Experimental Study of Passive Ramps for Control of Shock-Boundary Layer Interactions.

Abstract

Experimental results are presented from an investigation of the effects of sub-boundary layer streamwise vortices on incident oblique shock-boundary layer interactions (SBLI). Stereo particle image velocimetry (SPIV) is used to first quantify the mean, fluctuation, and gradient statistics in an evolving boundary layer with Reynolds number based on momentum thickness growing from 6,600 to 9,600. Oblique shock waves of varying strengths are produced by 7.75-deg, 10.0-deg, and 12.0-deg deflections of the Mach 2.75 free-stream. The resulting SBLI are quantified using the SPIV technique. Data are obtained in two-dimensional planes oriented in the streamwise direction, as well as in multiple spanwise planes located progressively throughout the interactions. A free shear layer, anchored by the impingement of the reflected shock wave, is visualized. Its highly unsteady nature is likely fundamental to the large-scale low-frequency oscillations previously observed in SBLI.

These measurements are used as a baseline for evaluating the effectiveness of ramp-like streamwise vortex generators for passive SBLI control. Sufficiently strong vortices located appropriately with respect to themselves and the boundary layer are found to essentially eliminate instantaneous boundary layer separation events in localized regions and, correspondingly, substantially reduce large scale fluctuations of the SBLI. A new "inverse" micro-ramp design is shown to produce vortices approximately twice as strong as those produced by a "standard" design. The spanwise influence of the inverse micro-ramps is also twice that of the standard design, and the vortex pair resides almost twice as deep in the boundary layer. This results in a substantial local reduction of the shape factor through the interaction region, and breaks any large recirculation zones occurring at the shock foot into much smaller localized structures. The net effect is quantified via spanwise integration of the displacement thickness. In the interactions presently considered, the standard micro-ramp design reduces the peak net displacement thickness by 4%-22% depending on the incident shock strength, while the inverse micro-ramps produce corresponding reductions of 17%-34%. Structures of the type proposed therefore demonstrate considerable promise as simple, passive, and physically robust methods to supplement or replace other active control techniques in practical applications.