Error with openFoam

When I'm using the feaTool Interface and try to solve a problem with openFoam I get the following error after pressing solve in the openForm GUI:

        Unable to perform assignment because the left and right sides have a different number of elements.
       
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool

I run everything as administrator and i also remember that quite a while ago, it worked.
Also, solving with openFoam in matlab itself works.
So before I uninstall and reinstall blueCFD, I'd like to check if there's another way of solving this problem
Thanks

I am unfamiliar with this error but if you attach a model file I can have a look at it. If it worked berfore there must have been some changes made.

Leon Robers wrote

When I'm using the feaTool Interface and try to solve a problem with openFoam I get the following error after pressing solve in the openForm GUI:

        Unable to perform assignment because the left and right sides have a different number of elements.
       
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool
        Line 0 in featool

I run everything as administrator and i also remember that quite a while ago, it worked.
Also, solving with openFoam in matlab itself works.
So before I uninstall and reinstall blueCFD, I'd like to check if there's another way of solving this problem
Thanks

here is the file I am using (Unit system is with mm and tons, fluid is air)
CFD_V5_2D.fea

interestingly, the example file with the flow over a backward facing edge (openFoam) works. But if I set up a perfectly simple pipe flow with one inlet and one outlet, openFoam gives again onother error:

Input must be a character vector or string scalar.
 
Line 35 in str2num
Line 0 in featool
Line 0 in featool
Line 0 in featool
Line 0 in featool
Line 0 in featool
Line 0 in featool

This is really confusing and I have no idea why the example file works and all others not.
Thanks a lot

So the issue with the pipe flow was that I set the boundary condition of the outflow to "neutral outflow/stress boundary". By changing it to "Outflow/pressure", it works fine.

Making the same changes to the model I uploaded still leads to the same error

Yes, you are right, for OpenFOAM simulation usually are more stable with prescribed pressue outflow conditions than the neutal/zeroGradient conditions.

Regarding the attached model, this was a bug in the mesh extrusion process affecting meshes with a large number of cells (OpenFOAM only supports 3D simulations so FEATool automatically extrudes 2D meshes to 3D). This issue has been fixed in the FEATool v1.12 release:

FEATool Multiphysics GitHub releases

Leon Robers wrote

So the issue with the pipe flow was that I set the boundary condition of the outflow to "neutral outflow/stress boundary". By changing it to "Outflow/pressure", it works fine.

Making the same changes to the model I uploaded still leads to the same error

Please use the link below if the first one doesn't work (seems Google may have blocked and disallowed sharing)

https://github.com/precise-simulation/featool-multiphysics

Precise Simulation wrote

Yes, you are right, for OpenFOAM simulation usually are more stable with prescribed pressue outflow conditions than the neutal/zeroGradient conditions.

Regarding the attached model, this was a bug in the mesh extrusion process affecting meshes with a large number of cells (OpenFOAM only supports 3D simulations so FEATool automatically extrudes 2D meshes to 3D). There is a temporary fix for this in the updated build below:

[removed link due to release]

Leon Robers wrote

So the issue with the pipe flow was that I set the boundary condition of the outflow to "neutral outflow/stress boundary". By changing it to "Outflow/pressure", it works fine.

Making the same changes to the model I uploaded still leads to the same error

perfect, thanks a lot. Now it works :)
it's amazing how fast and competent you're always answering!

Thank you for the kind words, it means a lot to me. For better or worse, its easier to know how to help when also responsible for the code, as well as not having overhead bureaucracy as in larger organizations.

Leon Robers wrote

perfect, thanks a lot. Now it works :)
it's amazing how fast and competent you're always answering!