Discussion Forum

I'm not sure if I'm understanding exactly what you wish to do, but if you want two different geometries that interact with each other but theit axes aren't colineal, then I'd model them separately in the same file, and then use coupling variables to link them.

I'm not particularly sure of what you wish to do, since as far as I know an axisymmetric model requires the input (forces, flow, or whatever) to be axisymmetric too, so you'll end up with the same result as using the same axis. Then again, I might be missing something.

If what you need are two revolved elements that interact with each other in a more complex way, then a 3D model sounds to me like a much more flexible (if quite slower) alternative. You could do one in 2D and then transport the result into the 3D object and go from there, if there's no feedback from the second part into the first.

Hi Mario,

Thank you very much for your reply. I have two geometries, in this case two coils with AC currents passing through each of them. Their central axis are not the same (as you say they are not collinear), so I need to somehow define separate symmetry axis for each of these coils. However, I am not quite sure how I can do this in an axial symmetry model.

I think what you are suggesting (modelling them separately and using coupling variables) might work. Could you please explain how I can use coupling variables to link them and what you mean by modelling them separately?

Regards,
Ibrahim

Regarding coupling variables: You can find an explanation of them in page 269 of the COMSOL Multiphysics User's Guide (it's actually page 283 if you're looking at the PDF version).

By model them separately, I mean that if you use the coupling variables, you don't need to model the two different geometries in their correct position relative to each other, as you will define the way they interact with each other mathematically.