[electron]

hi,

as i said, SHM stops working at this point. it tells me, there are a number of edges, and the same number of edges to retest, but it does not go on.

i don't know, what the problem is. i created a backgroundmesh, using gambit, converted it to foam with the fluentMeshToFoam command, and tried to run snappyHexMesh.

my first thought was, that there may be an intersection of the trisurfacemesh and the background, but i checked that.

i testet the geometry with a backgroundmesh, that was only a cube, not the complicatet geometry, i wanted to use, and it worked quite fine. this tells me, the problem must lie within the backgroundmesh.

can anybody tell me, what might cause that error in snappyHexMesh?

thanks

[wyldckat]

Greetings electron and welcome to the forum!

First check the base mesh converted from Gambit:

checkMesh -constant

You might even want to make a full check:

checkMesh -constant -allGeometry -allTopology

Any flaws in the converted mesh can prove to be the reason why snappyHexMesh quits early!

Best regards,
Bruno

[electron]

Thank you for the welcome and the answer!

i followed your advice and checked the mesh. Most of it seems to be ok, but it says, there are negative volumes. and then the checkMesh operation quits with a "floating point exception" error.

is there a tool in openFoam, which can repair such mesh?

best regards

[wyldckat]

Hi electron,

AFAIK, there is no tool in OpenFOAM that will fix meshes. The closest one I suspect that could help is modifyMesh, but I've never tried it.

With OpenFOAM, the standard method for these bad meshes is simple: do the mesh again with better quality parameters!

For more about snappyHexMesh: http://openfoamwiki.net/index.php/SnappyHexMesh

Best regards,
Bruno