I often get a problem with 'turbulent viscosity limited to viscosity ratio of 1.0 E+5' when using ke standard or ke RNG turbulence models. Have found ways of reducing it but don't really understand the fundamental reason why this occurs and what to do to eliminate it. Any one else know? Thanks.

It's a serious problem. Especially, whenever I used second order and higher accuracy schemes, I usually had this problem. In that case, at first, I reduced the under-relaxation factor until 0.1 (in some special or extreme cases). Secondly, used first order scheme for turbulence viscosity term.

If this change is not working, please remesh your domain. There is no way.

Fundamentals or background? I have no idea.

Good luck.

Kim.

HI Dave

The turbulent viscosity ratio is telling you that there two adjancent cells with a high differences in velocity. Therefore, you should plot a contour of the turbulent viscosity ratio to identify the cells with the highest values. Then, adapt the mesh for these cells.

Best regards

Alex

Hi, it depends of the problem you are treated. plot the contours of k or epsilon, normally it corresponds to erroneous value of k and epsilon. to solve this problem, i declare the zone where there is such value as laminar zone at the beginning and uncheck it when the calculation is near convergence. It is especially efficient for euler-euler calculation with un a free surface (air-water for instance).

Bye

Thanks everyone for responses.

[annu]

hi friends
i had similar problem regarding turbulent viscosity ratio, i had remeshed(refine the grid double the previous one) the grid, but viscosity ratio remains there.
please tell me how to sort out this issue.
your response will be highly appreciated.