Shock oscillation mechanisms of highly separated transitional shock-wave/boundary-layer interactions

In business jet and defence applications, cruise altitudes typically range from 15 to 16 km, significantly higher than the 10 to 13 km common for commercial aircraft. At these higher altitudes, the atmosphere is quieter, air density is lower, and engine sizes are smaller than those of commercial aircraft. These conditions result in a low chord Reynolds number on the transonic fan – the first bladed component at the engine inlet of the turbofan engine. Consequently, a laminar boundary layer persists on the suction side of the transonic fan during cruise. Additionally, the trend in compressor design is higher blade loading, leading to larger separations. These conditions combined induce a shock oscillation with frequencies approximately matching the blade's natural structural frequencies. This resonance causes excitation and vibrations, which, if not mitigated, can lead to fatigue and potential structural failure. This dissertation explores the dynamics of this shock oscillation mechanism in transitional shock-wave/boundary-layer interactions (SBLIs) with the work arising from the occurrence and mitigation of shock oscillation-induced unsteadiness in transonic fans operating at low Reynolds numbers. The focus of the study is to investigate a particular shock oscillation mechanism. The study uses experimental and numerical approaches, and demonstrates that a canonical research configuration can replicate the shock oscillation mechanism observed in highly loaded transonic cascades. Using Large Eddy Simulations (LES) and experimental methods such as high-speed Schlieren imaging, spark light shadowgraphy, and particle image velocimetry (PIV), the research identifies a specific mechanism driven by the interplay of natural laminar separation growth, transition through shear layer instabilities, and separation suppression through turbulence. Reynolds-averaged methods typically used in industry cannot be used to resolve the oscillation mechanism. The oscillation mechanism of interest – one which is significant for a laminar oncoming boundary layer and suppressed for a turbulent boundary layer, occurs for highly separated transitional shock-wave/boundary-layer interaction, where transition occurs after a thin upstream laminar section of the shock induced separation, upstream of the shock impingement location. Laminar SBLIs naturally exhibit a larger upstream influence, and so the influence of the SBLI grows in the upstream direction, carrying shear layer instabilities with it in this highly separated case. These instabilities generate turbulence that eventually suppresses this thin, long, laminar section of the separation bubble, causing its collapse. The turbulence advects downstream, suppressing the bulk separation and influencing the reflected shock movement. A key visual characteristic is that laminar separation shock forms during the growth phase followed by a turbulent separation shock due to shear layer thickening at transition, while a (single) turbulent separation shock is briefly visible during the collapse phase in the brief absence of the laminar separation. Suppression of this mechanism is achieved through boundary layer disturbances, with a fully turbulent boundary layer proving unnecessary for achieving a significant degree of stabilization. Large eddy simulations reveal the mechanism's sensitivity to free-stream turbulence levels, with oscillation amplitude and frequency affected. The characteristic length scale of the shock oscillation mechanism corresponds to the distance travelled by the laminar separation shock before the collapse of the upstream laminar section. Strouhal numbers based on this length scale collapse LES and experimental shock oscillation frequencies, validating the simulated oscillation mechanism dynamics. LES simulations and experiments show that boundary layer tripping suppresses the oscillation mechanism. This research laid foundations regarding this particular shock oscillation mechanism found in transitional shock-wave/boundary-layer interactions, and will have an impact on the application by providing engineers with answers with regard to the shock oscillation mechanism and methods which can be used. Future work should focus on full-span LES simulations of industrial configurations, parametric studies on Reynolds and Mach number dependencies, and high-speed PIV.

Author dc:creator

• Nel, Philipp

Advisors dc:contributor.advisor

• Schreyer, Anne-Marie
• Swoboda, Marius

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Identifier URI

https://doi.org/10.14279/depositonce-24824

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/25996