Elliptic Blending RSM

[StandardRANSUser001]

Hi Everyone,

I have some questions about the elliptic blending Reynolds-Stress model that is implemented in Code Saturne, and how it should be used.

My case runs with the incompressible solver in the GUI, and this is with SGDH and GGDH heat flux models. I have implemented user definitions of Rij = X, epsilon = Y, and alpha = Z, where X, Y, and Z are relevant conditions at the inlet. Unfortunately, I have been unable to get the EB-GGDH versions of the heat flux model working.

I have tried to run this case again as a compressible case, and I am unsure of whether the EBRSM is supported in the compressible module. The case is at Mach 0.3, and I have used the supersonic outlet condition (the outlet should be a Neumann condition).

I have been unable to get this running, with any heat-flux modules. For example, when working with the EB-GGDH heat flux model, I get an error with an uninitialised boundary condition: beta, and total_energy_alpha. Are there additional boundary conditions that I should implement when studying a compressible flow with elliptic blending? I am unsure what these two quantities represent.

Could you please advise whether EBRSM is compatible with the compressible module. Furthermore, there are several EBRSM-specific heat flux models in the drop-down menu, and I am wondering which models (if any) are compatible with the compressible solver.

Any information on this would be greatly appreciated.

Best regards,
Sean Hanrahan

[Yvan Fournier]

Hello,

I do not think the EBRSM module has ever been tested with the compressible module, though I do not know if tgere is any incompatibility. In any case, if using this midel in a compressible context requires specific/modified source terms this is probably missing.

Also, the compressible model uses a specific thermal variable, so using it with another thermal model probably does not work.

There is a new compressible algorithm in the 8.2 (and maybe 8.1) branch, but it is not available from the GUI as far as I know.
It might be compatible with more turbulence models than the "main" compressible module.

Best regards,

Yvan