Characterization of the Transonic Flow in a Turbine Cascade using High-speed PIV

Abstract

We present high-speed particle image velocimetry (HS-PIV) measurements of the compressible flow in a 2d turbine cascade in order to capture the shock–boundary layer interaction (SBLI) that is known adversely affect the heat transfer of cooled turbine stator blades by introducing hot spots along with performance degrading unsteady aerodynamics. Compared to similar HS-PIV measurements in compressor cascade, the present measurements needed to address a number of challenges, such as significant vibration of model and light sheet and unsteady laser flare of the blade’s surfaces. In addition, the lack of seeding in the boundary layer (BL), caused by high centrifugal acceleration loads on the tracer particles passing around the leading edge, prevented the acquisition of reliable velocity data in the BL of the blade’s suction side. Laser flare in the raw imagery was handled with a proper orthogonal decomposition (POD) based image filtering approach and was able to recover the mean flow field more reliably. However, the data exhibits higher noise levels in the variances which is presumed to be caused by the distortion of particle images during the reconstruction from a limited number of nodes. Estimates of the in-flow turbulence level in the absence of the cascade were obtained by processing 80 kHz imagery with different image strides.