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.

current carrying wire forces

Please login with a confirmed email address before reporting spam

Hi,

Im totally new to comsol. I've been trying to create the very simpliest caculation from my range of interest, the electromagnetism. I created the 2D simulation of two paralel wires through which a 1A dc current flows in oposite directions. The wires are in circular area of space filled with air. The surrounding infinite space is aproximated by annual area filled with infinite elements. The resulting field looks quite ok. What I'm concerned about is the resulting forces. The distance of wires is 0.7m, so the resulting teoretical force is 1.429e-7 N/m. When I compute the resulting force as a Lorentz force, I get 5.69e-7 N/m and with boundary integrated maxwell stress tensor I get 2.08e-7N/m.
The wires are 0.02 in diameter, so I should be fairly close to the line current aproximation. Mesh and basis function shape is chosen suficiently fine for the forces to converge(checked).

Please point out what am I missing.

Thanks a lot,
Hans

PS : file attached


3 Replies Last Post 2009年11月5日 GMT-5 12:10
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2009年11月5日 GMT-5 02:52
Hi

You can find several discussions about this subject on the forum, try a search (on the forum DB).
In summary the lorentz forces are normally rather correct, while Maxwell stress tensoer calulations are very mesh dependent , with a corect mesh tey are too correct, if not very wrong !

Use rahter fine meshes in AC/DC

Have a look around, try out, check the models and the doc, as there are a few examples there too, and come back, when you have got further questions

Good luck
Ivar
Hi You can find several discussions about this subject on the forum, try a search (on the forum DB). In summary the lorentz forces are normally rather correct, while Maxwell stress tensoer calulations are very mesh dependent , with a corect mesh tey are too correct, if not very wrong ! Use rahter fine meshes in AC/DC Have a look around, try out, check the models and the doc, as there are a few examples there too, and come back, when you have got further questions Good luck Ivar

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2009年11月5日 GMT-5 10:43
Hi

I had a very quick look at your model, I notice that you use a "thick" infinite element, quite close to the wires, I suspet this is disturbing your Lorentz force calculations.

I get better results if I have more "air" around the wires and have only a thin (I tried th=0.05m) infinite element at the outside of your 1m radius "currently infinite element " circle. Still they are different from the theory but only some 10-15%, I beleive one should try with an even greater air surrounding.

the difference in the maxwell tensor, is for me a meshing artefact, have a look at my simple model in the ModelExchange:

www.comsol.com/community/exchange/71/

the virtual displacements [correction work] is propbably a better approach

there are several nice examples in the AC/DC documentation and the model library

Good luck
Ivar
Hi I had a very quick look at your model, I notice that you use a "thick" infinite element, quite close to the wires, I suspet this is disturbing your Lorentz force calculations. I get better results if I have more "air" around the wires and have only a thin (I tried th=0.05m) infinite element at the outside of your 1m radius "currently infinite element " circle. Still they are different from the theory but only some 10-15%, I beleive one should try with an even greater air surrounding. the difference in the maxwell tensor, is for me a meshing artefact, have a look at my simple model in the ModelExchange: http://www.comsol.com/community/exchange/71/ the virtual displacements [correction work] is propbably a better approach there are several nice examples in the AC/DC documentation and the model library Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2009年11月5日 GMT-5 12:10
I've just done the same simulation according to first model in model library for AC/DC Module. Now it works without a problem. Forces are ok accurate for 3,5k DOFs with cubic elements. I have no idea what have i set wrong previously. I must have missclicked somewhere :). For me the only possible(but still unclear) way to explain the crazy results, I have obtained, is that i used Union button while i have been struggling through my first model definition and it somehow messed the input(dunno if it is possible). But still I have no explanation how it could have happened.

Many thanks for your time Ivar,
Hans
I've just done the same simulation according to first model in model library for AC/DC Module. Now it works without a problem. Forces are ok accurate for 3,5k DOFs with cubic elements. I have no idea what have i set wrong previously. I must have missclicked somewhere :). For me the only possible(but still unclear) way to explain the crazy results, I have obtained, is that i used Union button while i have been struggling through my first model definition and it somehow messed the input(dunno if it is possible). But still I have no explanation how it could have happened. Many thanks for your time Ivar, Hans

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.