CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   imbalance not converging investigation (http://www.cfd-online.com/Forums/cfx/75043-imbalance-not-converging-investigation.html)

mactech001 April 15, 2010 03:00

imbalance not converging investigation
 
Dear all,

i'm modeling CHT application, and my T-energy in solid and H-energy in fluid imbalances are not converging to <1% when i've Conservation Target=0.01. RMS H-Energy is not converging to my 1e-5 criteria as well.

Is there a way in Post i can investigate further where this divergence problem is happening LOCALLY in my model please?

Hope you hear any comments/suggestions. Thanks.

ghorrocks April 15, 2010 07:25

Yes. If it is a steady state model you can chop bits out of your model (using volumes and others) and work out balances on them. That might help you locate the imbalanced area.

Two suggestions as to the root cause of the problem:
1) In CHT simulations the solid region converges typically far slower than the fluid region. A sensible time step in the fluid region is a tiny time step for the solid region. If steady state use a solid time scale factor to accelerate convergence in the solid region. Values like 100 or 1000 are typical.
2) If your flow is transient then you are forced to just run your simulation for a very long time. In some circumstances you can be clever and assume it as a series of steady state steps. Depends on the model as to whether this will work or not.

mactech001 April 15, 2010 08:25

Quote:

Originally Posted by ghorrocks (Post 254753)
Yes. If it is a steady state model you can chop bits out of your model (using volumes and others) and work out balances on them. That might help you locate the imbalanced area.

i'm doing steady state mode. how do i chop bits out of the model in Post please?


All times are GMT -4. The time now is 05:57.