Multi Relaxation


I have the case of Couette flow in a parallel gap. For two parallel walls it works fine now, but the problem starts, when I implement a dimple with rectangular edges in the gap. After some time the calculation becomes instable. Now I red somewhere that the single relaxation model have problems with this case. Since the results are accurate for the case of the parallel walls I gues I have no mistake in the bounceback conditions.

Here is the question: Does someone of you have experience with that problem and what do you think about the multi relaxation model. Could it be helpful in my situtation?

Thanks for your hints.


you should check out this paper:
It deals with multiple-relaxation time models and discusses stability.



thanks for the hint.