Experimental and numerical results of LP turbine cascade tests performed to investigate wake interaction effects on boundary layer transition are presented. The data obtained at different inlet flow angles, turbulence levels and Mach numbers are compared and discussed with special focus on low Reynolds number conditions. For the boundary layer, calculated propagation velocities of disturbances are introduced to explain the transition process on the suction side of the blade over time.

Using a moving bar wake generator and surface-mounted hot films as well as surface pressure tappings, the effects of periodic wake passing were studied in the High-Speed Cascade Wind Tunnel on the aft-loaded LPT profile T106. Blade pitch was increased as compared with design point conditions to achieve a higher blade loading. As a result, a large separation bubble formed on the suction side of the surface and allowed unsteady boundary layer development to be studied in great detail. Starting at a characteristic Reynolds number, massive separation occurred on the suction side under steady state conditions, i.e. the boundary layer was unable to reach the back pressure at the trailing edge. By using the wake generator, it was possible to reduce this separation and thus decrease profile pressure losses by 50%.

The primary objective of the study was to provide unsteady ensemble-averaged hot film data together with information on the wake induced path, sufficient for the validation of numerical simulations. Such a simulation of the experiment was conducted using the Unsteady Boundary Layer Interaction Method, which takes into account the influence of boundary layer displacement on the velocity distribution and the time-dependent turbulence level in the outflow. The computations provide a good description of the wall shear stress in the transitional region and are in good agreement with the experimental data. By plotting propagation directions of boundary layer disturbances in space-time diagrams, it is shown that one characteristic direction is deviated around the so-called becalmed region and the temporarily separated region into the wake-induced transitional region.

This content is only available via PDF.
You do not currently have access to this content.