Particle image velocimetry study of shock-induced turbulent boundary layer separation
An experimental study was conducted to investigate the characteristics of a Mach 2 shock wave / boundary layer interaction, by using particle image velocimetry (PIV). The objective was to investigate how the global flow structure is related to the shock-foot dynamics. A major component of this work was the development of a new multi-camera, multi-laser PIV system, which enables the acquisition of wide-field and time-sequenced velocity fields. The wide-field images are obtained by placing four cameras side-by-side giving an effective resolution of 4k×1k pixels. Four-image time sequences can be acquired where the time between frames is 30 to 200 µs. The PIV system was used to characterize the upstream Mach 2 boundary layer. The measured mean and RMS velocity profiles agreed well with previous measurements in compressible boundary layers and this provided important validation of the PIV system. The wide-field PIV system was used to image the entire interaction, spanning the upstream boundary layer, intermittent region, separated flow and the reattachment region on the ramp face. The separation shock wave location inferred from the PIV images agreed well with the shock-foot position inferred from the pressure data. The instantaneous vector fields reveal that boundary layer separation is not immediately induced by the shock foot, but sometimes develops substantially farther downstream. Significant reverse-flow velocities are seen in the instantaneous images, but on average no reverse-flow was observed. The global structure of the interaction was found to depend strongly on the location of the separation shock foot. Ensemble averages, conditioned upon the shock-foot position, showed that when the shock is upstream, the scale of the separated flow, the velocity fluctuations, and the domain of perturbed flow, are all substantially larger than when the shock-foot is downstream. Perhaps most importantly, the conditional upstream boundary layer profiles, conditioned on the shock position, showed that the boundary layer is thicker when the shock is upstream and vice versa. Furthermore, the conditional measurements confirmed the results of a previous study that reported a correlation between velocity fluctuations in the upstream boundary layer and shock foot motion. A preliminary study was used to test the hypothesis that acceleration fluctuations in the upstream boundary layer correlate with shock foot motion. These results showed no meaningful relationship between upstream acceleration with the shock motion, but given certain limitations of the experiment this conclusion cannot be considered definitive.