PMSM Simulation from Pavel in ElmerFEM 9.0

[raback]

Hi

The combination of "Discontinuous Boundary" and "Mortar BC" is a rare one. The 1st keyword duplicates the nodes and thereby makes it possible to separate the rotor and stator at the interface. Otherwise one could not rotate the mesh.

The BC elements are duplicated such that you can create a mortar boundary between them. It may be that the index for that boundary is by construction given the index given by the "mortar bc". So in this case the index is given 1st and then the boundary is created when usually it is the other way around.

This is a little obscure. It may be easier to understand what is happening if the mesh is discontinuous at the interface already after meshing.

-Peter

[zmladen]

Hallo Peter,

thanks for the explanation. I think as well that it is better to use the two independent meshes. I have tried to do that with the symmetry boundary conditions. But it is not working as expected. I opened the case at this forum entitled "PMSM Simulation with Symmetry BC".

[kevinarden]

If you use two independent meshes then for me it is easier to have independent boundary elements on each one. In that case one boundary is the master and the other is the slave. The master defines the conditions and points to the slave.

Boundary Condition 2
Target Boundary = 2 ! Master
Mortar BC = 3 ! slave
...conditions
End

Boundary Condition 3
Target Boundary = 3 ! slave
End

[zmladen]

That is what I did in the example posted on this forum, see "PMSM Simulation with Symmetry BC". But it seems that the model is not working properly. It would be great if someone could check what I did wrong in the SIF file.

-Mladen