two folding wings built on either side were used for this operation. The drills were mounted on the wings, curved on a 28-foot radius, and the whole of the invert section drilled in one operation.

Trimming and scaling operations were performed coincidently with, and a short distance ahead of, the invert excavation. A horseshoe-shaped steel framework with an outside diameter of 50 feet, mounted on wheels which traveled on rails laid true to line and grade was used to measure the rocks projecting within the allowable clearances.

For the purpose of applying concrete lining, the circular section was divided into three parts-the lower 74° for the invert, to be placed first; above this section, on either side, the side wall portions of 88° each, to be placed second; and above the side wall sections the roof or arch section of 110°, to be placed last.

Specifications required an average lining thickness of 36 inches with a 24-inch minimum. Concrete for the invert and side wall sections was composed of 1 part cement, 2.1 parts sand, and 4.7 parts gravel. Aggregate was graded as follows: 4 to 4 inch, 32 percent; % to 11⁄2 inches, 32 percent; 11⁄2 to 3 inches, 36 percent. Concrete for the arch section was 1 part cement, 2.5 parts sand, and 4.3 parts gravel. The larger size of aggregate, 11⁄2 to 3 inches, was omitted from this mix. The slump of the concrete used in the various sections was: invert, 3%1⁄2 inches; side walls, 4%1⁄2 inches; and arch, 5 inches.

Concreting operations were started at the upstream portals and carried progressively toward the downstream ends of the tunnels. All concrete was mixed at the low level plant, located a short distance upstream from the tunnel inlets on the Nevada side of the river. Truck transportation was used exclusively, consisting of either two 2-cubic yard form buckets, carried on platform trucks, or 4-cubic yard agitator bodies.

The first step in the lining program involved the construction of continuous concrete shelves to serve as rail bases for a 10-ton electrically operated gantry crane, to be used for concrete placing. Invert forms, consisting of steel side. plates carrying curved steel screeds, were then set in place and pouring started. Concrete was hauled by truck to the forms where the gantry crane took up the buckets and deposited the contents, one on either side of the screeds. The concrete was then puddled into place and when sufficiently high the screeds were moved outward toward the side walls to form the required concave surface. The surface was finished by workmen on a movable platform suspended just above the concrete, using wooden floats and steel trowels.

Bulkheads for contraction joints in the invert were provided every 40 feet, except in those portions of the two outer tunnels which were later to be used as spillways, where the distance was 26 feet 8 inches. This same spacing of contraction joints also applied to the side walls and arch sections. 'The use of the invert for trucking in subsequent lining

work necessitated the placing of about 3 feet of fine rock spoil, to protect the concrete. After the invert section was placed, two concrete rail shelves were constructed on the the invert to support the side wall and arch form equipment. A huge structural steel framework or jumbo, weighing about 270 tons for an 80-foot section, was provided to support the wall forms which were composed of 4-inch steel plate. The jumbo was completely equipped to handle concrete with an electric crane operating on top of the form and a system of chutes, seven high on each side of the form, spaced horizontally 9 feet center to center.

Placing was started through the lowest row of chutes and carried progressively upward. Proper rotation was maintained to place each side of the 40- or 26-foot 8-inch section simultaneously. Men were stationed back of the forms to puddle the concrete against the rock. As the concrete rose in the forms, the chutes were withdrawn and the form openings covered with steel plates.

Temporary timber bulkheads were used to provide construction joints between pours. These bulkheads were framed to form a simple offset keyway 11⁄2 inches deep and 10 inches wide. A keyway, approximately 10 inches wide and 2 inches deep, also was formed on top of the side wall sections to receive the arch section. With the completion of two 40-foot pours, or three pours of the shorter section, the form was left in position for 12 hours, then moved to the next position.

The forms and concrete placing equipment for the arch section included an 80-foot length of form supported by a structural steel framework or jumbo, a separate gun carriage containing two 2-cubic yard pneumatic concrete guns with the necessary hoisting equipment, and a separate traveler connected to the gun jumbo to support two placement pipes. This equipment was operated on the same rails that were used for the side wall jumbo and was kept not less than 150 feet behind the completed side walls.

Four-cubic-yard agitator bodies were hoisted off trucks and the concrete divided between the two guns. Operation of the concrete guns followed standard procedure, using a compressed air supply. A large air receiver was used with each gun and the nominal pressure was about 100 pounds, although this often dropped nearly 50 percent at the finish of a 2-yard charge of concrete. From the guns, 6-inch discharge pipes carried the concrete to the arch forms.

Curing of the invert concrete was performed by the roadway which kept the surface moist. A 2 week's spray cure was provided by the specifications, but when this procedure was attempted on the side walls the water collected in the roadway, softening the road. Also, during the summer months the temperature in the tunnels was extremely high, and the humidity resulting from the spraying made working conditions unbearable. As a result, an asphalt oil coating, applied with compressed air (the Hunt process), was used for curing.

Concrete arch cofferdams, flanked on the river side by rock fills, were constructed at each portal to keep out water. Wooden trestles with ramps were provided over these barriers to reach the tunnel floors.

Low-pressure grouting was performed in the arch section. of the tunnels to fill the voids between the lining and rock. Three holes, 1%1⁄2 inches in diameter, were drilled through the lining in the crown, one in the center and one at an angle of 45° from the perpendicular on either side. These holes were drilled at intervals of 20 to 26 feet 8 inches along the center line. Special holes were provided at all points of overbreak. Grout was applied at from 50 to 100 pounds pressure, using a duplex, air-driven, piston grout pump. mounted on a truck with the mixing apparatus. Low-pressure grout holes were drilled through the invert and lower side walls to grout the temporary drainage sys

tems in the tunnel invert. Grout was applied under a pressure of 50 pounds, prior to the placement of the highpressure grouting.

High-pressure grouting, using pressures from 100 to 500 pounds per square inch, was accomplished through a ring of eight holes, drilled radially into the rock to a depth of 24 feet, holes in successive rings being staggered. Drainage holes, spaced 28 feet 8 inches apart, were drilled after grouting was completed, to drain water from back of the lining. A total of 122,000 linear feet of grout holes were drilled in the diversion tunnels and more than 200,000 cubic feet of grout applied.

Progress in driving and lining the tunnels made possible the diversion of the river through the two Arizona tunnels on November 13, 1932, less than 2 years after excavation of the tunnels was commenced in June 1931 and almost a

year in advance of the date contemplated at the time the contract was awarded.

COFFERDAMS

The

With the river diverted through the tunnels during the winter season of low flow, work was concentrated on the construction of the cofferdams to complete the river diversion program before the arrival of the spring floods. upstream cofferdam, an earth and rock fill dam, was located 600 feet below the diversion tunnel portals. Approximately 250,000 cubic yards of river silt and loose deposit were removed to secure an adequate foundation of consolidated sand, gravel, and boulders.

The structure as completed was 98 feet from base to crest and 30 feet above the top of the diversion tunnels. A freeboard of 13 feet was provided for a discharge of 200,000 second-feet through the tunnels. The dam was 450 feet in length, 750 feet thick at the base, and contained 516,000 cubic yards of earth and 157,000 cubic yards of rock. The upstream face of 3:1 slope was protected by 6-inch concrete paving, laid on a 3-foot thickness of rock blanket; and the downstream slope of 4:1 was covered by a heavy rock fill. The paving consisted of a 6-inch concrete slab, reinforced with-inch round bars on 15-inch centers, spaced both ways. The reinforcement was continuous through the construction joints which extended up and down the face on 16-foot centers. The slab extended down to and connected with a line of steel sheet-piling, driven to bedrock along the upstream toe. The cut-off wall was joined to the face paving by a concrete wall and rubber seal which was protected by a puddled clay fill and a 5-foot thickness of dumped rock. A V-shaped sheet rubber water stop, extending from the steel piling to the crest of the dam, was provided along the joints of the paving slab and the canyon walls. Three reinforced concrete percolation stops, projecting at right angles from each canyon wall, were provided on each side of the canyon.

The downstream cofferdam was a rolled earth-fill structure, 66 feet high, 350 feet long, and 550 feet thick at the base. It contained about 230,000 cubic yards of earth and 63,000 cubic yards of rock. This structure was also founded on consolidated material of sand, gravel, and cobbles, after the loose riverbed material had been removed. The downstream slope was 5:1, and the upstream slope, 2:1. The downstream slope was protected by a thick rock blanket.

A rock barrier, 54 feet high, 375 feet long, 200 feet thick at the base, and containing 98,000 cubic yards of rock, was provided about 365 feet downstream from the lower cofferdam, to protect the earth fill from the backwash of the river during flood discharges. The cofferdams, rock barrier, and the Nevada diversion tunnels were completed in March 1933, in advance of the spring flood flows.

With the river out of the canyon, work was concentrated

on the excavation of the dam site. Stripping the canyon walls of loose rock was performed by men, lowered from the top of the cliffs in rope slings, using bars, jackhammers, and blasting, beginning from the top of the canyon walls and working down along the required excavation line.

Crawler-mounted electric shovels, loading directly into trucks, were used in excavating the foundation for the dam and exposing the bedrock in the area between the limits of the cofferdams. This involved more than 800,000 cubic yards of gravel and 100,000 cubic yards of rock. Excavation of the river bed revealed conditions to be essentially as indicated by the exploratory drilling. A gorge about 75 to 80 feet below the rock benches on either side of the canyon was uncovered along the center line of the canyon.

FOUNDATION GROUTING

Wherever concrete was poured against rock, advance provisions were made for forcing grout into rock fissures and seams, at intervals and depths consistent with the nature of the rock formation and its location with respect to the dam. The main grout curtain or cut-off for the dam included three systems of holes. The first or "B"hole grouting was designed to provide a leakproof layer under the upstream part of the dam, below which the deeper and high-pressure grouting could be performed. These holes, spaced 20 feet on centers, were drilled 30 to 50 feet in depth and grouted before any concrete was placed. Neat standard portland-cement grout having a water-cement ratio ranging from 1.0 to 5.0, by volume, was injected to practical refusal at 300 pounds per square inch pressure. Grouting equipment consisted of a grout mixer, a mechanically operated sump, and two highpressure mud pumps mounted on a truck for mobility. More than 5,200 linear feet of "B" holes were drilled, taking a total of 6,664 sacks of cement.

A total of 1,418 feet of holes was drilled along two faults, situated just upstream from the dam, and 1,064 sacks of cement were applied under pressures ranging up to 500 pounds per square inch. This fault grouting conformed to the "B"-hole grouting, of which it formed a component part. Wherever springs or shattered areas were encountered in the rock foundation, holes were drilled and pipe provided for future grouting after the dam was practically complete. A system of piping over the entire abutment areas, with outlet boxes spaced approximately 10 feet apart in a checkerboard arrangement, was provided for grouting the abutment contacts after the reservoir is filled and the maximum deflection of the dam has taken place.

"A"-line grouting, which forms the cut-off curtain under the dam, was not started until at least 100 feet of concrete had been placed over the site of the hole, the concrete cooled, and the joints grouted. The following maximum. pressures governed the introduction of the grout:

Drilling was carried to a maximum depth of 150 feet from the main grout gallery in the dam which closely follows the rock foundation. Before concrete was placed in the dam, holes were drilled 5 feet into the rock and pipes leading to the gallery were embedded in the concrete on 5-foot centers. A total of 54,400 linear feet of holes was drilled and 60,024 cubic feet of grout applied.

"C"-line grouting, to provide a supplemental cut-off under the heel of the dam, followed the upstream outline of the dam from the lowest point of the foundation to elevation 775 on either abutment. The holes were drilled on a 10-foot spacing to a maximum depth of 100 feet and were inclined downstream. Grout was placed with the portable, truck-mounted equipment, under pressures varying from 500 to 750 pounds per square inch. A total of 7,325 linear feet of holes was drilled and 7,106 cubic feet of grout applied.

The grout curtain was extended into the abutments by deep drilling and grouting from the diversion and penstock

tunnels as described under high pressure grouting for the diversion tunnels.

THE DAM

Boulder Dam is a concrete arch-gravity structure in which water load is carried by both gravity and arch action. The length along the crest is 1,282 feet, the maximum height as measured from the lowest point of the foundation rock to the roadway crest is 726.4 feet, and the widths up and down the stream are 45 feet at the top and 660 feet at the base. A total of 3,250,000 cubic yards of concrete was placed in the dam, and 4,400,000 cubic yards are contained in the dam, power plant, and appurtenant works.

The technical design of the dam was based on three fundamental requirements; first, that the dam should be of the massive arch-gravity type; second, that the base of the dam should be located wholly within the area bounded by the two adjacent fault lines in the canyon rock; and third, that the maximum compressive stress in the dam should not exceed 30 tons per square foot.

In the preliminary design studies for the dam, it was attempted to proportion the dimensions so that the maximum stresses would not exceed 30 tons per square foot. Analyses

of 35 tentative plans, including all practical variations in horizontal curvature, many different thicknesses of crosssection at all elevations, and widely varying slopes of upstream and downstream faces, showed that owing to the great height of the structure, such an ideal was impossible. When the design of the dam was selected, detailed analyses of stress magnitude and distribution were made by the trialload method. The analyses of the adopted design, assuming. a straight line variation of stress, showed maximum cantilever compressive stresses of approximately 40 tons per square foot at the upstream edge of the base. However, the cantilever stresses at the downstream face of the dam and the maximum arch stresses in the horizontal elements were well below 30 tons per square foot. Resultants were at approximately the same location for both empty and full conditions of the reservoir, due to the great size of the structure and the relatively flat upstream slope.

The effect of nonlinear distribution of stress was found to increase the arch and cantilever stresses along the planes of contact between the concrete and rock from 15 to 25 percent at the locations of higher concrete stress. Earthquake stresses were considered but necessitated no modifications in the adopted dimensions of the dam.

The action of the dam and resulting stress conditions

under temperature and water loads were checked by detailed experimental investigations on two models of the entire structure, a slab model of the crown cantilever, and a slab model of an arch element at an elevation about halfway between the top and bottom of the structure. One model of the complete dam and two slab models were built of a plaster and celite mixture. The other model of the entire dam was constructed of a rubber-litharge compound.

Results obtained from the model tests furnished very satisfactory confirmations of the trial-load analyses and special stress studies, and indicated the adequacy of the design. The desirability of certain modifications in design details also was indicated. Long-radius fillets were added at the downstream face of the dam near the arch abutments to reduce the compressive stresses at the intrados, and a short-radius fillet was provided at the contact between the concrete and rock along the upstream face of the dam.

The large volume of concrete to be placed made logical the initiation of a concrete research program to investigate the unprecedented problems involved and to insure a satisfactory mass concrete structure. A cement and concrete research program was instituted, covering tests of ultimate compressive stress, permeability, Poisson's ratio, modulus of elasticity, sliding friction, bond at horizontal construction joints, variations of strength with age and curing temperature, proper gradation of fine and coarse aggregates, effects of vibrating fresh concrete, determination of proper mix, and numerous problems of a thermal

nature.

Two of the most important and difficult problems in the design and construction of the dam were the generation of heat in a large mass of concrete due to hydration of cement, and the volumetric changes occurring in mass concrete due to temperature changes and other causes. Provisions were made to include circumferential contraction joints in the dam and to lower the temperature of the concrete to normal, or subnormal values, by means of an artificial cooling system.

Numerous tests and studies were made to determine the best cement for use in the dam and to assist in the control of maximum temperatures, thus decreasing the total amount of heat to be extracted. By controlling the chemical composition of the cement, lower heat-generating qualities and greater durability were obtained. Increased workability of the concrete was effected by closer control of the fineness; and greater uniformity resulted from blending and control of the composition.

Mass concrete for the dam was designed to contain one barrel of cement per cubic yard of concrete. The maximum size of aggregate was limited to 9 inches to facilitate mixing and placing operations. The mix, as finally adopted, was composed of one part cement, 2.45 parts sand, and 7.05 parts gravel, by weight, graded as follows: