Figure 137.-Penstocks gate hoist structure-Elevation and sections.

Figure 138.-Penstocks gate hoist and stem storage platform, and gate erector platform-Plans and sections.

12-72-62 ADDED RECEPTACLE BOX. 02216

THIS DRAWING JUPERSEDES DB 357-0-107 PRET

UNITED STATES

DEPARTMENT OF THE INTERIOR
BUREAU OF RECLAMATION
COLORADO EVER STORAGE PROJECT

MOBLE RIVER IN GLEN CARTON UNIT - ARIZONA - VENN

GLEN CANYON DAM

loads:

The footings were designed for the following

(1) Dead loads.

(2) Penstock reactions.

(3) Fill material (partly saturated).

(4) Earthquake effects.

A minimum of 1 percent of vertical reinforcement was placed in the H-columns.

56. TAILRACE TRAINING WALLS. Tailrace training walls are provided on each side of the tailrace downstream of the powerplant. On the left side of the upstream 75 feet of the wall is a gravity section and the rest of the wall is cantilevered off the mass concrete under the river outlets and hollow-jet valves. It is shown on figure 114. It also serves as a retaining wall to retain the fill for the powerplant parking area. The right training wall is of the line-drilled type and was doweled to the rock, figure 139.

The left training wall was designed for the following loads:

(1) Parking area fill material (saturated).

(2) 100-ton trailer unit including impact from bumping curb.

(3) Earthquake effect.

57. TAILRACE SLAB. To insure a minimum tailwater for the turbines, a tailrace slab (fig. 140) was constructed between the tailrace walls. This slab sloped up on a 6 to 1 slope from the draft tubes to form a weir at elevation 3132, 180 feet downstream of the power plant. From the weir the slab sloped down until it was approximately 20 feet below riverbed. It was originally planned that the tailrace channel and weir would be constructed of riprap, but lack of suitable rock within economical haul distance dictated the change to a reinforced concrete slab. The riprap weir was tested in a hydraulic model.1

The maximum drawdown rate of the tailwater over the weir was established at 10 feet in 10 minutes. Studies showed that porous concrete drains at 10-foot centers each way provided adequate relief from uplift

pressures so that, in general, an 8-inch slab would be stable. Near the left training wall, studies indicated that saturation of the fill and bedrock behind the wall caused higher uplift adjacent to the wall and the slab thickness was increased to 12 inches. At the weir, the slab was made 12 inches thick to provide additional stability for the weir.

During the April through July runoff period in 1965, discharges up to 50,000 cubic feet per second were released in order to obtain a rated head at Lake Mead and maintain Lake Powell at about elevation 3490. The discharge was obtained by using combinations of the discharges from the turbines, the diversion tunnel outlets, and the river outlets. The river outlets were used to their design capacity when the tunnel outlets were closed.

During the night of April 20, 1965, the slab was undermined and portions of it sank from sight. At the time of the failure the reservoir water surface was at about elevation 3490, four of the eight units in the powerplant were operating, the diversion tunnel outlets and outlet No. 1 were closed, outlet No. 2 was 25 percent open, outlets No. 3 and 4 were 90 percent open, and the tailwater was 4 to 6 feet lower than had been predicted and used in the model studies. Operating the river outlets at reservoir water surface elevation 3490, instead of at normal water surface elevation 3700 as used in the model, caused the jets to impinge closer to the weir.

Observation of flows in the tail race area, after the failure, showed a reversal of flow due to a large eddy extending to the location of the weir. This reverse flow had an unexpected velocity estimated to be 15 feet per second. The operation of the outlets at the lower head, operation of only four powerplant units instead of eight, and the lower tailwater all intensified the eddy action over that predicted from the model. The scour from this reverse flow, caused by the large eddy, undermined the toe of the slab and rapidly removed large amounts of fine bedding material, causing collapse of the slab. The extensive use of the river outlets at near full capacity with the reservoir at about elevation 3490 and with only four of the powerplant units operating had not been anticipated and had not been checked in the model.

To date (June 1969) the floor slab has not been repaired or replaced, as observation has indicated no further significant erosion in the tailrace floor.