An impinging Shock Wave-Turbulent Boundary Layer Interaction (SWTBLI) at Mach 2 was investigated experimentally while implementing two-dimensional Shock Control Bumps (SCBs). The aim was to investigate the changes in the unsteady dynamics while changing the bump ramp angle, tail angle, spanwise shape, and the impinging shock location. Schlieren and oil flow visualisations were used to identify these changes. Unsteadiness was quantified through a spectral analysis based on Welch’s method, and a Canny-based edge-detection algorithm was developed to track the impingement and separation location in the Schlieren images.

An uncontrolled SWTBLI could successfully be generated, and the implementation of the baseline bump showed a replacement of the unsteady separation shock by a steady compression ramp shock originating from the leading edge. Spectral analysis confirmed this behaviour since the characteristic low-frequencies of the uncontrolled separation shock seemed to be removed for this compression ramp shock. An increase in the bump ramp angle showed the progressive generation of a separation shock upstream of the bump with spectral content trending towards the low frequencies of the uncontrolled interaction. On the contrary, no alterations were observed for an increase in tail angle. An upstream impingement produced similar behaviour as the increase in ramp angle; a separation shock was generated upstream of the bump with increased low-frequency spectral content. The downstream impingement, however, did not show any alteration in the same region. Major factors influencing the unsteady dynamics are identified as the impingement location and the ramp shape of the bump.

The developed edge-detection algorithm proved unsuccessful in quantifying the unsteadiness although it could detect the impinging and reflection shock of the interactions. Spatial standard deviation distributions of the interaction revealed increased deviation values in the impinging shock suggesting that the impinging shock fluctuates. Rather, it is suspected to be a form of noise intrinsic to the experimental technique. Therefore, this affects the detection of the edges and the calculated impingement and separation location. Future research is suggested to improve the noise mitigation method in the algorithm. Additionally, the benefits associated with the 2D-SCB would be most noticeable in an integrated approach with an industrial application. Finally, numerical simulations and/or different experimental techniques are suggested for future research to obtain a better quantification of the interaction and unsteady dynamics.

Supersonic flows cause thin panels to flutter, which is characterized by high-amplitude self-sustained oscillations, increasing the risk of fatigue failure. Flutter is known to be exacerbated when a shock wave impinges on the panel, creating a shock wave/boundary-layer interaction (SWBLI) which promotes separation of flow and leads to increased aerodynamic and thermal loading on the panel. This novel fluid-structure interaction (FSI), known as shock-induced panel flutter, poses a risk to the structural integrity of components used in high-speed aerial vehicles, such as rocket nozzles and supersonic engine inlets.

Due to the complex coupled interaction between flow and structural dynamics in flutter, experiments provide a better option to investigate the underlying physical mechanisms, rather than computational studies, which usually have to compromise between computational cost and resultant accuracy. The ST-15 supersonic wind tunnel facility at TU Delft is employed to conduct experimental measurements of shock-induced panel flutter, using high-speed Schlieren imaging to capture flow structures and stereographic Digital Image Correlation (DIC) to record panel displacements in separate campaigns. Tests are done at Mach 2 with fully-clamped thin panels, and the flutter response is excited using different shock strengths and impingement locations.

In both campaigns, accelerometers are used to measure spurious vibrations around the wind tunnel test section. This helps reveal the existence of external vibrations inherent to the facility, which are also found to drive the frequency of the panel flutter: at 760-770 Hz without an impinging shock, and 605-640 Hz with an impinging shock. The latter frequency is detected in both - shock motion and panel displacements – which establishes coupling between flow and structure despite measurements being non-simultaneous. Changing the shock impingement location does not have a significant effect on the degree of flow separation caused over the thin fluttering panel, which is always higher than the separation on a rigid plate at the same shock strength, thus proving that fluttering panels are not viable means of shock-induced separation control. A stronger impinging shock produces increased shock-induced flow separation but results in less energetic panel flutter, which is attributed to the higher post-shock pressure rise suppressing the panel. Flutter is found to be most energetic, and consequently, the panel is most susceptible to fatigue failure, when the shock impinges at 60% of the panel length.