[femm] Re: Current Density X-Y plot problems

[Dcm3c@xxxxxxx]

In a message dated 1/11/00 10:55:58 PM Eastern Standard Time, 
mperry@xxxxxxxxxx writes:

> Let me start off by saying that I am new to FEA and just blew the dust
> off my EM textbooks.
> 
> I am attempting to look at the current distribution a winding of a
> planar inductor. I set up a simple model where the winding is just a
> single turn copper section on each side of a gapped E core. In the
> materials section I set J to be equal `x' for one side and equal to
> `-x' on the other side (where x is the current density to yield the
> desired current for the geometry). Next I performed an analysis at
> frequency and commanded an X-Y plot to look at the magnitude of J.
> 
> All I can say is that the shape of the curve looks ok but the scale has
> me troubled. Way troubled! In the case where the current density is
> uniform in the Y axis should I not be able to integrate the current
> density over the length and back out the driving current?
> 
> The ice is a little thin here so any help to get my self on track would
> be very appreciated.

When you assign a volume current to a region with nonzero conductivity, this 
is really just the "source current". That is, the current that would be 
there at zero frequency. At non-zero frequencies, you get induced currents 
in the conducting regions that tend to opposed the defined source currents. 
The result is a change in the distribution of the currents (i.e. skin 
effects), as well as a reduction in the total amount of current flowing in 
the section (i.e. effects of inductive impedance).

Anyhow, when you define a current density to a conductive region, what you 
are really defining is the the voltage gradient applied to that region 
multiplied by the conductivity of that region. The total current in the 
region isn't constraint to be constant (this option is coming in the next 
version of the program).

Dave.