CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   convergence, fluent (http://www.cfd-online.com/Forums/fluent/46193-convergence-fluent.html)

rsi October 1, 2007 07:46

convergence, fluent
 
What should we do if 2 out of 10 CFD models do not converge? We need to compare CFD results amongst all 10 cases but 2 of them not converged. If we change under-relaxation factors for 2 of these, will our comparison be valid? What do people normally do when encounter this problem?

bart weisser October 1, 2007 14:21

Re: convergence, fluent
 
Can you give us details of your solution method?

*Transient, steady?

*Compressible, incompressible?

*Turbulent, laminar?

*Single component, multiple component (misible or not)?

*What is not converging (mass, momentum, or mass fraction)?

*Are residualrs normalized or not?

etc, etc ...

These would help determine where the problem likely lies.

BW.


Pirooz Darabi October 1, 2007 17:39

Re: convergence, fluent
 
Yes man,

Try be reducing the relaxation factors! Do not reduce them significantly though. And if did not work then you may have to refine your mesh or change your solution method.

Cheers Pirooz

rsi October 4, 2007 20:08

Re: convergence, fluent
 
steady state, incompressible, turbulent, single component, mass is not converging but when changed relaxn factor for pressure, the v-component is not converging.


Bart Weisser October 5, 2007 12:49

Re: convergence, fluent
 
On second thought ... it might be useful to look at your residual plots for the two quantities in question. I bet the residual is pretty low everywhere else except at some corner where it is exceedingly high. If that is the case I would suggest reviewing your boundary conditions or grid density in the offending vicinity. I hope that helps.

BW.


All times are GMT -4. The time now is 15:24.