Menu

Discussion Forum

New Discussion
Filter

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Modeling Electromagnet

Posted Dec 17, 2010, 2:58 a.m. EST Low-Frequency Electromagnetics, Studies & Solvers Version 4.1 2 Replies

mb
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi,

attached is a model of an electromagnet (DC). The excitation happens at the coil endings via "Electric potential" and "Ground".

I managed to solve that model in v4.0a with the direct solver PARDISO with a finer mesh and linear elements. Unfortunately, this method consumes a huge amount of RAM and it is only a submodel for further investigations for me. Now I want to solve that with the iterative solver (coarser mesh but quadratic elements).

When I set a constant damping factor (see solver 1) I get "Divergence of the linear iterations". Depending on what value I set for the factor, this error occurs earlier or later.

When I set automatic damping factor (see solver 2), the solution process stops after 1 iteration without error, but the stepsize is still very large and I don't believe it is converged. Moreover, the solution process always starts with damping factor 1 although I set a smaller value.

What do you think, is this a kind of bug in COMSOL or is there something wrong with the boundaries (they changed from v4.0a to v4.1 in AC/DC module)? What can I do solve that model. I already tested various mesh sizes and solver settings.


Any comment would be appreciated.

mb
2 Replies Last Post Dec 17, 2010, 8:38 a.m. EST
Matthias Mayer
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 17, 2010, 7:25 a.m. EST
Hello Marcel,

i hope i can help

I think your problem is the small conductivity in the air gap. Try a bigger value sigma= 1e-3 then should work your constant damping solver.

For a better result make a new simulation with sigma = 1e-9 and use your last result as start conditions.

Then you can compare your result with automatic solver damping.

kind regards

Matthias



Hello Marcel, i hope i can help I think your problem is the small conductivity in the air gap. Try a bigger value sigma= 1e-3 then should work your constant damping solver. For a better result make a new simulation with sigma = 1e-9 and use your last result as start conditions. Then you can compare your result with automatic solver damping. kind regards Matthias
mb
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 17, 2010, 8:38 a.m. EST
Hello Matthias,

thanks for your hint. Indeed convergence seem to improve with constant damping factor.
With damping factor set to automatic, it still stops after 1 iteration.

Thanks so far,
Marcel
Hello Matthias, thanks for your hint. Indeed convergence seem to improve with constant damping factor. With damping factor set to automatic, it still stops after 1 iteration. Thanks so far, Marcel

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.

Suggested Content