Discussion Forum

You are naively attempting to solve the problem in two stages, first the currents problem and then the fields, but there are at least two issues involved here: (1) You have not prepared the necessary connection between the solution of the first stage as an input to the second, and (2) your attempt at a solution to the first stage prevents properly accounting for how the magnetic fields would modify the current distribution later anyway (even if this is only a small effect). You would probably be better off attempting this model without using the EC step first. Do it in one stage, not two, and (in your case) focus on what you are currently using as the second stage. This reduces your first stage part of the problem to simply specifying the appropriate boundary conditions in (what is at present) the second stage. You may find this tricky to do right, since the software won't let you simply inject a current in one end. But if you think about it long enough and review the examples in the application library, you'll figure out how to manage this challenge. :-)

Indra,

I think your task can be done in one mf physics. The conductor would then be set up as a numeric coil domain. I don't think you need mef. Key is that the coil terminals are at the outer model boundary as is the case in your model.

Cheers Edgar