Air Demand and Flow Patterns of Low-Level Outlets: Accounting for Wall Roughness: Fid-Bau Portal

Air Demand and Flow Patterns of Low-Level Outlets: Accounting for Wall Roughness

Pagliara, Simone / Felder, Stefan / Hohermuth, Benjamin / Boes, Robert Michael

Low-level outlets (LLOs) are key safety elements of high-head dams, typically consisting of a pressurized inflow controlled by a vertical sluice gate that discharges into a free-surface flow tunnel. The transition from pressurized to free-surface flow generates a high-velocity water jet with considerable air entrainment and transport along the tunnel, resulting in subatmospheric pressures downstream of the gate. These conditions potentially aggravate serious safety issues such as cavitation, gate vibration, and, in combination with sediment transport, hydroabrasion. Sufficient air supply can mitigate these problems. Several empirical equations have been developed to predict the air demand of LLOs, incorporating the effects of flow patterns, air vent loss coefficient, and tunnel geometry. However, reported model and prototype air demand data scatter over one order of magnitude, with tunnel roughness identified as a potential reason for these differences. To date, the influence of wall roughness on the performance of LLOs has not been systematically investigated. In this study, physical model tests were conducted with varying wall roughness, representing finished concrete, abraded concrete, and unlined rock at prototype scale. The results showed a significant increase in the air-water mixture flow depth with increasing wall roughness, where excessive filling of the tunnel may trigger foamy flow, flow choking, and the formation of hydraulic jumps, resulting in severe degradation of LLO performance. Increased wall roughness also led to a higher air demand, suggesting a predominant effect of the invert roughness over the wall and soffit roughness. A novel empirical equation was derived for air demand, incorporating the effects of tunnel roughness. The equation showed good agreement with previous laboratory and prototype data, indicating that other design parameters were not affected by the tunnel roughness. Finally, design recommendations were updated to account for roughness effects in LLO design, thereby contributing toward a safer design of these structures.