Magnetodynamics - some success (pictures)

[madstamm]

1st of all the thank-yous:
- Elmer team (esp Juha and Peter for the hours spent getting the Whitney solver to run
- Salome team
- Pascal Krezel for the Salome pipe constructor: http://www.code-aster.org/forum2/viewtopic.php?id=14869
- Daniel Boyle for the well-executed sif file: viewtopic.php?f=3&t=1847&p=5651


After considerable time spent trying to find a good workflow to get multiple bodies into elmer with conformal meshes...
(now using Salome>Elmer)
...first success in modeling and calculating the classical magnetic problem of a gapped core with a 10-turn coil

Now my next task:
- how to make sure the physical dimensions are correct
- how to measure flux and inductance using vtk...

Screenshot-ElmerGUI.png

(177.37 KiB) Not downloaded yet

Screenshot-ElmerGUI postprocessor.png

(170.49 KiB) Not downloaded yet

[desA]

Brilliant. That is so incredibly impressive.

[raback]

Hi

Indeed, very impressive. One of the nastiest meshes I've seen done for Elmer, particularly using only open source tools.

Now my next task:
- how to make sure the physical dimensions are correct
- how to measure flux and inductance using vtk...

On the second topic I would think that it would be favourable to calculate the lumped quantities within ElmerSolver. Then you can be consistant on the FE basis that is used. Of course this might need some minor code development but probably nothing that significant.

It seems that in 3D magnetodynamics there aren't that many alternatives around. Hence the marginal utility of the little experimenting that Juha did has been really well taken into use. Particularly considering that the documentation is really rudimentary.

Peter