An Experimental Study of Three-Dimensional Inlet Shock-Boundary Layer Interactions.

Abstract

New experimental results are presented of a three dimensional inlet shock-boundary layer interaction (3DI-SBLI) generated by a six-degree full-span wedge installed in the 57 $times$ 70 -mm Michigan 'Glass Inlet' wind tunnel at Mach 2.75. Images from two traditional techniques, oil flow and Schlieren are compared. Oil flow lines, coincident with the local skin friction, show a complex three dimensional picture despite the two-dimensionality of the density gradient visualization. Analytical complexity regarding three dimensional flow has prevented many previous investigations from investigating these regions. However, significant discrepancy occurs when using insight gained in two-dimensional interactions on three dimensional problems.

Rather than rely on amalgamated two-dimensional explanations for the oil flow, Legendre's critical point theory is recalled to describe local skin friction topologies in the 3DI-SBLI. Extending this framework from a local interaction structure towards the far field, the `secondary flow separation concept' is proposed to couple the global flow topology to the skin friction lines. Using the concept, the influence of adding the third dimension becomes clear. Arranging points of `primary separation' in the flow, and the vortex structures produced by them, we can predict the locations of secondary separation and the global flow character downstream.

To investigate the link between skin friction and the flow field, the 3DI-SBLI is quantified using stereo particle image velocimetry (SPIV). Data have been obtained in three streamwise horizontal, three streamwise vertical, and ten spanwise image planes. These measurements are used to evaluate the secondary flow separation concept. Verification of the concept is crystalized by the measurement of streamwise vortical structures. For this case, one large vortex structure is carried along each sidewall under the swept-shock foot, while the two from in each corner region. The hypothesis predicts the interaction of two opposing vortical structures produce a secondary separation, and one appears on each sidewall and another along the tunnel centerline. The proposed theory is therefore useful as a simple physical model to describe complex 3DI-SBLIs as the sum of vortex interactions.