Hello,
i believe the statcurrent solver does not check for the "Getfamily()=1" elements in parallel (also not in serial but there it doesnt matters so much).
So the current densitys are singular values there at the orphan nodes.
best regards
Franz
Statcurrentsolve does not account for orphan node boundarys
-
- Posts: 196
- Joined: 29 Sep 2011, 12:25
- Antispam: Yes
-
- Site Admin
- Posts: 4823
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: Statcurrentsolve does not account for orphan node boundarys
Hi Franz
You're correct. There was a request to the opposite some time ago as somebody needed to solve 1D problems. I guess the orphan nodes had been forgotten. Now the system uses a function
which should be ok for all dimensions.
-Peter
You're correct. There was a request to the opposite some time ago as somebody needed to solve 1D problems. I guess the orphan nodes had been forgotten. Now the system uses a function
Code: Select all
IF (.NOT. PossibleFluxElement(Element)) CYCLE
-Peter
-
- Posts: 196
- Joined: 29 Sep 2011, 12:25
- Antispam: Yes