Elmer Discussion Forum

Another possibility for integration over internal boundaries is to use the postprocessor. You can use the "Integrate Variables" filter in ParaView (which you seem to use according to your screenshot) to do this.

I see. So what would actually be necessary is to insert addtional supporting points just for the current calculation or something like this. This could maybe be done by using another mesh for the postprocessing steps. The optimal sulution would be, if the nodes at the boundary would get automaticall...

I've been using Elmer to calculate current distributions and stationary magnetic fields for some months now. One problem I'm facing occasionally is the following: When I have a sandwich structure of e.g. thin layers with different conductivity, the current distribution "blurs" from one lay...

I'm sorry, but have to dig this out again. According to this http://www.elmerfem.org/forum/viewtopic.php?f=3&t=1390&hilit=fix+current&start=10#p4114 post, the "fix input current density = true" flag will introduce additional sources to achiece div(J)=0. My question is now: Does...

Solving stationary magnetic fields works fine so far, but now I need to extend my investigations to harmonic analysis to incorporate eddy currents and skin effect (using WhitneyAVHarmonicSolver). I'm not really sure if Elmer can do this. I used the mgdyn_harmonic testcase as a starting point. Up to ...

Thank you for your reply. I changing the IDs below 1000 and the solver still crashes. If I use the -autoclean option, it works, but I don't like that at all. It changes the boundary numbers and I have to look them up in the output every time I change the mesh; Very annoying. (BTW: Is there a section...

Ok. I've attached the testcase, including the gmsh mesh file. It's really just the mgdyn_harmonic test with another mesh. The boundary conditions are wrong, because they are different with my testcase (different surface numbers, ...), but it doesn't make any difference if I put the correct ones in. ...

Thanks to your help I've already figured out how to simulate static current distributions and its magnetic field. Currently, I'm trying to switch to a harmonic simulation with the WhitneyAVHarmonicSolver to incorporate eddy currents and skin effect. I'm using the mgdyn_harmonic testcase as basis, bu...

Yes, "Potential" is the StatCurrentSolvers primary variable. It works as expected, the "problem" was my fault. I thought the result is stored monotonically along the given line and missed the seperate columns for coordinates. Using the coordinates and values together as intended ...

You might skip the "Variable 1" option in SaveLine in order to get all relevant fields. If I skip it, the output is still garbage. (I put the "Variable" option in, to check if it changes anything.) I mean to plot the results you need Coordinate values as well. I haven't found ou...