Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
continuity identity pairs in 3D (magnetic field physic)
Posted 2015年11月13日 GMT-5 05:19 Low-Frequency Electromagnetics Version 5.1 5 Replies
Please login with a confirmed email address before reporting spam
I am working on a 3D frequency (several kHz) problem involving magnetic field and moving mesh.
Excitation is done with external current density flowing in a coil. Impedance boundary conditions are applied on conductive boundaries)
To help the mesh deformation I want to use identity pair.
I just want a continuity of variables along the identity boundary, (the identity pair is located in the gas domain), I use the default condition: continuity + magnetic insulation.
I have succeed to make a very close problem in 2d-axy but it seems that there are issues in 3D.
Someone has some background with this problem?
Thank you
Julien
Hello Julien Givernaud
Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.
If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.
Please login with a confirmed email address before reporting spam
I am facing a similar problem. I am working on a 3D model of an Induction machine. 2D seems to be working fine, but in 3D it appears that the Default Continuity i.e. Magnetic Insulation doesn't work. The flux seems to be limited to the stator iron, I expect it to cross the airgap and enter partly into the rotor as well. In 2D, that is what happens.
Did you manage to solve your problem?
Best,
Faisal.
Please login with a confirmed email address before reporting spam
No I have give up assembly geometry because of this problem.
Julien
Please login with a confirmed email address before reporting spam
Thanks for the reply.
If you eventually did not employ Assembly, would you kindly let me know how did you manage to make a part of geometry rotating and a part stationary? I am not sure how it is possible without that.
Best,
Faisal.
Please login with a confirmed email address before reporting spam
Julien
Please login with a confirmed email address before reporting spam
It is a known issue that, for vector element based physics, assembly pair continuity requires identical meshing of source and destination sides (and then of course no relative motion between sides) for the model to be numerically stable. The reason is that in Maxwell-Ampère's law, the equation of continuity is inherent, i.e. by taking the divergence of both sides of
curl(H) = J
one gets
div(J) = 0
At an assembly boundary with non-conforming meshes, the interpolation errors will result in significant current sources/imbalances and hence an inconsistent right-hand side to the first equation. I should add that there is no known stabilization method for this problem.
This is explained in the model
ACDC_Module/Motors_and_Actuators/rotating_machinery_3d_tutorial
which also shows how to get around the problem by using a magnetic scalar potential formulation (magnetostatic approximation) on both sides of the assembly pair (in the air gap region).
Best regards,
Magnus