[Getdp] issue with 2D axisymetric solenoid magnet resolution
Ruth V. Sabariego
r.sabariego at ulg.ac.be
Tue Feb 12 15:54:49 CET 2013
Indeed, that was the problem. ;-)
BTW, you should due to the axisymmetry, you should multiply CompZ[A] by X for getting the actual magnetic vector potential.
On 12 Feb 2013, at 15:49, Frederic Trillaud <ftrillaudp at pumas.ii.unam.mx> wrote:
> Hi Ruth,
> I think that I found my answer. Here is a solution using VolAxiSphVolume
> and VolAxi instead of VolSphVolume and Vol in the Jacobian method. The
> picture in my previous email was obviously the module of B and not the
> potential vector! (my bad). Here is a new picture with, I think, a
> relatively good solution.
> On Tue, 2013-02-12 at 11:01 +0100, Ruth V. Sabariego wrote:
>> Hi Frederic,
>> The problem came from your geometry: the physical line for the
>> symmetry and infinity where you define the dirichlet boundary
>> condition was not properly defined.
>> You have to use the geometrical lines and not the loops.
>> See the small correction in the geo file.
>> Best regards,
>> On 11 Feb 2013, at 21:05, Frederic Trillaud
>> <ftrillaudp at pumas.ii.unam.mx> wrote:
>>> Dear all,
>>> I have an issue with the solution of a 2D axi-symmetric problem
>>> consisting of a solenoid magnet. The magnet field is not as
>>> Here are the files to run the model in the example.zip attached
>>> solenoid2D.par -> parameters
>>> solenoid2D.geo -> gmsh geometry
>>> main.pro -> model in getdp
>>> magnetostaticProblem.pro -> formulation
>>> gmsh -2 solenoid2D.geo -msh solenoid2D.msh
>>> getdp main -msh solenoid2D.msh -solve potentialVectorResolution
>>> getdp main -msh solenoid2D.msh -pos fieldMap
>>> Dirichlet is applied on the axis of symmetry and the on the external
>>> shell. A fixed current density is given normal to the cross-section
>>> I have tried with different Jacobian volume, VolAxi and
>>> but without success in addition to what is included in the given
>>> Best regards,
>>> getdp mailing list
>>> getdp at geuz.org
>> Dr. Ir. Ruth V. Sabariego
>> University of Liege, Electrical Engineering and Computer Science
Dr. Ir. Ruth V. Sabariego
University of Liege, Electrical Engineering and Computer Science
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the getdp