[shimakasaei]

Hi all,

I am simulating a model with olaFlow and using sanppyHexMesh, I am decomposing my model via decomposePar and simple model. I faced a problem running parallel although the simulation do not have any problem running on one node. So, I believe that the problem should be related to decomposing or something. I have checked my decomposeParDict multiple times, but I did not get what the problem is!!!
Here is mu decomposeParDict:

\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
location "system";
object decomposeParDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

numberOfSubdomains 48;
method simple;

simpleCoeffs
{
n (6 2 4);
delta 0.001;
}

Here is the error that I faced:

--> FOAM FATAL ERROR:
No base point for face 36495, 5(25399 30237 30238 227042 25400), produces a decomposition that has a minimum volume greater than tolerance.
From function void Foam:article::crossEdgeConnectedFace(const label&, Foam::label&, Foam::label&, const Foam::edge&)
in file /home/OpenFoam/OpenFOAM-v1612+/src/lagrangian/basic/lnInclude/particleI.H at line 526.

FOAM parallel run aborting


Could anyone please give me a hint?

Best,
Shima

[Phicau]

Hi Shima,

maybe your case has only 1 cell in the Y direction and you are trying to decompose it into 2 domains in that direction?
If not, have you tried scotch decomposition method?

Best,

Pablo

[shimakasaei]

Originally Posted by Phicau

Hi Shima,

maybe your case has only 1 cell in the Y direction and you are trying to decompose it into 2 domains in that direction?
If not, have you tried scotch decomposition method?

Best,

Pablo

Thank you for your reply. I had more than one cell in y direction, but my problem solved as I changed the minTetQuality in Qualitycontrol part of the snappyHexMeshDict .

Best,
Shima.