TRANSACTIONS OF THE AMERICAN SOCIETY op CIVIL ENGINEERS (INSTITUTED 1852) VOL. LXVII JUNE, 1910 Edited by the Secretary, under the direction of the Committee on Publications. Reprints from this publication, which is copyrighted, may be made on condition that the full title of Paper, name of Author, and page reference are given. NEW YORK PUBLISHED BY THE SOCIETY I9IO It Avery Architectural and Fine Arts Library Gift of Seymour B. Durst Old York Library Entered according to Act of Congress, in the year 1S10, by the American Society ok Civil Engineeks. in the Office of the Librarian of Congress, at Washington. Note.— This Society is not responsible, as n body, hw the facts and opinions advanced in any of its publications. CO NT K NTS PAPERS NO. Y PAGE 1148 NOTES ON THE REPLACING OF THE SUPERSTRUCTURE OF THE HAR- LEM RIVER SHIP CANAL BRIDGE. By Horace J. Howe i Discussion: By Lincoln Bush 23 Martin Gay 24 St. John Clarke 20 Theodore Belzner 27 Frank W. Skinner 28 Horace J. Howe SO 1141 PRECARIOUS EXPEDIENTS IN ENGINEERING PRACTICE. By John Hawkesworth 32 Discussion: By Eugene W. Stern 46 W. W. Crosby 48 J. S. Branne 50 Andrews Allen 52 Guy B. Waite 55 J. H. Gandolfo 57 John Hawkesworth 59 1145 WATER SUPPLY FOR THE LOCK CANAL AT PANAMA. By Julio F. Sorzano 01 Discussion: By C. E. Grunsky «.)1 H. F. Hodges 96 Theodore Paschke 100 Allen Hazen 108 CM. Saville 112 Julio F. Sorzano 133 1140 THE IMPROVED WATER AND SEWAGE WORKS OF COLUMBUS, OHIO. By John H. Gregory 206 Discussion: By Joseph W. Ellms 324 Julian Griggs , 325 C-E. A. Winslow 328 R. D. Scott and R. F. McDowell 330 Samuel Tobias Wagner , 334 J. Corbett 385 W. R. Copeland 387 W. A. Sperry 342 J. W. Sale 845 C. P. Hoover 854 C. B. Hoover 359 A. Elliott Kimberly 379 IV No. PJL0I 1146 Discussion (Continued): Samuel Rideal 388 Alexander Potter 390 Gilbert Fowler 394 William Gavin Taylor 397 Allen Hazen 399 William B. Fuller 400 Emil Kuichling 402 Langdon Pearse 406 Rudolph Hering 407 George A. Johnson 415 George W. Fuller 419 John H. Gregory 426 1117 AGREEMENTS FOR BUILDING CONTRACTS. By William B. Bamford 438 Discussion: By Norman R. McLure 478 William V. Polleys 479 Charles H. Higgins 481 W. G. Wilkins 483 William A. Boring 486 J. C. Wait 488 W. L. Bowman 491 Charles A. Ruggles 493 Oscar Lowinson 496 E. T. Thurston, Jr 497 John Mason Brown 502 J. A. L. Waddell BIO DeWitt V. Moore 515 Leslie H. Allen 520 William B. Bamford 525 His UNDERPINNING THE CAMBRIDGE BUILDING, NEW YORK CITY. By T. Kennard Thomson 553 Discussion: By Henry (>orton Opdycke BH E. A. Yates 566 Ogden Merrill 567 Oscar Lowinson 568 J. C. Meem 569 T. Kennard Thomson 570 1149 THE EFFECT OF ALKALI ON CONCRETE. By Georjje Gray Anderson 572 Discussion: By R. A. Hart 587 Rudolph Hering 592 Richard L. Humphrey 598 George F. Morse 601 Richard H. Gaines 602 Thomas H. Means 606 F. E. Robertson 608 Philo EL Bates 608 E. P. (Joodrich 611 W. E. Belknap 612 ( J eorgk (»ray Anderson 617 V MEMOIRS OF DECEASED MEMBERS PA(iK Hknry Furlong Baldwin. M. Am. Soc. C. E 621 Clifford Buxton, M. Am. Soc. C. E 623 Wilson Crosby, M. Am. Soc. C. E 62."> John Hall Emigh, M. Am. Soc. C. E 62K Henry Cyprian Humphrey, M. Am. Soc. C. E 880 Henry Maynadier Steele, M. Am. Soc. C. E 681 John Joseph Horan, Assoc. M. Am. Soc. C. E 684 PLATES PLATE PAPER PAGE \. South Span. Harlem Ship Canal Bridere, Being Made Ready for Final Move, and North Span at Beginning of Erection 1143 5 IF. Old Draw-Span Ready for Final Move, and Old and New Draw- Spans 1148 9 HI. New Draw-Span in Readiness to Move, and Present Harlem Ship V Canal Bridge 1143 11 IV. Coteau Bridge, and Brunot Island Bridge on Falsework Floating on Barges 1113 29,^" V. Topographical Map, Showing Ridge at Escoval, Canal Zone 1145 120 VI. Map of Part of Lake Gatun, Canal Zone 1145 165 VII. Views of Dam, and of Scioto River Pumping Station and Water Purification Works, Columbus. Qhio 1146 213 VIII Scioto River Pumping Station and Water Purification Works: Plans Showing Location of Structures 1146 221 IX. Scioto River Pumping Station and Water Purification Works: General Plan, Showing Arrangement of Structures and Piping. . 1146 227 X. Plan of Scioto River Pumping Station 1146 829 XT. Scioto River Pumping Station: Section Through Engine-Room. Looking East 114G 231 XII. Water Purification Works: Floor Plans of Main Building 1146 283 XIII. Water Purification Works: Substructure of Head-House 1146 235 XIV. Interior Views of Filters and Filtered- Water Reservoir, Showing Construction, and Sciolo River Pumping Station 1146 239 XV. Plan and Sections of Lime Saturators 114<> 241 XVI. Plan and Sections of Mixing Tanks 1146 243 XVII. Plans and Sections of O te-House and Main Dividing Wall of Settling Basins 114(5 245 XVIII. Details of Filter Gallery 114H 217 XIX. Details of Filters 1146 24ft XX. Details of South Filtered-Water Reservoir 1146 251 XXI. Interior Views of Filter Gallery, Chemical Mixing Room, Lime Saturator House, and View of Settling Basins 114(i 858 XXII. East Side Sewage Pumping Station: Plan and Section Showing Machinery .' 1144) 281 XXIII. General Plan of Sewage Purification Works 114(i 888 XXIV. Sewage Purification Works: Location of Structures and Plan of Pipes and Conduits 114C 289 XXV. Plans and Sections of Septic Tanks 114H 291 XXVI. Views of Primary and Secondary Septic Tanks 1146 293 XXVII. Sections of Gate-House 1146 295 V I PLATE PAPER PACE XXVIII. Plan of Sprinkling Filter No. 3 1146 2*>7 XXIX. Details of Sprinkling Filters 1146 2W) XXX Views of Sprinkling Filters, During Construction 1146 301 XXXI. Plan and Sections of Settling Basins 1146 303 \ XXII. Views of Sprinkling Filters Nos. 1 and 8, and of Septic Tanks 1146 305 XXXIII. Details of Pump-House 1146 307 XXXIV. Views Showing Settlement of Street, Etc., in Thirty-third Street. in Front of Cambridge Building 1148 555 XXXV. Views Showing Cracks in Wall and Ceiling of Wain Floor, and Settlement in Area, Cambridge Building 1148 557 AMEKICAN SOCIETY OF CIVIL ENGINEERS I N S T I T U T ED 1858 TRANSACTIONS Paper No. 1143 NOTES OX THE REPLACING OF THE SUPERSTRUC TURE OF THE HARLEM SHIP ( ANAL BRIDGE.* By Horace J. Howe, M. Am. Soc. C. E. Wnii Discussion by Messrs. Lincoln Bush, Martin (Ian. St. John Clarke, Theodore Belzner, Frank YY. Skinner, and Horace J. Howe. The highway in upper Manhattan which was known as King's Bridge Road in Washington's time was the only connection by land from New York City to Albany and Boston. Going north from Fort Washington, its location is easily surmised to-day, winding along the base of the foot-hills to the present Ship Canal, thence branching to the right, over Farmer's Bridge (built in 1759), and then to the left and over King's Bridge (built in 1693), to Westchester County, cross- ing the tidal stream early named Spuyten Duyvil Creek. In the celebrated retreat to White Plains (1776), a dozen miles to the northeast, these bridges were useful; and the British likewise found them so, on their return from that battle ground, in their rapid march to capture Fort Washington. These operations were in a way typical of the border warfare in this vicinity during the ensuing seven years, or until the evacuation of the city (1783), and no doubt tested the bridges severely as to their loading and as arteries of' traffic. King's Bridge was 24 ft. wide and Farmer's Bridge 18 ft. wide. Happily, such conditions did not recur, and it was not until 1900 that a third bridge over the creek at Broadway came to their assistance. * Presented at the meeting of January 5th. 1910. 2 REPLACEMENT OF THE HARLEM SHIP CANAL BttlOGE Fk;. L, REPLACEMENT OF THE HARLEM SHIP ( ANAL BRIIXiE Owing to the march of progress, all three will be taken down in the near future, and the former stream will be filled in completely. * This is due primarily to the excavation by the United States Govern- ment of what is known as the Ship Canal (1894), a waterway which connects the Harlem and the Hudson, saves distance to points on the Harlem, and enables 1 000-ton barges to cut across to the East River and avoid the dangers of the trip around the Battery. It required a swing bridge with equal waterways of 100 ft. in the clear, and at each end an approach span 100 ft. long at King's Bridge Road, or Broadway, as it is now called. The moving of this recently built bridge and the substitution of the present one is the subject of this paper. The change was agreed to by four interested parties, and resulted, in short, in a bridge with: Three elevated tracks for the Interborough Rapid Transit Company; two surface tracks for the Metropolitan Street Railway Company; an additional track underneath (four in all) for the New York Central and Hudson River Railroad; and a roadway increased to 35 ft. in width, and sidewalks increased to 7 ft. in width. The City of New York, furthermore, took over the old superstruc- ture for use at University Heights, a mile up the river. The work was done by John B. McDonald (Contract No. 1 of February 21st, 1900), under the direction of the Board of Rapid Transit Commissioners. Permission from the War Department was duly obtained, subject to the condition of maintaining for the new spans at all times a clear waterway, 250 ft. wide and 15 ft. deep, and a minimum clearance of 24 ft. above high water of spring tides. Permission was obtained also from the Bridge Department of the City of New York, which considered the neighborhood carefully, and issued a permit allowing a "closing of the said Ship Canal Bridge to street travel for a total time of five days of which not more than three days shall be consecutive." , Thus the preliminaries were arranged, but direct contract work was delayed until 1905, which permitted a study of the situation in detail. The Harlem River is an inland stream, about 6 miles long, subject to tidal influences from both ends — at the west via the Narrow? and the Hudson River, and at the east via Long Island Sound and I Ml I REPLACEMENT OF THE II WM.K.M SJ11L' CANAL BRIDGE Gate. The mean rise near the bridge is 4.1 ft., and the extreme range fs perhaps 10 ft. The velocity of the current at mid-tide is somewhat more than 2 miles an hour; hut the tide and velocities are very variable and not to be anticipated. There is considerable summer traffic on the river. Soundings were taken at various points, and Messrs. Terry and Tench, the sub-contractors, decided to construct falsework for the new draw-bridge off shore at 21Gth Street (Manhattan), i mile above the site, 221st Street, and to construct the approach spans on shore near Broadway. It was thought that dredging would not be necessary. Pile-driving began at 216th Street, in May, 1905. There were 15 bents of 10 piles each, and 3 bents of 15 each at the center, all being spaced carefully with relation to the floor-beams and the pivot. Besides, two clusters were driven, one at each outside end, for protection against river craft, and some others to connect with the bank. The piles were of good size, as long as 55 ft. on the deeper side, and were driven practically to refusal, penetrating from 10 to 20 ft. into the river bed, but not to bed-rock. The capping and bracing were finished in September. South Approach Span. — Meanwhile, the framing, bracing, and deck- ing of the bents for the south approach span was completed by August 1st, and erection begun. The total force at this time was 50 men, using two scows (100 By 30 ft.), and a well-equipped steam lighter with a steel derrick. The riveting was completed by October 1st, and the special work for the surface tracks by October 14th. There were 14 000 field rivets. The general scheme was to move the old span out and the new one in, simultaneously, by land and water. The span, having been built with a panel projecting over the bulk- head, was at this time resting at that end on the cribbing of the scow, and at the other end on two sets of cribs on railroad trucks and tracks. All masonry parapets interfering with the sliding of the new span eastward to place were removed, anchor-bolts were cut, and steel wedges inserted, and a double track was lined and surfaced parallel to the bulkhead, and at its level. On October 17th, the new span, weighing 330 tons, was moved diagonally northwest about 50 ft. to a point 115. ft. west of its final position, thereby providing room for the two scows to act in tandem. (Fig. 1, Plate I.) All end posts were then freed, and the span was sup- PLATE I. TRANS. AM. SOC. CIV. ENGRS. VOL. LXVII, No. 1143. HOWE ON HARLEM SHIP CANAL BRIDGE. Fig. '2.— North Span. Harlem Ship Canal Bridge, at Beginning ok Erection REPLACEMENT OF THE II AIM. I . M SHIP CANAL BRIDGE ported at intermediate panel points by a scow at the north end and the two sets of trucks at the south end. On October 19th, at low water (5 a. m.),' the old approach span, weighing 240 tons, having boon cribbed up on trucks at one end, was wedged up on the scow at the other end. At 8 a. m., the north end rose clear of the pier. Some delay ensued, and it was not until 10.45 a. if. that the strain was put on the wire ropes from the drums of the stationary engine. Two pulley blocks were used in connecting with the trucks of the old span, and one three-sheave block with the new. Both were adequate, and the entire mass was moved out easily, until the old span was entirely clear of the masonry. Owing to the settle- ment of the tracks by this move, and to the fact that the tide had risen, the 4-in. clearance of grillage beams and abutment was reduced to scraping distance, and for a few minutes the success of the last half of the moving was a matter of conjecture. However, the span was moved in 14 min. A temporary plank roadway was begun at once, and the first vehicle passed over the bridge at 5.45 p. m. At noon the Barber Asphalt Com- pany began concreting the buckle-plate floor, and completed the east roadway on October 25th. North Approach Span. — The erection of separate cribs for the new 112-ft. span was begun on September 5th, 1905, about 300 ft. east of Broadway. (Fig. 2, Plate I.) The trusses were at 39-ft. centers, and there were five panels, the same as in the south span; the total weight was 380 tons. Cribs were erected under alternate floor-beams, leaving one panel overhanging the bulkhead. The foundation course was com- posed of twelve 10 by 16-in. pieces laid side by side; the second course was four 16 by 16-in. pieces, square with the first, and the top half of each crib was made up of three 12 by 12-in. pieces of diminishing lengths. The level of the top was made 10 in. higher than the ultimate position of the bottom of the floor-beams. Erection was begun on September 12th, and took two weeks. Then riveting began, and was finished on October 2Sth. There were 1G 000 field rivets, practically all driven by power. Meanwhile, the New York Central and Hudson "River Railroad Company completed the new north abutment. During this time, much care was taken in lining, squaring, and leveling the trusses .and floor-beams. Owing to the direction of the G i!i:rL ack.m i:\ r of j in; ii aulem ajiu* canal biuduk shore line, it was impossible to erect the span exactly square with its final position at Broadway, and it was necessary to roll it and float it 20 ft. to the south. *To do this, four tracks were laid, a pair of trucks was placed under each north post, and the weight was deposited on them. A scow, as before, took the other end, and, by the aid of a stationary engine with its lines, the move was easily made, and the two south panels extended over the water. There was not room enough for another scow, and so the weight was again dropped on the original cribs and the scow floated off. This left the span exposed to river craft. At the next ebb tide, the second panel point was supported by a scow. Likewise, the weight at the north end was transferred at its second panel point, and all was made ready for a run westward. There were two tracks on the shore end. On November 2d the span was moved westward to near Broadway by using a stationary engine and deadman. The next step — the track extension westward — was a puzzling problem, for the channel was only 40 ft. wide between pier and bulk- head, and the latter curved sharply to the south, west of Broadway, and it was important to insure the different clearances of the scows. Finally, a track of 200 ft. radius was laid out, carefully eased off at the approach, super-elevated and braced against the neighboring bank. It was foreseen that the cribs would have to turn on the trucks, m a manner similar to that of a car body, and, to provide for this, two iron bearing plates, well greased, were set in place, but not fastened. The north end of the old span required three trucks, one under each chord, and the new span required two double trucks, a pair under each point of support. All walls, anchor-bolts, steel wedges, etc., were regulated as before. On November 6th at low water (10.30 a. m.), a scow was floated under the south end of the old span, and 3 ft. west of the other scow. The clearance of the northwest corner of the pier was 8 in., and of the northeast corner 18 in. A stationary engine had been placed 200 ft. west, and lines had been passed to the trucks of both spans and attached. The derrick- lighter was placed east of both spans and connected with a deadman west of the bridge. It was calculated that the lighter would counteract i In' netion of the tide against the scows, and in general would preserve proper relations with the trueks. REPLACEMENT OF THE II MM. KM Mill' CANAL BRIDGE 7 The combined effect of the pumps and the tide caused the old span to rise clear of the pier at 2.00 p. m. An hour or so later, levels showed the new span to he an average of 10 in. ahove the bridge seat. At 3.25 p. m., the signal was given to move, and in 2 min. the com- bined mass had moved 70 ft., and the westerly truck was at the west end of the curve. An examination of the bearing plates at the west trucks of the old span showed that the upper plates had slid along the lower ones and were nearly off; also, that the axles of the east trucks under the old span, and of all the trucks of the new span, were bending or bowing up in the middle, under the excessive strain. This condition was partly caused by the undue pushing of the lighter. A halt was called, the cribs were jacked up, and the old span was moved separately to a clearance point, shortly after 5.00 P. M. At 5.45 p. M. the new span was moved to place, the lines were cast off, and pumps set to work. By 7.00 p. m. the south end was at rest. The north end was landed by using GO-ton jacks, after a hard struggle lasting until midnight. The liquid seemed to have affected the packing and caused leaks, while the tide caused much variation in the cribbing from time to time. The bridge had been closed to vehicles from 1.30 to 10.15 P. m., and to foot passengers from 3.00 to 5.00 p. m. Planking had been laid previous to moving. A week later, the old north approach span was loaded on two scows, and, with the help of two tugs, was floated to the south shore, west of the bridge, rolled on tracks far enough to clear the canal, and then cribbed up until required, a year later. The asphalt roadway was completed on November 17th, 1905. The Neiu Draw -Span. —The erection of the new draw-span began early in December, and continued until the strike of January 1st, 1000. Union labor had been getting $4.50 for 8 hours' work and wanted $5.00. The strike lasted until February 12th, when a compromise was effected. Little of importance occurred meantime. The weather was particu- larly fine, but later became an element of expense. Some temporary wood bracing was put into the old approach spans for protection, and odd jobs were attended to. Also, the New York Central and Hudson River Railroad began running trains under the north span. On February 14th, power-riveting began on the roadway stringers and floor-plates of the motor platforms. The force was soon increased 8 REPLACEMENT OF TIIK HARLEM SHIP CANAL BRIDGE to 45 men. All cross-girders had been lined and leveled some weeks before, but required constant examination. The concentration of load- ing had been near the middle bents, but, of course, decreased as the top work advanced. Two derrick-lighters were moved about the structure on both sides, and bumped the piles more or less; and the stationary engine and riveting guns added somewhat to the vibrations. On February 24th, the new south span was connected with (he viaduct south of it. The temporary terminal station at 221st Street was opened a month later. Cracks having been discovered in the 27 by $-in. connecting angle plates of the pivot and loading girders, new ones were ordered, and arrived early in March. After much cutting out of all rivets affected, the girders were riveted up with 1-in. rivets as before. It was decided that the material was good, but having been shipped in an exposed manner, attached to the girder, the damage had occurred in transit. The maximum force was 70 men, including 8 riveting gangs of 4 men each. In this span 122 000 field rivets and 7 000 bolts were used. By the end of March the main trusses, with sway and lateral bracing, were erected. The eye-bars had been easily connected, as the ends of the trusses had been raised 1J in.; but the cross-strut at the king-posts was a tighter fit. The adjustment of the machinery in connection with the electric motors proceeded until the last of April, and from this time until the middle of June, the draw-span remained on the falsework. Moving the Old South Approach Span. — On May 1st, 190G, the old south approach span, weighing 240 tons, was floated to University Heights on two scows, after having been run off the bulkhead on three tracks. There was some delay at one corner in detaching it from the shore crib. The contractors ordered fresh jacks, and the crib timbers wore cut and slashed in order to drop that end clear of its support. This was a typical operation. It was high tide when the destination was reached, and the scows wore straddling the end draw-pior. Rollers were ready on the pier, and there was no delay in working the span westward to the temporary cribbing on the bulkhead; whence, a few days later, it was rolled still farther west to the final position on three well-braced columns. Work al the Old T)rUw-Span. — The first thing done in the work at the old draw-span was tho chocking of tlio eye-ban of tlio top chord Fig. 2.— Old Draw-Span Floating hy New Span, en Route to University Heights REPLACEMENT OF THE HARLEM SHIP CANAL BRIDGE 9 in order to provide for compressive strains while in transit to University Heights. This was early in May, 1906. As the new draw was to rest on a center pivot, and was to have only a few balance wheels or "tippers" on the outside rim, it was necessary to dress the masonry down, and furnish filler plates to take the new casting (8 ft. 6 in. in diameter), and transfer the weight to the pier; but, as the old rack was to gear with two pinions with vertical shafts, as before, it was important that the center of the rack and the center of the pivot should correspond. Precise measurements were accordingly taken inside the old drum and resulted in finding and referencing the true point. This was then transferred by transit sidewise, up over the drum and thence to the roadway, where it became the governing point in lining up the spans. Late in May, the old anchor-bolts were cut, and the center-pin and casting were suspended, together with the radiating rods and struts. The old rack was left in place, and a new one was put in at University Heights. One pinion was disconnected. The old wooden trestle, used as an oiling platform, was torn down, as it was designed to oil the new shafting while the draw was closed. Four submarine armored cables were laid across the channels in a trench, up the center pier, and through drilled holes in the coping. Meantime, electric motors had been installed at the 216th Street site, with a view to the immediate operation of the bridge on its arrival at Broadway. All riveting had been completed, and the compressors had been removed. As it was necessary to use four scows to carry the draw, some of the piles of the platform had to be removed. This left two bents at each end and five bents under the center, supporting the entire weight of the draw and the platform. The removal of the piles was completed on May 22d. ■ Levels, taken two days later, showed that there had been no settlement. The total settlement for the year, however, had been 3 in. On the new draw as well as the old, the true center of the pivot was carried up precisely, and located on the roadway. Measurement to the curved end faces showed a discrepancy of i in., an error not at all serious in its relation to the clearances of the approach spans. Temporary plank sidewalks and half of the roadway were laid. In the early part of June, as the days went by and the impending move was not made, it was reported that the draw was creeping south 10 REPLACEMENT OF THE HARLEM SHIP CANAL BRIDGE ward on its falsework, now much reduced, as stated. Observation showed that some of the piles were out of plumb, and that the caps were sagging on the center bents. Fixed lines by instrument were established on shore, and sights were taken at regular intervals for the next week or more; but only temperature variations were to be seen. The average load on each pile was 12 tons. There was no disposition to delay the work, it may be imagined, particularly as the center pier at University Heights was completed; and operations were prosecuted with great energy by Superintendent J. F. Sullivan and his force of 50 men, up to June 14th, when the old draw was moved up the river to its present location. On that day at low water (10.15 a. m.) it was swung for the last time at Broadway through an angle of 35°, lining up northeast and southwest, and just clearing the approaches. Two scows with well-braced cribbing were floated under each arm and placed parallel to the channel, after allow- ing for the proper clearances at 207th Street. The suspended drum, etc., was braced and guyed to the adjoining scows. The weight was 900 tons, exclusive of cribbing, and the displacement of each scow was rated at 500 tons. (Fig. 1, Plate 11.) The scows had 4 ft. of water inside, and after the wedging and blocking up had been done, pumping was commenced by the aid of steam from four tugs, two lighters, and a boiler-scow. There were three 2-in. streams from the north and south scows, and two streams and an 8-in. centrifugal, from the other scows. 12.30 p. If. — The rim wheels were barely off the circular tread, and the tide had risen nearly 2 ft., as noted on the neighboring tide gauge of the Dock Department. 1.45 P. M. — The scows wore 3 ft. out of water, and the depth of water in the hold was 1| ft. 2.15 p. M. — The lower struts or arms to the rack (which had been sawed off) were clear, except at the west side, where much wedging and jacking was being done. 2.35 p. M. — All clear; the bridge drifted eastward, grazing first one approach and then the other. 2.45 p. m. — The lower struts cleared the southeast corner of the circular rack, and the tugs signaled "Go ahead." The bridgemen, being no longer needed at jacks, lines, etc., were relieved temporarily. Two tugs (each rated at 400 h.p.) were placed in front and two in the rear, and they were shifted about in going around the bend and approaching their destination. Full control was maintained through Fig. 2.— The Present Haklem Ship Canal Bridge REPLACEMENT OF THE HARLEM SHIP CANAL BRIDGE 1 1 out the passage, the only incident being the considerable amount of mud stirred up at one point by one of the tugs. The writer was stationed on one of the middle scows. There was no creaking of the top chords, or other indication of strain in the main trusses. Some of the braces about the drum looked well "tuned-up," while the diagonals were taut. The cribs were solid and well connected. (Fig. 2, Plate II.) 3.45 p, m. — Arriving at University Heights, the various clearances were found to be sufficient. No bumping against the timber protection pier occurred. The diagonals to the drum slid past the round coping, and the rim wheels were well up over the rack, and, in fact, afterward, they raised somewhat. The pumps were again applied, and the tide soon helped to drop the mass on the pier. As the pivot casting drew near to place, the holes were seen to match, and the anchor-bolts were dropped in. 10.30 p. m. — The pinion engaged the rack, and, steam having been maintained all the way up from Broadway, the span was turned one- quarter around, and left in the open position above the pier protection. Moving the New Draw-Bridge. — On the following morning (June 15th), the scows were taken to 216th Street and placed diagonally as before. The day was spent in adjusting the cribbing, calculating on the tides, as usual. A finder-ring for the new pivot casting was placed accurately on the old pier, and wedge plates to relieve the pivot, were provided on its east and west sides. At the end piers, the latch girders and wedge plates were centered and trued up generally. During the night rain began to fall and continued throughout June 16th, with the wind in the east. This had the effect of raising the low- water level 1.3 ft., and caused much delay in cribbing up. Later, the north scow grounded, in spite of this, showing the need of dredging, and causing further delay; so that, with high tide due by almanac at 6.40 P. m., it was 1.00 P. M. before pumping out began. The plant was the same as before, with an added centrifugal pump and a number of hand-pumps. All were needed, for it was not until 4.15 P. If. that the pivot was clear, although the ends had been so for half an hour. 5.20 p. m. — The pivot was 10 in. clear of its seat, the latter being at Elevation 10.94 above mean high water. The top of the pier at * Broadway was nearly the same (Elevation 10.89) ; but the wheel tread was about 1 ft. higher (Elevation 12.11), and this governed the lift at 216th Street, and the consequent time of Btarting. (Fig. 1, Plate III.) I? REPLACEMENT OF THE HARLEM BHIP CANAL BRIDGE 5.55 i\ M. — The pivot was 14 in. up, and there was not much more time, as the tide seemed to have turned. Tidal records showed that the flood tide was 0.4 ft. higher than usual. The order was finally given, and the immense structure (weighing with cribbing nearly 2 000 tons) was slowly pulled out into the stream, pivoting on the northwest corner and lining up, broadside to the north. The channel was none too wide, and there was some sweeping away of blocking into the river. One of the four tugs was detailed to butt out that corner, and thenceforth there was no stop or hindrance until near Broadway, where speed was reduced, lines were cast off, and the whole mass, grazing the approaches, was eased into its berth. The pivot casting slid over the rack with a margin of an inch or two. G.35 p. m. — The casting, somewhat out of level, hung over the pier 15 in. clear, and there was ample clearance at the approaches. Lines were made fast, and the tugs were used, as before, to furnish steam for pumping water into the scows. The two scows at the north end were loaded heavily, the water being within 18 in. of the deck. 7.00 p. M. — Electric connections had been made, and lights were turned on below and above. 7.45 p. m. — The north pinion engaged the rack, and, later, closed the draw. The south pinion was evidently about right. 8.15 p. M. — Owing to the failure of a centrifugal pump, one of the northerly scows had to be sunk, and the pivot casting was 4 in. above the pier on the north side, and 6 in. on the south side, an inequality which was magnified greatly at the ends of the draw. 9.30 p. m. — The span rested on the center pier. June 17th, 190G, Sunday: 12.30 a. M. — The scows had dropped down, and were free from (lie south arm. An hour later, the same was true for the north arm. 3.45 a. m. — The bridge was swung by hand-levers, and the end wedges were in place, but not adjusted. 5.20 a. M. — The span was swung half around by electric power, and it was found that it could be operated quickly and satisfactorily. The temporary wooden aprons were repaired, and the roadway was opened. 9.30 a. m. — Measurements at G8° Fahr. showed that the space between the draw and the approach span on the north side measured 25 in. and on the south side 2J in. The next week was spent in shipping away timber, thereby releasing the scows, removing the chocks from the top chords, taking down the pile platform, adjusting the aprons, wedges, and pinions, installing the • two circular feeder tracks at the pivot, and connecting with the third rails above. By the end of June, the approach S{rtnis had been lined up, and the rocker posts plumbed up. The asphalting was also well advanced. Ulil'LACKMKNT OF TIIK IIAIM.K.M SHIP CANAL BRIDGE It was decided to put in fixed aprons on the approach spans, and to close the span in one position only (without reversing), on account of the number of important rail connections. Later, it was found neces- sary to stiffen the rack with iron knees, at intervals. Much of the new equipment was done by the Bridge Department of New York City, in accordance with its practice on the Harlem. This is the only pivot-bearing draw on the river. The weight on the pivot is 1 500 tons. It is operated by two 60-h.p. electric motors, and. considering its weight, is handled more easily and quickly than any other. Occasionally, it delays traffic, while opening and closing, for a period of only 4 min. In addition to signals by whistle, and electric bells, there is a telephone connection (seldom used) between tower and engine-room. (Fig. 2, Plate III.) During 1907 and 1908, the shafting was partially renewed several times, due to causes which were exceedingly difficult to locate, but resulting in twisting the shafting. In June, 1907, the latch girders were firmly fixed by reinforced concrete around and about them for the entire distance between the pedestals, and thus the play at the ends of the draw was much reduced. The contract price for removing the old bridge and replacing it with the new one was about $250 000. Old North Approach Span. — In November, 1907, the old north approach span, which had been left on the south bank, was rolled toward the river and received by scows, in much the same manner as described previously. The scows floated broadside through the draw channel, and were afterward swung around, with one tug in front and one in rear. The top of the east pier at University Heights was over-run comfortably, and the rollers did the rest. As no water was pumped into the scows, several tides were necessary before the final position on the masonry was reached (November 29th, 1907). University Heights Bridge was declared open by Mayor George B. McClellan on January 8th, 1908. In all these operations, extending over a period of two years, there was no accident of a serious character to any of the force employed. The Rapid Transit Commissioners were represented by George S.« Rice, M. Am. Soc. C. E., Chief Engineer. Alfred Craven, M. Am. Soc. C. E., Deputy Chief Engineer, Sverre Dahm, M. Am. Soc. C. E., Gen- eral Inspector of Designs, and by C. V. V. Powers, M. Am. Soc. C. E., 1 1 REPLACEMENT OF THE IIAKLKM SHIP CANAL BKIIXiE Division Engineer. The writer acted as Resident Engineer, and takes this occasion to make grateful mention of his subordinates, who con- tinuously gave their best services. They were, from time to time, as follows : Messrs. R. A. Berry, L. C. Devery, J. D. P. Hogue, R. J. Smyth, 0. M. Torpey, and W. P. Vallely, Assistant Engineers; W. C. Martin, Rodman, and Theodore Belzner, Jun. Am. Soc. C. E., Inspector. The Rapid Transit Construction Company was represented by George H. Pegram, M. Am. Soc. C. E., Chief Engineer, and A. S. F. Berquist, M. Am. Soc. C. E., Inspector of Steel. Historical Notes. In searching for records of similar operations, the first, chronologi- cally, of any magnitude, is seen to be the first in magnitude with reference to the tonnage moved, if not in regard to the velocity of the current and the exposure to gales; and, later, in relation to special appliances used in raising to place. This was the work of Robert Stephenson in connection with the Conway and Britannia Bridges, as recorded chiefly by his assistant, Mr. Edwin Clark.* Mr. Stephenson had led up to those pioneer monuments by various achievements in his practice. He had acquired habits of close technical experimentation, and of co-operation with other engineers, architects, parliamentary committees, boards of directors, and, in fact, with men of all classes. He could say the right word at a dinner; as for instance when he noted the reduction of time between London and Dublin made possible by the bridges, and spoke of engineers as being instrumental in bringing about a better understanding between the English and the Irish. It required a very able man to make the jump to the first long- span wrought-iron bridge; for, as Mr. Clark says, the process of uniting wrought-iron plates by rivets had been hitherto a peculiar feature of boiler-Work or iron ship-building and had been little resorted to by the Civil Engineer, until Mr. Stephenson proposed the construction of these bridges. All wonderful things cease to become so, however, when the separate steps are studied; and perhaps the transition from an iron ship, with • ** The Conway and Britannia Tubular Bridp^." London, isr>0, i>\ Edwin Clark. Re*.i deal Engineer. REPLACEMENT OF THE HAUL EM SHU' CANAL BRIDGE 15 ribs and knees at short distances, to the rectangular tube, with web stiffeners and gussets every 2 ft., together with the cellular top and bottom, borrowed from a previous design of his, was not intrinsically so great an innovation, after all. The spectacle of an accident to an iron vessel and the evidence of its stanchncss in resisting the waves, and holding together, is said to have exerted a strong influence on Mr. Fairbairn, the manufacturer, and through him, on Mr. Stephenson, as to the ability of a similar iron framework to carry rolling loads. Later, when the method of floating the tubes was considered, there naturally followed the idea of doing without the use of scows, but, for good reasons, that method was not adopted. The Conway Bridge. — Work was wisely begun on the smaller bridge, at the Conway River (May, 1846), and, after two years of experiment and construction, the first tube was floated from its falsework, some GOO ft., to the site it now occupies. Steam riveting machines had been used throughout, and half-round "rimmers." The type of tube had been decided chiefly from tests on a model of one-sixth the size of the bridge. Not satisfied with this, Mr. Stephenson supported the tube at its ends, cut out the falsework, and loaded it with cars containing iron plates. The resulting deflection was so small that no similar test was necessary again. The floating of the tube (26 ft. high, 14 ft. wide, inside, 412 ft. long, and having a weight of 1 417 tons of 2 000 lb.) was done by six scows (each 98 ft. long, 25 ft. wide, and 8 ft. deep) arranged in two sets of three. This provided for an overhang at each end, sufficient to take the timber seats prepared in the recesses of the abutments, at some distance from the shore line. The tubes were 6 ft. 6 in. above the water. There were two guide-chains under water, with which it was sup- posed that the movement could be controlled in a manner similar to that of a ferry on a suspended cable, but they proved almost useless, and most of the hauling was done by two Manila hemp cables, each operated from a 50-man capstan. The force was about 350 men, on land and water. A steamer was at command, but seems to have acted only as a utility boat. Some progress was made against the rising tide, but more when it slackened at the flood, and enabled the tube to be easily hauled to place, clearing the abutments by 4 in. at each end. The average rise 16 REPLACEMENT OF THE HARLEM SHIP ( ANAL BRIDGE of the tide was 18 ft. and the elearance of the bridge was also 18 ft., so that a lift of about 12 ft. remained to be made. This was done by a hydraulic press (18-in. ram with a 6-ft. stroke) at each end. The weight of the tube was transferred first to a cast-iron frame inside; thence to eye-bars with pin connections; and thence to the cross-head of the press above. A small steam engine and pump, to force water into the press, were attached. Barring the pulling out of a cylinder head, and the delay in re- placing it with a stronger one, the work was continuous; but cast iron was either eliminated or heavily reinforced with wrought iron at the Britannia Bridge. Mr. Stephenson passed through on the first locomotive in April, 1848. It may be well to note here the early decision with regard to painting plates before riveting them together, which was, that close contact is thereby prevented. The claim was made that old pieces of riveted work were most sound at the joints. Where necessary, a mix- ture of red and white lead was used as a filler. The cellular floor and top chord, built up of plates and angles, wen 1 large enough to allow a man to paint them from the inside. The track floor was cheaply water-proofed with a mixture of coal-tai, lime, and turpentine. A corrugated, galvanized-iron roof was constructed over the whole tube. The price per (long) ton was £37, and the total cost of the Conway Bridge was £145 200. The Britannia Bridge. — The Britannia Bridge connects Wales and the Island of Anglesey, across Menai Straits. The two main spans are each 460 ft. in the clear, and were floated to place, whereas the approach spans were constructed on falsework at the site. The Government required 105 ft. clearance above high water, and therefore a hollow, center tower, 220 ft. high or more, was required on Britannia Bock. The foundations were begun on Good Friday, 1846, and the last stone was laid in June, 1840, the masons working 12 hours a day for 6 days in the week, in order not to hold back the tubes. The latter were 30 ft. high, 14 ft. wide, inside, and weighed 2 150 tons. Tn floating each tube 650 men were employed, about 400 being REPLACEMENT OF CHE ll.UM.I M SHIP OAK AX BRIDGE 17 sailors hired for the occasion. Eight scows were used, in much the same way as at the Conway Bridge, and were guided by Manila cables buoyed up by casks. Tarred Manila ropes were used for hauling, with ''cable-stoppers" and "messengers." The four tubes were handled in conformity with written instruc- tions, after Mr. Stephenson had made experiments with models on a small pond. The various rocks and shoals were indicated, together with the scows, tubes, and lines, and everything that it was possible to foresee was discussed. At the moving, he and others (including Captain Claxton, who had immediate charge) occupied a position on the top of the tube, and from that elevation signals by flags were given to each boat concerned. Although, in floating the four tubes, there were instances of heavy cables snapping, and of severe bumps, yet, in general, the men were equal to the emergencies, and landed the tube each time in the re- cesses of the masonry, before the tide fell. Two steamers were in attendance. The lifting machinery was considerably improved, the press having a 20-in. ram, lined with brass, and having a 6-ft. stroke as before. Leaks were stopped with oatmeal gruel and sal-ammoniac, and, later, by a second U-shaped leather collar below the first. Brickwork in cement followed closely, but, in spite of this, the first tube, during one lift, dropped 8 or 9 in., doing considerable damage. In the raising of each tube, 6 ft. per day was accomplished for about 16 days, and a single track was put in use in March, 1850. The fourth tube was floated in July, 1850. The Britannia Bridge was riveted by hand. The total cost was £601 S65, and of this, one-fourth was for masonry. Both the Conway and Britannia Bridges are still in use. Associated with Mr. Stephenson were Mr. I. K. Brunei, and Captain Claxton; and as these gentlemen were soon afterward engaged at the Chepstow and Saltash Bridges, in similar movements, brief mention of these may naturally be made next. The Chepstow Bridge. — The chief feature at the Chepstow Bridge Cover the River Wye), barring the pneumatic process used in sinking the cylindrical foundations, was the design and erection of the 300-ft. channel span. IS REPLACEMENT OF THE HARLEM SHIP CANAL BRIDGE The top chord was composed of one wrought-iron tube, 9 ft. in diam- eter, treated like a trussed beam of three panels, except that there were two suspended chains, with the track between. The height of the truss was 50 ft. Pins 7 in. in diameter were used at the ends. A second tube for the other track enabled lateral bracing to be introduced, and stiffened each. Following out the cautious method of the Conway Bridge, the first span was erected on falsework and tested with a live load of 2£ (long) tons per linear foot. The wrought iron in each truss weighed 4 GO (long) tons. It was then taken apart, and the tube itself, sup- ported by posts and chains, was floated (April, 1852), on six wrought- iron barges. It had to be done quickly, as the rise of spring tides was 40 ft. The use of crabs with two barrels enabled hauling-chains to be wound without difficulty. The tube was lifted by jacks, during the day, some 40 ft., and, later, it was raised 50 ft. more, to its final position. A single line was opened up in July, 1852. This was the first circular tube of large size introduced as a bridge member. The Saltash Bridge. — In the Saltash Bridge, a single-track struc- ture across the Tamar River, nea;- Plymouth (also on the Cornwall Kail way), the top chord tube was made elliptical (16 ft. 9 in. wide and 12 ft. 3 in. high). The bottom chord was double. The typo of truss was a bowstring, with the string curved equally with the bow. Mr. Brunei* (3d) states that "the mode of floating and lifting the superstructure had great influence in the preparation of the design" — meaning cylindrical foundations and cast-iron columns, as well a« truss design. There were two single spans of 455 ft., 56 ft. high, and each con- tained 1 200 tons of wrought iron. The bottom chord was of 7 by 1-in. bars, 20 ft. long. There was a solid ballast floor throughout. The work was begun in 1853, but was delayed, on account of the difficulty of founding a 35-ft. cylinder, by using compressed air, af a depth of 88 ft. below high water. This was the first successful attempt on a large scale. Meanwhile, the superstructure was manufactured, put together, and afterward tested at its falsework by a live load of 1 340 tons, uniformly distributed. The deflection was 5 or f> in., and was deemed satisfactory. *"T>>e Lifp of [nambard Kingdom Brunei, Civil Engineer," by [mnbsrd Brunei. B. C. (j., London, wo. REPLACEMENT 01' 'I'll 10 IFAHLKAI SHIP CANAL BRIDGE I!) It was as late as September 1st, 1857, before the first span was moved, and information in relation to it is comparatively meager. The methods used at the Britannia Bridge were followed. The actions of the 500 men engaged were regulated in accordance with printed instructions which had been issued, and with flag signals from the ''bridge," where Mr. Brunei and others were stationed during the passage. Mr. Stephenson was too ill to be there. The move was at first outward from the shore, then, after an inter- val in which the hawsers were shifted, up the river, endwise to. the center pier, then a quarter turn was made, and finally, an accurate placing by strong lines at the piers. There was a clearance of a few feet above the water. The tide was not swift and there v/ere no mis- haps mentioned. The lifting was done 3 ft. at a time, separately at each end. At the center pier, cast-iron columns (braced) were built up as fast as these lifts were made; but, at the shore end, the masonry proceeded more slowly. There was no cause for hurry, as the river traffic went through the other opening. There was finally a clear distance of 100 ft. above spring tides; and the total height above the foundations was 260 ft. to the highest point of the trusses. Three hydraulic presses or jacks were used under each end. The rams had a screw-thread, arranged with a nut in such a way that the tube was supported at all times. Timber packing was also used. By July, 1858, the full height was reached. The second span was floated successfully soon after. In May, 1859, the bridge was declared open by Prince Albert, after whom it was named the Royal Albert Bridge. The total cost was £225 000. The fiftieth anniversary has recently been celebrated, the old bridge remaining as it was originally. Oilier Bridges. — Passing by well-known examples of this method of erection — the Moerdyck Bridge (1875), the new Tay Bridge (1885), the Hawkesbury Bridge (1889), and others of 500 tons or less, in which about the same or similar principles were involved — we come (1S90) to the 523-ft., single-track, through bridge across the main channel of the Ohio River, below Pittsburg, on the Ohio Connecting Railway (Pittsburg, Cincinnati, Chicago, and St. Louis Railway), erected by the Keystone Bridge Company, C. L. Strobel, M. Am. Soc. C. E., Chief Engineer. The Government, having required the long span and also a clear 20 REPLACEMENT OF THE HABLEJM SHIP CANAL BRIDGE height of 75 ft. above mean water level, and having limited the time during which the river might be obstructed, it became necessary to build the falsework at a corresponding height, assemble and erect the bridge upon it, and then float the structure to its site, as quickly as possible. Nineteen bents of 5 piles each were driven and cut off 16 ft. above the water. On these were framed bents of 5 posts each, with an 80-ft. sill and a 32-ft. cap, all with the usual railroad trestle dimen- sions. At the base, across each alternate bay, 20-in. I-beams were added. Under these were duly placed nine ordinary coal barges (130 by 26 by 8 ft.). The weight of the bridge was about 900 tons. The trusses were 65 ft. high. The bracing of the bents into towers and the binding of them together by rods with turnbuckles was susceptible of careful study, and, once accomplished, eliminated the principal cause of un- certainty. At Hawkesbury, the Union Bridge Company had effected this by a single pontoon, 335 ft. long, well braced with timber and wire cables internally, but with the superstructure at a much lower level. In both cases, the hauling to place was done by cables. At Pittsburg, two scows were fastened to the bows of the end barges (leaving the three barges in the middle unattached), and on these scows four engines controlled the lines up to the bridge piers. The nine barges were braced together at the bow, and. after floating out to clear the piles, they were braced likewise at the stern. The current in the river was slight, and, although a couple of heavy showers occurred while the scows were being moved the few hundred foot neces- sary, there was no bumping against shore or against obstructions. The itinerary of the work was as follows:* 8.50 a. if aj August 10th. — Sufficient water had been pumped from the barges so that the mass rested on them. 11.15 a. m. — The barges had been floated out toward the center of the river, thereby clearing the bents of piles. 7.30 p. m. — Water was again admitted to the barges, and the span cleared the bridge seats by 2 ft. 8.20 p. m. — The span was about in. clear. The work was blocked up and left until the following morning. 7.30 \. m . , August 20th. — The lowering was continued, so that the bridge rested on its piers. The barges were removed later in the day. ; KiKjiticvrituj \( us, SepttMiihtM- 'JOth. l' s( -'<». REPLACEMENT OF TJIE HARLEM SHIP CANAL PR I DOE 21 There had been no distortion of bents or barges, nor any important casualty of any kind. In Canada, the Dominion Bridge Company has erected several bridges of note, in this manner. The Coteau Bridge (1890), across the St. Lawrence (Grand Trunk Railway) west of Montreal, has single-track spans of 175 to 223 ft. The Shubenacadie Bridge, in Nova Scotia (1901), across an arm of the Bay of Fundy (Grand Trunk Railway), has single-track spans of 215 ft., each weighing 150 tons. The Miramichi River Bridge, Newcastle, N. B. (1902), has single- track spans of 204 ft., each weighing 320 tons. The French River Bridge, Ontario, a single-track span of 415 ft. weighing 1 424 tons, was erected by the Canadian Pacific Railroad in 1907. This was slid endwise from the embankment upon a scow (155 by 33 by 12 ft.) and thence was hauled to place by lines from an engine on the scow. Local difficulties in all these cases were overcome successfully. Sand Jacks. — The use of sand in jacking up or in letting down bridges of moderate tonnage dates back to 1S78, or even earlier, in Europe. The most noteworthy instance of this sort in America was wit- nessed by the writer on Sunday, December 20th, 1903, at Newark, N. J., when the 220-ft. double-deck draw-bridge of the Delaware, Lacka- wanna and Western Railroad, across the Passaic River, was moved, complete and ready for use, to its present position, under the direc- tion of Lincoln Bush, M. Am. Soc. C. E., Chief Engineer. The bridge is center-bearing, and has four tracks. It was moved up stream some 40 ft. and lowered 10J ft. from its level at the old pier. Several years before, it had become necessary to renew the old draw, and as grade-crossing w r ork was in progress in the vicinity, the new draw was designed accordingly, and the lower deck used tem- porarily. The weight of the bridge was 1 017* tons, and the sand and timber- work 936 more, a total of 1 953 tons on the scows above the deck. A year or so before, there had been experiments on the action of sand in compression by loading two 83-ft. girders on a pair of plungers * Engineering News, Vol. so, p. B06. 22 REPLACEMENT OF THE HARLEM SHIP CANAL BRIDGE fitting into 4 by 6-ft. boxes. It was found that there was little swelling, that the most favorable angle of the side holes was 45°, and that the holes should be made funnel-shaped, with a diameter of 2 in. inside and 3 in. outside the box. There were also 2-in. holes in the bottom of the box. All holes were regulated by slides. The bridge rested on two scows under each arm, and there were four sand boxes of 12 by 12-in. stuff, each 52 ft. long, 12£ ft. high and about 4 ft. wide, inside measurements. The plungers had a clearance of \ in., and their pressure on the sand was less than- li tons per sq. ft. The rise of tide at Newark is generally somewhat more than 4 ft. but on the day the bridge was moved the wind and tide were perverse, and, unfortunately, there was a rain all day; so that it was with dif- ficulty that the sand was kept in prime flowing condition, tarpaulins being used to protect it as much as possible. In spite of all precautions, there was irregular settling, and time was lost in balancing the movement. The bridge had been raised by the tide, helped by four centrifugal pumps. A boiler scow and pump was held in reserve. The rate of movement was as follows: At 0.30 a. m., the old supports were cleared, " 0.35 a. M., began lowering at the new site, " 5.00 p. M., the span rested on the new pivot, . " 6.15 p. m., all scows had been released, " 8.30 p. m., the first train crossed over. The objection to this method is the likelihood of wetting the Band (which in this case was screened through a sieve of \-\w. mesh) and the consequent ignorance as to what is going on inside the boxes. Under such conditions, looking for evidence of strain in heavy timber is a responsibility which few engineers would care to take. Finally, other notation operations on the Harlem River should be mentioned, namely: At Macomb's Dam Bridge, at Madison Avenue, and possibly at the crossing of the New York Central and Hudson River Railroad — operations which deserve to be recorded in detail. DISCISSION OX I f A HI. KM Mill' CANAL HHIlNiK DISCUSSION Lincoln Rrsn, M. A.m. Six. ('. K. (by letter). In reference to tlie Mr. use of sand jacks for lowering the draw-bridge of the Delaware, Lacka- Bu!,,l wanna and Western Railroad over the Passaic River at Newark, N. J., on December 20th, 1903, the following comments are offered: On the day the bridge was moved and lowered, rain commenced at 1.20 a. IE., and continued until 5.00 P. M., at which time the bridge rested on the new center pier in exact position, the precipitation during this time being 1.23 in. In order to keep the sand dry while the -and boxes were being filled and for the same purpose after they were filled, tarpaulins had been provided before the operation. During the night prior to the move, one of the tarpaulins accidentally became partly removed on the south side of one of the four boxes, letting some water into the sand near the top of the box. This wet sand, however, caused a delay of only 20 min. in the lowering, and, aside from this -incident, there was not the least difficulty in keeping the sand dry, even with the heavy precipitation of 1.23 in. Mr. Howe mentions the irregular settling and lost time in balancing the movement. He probably refers to the lowering of one end of the bridge at a time, which was done in 2-ft. stages. The day was chilly, and the men were wet to the skin. The writer was reasonably certain that the operation could be completed on schedule time, by handling the lowering slowly and carefully, rather than by seeing how quickly it could be done, and, instead of lowering both ends of the bridge at the same time, he adopted the plan already mentioned. It was shown l>y tests made with test sand boxes, and in the actual lowering of the bridge itself, that the bridge could be lowered very readily at the rate of 3 in. per min. Observations on the tide movements at the bridge site for four months, and other reliable data, showed that at times the minimum low tide did not fall below zero or mean tide, and that the maximum low tide frequently fell to an elevation of 3 ft. below mean tide. These data also showed that at times the high tide did not go above mean tide, and that the maximum high tide at times rose to a point 4 ft. above mean tide. These observations showed that the minimum varia- tion between high and low tide was nothing, and that the maximum variation was 7 ft. The normal variation lor low tide at the bridge site is — 2.5 ft. (below mean tide), and the normal variation for high tide is +2.5 ft. (above mean tide). The work had to be done, as planned, in 12 hours, in other words, from low tide to low tide. When the four barges were run under the bridge, the low tide was — 1.0 ft., and when they were released at the next low tide, fell to only — 1 ft. instead of a normal low tide of 24 DISCUSSION ON HABLEM SHIP CANAL BRIDGE Mr. — 2.5 ft. The high tide rose to +3.0 ft. instead of a normal high Bu8h - tide of +2.5 ft. The tests made with test sand boxes showed, with most convincing evidence, that dry sand, carrying a load of about 2 500 lb. per sq. ft., produced very little lateral pressure; and no outward deflection of the timber, with a straight-edge on the sides of the boxes, was discernible. Neither was there any perceptible deflection in the sides of the timber boxes used in lowering the bridge, except with one box where wet sand occurred, and, as stated, this was relieved with a delay of only 20 min. by opening the sand holes lower down in the box where dry sand flowed out freely.* The fact that the work was done on schedule time, as planned, witli abnormal tide conditions and a heavy rainfall continuing throughout the operation, should be convincing evidence that this method of lower- ing a heavy structure, tempered with good judgment, is perfectly safe, economical, and efficient. Mr. Martin Gay, M. Am. Soc. C. E. — When the proposal was first made Uay * to run the trains of the Rapid Transit Railroad across the Harlem Ship Canal Bridge, it was thought feasible to add an upper deck to the old draw-span and to build new spans above the old approaches. This was objected to by the Department of Bridges, under the jurisdiction of which the bridge came, on the ground that the turn- table, though well designed and sufficient to carry its comparatively light load of some 700 tons, was not substantial enough to endure the racking and distortions due to a moving load of perhaps twice that weight. While this point was under discussion by the engineers of the Department of Bridges and the Rapid Transit Commission, several other interests came into view. For some years the Kingsbridge Railroad Company had held a franchise permitting it to cross the bridge, and its officers had discussed the question with sufficient serious- ness to cause the Department of Bridges to make a study of the situa- tion and to determine that it would be necessary to strengthen the floor system and practically rebuild it at a higher elevation, and also to change the grade of the approaching streets. This would have been a matter of considerable expense. Also, the New York Central Railroad, in connection with the plait for the elimination of grade crossings, had prepared to abandon the long curve with many crossings through Kingsbridge Village, and to cross from the east bank of the Harlem River to the east bank of the Hudson, following the north bank of the Ship ('anal and passing under the Ship Canal Bridge. To do this an additional space of 12 ft. *An BOOUrate and full presentation of this Operation and the tests mad« prior to it are given in WngineerinQ ZVeic*, December :i;st. 1906. DISCUSSION ON II MM. KM S 1 1 II ' CANAL HKIlXiK was required between the bulkhead of the canal and the north abutment Mr. of the bridge. (iav Also, at this time the Department of Bridges was planning a new bridge across the Harlem, near Fordham Landing, and had about determined to build a double-leaf rolling lift of the Scherzer type. It was to the combination of these various interests that we owe this admirable paper. Each party to the combination contributed its share of the cost of the project, either in money or in work, or both, and each got what it wanted. Mr. Howe has described the work of moving the bridge so clearly that little can be said on that head. Each of the six operations, that is, moving out an old span or moving in a new one, was not of itself a very complex problem, but, taken altogether, it was a work of considerable magnitude, and con- sidering the cramped space in which the contractor was obliged to maneuver the approach spans, the uncertainty of the tides, and the short space of time during which he could obstruct travel, the problem was one requiring nice calculation and unusual judgment. The time during which street travel could be obstructed was fixed by the Department of Bridges, whose duty it was to consider the con- venience of the public, after consultation with Messrs. Terry and Tench, as to the methods they proposed to use and the time they would require. The Department w r as not willing to deprive people of an opportunity to cross the stream for a longer time than necessary, and did not wish to impose oppressive conditions on the contractors. It was estimated that one day would be sufficient to move and re- adjust each of the approach spans, allowing for accidents and delays, and that 3 days would be required for the draw-span. Therefore, the permit allowed street travel to be interrupted for a total time of 5 days, but not for more than 3 days consecutively. That no very serious delays occurred is indicated by an analysis of Mr. ITowe's figures, which shows that the total time of interrupted travel was 3 days, 1C hours, 35 min., and that during the moving of the draw-span, tin interruption was less than the stipulated 3 days by 4 hours, 55 min. Each of the three new spans was put to use by vehicles and pedes trians as soon as it was placed in position, and the draw-span was opened to pass vessels without causing any delay to river traffic. The old spans, which have become a part of the University Heights Bridge, had to have considerable changes in order to bring them up t«> modern requirements, and they have done duty continuously since the formal opening of the bridge. In all operations of this nature on the Harlem River, of which there is any record, the rise and fall of the tide has been counted on as a factor and has been used to some purpose, and yet it is such an uncertain factor thai in almost every case, other means for raising or 2G DLSCL'SSION ON FIAIM.KM SIII1* CANAL BKIIXJK Mr. lowering- have been provided and usually resorted to, in order to ' supplement the deficiency of tidal power. As the movements of the tides, of greater or less extent, take place at more or less definitely known times, they must be reckoned with. On a non-tidal stream, however, where other power would of necessity be provided, the attend- ant uncertainty and anxiety regarding the behavior of the tide would be eliminated, and probably the cost would not be greatly increased. One occasion on which the lifting power of the tide was used to good advantage, but perhaps not with any great economy, was in the moving of the draw-span of the old Macomb's Dam Bridge, of which Mr. Howe has spoken. The bridge was built in the early Sixties, and was replaced in 1892 by the present structure, which was designed and the construction supervised by A. P. Boiler, M. Am. Soc. C. E. It was necessary to remove the old bridge to make way for the new one, and it was also necessary to provide for public travel. It was decided, therefore, to move the draw one block north of the old site, to the line of 15Gth Street. A pivot pier and approaches having been constructed at an elevation 12 ft. lower than that of the bridge, the problem was to move the draw-span about 200 ft. north, and to lower it 12 ft. to its new position without obstructing navigation. This was done by lifting the span on two scows and a cribbing of 12 by 12-in. timbers, and towing it to a pile and cribbing pier which had been built near the bank of the stream and out of the course of passing vessels. Here it was deposited safely at an elevation somewhat lower than that from which it had started. When the next flood tide raised the scows and bridge, a course of crib timbers was pulled off the pier, and as the span settled with the ebb tide and rested, a course of crib timbers was taken off the scows. As the next flood tide freed the span from the pier, another course of timbers was removed and so on, until it rested at the proper elevation to be floated across to the permanent pier and be placed in position. An instance of the tide not doing its duty, as expected, was at the moving of the Madison Avenue draw-span, where an east storm made an early flood which caught and held the scows under the span after it had been landed. Considerable damage might have resulted, but for the scuttling of one of the scows. Mr. St. John Clarke, M. Am. Soc. C. E. — As Mr. Gay lias mentioned laik<> ' a design prepared for this bridge, in which parts of the old bridge were used, it may be of interest to explain more fully the reason for making this design. When the crossing of the Harlem at this point was being con- sidered by the engineers of the Rapid Transit Commission, it developed tlnit the Bridge Department desired a wider bridge, and also wanted to use the old bridge at the lower crossing; that the New York Central wanted to change the northern span, so as to give more 1 head-room DISCUSSION OX HARLEM SHIP CANAL BRIDGE 27 and also room for more tracks; and that the Metropolitan Street Rail- Mr. way Company was considering a change in the floor system of the old bridge, so as to provide a slot and conduit in order to allow the surface cars to cross the bridge and continue up Broadway. Changes in the old bridge for any one of these interests would have been quite expensive; but to provide for these several requirements in a new bridge would not add very materially to its cost. The stand was taken by the other interests that the Rapid Transit Commission, or rather its contractor, had to build a new bridge in any case, and that, while doing so, it would be better to provide for all future uses of the bridge, the additional cost being small. It is hardly necessary to add that these interests displayed no great willingness to assume a fair share in the cost of the new bridge. It was because of this condition of affairs that the speaker, as General Inspector of Designs, considered the possibility of using the old bridge. A design was prepared utilizing the floor system, turn- table, and machinery, with the bottom chord of the old draw, new and deeper trusses being provided. This design complied with all the theoretical requirements of the Bridge Department, and it was shown that the old bridge could be altered to serve the Rapid Transit Railroad. When this was evident, and not before, it became possible to make a fair arrangement with the various interests involved. The new bridge was designed, and all those using it contributed to the cost. All were somewhat disappointed because the design for using the old bridge was not adopted, as it seemed hardly fair to these plans after they had served their purpose so well. Theodore Belzner, Jun. Am. Soc. C. E. (by letter). — The writer Mr. was connected with the construction of this bridge as Inspector of Belzne Erection, for the Rapid Transit Railroad Commission, and recalls the cracks which were discovered at the apex of the connecting angle plates, and during the riveting up of those replaced. In a general way, he tested the shop rivets of the connecting plates at the other end of the pivot girder, in order to determine their tightness as compared with the field-driven rivets, and found quite a number of the shop rivets loose. This was reported to Mr. Howe, and resulted in a conference between the engineers and inspectors. It was finally decided to cut off a few rivet heads and drill out the shanks. This method of removal, however, was abandoned after a few rivets had been partially drilled out, as it was slow and costly, and in some cases it was difficult to keep the drill centered on the rivets, especially on the innermost ones. Then the rivets were "backed out" and with much better results, on account of the reamed holes in the plates, etc. On May 15th, 190G, the writer reported the results of lii- examina- tion to Mr. Howe, which in part was as follows:* * Transactions, Am. Soc. C. E., Vol. LVII, p. a62. DISCUSSION OX HARLEM SHIP CANAL BRIDGE Mr. ''The bent plate of this member contained 87 long 1-in. shop rivets, iezner. 2 8 f which were condemned for looseness. The shop rivets were cut out and replaced by machine-driven field rivets. In cutting off the heads and 'backing out' the rivets, no material damage was done to the adjacent ones, but many of them became loosened while the field rivets were being driven. The rivets 'backed out' easily, some requiring only a few blows with an 8-lb. hammer. An examination of these rivets showed that they had not been upset properly, in some cases shoulders were formed under the heads (caused by oval-shaped holes through the bent plates), in other cases the rivets were slightly rusted and showed some scale. The result was that more than 60 field rivets had to be driven to replace the shop rivets." Referring to the twisting of the shafting of the New Ship Canal Bridge, the writer would like to ask if this occurred in driving home the wedges on the rest piers? If the writer remembers correctly, there was a slight "tilt" at the southeast corner of the bridge, and some difficulty in driving the wedges, especially at this point. Information as to the adjustment of this matter would be of interest. How does the efficiency of the shaft claw couplings compare with that of the flanged ones? Mr. Frank W. Skinner, M. Am. Soc. C. E. (by letter). — The opera- 1 ner " tions described in this paper are nearly or quite the most intricate and prolonged of any of similar character yet effected, and their complete success under difficult and involved conditions reflects great credit on the engineers and contractors for the preparation of the plans and for their courage and ability in executing them. The author's description of the moving of this bridge is so com- plete that there is no room for criticism or discussion, but some addi- tional facts concerning other work of a similar nature which he mentioned briefly, may be timely. The Hawkesbury Bridge, Australia, has seven through, double-track spans which were erected successively on falsework, 34 ft. high, on the deck of a Gl by 335-ft pontoon, 10 ft. deep, stiffened by bulkheads and trussing cables, and sunk on a timber grillage in shallow water. After the completion of each span, the water was drained out of the pontoon at low tide, and at high tide it was hauled to the piers, a distance of about 4 000 ft., by two cables operated by fixed hoisting engines. Bach span was lowered to the piers by admitting water ballast to the pon- toon. Despite several accidents, such as the breaking of the cables, the grounding of the pontoon, and the reversing of one span, the erec- tion was accomplished successfully. The single-track bridge across two branches of the Miramichi River, New Brunswick, had twelve 204-ft. -pans, each weighing about 225 ton-. Without much interruption of t ratlic, these were replaced. i.ii the same piers, by stronger 270-ton spans. Two Beta of pile false- PLATE IV. TRANS. AM. SOC. ClV. ENGRS. VOL. LXVII, No. 1143. SKINNER ON HARLEM SHIP CANAL BRIDGE. Fig. 2. — Brcnot Island Briikjk Across thk Ohio Hivkh. Completed Span and Falsework Floated on Ni>e Hahges. DISCUSSION ON HARLEM SHIP CANAL BRIDGE 2!) work, opposite each other, were established parallel and close to the old bridge, and the new spans were creeled successively on one of them, skidded across to the other, and riveted up while the next span was being erected on the first falsework. After the riveting was com- pleted the span was transferred to towers on the decks of a pair of pontoons, and was towed to position between the piers. Simultane- ously, the old span was lifted from its piers and removed on a Howe truss span on the decks of two other pontoons. In some cases the old spans were deposited on trucks on falsework, rolled ashore, and dis- mantled at leisure; in other cases they were deposited on piles driven in the river and there taken apart with an ordinary traveler. The spans were replaced with an average interruption of traffic of about 4 hours. The Coteau Bridge, across the St. Lawrence River, has 14 spans about 220 ft. long. These were erected successively on shore false- work about 3 miles up stream from the site, skidded to towers on the decks of a pair of pontoons trussed together, safely towed down through a 7-mile current, and deposited on their piers. The Midland Railway crosses the Shubenacadie River, Nova Scotia, at a point where it is subject to a 32-ft. rise of the tide. The current is very swift, and the bottom under three of the 215-ft., 330 000-lb. pin- connected-truss through spans was too soft to support falsework; there- fore these spans were erected on pile falsework at right angles to the bridge axis, transferred to towers on a pair of pontoons, and hauled to position by. hoisting engines located on the bridge piers. Each span was transferred from the falsework to its permanent position in less than 2 hours. The 232-ft. span of the Fraser River Bridge, at New Westminster, British Columbia, has two trusses 54 ft. deep, 20 ft. apart at one end and 135£ ft. apart, at the other. These were erected successively on falsework in the river near a completed span of the bridge. The first truss was moved transversely to a temporary support on falsework piers, and the second truss was erected and braced to it; the heavy floor-beams and transverse struts were assembled, and the 468-ton span was transferred to three large pontoons, floated on them to position, and lowered to its bearing on its three permanent piers. Other bridges floated to position in America, are the Belle [sle Bridge, Detroit, eleven 156-ft. spans of which were erected on shore and towed nearly a mile to the site; several draw-bridges; and the Harvard Bridge, Boston, -with 23 spans ranging from 76 to 106 ft., the separate girders of which were delivered by a gantry crane to a large scow, towed, to position, and lowered to their piers by admitting water ballast and by the falling tide. The anchor arm of the long-span Tnterprovincial Bridge, across the Ottawa River, and the Faidherbo Bridge, in Africa, were ereccted DISCUSSION ON HARLEM SHIP CANAL BRIDGE Mr^ od falsework supported on boats, and the Bismarck, Omaha, and Glas- ' gow Bridges, across the Missouri River, were all erected or replaced on falsework trusses which were moved from pier to pier on pontoons. H Mr.^ Horace J. Howe, M. Am. Soc. C. E. (by letter).— In comparing ' old and new methods, the advance is found to be along these lines: First. — The use of well-bonded cribbing, in preference to framed bents on the scows; Second. — The supply of abundant steam for pumping water in and out of the scows, thereby neutralizing any uncertainty as to the tide; Third. — The use of tugs of ordinary size; Fourth. — The use of improved hydraulic jacks or presses, for adjustments; Fifth. — The use of sand jacks for loads of 2 000 tons or more, under particular circumstances. The writer thinks that much might have been said in the dis- cussion as to the last two headings. When one calculates the cost of the time frequently wasted by a high-priced gang of men, on account of defective judgment as to jacks, and compares this cost with the rental of adequate power, he is led to think that here is a new field for the engineer. Mr. Bush has taken a step ahead, in this respect. He also is in line with Mr. Stephenson and others in informing himself by previous experiments on a generous scale of what might reasonably be ex- pected; and furthermore, in allowing generously for contingencies of all kinds during the critical period. Mr. Gay's experience with the Harlem, and his proximity to the work described, render his opinions of especial value. His estimate of the loss of time to the traveling public at the bridge proved to bo nearly exact. Mr. Clarke's design, it is needless to say, involved expert work of a high order. As one examines the bridge to-day, he is struck with its difficulties, and furthermore of the subsequent difficulties of im- pressing opposing interests as to its practicability. The writer would call Mr. Belzner's attention to the quality of I he hand riveting of Mr. Stephenson's time, superseding, in fact, the steam riveting at Conway. To-day, it would be quite a question bow long it would take a power-riveter to become as truly expert at hand-riveting as one of those <»ld-t imo riveters. Attention is also called to the avoidance of painting the plates before riveting, and the use of red and white lead as a filler, both practices being in line with the writer's experience. As to the twisted shafting, probably more work was done at that corner by the wedges, in lifting the bridge, due to slowness of action relative to the other wedges. The bridge seats are known to be level. DISCISSION <>\ IIAKLKM BHIP CANAL BRIDGE 31 The writer is informed that the "clutch" couplings have worked :\ir satisfactorily. On a draw-bridge subject to deflections and vibra- How ' tions, there seems to be a certain amount of "give" necessary to a line of shafting. The writer is indebted to Mr. Skinner for his illustration of the classic falsework used at the Brunot Island Bridge, and for his other references, bringing the subject substantially up to date. AMERICAN SOCIETY OF CIVIL ENGINEERS INSTITUTED 1852 TRANSACTIONS Paper No. 1144 PRECARIOUS EXPEDIENTS IN ENGINEERING PRACTICE.* By John Hawkesvvorth, Assoc. M. Am. Soc. C. E. With Discussion hy Mi-msrk. Elcknk W. Stkrw W. W. Crosby, J. S. Branne, Andrews Allen, Guy B. Watte, J. H. GandolpOj ind John Hawkesworth. In presenting this paper for the Society's consideration, the obser- vation may be made that it is not, in the strict sense of the term, a scientific paper. It would seem, however, that if engineers h?d no thoughts or aspirations in regard to their Profession, other than those applying to technical details, that Profession, and the Society which represents it, would fall short of the standards which inspire workers in other fields. In the many scientific papers which are continually being presented, attention is directed to methods and formulas by which one may build safely and efficiently. With these two watch- words of the Engineering Profession — "safety" and "efficiency" — is it not necessary to class the additional word, "honesty"? In the Engineering Profession, as in other professions, ordinary intentional dishonesty sooner or later brings its own punishment. It is not to such self-evident facts that the writer would draw attention. I nere is another kind of dishonesty to which one may, without premeditation, be made accessory, which is known to every engineer and architect, yet very rarely discussed. Possibly this is because some may believe it is the result of the present method of financing large * Presented at the nieetinc of February 18th, 1&10.