Turbulence Modeling Validation for Afterbody Flows

The aerothermodynamics phenomena over space launch vehicles or missiles are a challenging problem on space and aeronautical applications. These physical phenomena can strongly affect the engine's aerodynamic performances. The physical problem met on this geometry is essentially the result of the interaction of two merging flows; one issued from a propulsive jet at high speed and high temperature (flow 2), and the other (flow 1) caused by the ambient low speed stream (see Fig. 1). Several complicating issues are presented in this flow configuration : • The afterbody flow-field is dominated by the behavior of the turbulent boundary layer. Thus the quality of computational results depends strongly on the accuracy of the turbulence model used for CFD. • The afterbody configurations are often quite complex, since they include singleor double flux nozzles with a variable geometry, twin-engine with tails and non axisymmetric nozzles with base flows. All of these elements are important, and should be modeled. • On the other hand, a rather fair prediction of the flowfield can be accompanied by large errors in the calculation of wall properties, affecting mainly transfer coefficients, skin friction and heat flux. The complexity and the usual misunderstanding of these physical phenomena coupled with the turbulence modeling problem of complex compressible flows motivates the present study. The outline of this paper is as follow: the turbulence models used in this study are given in section 2, the main lines of the numerical method are presented in section 3, the nozzle startup process is examined in section 4, finally sections 5 and 6 are devoted to the comparisons between numerical and experimental measurements. It is found that an interesting estimation of the pressure coefficient on the afterbody is obtained via the multi-scale model