In spaceflight, a launch vehicle is a rocket used to carry a payload from the Earth's surface into outer space. By traveling supersonically, launch vehicles are exposed to harsh environmental conditions during the various phases of operation. However, the most extreme operational conditions are encountered within the rocket engines themselves. By their process, combustion temperatures can reach an excess of 3000 K followed by a 'mere’ 1100 K at the nozzle surface. Considering that these temperatures tremendously exceed the maximum operating temperatures of typical launcher materials, active cooling is required to enable continuous operation. Active cooling is often achieved by cooling channels, which are formed by constant cross-sectional, hollow tubes welded onto the inner surface of the nozzle. Even though the application of these cooling channels are quite common, detailed investigations on their more fundamental design properties are absent for flow conditions similar to those encountered in rocket nozzles. As such, it is decided eliminate the custom design aspects and reduce the complexity of the nozzle roughness elements to periodically placed, ribbed roughness elements.

The aim of the present study is to obtain quantitative insights on the influence of large, ribbed wall roughness elements on the mean flow, heat transfer and turbulence properties of a turbulent, supersonic boundary layer (M = 2.0). A total of fifteen test geometries, including one smooth and fourteen rough surfaces, consisting of various relative roughness heights, e/δ and pitches, p/e are tested using Schlieren, particle image velocimetry (PIV) and quantitative infrared thermography (QIRT). Heat transfer measurements were obtained by the heated-thin-foil method, providing a constant heat flux boundary condition. The QIRT setup was self-designed and constructed to yield an accurate mapping of the surface temperatures. It was observed that geometries at an e/δ of 0.2 and p/e of 10 resulted in optimal turbulence levels, whereas those with 0.2 and p/e of 25 in idealized heat transfer.