Running bubbleFoam for turbulent case...
 

Hello,

I tried to run bubblefoam for bubble column (the same case with is given in tutorial) simulation for turbulent flow but it is running for some time steps and then diverging. Can anybody help me?????

There is probably something wrong with your case.

Sorry, but I can't be more specific than you are. You said you did the tutorial case, but I checked, and that case is laminar, not turbulent.

Hey thanks Bernhard

yeah u r right the given case is for laminar. But i changed the model from laminar to kEpsilon in RASproperties in constant (with turbulence on).
Then i added the following schemes for epsilon and k in the fvSchemes:

laplacian(DepsilonEff,epsilon) Gauss linear corrected;
laplacian(DkEff,k) Gauss linear corrected;

after this when i run the simulation again, then it runs for certain timesteps and stops.

I have changed the number of grids from 3000 to 150000. That time also the same thing happens.

Can you suggest something, if possible????

Hi and welcome to the forum,

I also looked into this. The default bubbleFoam tutorial case seems to work fine with kEpsilon turbulence model enabled.

Again, as Berhard already said, be more specific. Which version of OpenFOAM is installed on your machine? What timestep did you use? When does the solution diverge? What changes have you done to the grid? It worked without any modifications for me. We can only assume you somehow changed the discretizations in the blockMeshDict.

By the way, what are grids? At least try to stick to proper english, please.

cutter

Dear Cutter
Currently I am using OpenFOAM 1.6.

For the details of simulation I have attached the files, which are modified accordingly for the case of turbulent.

Please find the attachment and suggest, if any modifications required.

Thankig you.

Ok, that's a really important information. I had no time so far to check your case and OF version. But this what I did to get it running with OpenFOAM 2.1.x (git release):

- use RASModel kEpsilon and enable turbulence in RASProperties
- add laplacian schemes for k and epsilon in fvSchemes (there may be better choices):
- add subdictionaries for kFinal and epsilonFinal in fvSolution:

Thank you very much for your reply.

I ll try this case with OpenFOAM 2.1

Hey dear cutter thank you very much for your suggestion. I think that problem was with OpenFOAM version. I am also getting the results for default case in Version 2.2.

Thanks a lot :)