The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

THE CONTRACTOR'S PLANT AND METHODS ON MARE ISLAND DRY DOCK NO. 2 (From The Engineering Record.) (Continued from Oct. 8.) The portion of the piles bearing against the walings of the crib acts as a continuous beam when the water is pumped out of the cofferdam. The length between the lowest waling and some point of support in the mud was also known to-be a part of this con- tinuous beam, if that point of support could be determined to be sufficiently rigid, as it evidently was from the driving. In order, however, to locate accurately the line of support for the lower end of the-sheet piles, the clus- ters of the round piles used to sup- port the guides of the crib were snapped off. The lowest grade at which these piles. broke was approxi- mately at the dredging line, which demonstrated that the part of the sheet piles below the lowest waling also acts as a continuous beam, with a maximum span of 12 ft. between the. lowest -supports, which was assumed to be the worst condition. A section of pile tested under a 12-ft. span, with.one end of the beams free and the other end partially fixed, car- ried a load of 5,000 pounds per square foot. with a maximum deflection of 1 in., while the maximum load on the sides of the cofferdam is only about 3,000 pounds per sq. ft. and is distrib- uted under more favorable conditions. After the sides of. the cofferdam had been made as tight as possible under water, pumps were started to remove the water from the enclosed space in order to locate further leaks and to determine the strength of the struc- ture. As the water level was lowered such few leaks in the sheeting as oc- curred were plugged, and eight 3 x 3- ft. flood gates, which had been placed in the sides and the outer end of the cofferdam, were caulked. Two 15-in. centrifugal pumps lowered the water to the fourth course of timbers from the top of the crib, and after that depth was reached two 8-in. centrifu- gal pumps were added. Three-quar- ters of the total leakage up to this time was through the butt joints be- tween the two timbers of each sheet pile. The hard driving had forced the oakum out of these joints, which it was then evident would have been better with a pin connection as a stop-water. The first row of these butt joints was plugged with shin- gles from the inside after the water had been drawn down, and then the water was lowered to below the sec- ond row, 26.5 ft. below mean tide distance of about 250 ft., TAE Marine REVIEW level, where one 15-in. and one 8-in. pump held it until those joints were also plugged. Meanwhile, a 1%-in. opening be- tween the regular sheet piles and a group of four patch piles, bolted to- gether, near the head of the dock scoured a V-shaped hole, 9 ft. deep, on both sides of the sheeting in the conglomerate formation. This hole, when filled with concrete in. bags on the inside, and clay on the outside of the sheeting, gave no further trouble. The following day another leak oc- currred near the head of the dock, where five of the sheet piles had a penetration of from only 2% to 3% ft., having landed on a pocket of dis- integrated sandstone. Irregular strata of shale and clay run diagonally under the sheeting close to this point, with occasional pockets of this disintegrated sandstone, which slakes rapidly in water or in air. The water had under- cut through one of these pockets, and the leak, which appeared out some 25 ft. from the shéeting in the cofferdam, was of such magnitude that the flood gates were opened and the cofferdam allowed to fill. Afterwards, 110 cu. ft. -of concrete: were placed in this hole inside of the sheeting, and 10 cu. yds. outside before it was filled. Seven-eighths of the total length of the sheeting walls were driven 12 to 14 ft. in good clay, in which no un- dercutting would. occur, and in this part of the cofferdam the only danger was from an opening in the sheet pil- ing acting asa jet to cut its way. down through the clay. These places had been detected and soon closed, so no trouble was anticipated from them. The decision was made, therefore, to puddle around the outside of the sheet- ing at the head of the dock for a where the piles had penetrated 7 to 10 ft. but were in the broken, irregular forma- tions of clay and shale. Meanwhile, the fissure in the disintegrated sand- stone, which had cut down 9 ft. below grade, had to be carried 8 ft. farther before the bottom of the pocket was reached. . This hole was then filled with concrete in bags by a diver, and the puddle placed over it. When the puddle had been placed around the head of the dock to a depth of 6 to 10 ft, the water was again pumped out of the cofferdam. Three pumps, an 8-in. and two 15-in. centrifugals, required nine days to lower the water level to the heads of the piles. It had been anticipated that less time than this would be neces- sary to unwater the cofferdam, but the strips of canvas over many. of the ea 31 cracks had rotted, since the canvas only lasts about two weeks under water. These openings were there- fore wedged up on the inside with shingles as the water was pumped down. ; During the period of pumping from the second course of timber in crib down, a patrol of four' men was con- stantly searching for signs of scour with long poles. The water level was held down to the pile heads for two days with no signs of a bottom leak, and no evidence of weakness in the cofferdam. A scour then occurred near the head of the dock in a pocket of conglomerate underlying a bed of blue clay. This conglomerate was washed out, leaving the clay sup- ported against the sheeting, and ap- peared 24 ft. inside the latter, Ef- forts to stop the leak were of no avail, so the flood gates were again opened in order to fill the cofferdam and prevent any serious consequences. At the time the cofferdam was flooded there were no evidences of any other scours or leaks, the water level being held down. by one 15-in. centri- fugal pump. The significance of this condition can scarcely be appreciated, as the sheeting offered a wall, 1,960 ft. long, with about 11 miles of joints in it, against which there was a sheer head of water of 48 ft. at high tide. Prac- tically the only leakage into the cof- ferdam was through cracks below the lowest course of walings; which did not produce sufficient water to flush out the accumulation of silt over the piles. In fact, four 4-in. syphons and two 4-in. discharge steam pumps sup- plying six lines of 4-in, hose were all discharging water into the off-shore end of the cofferdam for this purpose when the latter was filled the second time. The water which leaked into the cofferdam, added to that from the two 4-in. pumps and the four syphons, was not more than the 15-in. centrifu- gal pump could handle. The strength and general tightness of the long, high walls of the sheet piling having thus been fully deter- mined, it was evident that underscour- ing through pockets in the unusually irregular strata of clay and shale at the head of the dock was the chief danger to be overcome. By any other means, except in an open caisson, the mud and silt which had accumulated over the heads of the piles to a depth of as much as 6 ft. would have to be removed under water and the bottom of the cofferdam sealed with concrete in the same manner, all of which would have been done with great difficulty. The decision was eae ne made to