CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Foam::error::printStack(Foam::Ostream&) (http://www.cfd-online.com/Forums/openfoam-solving/110462-foam-error-printstack-foam-ostream.html)

wlamont December 11, 2012 15:35

Foam::error::printStack(Foam::Ostream&)
 
2 Attachment(s)
I am trying to run OF but I am getting an error after iteration 49. I created a mesh in Ansys mesher and exported to fluent (ascii mesh). I then used fluent3DMeshToFoam to create an OF mesh. I setup the case with appropriate BCs and calculated appropriate k and epsilon. After iteration 49 I get the following error:

Code:

#0  Foam::error::printStack(Foam::Ostream&) in "/opt/openfoam211/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
I have attached the full log file and the checkMesh report. The checkMesh report seems good to me. I understand the error is a floating point error. Likely a divide by zero but I can't seem to get rid of it. Any ideas or thoughts would be very helpful.

Thanks,

wlamont December 11, 2012 15:39

removed
 
removed.... thanks

immortality December 12, 2012 07:40

I have the same problem.how did you fix it?

wlamont December 14, 2012 10:18

I still haven't figured out what is wrong. I am still troubleshooting. Any thoughts?

ngj December 14, 2012 10:29

Hi Warren,

Well, in the very first time step you have 1001 iterations on the pressure, which is typically the maximum value of iterations unless otherwise stated by the user.

This results in a very large local mass conservation error (I never use steady state solvers, so have no idea if that is typically allowed) and secondly the values for epsilon, still in t = 41, is by far too large.

I conclusion: I would suspect that your boundary conditions are not sane or you are using an initial internal field, which is incorrect. Another advice typically given on this forum is to use "upwind" for all numerical schemes. This should stabilise the simulation if the divergence is related to the choice of scheme.

Kind regards,

Niels


All times are GMT -4. The time now is 23:56.