CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   STAR-CCM+ (https://www.cfd-online.com/Forums/star-ccm/)
-   -   Turbulent viscosity limited (https://www.cfd-online.com/Forums/star-ccm/152642-turbulent-viscosity-limited.html)

MissCFD May 5, 2015 14:48

Turbulent viscosity limited
 
Hello,

When I start my simulation, I have already turbulent viscosity limited at the first iteration what does that mean ? My simulation is initialized with constant variables and it seems to be a problem ... I have to initialize with an other simulation in order to my simulation works. Can somebody please explain it to me ?

Thank you

fluid23 May 5, 2015 16:58

Check your boundary conditions. 9 times out of 10 that is the problem on something like this.

It could also be your initialization. How did you initialize your solution?

fluid23 May 5, 2015 17:49

Sorry, I guess I didn't read carefully. It definitely sounds like you need to initialize differently. Try using Expert Initialization (Grid Sequencing) under Solvers. Basically this simplifies the governing equations and solves your flow over a series of ever finer meshes (using the previous course mesh results as your initial conditions). Finally interpolating the last of the grid sequencing results onto your actual mesh.

This can be an issue with very complex flows, rotating flows, etc... However, it should solve your problem and also cut down on your run time. I have saved litterally days on some rather beefy analysis using this option.

MissCFD May 5, 2015 20:25

Thank you very much for your response but I have already tested ''Expert Initialization'' and it doesn't work either, it give me outliers with very high-values' for velocity ...

Do you think that if I reduce initial conditions for turbulence intensity and turbulente length scale, it can work ? Because I took the same values that I put for my boundaries ...

Quote:

Originally Posted by MBdonCFD (Post 545131)
Sorry, I guess I didn't read carefully. It definitely sounds like you need to initialize differently. Try using Expert Initialization (Grid Sequencing) under Solvers. Basically this simplifies the governing equations and solves your flow over a series of ever finer meshes (using the previous course mesh results as your initial conditions). Finally interpolating the last of the grid sequencing results onto your actual mesh.

This can be an issue with very complex flows, rotating flows, etc... However, it should solve your problem and also cut down on your run time. I have saved litterally days on some rather beefy analysis using this option.



All times are GMT -4. The time now is 12:14.