simulation of RSM in compressible mixing layer

lars

Hi. recently, I trying to compute the compressible mixing layer with Reynolds Stress model. In my code, FVM is used.( LU-SGS, 2nd order TVD Scheme) convection term is treated implictly. and viscous term is treated explicitly. inlet condition is given, for example, U velocity is given with hyper-tangent function.

I succeeded in computing several case with incompressilbe RSM model and RSM+Sarkar's Dilatation dissipation model. At several convective mach number case, reasonable result is obtained. ( reduced growth rate )

But, I trying to compute the RSM with compressible pressure-strain model.( Int. J.of Heat and Fluid flow, 2000, H. Fujiwara et al. "A turbulence model for the pressure-strain correlation term accounting for compressibility effects") In this model, pressre-strain term is reduced with increasing turbulent mach number.

With this model, the computations is very unstable. and resutl is very bad.

Do you think wht's the problem?....

Please, advice me....