Page - 835 - in Book of Full Papers - Symposium Hydro Engineering

The sudden increase of Qa,o for a/amax ≈ 0.05 is caused by the transition to spray flow. At large a/amax the air flow rate is decreasing again, especially for large HE. This is due to the transition to foamy flow with the tunnel almost flowing full at the tunnel end. The reduced cross-sectional area above the air-water mixture for foamy flow allows less air to be dragged along with the flow. Qa,u is negative for moderate a/amax, low ζ value and long tunnel, indicating a net air flow out of the tunnel at the downstream end. For foamy flow conditions Qa,u increases to zero, whereas spray flow leads to a slight decrease of Qa,u (Fig. 2b). An increase in ζ drastically reduces Qa,o, while simultaneously increasing Qa,u, resulting in a net air flow into the tunnel from the downstream (Fig. 2). The decrease in Qa,o and the increase in Qa,u are even more pronounced for short tunnels as air can enter more easily from the downstream tunnel end. The minimal air pressure in the gate chamber pa,min depends on Qa,o and ζ. With all other parameters kept constant, an increase in ζ leads to a simultaneous decrease of Qa,o and pa,min. The magnitude of decrease of each quantity in turn depends on a/amax, HE and L. 3. PROTOTYPE MEASUREMENTS 3.1. TEST SITES The two BOs of the arch dam Malvaglia and the BO and middle outlet (MO) of the arch dam Luzzone in the Canton of Ticion, southern Switzerland were equipped with measurement devices. All outlets are usually operated once a year to flush sediments out of the reservoir. Additionally, a mandatory function test requires the test of small gate openings once a year. All three outlets are equipped with sluice gates. Luzzone features a maximum static head Ho,m of 224 m w.c. for the BO and 120 m w.c. for the MO. Malvaglia has two BOs, an old and a new one with Ho,m = 94 m w.c. and Ho,m = 91 m w.c., respectively. The new BO was specifically built to flush sediments from the intake vicinity. The old BO features a circular tunnel with a diameter of d = 4.3 m, whereas the new BO has a 2 m wide, 2.5 m high horseshoe profile. Both outlet tunnels join a common outlet tunnel (d = 5.6 m) to which the spillway is also connected (Fig. 3). The tunnel conjunction is well aerated what leads to effective tunnel lengths of L = 67 m and L = 82 m for the old and new BO. Up to now only measurements during a function test in Malvaglia could be conducted. 3.2. PROTOTYPE INSTRUMENTATION The instrumentation applied at Malvaglia Dam and their accuracy are listed in Table 1, and a sketch of the measurement setup is shown in Fig. 3. 835