PIV MEASUREMENTS IN A HIGH-SPEED LOW-REYNOLDS LOW-PRESSURE TURBINE CASCADE

Particle image velocimetry (PIV) measurements in the blade-to-blade (B2B) plane and cascade outlet plane (COP) of a high-speed low-pressure turbine (LPT) cascade were performed. The measurements were performed at engine-representative outlet Mach number (0.70-0.95), and Reynolds number (70k-120k) under steady flow conditions. The freestream turbulence characteristics were imposed by means of a passive turbulence grid. The PIV results on the B2B plane were compared against five-hole probe (5HP) and Reynolds-averaged Navier-Stokes (RANS) computations to assess the validity of measurement and simulation techniques in the engine-relevant LPT cascade flows. The PIV captured the wake depth and width measured by the 5HP whereas the RANS displayed an overprediction of the wake Mach number deficit. The 5HP was found to impose sinewave fluctuations of the measured flow angle downstream. The variation of fluctuation captured by the PIV was around three times lower. In addition, PIV provided an estimation of the turbulence intensity (TI) in the cascade passage, displaying a decay along a streamline following the passage curvature. For the highest Mach number investigated, a peak in the TI was measured past a shock wave. Measurements of the outlet flow field highlighted a high TI in the secondary flow region whereas high degree of anisotropy (DA) was registered in the boundary of the secondary flow and freestream regions. The contribution of the streamwise fluctuation component was found to be less than the crosswise and radial components in the freestream region. Increasing the cascade outlet Mach number, the contribution of streamwise fluctuation to the DA was observed to decrease.