Questions regarding the use of Code_Saturne: turbulence models

[Stéphane Blanchet]

Hello,
Sorry to bother you again, I have a few more questions about the turbulence models, more particularly about the Rij models and the k-epsilon linear production model.
- Is there any case where one prefer to use one of them? You told me that the k-epsilon linear production was used for jet-impacting flows, but what about the Rij models? And is there any reason if the LRR is rarely used?
- What is the difference between the two k-epsilon models?
Thank you,
Stéphane

[Alexandre Douce]

Indeed, "k-eps linear production" is suitable for jet-impacting flows (in fact the flow near a stagnation point), unlike the standard k-eps. But of course RSM can handle jet-impacting flows, but it is more costly in CPU time.
More generally, RSM respond correctly to streamline curvature like rotating flows, flow near a stagnation point, flows affected of buoyancy, computational domain with strong curvature of the boundary wall...
RSM LRR is a legacy model (1975), kept for backward compatibility, prefer the more recent RSM SSG (1991).

[Stéphane Blanchet]

Thank you for your answers! You used RSM instead of Rij, does it mean Rij-LRR (or SSG) is the same model as the RSM LRR (or SSG) model used in Fluent. I tought the RSM LRR model used in Fluent was a low-Re model, whereas the Rij-eps LRR is a high-Re model in Code_Saturne.

[Alexandre Douce]

For me, RSM (Reynolds Stress Model) stands for Rij-eps. For Fluent, I don't know, but in Code_Saturne, Rij-eps LRR and Rij-eps SSG are both high Reynolds model. The low Reynolds turbulence model of Code_Saturne is the v2f model.