Validations of CFD against detailed velocity and pressure measurements in water turbine runner flow

This work compares CFD results with experimental results of the flow in two different kinds of water turbine runners. The runners studied are the GAMM Francis runner and the Holleforsen Kaplan runner. The GAMM Francis runner was used as a test case in the 1989 GAMM Workshop on 3D Computation of Incompressible Internal Flows where the geometry and detailed best efficiency measurements were made available. In addition to the best efficiency measurements, four off-design operating condition measurements are used for the comparisons in this work. The Holleforsen Kaplan runner was used at the 1999 Turbine 99 and 2001 Turbine 99 - II workshops on draft tube flow, where detailed measurements made after the runner were used as inlet boundary conditions for the draft tube computations. The measurements are used here to validate computations of the flow in the runner. The computations are made in a single runner blade passage where the inlet boundary conditions are obtained from an extrapolation of detailed measurements (GAMM) or from separate guide vane computations (Holleforsen). The steady flow in a rotating co-ordinate system is computed. The effects of turbulence are modelled by a lowReynolds number k-(omega) turbulence model, which removes some of the assumptions of the commonly used wall function approach and brings the computations one step further. The computational results are compared to the measurements with respect to detailed velocity and pressure distributions at several measurement sections and several operating conditions. The computational results are also compared to coarse grid wall function computations by the TASCflow CFD code, which uses a similar methodology as the CALC-PMB CFD code used in the present work. The comparisons show where the computational method is sufficient and where it is not sufficient.The behaviour of the computational method is similar for both kinds of water turbines, which shows that experience of computations in water turbines will ultimately give quantitatively correct results