Investigations on Impinging Shock Wave and Boundary Layer Interactions in Hypersonic Flow

Abstract

Shock boundary layer interactions (SBLIs) rank among the most fundamental problems in high-speed aerodynamics and are a primary concern in hypersonic flows. Shocks at hypersonic speeds are typically strong enough to separate the boundary layer, which can lead to dramatic changes in the entire flowfield structure with the formation of strong vortices and complex shock patterns characteristic of the SBLI phenomenon. Their presence in the systems in which they occur results in large-scale unsteadiness and high aerodynamic heating loads which curtail the performance of these systems and in many cases, pose insurmountable problems which limit the practical realisation of such systems. Following a wide-ranging literature survey which lays bare the dearth of experimental data and computational assessments in hypersonic SBLIs in high Reynolds number flows, an experimental campaign was undertaken on a configuration consisting of an impinging oblique shock interacting with the boundary layer on a flat plate. The effects of the impinging shock strength, the shock impingement locations on the flat plate and the flow Reynolds number were investigated for a range of configurations and flow conditions. The HST2 shock tunnel facility at LHSR (IISc) was used in the straight through and reflected modes of operation, and a Ludwieg tunnel constructed from the HST2 facility enabled the investigation of Reynolds number effects. The experiments were performed over a vast Reynolds number range between 1.13 million/m to 44.4 million/m, and it is this parameter whose effect is of primary focus in the present study. Of the 49 experimental runs performed, 4 were at a nominal Mach number of 6 (straight through mode of shock tunnel) while 45 runs were at a nominal Mach number of 8 (reflected mode in shock tunnel and all Ludwieg tunnel runs). The flow enthalpies for the present study fall in the low enthalpy range (varying from 0.3 MJ/kg for the Ludwieg tunnel runs to 1.67 MJ/kg for the reflected mode of shock tunnel), thus allowing the analyses to proceed with the assumption of negligible variations in the molecular composition and intrinsic characteristics of the flowing gas. The flow diagnostics consists of primarily Schlieren visualizations, along with augmented pressure distribution data obtained from PCB pressure sensors at a few locations on the centre line of the flat plate. The experiments revealed a nonlinear scaling of the separation bubble with the shock impingement location, and a clear influence of the flow Reynolds number on the extent of the interaction, with the separation bubble size decreasing dramatically under a range of flow conditions as the Reynolds number is increased. A correlation law was formulated to account for the effect of the flow Reynolds number on the phenomenon for the Ludwieg tunnel cases with finite separation. The correlation law proposed a factor of 𝑅𝑒𝐿19 to be included in the scaled separation term, and this law provides an accurate quantitative relationship between the separation length, point of shock impingement, flow Mach number, flow Reynolds number, and the imposed pressure ratio for the test conditions used in the present study. Following the experimental study, three dimensional steady (time-invariant) simulations of the full test section with the test model were performed for a few configurations with the aim of assessing the fidelity and accuracy of RANS simulations for the particular case of leading edge separation. The computational results have shown good agreement with the experimental data procured, and the three-dimensional nature of the interaction over the flat plate is captured. The thesis concludes with a summary of the results obtained, and with suggestions for future work.