 532-9 
 624.1 
 568.2 
 582.3 
 578-1 
 
 406.7 
 506.7 
 509.0 
 550-9 
 531-7 
 487.9 
 
 515-0 
 564.6 
 540.0 
 543-6 
 584-3 
 355 3 
 542-6 
 
 33 
 35 
 36 
 8 
 10 
 
 The gain also seems to be greater and more permanent 
 with the Portland than with the natural cements. 
 
 The effects of using a 10 per cent, seem not so good as with 
 a 3 per cent, solution. The tests were not carried out for the 
 
SALT IN CEMENT MORTAR. 
 
 full thirteen weeks, and cannot be directly compared. The 
 preceding tables give the average breaking strains of natural 
 and Portland cements, with fresh and salt water, and the per 
 cent, of average gain of salt over fresh water. 
 
 With those of fresh and salt water, the relative strength for 
 the seven weeks tested can be seen from the tables and 
 diagrams. 
 
 Valuable results might, no doubt, be obtained by a study of 
 the chemical analysis of the different brands of cement in 
 connection with the effect of salt water upon the same. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 70 O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 F.W'ITtS' 
 
 
 
 
 
 
 
 
 
 /. 
 
 
 
 
 
 
 
 boo 
 
 
 
 
 
 
 \ 
 
 
 // 
 /'/ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 N 
 
 
 
 \ 
 
 / 
 
 / 
 
 
 
 
 3%S«T 
 
 
 
 Soo 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *oo 
 
 il 
 
 il 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 0O 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 zoo . 
 
 i 
 
 -4— 
 
 
 
 Bl 
 
 JCKE 
 
 :ye 
 
 •POF 
 
 ;TL^ 
 
 ,ND 
 
 CEf 
 
 yiEN 
 
 r 
 
 
 
 
 J— 
 
 
 
 
 
 BEL 
 
 LEFC 
 
 )NTA 
 
 iNE, 
 
 OH 
 
 10. 
 
 
 
 
 lOO 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 2 
 
 3 
 
 
 
 6 
 
 7 
 
 3 
 
 9 
 
 10 
 
 II 
 
 t 
 
 /2 
 
 /3 
 
 ZOO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 UTI 
 
 CA 
 
 J. c 
 
 LARI 
 
 : C[ 
 
 :mei 
 
 
 
 
 
 
 
 2.0 0 
 
 
 
 
 
 UTK 
 
 :a, 1 
 
 LL. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 f. wnrrD 
 
 /oo 
 
 
 
 
 
 
 
 
 ..'■^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2. 
 
 
 •St 
 
 J- 
 
 
 7 
 
 8 
 
 9 
 
 
 II 
 
 
 /J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 HOf 
 
 
 \N 
 
 ROS 
 
 EN[ 
 
 )ALE 
 
 c 
 
 EME 
 
 NT 
 
 
 
 
 
 4-00 
 
 
 
 
 KIN 
 
 GST 
 
 3N, 
 
 N.Y. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3O0 
 
 
 
 
 
 
 
 
 
 
 / 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 lOo 
 
 
 
 
 
 
 
 
 
 7 
 
 
 
 
 
 
 
 '~f 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 lOO 
 
 .■// 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 4- 
 
 
 6 
 
 7 
 
 8 
 
 9 
 
 /o 
 
 II 
 
 IZ 
 
 /3 
 
176 
 
 PORTLAND CEMENT. 
 
 No. 3. 
 
 Hot Tests of Cement — Abstract from the Graduating 
 Thesis of Frank Woolston, '94, Engineering 
 Department, State University of Iowa. 
 
 The cements used in making these tests were purchased on 
 the Cliicago market by Musser & Co., of Iowa City. No 
 mention was made at the time of purchasing that they were to 
 be used for testing. 
 
 The briquettes were made in the Jameson Briquette Ma- 
 chine. The cement was used neat and iinsifted, in the condi- 
 tion tliat it was taken from the barrel. As small amount of 
 water as possible was used. The actual amount will be 
 found in Table 11. After being made, the briquettes were 
 placed in galvanized iron pans and covered with a very damp 
 cloth for tvvent3'-four hours. At the end of twent3'-four hours 
 one-half the briquettes were placed in the hot water bath and 
 the other half in a cold water bath. The water was changed 
 every twenty-four hours. The hot water bath was kept at a 
 constant temperature of 200'^ Fah. b}' means of a gas stove 
 and a water jacket. 
 
 The briquettes were broken every twenty-four hours for 
 one week, and then every week until a total of ninety days 
 had expired from the time of immersion. Ten briquettes from 
 the hot bath and ten from the cold bath were broken each 
 time, and the breaking strains shown in the diagram are the 
 average of these ten. 
 
 A chemical analysis of Dyckerhoff, Germania, South Bend, 
 Buckeye, and Louisville was made by Mr. Earl Durfee, a 
 student in advanced chemistry. This work was done under the 
 supervision of Dr. L. W. Andrews, Professor of Chemistry-.* 
 
 *The method of analysis is 
 The student is referred to the 
 
 given in full in the thesis, but omitted 
 article bj Dr. Andrews, page 91. 
 
 here. 
 
HOT TESTS OP CEMENT. 
 
 177 
 
 CHEMICAL ANALYSIS. 
 
 1. 
 
 Dyckeinott, 
 
 
 German Portland. 
 
 
 T T 
 
 11. 
 
 Germania, 
 
 
 German Portland. 
 
 
 TTT 
 111. 
 
 ooutli liencl, 
 
 
 American Portland. 
 
 
 IV. 
 
 
 
 American Portland. 
 
 
 V 
 
 
 
 American Natural. 
 
 
 
 I 
 
 II 
 
 Ill 
 
 IV 
 
 V 
 
 Sil., . 
 
 . . 20.25 
 
 22.36 
 
 19.26 
 
 20.80 
 
 18.92 
 
 Cal. ox., 
 
 . . 58.03 
 
 64.38 
 
 60.25 
 
 57-82 
 
 46.90 
 
 Fr. ox., 
 
 . . 4-03 
 
 4-15 
 
 3-39 
 
 4.64 
 
 1.91 
 
 Al. ox.. 
 
 . . 12.39 
 
 2.83 
 
 14-54 
 
 12.31 
 
 11.02 
 
 Mg. ox. 
 
 . . -74 
 
 1.87 
 
 Trace 
 
 4.84 
 
 -97 
 
 By an examination of the analyses, we find that III. has an 
 excess of free lime and ahiminum oxide. The free lime is 
 plainly visible in the shape of large white specks. This 
 cement cracked badly in setting when in the "hot test" room 
 of the laboratory, which had a very moist temperature of about 
 80° Fah. 
 
 By placing the briquettes of this cement in a cool, dry place 
 it set in twenty-four hours so that four briquettes attained the 
 following breaking strength in twenty-four hours without 
 signs of cracking: 147, 172, 98, and 115 pounds. 
 
 The briquettes immersed in the cold water bath showed 
 signs of cracking in twenty-four hours, and in three weeks' 
 time went to pieces, after having swehed to one and one-half 
 their original size. The briquettes placed in hot water went 
 to pieces entirely in one hour. 
 
 The barrel with what cement remained in it was then 
 moved into the '-hot test" room; temperature, 80", and very 
 moist. The cement soon swelled until the hoops were broken. 
 
 The first setting of this cement was so rapid that it was 
 with difficulty the briquettes were made. No. IV. seemed to 
 be the reverse of No. III. It was found to be almost impossible 
 to make briquettes with it on account of its almost total lack 
 of cementitious qualities. Some of it was mixed with water 
 
178 
 
 PORTLAND CEMENT. 
 
 and placed on a glass plate. It simply dried out without 
 showing any cementing qualilies. 
 
 Briquettes, after twenty-four hours under a damp cloth, 
 were placed in a dry, cool room; temperature, 65° Fah. 
 Could be easily crushed by the hand at the end of two weeks. 
 Some that was used neat, as plaster in the laboratory, had 
 about the consistency of sand at the end of three weeks. 
 
 This cement is shown to have had an excess of mao-nesium 
 oxide, and a slight excess of ferric ovide, with a specihc gravity 
 lower than the other Portland cements. Both companies were 
 notified of the results. 
 
 Diagrams. — In the diagrams, the dotted lines sh ow^ the 
 results of the ''cold icatcr'' tests, and the ///// line the results 
 of the ''hot water'' tests. 
 
 By reference to these diagrams, we find that with the best 
 Portland cements tested, such as I. and II., there is but little 
 difference between the "hot" and '-cold" tests, as the two 
 lines cross and recross. This would show that the effect of 
 the " hot" test was very slight upon either of these cements. 
 
 In the case of II. the earlier strength of the "hot" tests wvas 
 even less than 'in the cold " and at seventy-six days the 
 strength of the "hot" tests exceed that of the "cold" tests. 
 
 In I. the "hot" testis the weaker at the end of two weeks 
 but at sixty-two days nearly equals the maximum reached by 
 the "' cold" test. 
 
 The Mankato cement (natural) shows an even, uniform 
 increase. The "hot" rising somewhat faster at first but the 
 "cold" steadily increasing until at the end of ninety days the 
 two lines are but a few pounds apart. 
 
 At two weeks the Mankato briquettes showed sHght checks 
 but the strength was not effected and the checks did not 
 increase. 
 
 Louisville (V.) shows a much more marked difference, the 
 "hot" increasing rapidly in strength during the earlier stages. 
 The "cold" show a uniform increase and the two lines would 
 probably have crossed had the time been extended. At the 
 end of seven days the "Uiot" test showed the same strength as 
 the " cold " at the end of seven weeks. 
 
HOT TESTS OF CEMENa\ 1 79 
 
 The Milwaukee cement (natural) exhibits a marked differ- 
 ence of behavior in the two tests. In the "hot" test the 
 strength increases rapidly during the earlier stages and then 
 decreases until at the end of six weeks the briquettes are so 
 weak as to make it impossible to test them, although at the 
 end of ninety-seven days they are still intact. The increase 
 in size is very marked. The briquettes in cold water grad- 
 ually and continuously increase in strength. 
 
 An the mixing of the briquettes of the Milwaukee cement 
 an excess of water was used, and it is to be regretted that the, 
 cement was not analyzed in order to ascertain if any of its 
 constitutents could account for its faiku'e under the "hot" test 
 at such a late day. 
 
 Conclusions. — We find that the cements that show the best 
 results under the " cold" tests, give practically the same results 
 under the "hot" tests; there being no marked difference 
 between the two lines at any time. Numbers I. and II. show 
 very little differences and these differences constantly changing. 
 
 The strength of the natural cements is very much hastened 
 during the earlier stages by the "hot" test but there is a 
 gradual increase in the "cold" test that sooner or later makes 
 the lines cross. 
 
 Some' exceedingly faulty cements are exposed at once by 
 the " hot" test as some of those, that were tested, were reduced 
 to a lifeless powder after one hour in the hot water, and these 
 same cements stood the ordinary twenty-eight day test well. 
 
 From the results obtained we may infer that no thoroughly 
 good cement need fear the "hot" test. The strength of the 
 briquettes may not reach the maximum of those reached by 
 the "cold" tests but they wih approximate them. 
 
 Cements that contain an excess of free Hme will not stand 
 the "hot" tests, and when but a limited time is possible for 
 the testing the " hot" test has every advantage over the " cold." 
 In the case of the Louisville and the Milwaukee results were 
 reached in one week by the "hot" test that exceeded those 
 reached by the "cold" tests in over seven weeks. 
 
i8o 
 
 PORTLAND CEMENT. 
 TABLE IL 
 
 
 
 
 
 
 S > 
 
 iJ c 
 
 
 
 
 & S 
 
 
 Cement. 
 
 
 
 Reji.^rks. 
 
 
 V 8 
 
 Cu. ce 
 per lb. 
 
 
 Djckerhoff . 
 
 91.8 
 
 120 
 
 Residue dark, sharp, and heavy. 
 
 Germania . . 
 
 90.6 
 
 140 
 
 " light colored, soft and light. 
 
 Milwaukee . 
 
 79-7 
 
 162 
 
 " " " sharp, and coarse. 
 
 Louisville . . 
 
 72.8 
 
 190 
 
 " color as cement, sharp, coarse, heav\. 
 
 Mankato . . . 
 
 76.7 
 
 
 " medium weight with dark specks. 
 
 South Bend . 
 
 91.7 
 
 118 
 
 " dark, sharp, heavy, with white specks. 
 
 Buckeye . . . 
 
 84.6 
 
 
 " soft, light, and fine. 
 
l82 
 
 PORTLAND CEMENT. 
 
PORTLAND CEMENT. 
 
INDEX. 
 
 A. 
 
 Abrasion, 65 
 Adhesion, 62 
 
 Adulteration, detection of, 95 
 Aggregate, 116 
 
 — gravel, 116 
 
 — machine broken, 117 
 
 — requisites for good, 114. 
 
 — reduction, 115. 
 
 — screening, 115. 
 
 — size of, 115. 
 Artificial Stone, 132. 
 
 — — Frear, 136. 
 
 — McMurtie, 136. 
 
 — — Ransom, 136. 
 
 — — Sorel, 137. 
 
 — — Victoria, 135. 
 Allentown, 14. 
 Aluminate, 97. 
 American Natural, 72. 
 
 — Portland, 73. 
 Aspdin, John, 13. 
 
 B. 
 
 Backs, 23. 
 
 Bauschinger's Calipers, 63. 
 Bellefontaine, 14. 
 Beton Coignet, 132. 
 Blowing, testing for, 63. 
 Bohme Hammer, 47. 
 Blowey, 63. 
 
 Breaking strength, average, 72. 
 
 — — variation, 54. 
 Brick machine, 25. 
 
 Brick, pressed, 34. 
 
 Briquettes, amount of material, 80. 
 
 — — water, 55. 
 
 — Bohme's Hammer, 85. 
 
 — Machines, 46, 142. 
 
 — — Bohme's Hammer, 47. 
 
 — — Jameson, 50. 
 
 — — — capacity, 52. 
 
 — making, 45. 
 
 — machine-made, advantages, 
 
 54- 
 
 — hand-made, 84. 
 
 — English method, 54. 
 
 — molds, 46. 
 
 — - mixing mortar, 55. 
 
 — placing, 53. 
 
 — • porcelain sink, 56. 
 
 — sand, 56. 
 
 — temperature, 55. 
 Broken stone, by machine, 117. 
 
 — — increase in bulk, 117. 
 
 — — by hand, 116. 
 Burning, 27. 
 
 — correctly, 30. 
 
 — over, 27-30. 
 
 — reburning, 31. 
 
 — under, 30. 
 
 C. 
 
 Carbonate of lime, 18,97. 
 Carbonic acid, 27, 30, 97. 
 Calcium aluminate, 97. 
 
 — ferraluminate, 97. 
 
 — metasilicate, 97. 
 
 — orthosilicate, 97. 
 Cavities, 23. 
 
INDEX. 
 
 Cement, i. 
 
 — constituents, 4. 
 
 — — relative financial 
 importance, iS. 
 
 — choice, 10. 
 
 — curbing, 130. 
 
 — and limes, difference, 3. 
 — • engineering definition, 7. 
 
 — knowledge of, 13. 
 
 — mortar, strength of, 3. 
 
 — natural, 7, 8. 
 
 — plaster composition, 113. 
 
 — as plaster, 113. 
 
 — tests, Remlej', 141. 
 
 — setting, 3. 
 
 — tester, £42. 
 
 — tests, hot 176. 
 
 — tests. State University of 
 
 Iowa, 141. 
 
 — tests. State Universitj^ of 
 
 Iowa, hot, 176. 
 
 — tests, State University of 
 
 Iowa, fresh and salt 
 water, 169. 
 
 — tests. State University of 
 
 Iowa, salt and fresh 
 water, 169. 
 
 — use of, 2. 
 
 — works established, 13. 
 
 — work wet, 3. 
 Cementitious, 28. 
 Characteristics, 94. 
 Chalk, 18. 
 
 — weight, 18. 
 Checking, 71. 
 Chemical analysis, 86. 
 
 — reactions, 3, 98. 
 Chutes, 23. 
 Classification, 91. 
 Clay, 17, 18. 
 
 Clinker, 19, 31. 
 Clips, 52, 79. 
 
 — blotting paper, 53. 
 
 — rubber, 53. 
 
 — — buffer, 53. 
 Color, 28. 
 Composition, 94. 
 
 Concrete, 113. 
 
 — • depositing, 123. 
 
 — — under water, 124. 
 
 — machine mixing, 119. 
 
 — materials, 1 13. 
 
 — mixing, 118. 
 
 — — platforms, 118. 
 
 — pavement foundations, 126. 
 
 — proportions, 126. 
 
 — sidewalks, 127. 
 
 — quantities, 139. 
 Constituents, relative financial im- 
 portance, 18. 
 
 Cracking, 63. 
 
 Cross strain, 62. 
 
 Crushing tests, objections, 43. 
 
 Crystallizing, 97. 
 
 Curbing, 130. 
 
 1). 
 
 Disintegrating, 63. 
 Dobbs, Edgar, 13. 
 Double calcination, 19. 
 Double working, 22. 
 Down draft, 33. 
 Drying floors, 19, 31. 
 Dry process, 19, 24. 
 Drying rooms, 31. 
 Dusting, 34. 
 
 E, 
 
 Eddystone Lighthouse, 13. 
 Egypt, 14. 
 Egyptian, 12. 
 Emery, 66. 
 
 Estimating quantities, 137. 
 
 — — amount mortar re- 
 quired per cu. yd. of 
 masonry, 138. 
 
 — — amount material 
 per cu. yd. of mortar, 138 
 
 — — cement, 137. 
 
 — — concrete, 139. 
 
 — — lime, 137. 
 Expert, 27. 
 
INDEX. 
 
 189 
 
 F. 
 
 Faija, Henry, warm test, 64. 
 Fineness, 44, 71, 81. 
 
 — French specifications, 89. 
 
 — percentage, 44. 
 Foreign Portland, 73. 
 Floury, 40. 
 
 Flux, 35. 
 France, 14. 
 Frear stone, 136. 
 Free lime, 58. 
 Fresh cement, 64. 
 Fuel, 17. 
 
 — amount, 30. 
 
 — saving, 34. 
 
 — under control, 34. 
 
 — -waste, 30. 
 Fusing, 27. 
 
 G. 
 
 German Association of Cement 
 
 Makers, 67. 
 German Portland, 68. 
 
 — — established, 14. 
 
 — — production, 14. 
 Gilmore, 71. 
 
 Glass, liquid, 134. 
 
 — powdered, 28. 
 Gravel, 116. 
 Gravity, 19 
 Grinding, 39. 
 
 — ball pulverizers, 25. 
 
 — millstones, 25. 
 
 — machine for abrasion, 66. 
 
 — power required, 40. 
 
 — vertical runner, 25. 
 Gypsum, 67. 
 
 H. 
 
 Hardening, chemical processes, 91, 97 
 
 — dependent on, 94. 
 Harrows, 22. 
 
 Heat test, 83. 
 
 — • utilized, 30. 
 Hennepin Canal, 133. 
 Historical data, 11. 
 Hot tests, 176. 
 
 Humidity, 30. 
 Hydration, 97. 
 
 I. 
 
 Indiana, 16. 
 Iowa, 16. 
 
 J- 
 
 Jameson, Briquette Macliine, 5. 
 K. 
 
 Kiln, 19. 
 
 — Bock tunnel, 36. 
 
 — continuous, 30, 35. 
 
 — dome, 28, 29. 
 
 — — disadvantages, 31. 
 
 — Hoffman, 31. 
 
 — — advantages, 34. 
 
 — — dampers, 33. 
 
 — — flues, 33. 
 
 — — method of operating, 33 
 
 — iron tube, 39. 
 
 — lime, 5. 
 
 — lining, 31. 
 
 — wrought iron, 38. 
 Knives, 22. 
 
 L. 
 
 Leland Stanford, Jr., University, 134 
 Leon, Mexico, 17. 
 Limes, i. 
 
 — and cements, power of absorp- 
 
 tion, difference, 3. 
 
 — constituents, 4. 
 
 — free, 58. 
 
 — hydraulic, 7. 
 
 — — constituents of, 4. 
 
 — kiln, 5. 
 
 — manufacture of, 4, 5. 
 
 — mortar, 2, 107. 
 
 — — setting, 2. 
 
 — — strength, 3. 
 
 — slaked, 6. 
 
 — use of, 2. 
 
 Locks, Hennepin Canal, 133. 
 Louisville, 15, 17. 
 
190 
 
 INDEX. 
 
 M. 
 
 Machines, spring balance, 80. 
 Magnesia, 96. 
 Manufacture of cement, i. 
 Market, 18. 
 
 Masonry repairing, 131. 
 
 Maximum and minimum tensile 
 
 strength, 72. 
 McMurtie stone, 136. 
 Mexico, 17. 
 
 Microscope, polarizing, 97. 
 Milw aukee, 15. 
 Missouri, 16. 
 Mixing machine, 26. 
 
 — for tests, 75. 
 
 — mortar, 86. 
 Molds, 79, 142. 
 Monolithic structures, 134. 
 Mortar, 107. 
 
 — amount of water, 112. 
 
 — brash, i 10. 
 
 — harsh, 110. 
 
 — lime, 107. 
 
 — mixing, 112. 
 
 — smooth, no. 
 
 N. 
 
 Natural cement, 108. 
 
 — — cost, 9. 
 
 — and Portland, 146. 
 
 — — difference, 8. 
 
 — cement, production in U.S., 15 
 Napus quartz, 66. 
 
 Needles, felted, 58. 
 Needle test, England, 59. 
 
 — — German, 60. 
 
 — — Gilmore, 58. 
 
 — — unsatisfactory, 60. 
 
 — — S. U. I., 61. 
 
 — — Vicat, 59. 
 New York, 16. 
 
 P. 
 
 Pacific, 17. 
 Packing, 19, 41, 80. 
 Panama, 17. 
 Pans, 56. 
 
 Patents, 13. 
 Pennsylvania, 16. 
 Plates, 19. 
 
 Plastering with cement, 113. 
 Plaster of Paris, 93. 
 
 — — amount, 109. 
 
 — patent cement, 113. 
 Portland, constituents, 16. 
 
 — definition, 7. 
 
 — cost, 9.- 
 
 — German, 14. 
 
 — manufacture, 16. 
 
 — and natural, difference, 8. 
 
 — location raw materials, 16. 
 
 — patentees, 13. 
 
 — production in U. S., 15. 
 
 — raw materials, 9. 
 
 — vmiformity, 9. 
 
 — value, 18. 
 Pozzolana, 12. 
 
 Proportions natural cement and 
 
 lime, 109. 
 Pug mill, 26. 
 
 Q. 
 
 Quality, effect, i. 
 
 — cement, 2. 
 Quarries, 18, 19. 
 Quartz, crushed, 56. 
 
 — NajDus, 66. 
 Quicklime, 6. 
 
 R. 
 
 Ransom, Smith 61 Co., 134. 
 Ransom stone, 136 
 Raw materials, 18. 
 
 — — crystalline limestone, iS 
 
 — — hard working, 18. 
 
 — — reduction, 19. 
 Regauging, no. 
 Reheating, 31. 
 
 Reid, Henry, 135. 
 Remley cement tests, 141. 
 Repairing masonry, 131. 
 Rosendale, 14. 
 Reservoirs, 23. 
 Resistance, 65. 
 
Results, maximum, 2. 
 Roman cement, 14. 
 
 S. 
 
 Sampling, 76. 
 Sand, 56. 
 
 — bank, no. 
 
 — excessive amount, 108. 
 
 — German standard, 83. 
 
 — river, no. 
 
 — standard, 56, 78. 
 
 — — price, 78. 
 
 — Vitruvius, in. 
 Sajlor's cement, 14. 
 Salt water, effect of, 169, 
 Screening, 115. 
 Screen for broken stone, 104. 
 Semi-Avet, 26. 
 Setting, 7, 58, 76, 81. 
 
 — causes of, 91. 
 
 — chemical processes, 91. 
 
 — conditions affecting, 91. 
 
 — French specifications, 89. 
 
 — German tests, 82. 
 
 — incipient, 1 10. 
 
 — quick, 76, 109. 
 
 — slow, 65, 76, 109. 
 Shipping, 19. 
 Sidewalks, 127. 
 Sieves, 77. 
 
 — cost, 78. 
 Sifting, 40. 
 
 Silicating solution, 135. 
 Slabs, 56. 
 Slaking, 6. 
 
 — air, 6. 
 
 — cause of, 92. 
 
 — drowning, 6. 
 
 — sprinkling, 6. 
 Slate, 18. 
 Slurry, 24. 
 Smeoton, John, 13. 
 Sorel stone, 137. 
 South Bend, Indiana, 14. 
 Specifications, 69. 
 
 — American Society of Civil 
 Engineers, 69. 
 
 INDEX. 
 
 Specifications, Austrian, 74. 
 
 — — for Portland, 89. 
 
 — — ■ — Roman, 89 
 
 — English, 89. 
 
 — — ■ for expansion, 90. 
 
 — — — tensile strength, 90 
 
 — French, 86. 
 
 — German, 74. 
 
 — — standard, 80. 
 Stability, 63. 
 St. Leger, Maurice, 13. 
 Stetten, 14. 
 Stone, artificial, 132. 
 
 — crushers, 39, 99. 
 
 — — description, 104. 
 
 — — Farrel Marsden, 103. 
 
 cost and capacity, 105. 
 
 — — Gates, 105. 
 
 — — — cost and capacity, 
 106. 
 
 — — location of, 117. 
 Storing, 41. 
 Storage, 19. 
 
 Strength, compressive, 81. 
 
 — French specifications, 87. 
 
 — test for, 81. 
 
 — tensile, 81. 
 Sylvester's process, 136. 
 
 T. 
 
 Tanks, 57, 142. 
 Temperature, 30, 75, 142. 
 
 — increase, 25, 57. 
 Tensile strength, 42. 
 Testing, 42. 
 
 — applying strain, 53. 
 
 — compression, 85. 
 
 — heat, 64, 83. 
 
 — machines, 80, 99. 
 
 — — Fairbanks, loi. 
 . — • — Tini'is Olson, 100. 
 
 — — Riehle Brothers, 99. 
 
 — method, 2. 
 
 — German number of 
 briquettes, 84. 
 
 Test pieces, neat, 86. 
 
192 
 
 INDEX 
 
 Test peices, treatment, 86. 
 Transportation, water, 18. 
 Texas, 17. 
 
 U. 
 
 Ulster Co., N. Y., 14. 
 
 United States, 14. 
 
 Use, economical, 108. 
 
 Use of Portland cement, 107. 
 
 Utica, 15. 
 
 V. 
 
 Virginia, 16. 
 Vanne aqueduct, 133. 
 Vicat needle test, 59. 
 Victoria stone, 135. 
 Vitrifaction, 13, 28. 
 Vitruvius on lime, 11. 
 
 — — sand, 1] I. 
 Volume, constancy, 81. 
 
 W. 
 
 Wash mill, 20. 
 
 — speed, 22. 
 Walks, 127. 
 
 Walks, foundation, 12S.' 
 
 — frames, 128. 
 
 ■ — manipulation, 128. 
 
 — materials for, 127. 
 
 — mixing concrete, 128. 
 
 — outfit, 130. 
 
 — quantities material required, 
 
 131- 
 
 Warm test, Faija, 64. 
 Warner's New York, 14. 
 Water, amount, 22. 
 
 — — in mortar, 112. 
 
 — excessive use of, 112. 
 
 — on stone or brick, 112. 
 Weight, 28, 76, 80. 
 
 — loss allowed, 81. 
 Wet process, 19, 20. 
 
 — — time, 24. 
 Wetting stone or brick, 112. 
 Wire test, 71 
 
 Work, arrangement, 19, 
 
 Y. 
 
 Yankton, S. D., 14.