[femm] Re: No Subject

[Dcm3c@xxxxxxx]

In a message dated 10/26/99 6:14:54 PM Eastern Daylight Time, 
mathewsd@xxxxxxxxxxxxx writes:

> The simplest model of my system is an iron disk with a cylindrical air
> slot that has copper plated walls. The slot contains a solenoid which
> is fed by a sinusoidal current. The induced currents in the copper
> plating generate a B field which isn't negligible. I need to know
> that current so I can calculate heat dissipation. I also need to know
> B "everywhere."
> My plan is to solve for the A vector and then use it to calculate B in
> the slot, but I'm worried about how to handle the boundary conditions
> at the copper. Do I calculate the induced currents in the inner and
> outer copper walls first and then use them as two extra sources in a
> second calculation, or what? I'm willing to assume constant, very high
> permeability. 
> I have been using PDEase and just learned about FEMM in the last week
> or two.

This seems like a relatively straightforward harmonic problem--perhaps I'm 
missing something. You don't have to worry about defining any boundary 
conditions on the copper. Just model the copper plating explicitly as one of 
the regions in your problem, along with the solenoidal winding, iron core, 
and surrounding air. There is no particular reason to assume infinite 
permeability in the iron--it's just another region in the problem. When you 
solve the problem at a given frequency, the resulting B field reported is 
what you desire--the sum of field contributions from both the source currents 
and eddy currents.

Dave.
--
http://members.aol.com/dcm3c