REESE LIBRARY OF THE UNIVERSITY OF CALIFORNIA. Deceived J^MEN T. PAGE Relation of Foundation to Bridge Design. Roman and Other Ancient Founda- tions. Bridge at Shuster, Persia. Roman Arch at Trezzo. Four Ancient Methods for Foundations. Method of Open Caissons. Method with Piles and Concrete Capping. Method of Encaissement. Method of Coffer-dams. Caesar's Bridge over the Rhine. Pneumatic Caissons and Coffer-dams appli- cable to Different Cases. Origin of Coffer-dams and Primitive Types. The Hutcheson Bridge at Glasgow. Robert Stevenson's Specifications for Coffer- dams on Hutcheson Bridge. Old Directions for Triple-puddle Coffer-dam in Forty Feet ( !) of Tide-water. W. Tierney Clark's Account of the Great Coffer- dams for the Buda-Pesth Suspension Bridge. Character of Puddle used. Class of Work to which Coffer-dams should be applied. Value of Actual Examples.. i ARTICLE II. CONSTRUCTION AND PRACTICE. CAY/. 1 COFFER-DAMS. Definition of Coffer-dam. Simple Clay Bank. Drag Scraper for removing Soft Bottom. Excavating Spoon. Larger Dredges mentioned. Crib and Embank- ment used on Chanoine Dams on Great Kanawah River. Improvised Na- smyth Sheet-pile Hammer. Failure on Ohio River because of Porous Bottom. Crib Coffer-dam with Puddle Chamber, C., B. & Q. R. R. Cribs without Puddle Chambers, Can. Pac. Ry. Cribs of Old Plank, Santa Fe Ry. Crib for Arkansas River, St. L. and S. F. Ry. Sheet Piles used on Santa Fe. Sheet Piles used on Union Pacific Ry. Coffer-dam on Grillage, Union Pacific Ry. Circular Coffer-dam of Staves at Fort Madison, la. Circular Coffer-dam pailure at Walnut St., Phila. Probable Cause of Failure. Form of Construc- tion to adopt. Use of Puddle. Cutwaters. True Economy of Construction. . 13 ARTICLE III. CONSTRUCTION AND PRACTICE. CRIBS AND CANT AS. Stopping Leaks. Canvas Bulkhead at Keokuk, Iowa. Canvas Funnel for Springs. Anchoring Cribs and Crib Coffer-dam at St. Louis. Timber Casings cov- vii Viil TABLE OF CONTENTS. PAGE ered with Canvas, Melbourne. Strength of Water-soaked Timber. Polygonal Crib for Harlem Ship Canal Pivot Pier. Polygonal Crib for Arthur Kill Bridge. Octagonal Crib, Coteau Bridge 28 ARTICLE IV. PILE DRIl'ING AND SHEET PILES. Historical Forms of Pile Drivers. Simple Sheet-pile Driver. Large Pile-driving Derricks. Machinery for Pile Driving. Cost of Outfits. Nasmyth Hammers of Various Types. Loads on Guide and Foundation Piles. Pulling Piles and sawing off under Water. Forms of Sheet Piles. Wakefield Sheet Piling. Shoes for Sheet Piling. 40 ARTICLE V. CONSTRUCTION WITH SHEET PILES, Water and Puddle Pressure. Calculation of Sheet Piling. Size of Wales and Struts. Width of Puddle Chambers. Guide Piles and Guides. Ann Arbor Sheet-pile and Puddle Coffer-dam, M. C. Ry. Failure with Sheet Piles at Arthur Kill Bridge. Successful Method adopted. Sewer Coffer-dam for Boston Sewerage System. Wakefield Sheet Piling. Harper's Ferry Coffer- dam. Momence, 111., Coffer-dam, C. & E. I. Ry. Sheet Piling for Charles- town Bridge Piers. Polygonal Sheet-pile Reservoir Coffer-dam at Fort Monroe, Va 54 ARTICLE VI. CONSTRUCTION WITH SHEET PILES. Combinations of Various Forms of Sheet Piles. Sheet-pile and Puddle Coffer- dam, Walnut Street Bridge, Chattanooga. Framing of Cumberland, Md., Coffer-dam. Sandy Lake Coffer-dam and Pile-driving Plant. Driving Sheet Piles with Water Jet. Use of Sheet Piling on Foundations of Main Street Bridge, Little Rock. Concrete Piers at Little Rock. Removal of Old Pier at Stettin, Germany. Removal and Repair of Pier in Coosa River, Alabama. Floating Coffer-dam for P. & R. R. R. Bridge over the Schuylkill. Use of Six- inch Sheet Piles at St. Helier, Jersey. Stock Rammer to stop Leaks. Single- pile Coffer-dams, Putney Bridge. Twelve-inch Sheet Piling, Victoria Docks. Tongue and Groove Sheet Piling, Topeka, Kansas. Use of Dredging Pump at Topeka - 66 ARTICLE VII. METAL CONSTRUCTION. Thin Steel Shells. Hawkesbury Oblong Metal Piers. Vertical and Inclined Cut- ting Edges. Water-tight Construction. Pivot Pier of Clustered Cylinders. TABLE OF CONTENTS. ix PAGE Double-cylinder Pier. Russian Ornamental Cylinder Piers. Lighthouse Cylinders. Calculation of Thin Metal Cylinders. Forth Bridge Metal Coffer- dams. Forth Bridge Circular Granite Piers. Combined Metal Coffer-dam and Pier Base. Metal Sheet Piles 80 ARTICLE VIII. PUMPING AND DREDGING. Amount of Pumping indicates Success. Bascule for Pumping. Chapelet for Pumping. Bucket Wheel used at Neuilly. BQX Lift Pump. Metal Lift Pump. Diaphragm Pump. Steam Siphons. Van Duzen Jet. Lansdell Siphon. Pulsometer Steam Pump. Maslin Automatic Vacuum Pump. Comparative Efficiency of Centrifugal and Reciprocating Pumps. Tests of Centrifugal Pumps. Direct-Connected Engine and Centrifugal Pumps. Use of Electric Power. Suction-pipe Details. Type and Capacity of Pump. Methods of Priming. Double-suction Pumps. Dredging Pumps. Clam shell and Grapple Dredges. Sand Diggers and. Elevator Dredges. Dipper Dredges. Cost of Dredging 9: ARTICLE IX. THE FOUND A TION. Character of Foundation. ^-Kind of Bottom. Soft Bottom. Pile Foundation. Soft Material overlying Hard Bottom. Clean Smooth Rock. Sloping Rock. Rough Rock. Concrete Levelling Course. Concreting under Water. Mono- lithic Concrete Piers. Concrete Piers at Red River. Monolithic Concrete on Illinois and Mississippi Canal. Requirements for Good Concrete. Compo- sition of Concrete. Contractor's Plant. Cableways 106 ARTICLE X. LOCATION AND DESIGN OF PIERS. Location at Fixed Site. Location at New Site. Government Requirements. Examination of Site. Test-boring Apparatus. Mississippi River Commission Boring Device. Economical Length of Spans. Ottewell's Formula for Eco- nomic Span. Morison's Design for Piers. Omaha Union Pacific Piers. Rus- sian Piers. Obstruction caused by Piers. Cresy's Experiments on the Obstruction caused by Piers. Correlation of Theoretical Form and Archi- tectural Design 120 TABLE OF COFFER-DAMS.- , & X River and Location. C-en,. *' Character of Bottom. i 5 2 6 3 8 4 9 5 U 6 15 7 17 S 18 9 18 10 18 II 20 12 2O I 3 20 14 20 15 20 I ( ) 2O 17 20 IS 2 4 I 9 3 20 72 21 33 22 36 23 38 24 39 25 59 2t> 60 27 60 28 62 29 62 30 63 3i ( >3 32 64 33 66 34 67 35 67 36 70 37 72 3S ; 74 39 74 40177 -41 77 42 77 43 78 44|86 River 200 feet wide Ohio None. 12' -f Slight. 9' + Tide. 40' Swift. 54' Swift. 34' - Moderate. 20' -f- Moderate. 6' + Swift. 20' -i Swift. 21' - None. 15' -)- Moderate. 7' -j- Moderate. 6' + Moderate. 6' -+- Moderate. 7' -j- Moderate. 6' -j- Moderate. 6' Swift. 19' Moderate. Deep. None. 12' -f- Swift. ; 22' Swift. 15' Moderate. 25' Tide. 28' Moderate. ! 28' Moderate. 6' -f- Tide. 30' - Tide. 10' Moderate. 7' Moderate. 6' -+- Swift. 6' + Tide. 6| + None. 20' Swift. Moderate. 10' -(- Swift. 8' -f Moderate. 6' -\- Moderate. 25' -j- Moderate 10' -j- Swift. 8' + Tide. 13' + Moderate. Deep. Tide. 35' Swift. 6' + Tide. 15' + Cemented gravel. Gravel, sand, mud. Sand & gravel over clay. Gravel over clay. Gravel over hardpan. Gravel. Soft. Rock. Rock. Sand. Gravel over rock. Soft. Sandy. Gravel over soapstone. Rock. Soft. Soft. Mud over rock. Rock. Rock. Rock. Rock. Clay over rock. Rock. Gravel. Mud and clay. Sand and gravel. Sand and mud. Rock. Rock. Soft. Soft. Gravel over rock. Sand over hardpan. Sand. Sand. Clay. Gravel over rock. Rock. Earth over rock. Mud. Rock. Sand. Rock. Danube at Bud a Pesth . - Western part United States Western part United States Western part United States Western part United States Payette and Weiser, Union Pacific. Mississippi Fort Madison Schuylkill near Philadelphia, Pa... U S Canal Keokuk Arthur Kill Bridge Coteau Bridge C Pac Ry Ann Arbor, Mich., M. C. Ry Arthur Kill Bridge Boston Harbor sewer Illinois River La Grange Kankakee at Alomence Potomac at Harper's Ferry C harlestown Bridge Boston Cumberland Md Mississippi Sandy Lake Arkansas Little Rock . . Parnitz, Stettin, Germany Coosa, Gadsden, Ala Schuylkill P & R R R St. Helier Bridge, Jersey, Eng. . . . Thames at Putney Victoria (B C ) Docks Kaw at Topeka SYNOPSIS OF EXAMPLES. Form of Construction. Inside Dimensions. Kind of Puddle. 'Vuddle* 8 Remarks. i ' Earth bank. TO' X 60'? Clay and gravel. 5' 4- No leaks. 5 Sheet piles. 20' X 58'? Clay. 3' 6 2 Sheet piles. Large. Clay, sand &. gravel. 3~6' Typical. 8 3 Sheet piles. 72' X 136' + Clay and gravel. 2-5' Difficult. 9 4 Earth bank. 90' X 33' Clay and gravel. 19' 4~ J 4 5 Earth bank.? 200' X 600' Clay and gravel. Failed. 15 6 Crib. Medium. Clay. 3' 4- 17 7 Crib, single. 24' X 43' Concrete inside. 18 3 Crib, single. 1 6' X 34' Concrete inside. IS Q Crib, single. 17' X 43' Clay outside. Special. 18 10 Crib, single. Medium. Clay outside. 20 II Sheet piles. Medium. Typical. 20 12 Sheet piles. Medium. Clay outside. 20 13 Sheet piles. Medium. Clay outside. 20 14 Sheet piles. Medium, Clay. Equal depth. 20 15 Box or crib. 12' X 36' None. On grillage. '< 20 16 Staves. 36' diam. None. On grillage. 20 17 Sheet piles. So' diam. None. Failed. 24 18 Canvas on plank. So' long. Rotten manure. Bulkhead. 30 19 Crib, double. 28' X 64' Clay. 3'o" Canvas used 32 20 Box and canvas. Square. Clay outside. Movable. 33 21 Polygon crib. [47' diam. Clay. 4' 6 " 36 22 Polygon crib. 44' diam. Clay and gravel. 5' o" 38 23 Crib, single. i 34' diam. Concrete inside. 39 24 Sheet piles. 13' X 44' Clay and gravel. 2' 3" 59 1 25 Sheet piles. Large. None. Two trials. 60 j 26 Sheet piles. 12' wide. Clay. 6' to 8' 60 27 Sheet piles. Medium. None. 62 28 Sheet piles. Medium. Gravel. Two trials. 62 29 Sheet piles. Medium. Gravelly clay. 63 Sheet piles. 18' 6" X 119' Concrete inside. 63 3i Sheet piles. 44' diam. Sand and concrete. 7' -f- 64 32 Sheet piles. Large. Clay. ()' o" 66 33 Sheet piles. 15' x 50' None. 67 34 Sheet piles. 829' long. Clay. S' 67 35 Sheet piles. 1 6' X 38' Earth outside. /o 36 Sheet piles. 23' X 55' Clay. 2' to 4' Removal. 72 37 Sheet piles. 28' X 28' Clay. 12' -4- Removal. 74 38 Sheet piles. 16' X 42' Clay and gravel. 8' 4- Movable. 74 39 Sheet piles. Medium. Clay outside. 77 40 Sheet piles. Medium. None. 77 4 1 Sheet piles. 500' long. Clay. 2-7' 77 42 Sheet piles. 18' X 55' Clay outside. 78 43 Metal. 60' diam. Concrete seal. 86 44 LIST OF ILLUSTRATIONS. NUMBER PAGE The Pont du Gard, Mimes, France Frontispiece. 1. Bridge at Shuster, Persia, over the River Karun 2 2. Bridge over the Adda at Trezzo Milanese 3 3. Caesar's Bridge over the Rhine 4 4. A Primitive Solution. (Earth-bank Coffet -dam .) 6 5. Coffer-dam in Tide- water. (Sheet Piles and Puddle.} 8 6. Buda-Pesth Suspension Bridge. (Puddle Coffer-dam.} 9 7. Buda-Pesth Suspension Bridge, Plan of Coffer-dam No. 3 n 8. Scraper Dredge. (For Drag Dredging, C. & M. V. Ry.) 14 9. Coffer-dam at Dam No. IT, Gt. Kanawah River. (Earth and Crib.} 15 TO. Crib Coffer-dam, C., B. & Q. R. R. ( With Puddle Chamber.} 16 11. St. Lawrence River Bridge, C. P. Ry. (Crib and Coffer-dam] 17 12. Arnprior Bridge, C. P. Ry. (Crib and Coffer-dam} 18, 19 13. Crib Coffer-dam, A., T. & S. F. Ry. (A'o Puddle Chamber.} 21 14. Coffer-dam on Grillage, Payette and Weiser Rivers, U. P 22, 23 15. Coffer-dam on Grillage, Fort Madison Bridge, A., T. & S. F. Ry 24 16. A Crib Coffer-dam after a Flood. (Showing Plant.} 25 17. Apparatus used to force Clay into Crevice of Rock. (Leak.} 29 18. Details of Canvas and Plank Btilkhead, Keokuk, la 31 19. Inside View of Bulkhead, Lock pumped Dry, Keokuk, la 34 20. Canvas Funnel for closing Leaks. (Springs.) 35 21. Cribs for anchoring St. Louis Coffer-dam. (Crib and Puddle.} 36 22. Polygonal (Crib} Coffer-dam. Harlem Ship Canal Bridge 38 23 Details Coffer-dam, Arthur Kill Bridge. (Crib and Puddle.) 37 24. Coffer dam for Pivot Pier, Coteau Bridge. (Crib.} 38 25. Perronet's Pile Driver. (Historical; Man Power.) 41 26. Perronet's Bull-wheel Pile Driver. (Historical; Horse .Power.) 41 27. Sheet-pile Driver. (Hand-power Derrick.) 41 28. Pile-driver Derrick for Use on a Scow 42 29. Lidgerwood Pile-driving Derrick 43 30. Hammer with Nippers. (For Horse Power.) 43 31. Pile-driving Scow, New York State Canals. (Steam.) 44 32. Warrington-Nasmyth Steam Pile Hammer 45 33. Warrington-Nasmyth Hammer, Fair Haven Bridge 46 34. Cram-Nasmyth Steam Pile Hammer 47 35. Machine for sawing off Piles under Water 48 36. Pile-pulling Lever. (Hand Power.) ' 49 37. Pile-pulling Scow. New York State Canals. (Steam.) 50 xiii XIV LIST OF ILLUSTRATIONS. 38. Sheet Piles and Sheet-pile Details 51 39. Charlestown Bridge. Driving Wakefield Sluet Piling 52 40. Arrangement and Diagrams of Sizes for Sheet-pile Coffer-dam x 55 41. Sheet-pile Guides and Clamps 57 42. Coffer-dam for Ann Arbor Bridge, M. C. Ry. (Sheet Piles and Puddle.) 58 43. Sewer Coffer-dam, Boston Sewerage System. (Sheet Piles and Puddle.} 59 44. Wakefield Sheet Piling. (Details. ) Ci 45. Type of Momence and Harper's Ferry Coffer-dams. (Sheet Piling.) 62 46. Coffer-dam on Charlestown Bridge. (Sheet Piling.} 03 47. Resevoir Coffer-dam, Fort Monroe, Va. (Sheet Piling.) 65 48. Compound Sheet Pile 67 49. Chattanooga Bridge, Bed-rock Pier No. 3 68 50. Framework of Coffer-dam, Cumberland, Md. (Sheet Piling} 69 51. Sandy Lake Coffer-dam. (Sh^'t Piling.} 7 o 52. Coffer-dam and Concrete Pier, Little Rock, Ark. (Sheet Piling 71 53. Removal of Masonry Pier at Stettin, Germany. (Sheet Piling.} 73 54. Coosa River Coffer-dam. (Sheet Piling.} 75 55. Stock Rammer. (For packing Clay to stop Leaks.) 77 56. Topeka Bridge Coffer-dam. (Sheet Piling.} 78 57. Havvkesbury Bridge, Caisson No. 6. (Metal Shell.') Si 58. Group of Cylinders for Pivot Pier. (Metal Shells.} 82 59. Pier of Two Cylinders, Victoria Bridge. (Metal Shells.} 83 60. Circular Saw for cutting off piles under Water 84 61. Cylinder-pier Bridge, Riga-Orel R. R., Russia. (Metal Shells.) 85 62. Cylinder Piers, with Diaphragm. (Metal Shells.} 86 63. Circular Granite Pier, Forth Bridge 87 64. Forth Bridge. (Metal Coffer-dam } 88 65. Forth Bridge. (Circular Granite Pier and Metal Coffer-dam} 90 66. Old Bascule Pump. (Hand Power.) 93 67. Old Chapelet, Side Eelevation. (Water-power Pump.) 94 68. Old Chapelet, End Elevation. (Water-power Pump.) 94 69. Hand Pump, Soldered Joints 95 70. Hand Pump, Screw Joints 95 71. Diaphragm Pump. (Hand Power.) 95 72. Van Duzen Jet Pump. (Steam Power.) 96 73. Lansdell's Siphon Pump. (Steam Power.) 96 74. Pulsometer Steam Pump 97 75. Section of Pulsometer 97 76. Centrifugal Pump, directly connected to Engine 98 77. Suction Details for Pumps ; 99 78. Centrifugal Pump, Double Suction 100 79. Dredging Pump 100 80. Dredging-pump Piston 101 81. Lancaster Grapple. (Derrick Dredge.) 102 82. Sand Digger. (Light Elevator Dredge. ) 103 83. Osgood Dipper Dredge, New York State Canals IO ^ 84. Osgood Dipper Dredge, Details, New York State Canals Io ^ 85. Metal Tube for Concreting Io -. 86. Metal Bucket for Concreting TO S LIST OF ILLUSTRATIONS. XV 87. Concrete Piers. Red River Bridge 109 88. Concrete Forms, Red River Bridge no 89. Concrete Forms, Illinois and Michigan Canal in 90. Stone Crusher and Concrete Mixer, I. and M. Canal 112 91. Double-drum Guy Derrick, Am. Hoist & Derrick Co 113 92. Single-drum Horse Power, Con. Plant Mfg. Co .' 114 93. Double-drum Hoist Engine, Lidgerwood Mfg. Co 114 94. Crocker-Wheeler Electric Hoist 115 95. Lidgerwood Cableway Carriage and Skip 1 16 96. Lidgerwood Cableway at Coosa Dam. (Span 1012 Feet) n3 97. Hand Drill and Swab 121 98. Steam-power Well Driller 122 99. Test-boring Apparatus, Mississippi River Commission 123 100. Clamp and Maul. (Test Boring.) 124 101. Pier of Omaha Bridge, Union Pacific System 126 102. Russian Pier, Russian State Railways 127 103. Cresy's Experiments on the Form of Piers. 128 104. Cresy's Experiments on the Form of Piers 130 ARTICLE I. NTT = THE COFFER-DAM PROCESS FORMERS. ~~ HISTORICAL DEVELOPMENT. HE continued increase in the weight of our bridge super- structures and of the loads they have to carry has led to increased care, to a very gratifying degree, in the prep- aration of the foundations for bridge piers and abutments. An old authority very truly states "The most refined elegance of taste as applied in the architectural embellishment of the structure; the most scientific arrangement of the spans and disposition generally of the superior parts of the work; and the most judicious and workmanlike selection and subsequent combination of the whole materials composing the edifice, are evidently secondary to the grand object of producing certain firm and solid bases whereon to carry up to any required height the various pedestals of support for the spans of the bridge." There is every reason to believe, from the bridges of the Romans still extant and of those of ancient and mediaeval times of which there are remains or records, that the foundations were carefully considered. The most ancient form was likely begun by dumping in loose stones until the surface of the water was reached and the masonry could then be com- menced without the necessity for any method of excluding the water. The oldest civilizations were in tropical or semi-tropical countries where the streams are dry beds for many months in the year and suitable foundations were easily made without water to contend with. Where the bottom of the stream was rock, the engineering could be very little improved upon to-day, and even where there was shallow water on rock bottom, the piers were well founded in the shallowest places, the bridge often winding across the stream in serpentine form, similar to the bridge over the river Karun, at Shuster, Persia. Fig. 1. The arch was developed to such an extent by the Romans, and the spans were increased to a length which rendered the construction of piers in the water unnecessary for short bridges, the abutments or skewbacks being without the stream on either bank. The difficulty of founding piers in midstream was doubtless the con- trolling cause for the larger spans, such as the one built at Trezzo, over the river Adda, by order of the Duke of Milan, sometime prior to the year 1390. The span at low water was 251 feet, the single arch being of granite in two THE COFFER-DAM PROCESS FOR PIERS. 3 courses. The placing of a middle support was doubtless found to be imprac- ticable and caused the design of an arch which has never been equaled or eclipsed. Fig. 2. The construction of roads has ever been the harbinger of civilization, and with the spread of civilization came a demand for the improvement of means of communication. The engineer was called upon to construct better and greater bridges in a permanent manner, which led to the origin and development of the four methods for founding in water that were used in FIG. 2. BRIDGE OVER THE ADDA, AT TREZZO, MILANESE, A PROBABLE RESTORATION. THIi SHADED PORTION OF ARCH RINGS IS A I, I, THAT REMAINS. olden times. These may be classified as, first, the method with open cais- sons; second, the use of piles with a capping of coarse concrete about the tops; third, the use of piles after the manner of the French encaissement; and fourth, the use of coffer-dams. A fifth method might be added, in which the bridge was built on dry land adjacent to the stream, and the river diverted to a new channel afterwards excavated under the completed struc- ture. This is, however, an avoidance rather than a solution, unless the river is to be diverted in the course of its improvement. The first method, as described in old treatises or accounts, consisted of little more than baskets formed of branches of trees, weighted with stone to sink them, and after sinking filled with loose stone to near low water level, 4 THE COFFER-DAM PROCESS FOR PIERS. where the masonry could be commenced. These baskets were similar in construction to the mattresses used in the bank revetment of the Mississippi or the bamboo casings used by the Japanese to hold stones in place on bank protection. An improvement was effected by using in place of baskets, boxes or small open caissons which were sunk and filled in the same manner, several being used for one pier. This was the method used at Black friars bridge and also at Westminister bridge, over the Thames, and has been much used in recent times, the caisson being built large and strong enough for the entire pier, which is built up as the caisson sinks. The second method consisted of driving piles over the area of the foun- dation until the heads were below low water level, and spaced at distances FIG. 3. TEN DAYS TO CONSTRUCT; LENGTH ABOUT A QUARTER MII.E; DEPTH WATER 16'; WIDTH, 25'; BEAM, 2 X THICK; ABOUT 50 PIERS. apart as required by the nature of the bottom, similar to the methods in vogue to-day. The heads of the piles were not driven to the same level, however, and were incased in a form of coarse concrete such as was used by the Romans, but what is now called beton. This was leveled up and on it was laid the stone for the footing course of the pier. The third method of encaissement was probably an improvement of the dumping in of loose stone on which to place the pier, and consisted in inclos- ing the space for the pier with sheet piling, after which the loose material was removed from the bottom as much as possible and the stone dumped inside until nearly up to low water, at which time the pier could be begun. THE COFFER-DAM PROCESS FOR PIERS. 5 These last two methods doubtless met with much favor owing to the familiarity with pile driving, in which the Romans especially were proficient. Caesar's bridge over the Rhine was built entirely on piles, and in a view of it after the old print in the Museum de St. Germaine, is pictured a pile driver on a float in position for driving. Fig. 3. This third method was the early type of the crib which has been such a factor in the building of the earlier foundations over our American rivers. Crossed timbers laid up crib fashion with rectangular openings or cells between the timbers were sunk and filled with broken stone on which to build the pier. These methods were all deficient in affording no means of seeing or mak- ing a careful examination of the bottom on which the foundation was to be placed, and with the advent of more permanent structures of greater magni- tude the coffer-dam came into use. This allowed the bottom to be freed from water and after a careful examination and preparation of the founda- tion, the work could proceed in the dry until above water level. The pneumatic caisson is now in general use for all foundations that must go to any considerable depth below the water and has even been used in some instances where the depth was slight, but where for various reasons it was deemed expedient louse compressed air caissons. Recent expressions from some engineers of high standing would indicate that they do not con- sider it good practice to use coffer-dams in any case, one making the state- ment that he had not used a coffer-dam for thirty years, while another seemed to think it a matter to be left to the pleasure of the contractor. That the use of this method has gotten into disfavor seems to be beyond question and it will be the purpose of the succeeding pages to learn to some extent why this is so, but mainly to show from successful examples how to pro- ceed, that success instead of failure may result. Any attempt to account for the origin of the coffer- dam process would be futile, inasmuch as the savage, wishing to free a space from water, doubtless banked up earth about the area and. scooping out the water with his hands, laid the ground bare for inspection. From so simple a beginning, the first method likely to occur to a mind capable of reasoning, can readily be imagined the course of develop- ment of coffer-dams. The most simple form in use at the present time, where the water is quiet, is shown in the Fig. 4, and consists principally of a bank of earth which is made thick enough to be nearly or quite impervious to water, the earth being prevented from caving into the excavation by piles supporting a timber casing. Some of the recorded examples of the early use of this process are of interest in illustrating the care which was bestowed upon their construction in important works and will call attention to that incess- ant care which is necessary to success in any work of this character. 6 THE COFFER-DAM PROCESS FOR PIERS. Robert Stevenson, the great English engineer, thought it not beneath his dignity to give full instructions as to the construction of the coffer-dams for the Hutcheson bridge over the Clyde at Glasgow. The bridge consisted of five arch spans, the total length between the abutments being 404 feet and the width 38 feet. The four piers were from 11 to 12 feet in thickness, being designed to take up the arch thrust, and 48 feet in length at the foot- FIG. 4. A PRIMITIVE SOLUTION. ing. The specifications written at Edinburgh in April, 1828, are so explicit that they will be quoted in full on this point: "It having been ascertained by boring and mining that the subsoils of the bed of the river consist of gravel, sand and mud to the depth of 27 feet and upwards, it becomes necessary to prepare foundations of pile work for the bridge; and, therefore, THE COFFER-DAM PROCESS FOR PIERS. 7 to insure the proper and safe execution of the works, coffer-dams are to be constructed around each of the foundation pits of the two abutments and four piers of such dimensions as to afford ample space for driving piles, fix- ing wale pieces, laying platforms, pumping water, and setting the masonry; and likewise for the construction of an inner or double coffer-dam should this ultimately be found necessary. The framework of the coffer-dams is to consist of not less than two rows of standard or gauge and sheeting piles, kept at not less than three feet apart for the thickness of a puddle wall or dyke, which space is to be dredged to a depth of not less than nine feet under the level of the summer watermark above described, before the sheeting piles are driven. The gauge or standard piles are to measure not less than 24 feet in length and 10 inches square. They are to be placed three yards apart and driven perpendicularly into the bed of the river to the depth of sixteen feet under the level of the summer watermark, thereby leaving eight feet of their length above that mark. Runners or walepieces of tim- ber nine inches square are then to be fitted on both sides of each row of gauge piles, to which they are to be fixed with two screw bolts of not less than one inch in diameter, passing through each of the gauge piles. One set of these inside and outside walepieces is to be placed at or below the level of summer watermark, and the other set within one foot of the top of each row of said piles, the whole to be fixed with screw bolts in the manner above described. The walepieces are to be four and one-half inches apart in order to receive and guide the sheeting piles. This is to be effected by notching the walepieces into the gauge piles. The sheeting piles are to be 21 feet in length, 4^ inches in thickness, and not exceeding 9 inches in breadth. They are to be closely driven, edge to edge, along the space left between the walings, and each compartment of the sheeting between the gauge piles is to be tightened with a key pile. The coffer-dam frames are to be properly connected with stretchers and 'braces before commencing the interior exca- vation. Each coffer-dam is to be provided with a draw-sluice, fourteen inches square in the void, with a corresponding conduit passing through the puddle dyke at the level of summer watermark. To render the coffer-dams water tight the whole excavated space between the two rows of piling is to be carefully cleared of gravel, sand or other matters, to the specified depth, and clay well punned or puddled is then to be filled in and carried up to the level of the top of the sheeting piles. But if it shall, notwithstanding, be found that the single tiers of coffer-dam do not keep the foundation pits suf- ficiently free of water for building operations, the water must either be pumped out and kept perfectly under by steam or other power, or else excluded by the construction of a second tier of coffer-dam similar in con- struction to the first. For the foundation pits of the two abutment piers on either side of the river it is not expected that more will be required on the 8 THE COFFER-DAM PROCESS FOR PIERS. landward side for keeping up the stuff than a single row of gauge and sheet- ing piles; but if the engineer shall find other works necessary upon opening the ground they must be executed by the contractor and shall be paid for agreeably to the contract schedule of prices for the regulation of extra and short works. The stuff within the coffer-dams is to be excavated to the depth of ten feet under the level of summer watermark for each of the piers and eight feet for each of the abutments." The present practice of leaving all this to a contractor, whose idea is too often to sacrifice everything to cheapness, appears in very unfavorable con- trast to this careful description. An article on founding by means of coffer-dams, published in 1843, gives directions for placing a coffer-dam in forty feet of tide water; and FIG. 5. COFFER-DAM IN TIDEWATER. although the engineer of to-day would use some other method for such a depth, an illustration, Fig. 5, and short description of it are given, as ideas may be gained for application to ordinary works. The water was assumed at ten feet deep for low tide, twenty-eight feet at high tide, with twelve feet of sand and gravel to be removed to expose the clay on which the pier was to rest. Four rows of piles were to be driven around the area, the outer row to within one foot of low water, the two rows in the middle to within three feet of high water, the inner row to eleven feet above low water, and all to be down five feet into the clay. The outer row of piles to be six by twelve inches, the two rows in the middle twelve by twelve inches, and the inner row eight by twelve inches; all driven close together and to have walingpieces, braces and brace rods as shown in cross section. The rows to be six feet apart and to be filled in between with a puddle of clay mixed with sand and gravel. The report of W. Tierney Clark, the engineer of the Buda Pesth suspen- sion bridge, gives an account of what are probably the largest bridge coffer- THE COFFER-DAM PROCESS FOR PIERS. 9 dams ever constructed. Some other method would now be used for such a location, but this fact will not detract from the lessons that may be drawn from them. The Danube was crossed at Buda Pesth previous to the year 1837 by means of a bridge of boats which had to be taken up during the winter and the passage made by ferry or on the ice, so that for six months of each year there was great risk in crossing and frequent loss of life. The building of a permanent bridge was brought about through the efforts of the Count Szechenyi, who, as a member of a committee, proceeded to England in 1832 and after a careful investigation of existing works decided upon the con- struction of a suspension bridge. The greatest question for solution was the founding of the two towers in a river like the Danube, where the ice Trans\crtt Sfctic J Ccfler Dam' FIG. 6. BUDA PESTH SUSPENSION BRIDGE. throughout the long winter wrought havoc with everything in reach. The ice in the river in February, 1838, was from six to ten feet thick near the site of the proposed piers. On March 9 a movement occurred across the whole river and for a length of 350 yards, the whole moving in a solid mass. On March 13 it moved again 400 yards and three hours later a gen- eral breaking began. The ice piled up on the shoals causing a sudden rise to twenty-nine feet five inches above zero, and while it was at this height for only a few hours, it is recorded that a great part of Buda and two-thirds of Pesth were destroyed and many lives lost. The extraordinary design of the coffer-dams can the more readily be understood after this description, it being doubted by many persons at that time whether piers could be placed in the river by any means. Fig. 6. 10 THE COFFER-DAM PROCESS FOR PIERS. The drawings reproduced are of coffer-dam No. 3 which was about 72 feet in width and about 136 feet in length inside the puddle walls, there being two puddle chambers, each five feet in width. From a point about thirteen feet above the clay on which the tower was to rest, was an inside wall of sheet piling, this space being nearly filled after excavating, with a bed of concrete. The piling of each row, from forty to eighty feet in length, was all carefully sized to fifteen inches square, shod with iron and driven close together, penetrating twenty feet below the bed of the stream or forty feet below the zero level. The framing of the ice breaker and the bracing within the dam was of enormous strength. The number of piles driven in the four coffer-dams reached the enormous total of 5,224, and of the 1,227 driven in darn No. 3, 16^ P er cent, were drawn and redriven. These piles and the timber were obtained from the forests of Bavaria and Upper Austria. Fig. 7. The first pile on dam No. 3 was set on April 8, 1842, but owing to the difficulties encountered it was not finished until three years later April 4, 1845. From six to seven days were occupied at the first in driving a pile to a depth of five or six feet into the clay, but as the work progressed the difficulty increased, the operation of driving one pile consuming from twelve to fourteen days, many piles breaking short 'off so they could not be with- drawn, and the gravel was dredged out from behind and a second row driven. The report further describes the difficulty of the work: "The dredging for the No. 3 dam was carried on to the average depth of forty-four feet from the top of the outer row of piles, leaving about ten feet of gravel to drive through, and extra piles were driven where the gravel found its way between the piles, as well as where it was known the piles were not driven to the proper depth, or were broken or otherwise injured. As the gravel was dredged out to the above depth, the inner and middle row of piles were driven, and a great part of them got down as was supposed to the requisite depth. The work was carried on in the above manner until the 7th of No- vember, when from the appearance of several piles which were pulled up, and from other causes, it became apparent that the outer row was in a much worse state than had been expected and was almost a matter of certainty, that those piles which had taken ten or twelve days to get down were not driven to the proper depth by at least three or four feet, having upset or lost their points to that extent. There was likewise every reason to believe that many of them were broken or dangerously crippled. Added to this the Danube was rising, and at the late time of the year, with winter rapidly approaching, the general appearance of the dam was anything but satisfac- tory. Upon mature consideration the only course appeared to be to drive a much greater number of piles than was at first calculated upon, and another complete row of piles was driven all round at intervals of fifteen inches apart, and in some cases double and triple piles were driven during the progress of 12 THE COFFER- DAM PROCESS FOR PIERS. the dredging. At the commencement of the driving a few were got down to the depth of fifty-seven or fifty-eight feet, being from three to four feet in the clay; but as the gravel began to get compressed many of them would not penetrate more than fifty- four or fifty-five feet, the sharp angular gravel overlying the clay appearing to be compressed into a substance as hard as rock." The puddle used was clay mixed with about one-third clean gravel, it having been found to set quite solid, from experiments made by sinking specimens in the Danube. When leaks occurred they were closed by driv- ing square timbers down thirty or forty feet into the puddle to pack it or by driving new piles to close the cracks and in some cases by driving sheet piling. Experiences of this nature led to the disuse of coffer-dams for founda- tions to such depths, but a very small percentage of the care exercised and the persistence shown in this work would lead to greater success on ordi- nary foundations. The class of work to which coffer-dams may still be applied will be shown in the succeeding pages and the examples from actual practice will show in some measure the care that must be exercised in the first construction to prevent failure, and the expedients adopted to overcome unavoidable acci- dents. "In every man's mind, some images, words and facts remain, without effort on his part to imprint them, which others forget, and afterwards these illustrate to him important laws." ARTICLE II. THE COFFER-DAM PROCESS FOR PIERS. CONSTRUCTION AND PRACTICE. HE exact definition of the term coffer-dam "a water-tight inclosure, from which the water is pumped to expose the bottom and permit the laying of foundations" is the class of structure which is to be considered, although in the construction of them cribs or caissons may be employed and utilized; the essential purpose being to form an inclosure as nearly watertight as possible in order that the expenditure of power for pumping out the water may be of small amount. The attainment of this when the water is shallow and has little current we have seen to be easily accomplished by means of a bank of clay or clayey gravel. This form may also be employed in still water up to about four feet in depth by the addition of sheet piling or a casing supported by ordinary piles to prevent the embankment from caving into the excavation. Where the bottom is of soft mud or porous material over a solid clay or gravel, as much as possible of the porous material should be removed before forming the embankment, thus preventing leakage underneath. In very shallow water this can be accomplished by shoveling and with large hoes or scoops, but with several feet of water to contend with, some form of dredge or scraper must be employed. A very convenient form of scraper used by M. L. Byers on the Cinti. & Mus. Valley Railway is described in Vol. 31 of the ''Transactions of the American Society of Civil Engineers," and consists of old boiler iron, strengthened by three ribs of light iron rail as shown in Fig. 8. This was operated by a double drum 20 horse-power Mundy hoisting engine, with the towing line running directly from one drum to the scraper and the back line from the other drum over a sheave to the front of the scraper. The excavating averaged about forty-five yards of material each day during twelve days' work. The weight of the device was about one thousand pounds. Where the material is very soft, a hand dredge called a spoon will accom- plish the work at about the same cost as excavating on dry land. The spoon usually consists of a long pole, having a cutting ring fastened at one end, and to this ring is attached a canvas bag to contain the excavated ma- 13 THE COFFER-DAM PROCESS FOR PIERS terial. The ring is hung from a derrick with a set of falls, being guided with the pole, as it is dragged forward by the derrick through the material to be excavated. Excavating may be done on all the larger rivers by employing the sand or gravel diggers which are most always to be found, the dredging being accomplished by means of a series of buckets on a belt or on chains operated through a well in the bottom of a barge. Dredging by machinery on a large scale will be considered later on in some detail. The method of embankment is sometimes employed for greater depths than four feet and in some instances successfully. The Chanoine dams on the Great Kanawah River required substantial foundations beneath the water, and to accomplish this Addison M. Scott, the resident engineer, employed log cribs about the spaces, with earth banked up on the outside. This work is described in the report of the Chief of Engineers for 1896. The site of the navigation pass of dam No. 11 including the center pier, re- quired a coffer-dam 90 feet wide and 330 feet long inside. (Fig. 9.) This area including the necessary room for the cribs, was dredged out to hardpan from 20 to 24 feet below low water. The log cribs which contained about 84,000 lineal feet of logs, were sunk in sections 19 feet wide and 20 feet long. They were sheathed up to about three feet above low water, with sheet piling in three layers, on the Wakefield system. The driving was accomplished by attaching an eighty-pound weight to an Ingersoll-Sergeant drill run by steam and utilizing the reciprocating motion by attaching the drill with clamps to the tops of the sheathing, following it down as it was driven, after the manner of the Nasmyth steam pile hammer. This device, which is one of FIG. 8. SCRAPER DREDGE. THE COFFER-DAM PROCESS FOR PIERS. 15 the most ingenious ever devised for the purpose, was arranged by the con- tractor's engineer, S. H. Reynolds, and was a complete success. The tops ol the cribs were ten feet above low water, and the bottoms rested on the hardpan, making a total height of from thirty to thirty-four feet. The cribs were tilled with sand and gravel that had been dredged out, FIG. 9, COFFER-DAM AT DAM NO. 1 GREAT KANAVVAH RIVER. but the outside was banked up with selected clay and dredged material, which was protected by a layer of riprap up to about low water. When the coffer dam was first pumped out several leaks were developed, but after one week in perfecting the details the pumps were started regu- larly and no serious trouble was had afterward. This is only one of a ser- ies of coffer-dams which have been constructed on the several dams in this river, and owing to the care exercised good results were obtained uniformly. The construction of a similar piece of work on the Ohio river was begun by Major R. L. Hoxie, corps of engineers, and is described in the report of 1895: "It was originally planned to enclose the site of the dam and lock within a coffer-dam, and work was commenced upon that basis. But on 10 THE COFFER-DAM PROCESS FOR PIERS. attempting to pump out the inclosure, it was found that water came in in large quantities, not only under the dam but from springs in the bottom, and all attempts to close these by dumping clay and gravel was a failure. The area inclosed by the dam was about 600 by 200 feet or about three acres of river bottom. The deposit of sand and gravel overlying the rock was about thirty-five feet thick, the rock being forty-five feet below the water level, while the plans required an excavation twenty feet deep below this water surface. The bottom deposit had been worked over for years by sand- diggers who threw back the large stones and coarse gravel after removing FIG. 10. CRIB COFFER-DAM; CHICAGO, BURLINGTON AND QUINCY RAILROAD. the fine sand, this work resulting in a very permeable bottom, with possible channels of comparatively large dimensions extending to unknown distances beyond the limits of the coffer-dam." This is perhaps the most frequent source of failure of a well constructed coffer-dam and should always be guarded against by removing as much of the porous material as possible, by dredging before the construction of the coffer-dam is begun. Cribs are very easy to construct, usually very substantial, and easy to THE COFFER-DAM PROCESS FOR PIERS. 1 7 make use of by floating to position and then sinking in place. A very sim- ple form that has been used on the Chicago, Burlington & Quincy railroad is described by E. J. Blake, chief engineer. Where the water is shallow they have been built in the form shown (Fig. 10) of fence boards spiked one piece on another; with deeper water they are made of heavier timber 2"x8" or 2"xlO". They are built on the water and are tied across at inter- vals by pieces spiked through the wall, w T hich pieces should be carefully fitted to prevent leakage. In some cases where the bottom is soft, instead Of dredging, a bottom is added to the crib to prevent the filling from squeez- ing its way out from under the edge. When the crib has reached bottom, being sunk by weighting it down if n FIG. 11. ST. LAWRENCE RIVER BRIDGE CRIB AND COFFER-DAM, CANADIAN PACIFIC RAILWAY. necessary, the chambers are filled with clay puddle and clay is banked up around the outside to prevent water running under. The crib" is made large enough so that the excavation will leave an easy slope to the inner edge of the timber work. This form can be made to conform readily to the contour of the bottom by starting the layers of timber at different elevations. No leakage has been experienced except what, can readily be kept under control with ordinary sized centrifugal pumps. The cost of construction is gener- ally a minimum, as there are usually plenty of old timbers available for use from the railroad yards. iS THE COFFER-DAM PROCESS FOR PIERS. Cribs constructed in a similar manner but with only one wall of timber have been used successfully on the Canadian Pacific Railway by P. Alex. Peterson, chief engineer. The bracing is very efficiently attached by dovetailing it into the sides, while the form of the crib enables it to withstand the force of the current and the ice. The projections on the inside are to prevent the water from forcing its way up between the sides and the concrete filling when the dam is pumped out. These projections answered their purpose very effectually , and when the dam was pumped out it remained dry enough to lay the masonry without any additional pumping. Illustrations are given of a crib of this character which was used on the St. Lawrence river (Fig. 11) similar ones being used for the other piers of the same bridge, and of the crib used for the Arnprior bridge. (Fig. 12.) This shows the concrete which was deposited on which to found the masonry, and 3 I E FIG. 12. ARNPRIOR BRIDGE CRIB AND COFFER-DAM, CANADIAN PACIFIC* RAILWAY. which formed a watertight bottom so that the crib could be pumped out for the laying of the stone. The practice on the Atchison, Topeka & Santa Fe railroad has been in some respects similar to what has been given. C. D. Purdon, assist- ant chief engineer, states that cribs built of old timbers are used when such material as stringers 7"xl6" is plentiful, each course being stepped in about one-half an inch to give a batter. For use in sand when rocks and drift are likely to be encountered a crib is made by constructing a frame of old bridge timbers and sheathing it with plank. (Fig. 13.) This is sunk by digging out the sand, which is shoveled first into box A, then to boxes B, then to C, and then outside. The suction pipe is shown in dotted lines, the pumping being accomplished with a centrifugal pump. This plan works very sue- 20 THE COFFER-DAM PROCESS FOR PIERS. cessfully on the streams in Colorado and New Mexico where the water is mostly in the sand and but little shows as surface water. The Arkansas river bridge of the St. Louis & San Francisco railroad at Tulsa was built over a bottom of gravel and riprap above rock, which was quite level and about seven feet below low water. Cribs were constructed for coffer-dams similar to the one just described and set on the bed of the stream. Clay from the bank was dumped outside and as the crib was dug out and sunk, the clay followed down and kept out the water. When the bottom is of clay or of sand without obstructions, sheet piles, either tongue and groove or the Wakefield, are driven around a crib. Geo. H. Pegram, chief engineer of the Union Pacific system, has made the construction of coffer-dams conform to available material and local condi- tions. At the crossing of the Republican river in Kansas, where the bottom was sandy, a single thickness of four-inch V-shaped tongue and groove sheet- piling, with the usual guide piles and wales, served to form a watertight structure. Where a gravel bottom overlaid a hard soapstone, as on some work in Idaho, with seven feet of water to contend with, the coffer-dam was made of Wakefield piling, formed of 1^-inch sized plank. The joints were tightened with cement; and sand, gravel and straw placed outside to prevent leaking. Wakefield piling has also been used for clean rock bottom, placed in two rows about the depth of the water apart. Intermediate cribs filled with rock were used to sink them. The ends of the piling were sharpened and driven on the rock until broomed up and rendered nearly watertight, when gravel mixed with straw was placed around outside to close any remaining leaks. In cases where ordinary piling has been driven and a grillage laid upon them to receive the masonry, a coffer-dam is constructed as shown (Fig. 14) in which to lay the masonry. The construction of this is fully shown in the different views given. Another form of coffer-dam for the same purpose was constructed by Octave "Chanute in laying the masonry of the pivot pier for the Fort Madison bridge over the Mississippi river, on the line of the Atchison, Topeka & Santa Fe railroad. (Fig. 15). This is described in the Engineering News of June 2, 1888, by W. W. Curtis, resident engineer : "The grillage (for the pivot pier) is four feet, three inches thick, the upper fifteen inches being dressed to an accurate circle of the desired diameter. The coffer-dam was footed against these two courses and was formed of 3"x8"pine plank staves, dressed on the sides to a slight bevel around which were placed seven wrought iron hoops 4"x T y, 5"x ", and 6"xyV'> similar to those used for water tanks, and screwed up tight. Inside of these, circular braces of plank were fitted. As a water pressure of nineteen feet was to be resisted, additional security against leakage was obtained by placing a string of candle wicking vertically between r TTTT 1 Half Plan. FIG. 14. COFFERDAM ON GKILLAGK ; PAYETTE AND WEISER RIVER BRIDGES. UNION PACIFIC SYSTEM. 6' O" 6'0' >4"x6"*7' 3"xl2"xiz' 3 * 12" x 12' View. FIG. 14 COFFER-DAM ON GRILLAGE, PAYETTK AM) WEISER RIVER BRIDGES. UNION PACIFIC SYSTEM. 24 THE COFFER-DAM PROCESS FOR PIERS. each stave. When the caisson was submerged to about full depth it became necessary for the steamboat to assist it into final position. A 12"xl2" post was bedded in the concrete in the center of the pier, with four braces run- ning to the circular bracing of the sides. This makes a very cheap coffer- dam and was found to work very well." An attempt to use a form similar to this was made in constructing the Walnut Street bridge at Philadelphia. This is described by Geo. S. Web- ster, chief engineer Bureau of Surveys, in the Engineering News of March 15, 1894: "In founding the river piers, the Robinson coffer-dam was first tried, but was abandoned after three of them had failed by collapsing. This Section o-f Pier FIG. 15. COFFER-DAM ON GRII,I,AGB, FORT MADISON BRIDGE, ATCHISON, TOPEKA AND SANTA FF, RAILWAY. dam may be briefly described as follows : A circular platform about eighty feet in diameter supported upon piles at an elevation of about four feet above high water was first constructed. Square piles of 12"xl2" yellow pine were then prepared by spiking a 3"x4" timber flat, along the middle of one side, and two others along the edges of the opposite side, forming a tongue and groove on each pile. The tops were squared off and the bottom ends pointed to a wedge shape. These piles were then driven close together against the edge of the circular platform and down to rock. Mr. Robinson's idea was that the mud overlying the rock would hold the piles in position at the bottom, and if the top ends were held by an outside hoop, the dam would be secure without internal bracing to resist collapsing pressure. In the first trial the hoop was made of boiler iron some four feet or more in width. In the second dam it was formed of a heavy steel railway rail, and in the third THE COFFER-DAM PROCESS FOR PIERS, 25 dam the hoop was the same as in the second, but it also had a number of radial rods in addition. The first dam was pumped out and held for nearly an hour before collapsing, but the others collapsed before being entirely pumped out. After the third failure this form of dam was abandoned." It would seem likely from a comparison of the two cases, one being en- tirely successful and the other a failure, that had the Walnut street dam been supplied with additional bands lower down and provided with some means of tightening, with several internal bracing ribs of timber, it would have proven a success. These bands and ribs could likely have been placed by a diver. The uncertainty which always exists regarding any construction under FIG. 16. A CRIB COFFER-DAM AFTER A FIvOOD. water makes it imperative that every precaution should be taken to guard against troubles that might arise, by making the construction of no doubtful form and in no doubtful manner from its first inception. The nature of the bottom will always indicate the method of construc- tion which should be adopted in a given case, but it would be rarely that the preliminary dredging could be dispensed with. It is true that there are cases where there is a deposit overlying a seamy rock, and the water will find its way along the seams, bubbling up in springs inside. Resource must be had to cutting off the flow, by puddling on the outside, sometimes ex- tending the operations a distance of a hundred feet or more away, until enough of the flow has been stopped so that the water can be kept down b}' a reasonable amount of pumping. 26 THE COFFER-DAM PROCESS FOR PIERS. The next precaution after dredging, is the building of some form of coffer- dam which shall effectually exclude any flow through the sides of the dam. TJiis we have seen to be accomplished in many cases by means of a bank of clay, or a row of sheet-piling, and in some cases by a single walled crib. But in the last two methods a supplementary bank of clay or clayey gravel on the outside is necessary to prevent leakage. This bank may be protected from wash by covering it with clay, sand or gravel in gunny sacks, or by ripraping up to about low water, as was done on the Kanawah dams. Double walled cribs and coffer-dams constructed with two rows of water- tight sheet piling, require to be puddled with a carefully selected material. While clay can be used with a good degree of success, it will be found better to use a clayey gravel or to mix the clay and gravel, as was done at the Buda-Pesth bridge. When a small leak starts through a pure clay puddle, it washes out the clay in considerable quantities and a dangerous leak is soon developed. With the admixture of gravel, however, a leak is stopped almost as quick as started by the heavier gravel falling into and closing the void. It will generally be found advantageous to use a bank of clay outside of a double walled dam, unless it might be a case where sheet piling has been driven to rock, and even then a certain amount of material in sacks should be used to prevent wash or the cutting out of the earth around the sheeting. Whatever excavation is taken out of the interior of the coffer-dam after it has been pumped, should be dumped at the upstream end and corners, or to fill any holes or pockets there may be around the sides or ends. Cutwaters should be added to all coffer-dams which are built in rivers having a swift current or a heavy flow of ice, as was the case at Buda-Pesth and on the Canadian Pacific examples. They must also be used in rivers where the run of drift with each rise is of large amount. For the purpose of preventing wash around a dam, a cutwater of plank supported by a frame of timber may be constructed separate from the main structure, or a V-shaped row of sheet piling driven up stream. On rock, a timber crib of triangular shape, built of round logs, may be sunk up stream and filled with broken stone. Such a crib can be utilized in anchoring the main crib of a coffer- dam, as was done at St. Louis, and which will be described in future pages. More fitting language cannot be found for closing words than those used in Wellington's monumental work on railway location: ''The uncertainty as to the exact requirements to be fulfilled by the works when completed is a disadvantage, indeed, which cannot be escaped; but the more difficult it is to reach absolute correctness, the greater need we have of some guide which shall reduce the unavoidable guess-work to its lowest terms, and so save us from the manifold hazards which result from not only guessing at facts, but THE COFFER-DAM PROCESS FOR PIERS. 2/ at the effect of those facts. Whatever care we use we can never attempt with success to fix the exact point where economy ends and extravagance begins ; but what we can do is to establish certain narrow limits in either direction, somewhere within which lies the truth, and anywhere outside of which lies a certainty of error." ARTICLE III. THE COFFER-DAM PROCESS FOR PIERS. CONSTRUCTION AND TRACT ICE. I/S HEN for some reason the necessary care has not been exercised in the construction of a coffer-dam and in puddling it, or where there were discovered conditions not known before the con- struction began, which rendered the work unsatisfactory or leaky, it will usually be found that the mode of repair which seems most expensive will in the end prove the cheapest and most expeditious. If the puddle proves leaky, and it be decided that the material was of too porous a nature, the best remedy is to dig out and replace it with better. Should it be found that the porous bottom had not been removed to a sufficient depth, it may be found necessary to dig out the pud- dle chambers and puddle deeper, or the leaks might be stopped by banking up outside of the dam with clay or clayey gravel, or perhaps sand in sacks would do some good. Gravel will allow the percolation of water even where the head is small, and when a pressure of from four feet upwards is brought upon it, the leak- age becomes considerable and difficult to control, so that pure gravel is of little service in stopping leaks. Hay, straw, oats, crushed cane stalks, rotten stable manure, and similar materials, mixed with the banking material, are very efficacious in pro- ducing tightness, and when applied to local leaks will assist in closing them. Where sheet piling have been used to exclude the water and leaks still occur, they can often be closed by driving more sheeting to lap the cracks, which may have been widened out lower down as the sheet piles were first driven. This, we have seen, produced satisfactory results at Buda-Pesth, where leaks were also closed by driving square timbers into the puddle to compact it. Clay can also be forced down through pipes directly to where the leakage occurs. The use of this at the Government Lock at Sault Ste. Marie is de- scribed in the Engineering News of September 26, 1896: "The only diffi- culty encountered in the work of excavation was due to a leak in the coffer- dam, which flooded the lock pit and delayed the work considerably. The cause of this leak was found to be a crevice in the rock passing underneath the coffer-dam, and despite all efforts to close it, the flow of water rapidly enlarged the break until about fifty feet of the clay in the coffer-dam had 28 THE COFFER-DAM PROCESS FOR PIERS. 2 9 been washed away. The large break was closed by driving additional sheet piling and filling in with brush, hay, and clay in sacks. This, however, failed to entirely stop the leak through the crevice, and it was determined to fill the cavity with clay. For this purpose a 3-inch pipe was driven down through the coffer-dam until its lower end penetrated the crevice. In this A. Cutting theClay Cylinders III Is B. Inserting the Cylinders | ! !; intoTheTutae. C. ForcincjtheOay down ' V'sSBfikfe^. .'.:. .-rL ~- - G.1y_ APPARATUS USED TO FORCE CLAY INTO CREVICE OF FOUNDATION ROCK AND CLOSE LEAK IN COFFERDAM. pipe small cylinders of clay about one foot long were placed and forced down into the cavity by means of a plunger working in the pipe. The apparatus is shown in the illustration (Fig. 17). As will be seen, the plunger, or ram- mer, is an iron rod to the top of which is fastened a block of wood sliding between the guides of an ordinary pile driver. The hammer of the pile driver is the weight which pushes down the rammer. This apparatus was 30 THE COFFER-DAM PROCESS FOR PIERS. designed by E. S. Wheeler, engineer in charge of the work, and was used not only to fill the crevice, but all along the coffer-dam for the purpose of compacting the clay filling. The apparatus proved most successful for the purpose for which it was intended." The use of rods for bracing in double walled coffer-dams is very often the cause of considerable leakage, the w r ater following along them through the puddle. This may be stopped by wrapping a band of hay or straw around the rod next to the timbers, or by a wrapping of coarse cloth, or by a wood washer having a hole slightly smaller than the rod, which is forced through. The walls of the dam must always be made tight, and this we have seen to be effected by careful framing of sides and bracing, and it will be seen in a later example how round struts between the two walls allowed the puddle to flow around them and close up much better than if the braces were square timbers. The use of candle-wicking between the staves proved successful at Fort Madison, and calking is very often resorted to at the first, and also to close up local leaks. The use of this and the use of a stiff grease between the layers of a crib will be referred to in another part of this article. Th.e use of tarpaulins to make a watertight piece of work is described in the Trans. Am. Soc. C. E., Vol. 31, by Montgomery Meigs, engineer in charge of the government work at Keokuk, Iowa. "The upper one of three locks was twice repaired by separating it from the river by an ordinary plank and mud coffer-dam. But as this work had to be done after the close of navigation, it was found to be very unsatisfactory on account of the freezing of the puddle, and on one occasion the partly puddled dam froze and upset. After this experience it was determined to use some other method than puddle to produce tightness. There was available for drainage a 50-H. P. suction dredge, with 14-inch suction, and a rotary Van Wie pump, and plenty of 12-inch discharge pipe mounted on pontoons. It was proposed to drain the lock with this dredge, allowing the boat to settle in the mud at the bottom of the lock as the water left it, and to complete the work with a 3- inch discharge Pulsometer. The lock being 350 feet long and 80 feet wide, a flat place on the bottom was selected, the dredge placed over it and the necessary length of discharge pipe placed in position on its pontoons. The point selected for a bulkhead (Figs. 18 and 19) was just outside the lock gates, about forty feet below the lower mitre sill, where there was a smooth rock bottom, the ends of the dam abutting against the flaring ashlar wing walls of the lock approach. "The bulkhead was constructed with thirteen bents eight feet apart, of the size timber shown, with light diagonal bracing. After being built 2^ miles from the lock it was towed to position and sunk by weighting it with old railroad rails, enough being used to overcome the buoyancy after the sheath- THE COFFER-DAM PROCESS FOR PIERS. 3 I ing was added. A diver was employed to see that the bottom was clear of obstructions and to guide the bulkhead to a solid bearing. The sheathing was also guided to place by his assistance. "The canvas sheet, which was designed to give tightness to the apron, was of two breadths of ten feet and one breadth of six feet wide, sewed together edge to edge for convenience, and about four feet longer than the ^^ ; ^vy^^ ,-, ^^o tvjw ' Section / - ' ^-^ ' " ~ ~ "'""" FIG. 18. DETAILS OF CANVAS AND PLANK BULKHEAD. extreme length of the apron. Some old j^-inch and ^6-inch chain was sewed to one edge continuously to act as a sinker and insure the lower edge of the canvas sheet hugging the bottom tightly. A few stones laid on it would have answered the same purpose, but not so well. The canvas was 12-otince duck. 32 THE COFFER-DAM PROCESS FOR PIERS. "The sheet was spread under water by the diver. It lapped on the bot- tom about twelve inches, covered the face of the apron and extended some inches up the face of the wing walls at the end of the dam. Cleats were nailed on the angle between the apron and the wing walls. These were of Ix4-inch strips, nailed with 2-inch wire nails about twelve inches apart. The upper edge of the canvas was also lightly cleated to the planking in a similar manner. No other nails were driven in the canvas, which was designed to be cut up into tarpaulins eventually. Where the plank touched bottom no beveling was used, but one ragged hole was stopped with the beveled "stop waters" which were made use of. The dam was pumped out in about six hours and the leakage w r as so small that a 3-inch discharge pulsometer kept out the water, and was then run only at intervals. Small leaks were stopped by dumping rotten stable manure in their vicinity." It is interesting to note that the bulkhead stood a pressure of twelve feet of water. Experiments made to determine what pressure 12-ounce duck would stand, show that the clean canvas begins to leak at two pounds pressure, and at five pounds pressure the leakage becomes a marked amount. With mud on the canvass the leakage becomes noticeable at from five to seven pounds, and of a considerable amount at fifty pounds pressure, these pressures being on a circle 4^ inches in diameter. The canvas did not rupture at 800 pounds. The suggestion is made to use an inverted funnel of canvas to stop the leakage of springs on rock bottom. (Fig. 20.) The canvas to be spread out over the bottom and weighted down with concrete, and the top wired to a pipe into which the water may rise until the pressure head is overcome or the pipe can be plugged. Arrangements of this nature, but without the canvas funnel, have been frequently used. An iron pipe set on end is fitted over the leak, and after concreting around to make it watertight, the water rises inside until the pressure is balanced. A watertight wooden box may also be used for the same purpose. The founding of a new inlet tower in the Mississippi at the St. Louis water works was accomplished by using a coffer-dam and it was the inten- tion to form a junction with the bottom by using a canvas curtain. When the coffer-dam was floated into position and the divers were sent down to spread the canvas and weight it down with stones, it was found to be dam- aged so as to be useless. This was supposed to be due to the action of the swift current, but was most probably due to some accident such as fouling on a snag or against a barge. The anchoring of the crib for this dam is related in the Engineering News of July 4, 1891. The dam was to be located near the head of a stone dike about twenty feet in height and on solid rock bottom which was uneven and worn into grooves by the action of the current, which had a velocity of THE COFFER-DAM PROCESS FOR PIERS. 33 between six and eight miles per hour. The bottom was leveled off by blast- ing, to receive the crib, which was to be sunk in from fifteen to eighteen feet of water. The three triangular cribs shown (Fig. 21) were sunk and filled with stone and were used to hold the dam in place while building and while being sunk. Steel cables 1% inches in diameter were used as anchors. The large crib also served as a protection from the current and drift. The size of the crib was 38x74 feet outside and the height 22 feet. The 12xl2-inch yellow pine timbers were drift-bolted together with from one to two feet spacing of bolts, and all the joints between the timbers were calked. The bracing consisted of 12-inch square timbers, of which there were three rows, the braces in each row being four feet apart vertically. These were cut out as the masonry was built up and bracing against the stone work sub- stituted. There were four sets of diagonal bracing as shown. The space between the walls, which was three feet, was partly filled with concrete in sacks, and puddle placed on top. Sacks of clay were banked up around the outside, and then the dam w r as pumped dry with a 10-inch pump. Inside was found eight feet of mud and sixty sacks of concrete which had been washed there by the swift current. The amount of timber used was 125,000 feet, B. M., and about 12,000 feet of ^s-inch round iron for drift bolts. The puddle chamber required 1,000 sacks of concrete and 1,00 barge loads of clay, while 10,000 sacks were used for banking up clay on the outside. This work was constructed under the direction of C. V. Mersereau, Division Engineer, under S. B. Russell, Principal Assistant Engineer. The Queen's bridge at Melbourne, Australia, is a plate girder structure, with four piers of eight cylinders each. The bottom was a reef of bluestone which had been shattered by blasting and which was silted over with about three feet of very soft silt. The use of ordinary puddle coffer-dams was thought to be too expensive as the bridge was 100 feet in width, and it was proposed to use a single wall of timber protected by tarpaulins. The account of this work is taken from the Engineering News of April 4, 1895, which is an abstract of a paper by W. R. Renwick, engineer in charge. To insure as light a construction as possible experiments were made on the strength of Oregon pine, and it was found that tests of water soaked tim- ber showed a loss of strength of as much as 33 per cent., when compared / with tests of seasoned timber. The break, too, of the water- soaked pieces was very short. This strength being the one adopted, a very low factor of safety was used. A separate dam was constructed around each tube, but with one side to open as a door to allow its removal and use for another THE COFFER-DAM PROCESS FOR PIERS. 35 place. The frame was made from 12x12 Oregon pine, with the sticks placed closer together near the bottom to resist the greater water pressure, and 12x12 pieces were run up the corners, the frames being notched in. These also served as spacers for the side timbers and as door frames. The sheeting on the outside was of 4x12 rough timber, and outside of this at the top and bottom were wale pieces, 6x12, bolted through the frames with 1-inch bolts to hold the sheeting in place. The tarpaulin was passed completely around the dam, being tacked to the waling pieces, and so arranged as to allow the door to open. When the dam had been placed around a tube the sheeting was driven down to rock, through puddle which had been dumped on the bottom, and FIG. 20. CANVAS FUNNEL FOR CLOSING LEAKS. the pumping was readily done with pulsometer pumps. The only serious leaking was where the 1-inch bolts passed through the joints between the sheeting, but these were plugged with soft wood plugs, and in other work the bolts were flattened to three-eighths of an inch where they passed between the plank. The dams were removed by first drawing the sheeting up to its original position, when the door was opened and the crib taken to another tube. The depth of water was about fifteen feet, but while this was successful in this instance, the method should not be copied unless the condi- tions are favorable, nor unless the cribs are made practically watertight in themselves. This was the case in the above work, as one of the tarpaulins was acci- dentally torn off and the dam still excluded the water, so that the tarpaulin was only a wise precaution. Why the cylinders were not made watertight 36 THE COFFER-DAM PROCESS FOR PIERS. and used as their own coffer-dam is not stated, but this possibly could have been done. The use of tarpaulin in closing accidental leaks could doubtless be made use of frequently, but as the sole dependence for producing tightness it should be used with extreme care, in a gentle current and well protected from damage. The pivot pier of the Harlem Ship Canal bridge was founded in a polygonal coffer-dam, from the plans of William H. Burr, consulting engi- neer. The work is described in the Engineering Record of July 24, 1897: "The rock bottom secured by the canal excavation being an acceptable sur- face for the masonry of the pivot pier it was constructed in a polygonal FIG. 21. CRIBS FOR ANCHORING ST. LOUIS COFFER-DAM. double- walled coffer-dam with thirteen sides twenty-five feet high and sixty feet in extreme diameter. The great dimensions of the coffer-dam would have made it difficult to build and launch it on shore. Consequently it was built partly on a detachable raft. As shown in the illustration (Fig. 23) the inside wall was built up of timbers lapped and halved at the angles; the outer wall timbers were carefully butt-jointed and secured by cross-struts and 1-inch bolts to the inside walls. The rough-sawed horizontal surfaces of the inner wall were bedded in stiff grease and the joints calked, which notably resisted the penetration of the water. Each course of timber was secured to the one below it by 24 -inch drift bolts spaced about four feet apart. When the bottom was thoroughly cleaned the concrete was dumped in place by a special steel bucket. Concreting was carried on night and day :J n A r o ZQ Hd. Bottom to H.W. /2\IZ'-+*-6-/0'*U+*-J-8"x/Z-+ Drift bolt,-, j'x/8" about 3ft.a.pa.rt. THE COFFER-DAM PROCESS FOR PIERS. and was completed before puddling was begun. Considerable difficulty was occasioned by the irregularities of the bottom which the coffer-dam could not be made to fit closely. Divers were sent down and filled in bags of sand, as at S, and riprap R was piled up outside to protect it. Then the space between the walls was filled with puddle." Another polygonal dam was constructed for the draw pier of the Arthur Kill bridge, by Alfred P. Boiler, consulting engineer. The following account is taken from Vol. 27 of the Transactions Am. Soc. C. E.: "It was Halt '(Section Through Pier and Co/ftrc/Q/n ~K HALF -" C ^ HALF v ELEVATION. SECTION. FIG. 22. POLYGONAL COFFER-DAM, HARLEM SHIP-CANAL DRAW-BRIDGE. PLAN. FIG. 24. COFFER-DAM FOR PIVOT PIER OF THE COTE A U BRIDGE. necessary to use as little space as possible for the dam, and to construct it without interior bracing, so that a double-walled twelve-sided polygon (Fig. 22) with walls four feet apart in the clear was used. The rock bottom was overlaid with two feet of clay and the clay with eighteen inches of sand and mud, the depth of water over the rock being twenty-eight feet at high tide. The square hemlock timbers used in the walls were halved together and the walls braced together by bolts and round timbers for struts, the round tim- bers allowing the puddle to run around them and pack well as thrown in. THE COFFER-DAM PROCESS FOR PIERS. 39 Clamp timbers 4x6, in two lengths, were held in place by the bolts and the struts were braced against 6-inch plank. The dam was built to one-third its height on shore, then towed to position and built up until grounded. Between the timbers and the joints candle-wicking was placed, and the courses drift bolted together every three feet and spiked at the joints. The rock was dredged bare before placing the crib, which was filled with a hard, gravelly clay between the walls after being sunk in place. A rich Portland concrete was dumped inside, from triangular buckets, to seal the bottom and then the dam was pumped out with a 6-inch pump and kept dry by pumping at intervals. In one place the concrete was not thick enough and a spring came up through a fissure in the rock. This was boxed in and led to the sump. The material used was 140,000 feet of timber, 15,000 pounds of iron, and 600 yards of puddle." A piece of work similar to the Canadian Pacific example was an octag- onal single- walled dam used in the construction of the Coteau bridge on the Canada Atlantic Railway. This is illustrated in the Engineering 'Nc-cs of May 30, 1891 (Fig. 24), and was braced thoroughly with cross-timbers built into the sides. The bottom being of rock it was partly filled with concrete to make it watertight. The different forms of sheet piling will next be taken up, together with pile driving machinery and the methods of driving both sheet and guide piles. After this will be described the use of sheet piles for forming water- tight coffer-dams, by reference to actual constructions of that character. ARTICLE IV. THE COFFER-DAM PROCESS FOR PIERS.* PILE DRIVING AND SHEET PILES. N no department of engineering have ancient methods been more rigidly adhered to than in that of pile driving. The form of the pile-driver derrick has remained so characteristic that a person but slightly familiar with the subject would have little difficulty in recognizing the pile driver in the picture of Caesar's Bridge (Fig. 3) in the first article. The bridge of the Emperor Trajan over the River Danube is an instance of the early use of piles. This bridge was constructed in the first century, and when the piles under water were examined in the eighteenth century they were found in some cases to have become petrified to a depth of three-fourths of an inch from the surface, beyond which the timber was in its original state. Before derricks were used it is probable that piles were driven by a large maul of hard wood, which is termed by Cresy a "three-handed beetle." The block of hard wood was hooped with iron and had two handles radiating from its center, to be worked by two men, while a third man assisted in lift- ing it by means of a short handle opposite. Wooden mauls are still used where sheet piling is to be driven into a soft bottom, and heavy iron mauls or sledges are also used; but as has been fre- quently stated such a soft bottom should be dredged and some more elab- orate apparatus used to drive the piles into a harder substratum. The most primitive form of the pile-driving derrick is similar to the one used in 1751 by the celebrated French engineer, Perronet, at the brdige of Orleans (Fig. 25). This was arranged with a number of small ropes splayed out from the end of the lead line, so that a number of men could pull down at one time, the drop of the hammer, of course, being limited by the reach of the men's arms. The windlass shown was for the purpose of raising the pile into place between the leads. The same engineer improved upon this derrick by adding a large bull- wheel to the windlass, on which was wound a rope to be pulled by a horse from the side, as shown in Fig. 26, thus winding up the lead line on the * The subject of pile-driving has been restricted to the ordinary methods and operations ; such unusual processes as gunpowder pile driving and the like have not been .referred to. Pile-driving, with the assistance of the water- jet, has been described on page 70, in the account of the Sandy Lake coffer-dam. The ordinary operations of pile-driving, as practiced on that work, are also described in some detail. 40 THE COFFER-DAM PROCESS FOR PIERS. windlass. This same apparatus is in use down to the present time, except that one seen recently had the windlass at right angles to the one illustrated. The ram or Hammer used in olden times was of oak, bound with iron, and weighed for the work at Orleans 1,200 pounds for the main piles which were nine to twelve inches in diameter and which were driven three to four feet apart, center to center, to a depth of six feet into the bed of the river; the ram for the sheet piles only weighed half as much, the sheet piles being about twelve inches wide by four inches thick. At the bridge of Saurnur, which was built about the year 1756, De FIG. 2o. PERRONET'S PILE DRIVER. FIG. 26. PERRONET'S BULL WHEEL PILE DRIVER. Cessart employed a driver with a bull-wheel, in the periphery of which were set pins, to form handles for the men to pull upon and rotate the wheel. Eight men, by making three turns of the wheel, raised the ram weighing 1,500 pounds six feet, when it was unhooked and allowed to drop. The piles cost from two to five dollars each in place. A very simple form of pile driver is shown in Fig. 27 and was described in the Engineering News of March 16, 1893, by Julian A. Hall. The hammer is hewed out of a section of a hardwood log, and has pieces bolted on the sides to hold it in the leads, which should give plenty of clearance. The derrick was constructed of very light timber, the ver- ticals being 4-inch sawed stuff and the bottom timbers 6x6 inches. The rope passes over the sheave A and down over the tops of the steps B B, on which the men stand to pull the line and thus operate the hammer. This was a very inexpensive apparatus and was found to work well. Where there is already in use a heavier hammer of cast iron it can be used by striking light FIG. 27. SHEET PILE DRIVER. THE COFFER-DAM PROCESS FOR PIERS. blows. The construction of the ordinary pile driver derrick is a simple piece of framing, when good straight timber is easily obtained, the essential features being to keep the leads free from any obstruction for the hammer and to have efficient bracing. For bracing a derrick under twenty-five feet a straight-back brace or FIG. 28. PILE DRIVER DERRICK FOR USE ON A SCOW. ladder having two horizontals running to the leads, and two side-braces will be sufficient. But for a higher one, either additional long braces should be used or diagonals introduced between the leads and the ladder. The use of long braces is shown in Fig. 28, which is the design of pile-driver such as is THE COFFER-DAM PROCESS FOR PIERS. 43 FIG. 29. LIDGER- \VOOD PILE DRIV- ING DERRICK. used about harbors or rivers on heavy work. It would be mounted on a scow or flat-boat sixty feet in length, twenty-five feet in width and of about six feet in depth. The design of smaller derricks can be approximated from this one, the bracing being used in proportion. It will be noticed that the guides for the hammer are 4x4 inches lined with a steel plate. Two lines are provided, one being for the operation of the hammer and the other for pulling piles into place. Especial attention is called to the hooks at A, as these are seldom shown in the plan of a derrick and they are of constant use for clamping and guiding piles. A timber laid across is wedged tight against the pile to draw it to line, and can be used to correct a stick which is beginning to slant badly. Similar clamps of course are used on the oppo- site side of the leads. Where a pile begins to sliver or split in driving, if the sliver is spiked down and the clamps used to hold it in place, the trouble can usually be corrected before the pile is badly damaged. The use of diagonal bracing between the leads and ladder is shown in the Lidgerwood derrick (Fig. 29) in which a diagonal is introduced between each pair of horizontals. This form of bracing is very satisfactory and equally as good as the other method. The diagonals on a very large driver may be extended over two panels and planks spiked down to the horizontals to form a platform for the workmen. In smaller derricks the diagonal bracing is most always omitted, dependence being placed in the stiffness of the leads and the bracing from the ladder and horizontals, as was done in the derrick shown in Fig. 4. The power for driving with a small hammer weighing from 500 to 1,500 pounds, may be furnished by laborers pulling, but this is a slow operation and horse power is nearly always used where steam is not available. The power is furnished from a drum with a long lever, to which the horse is hitched and winds up the hammer by walking in a circle about the drum, the frame of which is firmly fastened in place. This is called a "horsepower" apparatus and works slowly, but is a cheap and satisfactor} 7 way where a very few piles are to be driven. To the hammer line are attached the tongs or nippers, which engage the pin in the top of the hammer (Fig. 30), and when the hammer has reached the proper height it is dropped by pulling a tripping rope and releasing the tongs, or if the hammer is hoisted to the top of the leads, the top arms of the tongs are FIG. 30. HAM- MER WITH NIPPERS. 44 THE COFFER-DAM PROCESS FOR PIERS. pushed together by the wedges on the leads and the hammer released auto- matically. This is a slow method on account of waiting until the tongs run down again and engage the hammer. The horse power, of course, has a ratchet, so that the rope runs down free and usually the blows are hurried by overhauling the line. With the addition of a hoisting engine all this is changed and pile driving becomes one of the most stirring operations of the contractor. When the hammer is hoisted up, the friction lever is released and the hammer descends carrying the rope with it, as the tongs are done FIG. 31. PILE DRIVING SCOW, NEW YORK STATE CANALS. away with and the line attached directly to the hammer. A good engine man will catch the hammer on the rebound and materially lessen the time between the blows and likewise the cost of driving. With a heavy hammer shorter drops are made, thus causing much less damage to the pile, which would split badly under the high drop from the use of tongs. For the smaller-sized hammers from 1,000 to 1,500 pounds an engine of 10-horse power is mostly used, as it is usually thought best to have a surplus of power in case ol need; while for a 3,000 pound hammer a 20-horse-power engine would likely prove the best and most economical, but not infrequently a 25-horse-power hoist is employed. The cost of an outfit will vary greatly and the only satisfactory way is to get prices from responsible firms, but for preliminary estimates the cost of a 10-horse-power hoist with single cylinder and single drum may be taken at about $900, and for a 20-horse power at $1,270. Preliminary prices for other THE COFFER-DAM PROCESS FOR PIERS. 45 sizes of single cylinder, single drum hoists, may be obtained from the formula: Cost =i/81,000xhorse power. The double cylinder engines will cost about 10 per cent, more and double drums about 10 per cent, additional to this. Pile driver derricks will vary much in cost owing to the location, on account of the cost of timber, but a minimum cost for a first-class derrick will be $6 per vertical foot and a maximum of $8. Being such a simple structure the easiest and safest way will be to make an estimate for each case. In the selection of an engine it is well to remember that with a double drum a second pile may be hoisted into place, while the first one is being driven, as all der- ricks are, or should be, provided with two sheave wheels at the top for this purpose. While a single-drum engine has a spool for this purpose, it cannot be used very satis- factorily. A pile driver on a scow is shown in Fig. 31, such as was used in driving piles on the New York State canals. Another pile is just being hoisted into position. The hoisting engine has no protection, but a shed or house is mostly provided as a protection from the weather. While little change has ever been effected in the de- sign of pile driving derricks, the adoption of steam hoists was a great improvement, as \vas also the invention of the steam pile hammer by James Nasmyth. The principle is the same as that of steam forging hammers, and was applied by Nasmyth to pile driving in 1845, the ham- mers of this class bearing his name to-day. His idea was that the drop-hammer was calculated more for de- struction than for useful effect and he termed it the "artillery or cannon ball principle." Besides this the action of the drop-hammer even with the use of the "monkey" engine was somewhat slow. Samuel Smiles says that "in Xasmyth's new and beautiful machine he applied the elastic force of steam in raising the ram or driving-block, on which, the driv- ing-block being disengaged, its whole weight of three tons descended on the head of the pile, and the process being repeated eighty times in a minute the pile was sent home with a rapidity that was quite marvelous as compared with the old method. In forming coffer-dams for piers and abutments of bridges, quays and harbors, and in piling the foun- FIG. 32. WARRING- TON - NASMYTH STEAM PILE HAMMER. 4 6 THE COFFER-DAM PROCESS FOR PIERS. dations of all kinds of masonry the steam pile driver was found of invalua- ble use by the engineer. At the first experiment made with the machine Mr. Nasmyth drove a 14-inch pile fifteen feet into hard ground at the rate of sixty-five blows per minute. The saving of time effected by this machine was very remarkable, the ratio being as 1 to 1,800; that is, a pile could be driven in four minutes that had before required a day. One of the peculiar features of the invention was that of employing the pile itself as the support of the steam hammer part of the apparatus while it was being driven, so that FIG. 33. WARRINGTON-NASMYTH HAMMER, FAIR HAVEN BRIDGE. the pile had the percussive force of the deadweight of the hammer as well as the lively blows to induce it to sink into the ground. One of the most ingenious contrivances of the pile driver was the use of steam as a buffer in the upper part of the cylinder, which had the effect of a recoil spring and greatly enhanced the effect of the downward blow." Many modifications of this hammer have been manufactured, and one much used at present is the Warrington-Nasmyth hammer, made by the Vulcan Iron Works. This hammer (Fig. 32) is made in three sizes, the weight of the striking parts being 550 pounds for sheet pile work, 3,000 THE COFFER-DAM PROCESS FOR PIERS. 47 FIG. 34. CRAM-NASMYTH STEAM PILE HAMMER. pounds for medium pile work, and 4,800 pounds for use on heavy work. This machine is provided with a positive valve-gear, a short steam passage to avoid the waste of steam, a wide exhaust opening to prevent back pres- sure as the hammer drops, a piston-head forged on the rod, and channel bars on the sides to allow the pile to be driven lower than the leads of the derrick. The hammer is attached to the hoist rope, but this is left slack when the hammer is resting on the head of the pile, steam is turned on and the ham- 4 8 THE COFFER-DAM PROCESS FOR PIERS. mer pounds automatically at the rate of sixty to seventy blows per minute until the pile is driven. The bottom casting which rests on the pile is a bonnet which encases the top and prevents brooming or splitting. The hammer should have plenty of play in the leads, and the steam pipe should extend half way up the derrick to save length of hose. This FIG. 35. MACHINE FOR SAWING OFF PILES UNDER WATER. hammer has a record of as high as seventy-five to one hundred piles per day, and one account gives the record of 3,UOO lineal feet of piling per day at a cost of $50, the number of men employed being sixteen and the coal con- sumption one ton. This hammer is shown in Fig. 33 in use driving piles for bridge work on the Fair Haven bridge. THE COFFER-DAM PROCESS FOR PIERS. 49 Another form of the Nasmyth hammer is the Cram (Fig. 34) which is very simple in construction. The driving head is hollow and the steam enters through a hollow piston rod, causing the head or cylinder to rise on the rod. Four sizes are made, with hammers of 430 pounds, 2,000 pounds, 3,000 pounds and 5,500 pounds. The small hammer which is listed at $300 is used for sheet pile work by inserting a "follower" of oak which fits the base or pile cap, and which has a slit in the lower end to fit the sheet pile. The number of blows per minute is the same as other steam pile hammers and an average of eighty-three piles per day of ten hours is reported, where they were driven seventeen feet into sand and oyster shells in the Passaic river, the largest day's work being 121 piles, or nearly double the best work with an ordinary hammer. Mention has been made of the use of a rock drill as a Nasmyth hammer, on the Great Kanawah river coffer-dams; and where any amount of driving is to be done it will certainly be wise to use a hammer of the Nasmyth type. FIF. 36. PILE-PULING LEVER. AFTER CRESY. The guide piles of a coffer-dam should always be driven with the idea of using them as a support for pumps, engines, derricks, and the like, although it will often be found cheaper to rig up on flat-boats when there is danger from floods. In determining what load a pile will carry from this source, or when driven as a foundation pile to support the masonry, Wel- lington's formula is at once the most accurate and the easiest to remember and use. For a drop-hammer, multiply twice the weight of the hammer in pounds by the drop in feet and divide by the last sinking in inches plus one, and the result is t he load in pounds the pile will carry, with a factor of six for safety. This is easily remembered as 2 wh over s + 1, and is always ready for use. For the steam-hammer the form is 2 wh over s+0.1, the "wh" representing the dynamic effect of the hammer. Where piles have been firmly driven and they are to be removed when the work is done they can be cut off under the water by a machine similar to Fig. 35, which can be operated from a barge. The description in the Engi- neering News gives but little information in addition to the drawing. The shaft works in cast-iron sleeves attached to a timber, which slides in the 50 THE COFFER-DAM PROCESS FOR PIERS. leads, this being operated by the winch shown in side elevation. The final adjustment is made by the hand- wheel on the 3-feet adjusting screw. Where the piles are not so solidly driven they can be pulled out with a lever, an old form of which is given by Cresy (Fig. 36). In place of the pin and links, a chain closely wrapped around the top of the pile is usually made use of. The apparatus used on the New York State canal work (Fig. 37) con- sisted of a strong frame mounted on a scow, from which was suspended a \eavy set of falls to attach to the chain wrapped around the head of the pile, f he pulling was done by an engine placed on the scow. The construction of coffer-dams with sheet piling has led to the use of a FIG. 37. PII,E-PUI,IJNG SCOW, NEW YORK STATE CANALS. number of forms of sheet piles, some of which are driven only as a protection to the puddle, while others are nearly or quite watertight in themselves. The principal forms are shown in Fig. 38, the simplest form being plarik of some considerable thickness (a) for which Stevenson specified 4^ inches by not exceeding 9 inches in width for the Hutcheson bridge. The points are sharpened as at (i) so they will draw together in driving, and as at (j) to cause them to drive straight and easy. The same principle is embodied in the patent metal point shown at (k), which is used to protect the point when driving through coarse gravel. The piles at Buda-Pesth were increased to fifteen inches square in order to resist the pressure brought upon the sides of the dam by the puddle, the water, and also by the ice. Flat plank are also used by driving two or more THE COFFER-DAM PROCESS FOR PIERS. rows as at (b),. the second and third rows being used to close the cracks in the main row of piles and retain the puddle. An example of this will be given in the next article, where it was used on the Michigan Central Railway. The extra rows may be of thinner plank if they can be driven. Mention has already been made, incidentally, of the use of V-shaped tongue and groove piling (c), on the Union Pacific Railway. This may be I . I "A V FIG. 38. SHEET PII.ES AND SHEET PILE DETAILS. made on a beveled saw table, the saw cutting half through the plank from opposite sides at each cut. This will produce a reasonably tight wall, if care is used in driving and if the points are sharpened to draw them together and make tight joints. Ordinary tongue and groove piling (d) is frequently used, but a more frequent form is that shown at (e), like that used on the Robinson circular dam. The two pieces forming the groove and the piece for the tongue are THE COFFER-DAM. PROCESS FOR PIERS. FIG. 39; CHARLESTOWN BRIDGE. DRIVING WAKEFIEI.D SHEET PII.ING. spiked to the 9x12 with 6-inch spikes sloping upward. A sheet pile dam on another pier of the Arthur Kill bridge, employed piling in which the grooves were made by making two saw cuts and cleaning out between with a chisel, the tongue being formed in the same manner as at (f), the tongue being spiked in one side. A method which is not often employed is shown at (f) two grooves being made in the sheet pile and a key driven after the piles are down. Should the piles not drive in perfect line, and the groove fail to match, the method will not be found to be a success. Sheet piling formed of two or more plank bolted together is being exten- tively used, one of them (g) being formed by two planks sawed with beveled THE COFFER-DAM PROCESS FOR PIERS. S3 edges and bolted together to form a pile similar to (c). This forms a pile which will drive easily on account of having some size and which will require fewer supports in the shape of waling pieces. Several examples already given describe the use of Wakefield patent sheet piling (h), the method of sharpening being shown at (h'). This is constructed of three layers of plank from one to four inches thick, owing to the pressure to be sustained. The center plank must be sized to keep the tongue and groove uniform, and the plank are bolted together with six bolts for a length of from sixteen to twenty feet, two bolts near each end and two intermediate. For long piles, spikes should be driven between the bolts. The bolts vary from f inch for 1-inch plank to 3^-inch for 4-inch plank. A coffer-dam constructed with this piling is shown in process of construction in Fig. 39, for the foundations of Charlestown bridge near Boston. A description of this will be given in the next article. Pile shoes for use on round or square piles are shown at (1) and (m), (1) being patent forms. Straps of bar iron are used in many cases with suc- cess, for main piles, and sheet iron of J /i -inch thickness, bent to a "V" and spiked on, is often all that is necessary when shoes must be used on sheet piles. The thickness of sheet piling should be sufficient to prevent the plank from bulging and should be calculated to stand a water pressure due to the depth, and for a span equal to the distance between the waling timbers or other supports. This would necessitate wales every six feet for 3-inch plank under five feet head, or wales every three feet for a 21-feet head. Plank 4y 2 inches thick would require wales every seven feet under a 9-feet head, or every five feet for an 18-feet head. Timbers nine inches thick will carry nine feet under a 20-feet head, while the 15-inch timbers of the Buda- Pesth dam would carry twelve feet under a 33-feet head. Good timber should always be employed if it can be procured, or if faulty stuff must be used, allowance must be made by using thicker piles and by placing the wales closer together. It I**LT? ARTICLE V. THE COFFER-DAM PROCESS FOR PIERS.* CONSTRUCTION WITH SHEET PILES. ATER pressure against the sides of a sheet pile coffer-dam is seldom provided for in an accurate manner, the thickness of the piling being usually decided upon from past experience, as is also the size and spacing of the guide piles and wales. &-- , These are points where guess-work should be eliminated, as otherwise good coffer-dams are often seen, where the pressure has so bulged the plank as to cause leakage. While this may perhaps be corrected by additional bracing, simple calculations may easily be made to determine the size beforehand. The pressure against a coffer-dam may act as at (a), Fig. 40, the sheet piling being in the condition of a beam fixed at one end and loaded with a gradually increasing weight, as shown by the dotted lines, due to the pressure of water or puddle at 62 .4 pounds per cubic foot. Then the load on a width w of the wall is 124. 8 w d 2 and the moment of the pressure is 83. 2 zv d* . Taking the allowable unit stress on wet timber at 1,000 pounds per square inch, the thickness / of the sheet piling may be obtained from the formula in which d is to be taken in feet and the resulting value of / will be the thickness in inches of the sheet piling. This formula has been expressed in a graphic manner in diagram (d), Fig. 40, from which, knowing the depth of water 2d, the thickness of piling may be read directly without calculation. The addition of a strut, as at (b), Fig: 40, places the sheet piling in the condition of L beam supported at the upper end and fixed at the lower end, but for practical reasons, it is best to consider it as merely supported at both ends. The load will be the same as in the former case, 124.8 w d 2 , but the * The assumption that the pressure of puddle will be the same as water-pressure is made advisedly. It is true that very wet clay, approaching a fluid condition, will exert a much greater pressure, but it would then be useless as puddle. Dry clay would exert a pressure of less than half that due to water, so it has been assumed that wet clay or puddle would exert the same force as water. Should it exceed it for a short time no damage would be done, owing to the low unit-stress adopted. 54 55 /0 4o 3 4 5 *6 ?. 5 9 /o 77i/c /r/\ e^5 Piling m lr\ct\ e5 . &ZE PlL //V <5 . blA GRA A] (a ). f. *>. P/Z. IHG.blA GRA MS (btc). W $35 30 60 In Q O O 5 3*t56?$9/0 ^ r v ) GUIDKS WITH SEPARATORS. (c) SHEET-PILE CLAMP FIG. 41 SHEET PILE GUIDES AND CLAMPS. spacing blocks as they are reached, and substituting bolts through the sheet piles, firmly connecting the piles and wales together, A very satisfactory method is described in the Engineering Neiu s of May 12, 1892, which was used by A. F. Walker. Having occasion to do a large amount of work it was desirable not to go to the expense of squared guide piles. Round guide piles (P) were first driven seven feet apart, and cut off to a level. Caps were then drift-bolted to the tops, allowing them to project slightly beyond the face of the round piles, thus forming a perma- nent support for the top of the sheet piles. Near the ground line was placed the clamp, consisting of two sticks (X) and (Y), connected by three bolts and drawn together as tight as the intervening piles or pile and gauge block (G) will permit. The stick (Z) is then forced forward by the wedges (W) until THE COFFER-DAM PROCESS FOR PIERS. 59 the space between (Z) and (Y) is the same as the thickness of the piles. The pieces (X) (Y) (Z) are slotted for the middle bolt, and this permits of some adjustment. When one of the piles partially closes this slot, a notch is cut in the same large enough to receive the bolt, and the bolt is then slipped up to it and tightened. This allows of the next pile being driven as close as the others. When one panel has been completed the nuts are removed and the clamps moved forward to the next one, a notch being cut in the end pile to receive the end bolt of the clamp. The piles are sharp- ened flatwise with a little more slope on the side facing the guide piles, giving them a tendency to drive away from the guide pile at the foot and bear against the cap at the top. A slight bevel is also given to the edge to make the foot crowd the adjoining pile. During the first half of the driving, the joint is held a little open at the top, but during the latter half, pressure FIG. 43 SEWER COFFER-DAM. BOSTON SEWERAGE SYSTEM. is brought to crowd it toward its neighbor, and the joint will close as tightly as possible. The use of single pieces of timber as wales, against which the sheet piling is driven, is illustrated in the use of method (b) of Fig. 38, by Benj. Doug- las, Bridge Engineer of the Michigan Central Railway. The coffer-dam (Fig. 42) was built without guide piles, the wales being 12x1 2-inch timber bolted against the outside of the sheet piling, by the brace rods one inch in diameter. The wales are held in place vertically by bracing of 2x1 2-inch pine plank, which are spiked on as verticals and diagonals to form a truss and also to stiffen the framework in general. The sheet piling is 5x12, and after being driven into the hard gravel bottom, the cracks were lapped by I -inch boards The bottom was uneven 60 THE COFFER-DAM PROCESS FOR PIERS. and accounts for the difference in height, the excavation at the high end being dumped outside at the low end, to assist in making the dam tight. The puddle chamber was 2 feet 8 inches wide and was filled with clayey gravel. The plan also shows the grillage in place for receiving the founda- tion courses of the stonework. This is formed by 12x12 timber crossed, and drift-bolted together with 1-inch round and 18-inch long drift bolts. The account of the Arthur Kill bridge foundation in Vol. 27 of the ''Transactions of the American Society of Civil Engineers," by A. P. Boiler, Consulting Engineer, covers a very interesting experience with sheet piling on Pier No. 5: "This pier is near the edge of the marsh forming the Staten Island shore, which is barely flooded at extreme high tides. Borings indicated about thirty feet from the surface to hard bottom, consisting of mud, mud and clay, clay and shale to the bottom of shaley clay, in which the pier was to be founded. Experience on other work of a similar char- acter, indicated that the founding of this pier would be accomplished with little difficulty. The area of the foundations was inclosed with a tongued and grooved sheet pile dam of 4-inch yellow pine plank. But it was found impossible to hold the plank at a depth of fifteen feet, the mud and clay becoming puddled with water, and despite all efforts at bracing, the plank shoved inward to such an extent, as to spoil the whole dam before we were half way down. A second dam was therefore driven around the first one, but this time with 10xl2-inch tongued and grooved timbers, in one length to reach the extreme bottom. These timbers were grooved by slitting the grooves out at the mill with a circular saw and chiseling the blank so formed free. The tongue was an independent spline, 2^x4 inches, of dry wood and nailed in one groove. The timbers were shaped at the feet to drive close. This dam was hard driving, but was finally accomplished, when digging was resumed and the old dam removed piecemeal as we could get in the braces. The bottom was reached within a perfect dam, with only one bad leak in the northwest corner, due to the shattering of a small piece of one tongue during the driving. As it was impossible to stop this leak from the inside, and the outside was inaccessible, to prevent washing the concrete, the leak was led off in a box at the side of the dam to the sump well, and the footing course of concrete, filling the whole area of the dam about seven feet deep, was gotten in place." This example emphasizes in a very decided manner many of the state- ments that have been made heretofore. While no doubt the removal of the old dam was attended with much expense, its inclosure entirely within the new sheet piling rendered the prosecution of the work comparatively certain. An example of the driving of sheet piling on a slant, to prevent crowd- ing in at the bottom is shown in Fig. 43, which is a cross-section of a sewer coffer-dam used on the Metropolitan Sewerage Systems of Massachusetts by THE COFFER-DAM PROCESS FOR PIERS. 61 Howard A. Carson, chief engineer, and described in the Engineering News of Feb. 8, 1894. The outlet into the ocean at Deer Island begins at a point about sixty feet inside the high water line and about 1,850 lineal feet is from five to ten feet below high water. This necessitated the coffer-dam, which was con- structed with bents every six feet and with 2-inch plank inside the high water line, but for the remaining distance of 4-inch matched plank. The excavation was done by means of buckets, traveling derricks and dump cars, the latter being emptied at the sides and ends of the trench. The leakage from the ocean was kept out by using centrifugal pumps, which pumped a maxi- mum of 46,000 gallons per hour. The concrete, which has large boulders imbedded in its surface the size of paving stones, was car- ried up to the level of the ocean bottom. From the middle of June, 1893, when the work was begun, to the end of September, 526 feet of trench was completed. The size of the trench was 14 feet average depth and 10. 8 feet aver- age width, which made the ex- cavation average 5.6 yards per lineal foot. The cost for the trench, including coffer-dam, sheeting left in and back filling was $44.00 per lineal loot. Casual mention has been made in several places of the use of Wakefield sheet piling which was illustrated at h and h' of Fig. 38 and which is further shown in Fig. 44. View No. 1 is of a corner which is formed as in the plan 44 _ WAKEFIELD SHEET PIWNG . 62 THE COFFER-DAM PROCESS FOR PIERS. No. 2, a tongue being bolted on the side of a pile, when the corner is reached as in No. 3. Any angle is turned by a similar method, which is shown by No. 4, or the piles may be driven to form a curve. The essential features of the system ?re the triple lap or long tongue and groove which excludes the water, and the use of ordinary plank, which can be easily obtained. The center planks should be sized to a uniform thickness, to insure the tongues fitting the grooves, and to make driving easy, while the three plank are to be bolted and spiked together to cause them to act as a compound beam and not as separate plank like the system of (b) Fig. 38. It is recommended to use a 2^ -inch tongue on 1-inch boards and ^-inch bolts. For 1^-inch plank a 3-inch tongue, for 2-inch and 2^-inch plank a 3%- inch tongue and ^-inch bolts, while for 3-inch plank a 3^-inch tongue and ^i-inch bolts are to be used, and the same size bolts for 4-inch plank, but a 4-inch tongue. Two bolts are to be staggered in every five to eight feet of the length of the pile and spikes used between the bolts on long piles. FIG. 45 TYPE OF MOMENCK AND HARPER'S FERRY COFFER-DAMS. The La Grange lock on the Illinois river was inclosed with this piling, under the direction of Major W. L. Marshall, Corps of Engineers. It was intended to back the sheeting with earth, but as both dredges broke down the water tightness was entirely dependent on the Wakefield piling, and under a 7-feet head no leaks were developed. The piles were made of three plank 3x12 inches by 22 feet long and with a 3-inch tongue; they were driven by three pile drivers with hammers of from 2,800 to 3,000 pounds through sand and mud, and in one place a layer of shells. There was no difficulty experienced in driving the piles without special appliances. The. use of 1-inch boards in this form (Fig. 45) is described by H. F Baldwin, chief engineer of the C. & E. I. Railway: "In constructing our second track over the Kankakee river at Momence, 111., it was necessary to extend the piers in that river. The bottom is limestone and the surface is very irregular. We tried several days and finally succeeded in constructing a coffer-dam with two parallel walls of sheet piling. We then tried the THE COFFER-DAM PROCESS FOR PIERS. 63 Wakefield triple lap piling, constructed of 1-inc'i boards sharpened to an edge, 2 ^-tongue and groove, which were driven with sledges until the piles, which were soft pine, conformed to the uneven surface of the rock. This piling was driven around cribs loaded with stone, and after the piling was driven, gravel was put outside the coffer-dam, after which no trouble was experienced in pumping out the water." The work on the foundations of the new B. & O. R. R. bridge, over the Potomac river at Harper's Ferry was similar in many respects to the above, and the system was found to be very satisfactory. Reference was made to the use of this piling on the Charlestown bridge FIG. 46 COFFER-DAM ON CHARLESTOWX BRIDGE. at Boston and the driving of the piles shown in Fig. 39. The work was under the charge of Jno. E. Cheney, Consulting Engineer, and was success- fully carried out. The piling were driven principally as forms for concrete foundations and but little care was taken to make the dams watertight. After the concrete was deposited they were used as coffer-dams against a 6 or7-feet head of water. They were 18 feet 6 inches by 119 feet (Fig. 46) and in some cases were thirty feet below low water or forty feet below mean high water. The piling was made of 2-inch plank and driven with an ordi- nary pile driver. The pumping was done with a 20-inch centrifugal pump OF TH* UNIVERSITY 64 THE COFFER-DAM PROCESS FOR PIERS. and in some cases a 12-inch Follansbee pump of the propeller type was used. The construction of the sewerage system at Fort Monroe, Va., under Capt. Thos. L. Casey, Corps of Engineers, is described in the report of the Chief of Engineers of 1896. The work was done on the general plans of Rudolph Hering, Consulting Sanitary Engineer. One of the special diffi- culties encountered "was the building of a sewage tank fifty feet in diameter, with walls of brick two feet in thickness, exteriorly diminishing to three feet at the center, the inferior reference of which was twenty feet below low water. As described in the report referred to, this was accomplished very successfully by excavating a large area to the reference of ground water, some five or six feet below the surface, and then driving by the pile driver and water jet combined, two concentric twelve-sided polygons of Wakefield sheet piling 28 feet in length, 30 and 22 feet from the center, about the circumfer- ence of the shallow excavation. (Fig. 47.) The material, consisting of fine water-soaked sand, with a small admixture of clayey matter and fine gravel, was then excavated between the polygons to a reference of 20 feet, trans- verse shoring braces bearing upon stout stringers being put in at intervals as the work proceeded. The material did not vary much in its general nature, but a number of old piles were taken up, some of which did consid- erable injury to the sheet piling when driven, as shown in the subsequent excavation. The water was controlled by a powerful steam pump having its point of suction fixed, the water being permitted to flow toward it through- out the circumference. It was noticed that ground water came through the sheeting very freely at first, but that it constantly ceased to flow to any great extent at a height of a few feet above the point of excavation as this continually descended, owing to the rapid drainage of the strata. The interior core, in fact, t ecame quite dry, so that in excavating after the walls were laid, no water was encountered until the bottom of the external concrete ring had been virtually laid bare. Upon attaining the reference 20 feet, the excavation ceased and hand-mixed concrete was deposited directly upon the bottom . as this was considered to be sufficiently firm, the pump being stopped temporarily in order to prevent a flow. The concrete was rammed firmly against the outer sheeting externally and against plank forms with triangular cross-section resting against the inner sheeting internally, until six feet in depth had been put in place. The portion of the ring at the pump suction was filled rapidly with concrete in bags. The 2-feet brick wall was then carried up from the axia) line of the concrete ring, the space between the wall and the outer sheeting filled with sand, except about six inches at the base of the wall, which was of concrete. The braces were removed as successively attained, the inner prism of dry sand being held securely by the sheeting and the extreme top struts, which were left in place until the inner core was completely excavated. On the completion of the latter work to reference 20 feet, the water which came in freely from with- THE COFFER-DAM PROCESS FOR PIERS. 6$ out under the concrete ring at several points was conducted in a peripheral trench to the fixed point of pumping. No water came upward and the middle portions of the bottom became perfectly dry. The inner .sheeting was cut off at the base of the ring, boards were placed transversely over the peripheral trench, a duck tarpaulin coated with hot asphalt laid down, and concrete rammed in place until the concave bottom with sump channel had been completed, leaving only the pipe, through which the ground water had been pumped continually, night and day at about 1,000 gallons per minute, penetrating the concrete. In order to fill this pipe, it was cut off above the level of permanent ground water, and after the water FIG. 47. RESERVOIR COFFER-DAM. FORT MONROE, VA. within had attained the level of ground water in the surrounding area and had become perfectly quiescent, neat cement in paper bags was dropped within, being retained at the bottom by the closed valve; the bags were readily broken up by a long pole thrust down the pipe. The latter was then cut off at the level of the bottom and a coating of cement plaster applied throughout. The resultant leakage through the bottom did not exceed about a gallon a minute and this will be greatly reduced by the infiltration of sand from beneath." Further illustrations of the use of sheet pile coffer-dams will be given; then the operations of dredging, pumping and concreting described at some length. ARTICLE VI. THE COFFER-DAM PROCESS FOR PIERS. CONSTRUCTION WITH SHEET PILES. VARIOUS combinations of the sheet piling shown in Fig. 38 may be made, when occasion demands, or modifications may be made that will perhaps render the available material more effective. For example, the form (g) may be modified to the form shown in Fig. 48, which has the advantage of a wider lap, and should the piles not draw tight together in driving, no crack will be left open to admit the water. Then the piles of this form will act as guides to the ones being driven, similar to the ordinary tongue and groove piling. Other combinations and arrangements will readily suggest themselves as necessity may demand. The use of sheet piling is often accompanied by a great deal of trouble in producing tightness, and as a matter of precaution, the very best method possible should be adopted in making the piling. The coffer-dams constructed at Chattanooga for the Walnut street bridge over the Tennessee River, under Edwin Thacher, Consulting Engineer, were described in the Engineering A r ews of May 16, 1891. Four piers were founded by this method, but the account of pier number two will fully illustrate the work. The bed rock which was level, was covered by cemented sand, gravel and boulders, of which 320 yards were removed. The coffer-dam was built eighteen feet high, or eight feet above low water, to provide for a future rise. The inside was made large enough to allow of a space of four feet all around the base of the pier, and the space between the sheet piles for a puddle chamber was made nine feet. This was filled to an average of twelve feet with a clay puddle, of which there was 900 yards used. As a protection, there was placed outside the dam about 450 yards of puddle, and a breakwater was built up stream. About 38,000 feet of timber was used in the dam and breakwater. After the dam was completed a rise of thirty feet washed out about half the puddle, and one end was crushed by a raft, but the repairs were made v\'ithout serious trouble. No extra amount of pumping was required on any of this work except pier number three, where the seams in the bed rock re- quired pumps with a capacity of 5,000 gallons per minute, and these did not suffice to keep the water down, until the seams were closed by laying sacks of concrete over them and weighting them down with large stones. The location of these seams is shown in Fig. 49. The framework and wales for a sheet pile coffer-dam, used in founding 66 THE COFFER-DAM PROCESS FOR PIERS. 67 the pier for the Baltimore street bridge at Cumberland, Md., are shown in Fig. 50, and this was described in the Engineering News of July 21, 1892, by H. P. Le Fevre, engineer in charge. The frame was built in place on two canal-boats and after completion was suspended from the old Bollman truss which the new bridge replaced. The depth of the water was four feet, and about six feet of very loose quicksand and small round pebbles overlaid the hard bottom. After the boats were removed, the frame was lowered to its place, the sheet piling driven and the dam pumped out with a six-inch pump. The foundation was laid on the hard bottom under the quicksand, after this had been removed. The grillage was made of two courses of 15x1 5-inch clear white oak, around which was built a framework, and the open spaces of the grillage were then filled with a concrete, made up of one part of Cedar Cliff cement to two parts of sand and four parts of hydraulic limestone, broken to pass V \ II I, \ !! i \ n ' n I FIG. 48 COMPOUND SHEET PILE. through a two-inch ring. Upon this was laid the footing courses of the masonry. Another ordinary sheet pile coffer-dam which gave good satisfaction, was used at the Sandy Lake dam on the Mississippi River, by Major W. A. Jones, corps of engineers, and as the account contains so much of value, it will be quoted in full from the 1894 report of the Chief of Engineers. "The coffer-dam is composed of two rows of round piles, twelve feet from center to center of piles, with the exception of sixty-two feet of the east end of the upper part, where they were driven fourteen feet from center. The piles in each row are eight and one-half feet from center to center, cut off at an elevation of 1217 feet above sea level and capped with 12x12 inch timber. The inside row of sheeting is 4x12 inch, and the outside 6x12 inch plank. The sheeting is cut off at an elevation of 1218 feet above sea level, or two feet below the flowage line. One-inch rods of round iron, eight and one-half feet apart, pass through the caps to prevent the filling from spreading the two lines of sheeting at the top. In May, 1892, when a flood occurred, the outside of the cofferdam was raised three feet by splicing three-inch planks to the outside row of sheet- 68 THE COFFER-DAM PROCESS FOR PIERS. ing and then filling the triangular prism thus formed with earth. The cross section of Fig. 51 gives an idea of the dam above the bottom, while the longitudinal section shows the framing down to where it rests on the bot- tom, the frames being joined by the one-inch lateral rods of iron. The total length of the coffer-dam is 829 feet, of which 742 feet is like that shown in cross section and the other 87 feet like that shown in the longitudinal section. The number of round piles driven in the foundation is 1.605. The driv- ing was commenced on November 12,1891, and completed on August 2 1 , 1 893 . The material in the foundation is sand, excepting in the lower right hand corner, where there is some blue clay overlying the sand. The sand in the foundation is not as compact as it is usually found in the bed of streams. In the south half of the dam, the surface settled from four to six EI.O EI.-S 1 I 1 \ Bed of River. ivest. fast FIG. 49 CHATTANOOGA BRIDGE BED ROCK PIER NO. 3. inches during the driving. As the surface was settling, the driving became harder all the time. In the north half, which embraces the navigable pass, there was some settlement, but it was not as noticeable as in the south half. The surface had probably settled by the jarring of the hammers while the first half was being driven. The penetration of the piles is also greater than it usually is in sand foundations in the bed of streams. The piles were all of Norway pine and well seasoned. Two Mundy steam hoisting engines were used in driving, one a single cylinder and the other a double cylinder engine. In operating the hammer an inch and a half manila rope was attached to the pin connecting the lugs of the ham- mer, then passed over the sheave at the top of the leaders, and next around the drum of the hoisting engine. When the hammer falls, it pulls the rope with it and unwinds it from the drum. This is what is termed driving with a "slack line." The blows are ;o THE COFFER-DAM PROCESS FOR PIERS. more rapid and keeps the material around the piles looser than it would be in the case of using nippers. Iron rings of 5/gx2^ inches Norway iron were used to protect the head of the pile. It is a well-known fact in pile driving that it is very important to keep the material from settling around the pile, once it has been loosened, until the pile is down; for when the material has settled, or even partially, the penetration is diminished. The greatest load on a bearing pile is about 13^ tons. Sheet piling was driven by a pile driver, assisted by a jet of water from a steam force pump. In driving all sheet piles a cast-iron cap or follower was used which fitted over the head of the pile. On the upper side of the fol- lower there is a wooden block of some seasoned or close grained wood which receives the blow of the hammer. This device saves the head of the pile from being battered or splintered, and the pile can be driven to a greater depth than it could be without it. '21' m FIG. 51 SANDY I.AKE COFFER-DAM. In first using the jet on a sheet pile, a groove was made in the inner edge to receive a half-inch gas pipe, which was connected to the force pump by means of an inch and a half hose. The aperture at the lower end of the gas pipe was reduced to a diameter of about three-eighths inch. The water was thus forced to the bottom of the pile, and the sand loosened. This worked well until the sheet pile struck gravel, when the nozzle of the pipe would become battered or filled with gravel. The pressure in the hose would then burst a coupling somewhere. Another source of trouble was the frequent breakages in the connection between the pipe and the hose, on account of the jarring of the hammer. This plan after awhile was abandoned and the nozzle of the pipe was thrust by hand under the point of the pile. The piles are driven in the ground from 12 to 14 feet. The construction of the Main street bridge at Little Rock, Arkansas, in- volved the construction of two coffer-dams, for piers No.. 5 and No. 6. This work was done under the direction of Edwin Thacher, Consulting Engineer, whose original specifications called for pile foundations for these piers, the piles to be driven to bed rock and cut off four feet below low water, to re- ceive a grillage of 12x1 2-inch timbers to receive the masonry. The size of THE COFFER-DAM PROCESS FOR PIERS. FIG. 52 COFFER-DAM AND CONCRETE PIER, LITTLE ROCK, ARK. the grillage being 12 and 13 feet wide by 34 feet long and resting on forty- eight and sixty piles respectively, the piles being of good sound oak or pine at least eight inches in size at the small end and not less than twelve inches at the butt when sawed off. The coffer-dams were constructed, as can be seen from the view in Fig. 52, by driving guide piles, to the top of which are drift bolted square guide timbers. The sheet piling of three-inch tongue and groove stuff was driven against the outside of this timber, and the excavation banked up against 72 THE COFFER-DAM PROCESS FOR PIERS. the outside. They gave excellent satisfaction and caused little trouble as the water was shallow. The piers were constructed of Portland cement concrete, the facing of two inches thickness being a mortar of one part of cement to two parts of sand while the balance was of concrete of one part cement, three parts sand and six parts of broken stone. Where sheet piles are to be driven on rock bottom or through earth or gravel to rock bottom, they should be driven hard enough to broom up and form a close joint with the rock. This has been accomplished also by driv- ing the piles with a thin edge until they fit the rock bottom, when they are drawn and after cutting them to conform to the contour of the rock, they are redriven, thus forming a tight joint. This method while very good, is too expensive for general adoption. Coffer-dams are quite frequently constructed for the repair or removal of existing piers. A pier which was constructed in 1840 in the river Parnitz, at Stettin, Germany, became an obstruction to navigation and it was de- cided to remove it. . The work was described in the Engineering News of July 14, 1892. Its exterior showed a facing of granite laid in hard Roman cement, and soundings revealed the existence of a course of sheet piling around the pier, with a protection of rip-rap at its foot. The original drawing of the pier showed a pile foundation. The specification prescribed the use of the old course of sheet piling, shown at A on accompanying cuts, for the construc- tion of the coffer-dam. Owing to the belief that the existing sheet piling, after having served such a length of time, would not be sound enough to permit of its use in the erection of a coffer-dam, local contractors could not be found and the work was let to an outside contractor. The preliminary work was begun by picking up the rip-rap around the foot of the pier with a claw dredger mounted on a raft. Some of the stones weighed as much as a ton. The bottom of the river, after the rip- rap had been cleared away, was found to be covered with a layer of concrete, consist- ing of pieces of brick and cement. This was brought up in large slabs. The pier itself was found to be of rubble masonry, composed of irregular shaped granite blocks with the interstices filled with brick, laid in cement mortar. The single stones were detached and swung off by the claws of the dredger. Their average weight was about one and a half tons. After the masonry had been pulled down to nearly the level of the water a row of sheet piling, shown at b in Fig. 53, consisting of piles seven inches thick, was driven to a depth of nearly ten feet. The space between the old and new sheet piling was filled with blue clay. To keep the interior free from water two pumps were employed. After putting in the necessary bracing the work of removing the masonry to the bed of the river was con- THE COFFER-DAM PROCESS FOR PIERS. 73 tinued. A shell of the latter, however, was left standing. Then the timber platform on which the masonry had been resting and the layer of concrete below were taken out, exposing a layer of clay underneath. While attempt- ing to pull one of the foundation piles a stream of water rushed through the opening thus formed, so that this plan had to be given up and blasting re- cJecr/orT, ReaUy for D'asting. Sectfot\ f OriOir\Ql Pier. Z6' 9" Plor\, R?<3dy for Blasting^. FIG. 53 REMOVAL OF MASONRY PIER AT STETTIN, GERMANY. sorted to. To do this the tops of the piles were bored to a depth of thirteen feet and filled with 8.8 pounds of dynamite each. The initial charges con- sisted of 10.6 ounces in air-tight canisters. The shell of masonry left stand- ing received four cubical charges of 8.8 pounds each. In all sixty-eight charges, consisting of 616 pounds of dynamite, were used. The electric current for the blast was divided into three currents, each being attached to 74 THE COFFER-DAM PROCESS FOR PIERS. an induction apparatus. The blasting, however, did not prove to be as effective as was anticipated, owing to the dissolving action of the water, and several charges were taken out intact. The clearing away of the wreck was almost entirely done by the claw dredges. The piles, which were split and loosened in their sockets by the force of the explosion, were pulled up by windlasses mounted on flat boats. The work of removing the pier lasted nearly nine months and the cost was about $8,700. Another example of the removal of a pier was at Gadsden, Alabama, where a pivot pier in the Coosa river had tilted. The pier had been built originally in a water-tight caisson and was supposed to have been founded on solid rock, but by some error a layer of gravel was left underneath and eventually the pier tilted down stream seven feet, nearly throwing the swing span into the river. After the span had been blocked up to allow the passage of trains, a coffer- dam was built around the pier to give plenty of clearance to the old caisson. (Fig. 54.) This was constructed by driving three rows of sheet piling through sand and gravel to bed rock and puddling between them. The sand and gravel over the rock was not removed from the bottom of the puddle chamber before puddling and a great deal of trouble was expe- rienced all through the work by leakage through the porous gravel. It is probable, too, that a poor joint was made between the sheet piling and the rock. Bents were erected upon the sides of the coffer-dam and by driving piles into the puddle and inside the dam, to carry a truss on each side of the span, which carried the drum and supported the main trusses at the center. When this had been tested by loading with trains of ore upon the bridge and found to be satisfactory, work was at once begun upon the removal of the old pier, by means of two fixed derricks on the false work and one float- ing derrick. The stones were marked as they were removed to insure their return to proper places when the pier was rebuilt, and were taken to the shore until needed again. When the masonry was all removed the grillage was broken up and taken out, after which the gravel inside the coffer-dam was cleaned out down to bed rock. New footing courses were laid to take the place of the gravel and old grillage, and the old stonework relaid by placing each course in its former position as nearly as possible. The pier was about 80 feet high and contained about 1,100 yards of masonry. The work occupied from Sept. 15 to Dec. 25, 1888, and was done under the di- rection of Cecil Frazer. The description is tiken from the Engineering News of April 13, 1893. The construction of the piers for the Philadelphia and Reading railroad bridge over the Schuylkill, was accomplished by the use of a floating coffer- dam, the foundations being laid upon the bed rock. THE COFFER-DAM PROCESS FOR PIERS. 75 When in position for work the dam is rectangular in shape, 62 feet long and 36 feet wide, outside dimensions, and 16 feet high. Each side consists of timber crib work 10 feet wide, making the inside dimensions 42x16 feet. At each corner there is a movable timber extending vertically from the bot- tom of the crib to some distance above the top. These timbers or spuds are shod with iron on the bottom, and serve to hold the dam in position while the sheet piling is being driven. The dam is divided vertically j6 THE LOFFER-DAM PROCESS FOR PIERS. through each short side into two equal parts, which can be floated separ- ately to any desired position and afterwards joined together. Watertight compartments are built in each section to assist in floating it, and these com- partments are also used to hold stone when it is desired to sink the cribs. When the two sections are united and placed in required position the spuds are dropped and the crib work is sunk by letting water into the water- tight compartments, and putting in the necessary amount of stone. Any irregularity in bearing between the bottom rock and the bottom of the crib is then corrected by a diver, who blocks up where required. Close sheet piling of jointed plank three or four inches thick is then put on the outside and spiked to the cribs. Puddle, composed of clay and gravel, is then thrown around the bottom outside, and the dam is ready to be pumped out. When the masonry reached the height of the braces they were taken out and the dam was braced against the masonry. The maximum depth of water encountered at Falls bridge was thirteen feet at ordinary water level. Several freshets occurred during the progress of the work which did some damage to the dam. At one time, when a dam was ready to be pumped out, a rise in the river moved it down stream about thirty feet, tearing off the sheet piling. It was drawn back to place and successfully completed. To make a complete shift of the dam from one pier to the next, with a gang of six men, required about six or eight days, divided as follows : To take the dam apart and reset it, about three days ; to sheet pile, about two days ; to puddle, about one day ; and pump- ing, out and puddling meanwhile required about one to two days, 'depending on the amount of the leakage. At each shift, a portion of the plank sheet piling, perhaps 10 per cent, had to be replaced by new stuff. The pump tised was located on a small steamboat, and was run by a steam engine. The amount of pumping required after the dam was once pumped out varied for the different piers ; some dams required little pumping and others a good deal. Only one of the foundations required much leveling off of the river bed, and this one also gave considerable trouble to keep the water out, but the leaks were finally stopped by using gunny bags around them ; the bags being drawn into the crevices by the force of the water, thus holding the puddle. The floating dam was used for the three piers in the river channel, the two piers near the shore being put in with ordinary dams. The floating dam is still in good condition and could be used again if needed. The original dam of which the one used at the Falls bridge is an enlarged copy, was used for twenty-three or twenty-four settings. The foregoing account is taken from the Engineering News of May 24, 1S94, the description being by W. B. Riegner, who states also that the cost of the coffer-dam, including one set of sheet piling, was $3,000, while the THE COFFER-DAM PROCESS FOR PIERS. total cost for five coffer-dams, including the two crib coffer-dams at the sides of the river, was $14,000. The subject of subaqueous foundations has been very fully treated of in a series of lectures by W. R. Kinipple, M. Inst. C. E., before the Royal Engineers' Institute at Chatham, England. The use of six-inch pitch pine close sheeting was made use of by him, for a quay wall in the harbor of St. Helier, Jersey. They were driven to rock or as deep as possible with a 2,800-pound hammer, and the tops cut off a few feet beneath half tide level, and clayey material banked up against the outside. The bottom through which the sheet piles were driven was sand and clay. The rock was laid bare to a depth of as much as thirteen feet below low water and in sections which contained about 900 tons of water to be pumped out ; this was done with a sixteen-inch centrifugal pump in about forty-two minutes. Several leaks were developed under the piles, but they were promptly stopped by "stock ramming." The stock rammer which is shown in Fig. 55, is 3 inches in diameter, 3^ feet long and banded top and bottom with iron. A ^-inch air hole is bored up from its foot a distance of twenty to thirty inches, and covered on the bottom with a sole leather flap, so that air is let in and suction prevented as it is with- drawn. The sheet piles have 3^ -inch holes bored through their sir 1 as, and cylinders of clay are inserted 3x9 inches long, similar to the work at Sault Ste. Marie. The stock rammer is inserted and driven by mauls as far as its length will permit when it is drawn out, and other charges inserted until no more clay can be driven. The hole in the pile being filled with a wooden plug. The piers for the Putney bridge, over the Thames, were built by the same engineer, with single pile dams to a great depth, by using fourteen- inch square piles, with elm wood tongues, and driving them down through the mud and clay to the stiff clay bottom, so that practically watertight work was secured In the construction of the docks at Victoria, British Columbia, he con- structed a coffer-dam 500 feet in length, in a depth of thirty-five feet of water, the bottom being of rock and overlaid in places with sand and shells several feet in thickness. At the center the sand and shells overlaid a bed of clay. Three rows of close 12xl2-inch sheet piling were driven with two puddle iron coUat- FIG. 55 STOCK RAMMER. 78 THE COFFER-DAM PROCESS FOR PIERS. chambers of seven feet each between. The guide piles were 15x15 inches and the wales were 12x12 inches. Where the dam rested on rock at the ends, heavy shoes were used on the piles and concrete deposited around their feet to make the work watertight. The dam was completed in October, 1879, and remained thoroughly tight until the dock was completed over seven years later. The arch bridge at Topeka, Kansas, over the Kaw river, which is being constructed on the Melan system, of concrete and steel, by Keepers and Thacher, the designing engineers, is a most interesting piece of work. The coffer-dams were required by the specifications to be watertight, and to FIG. 56 TOPEKA BRIDGE COFFER-DAM NO. 4. "A" shows puddle to stop leak. effect this 4x12 inch tongue and groove sheet piling was used. The size of the coffer-dam for pier No. 4 was 18x55 feet in the clear (Fig. 56) and the piling was driven about sixteen feet into the sand bottom or twenty-two feet below low water. The driving was done by a 1,600-pound hammer with thirty-six feet leads ; the power being furnished by a 15 H. P. hoisting engine. No puddle was used around the outside except to stop leaks, and the dam was kept clear of water with a No. 6 Special Van Wie sand pump. The capacity of the pump was 3,000 gallons per minute of water, and from sixty to eighty yards of sand per hour. It; was operated with a 15 H. P. THE COFFER-DAM PROCESS FOR PIERS. 79 engine. The other piers were handled in a similar manner and with no par- ticular trouble. The growing scarcity of timber will doubless lead to the use of metal at some time in the future, to replace sheet piling for coffer-dams, but where timber is abundant and reasonable care is exercised in its use, it will con- tinue to be of great service in obtaining foundations by this method. ARTICLE VII. THE COFFER-DAM PROCESS FOR PIERS.* METAI, CONSTRUCTION. HIN steel shells have been used extensively for foundation work, but in the majority of cases they have been retained as essential fea tures of the permanent construction. This is more particularly the case in locations where stone is scarce or expensive and it becomes necessary to substitute some other material for foundations. Tubular steel piers are con- structed of two tubes, ranging from 24 inches to several feet in diameter, or in the case of pivot piers, from 15 feet, with a single tube for a pier, to 30 leet or more. In a number of instances the steel shells for ordinary piers have been made oblong, in the general form of a stone pier, and braced internally to hold them in shape during sinking, after which they are filled with concrete. The metal shells for the Hawkesbury bridge in Australia were of this character, 20 feet wide, 48 feet long and with rounded ends. Each one was provided with three dredging wells, each 8 feet in diameter, through which the dredges shown in the view (Fig. 57) were operated. While these piers were not used as coffer-dams, they were made water-tight by boiler riveting, so that by pumping water in and out the displacement could be kept con- stant, and in this way control the pier in an average tide of five feet. These piers were sunk, by dredging out the material from the inside, to the great depth of from 135 feet 8 inches to 197 feet below the pier tops, or a distance of 155 feet below low water. Both inclined and vertical cutting edges were used, with the result that * Metal caissons have been used much more frequently in this country than have metal ' offer-dams, the reason being the cheapness of timber and its more easy application. In England metal coffer-dams are more frequently used. The example given in this article the Forth bridge coffer-dams might have been supplemented by reference to those used on the Clarence bridge at Cardiff, the construction used being illustrated and described in Engineering, and is especially notable for the design of the bracing. 80 THE COFFER-DAM PROCESS FOR PIERS. 81 the inclined ones were of frequent trouble and the vertical ones none what- ever. "If it is essential to increase the bearing surface at the bottom of the caisson to an area which is not required in the upper portion, this end can be secured by a vertical cutting shoe of considerable height, with a step or steps into the smaller diameter. This is quite as efficient to secure the end in view as a long incline on the cutting shoe, and has decided advantages. In the denser material the vertical sides leave the ground undisturbed for some height close to the skin of the caisson, and a vertical guide is secured riG. 57. HAWKESBURY BRIDGE. Caisson No. 6 in Process of Sinking, Showing Excavator and Shore Chains for Maintaining Vertical Position. which is entirely wanting in the case of an inclined shoe. This guide is valuable in cases where the soil may differ in density under the shoe, and particularly so if the excavation has been carried too far below the bottom of the shoe. With an inclined shoe and a slip of soil into the dredging well from one side more than another, experience in deep dredging has shown that there is a decidedly greater tendency to a horizontal movement than with a vertical shoe. The former has a flare to direct this sidewise motion in the first place, and nothing but a certain amount of disturbed material above the shoe to resist this tendency." The above account is from the Engineering News of January 5, 1889, 82 THE COFFER-DAM PROCESS FOR PIERS. the work having been done under the direction of J. F. Anderson, of the firm of Anderson & Barr. The shells were filled with concrete up to low water and masonry built from low water up to the top of the piers. Such work may be made water-tight by riveting according to ordinary boiler-maker rules, or if extra thick plates are used this can be exceeded and the rivets spaced some farther apart. The joints may be made with ordinary laps and calked, or a very much better appearance may be obtained by the use of butt joints, and if desirable to avoid calking, then a calking strip may be used to make the joints tight. This is merely a cloth or canvas strip, thoroughly saturated with paint paste, and is laid between the metal surfaces, and the riveting draws the plates upon it and a tight joint will result. The shells will be filled with concrete as soon as the piers are in place and the foundation prepared, so that only a temporary use is required of the strip. When metal cylinders are used simply as casings for concrete they need not be made water-tight, as they can be dredged out and have the concrete deposited through the water. The metal should never be less than one-quarter inch in thickness, and on first-class work five-sixteenths to one-half inch is preferable. Railroad work of this character is usually constructed of three-eighths inch metal for ordinary depths. The pivot pier of the bridge over the Little Bras d'Or river in Cape Breton was constructed of seven metal cylinders braced together. The cen- ter tube was 4 feet in diameter, while the six outside cylinders were 3 feet in diameter. (Fig. 58). The center pivot, about which the span revolves, rests on the center tube, while the track is supported by the other tubes, but resting directly on rolled beams covered with three-eighths inch plate. The tubes rest on a clump of piles, cut off at the bed of the stream, with FIG. 58. GROUP OF CYLINDERS FOR PIVOT PIERS. COFFER-DAM PROCESS FOR PIERS. one pile extending up into the center of each tube about six feet, around which the concrete was deposited, thus preventing displacement. Concrete and stone were placed on the outside up to 15 feet, as a protection. This work was described by Martin Murphy. in Trans. Am. Soc. C. E., Vol. 29, who also describes a pier for the Victoria bridge, over Bear river, constructed with two tubes, resting on piles cut off at the bed of the stream, but having four piles inside each tube. (Fig. 59.) Around the outside are timber, concrete and broken stone as a protection. The saw used for cut- ting off the piles under the water was very much simpler than the one shown in Fig. 35, and is illustrated in Fig. 60. Cylinder piers on European work are often of very elaborate construc- tion. The bridge on the Aa, at the crossing of the Russian Riga-Orel rail- way, is supported on elegant cylinder piers, with moulded caps, steel cut-waters, and are braced together with cylinders transversly. (Fig. 61.) This forms a very efficient construction, but so expensive to manufacture that it is usually re- placed by bracing of struts and rods, as in Fig. 59, or by a metal diaphragm (Fig. 62), stiffened with angles. Cylinders of water-tight construction and of large diameter may be used as coffer-dams, where they are sunk into impervious strata, or by sealing them with concrete around the bottom where they are placed upon smooth rock bottom. In the construction of light-houses such cylinders have been placed upon clean rock bottom through from 12 feet to 18 feet of water and concrete deposited around the circumference of the base outside and inside to make them water-tight, after which they were pumped out and the foun- dation laid. To withstand the pressure of any considerable depth of water the thick- ness and strength should be calculated and the construction carefully designed. Unless the depth of water exceeds ten feet, or the diameter of tube exceeds six feet, the minimum thickness it is advisable to use, will be sufficient for strength. This refers only to quiescent pressure, and any shock must be carefully considered and taken account of, by interior bracing if necessary. The most thorough discussion of the strength of thin, hollow metal cylinders is given in " Elasticitat and Festigkeit," by C. Bach. This con- siders the cylinder to have sides of a greater thickness than is true with FIG. 59. PIER OF TWO CYLIN- DERS, VICTORIA BRIDGE. 8 4 THE COFFER-DAM PROCESS FOR PIERS. pier shells, and having one radius given, the radius to the other side of the plate is found from the formula, the stress being variable from the inside to the outside of the plate. For thin cylinders the stress may, without appreciable error, be assumed to be uniform over the cross section of the plate, and the thickness t in inches be found from the formula t .001 r h where r is the radius of the cylinder in feet and h is the depth of the water foundation Pi/ing for Victoria. Bridge JL iflhl i nr t O - --16' 0-- - > CwJ 2tf j , Jd V-,' J6V/d. L L r n j i u L -i= =r 1 1 U n i UA 1 ' 1 | J J 1 ! i J j, r^ L. 1 ^h JjS-w^p J ^ v |r =- =r ====- =' = FIG. 60. CIRCULAR SAW FOR CUTTING OFF PILES UNDER WATER. to the section in feet, and t in no case to be used less than one-quarter incn in thickness. This is on the assumption that the metal will stand 5,000 pounds per square inch in compression with safety. For large cylinders, or for rec- tangular shells, girders and stiffeners or ties and struts must be added to prevent distortion. The foundations for the great Forth Bridge, which were constructed 86 THE COFFER-DAM PROCESS FOR PIERS. under the direction of Sir John Fowler and Sir Benjamin Baker, required the use of various methods to reach solid bearing, as the enormous weight to be carried required the most substantial piers obtainable. The use of coffer-dams of metal for the Inchgarvie piers is described by Engineering: The site of the two north or shallow piers being wholly sub- merged at high water, and about half in the case of the northeast and three- fourths in the case of the northwest pier, submerged also at low water, the preliminary work was tidal, and between spring tides no work could be carried on at all at this place. When it is con- sidered how exposed the position was there the work having to be car- ried on upon a narrow ledge of rock attacked by wiiid and waves from all sides it will be un- derstood that the pro- gress could not be very rapid. The conditions of the contract here re- quired that the rock should be excavated in steps, and that the rub- ble masonry comprising the foundation of the cir- cular granite piers (Fig. 63) should be bound by an iron belt 60 feet in diameter and 3 feet deep; the highest portion of the rock upon which this belt rested to be 2 feet below low water; the belt, or at any rate a part of it, to be brought down to form a protection for the foundation rubble masonry upon the lower steps. It was therefore decided to cut a chase 8 feet wide (3 feet to the inside and 5 feet to the outside of the 60 feet circle) out of the rock where it was higher than 2 feet below low water, to make the 60 feet belt of three thick- nesses of one-half inch plate and to carry the center plate downward, after it had been cut, in such a manner as to fit as nearly as possible the natural contour of the rock. (Fig. 64 A.) A light staging was, therefore, erected above high water, the correct center of the pier placed upon it, and by means of a trammel-rod 30 feet in length, from the end of which a pointed sounding-rod was suspended, a correct reading was taken every 6 inches on FIG. 62. CYLINDER PIERS, WITH DIAPHRAGM. OF THK UNIVERSITY 87 COFFER-DAM PROCESS FOR the circumference of the 60 feet circle, after a diver had been around to clear out any loose stones lying in the line, or picking off any sharp points projecting. These readings were plotted and the center plates cut to it. In the meantime work had been done upon the chase; and, when nearly cut down to the right level, the belt was put together on the staging exactly above the site of the pier. The plates, projecting downward and forming the shield, were stiffened by I bars vertically over the butts, and where required to be carried down to a considerable depth, as in the case of the northwest pier, they were further stiffened by horizontal circular girders and stayed to the rock by bars of angle iron. The whole belt was now riveted up, and when ready received two coats of red lead paint, and was lowered down to position by means of hydraulic jacks. (Fig. 64B.) The top edge of the 3 feet belt was then leveled all round, and corrected where FIG. 63. CIRCULAR GRANITE PIER AS FOUNDED BY COFFER-DAM. FORTH BRIDGE. necessary. A heavy angle iron 6 inches by 6 inches by J/Q inches ran round the inside of the 3 feet belt, and upon this was now set a single tier of temporary caisson, 10 feet in height, .and consisting of fourteen segments of about 30 cwt. each in weight. This helped to keep the belt down to the rock, and a number of heavy blocks of stone were placed on the top of the caisson for the same purpose. A sluice door in the lower part was kept open to admit of the tide flowing in and out. Steps were now taken to make good the joint between the 3 feet belt and the shield and the bed-rock. This was done in the following manner : A number of concrete bags, about 14 inches by 30 inches, and 8 inches to 9 inches thick, were prepared and passed down to a diver, who laid them round the outside of the belt at a distance of about 4 inches. A second row was next laid round the outside of the first row, and tolerably close up, THE COFFER-DAM PROCESS FOR PIERS. the space between the two being made up by clay puddle well stamped down. Any split or hole or crevice in the rock was also filled with clay. Upon these two lower rows other bags were now laid crosswise ; upon these, two rows lengthwise, and a fourth row crosswise on the top, which was laid close up to the belt. This was done in sections of about 15 feet to 16 feet length all along the shield, but round the outside of the treble belt only two bags deep were laid. On the inside also a single row of clay bags, AI.IV. IROH Section Through Shield cS/miv/rvgMocte of Aiding Joint to RocK Cone, bags loaded ly/tA sane/ bags. Cement groi/t. w I ROM COFFER-DAM . //. IV. PIER-* Outside View o/c5/i/e/c/. FIG. 64. FORTH BRIDGK. METAL COFFER-DAM. backed by a row of concrete bags, and loaded with stones, was laid round the complete circle. Cement grout, without intermixture of sand, was now prepared and passed down to the diver but only at slack tide, high water or low water who lifted off one or more of the top bags and poured the grout into the narrow space left, until it overflowed. He then replaced the bag and proceeded to the next division, until all was done. Forty-eight hours were allowed to elapse for the setting of the cement ; the sluice valve THE COFFER-DAM PROCESS FOR PIERS. 89 was then closed and the caisson pumped out gradually. When leaks were discovered the diver descended to examine the outside, and where neces- sary, cut out some of the grouting and replace it by new. As it was not considered that this cement joint would be able to stand the full pressure of the tidal rise the coffer-dam was worked as a half tide one, it having to be pumped out every tide as soon as the water had fallen below the top edge of the temporary caisson. In addition to the hydro- static water pressure, the caisson had to stand the heavy seas thrown against it, whether coming from east or west. Under these circumstances it was often considered advisable not to pump out the coffer-dam, but leave the sluices open and allow the tidal flow free access. Under such condi- tions it will be easy to see that, during a season of bad weather, much delay could not be avoided, and though the work of excavation had been com- menced in the summer of 1883 it was not till the middle of April of the following year that the first 'rubble masonry could be laid in this pier. In working the excavation no blasting was done within one and a half feet of the iron belt, but the rock was quarried up to within 6 inches and the rub- ble then built in at once. Any steps cut in the deeper portion were invar- iably at least twice as broad as they were deep. The deepest point to which the excavation had to be carried in this pier was 8 feet below low water. The coffer-dam or caisson for the northwest pier, Inchgarvie, was done in the same way precisely as described for the northeast, only that owing to the experience gained by the divers and other men engaged upon the work the progress was much more rapid. In the northwest pier the depth of the shield was 15 feet below low water, and extended to nearly one-half of the circumference. There was, therefore, in addition to the vertical I bars which covered the butt joints of the shield plates, three horizontal circular girders, carried at a distance of 4 feet 6 inches from each other, and from these a number of horizontal tie bars with cross-bars at the ends were carried radially and level to the rock opposite and pinned to it, and afterward built into the solid rubble masonry. (Fig. 64B.) This mode of making the joint between the rock and the iron belt was simple and quite effective. Most of the leaks were due to natural crevices in the rock, running from the inside to the outside at a considerable depth. These were circumvented by building small clay dams round, and leading the water by a chute to the pump. Leaks were also caused by the action of heavy waves running up to the temporary caisson at low water with great violence, and shaking the whole fabric. The whole of the northeast pier was built in a half-tide caisson, as the work was not pressing ; but in the case of the northwest pier, so soon as the 9 o THE COFFER-DAM PROCESS FOR iIERS. rubble masonry inside had been brought up to low water level a second tier of temporary caisson was added, and the work could then be carried on at all states of the tide. While tidal work was carried on in these two coffer- dams the amount of water which had to be pumped out every tide was 250,000 gallons in the one case and 340,000 in the other. The time occu- pied was 50 to 55 minutes, but work was, of course, commenced so soon as the higher parts were laid dry. For pumping out smaller quantities of water collected through leaks, pulsometers or small centrifugal pumps were used. An exterior view of the work is shown in Fig. 65, and while the method FIG. 65. FORTH BRIDGE, CIRCULAR GRANITE PIER AND METAI, COFFER-DAM. was successful and worthy of much study, the expense would only be justifiable where the metal would be retained as part of the permanent foundation, which was the case on this work. In man} 7 cases such a shell could be designed of the proper size for the footing course, and after use as a coffer-dam in obtaining the foundation it could be filled with concrete and serve as a base for the pier. Being made in sections vertically, portions projecting above low water could be removed and used on still other piers. Metal sheet piles are seldom used for any class of work, and need not be THE COFFER-DAM PROCESS FOR PIERS, 91 discussed at length in this connection. On some harbor work at Cuxhaven Harbor, Germany, hollow metal sheet piles, of elongated elliptical section, were used, and after being driven were filled with concrete. Whatever the class and form of material it may be decided to use, in securing a foundation bythe coffer-dam method, the temporary construction should be so related to the permanent foundation that as much as possible of the material used and labor employed will be of service in the finished structure. ARTICLE VIII. THE COFFER-DAM PROCESS FOR PIERS.* PUMPING AND DREDGING. HE degree of success which has been attained in the building of a coffer-dam will be evident when the pumping process is begun. After having been pumped out, if the leakage is so small as to require only a small amount of pumping to keep it free from water, it may reasonably be consid- ered a success. The pumping should not exceed what can be done by a steam siphon, a small pulsometer, or by running a centrifugal pump intermittently. Should leaks develop which cannot readily be contended with, then repairs must be made. The use of pumps for this class of work on ancient bridges is described by Cresy. The bascule, used by Perronet at the bridge of Orleans (Fig. 66), is one of the most primitive forms. It consists of a see-saw apparatus, at each end of which ten men were placed, and 150 motions were given it in each quar- ter of an hour. Four cubic feet of water were raised three feet each time, or about 300 gallons per minute. Various other kinds of pumps were used at this bridge, among them the chapelet, which is similar to a modern chain pump, worked by hand. Then the same device was employed, but geared to be operated by horses on a platform. A chapelet operated by a water wheel was also used (Figs. 67 and 68). The large wheel had 124 cogs, while the pinion had 15, which caused the raising of over sixty-six buckets on the chain for each turn of the large wheel. At 180 turns of the wheel per hour, with each bucket lifting 290 cubic inches of water, the capacity was about 250 gallons per minute. A great bucket wheel was employed by the same engineer at the Neuilly bridge, 16 feet 6 inches in diameter, 4 feet 6 inches wide, with sixteen buckets. * Attention is called to the numerous references in other articles of the pumping plants actually employed on coffer-dams, and especially to the plant used at Topeka, page 78. Great care should always be given to the selection of a pumping plant of the proper type and proper size, as the statements regarding capacity are often misleading. The outfit should be, if needed, one able to take care of the dredging, if the material is such that it can be pumped. 92 THE COFFER-DAM PROCESS FOR PIERS. 93 FIG. 66. OIf Rear Bent and Platform* Front Elevalion CONCRETE MIXING PLANT. Side -Elevation FIG. 90. STONE CRUSHER AND CONCRETE MIXER. IIJvINOIS AND MICHIGAN CANAI, enough to be struck off by a straight edge, the top of the frame being dressed and leveled for that purpose. " After the pier had been completed the frames were removed and the braces running through the piers cut off by a chisel inside the concrete. Then, to make a smooth surface, the pier was thoroughly wet and plastered with a mixture of one part sand to one part cement, after all the rough or THE COFFER-DAM PROCESS FOR PIERS. loose portions had been scraped off. This was mainly done for appearance," The mortar for the body of the pier was made of one part Alsen's Ger- man Portland cement and four parts of sand. There was used about l^i barrels of cement to a cubic yard of completed pier. In mixing the mortar eleven ordinary pails full of water were used to one barrel of ce- ment, which caused the water to just ap- pear on the surface when the tamping was done. The lock walls on the Illinois and Mis- sissippi canal have been constructed of monolithic concrete under Captain W. L. Marshall, Corps of En- gineers. The work was executed under L. L. Wheeler, en- gineer in charge, from whose account in the report of the Chief of Engineers for 1894, the followingis taken: "The rules adopt- ed for the work were adhered to and are worthy of careful study. "I. The forms or molds of the walls will be divided by vertical FIG. 91. AMERICAN HOIST AND DERRICK CO. GUY DERRICK. DOUBLE DRUM partitions perpendicu- lar to the longest axis of the mass, and the walls be constructed by filling alternate sections. "II. The sections will be filled in horizontal layers, well rammed, each layer to be deposited before the 'initial set' of the previously deposited layer. When the work of filling a section is begun it must proceed without inter- mission to completion, working night and day if necessary. 114 THE LOFFER-DAM PROCESS FOR PIERS. "III. The facing and backing must go on simultaneously in the same horizontal layers, using the same cement in the facing as in the backing, with no defined lines of demarcation between the facing which contains no stone and the concrete backing. "IV. When the top surface of the coping is reached it will be finished after ramming by cutting off the excess by a straight edge, and rubbed smooth and hard by a float. No plastering or wet finishing coat will be allowed. "V. The facing of the walls will not be finished by plastering or washing with cement after the forms are re- moved, nor dressed in any manner beyond chiseling away rough ridges should the plank forming not be smooth. FIG. 92. SINGLE DRUM HORSEPOWER. CON- "VI. The concrete shall TRCATORS PI, the total cost of the structure in dollars by y, while a and b are constants. ' From the estimated cost of a large number of spans, a curve of costs was plotted and the following equation of a parabola deduced : c = a FIG. 100. CLAMP AND Since there are n spans and n + 1 piers, the total cost of the structure would be Then by substituting the value of c from (1), reducing and making the first THE COFFER-DAM PROCESS FOR PIERS. 125 differential coefficient equal to zero the cost of one pier is obtained, which will make the total cost of the structure a minimum or Or when the cost of a pier has been estimated, the economical length of span may be found by a transposition of the above formula : s = y"ab + 400 + pb (4) The values of a and b may be found by substituting in equation (1) com- puted values of the cost of a number of spans for an actual loading. Values of 5, p and c, may then be computed and tabulated for spans from 100 feet upwards, as formula (1) is not true for shorter lengths. In an actual calculation for B. & O. R. R. loading, which consists of two 125-ton engines followed by a 4,000-pound per lineal foot trainload, a was found equal to 1950 and b to 3.05. Assuming a case where the length of the bridge is 700 feet, where the height of the piers will average 25 feet, and the average cost of piers and abutments be $4,310, then from formula (4) the economical span will be found equal to 140 feet. The total cost of the structure will be found, by using formula (1), and the cost of piers as above, to be $59,700. While with only four spans of 175 feet, the total cost would exceed $60, 800, and with six spans of 117 feet, would be about $61, 400. Should there be any doubt as to the ease of obtaining foundations, the prudent engineer might deem it wise, however, to build the four-span structure and avoid the risk and delay which would be caused by another foundation in the river. After deciding upon the number and location of the piers, they must be designed with reference both to their being as slight obstructions to the water as possible and to their architectural appearance. The design of piers has been given particular attention by Geo. S. Morison, Consulting Engineer, whose work on the bridges across our great rivers is notable for its strength, simplicity and finished appearance. In a recent lecture he describes the process of the design of some large piers : "Fourteen years ago I had occasion to design a bridge pier for a bridge across one of our Western rivers, and I tried to make an ornamental pier. When the plans were completed I did not like them. One change after another was made, all tending to simplicity. Finally the plans were done. From high water down, the pier was adapted to pass the water with the least disturbance; it had parallel sides and the ends were formed of two cir- cular arcs meeting. Above high water the ends were made semi-circular instead of being pointed. The pier was built throughout with a batter of one in twenty-four. A coping two feet wider than the body of the pier pro- jected far enough to shed water, and the projection was divided between <;. 101. I'IKK OK OMAHA IJRIIXIK, UNION PACIFIC. FHE COFFER-DAM PROCESS FOR PIERS. 12; the coping and the course below. Another coping with a less projection surmounted the pointed ends where the shape was changed. It was as simple a pier as could be built, and in every way fitted to do its duty. I had started to make a handsome pier. The pier that was exactly what was wanted for the work, was the only one that satisfied the demands of beauty. Forty-three piers of precisely this design (no change having been made except in the varying dimen- sions required for different structures), besides eight oth- ers in which only the lower parts are modified, are now , standing in eleven different bridges across three great Western rivers. In designing a pier it must be remembered that the portion of the pier below the water has more to do with the- free passage of the water than that above water. In a deep river the model form of the pier should begin near the bottom of the river and not at low water. Many rivers in flood time carry a great amount of drift. A pier like that which I have described catches but little of this drift. If, however, a rec- tangular foundation termin- ates but little below water, that foundation may uoth disturb the current and catch the drift. ' ' The piers of the Omaha bridge, which carries the Union Pacific across the Missouri river, are illustrated in Fig. 101, and were constructed as described and are among the most beautiful piers in this country. In Europe, where money is more lavishly expended on great works of engineering, piers of great architectural beauty are much more frequent. The Russian Government railways, which have seemingly been constructed without regard to expense, have many beautiful examples of bridge masonry and piers; the view of one of them (Fig. 102) with curved ends, shows the elegant and massive character of the masonry. While extremely FIG. 102. RUSSIAN PIER, RUSSIAN STATE RAILWAYS. FIG. lOo. CRESY'S EXPERIMENTS ON THE FORM OF IMKKS. THE COFFER-DAM PROCESS FOR PIERS. 129 simple in design, the cut stone coping and the moulded corbel course below give it a finish which cannot be surpassed. The design of piers for strength and stability is fully treated in Baker's Masonry Construction, but some experiments, which were made with ref- erence to the proper form to occasion the least resistance, will be quoted at length from Cresy. The introduction of piers into a channel gives rise to a great disturbance in the velocity and flow of the water. Rapid currents are formed which cause the bed of the stream to become washed and the foundations to be endangered; eddies are created which are likewise undesirable, and it becomes necessary to adopt such a form for the ends of the piers that the disturbance to the flow shall be small. M. Bossut, in a French work on jetties, thought to have solved this problem by mathemathics, his conclusion being that the starling should be triangular, the nose being a right angle. M. Dubuat, in his "Principles of Hydraulics," gave another solution which was more nearly the truth, in that he arrived at the conclusion that the faces of the starling should be convex cnrves. The true form is most nearly reached when these curves are tangent to the sides of the pier, and further than this, regard must be paid to giving enough solidity to the starlings to protect them from ice and drift. A happy medium would seem to be reached, by making the curves with a radius equal to one-sixth of the circumference, described on the sides of an equilateral triangle. Experiments were made with models of different forms, which were placed in a rectangular canal between boards of 50 centimeters in length, in which the water flowed about 40 millimeters in height, the models being 15 centimeters in thickness. By means of a fall, the water was given a velocity of 3 meters 9 centimeters per second, the contraction, eddies and currents being carefully measured. The first experiment was made on a pier (Fig. 103 A) with rectangular starling. An eddy was formed before the pier 34 millimeters high, in a nearly circular band A, falling nearly vertical at the corner. There were two other currents along the faces of the pier, the height of which can be seen in the cross-sections. The second experiment (Fig. 103B) was with a triangular starling, the nose being a right angle. It formed a less obstruction than the square end, but the fall at the shoulder was as deep and more dangerous, while eddies were formed as seen in the sections. The third one (Fig. 103C) had a semi-circular starling. The eddy was not so wide, but nearly as high. The fourth model had a triangular starling, with an angle of 60 degrees at the nose. (Fig. 103D.) The eddy was less, as was also the fall at the shoulder. 130 THE COFFER-DAM PROCESS FOR PIERS. The starling in the fifth was formed by two circular arcs, tangent to the sides and described on the sides of an equilateral triangle. (Fig. 103E.) The eddy was small and there was no fall at the shoulder. FIG. 104. CRESY'S EXPERIMENTS ON THE FORM OF PIERS. The sixth (Fig. 103F) was a model, the plan of which was an ellipse, of which the small diameter was one-fourth the length, and the eddy was less than any of the others. The seventh model (Fig. 104 A) had a starling with concave faces, such as is sometimes used where the wing wall meets an abutment. It produced the most dangerous currents of all. THE COFFER-DAM PROCESS FOR PIERS. 131 The eighth (Fig. 104B) was of the same form as Fig. 103E, but the water was supposed to mount the springing of the arch. The ninth and tenth experiments (Figs. 104C and 104D) were on the same forms as Figs. 103E and 103F, but the current had a velocity of 4 meters 87 centimeters per second, such as a river would have in its overflow. The eddy (Fig. 104C) rose to nearly twice the height, as was the case with the lesser velocity, and while there was no fall, the inclination formed along the faces was more rapid. The effect with this velocity on the elliptical pier (Fig. 104D) was sim- ilar to the lesser velocity but more marked. It may thus be concluded that the elliptical section offers the least resistance to the current and occasions the least contraction, while the form with convex starling comes next, and of piers with triangular starlings the one with the 60-degree nose is the best. Where ice is to be provided for, the nose is often inclined to allow large cakes to mount it and break in two, without doing further damage. For any large or important structure, the design of the piers should receive a great deal of study, and be designed not only with reference to their theo- retical form, but with reference to the form of pier which has shown the best results practically and has been found to be best suited to the velocity of the stream in which they are to be built, and to best withstand the drift and ice that may be met with, giving at the same time all the consideration possible to the architectural effect and to the harmony with the entire structure. APPENDIX. SELECTIONS FROM SPECIFICATIONS. SPECIFICATIONS FOR COFFER-DAMS AND FOUNDATIONS, OHIO RIVER MOVABLE DAMS. MAJOR W. H. HEUER, U. S. Engineer. GENERAL DESCRIPTION. The site of Dam No. 2 is on the Ohio River, distant from Pittsburg, Pa., ioi miles, and adjacent on the right bank to the Pittsburg, Ft. Wayne and Chicago Railway. It has Neville Island on the left bank, and is accessible by street cars from Pittsburg. The lock is to be located on the left bank of the Ohio River, immediately behind Merrimans dyke. It will be in general dimensions the' same as locks Nos. i and 6, viz., no feel wide and 600 feet long. SPECIAL DESCRIPTION. The river bed at No. 2 consists of gravel throughout, and the excavations will be made to a depth sufficient to insure a permanent and enduring foundation, which will ordinarily be 14 feet below the gate sill, but may be otherwise, as the Engineer, in his judgment, may direct. The work will conform to the drawingsexhibited, and to such others, in explanation of details or modifications of plans, as may be furnished from time to time during con- struction. CONTRACTOR TO FURNISH ALL MATERIAL AND WORK. It is understood and agreed that the contractor, under his contract prices for work in place, is to furnish and pay for all materials, stone, cement, sand, earth, timber, material for coffer-dam and protec- tion cribs, excavation, lock-filling and discharging valves (set in masonry), flushing valves, anchor bolts, lock-gate tracks, and everything entering into or connected with either the permanent or temporary construction, and he is also to supply and pay for all work : skilled and otherwise, required to prepare and place the materials, and complete the work according to the drawings and these specifications. CONTRACT TO INCLUDE The contract will cover the construction and completion of the foundation, masonry and timber work of the lock, including both land and river walls, the gate-recess walls, the foundations of the lock-gate tracks, the guiding walls above and below the lock, the pipe and flushing conduits, the drift chute, the founda- 133 134 APPENDIX tions for the power-house and lock-keepers' residence, and every such other permanent construction as shall be shown upon the drawings. It shall also include the clearing of the land necessary for the proper execution of the work embraced, in this contract, all pumping and bailing, dredging and excavation, puddling and embankment, the con- struction of all coffer-dams, stone masonry, concrete and brick masonry, timber work and iron work, and all such other work which, in the judgment of the Engineer, is nec- essary and included in the proper completion of the contract. TOOLS, MACHINERY, BUILDINGS, ETC. The contractor, without cost to the United States, shall furnish all appliances, dredges, pumps and pumping machinery, boats, tools, derricks, tramways, foot walks, roads and landings, and all needful temporary buildings and shops. COFFER-DAM. SHEETING. The coffer-dam, about 1500 feet in length, shall be built as shown gen- erally by the drawings exhibited, and as directed by the Engineer. It shall be 14 feet high above the sill of the lock, and shall consist of two walls or rows of plank sheeting, spaced 12 feet apart in the clear, driven or set firmly from i to 2 feet into the river bed, and supported laterally by horizontal longitudinal stringers, the latter being spaced at varying intervals, increasing in width from the bottom to the top, and to be suffi- ciently and firmly bolted together transversely with iron rods passing through the coffer-dam horizontally from the rows of stringers on the one side to the corresponding rows on the other, against which the vertical plank sheeting shall be securely spiked. FILLING AND DECKING. The interior, or space between the walls of sheeting, shall be filled with heavy dredged river-bed or other material not liable to wash, and to be covered over with a suitable decking of plank (to protect it from injury in case of being submerged by floods), all complete as shown on the drawings. PILING AND CRIBS TO PROTECT. At the upper outer corner of the coffer-dam shall be placed a crib built of framing timber and filled with riprap stone ; from the upper corner of the crib, at an angle of 45 with the axis of the current, a line of piling, spaced 5 feet apart, firmly bolted together with waling pieces, shall be driven to the shore to form a protection to the coffer-dam ; also outside and along the coffer-dam, from the upper outer corner to the lower corner, clusters of piles, firmly bound or bolted together, shall be driven at intervals of about 80 feet. The tops of all piling shall be sawed off to a uniform height, of 2 feet above the coffer-dam. Protection cribs shall be placed at such other points along the coffer-dam as may be shown upon drawings. How PAID FOR. Bidders will state a price per lineal foot of coffer-dam completed. No payment will be made for any portion thereof until the entire coffer-dam is completed. Drawings will be furnished, showing the general type of the coffer-dam and its manner of construction, and every detail necessary for intelligent bidding. Should any work on the outside of the coffer be necessary, such as gravel filling or riprapping, it shall be paid for at the price bid for gravel filling, riprapping, etc. If, owing to the nature of the river bed, it shall be found impossible to drive the plank sheeting to the required depth, then the contractor, after driving the sheeting as deep as possible without injury, and in lieu of driving it to its full depth, shall fill around the outside of the walls with the same material as is used in filling the coffer-dam, to the height of 4 feet above the surface of the river bed, and for which no extra compensation will be allowed. REMOVAL OF. The contractor will be required to remove the coffer-dam and its be- longings at his own cost. The time and manner of the removal of the coffer-dam, or any part thereof, and the place to deposit the materials, shall be prescribed by the Engineer. To BELONG TO THE UNITED STATES. It is understood and agreed that the payments made for the coffer-dam, including the crib and pile protection, shall cover the entire cost thereof to the United States, and by virtue thereof they shalj become the property of the United States. The contractor, however, must maintain the same and make all needed repairs to same during the existence of the contract, without expense to the United States. DEPOSIT WITHIN THE COFFER-DAM. Material washed or left in the space enclosed by the coffer-dam by freshets shall be removed by the contractor, as directed, at his price for common excavation, which price shall cover all necessary cleaning and scrub- bing. No payment will be made, however, for removing material washed into the enclos- APPENDIX. 135 lire from the coffer-dam itself or from any deposit made by the contractor on or above the works. MATERIAL AND WORKMANSHIP. TEMPORARY PILING shall include all piles driven for the protection of the coffer- dam and "deadmen" for derricks. They shall be of good quality, round oak timber, not less than 12" diameter at the butt, and of length varying from 20 to 25 feet, and longer if necessary. SHEET PILING. In excavating for foundations, should quick-sand or fine-sand- carrying water be encountered, close sheet piling will be required to be driven to what- ever extent the Engineer may direct. SHEETING. The sheeting shall include the walls and decking of the coffer-dam, in- cluding the stringers ; also such shoring as may be directed by the Engineer to remain in the finished structure. It shall consist of the best quality of hemlock obtainable, and must be in all cases satisfactory to the Engineer in charge. GRAVEL OR EARTH FILLING. Gravel or earth filling will include all material used in filling the land-wall enclosure, back of the guiding walls, etc. It does not include any filling in the construction of the coffer-dam. STONE FILLING shall include all stone placed in the protection cribs or any riprap stone ordered for the protection of the work. CRIB WORK shall be built of hemlock framing timber framed together in square bins and securely bolted together by iron drift-bolts. The interior of the cribs shall be filled with riprap stone, and should the Engineer deem it necessary such riprap stone shall be placed on the outside of the crib. The whole to be built as shown by the drawings. FRAMING TIMBER. For all temporary crib work, also the permanent crib at the head of the upper guiding wall, framing timber shall be used. No stick shall be less than 10" X 10'' in section. " Framing timber " is a commercial term for a class of timber hewn to various sizes. EXCAVATION. To INCLUDE. It shall include the removal of all gravel or other material to the depth required for the lock and its upper and lower entrances, the gate recesses, Poiree- dam and gate-track foundations, for the foundations of all walls, and for all conduits or wells, and all such other material as may be found necessary in the judgment of the Engineer to be removed for foundations and otherwise in permanent construction. It will include all dredging and all material excavated of whatever nature, however removed, for foundations and for site of coffer-dam. LINES, SLOPES, AND GRADES FOR. All excavations shall conform to such lines, slopes, and grades as may be given by the Engineer, and anything taken out beyond such given limits will not be paid for by the United States. MATERIAL TO BE DEPOSITED. Excavated material is to be deposited as and where directed by the Engineer. It shall be deposited in such manner as not to interfere with present or proposed navigation. Material of any kind deposited by the contractor in absence of, or in disregard of. instructions, shall, if required by the Engineer, be removed by the contractor at his own cost. SHORING. All excavation for foundation shall be securely shored and thus main- tained until the foundation has been sufficiently advanced to dispense with the same, when it may remain or be removed at the discretion of the Engineer. DREDGF.S AND PUMPS. The contractor will be required to employ, at the same time, not less than two suitable steam dredges at excavating and filling; and for pumping he must keep at least three good sufficient pumping outfits, with pumps, engines, and boats complete, in or always ready for operation. The dredges must be equipped to do effective work to a depth of 28 feet. APPENDIX. FOUNDATIONS. CHANGES OR MODIFICATIONS OF. The character of the river bed and of the proposed foundations for the different parts of the work is shown in general on the drawings and cross-sections exhibited, and it is understood that the United States shall have the power to make any changes in the plans of the foundations that may, in the judgment of the Engineer, be considered advisable after examinations made, as the excavations proceed within the coffer-dam after it is pumped out and it is understood and agreed that the contractor shall have or make no claim against the United States on account of any such changes in or modifications of the plans of the foundations, or on account of any increase or decrease in the depth of same, under or over those referred to herein or shown on the drawings exhibited. MASONRY. CEMENT. Cement will be of uniform quality, setting well both in air and water, and free from anything that will cause the mortar to swell, crack, or scale. It shall be put up in strong, sound barrels, well lined with paper so as to be reasonably protected from air and moisture. The average net weight of the barrels shall be not less than 265 pounds, unless expressly so stated in the proposal. Each barrel must be labeled with the name of the brand and of the manufacturer. In general, ten barrels of every one hundred will be tested. The cement must stand .the following tests ; Fineness At least 85 per cent must pass a sieve of 6400 meshes to the inch. Setting Cement must be moderately slow setting ; it must not begin to set within fifteen minutes, as determined by Vicat needle 1/12 inch in diameter with 1/4 pound load, and it shall not bear the weight of one pound on wire 1/24 inch in diameter within thirty minutes, but must bear such weight within one hour and a half. Strength The minimum tensile strength per square inch of briquettes of neat cement mixed with about 33 per cent of water by weight, and exposed in air for one hour, and the remainder of 24 hours in water, shall be not less than 50 pounds ; with longer time, whether in air or water, there must be a decided increase of strength ; it must also test to the satisfaction of the Engineer when mixed with sand. The tests for setting will be made at a temperature of air and water of about 75 Fahrenheit. All other tests will be made at a temperature above 6o 3 Fahrenheit. The cement will be subject to inspection at all times, and must be kept well housed. SAND. The sand used must be clean, sharp, washed, river sand, satisfactory to the Engineer. MORTAR. To be composed generally of two parts of sand to one of cement ; when required, and whenever thought necessary by the Engineer, it shall be made richer. It must be thoroughly mixed and used before it has begun to set. If required by the Engineer, the mortar beds will be protected from the sun. POINTING.' All face work is to be pointed, as fast as the work progresses, with stiff mortar, mixed, one of sand to one of Portland cement, thoroughly hammered in and finished with proper tools ; before the final acceptance ot the work all face masonry which at that time does not appear properly pointed shall be repointed by the con- tractor to the satisfaction of the Engineer, without extra cost. FROST. Masonry will not be executed during freezing weather, nor when, in the judgment of the Engineer or his agent, it is likely to fieeze before the rnortar shall set. To guard against injury from frost all new a^nd unfinished work shall be properly protected by the contractor at his own cost. VOIDS AND OPENINGS. Due regard shall be had in the construction of all masonry w r alls to leave all necessary voids or openings for conduits or wells, or for such other purposes as may be required by the Engineer. ASHLAR. It shall comprise such part of the walls as is built of stone, with point- dressed face, and beds and joints smoothly and squarely dressed. QUALITY OF STONE. All stone shall be perfectly sound, strong, hard, free from in- jurious seams, and in all respects satisfactory to the Engineer. Stone to be such as APPENDIX. 137 can be truly wrought to such Hues and surfaces, whether curved or plain, as may be re- quired. No stone shall be used which weighs less than 135 pounds to the cubic foot. SAMPLES OF STONE Each bidder must deposit at this office, all charges prepaid, before the bids are opened, a 6-inch cubical block of the stone he proposes to furnish, and state the quarry from which it was obtained. The quality of the stone must be at least equal to that of the sample. The sample must be truly squared, and dressed on four sides ; one side must be hammer-dressed, one side smooth-dressed and rubbed, and one side pitch-dressed. The other two sides are to be left with quarry face. STONE MAY BE REJECTED. The United States reserves the right to reject any stone not deemed suitable, or which, during the execution of the contract, shall be found defective. The beds of the stone must be their natural quarry beds. No lewis or dog holes, letters, or marks of any kind will be allowed on any dressed face of stone, but each face shall have left on it a boss for lifting, to be removed by the contractor after the stone has been set. DRESSING OF STONE. Stone must be accurately cut, square and true, and the faces must be pitch draughted and point-dressed to a plane with the draught, forming an approximately smooth surface. The beds must be smoothly and squarely dressed, full length and width. The vertical joints must be dressed to a depth of not less than 18 inches from the face, and the allowance for joints must not exceed 3/8 inch. One- third of the stone in each course must be headers. All stones not accurately dressed will be rejected. All dressed stone must have the dimensions plainly marked on one end. DIMENSIONS. The cut-stone stretchers must be not less than 3 feet nor more than 5 feet long, and their width must be not less than \\ times the height of the course to which they belong. The width of the headers must be not less than \\ times their height, and their length must be at least double their breadth, unless otherwise ordered. The thickness of courses includes the joint, which will be 3/8 inch. . LAYING STONE MASONRY. The faces of the walls shall be accurately laid to the lines indicated on the drawings, or as, directed by the Engineer. All stones to be well laid to proper lines, in full beds of mortar, and settled in place with a wooden maul ; the use of grout is prohibited. No dressing, except in special cases, and by permission of the Engineer, will be allowed on backing after it is laid in the wall. The bond of stone shall in no case be less than 9 inches. The walls will be laid in horizontal courses throughout, each course to be of uniform height through the wall. Heights and arrangements of courses to be determined by the Engineer. When laying masonry the site for the stone shall be thoroughly cleaned with a scrub broom and moistened ; and the stone shall always be cleaned and well moistened before being set. Not more than three unfinished courses of face stone will be permitted upon the wall at the same time, without special permission from the Engineer in each case. Proper machinery must be used in handling the stone , face stone shall not be disfigured by use of plug or grabs. Any stone chipped or spalled shall be rejected. Stones having defects con- cealed by cement or otherwise will be rejected on that account alone. COPING. The coping will be of the same class and quality of stone described in ashlar masonry. It will be carefully and truly cut to forms and dimensions given, from the best stone , it will be cranda'lled on all outer faces ; the exposed edges of the coping to be rounded to a radius of 3 inches and chiseled smooth where required. Beds and vertical joints to be pointed true and full throughout and be laid with 3/8- inch joints. The coping is to be doweled as required by the Engineer with round iron. The dowels to be furnished and placed by the contractor. The drilling for and placing of the dowels will be covered by the price for " Bolt Holes in Masonry." The dowels will be set in Portland cement. RUBBLE STONE. QUALITY AND DIMENSIONS OF. Rubble stone must be sound, hard, and durable, free from seams, scale, earthy matter, and other defects. Rubble stone shall in gen- eral be not less than 3/4 of a cubic foot in size. It must be in fair shape for laying in the face of the walls without dressing. No spalls will be allowed. LAYING. The stone must be laid on their natural bed in full beds of hydraulic 138 APPENDIX. cement mortar, with all joints and voids well filled with mortar. Leveling up under stones with small chips or spalls will not be allowed. The stone shall be carefully selected for the outer face so as to have vertical joints and present a good face of broken rough masonry. CONCRETE. COMPOSITION OF. Concrete shall be composed of satisfactory cement and river gravel ; the latter, should it be of an approved quality, shall be taken from the various excavations of the lock and its walls. This gravel generally has a sufficient volume of sand to fill all voids , should there be a deficiency of sand in any portion of the gravel the contractor will be required to supply said deficiency by good, sharp, washed, river sand. The quantity of cement to be used will generally be about 20 per cent greater than the volume of voids in. sand and gravel. MIXING AND PLACING OF. The concrete is to be well and rapidly mixed by machinery, as may be required by the Engineer, unless otherwise specified. It will be deposited in layers not more than 8 inches thick ; wherever and whenever required, the layers will be thinner than 8 inches, and all thoroughly rammed by such process as the Engineer may approve. RIVER WALL. In the river wall of the lock the concrete shall be laid in courses of a thickness corresponding to the adjoining courses of ashlar masonry. It shall be filled in flush with the top of each course before the next course of ashlar above shall be laid. Before putting in the concrete of any course the bed and adjoining course of ashlar shall be thoroughly wetted so that no dry surface may come in contact with the fresh concrete, destroying its power of adhesion by absorbing its moisture. In order that the work once begun may progress without delay all cut stone needed for the ashlar facing shall be on the ground when the concrete foundation has been completed. TIMBER IN PERMANENT CONSTRUCTION. To CONSIST OF all timber used in the timber facing of the lock walls and the guide walls ; all timber cribbing in the gate-track and Poiree dam foundations ; the oak sheeting at the head of the guide walls ; and such other timber in permanent con- struction as shall be shown upon the drawings. GENERAL QUALITY AND DIMENSIONS. All timber must be first class, and any of inferior quality will be rejected. Sap-wood in any stick will cause its rejection. The timber must be free from black or rotten knots, wane edges, wind shakes, dose, or other imperfections. Firm sound knots, if not too numerous, will not be considered defects. Timber must be full to size, true, and out of wind, and when required must be sawed large enough to dress down to required dimensions. The timber will be inspected on arrival at the work, and if found to be defective will be rejected. OAK. Oak timber must be taken from the best quality live white oak sawed timber. WHITE PINE. Shall consist of the best quality of clear white pine obtainable. HEMLOCK. Shall be the best quality of hemlock obtainable. FRAMING, ASSEMBLING, AND PAINTING. All timber must be accurately framed, fitted, and assembled, according to detailed drawings and directions. As the timber is framed it shall be painted about the ends and elsewhere as may be required to pre- vent checking. The paints for this will be furnished and applied by the contractor, and covered in his price for " Timber in Permanent Construction " TIMBER FACING, UPRIGHTS. AND SHEETING shall be constructed of oak, and shall consist of uprights spaced at intervals of 6 feet, center to center, anchored to the con- crete masonry by tee-head screw bolts as shown on drawings. To the uprights shall be bolted, with wrought-iron screw bolts, oak sheeting 6 inches thick. NOSING TIMBER shall extend along the top of the guide wall, forming a cap to the uprights and securely bolted to them, as shown on the drawings. The top surface of the nosing shall be flush with the top of the concrete masonry wall. OAK SHEETING. This refers to the sheeting on the upper faces of the protection APPENDIX. 139 crib for the upper guiding wall at the upper end thereof. It shalf be spiked on and firmly held in place with iron bands or straps bolted to the framing timbers of the crib, if, in' the judgment of the Engineer, this may be deemed necessary. SUPERVISION AND MEASUREMENT OF WORK. INSPECTION, REJECTED MATERIAL, ETC. The works will be conducted under the direction of the local or resident Engineer, who shall have power to prescribe the order and manner of executing the same in all its parts ; of inspecting and rejecting ma- terials, work, and workmanship which, in his judgment, do not conform to the draw- ings that may be furnished from time to time, or to these specifications. And any material, work, or workmanship so rejected by him shall be kept out of or removed from the finished work, and no estimate or payment shall be made until such material, work, or workmanship be so removed. When so required rejected material shall be piled up in sight near the works and kept there until the Engineer gives permission to have it removed. The United States will keep inspectors on the work who will receive instructions from the resident Engineer. They will have power to object to any materials, work, or workmanship. Any material, work, or workmanship objected to by the inspectors shall be kept out of or removed from the finished work, unless in each particular case the objections of the inspector shall be overruled by the local or resident Engineer ; and, unless the objection be so overruled, no estimate or payment shall be made until such material, work, or workmanship be so removed. The local or resident Engineer shall have power to overrule or rescind any or all objections or decisions of the inspector. The decision of the United States Engineer Officer in charge of the works shall be final and conclusive upon all matters relating to the work and upon all questions arising out of these specifications, and from his decision there shall be no appeal. FAILURE TO PROSECUTE OR PROTECT WORKS. If at any time the contractor shall refuse or fail to prosecute the work or provide for carrying on the same as directed by the Engineer, or fail to properly protect any part of the work, permanent or tem- porary, the Engineer shall have power to employ men, to purchase or otherwise provide materials, tools, machinery, etc., and put the work in proper advancement or condition, and the entire cost of so doing shall be deducted from payments to be made under this contract. COMPLETE WORK REQUIRED. The contractor is not to take advantage of any omissions of details in drawings or specifications, or errors in either, but he will be required to do everything which may be necessary to carry out the contract in good faith, which contemplates everything complete, in good working order, of good material, with accurate workmanship, skillfully fitted and properly connected and put together. Any point not clearly understood is to be referred to the Engineer for decision. CHANGES. Should any changes in the details of the shape, arrangement, or fitting of the parts be deemed necessary or advisable in the progress of the work, they must be made by the contractor, and a fair allowance will be paid for any changes which, in the judgment of the Engineer in charge, materially increases the cost of the work. MEASUREMENT. Measurement of all work and material shall be made in place, unless otherwise specified. COFFER-DAM. The price per lineal foot of coffer-dam shall include all material, lumber, iron, and gravel entering into its construction. A profile of the location will be furnished, showing a section of the river bed over which the coffer-dam is located, so that the contractor may estimate the amount of each kind of material required. PILING. Temporary piling shall be measured in lineal feet, and measurement shall be allowed for total length of piling used. SHEETING This will include all lumber used for temporary purposes, in shoring of excavations, or for forms necessarv to sustain any concrete masonry until it has become sufficiently hardened. Sheeting required by the Engineer to remain in the finished structure shall be paid for at the contractor's price per thousand feet B. M. All temporary sheeting not remaining in the finished structure shall be included in the contractor's unit price for material in place, and no estimate will be made thereof by 140 APPENDIX. the Engineer. Coffer-dam sheeting will be included in the contractor's price per lineal foot of coffer-dam. FILLING. Gravel filling will be measured in the fill, and will not include any filling placed in the coffer-dam as coffer-dam filling. Stone filling shall include all riprap work, either temporary or permanent. EXCAVATION. Excavation will be measured in excavation by cross-sections. MASONRY. All masonry, ashlar, rubble, brick, concrete, etc., will be measured by the cubic yard in place. Prices for masonry will include all required pointing. No payment will be allowed for voids or openings. BOLT HOLES. All holes drilled in rock or concrete or other masonry will be measured by the running foot as drilled. TIMBER IN PERMANENT CONSTRUCTION. Timber in permanent construction will include all timber used in any part of the permanent construction ; unless otherwise particularly specified, will be classed under the following heads : Oak in Permanent Construction. Pine in Permanent Construction. Hemlock in Permanent Construction. APPENDIX. 141 EXTRACTS FROM TOPEKA (KANSAS) MELAN ARCH BRIDGE SPECIFICATIONS. By permission of EDWIN THACHER, M. Am. Soc. C. E. PILING IN PERMANENT WORK. Piling and lumber for coffer-dams to be sound white oak, yellow pine, or other woods equally good for the purpose, the quality to be acceptable to the superintendent. The piles shall be straight-grained, trimmed close, and have all bark taken off, and shall be at least 10 inches in diameter at the small end and 14 inches in diameter at the butt when sawed off. The heads shall be cut off squarely at right angles to the axis of the pile, and all piles shall be fitted to and driven with a cast-iron head. The piles shall be driven with a hammer weighing not less than two thousand two hundred and fifty (2250) pounds, and until they do not move more than three-eighths (3/8) of an inch under a blow of the hammer falling twenty-five (25) feet. No pile shall be driven less than twenty-six (26) feet below low water, and if necessary to attain this minimum depth jets shall be used in addition to hammer. The number and arrangement of the piles for each foundation are shown on the plans, and must be carefully carried out by the contractor. The piles shall be cut off at an elevation of about six (6) inches below low water. A slight variation will be allowed, but no piles must be cut off at a higher elevation. Inspection of piling and lumber, except at bridge site, shall be at con- tractor's expense. COFFER-DAMS. After the bearing piles have been driven, a permanent coffer-dam, of the dimensions marked on the plans, of Wakefield (or other equally satisfactory) sheet piling, shall be used around each foundation. The earth inside thereof shall be excavated to the depth shown on plans and replaced with concrete as hereinafter specified. During the placing of the concrete the water shall be kept out of the coffer-dams unless the bottom is so porous that it is impracticable in the opinion of the superintendent to do so in which case some of the concrete may be placed in position by means of chutes under the direction of the superintendent until the bottom is well calked, after which the water shall be pumped out and the remaining concrete placed in position. The con- tractor will be required to make the sides and ends of the coffer-dams water-tight, and no leak through them will be considered sufficient cause to require any concrete to be placed by means of chutes. CENTERING. The contractor shall build an unyielding falsework, or centering, of the form and dimensions shown on the plans; particular care must be taken to drive the piles sup- porting it to a solid bearing. The estimated load upon each of these piles is twenty (20) tons. The contractor must, however, satisfy himself as to the load each pile will have to bear, and as to its supporting power. In case of any settlement the contractor shall take down and rebuild the centering and arch. The lagging shall be dressed on both edges to a uniform size so that when laid it will present a smooth surface, and this surface shall be built at the proper elevation to allow for settlement of arch, so that when the centering is struck the arch ring will come to the elevations shown on plans. 142 APPENDIX. The top surface of the lagging shall be covered with W. Field's Building Paper of medium weight, known as Double Saturated Water-proof Oiled Sheathing Paper (or other equally good) to prevent the concrete from adhering thereto, No center shall be struck until at least twenty-eight (28) days after the completion of the arch. Great care shall be used in lowering the centers so as not to throw undue strains upon the arches, nor shall any center be struck before ihe adjoining arch has been completed for a sufficiently long time, in the opinion of the superintendent, to be uninjured thereby. NOTE. For the above reasons it is probable that the five centers will be in use at the same time. PORTLAND CEMENT. The Portland'cement shall be a true Portland cement, made by calcining a proper mixture of calcareous and clayey earths; and the contractor shall furnish one or more certified statements of the chemical composition of the cement and of the raw materials from which it is manufactured. Only one brand of Portland cement shall be used on the work, except with permission of the superintendent, and it shall in no case contain more than two (2) per cent of magnesia in any form. The fineness of the cement shall be such that at least 98 per cent shall pass through a standard brass cloth sieve of 74 meshes per linear inch, and at least 95 per cent shall pass through a sieve of 100 meshes per linear inch. Samples for testing may be taken from each and every barrel delivered as superin- tendent may direct. Tensile tests will be made on specimens prepared and maintained, until tested, at a temperature of not less than 60 degrees Fahrenheit. Each specimen shall have an area of one square inch at the breaking section, and after being allowed to harden in moist air for twenty-four hours shall be immersed and retained under water until tested. The sand used in preparing the test specimensshall be clean, sharp, crushed quartz, retained on a sieve of 30 meshes per linear inch and passed through a sieve of 20 meshes per linear inch, and shall be furnished by contractor. No more than 23 to 27 per cent of water by weight Shall be used in preparing the test specimens of neat cement, and in making the test specimens one of cement to three of sand, no more than II or 12 per cent of water by weight shall be used. Specimens prepared from neat cement shall after seven days develop a tensile strength not less than 400 pounds per square inch. Specimens prepared from a mixture of one part cement and three parts sand (parts by weight) shall after seven days develop a tensile strength of not less than 140 pounds per square inch, and after twenty-eight days not less than 200 pounds per square inch. Specimens prepared from a mixture of one part cement and three parts sand (parts by weight) and immersed, after twenty-four hours, in water to be maintained at 176 degrees Fahrenheit, shall not swell nor crack,, and shall after seven days develop a tensile strength of not less than 140 pounds per square inch. Cement mixed neat with about 27 per cent of water, to form a stiff paste, shall, after 30 minutes, be appreciably indented by the end of a wire one-twelfth inch in diameter, loaded to weigh one-quarter pound. Cement made into thin cakes on glass plates shall not cr^ck, scale, or \varp under the following treatment: three pats shall be made and allowed to harden in moist air at from 60 to 70 degrees Fahrenheit; one of these shall be subjected to water vapor at 176 degrees Fahr. for three hours, after which it shall be immersed in hot water for forty-eight hours; another shall be placed in water at from 60 to 70 degrees Fahrenheit, and the third shall be left in moist air. Samples of one-to-two mortar and of concrete shall be made and tested from time to time as directed by the superintendent. All cement shall be housed and kept dry till wanted in the work. Storage rooms and rooms and apparatus for the tests shall be furnished by the con- tractor, and all tests shall be made entirely at his expense, and under the direction and to the satisfaction of the superintendent. APPENDIX. 143 PORTLAND CEMENT CONCRETE. The concrete shall be composed of clean, hard, broken limestone (or gravel with irregular surfaces) and cement mortar in volumes as hereinafter described. The sand shall be clean, sharp, Kansas River sand, washed entirely free from earth and loam. If obtainable, a mixture of coarse and fine sand shall be used. Approved mixing machines shall be used. These machines must be kept clean and no accumulations of old mortar shall be allowed to form in them. The ingredients shall be placed in the machine in a dry state and in the volumes specified and be thoroughly mixed, after which clean water shall be added and the mixing continued until the wet mixture is thorough and the mass uniform. No more w r ater shall be used than the concrete will bear without quaking in ramming. The mixing must be done as rapidly as possible, and the batch deposited in the work without delay, and before the cement begins to set. Stone must be entirely free from earth and earthy surfaces. Thin splints or leaves of stone, easily broken with fingers, will not be allowed to go into the work. The quality of stone and the crushing must be acceptable to the superintendent. The grades of concrete to be used are as follows (parts by volume): For the arches: one part Portland cement, two parts sand, and four parts broken stone (hazelnut size, from one-half inch to one inch), except for the exposed faces and soffits of the arches, which[shall have at least one inch in thickness of mortar composed of one part Portland cement and two parts sand. For the piers, abutments, spandrel and wing walls: on the exposed surfaces for at least one inch thick one part Portland cement and two parts sand; for the next seven (7) inches one part Portland cement, two parts sand, and four parts broken stone of hazelnut size. For the remaining portions- one part Portland cement, four parts sand, and eight parts broken stone of size to pass through a three-inch ring; except such por- tions of the interior of the piers and abutments as are above the top of the cornice, or elevation 15.75 ft. above low water, which shall be composed of one part Portland cement, three parts sand, and six parts broken stone which will pass through a two and one-half inch ring. No plastering of surfaces will be allowed nor any practice that will develop planes or surfaces of demarkation other than those hereinafter described. Immediately after the removal of any forms or centers, sand and cement shall be sifted on the surfaces and the surfaces rubbed hard with a float as may be directed by the superintendent. During warm and dry weather and whenever the superintendent shall direct, all newly built concrete shall be kept well shaded from the sun and well sprinkled with water at the surface for several days or until well set. There must be no definite plane or surface of demarkation between the facing and the concrete backing. The facing and the backing must be deposited in the same layer and well rammed in place at the same time. In connecting old concrete with new, in the planes hereafter described, the old concrete shall be cleaned and roughened and soaked with water, and at the points of contact a mortar composed of one part cement and two parts sand shall be used and shall be laid in the same manner as specified for laying the facing. NATURAL CEMENT CONCRETE. The concrete around piles, to take the place of the earth excavated from the coffer- dams, shall be composed of one part natural cement, equal to the best Fort Scott, Kas., cement, three parts sand, and six parts of broken stone of the size to pass through a three-inch ring. This concrete may be mixed by hand on platforms adjoining the foundations and shoveled directly into the coffer-dams, care being taken to deposit it in uniform layers of about six inches each and to carefully ram each layer. 144 APPENDIX. PIERS, ABUTMENTS, AND SPANDRELS. All piers, abutments, spandrels, and wing walls shall be built in timber forms. These forms shall be substantial and unyielding, of proper dimensions for the work intended and closely pointed, and all surfaces that come in contact with the concrete shall be smoothly dressed and well oiled with linseed oil to prevent the concrete from adhering to them. That portion next to the exposed faces of the work need not be oiled, but shall be covered with oiled paper, the same as that specified for the centers. Molds, to form molding and panels, smoothly finished and well oiled with linseed oil, shall be properly placed in the forms so that the finished work will appear as shown on the plans. Extreme care must be used to place them in proper position before placing any concrete or mortar in them. CONTINUOUS WORK. The following divisions shall constitute sections for continuous work, viz.: each footing course of piers or abutments ; each pier or abutment from footing course to cornice ; each pier or abutment from cornice to springing line of arch ; each spandrel wall from keystone to pier or abutment ; each pier or abutment spandrel wall ; that portion of the piers or abutments above springing line of arch shall be considered part of the longitudinal sections of the arch previously described. Each of the above sections shall be carried on continuously night and day if nec- essary ; that is, each layer shall be well rammed in place before the previously de- posited layer shall have time to partially set. Care shall be taken to make the joints (for expansion) in each spandrel wall over piers as indicated on the plans. CONCRETE IN COFFER-DAMS. The natural cement concrete in the coffer-dams shall extend from depths marked on plans to one foot below low water. Upon this concrete the footing courses of piers and abutments shall be founded. The sheet piling of coffer-dams shall be cut off at least down to low-water mark, neatly and evenly, by the contractor before the completion of the work. APPENDIX. H5 EXTRACTS FROM KATTE'S MASONRY SPECIFICATIONS. By permission of WALTER KATTE, M. Am. Soc. C. E. EXCAVATIONS will be classified under the following heads, viz.: earth, hardpan, loose rock, solid rock, and excavation in water. EARTH will include clay, sand, gravel, loam, decomposed rock and slate, stones and boulders containing less than one cubic foot, and all other matters of an earthy nature, however compact, excepting only " hardpan," as described below. HARDPAN will consist of tough, indurated clay or cemented gravel which, in the opinion of the Engineer, requires blasting for its removal. LOOSE ROCK. All boulders and detached masses of rock measuring over one (i') cubic foot in bulk, and less than one (i) cubic yard ; also all slate, shale, soft friable sandstone and soapstone, and all other materials excepting rock, solid ledge, and those described above ; also stratified rock in layers of not exceeding eight (8") inches in thickness, when separated by strata of clay, and which, in- the judgment of the Engineer, may be removed without blasting, although blasting may occasionally be resorted to. SOLID ROCK will include all rock found in ledges, or masses of more than one (i) cubic yard, which, in the judgment of the Engineer, may be best removed by blasting , with the exception of stratified rocks described under the head of Loose Rock. In rock excavations the " bottom " must in all cases be taken down truly to sub-grade ; and when so ordered by the Engineer ditches must be formed at the foot of the slope. The contract price for excavation will apply to pits required for foundations of masonry when water is not encountered, and the price for EXCAVATION IN WATER will only apply to foundation pits under water and deep- ening of channels in running water ; it must cover all classes of material, and include drainage, bailing, pumping, and all materials and labor connected with such excava- tions ; also the necessary dressing of the rock. CEMENT must be of the best quality of freshly burned and ground hydraulic cement, and be equal in quality to the best brands of Cement. It will be subject to test made by the Engineer or his appointed inspector, and must stand a proof tensile test of fifty (50) pounds per square inch of sectional area on specimens allowed a set of thirty (30) minutes in air and twenty-four (24; hours under water. MORTAR will in all cases be made of one part in bulk of the best hydraulic cement to two parts in bulk of clean, sharp sand, well and thoroughly mixed together in a clean box of boards, before the addition of the water, and must be used immediately after being mixed. No mortar left over night will, under any pretext, be allowed to be used. The sand and cement used will at all times be subject to inspection, test, acceptance, or rejection by the Engineer. CONCRETE. Concrete shall be composed of fragments of hard, sound, and accept- able stone, broken to a size that will pass through a two (2") inch ring in any direction, thoroughly clean and free from mud, dust, dirt, or any earthy admixture whatever ; mixed in the proportion of two (2) parts in bulk of the broken stone to one (i) part of fresh-made cement mortar of the quality herein described ; and is to be quickly laid in sections and in layers not exceeding nine (9) inches in thickness, and to be thoroughly rammed until the mortar Hushes to the surface ; it shall be allowed at least twelve (12) hours to " set" before any work is laid on it. APPENDIX. FOUNDATIONS. GENERAL DESCRIPTION. Foundations for masonry shall be excavated to such depths as may be necessary to secure a solid bearing for the masonry, of which the Engineer shall be the judge. The materials excavated will be classified and paid for. as provided for in these specifications, under the general head of Excavations ; and in case of foundations in rock, the rock must be excavated to such depth and in such form as may be required by the Engineer, and must be dressed level to receive the foundation course. When a safe and solid foundation for masonry cannot be found at a reasonable depth (to be judged of by the Engineer), there will be prepared by the contractor such artifi- cial foundations as the Engineer may direct. All materials taken from the excavations for foundations, if of proper quality, shall be deposited in the contiguous embankment; but any material unfit for such purpose shall be deposited outside the roadway, or in such place as the Engineer shall direct, and so that it shall not interfere with any drain or watercourse. TIMBER. Timber foundations when required shall be such as the Engineer may by drawings or otherwise prescribe, and will be paid for by the one thousand feet, board measure. The price, covering cost of material, framing and putting in place, and all wrought- and cast-iron work ordered by the Engineer, will be paid for per pound, the price including cost of material, manufacture, and placing in the work. PILING. All timber used in foundations or foundation piling shall be of young, sound, and thrifty white oak, yellow pine, or other timber equally good for the purpose, acceptable to the Engineer. Piles must be at least eight (8") inches in diameter at the small end and twelve (12") inches in diameter at the butt when sawn off; they must be perfectly straight and be trimmed close, and have the bark stripped off before they are driven. They must be driven into hard bottom until they do not move more than one- half inch under the blow of a hammer weighing two thousand (2000) pounds, falling twenty-five (25') feet at the last blow. They must be driven vertically and at the regular distances apart from centers, transversely and longitudinally, as required by the plans or directions of the Engineer; they must be cut off squarely at the butt and be well sharpened to a point, and when necessary, in the opinion of the Engineer, shall be shod with iron and the heads bound with iron hoops, of such dimensions as he may direct, which will be paid for the same as other iron work used in foundations. The necessary length of piles shall be ascertained by driving test piles in different parts of the localities in which they are to be used; and in case a pile shall not prove long enough to reach " hard bottom " it shall be sawed off square, and a hole two (2") inches in diameter be bored into its head twelve (12") inches deep; into this hole a circular white-oak trenail twenty three (23") inches in length shall be well driven, and another pile similarly squared and bored, and of as large a diameter at the small end as can be procured, shall be placed upon the lower pile, brought to its proper position, and driven as before directed. All piles, when thus driven to the required depth, are to be cut off truly square and horizontal at the proper height given by the Engineer, and only the actual number of lineal feet of the piles left for use in the foundations after being sawn off will be paid for. COFFER-DAMS. Where coffer-dams are, in the opinion of the Engineer, required for foundations the prices provided in the contract for timber, piles, and iron in founda- tions will be allowed for the material and work on same, which is understood as cover- ing all risks from high water or otherwise, draining, bailing, pumping, and all materials connected with the coffer-dams. Sheet piling will be classed as plank in foundations, and will be paid for per one thousand (1000') feet board measure if left in the ground. TIMBER. All timber must be sound, straight-grained, and free from sap, loose or rotten knots, wind shakes, or any other defect that would impair its strength or durability; it must be sawed (or hewed) perfectly straight and to exact dimensions, with full corners and square edges; all framing must be done in a thoroughly workmanlike manner, and both material and workmanship will be subject to the inspection and acceptance of the Engineer. APPENDIX. 147 SPECIFICATIONS FOR STEEL COFFER-DAM. DESIGN. The shell shall be made of elliptical shape for ordinary piers and circular for pivot piers. It shall be made not less than four feet larger than footing of pier in plan, to allo;v for variation in position during sinking. The plates used shall be as large as can be handled with ease in the shop, during shipment, and during erection. The splices may be either lap or butt joints, provided a good tight job will result, and the rivets must be spaced according to bcilermaker's rules. The joint may be made tight by calking or by the use of a calking strip, but in either event the result must be guaranteed. The shell must be stiffened by horizontal stiffening angles, girders, or trussing, to resist deformation during the placing and to resist both the quiescent and a maximum unbalanced earth or water pressure, or a wind pressure. The bottom plates shall be re-enforced with narrow plates inside and outside, to form a wedge-shaped cutting edge; and when there is rock or hard bottom the plates shall be cut to conform to its contour as nearly as possible. The top shall be properly stiffened, and if necessary provided with connection holes for additional sections. The factor for safety shall in no case be less than four, and in case the shell will be subject to shock, not less than six. No metal of a less thickness than 1/4 inch shall be used for temporary work, nor less than 3/8 inch for permanent work in fresh water or 1/2 inch in salt water. MATERIAL. The entire shell shall be constructed of the grade of steel known as soft medium, except rivets, which shall be of bridge quality of iron. The steel may be made either by the Bessemer or open-hearth process, and the phosphorus shall never exceed 0.08 per cent. Soft medium steel shall have an ultimate strength of from 55,000 1065,000 pounds per square inch, as determined from standard rest pieces ; an elastic limit of not less than one-half the ultimate strength ; an elongation of not less than 25 per cent in 8 inches , and a reduction of area at fracture of not less than 50 per cent. Samples to bend cold 180 degrees to a diameter equal to the thickness of the sample, without crack or flaw on the outside of the bent portion. ERECTION. The erection must be done in a first-class manner, and all rivets must have full heads. The shell shall be placed in position within one-half the distance allowed for error in the design of the coffer-dam. Only a reasonable variation will be allowed for difference in level. PAINTING. All the metal work shall be thoroughly cleaned of rust or scale at the shops and coated thoroughly with hot asphaltum. Before erection, in the field, it shall be given a second coating of hot asphaltum. SEALING. When in position on the bottom, if the coffer-dam has not been sunk through impervious strata, it shall be sealed by concreting around the circumference inside with concrete passed through a tube. REMOVAL. Should the coffer-dam not form a part of the permanent foundation it shall be taken apart, at the joints designed for ihe purpose, and carefully removed in such a manner as not to injure the foundation, and so as to be used again if required. I4 APPENDIX. HEALD & SISCO STANDARD IRON HORIZONTAL CENTRIFUGAL PUMPS, Horse-power No. Capacity in Gallons per Minute. required for Each Foot of Lift. Minimum Diameter and Face of Pulley in Inches. Floor Space required, in Inches. Shipping- Weight. Pounds. Price of Pump, Oilers and Wrench. Price of Pump and Primer. No. Quantity. It 50 to 70 .024 6X6 17 X 30 168 $45 $55 I* If 75 to 100 037 7X8 21 X 33 232 60 70 If 2 no to 150 054 8 X 8 23 X 37 306 75 90 2 2| 175 to 250 .086 8X8 24 X 38 348 90 105 at 3 250 to 350 .124 8X8 25 X 39 400 no 130 3 4 450 to 600 .223 10 X 10 30 X 40 545 130 155 4 5 750 to 900 372 15 X 12 34 X 54 826 165 195 5 6 1000 to 1400 .496 15 X 12 37 X 55 965 200 240 6 8 I7OO to 22OO .844 2O X T2 45 X 63 1500 310 375 8 10 22OO to 3OOO 1.093 24 X 12 5i X 71 2170 395 470 10 12 3000 to 4000 1.49 30 X 14 62 X 78 3050 500 12 15 4800 to 6000 2.38 40 X 15 77 X So 7100 850 15 is 4800 to 6000 2.38 30 x 15 60 X 68 3150 710 15 18 7500 to loooo 3-73 40 X 15 93 X 103 9000 1300 18 *i8 7500 to 10000 3-73 30 X 16 62 X 70 3500 1150 18 22 I2OOO tO I4OOO 5-96 48 X 20 126 X 130 12000 22 * Refers to low-lift pump. The number of pump is also diameter of discharge opening in inches. Where more than 25 feet of discharge pipe is attached to pump, use one or two sizes larger than pump discharge. For No. 12 and larger sizes a foot valve or flap valve and ejector for priming is recommended. LIST OF HEALD & SISCO HYDRAULIC DREDGING AND SAND PUMPS. Num- ber of Pump. Diameter Suction and Dis- charge Open- ings. Cubic Yards of Material they will Raise per Hour Horse- power recom- mended for 10- Foot Diameter and Face of Pulley. Floor Space Required. Inches, Shipping Weight. Pounds. Will Pass Solids, Diameter. Inches Price of Pump Complete, with Suction and Discharge Elbows, Flap Valve and Num- ber of Pump. Inches. Lift. Ejector. 4 4 30 to 50 6 12 X 12 40 x 31 800 2 $2IO 4 6 6 60 to 80 12 20 X 12 68 X 40 I7OO 4i 3OO 6 8 8 125 to 150 22 24 X 14 72 x 48 3400 6 475 8 10 10 2OO tO 3OO 35 30 x 15 94 X 54 4200 8 600 10 12 12 300 to 375 45 36 X 20 114 X 66 9OOO 10 850 12 15 15 400 to 500 75 42 X 24 154 X 78 I2OOO 10 1450 15 18 20 18 2O 500 to 700 125 48 X 30 160 X 80 13500 IO 1900 18 20 22 22 .... 22 APPENDIX. 149 NUMBER OF REVOLUTIONS AT WHICH PUMPS SHOULD RUN TO RAISE WATER TO DIFFERENT HEIGHTS. No. 5 Feet. 10 Feet. 15 Feet. 20 Feet. 25 Feet. 30 Feet. 35 Feet. 40 Feet. I* 428 604 739 854 955 1045 H3I 1208 If 348 491 60 1 695 777 850 920 982 2 3O2 426 522 60 3 674 737 798 8 5 2 2* 302 426 522 60 3 674 737 798 8 5 2 3 302 426 522 60 3 674 737 798 8 5 2 4 285 402 493 569 637 697 754 805 5 256 362 443 512 572 626 678 724 6 214 302 368 427 478 523 566 604 8 183 259 317 366 409 448 485 517 10 168 238 291 336 376 411 445 475 12 133 188 230 266 298 326 352 376 15 105 148 181 2OQ 234 256 277 295 *I5 151 213 261 301 337 369 399 426 18 105 148 181 209 234 256 277 295 *iS 151 213 261 301 337 369 399 426 * Refers to low^lift pumps. Above table gives correct speed of pumps as employed under usual conditions of pumping. If water must be forced through a number of bends and elbows, or a great length of piping, the above speed must be somewhat increased. Use large pipes and easy bends wherever practicable, as they save power. TABLE OF SIZES, LIDGERWOOD SINGLE-CYLINDER, SINGLE-DRUM HOISTING-ENGINES. d u Dimensions of Cylinder. 3 w , >> _ f.% 0^ B X Diameter. Inches. . ti ji-g 2JS p 11 Diam Bod between Flanges. Inches. So^jS <> Q Width. Inches. tl Diameter Shell. Inches. Height She Inches. Number of 2-inch Tubes. Estimated Weight C Lbs 4 5 8 I2OO 1000 ,0 20 22 38 60 28 63 40 3550 6 63- 8 I5OO 1250 10 20 22 38 60 28 69 40 3950 8 6|- IO 1750 1500! 12 20 24 41 73 30 72 44 4850 IO 7 10 2500 i8oo ! 12 2O 24 41 73 32 75 48 5050 n 7 10 2500 2000 14 22 26 45 73 34 78 52 5350 12} 81 IO 4000 2500; 14 23 29 47 73 36 75 57 6550 15 IO 4OOO 28OO 14 23 29 47 73 36 81 57 6750 20 8* 12 6OOO 4000 1 6 26 33 54 84 40 84 80 8500 25 IO 12 8000 5000! 16 26 33 54 84 42 90 88 9500 ISO APPENDIX. TABLE OF SIZES, LIDGERWOOD DOUBLE-CYLINDER, DOUBLE-DRUM HOISTING-ENGINES. a aj Dimensions of Cylinders. Dimensions of Hoisting-drums 11 w be tf|* Dimensions of Boiler. Dimensions of Bed-plate. .9- ^ u & o o(/> ]?= g ~ rt a 3 Diam. Stroke. Diam. Length. "bVbflw "re^ 1 B.y Diam. Height. ber of Width. Length E CO Inches. Inches. Inches. Inches. jj > -<<-i 2 .2 Inches. Inches. 2-inch Inches. Inches. E D jgj