CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Siemens (https://www.cfd-online.com/Forums/siemens/)
-   -   Help: MARS scheme does not converge! (https://www.cfd-online.com/Forums/siemens/56229-help-mars-scheme-does-not-converge.html)

Jimmy Fox August 7, 2007 15:52

Help: MARS scheme does not converge!
 
Hi, I am using MARS differencing scheme for UVW-mom (with standard K-e, incompressible, hexa mesh) to simulate oil flows through a spool. The case seems not complicated, but it is difficult to converge. Any sugguestions about it?

Kevin August 7, 2007 22:01

Re: Help: MARS scheme does not converge!
 
Reduce the Blending Factor value to around 0.2. This will make it converge faster but still maintain a second order accuracy.


matej August 8, 2007 02:59

Re: Help: MARS scheme does not converge!
 
and then try to move the blending back to see if you came to the same solution....

Jim August 8, 2007 05:16

Re: Help: MARS scheme does not converge!
 
Does it converge with UD?

murash-v August 8, 2007 07:06

Re: Help: MARS scheme does not converge!
 
I think you should try make Blending Factor value greater than 0.6. But it make more time! And try to use double presccision for solver. Does *info file contain any WARNINGS?

Jimmy Fox August 8, 2007 11:06

Re: Help: MARS scheme does not converge!
 
Thanks for your suggestion. I am trying the blending factor of 0.2 and will get results soon.

Jimmy Fox August 8, 2007 11:08

Re: Help: MARS scheme does not converge!
 
This case converges with UD (also double precision, without any wornings) very fast.


All times are GMT -4. The time now is 11:33.