Experimental investigation of induced supersonic boundary layer transition

  1. Bottini, Henny
Supervised by:
  1. Guillermo Paniagua Director
  2. Salvador de las Heras Director

Defence university: Universitat Politècnica de Catalunya (UPC)

Fecha de defensa: 28 March 2014

Committee:
  1. Tolga Yasa Chair
  2. Roberto Castilla López Secretary
  3. Juan Pedro Solano Fernández Committee member

Type: Thesis

Teseo: 117029 DIALNET lock_openTDX editor

Abstract

Turbulence onset within an initially laminar flow is one of the most common phenomenon in Fluid Mechanics, yet is an open field of research. This is due to the many and diverse causes that can trigger turbulence, which often add to each other, change their effect upon the flow velocity, and are difficult to single out in real-world situations. This is why laminar-toturbulent transition experiments have been and still are a fundamental tool for the researcher. This thesis work has addressed the study of turbulence onset in supersonic flows from an experimental point of view. Two test campaigns have been carried out each in a dedicated supersonic wind tunnel. The first aimed at tracking turbulence onset triggered by a single tetrahedral roughness element. It has been carried out in a multiple-Mach-number supersonic wind tunnel whose qualification campaign has been completed within this thesis work. Two roughness heights were tested, both for two Mach numbers, 1.6 and 2.3. The second test campaign has investigated the effects of an oblique shock wave impinging onto a Mach-2 transitional boundary layer. The transitional state of a boundary layer is that state during which the boundary layer passes from fully laminar to fully turbulent. This passage can be either induced or natural, and it was represented in this thesis work by the trigger wake and by the boundary layer flow downstream of the release point of an electric spark, respectively. Surface sensors, as thin-films and piezoelectric pressure transducers, were used to measure steady and unsteady highfrequency flow evolutions. Different wall temperatures were set for the thin-films insert as to allow the experimental estimation of the recovery temperature. Convective heat-flux trends have been extracted from the steady measurements, which, together with the recovery temperature trends, allowed the calculation of the Stanton number trends. All these quantities let to conclude on the general state of the boundary layer investigated in the first test campaign. The post-processing of the unsteady measurements yielded temperature and pressure fluctuations spectra and RMS streamwise evolutions, along with spectral time evolution at a given position. For the first test campaign, they allowed the characterization of the unsteadiness produced by the roughness within the supersonic boundary layer at different downstream locations. In so doing, they helped conclude on the state of the boundary layer, thus on the effectiveness of the roughness in triggering transition to turbulence. For the second test campaign, they allowed to single out the unsteady effects of the shock impinging downstream of the single roughness and downstream of the electric spark release point. In this way, differences in the shock effect between the roughness configuration and the clean-plate configuration have been highlighted, and the effects of different spark release frequencies compared.