CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Error in subroutine FNDVAR (http://www.cfd-online.com/Forums/cfx/66903-error-subroutine-fndvar.html)

magnec July 28, 2009 08:05

Error in subroutine FNDVAR
 
Hi there, I saw somebody also met similar error but there was no answer

CFX 11.0, the error occurs at the first step when applying Reynolds Stress model (BSL) in a cavitation problem, regardless of choosing a initial results file or not:

Error in subroutine FNDVAR :
Error finding variable VISCEFF
GETVAR originally called by subroutine ASS_STRESS

ERROR #001100279 has occurred in subroutine ErrAction.
Message:
Stopped in routine GV_ERROR


The non-cavitating computation has no problem when there is only one material of water. I also tried to add the vapor material in the non-cavitating computation but this error still occured.

Thanks for the help~

ghorrocks July 28, 2009 08:51

Reynolds Stress model AND cavitation? You really like tough ones, don't you! Have you tried using a simpler turbulence model? I bet it works if you use SST or any of the 2 equation models. I don't think you will find many people have used RSM with cavitation so you are probably in untested territory...

I suspect this is a bug in CFX - you are using an old version, maybe it is fixed in V12. If it still occurs in V12 I would report it as a bug.

Glenn Horrocks

magnec July 28, 2009 21:48

Hi, Glenn

Yes, i have already tried some simpler model like k-e, k-Omega and SST and results were got. but these models do not have a good performance on unsteady behavier, quite different from the experiment.

Anyway, i will try in CFX 12.0. Thank you!

ghorrocks July 28, 2009 22:46

Hi,

The two-equation turbulence models and RSM all use Reynolds averaging and therefore handle the transient aspect of turbulence the same way. The only reason to use RSM models is because your turbulence is anisotropic.

If you need better transient accuracy than the 2-equation models you should go to SAS, DES or LES.

Glenn Horrocks


All times are GMT -4. The time now is 16:23.