INFLUENCE OF COMPLEX GEOMETRY ON HYPERSONIC BOUNDARY-LAYER TRANSITION

<p dir="ltr">Understanding the behavior of high-speed boundary layers is a highly sensitive process and a critical part of vehicle design. When a vehicle is designed, there is an ideal specification set by the engineers, whereas the actual vehicle has the as-built geometry which can differ in subtle, but measurable ways. This can include gaps between control surfaces and the wing, steps between skin panels, or geometric error from the manufacturing process. Part of the current work studies the effect of geometric errors from the manufacturing process, or as-built geometry, and focuses on a 7° half-angle cone that uses 3D printed trips to transition the boundary layer from laminar to turbulent. Previous work for this geometry includes experimental data collected from the Boeing/AFOSR Mach 6 Quiet Tunnel and computational work investigating idealized trips. High fidelity computations, using the NASA code OVERFLOW, are carried out on a clean configuration, as-designed configuration, and as-built configuration to investigate and quantify the differences between them and their effect on high-speed boundary-layer transition. The low-dissipation scheme used in the present work captures transition on a clean configuration cone through second-mode instability. The differences between the trips show subtle, but noticeable differences between the form of the vortical pairs and indicate a more complicated transition mode.</p><p dir="ltr">Examining novel designs for high-speed inlets and predicting their behavior can provide insight into the current understanding of the fundamental mechanisms. Any unexpected phenomena that affect the boundary-layer can greatly impact the performance of the inlet. One novel design, the INlet, a joint effort between Indiana (IN) universities of Purdue University and the University of Notre Dame, is studied to illuminate a suspected boundary-layer instability presenting on the forebody. Previous experiments led to the hypothesis that this is due to a separation bubble resulting from a separation region immediately following the leading edge when tested at negative angles of attack. However, results generated from the CREATE-AV code KCFD show this instability at positive angles of attack as well. Supporting evidence from volume and surface data are used to identify the cause to be related to Görtler-like vortices and crossflow instabilities generated from the curvature of the forebody.</p><p dir="ltr">Overall, the investigations into these two projects show the utility of computational analysis and its ability to capture physics in a way experiments cannot. The cone results highlight the need to be diligent when setting up experiments and computations especially when studying boundary-layer transition. The non-linear and sensitive nature of the flow instabilities allows small errors to accumulate and compound. Studying novel and complex designs that have a foundation in analytical methods helps highlight and improve the current understanding of the physics and assumptions involved.</p>