CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   calculation process problem with CFD (http://www.cfd-online.com/Forums/main/9235-calculation-process-problem-cfd.html)

Hellen May 27, 2005 00:30

calculation process problem with CFD
 
I am using FLUENT to calculation airflow with standard two equation model. The plot of the residual has reduced to 10e-4. However, suddely, the residual of the variables has gone up to 10e-1. I cannot quite understand this. Does that mean the results are not converged? why is it happen?

Thanks

zxaar May 27, 2005 01:16

Re: calculation process problem with CFD
 
in fluent with two equation models it could some times happen because of the coupling of the two variables (k and e or k and omega), to check this run the calculationsd with out turbulence model and if you get converged solution, then for the two turbulence quantites, go to multigrid control and change the pre sweeps to 1 (default is 0) and make post sweeps to 2 (default is 1) and if u r getting warning of turbulent viscosity limited etc when u start the calculations, then restrict the viscosity ratio to some low value say 100, till you start to get convergence with the two turbulence quantitities and then slowly remove the restriction from viscosity ratio.

this should work fine.

Hellen May 27, 2005 04:06

Re: calculation process problem with CFD
 
Thanks for the detailed information


All times are GMT -4. The time now is 18:54.