File(s) under embargo
until file(s) become available
EFFECT OF ANGLE OF ATTACK ON INSTABILITY AND TRANSITION ON A FINITE-SPAN COMPRESSION RAMP IN QUIET HYPERSONIC FLOW
This research focuses on experiments on compression-induced shock wave/boundary-layer interactions conducted in the Boeing/AFOSR Mach-6 Quiet Tunnel (BAM6QT) at Purdue University. The BAM6QT facilitates a low-freestream-noise hypersonic test environment more similar to that experienced in flight than a conventional wind tunnel. Measurements were captured on two sliced 7° half-angle cones with finite-span compression ramps. On the first, the slice was cut parallel to the axis of the cone to build upon previous measurements in hypersonic flow. While similar geometries have been analyzed for over 30 years in experiment and computation, there are significant gaps in understanding of the underlying mechanisms leading to instability and transition on the ramp. Further, in low-noise Mach 6 flow, the boundary layer separated at the leading edge of the slice, which is unlikely to occur on a real flight vehicle. Thus, on the second model, the slice was cut at a 4° incline to the
cone axis to facilitate the growth of an attached laminar boundary layer on the slice. Using this configuration, the ramp-induced boundary-layer thickening initiated between the slice leading edge and the ramp leading edge, allowing the investigation of a ‘naturally’ formed separated region.
Data were captured at angles of attack ranging from 0° to 6°, on compression ramp angles ranging from 10° to 20°, and for freestream Reynolds numbers of 2.5×10^6/m to 12×10^6/m. To analyze the mean-flow behavior of the separation bubble as it changes with the above parametrics, time-averaged schlieren visualization was used to provide off-surface visualization of the flowfield, allowing estimates of reattachment position and separation bubble size. In all cases, reattachment position was shown to move upstream with an increase in angle of attack, an increase in ramp angle, and an increase in Reynolds number. However, on the model with the inclined slice, the Reynolds number impacted reattachment location to a much lesser extent.
Heat transfer measurements on the ramp revealed regions with the most significant aerothermal loading. Streamwise streaks of high heating originating at the ramp edges and centerline were observed to increase in magnitude with an increase in Reynolds number, angle of attack, and ramp angle. On the model with the inclined slice, many streaks of high heating were observed that increased in quantity and magnitude with angle of attack and ramp angle. Root mean squared pressure fluctuations computed from surface pressure measurements were shown to follow similar trends to centerline heat transfer results for both models. Angle of attack, ramp angle, and slice angle are shown to play a dominant role in transition. Finally, the importance of quiet tunnels is made remarkably clear, as the BAM6QT operating in its conventional-noise configuration resulted in drastically different results.
For measurement of shock wave/boundary-layer instabilities, schlieren frames were captured at 100,000 fps to allow measurement of low-to-mid-frequency fluctuations of the recirculation zone edge. Shear layer flapping frequencies were found to occur at around 1100–1200 Hz, which increased with angle of attack to up to 1600 Hz. It is likely that this is an inherent instability in the separation bubble itself, rather than a function of freestream disturbances, and may be indicative of an ‘expansion and relaxation’ effect known as bubble breathing. Additional measurements using low-frequency-capable pressure sensors must be captured to determine whether this breathing effect manifests on the model slice or ramp.
- Master of Science in Aeronautics and Astronautics
- Aeronautics and Astronautics
- West Lafayette