Hi Foamers,

I ve a problem to set up a case with topological changes (OpenFoam 1.6-ext). I like to simulate a spark ignited engine with a pent roof, including valves. So far I managed to adapt a mesh motion class to my requirements. In order to close and open the valves a small curtain surface around the valve is introduced by the attachdetach meshModifier. The mesh works fine. I ve tested it with a pure mesh motion solver, but with egineDyMFoam (adjusted to the engineTopChangerMesh class) it doesn’t t work.
Segmentation fault when the turbulence->correct() is called. Till this step everything works fine.

Remark: Without the attachdetach meshModifier( turbulence on) it works and with attachdetach (turbulence off) it works as well. It seems that I ve a problem with the boundary conditions of the attachdetach faces for the turbulence model.
Furthermore, with 1.5-dev it works without a problem. There, the turbulence model doesn’t t ask for boundary conditions for the attachdetach faces but in 1.6-ext it does.

Does anyone has used the attachdetach approach in 1.6-ext so far with a turbulence model?

Best regards
Peter

The problem is solved. Here is a big fix proposed by Mr Jasak. It will be corrected in the next release of 1.6-ext

It is to do with the state of turbulence on newly created patches after detach, where I have no data to map from.

Here is a quick fix for you:
- go to the turbulence model you are using (I assume compressible k-e)
- look for the beginning of the ::correct() function in kEpsilon.C
- add the following two lines at the very beginning:

k_.correctBoundaryConditions();
epsilon_.correctBoundaryConditions();

- recompile.

Great work!

Thank you very much!