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.
Induced current problem in AC-DC module
Posted 27.08.2010, 08:39 GMT-4 Version 5.1 11 Replies
Please login with a confirmed email address before reporting spam
I'm working on COMSOL's ac-dc module and I'm trying to find the magnetic field which is created by an electrode pair (dipole). The working environment of the pair has nonzero conductivity and relative permittivity.
I first found the current density between the electrodes in the 3D electric module and I tried to take the current density value (ie Jx_emqvw) as the external surce of the induction (magnetic) module. Then I started the solver for both electric and magnetic module. However, total current density becomes nearly zero everytime because the induction currents(magnetic) module creates an induced current density equal to the external current density entered by me (ie Jx_emqvw).
What can I do to prevent COMSOL from creating an induced current density?
I would be very glad for your answers.
Regards,
Emin
Please login with a confirmed email address before reporting spam
I'm not sure I fully catc your idea, first of all why not use the quasi static electromagnetic module so you have electric currents AND magnetics coupled correctly.
I'm woundering if you approach is fully coorect, as a current generates normally a magnetic field that will act on the current it depends on the way you segregate this, if you solve for one and then apply for the next you must loop several times as when solving for the first you are not telling your model to adapt to the effect of the other, before you done a full loop. (I'n not sure I'm eay to follow here)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
I did use the quasi static electromagnetic module before I tried this, and all the magnetic flux is created at the boundary. I thought that it is not much sense and I decided to try both modules seperately.
Also, I tried two boundaries which I take the smaller one as continuity and the larger one as magnetic insulation with infinite elements. So, the results was nearly the same, very small magnetix flux and at the boundaries.
I don't fully understood about the procedure that you mentioned involves the loop.
Regards
Emin
Please login with a confirmed email address before reporting spam
Please login with a confirmed email address before reporting spam
I guess what you observe may be the skin effect. AC current flows in a layer at the surface of a conductor. It is exactly what you observe: The induced current cancels the exciting current inside the conductor. The thickness of the skin layer depends on frequency and material conductivity. Search for 'skin depth'.
Regards
Edgar
Please login with a confirmed email address before reporting spam
Thanks for your advice. It may be the reason of my problem.
I create two boundaries, as one inside another, and give the boundary condition "continuity" inside. Is it the reason of the induced current? Does the continuity boundary act as a perfect electric conductor?
How may I solve this problem?
Regards
Emin
Please login with a confirmed email address before reporting spam
no, a continuity boundary condition taks care that the field quantities in your model make a continuous (not step-like) transition across the boundary.
Can you post your model, then myself or other can take a look. What version of COMSOL are you using?
Edgar
Please login with a confirmed email address before reporting spam
Sorry for my late answer. I uploaded the file to megaupload because I can not upload here due to the size.
www.megaupload.com/?d=G9AX54HD --> 2 modules (electric curr. and induction curr.)
www.megaupload.com/?d=EWMSFFGV --> 1 module (electric & induction currents)
The problem I talked about can be seen at the upper one. And the step-like changes can be seen at the lower one (1 module).
I'm using COMSOL 3.4 by the way.
Regards,
Emin
Please login with a confirmed email address before reporting spam
Regards
Emin
Please login with a confirmed email address before reporting spam
I must admit that I don't understand your models. In general, if you don't want induced currents you must set the conductivity of the domain to zero or you must use DC currents.
Regards
Edgar
Please login with a confirmed email address before reporting spam
I don't have time now to see your models, but if you have problems with too big induced current, try this solution from knowledge base:
www.comsol.com/support/knowledgebase/916/
This trick uses potential differences in applying currents to the wire.The example is showing some coil but there is no problem to use this solution in other modules and with straight wires. I use it any time I want see skin effect and it works preety good.
Please login with a confirmed email address before reporting spam
Could you please send that case to me ? Because the site you have showed cannot be found any more. Expect your reply !
Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.