Ph.D. – Boundary Layer Stability

Surface-Roughness-Induced Transient Growth in Hypersonic Boundary Layers

Cone with sharper distributed roughness element (DRE) nosetip

The Transient Growth Cone (TGCone) model with distributed roughness element nosetip

All hypersonic vehicles from the Orion capsule to the X-51 feature rough surfaces. This surface roughness can be categorized as isolated, like fasteners, joints between control surfaces, or like the protruding gap filler between thermal protection system (TPS) tiles that prompted a never-before-attempted EVA to the underside of the space shuttle. Surface roughness can also be distributed, like the marks left by a machining process, sandpaper, or an ablative heat shield.

Roughness generates disturbances in the boundary layer–the thin region of fluid near a solid surface–through a process called receptivity. These disturbances can then either decay or grow through several instability mechanisms. If they grow sufficiently large, they can cause the boundary layer to transition from laminar to turbulent flow, which is undesirable at hypersonic speeds where turbulent boundary layers increase the temperature loads on flight vehicles.

Boundary layer transition visualized

Boundary layer transition visualized; Image credit: Y. Kohama

In particular, my work focuses on the transient growth instability mechanism. Surface roughness can create streamwise vortices that stretch behind the roughness. These vortices redistribute momentum in the boundary layer, pushing low-momentum fluid upward in some locations and pulling high-momentum fluid toward the wall in others. This creates a spanwise variation of low- and high-speed streaks in the boundary layer.

Streaks forming in oil-flow visualization of a hypersonic boundary layer

Signs of streak formation in early oil-flow visualization tests of the TGCone in the M6QT

Transient growth has been measured in low-speed boundary layers but has not been a subject of study in high-speed boundary layers until recently. Several calculations of optimal disturbances in hypersonic boundary layers have been conducted, but my work represents the first attempt to deliberately measure transient growth experimentally. I conduct my work in the Texas A&M Mach 6 Quiet Tunnel (M6QT), a low-disturbance hypersonic facility, under the auspices of the AFOSR/NASA National Center for Hypersonic Laminar-Turbulent Transition Research.

Stability of Swept-Wing Boundary Layers

Some of my early projects at Texas A&M dealt with the stability of the boundary layer on a swept wing, like those found on most commercial transports. Due to the swept angle of the leading edge, flow on these wings is three-dimensional and the pressure gradient across the wing results in a secondary flow inside the boundary layer. This velocity profile contains an inflection point, providing the source for an inviscid instability, which, in flight, typically assumes the form of stationary near-streamwise counter-rotating vortices. These vortices ultimately lead to turbulent breakdown.

White Knight Flight TestsMy work with this problem was chiefly computational in nature. I wrote my own eigenvalue solver for 3D swept wing flows, which was later utilized to estimate boundary layer stability on wind turbine blades. I also performed computations in complement to a series of Northrop Grumman flight tests utilizing a novel swept wing design carried beneath the Scaled Composites White Knight aircraft. After computing the laminar basic state flow for a 2D airfoil slice in FLUENT and matching the pressure distributions to those obtained in flight, I used the LASTRAC stability code to calculate the most unstable stationary crossflow wavelengths on the model. The results compared favorably to those obtained from hotfilm measurements during the flight tests.