Measurements of entropy-layer instabilities over cone-ogive-cylinders at Mach 6

Predicting the onset of boundary layer transition is critical in hypersonic flight. To improve transition prediction methods, it is necessary to understand the underlying instability mechanisms that cause transition. Entropy-layer instabilities are of particular interest in the design of blunt reentry vehicles and other blunt supersonic and hypersonic vehicles. Entropy-layer instabilities from outside the boundary layer may enter the boundary layer and have a significant effect on transition. There is little experimental data for entropy-layer instabilities. Experimental measurements of what appear to be entropy-layer instabilities have been made in the Boeing/AFOSR Mach-6 Quiet Tunnel (BAM6QT) using surface pressure transducers and hot-wire anemometry. A long cone-ogive-cylinder model with interchangeable cone-ogive noses was used to generate the shock curvature that resulted in an entropy layer conducive to instability growth. The nosetip angles of the cone-ogive range from 25 to 40 degrees, with a majority of the measurements taken with the sharp 30 to 35-degree nosetips. Surface measurements of the entropy-layer instabilities using the 30 to 35-degree configurations show disturbances between 15 and 50 kHz. As the nosetip angle increases, the frequency of the instability decreases slightly. Results also show that the instability magnitude as measured on the model surface increases with downstream distance, then decreases, before starting to increase again. The decrease is likely due to stabilization that occurs during the entropy-layer swallowing process. Off-surface measurements using hot wires have also been made for each of the cone-ogive-cylinder configurations. These measurements show the location, frequency, and relative magnitude of the entropy-layer instability. As the instability progresses downstream, it grows inside the entropy layer, then at a certain distance downstream, the instability approaches the model surface and enters the boundary layer. Results show a smooth variation of the location of this instability descent with nosetip angle. As the angle increases, the instability approaches the model further upstream. Cross-correlations between the surface transducer and hot-wire anemometry measurements confirm that the same instability is being measured at both locations. Cross-correlations between axially-displaced surface sensors were used to calculate an instability convection velocity that is approximately equal to the numerically-calculated flow velocity. And cross-correlations between azimuthally-displaced sensors show that the instability is primarily axisymmetric. The model angle of attack for all measurements was nominally zero. However, the actual angle of attack may vary by up to 0.1 degrees. The experimental results were also compared with mean-flow computations for several of the model configurations