Numerical Investigation of a Generic Scramjet Configuration

Abstract

A Supersonic Combustion Ramjet (scramjet) is, at least in theory, an efficient air-breathing propulsion system for sustained hypersonic flight at Mach numbers above approximately M=5. Important design issues for such hypersonic propulsion systems, are the lack of ground based facilities capable of testing a full-sized engine at cruise flight conditions and the absence of general scaling laws for the extrapolation of wind tunnel data to flight configurations. Therefore, there is a strong need for the development and validation of CFD tools to support the design process of scramjet-powered vehicles. Specific challenges for the applied CFD solvers include the accurate prediction of skin friction and heat transfer caused by turbulent boundary layers, the mixing and combustion of fuel in compressible turbulent shear layers, and the modelling of chemical nonequilibrium effects, which can be of significant importance for the prediction of engine performance. Due to the uncertainties associated with the modelling of turbulent compressible and chemically reacting flows that are still associated with current CFD solvers and the limitations of experimental methods to comprehensively characterize the flow properties, a close interaction of CFD and experimental investigations is necessary to further improve the understanding of flow phenomena iinside scramjet engines. The aims of this thesis are, in this context, to assess the applicability of, to further develop, and to validate the DLR TAU flow solver for the CFD analysis of the complete flow-path of a scramjet vehicle. The basis of this validation and of the identification of critical modelling assumptions is the recalculation of a series of wind tunnel tests of the HyShot II generic scramjet configuration that were performed in the High Enthalpy Shock Tunnel Göttingen (HEG) at the German Aerospace Center, DLR. Appropriate models for turbulent hydrogen combustion are implemented in the TAU solver. These include models for the thermodynamic and transport properties of the participating species, appropriate reaction mechanisms and an assumed-PDF model of the turbulence-chemistry iinteractions. The complete modelling strategy is validated using a representative test case consisting of detailed experimental results for a lifted flame in a co-axial burner configuration. The TAU code is then applied to determine the free stream conditions in the HEG test section, to simulate the HyShot intake flow field, and for a numerical investigation of the turbulent reacting flow in the combustor. The sensitivity of the numerical results to relevant modelling parameters such as, for example, the choice of the turbulence model, the Schmidt number assumption and the influence of turbulence-chemistry interactions were assessed to draw conclusions regarding the uncertainty margins of the CFD predictions. Finally, further numerical investigations are performed to assess the applicability of the pL-scaling law for scramjet combustors.