Figure 102.-Spillway discharge curves for one 40- by 52.5-foot radial gate.

Crest - El 3648 00

Sili-E1 364750

£1 3635 0

(4) The spillway approach channels were greatly reduced from their original size and still provided extremely smooth flow conditions.

(5) Flow through the crest sections was excellent and no adverse pressure conditions were noticed. The maximum discharge of 138,000 cubic feet per second per tunnel was obtained at reservoir elevation 3711, the value used for design purposes.

(6) The preliminary tunnel transition was too abrupt. A surface fin formed in the center of the tunnel and pressures on the sidewalls were in the cavitation range.

(7) The longer recommended transition (fig. 104) was adequately streamlined and provided smooth flow conditions with no adverse pressures on the sidewalls.

(8) Flow in the 41-foot-diameter tunnels was excellent at all discharges.

(9) The preliminary downstream circular-to-rectangular tunnel transition was too short, as indicated by severely subatmospheric pressures in the lower corners. Increasing the transition length from 70 to 100 feet increased the pressures to a satisfactory value. This transition was eliminated in the recommended design.

(10) The preliminary deflector buckets, which were rectangular in cross section, were replaced by a bucket in which the circular invert of the tunnel intersected the vertical curve of the bucket. This type of bucket also eliminated the need for the circular-to-rectangular transition.

(11) The deflector buckets were moved upstream to the tunnel portals, eliminating about 200 feet of open channel.

(12) The outside walls of the buckets were turned inward 7 feet to direct the flow in a more favorable pattern at their impact points.

(13) The outside wall of the left bucket was extended 32.5 feet downstream from the lip to deflect the flow from the canyon wall.

(14) Pressure measurements on the wall and invert of the left bucket showed that pressures equivalent to 211 feet of water would be considered in the structural design of the bucket.

44. STRUCTURAL DESIGN. (a) Intake Structure. -The layout of the intake structure is shown on figure 103. The design was based on concrete having a compressive strength of 3,000 pounds per square inch at 28 days. The allowable working stresses, except as noted below, are shown on figure 71.

The crest and piers were designed for earthquake, hydrostatic pressure, uplift, and service bridge loadings. Earthquake loadings of 0.1 g horizontally and 0.05 g vertically were used. The horizontal component of thrust from the radial gates is carried directly to the rock through a massive beam above the portals of the tunnels. This beam was designed to support the vertical component of the thrust as well as the reaction of the weight of the gate and trunnion assemblies. The bridge was designed for the load imposed by the gate hoists and for a 20-ton truck-mounted crane.

The period of vibration of the center pier was found to be in the earthquake resonance zone and, therefore, this pier was designed for 0.7 g earthquake; however, the allowable stresses in the reinforcement were increased to 40,000 pounds per square inch for this condition. A split sewer-pipe drainage system was placed in the walls and on the foundation of the crest structure for uplift pressure relief.

(b) Tunnel. -The arrangement of each tunnel and transition is shown on figure 104.

The thickness of the tunnel lining for the circular sections was made 0.8 inch per foot of the tunnel diameter. The lining was reinforced where required for tension caused by the dam and reservoir waterloads. The stresses in the vicinity of the tunnels were determined by Newmark's method of computation of stresses in elastic foundations. Stresses in the lining caused by the stresses in the rock were determined by both analytical and model studies.2

Moments, thrusts, and shears in the noncircular transition section of the tunnels were determined by photoelastic model studies of various sections for hydrostatic load, dead load, grout load, and rock load based on the assumption that the height of the rock acting on the structures was equal to 0.35 of the excavated width. A grout load of 3,600 pounds per square foot over a 90° sector of the arched roof, or between spring lines, whichever was least, were used for design. The results of these studies were used to determine final concrete thickness and reinforcement.

2 Interoffice Memorandum to Chief, General Engineering Branch, from H. Boyd Phillips and Ira E. Allen, "Electrical Analogy Tray Study of Pressures on Spillway Tunnel Lining, Glen Canyon Dam," October 29, 1956 (unpublished).

3 Interoffice Memorandum to Chief, Dams Branch, from H. Boyd Phillips and Ira E. Allen, "Stress Analysis of Transition Section, Spillway Tunnel-Glen Canyon Dam," November 19, 1957 (unpublished).