This seems like an OpenFOAM convergence/mesh quality issue and nothing really directly to do with FEATool, as I was able to solve it by decreasing the grid size (gridmax = 0.08), or using quad cells (gmsh grid generator with the 'quad' = true property).
If no other users chip in here, you might get more detailed answers for questions diretly relating to external tools (OpenFOAM, SU2, FEniCS, Gmsh, etc...) by contacting the developers and support options for these tools directly. For OpenFOAM you can export the case files manually on the command line like:
openfoam( ..., 'casedir', 'openfoam_casedir_path', 'clean', false )
or alternatively use the "Export" button in the FEATool GUI. Alternatively, paid technical support and and consulting for custom issues is also available.
Leon Robers wrote
Lately, I experienced a really strange error while using openFoam. I'm evaluating all kind of geometries but approximately one out of ten leads to a really strange error and the abortion of the simulation. This always occurs between the 25 and 45 iteration and leads to a jump in the residual values and completely wrong results. Furthermore, I realized that the number of iterations for the pressure (row beginning with "GAMG") jumps to 1000 in the iteration before the simulation crashes.
I uploaded a zip folder containing one geometry which leads to this error, the error description, the corresponding residual plot, the results and the matlab file which sets up the simulation. Furthermore, i uploaded a very similar geometry where this error doesn't occur.
Also, reducing the inflow velocity for example to 5000 mm/s instead of 10000 leads to a converging simulation in this specific case, but other geometries still do not converge
My Unit system is mm, s, MPa, tonne, N
Any help is highly appreciated! Thanks a lot
Leon
Program versions:
Matlab 2018b,
FEAtool Version 1.11.1 build Number 20.01.014 (the one you uploaded lately to fix the meshing bug)
blueCFD-core 2017
Windows 10