Hello,
I have found a discrepancy that seems to be due to some bug:
When I calculate the magnetic energy using the AV Whitney module I get a result which is different depening on the used version of Elmer (parallel or serial).
For example:
serial:
MagnetoDynamics: (Electro)Magnetic Field Energy: 0.26580320720046341
parallel:
MagnetoDynamics: (Electro)Magnetic Field Energy: 0.13963285900101352
The correct value is about 0.240 (measured), so the serial version seems to be the "right" one. However, it is surprising that the serial value is higher than the measured since the A-V formulation should normally give too low values [www.sam.math.ethz.ch/~hiptmair/org/macs ... /dular.pdf, p. 23]. Propably my model is not fully representative of the reality.
Please tell me if you know how this (parallel bug) could be fixed.
Best Regards,
Eelis
parallel AVWhitney: Wrong magnetic energy
-
- Site Admin
- Posts: 4812
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: parallel AVWhitney: Wrong magnetic energy
Hi Eelis
This is a simple parallel reduction issue. I think Juha is working with the fix.
-Peter
This is a simple parallel reduction issue. I think Juha is working with the fix.
-Peter
-
- Site Admin
- Posts: 4812
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: parallel AVWhitney: Wrong magnetic energy
Hi Eelis, I believe this is now fixed. -Peter
Re: parallel AVWhitney: Wrong magnetic energy
Yes, I confirm. Now it works.
Thank you for the fix!
-Eelis
Thank you for the fix!
-Eelis