[Getdp] A-V formulation in "Inductor example"

Matti Pellikka matti.pellikka at tut.fi
Wed Nov 6 16:32:26 CET 2013


Hi, Yea I uploaded a new version of the file
http://onelab.info/files/getdp/Magnetodynamics/GMSH_GETDP/indheat.pro ,
please make sure you didn't just get the old version from your browser's
cache. Here's the diff (upper one is the new version):

155c155
<       { Region Omega; Value 0. ; }
---
>       { Region Omega; Region BdOmega; Value 0. ; }
222c222
< EntityType EdgesOfTreeIn ;
---
> EntityType EdgesOfTreeIn ; EntitySubType StartingOn ;
435c435
<       { Name e; Value{ Local{ [ -(Dt[ {a} ] + {e}) ] ;
---
>       { Name e; Value{ Local{ [ {e} ] ;

The problem with the old version was that the commands to generate a tree
was slightly wrong but worked anyhow on an old version of GetDP. Also, the
electric field produced by the old version was an unphysical one, as it was
just the "curl-free" part of it. Note also that the A-V formulation doesn't
satisfy the equation div J = 0 exactly (in contrast to the T-Omega
formulation, which does). Therefore, some electric current seems to flow
from the conductor to the air. A finer mesh should enforce the equation
more strictly.


-- Matti


On 6 November 2013 11:07, Vadims Geža <vadims.geza at gmail.com> wrote:

> Hi Matti,
> Did you mean that you fixed it? I still have the same results. Can you
> give some comments on this problem?
>
> Vadims Geza
>
>
> 2013/11/5 Matti Pellikka <matti.pellikka at tut.fi>
>
>> Hi Vadims, Indeed, thanks for reporting. It should be fixed now.
>>
>> --
>> Matti Pellikka, Doctoral student, Electromagnetics
>> Tampere University of Technology, Finland
>>
>>
>> On 5 November 2013 17:20, Vadims Geža <vadims.geza at gmail.com> wrote:
>>
>>> The link in the previous post was wrong. I am talking about other
>>> example with cohomology conditions.
>>> http://onelab.info/wiki/Magnetodynamics_with_cohomology_conditions
>>>
>>> _______________________________________________
>>> getdp mailing list
>>> getdp at geuz.org
>>> http://www.geuz.org/mailman/listinfo/getdp
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/getdp/attachments/20131106/70741d58/attachment.html>