Re: [femm] Motion Induced Eddy Current Simulation

[David Meeker <dmeeker@xxxxxxxx>]

STEVE ABOAGYE wrote:

> Hi,
> I need help on modeling a motion induced eddy current problem. The
> setup invloves a reciprocatory motion of a stage that has permanent
> magnet pole pieces that induce eddy current in adjacent material. I have
> seen some messages on eddy currents but cannot glean anything to help me
> yet. Thanks for the anticipated help.
>
> Steve
This can be difficult to do in FEMM. If the amplitude of the motion is
small, and if the geometry is amenable to a 2D or axisymmetric analysis,
you can sometimes model this sort of situation via an AC model that
models only the part of the field that is changing with motion. For
example, if you had a cylindrical magnet that was moving axially in a
sinusoidal manner at a frequency w and an amplitude d, you might replace
the magnet by a point current of d*Hc and the upper corner of the magnet
and a point current of -d*Hc at the lower corner of the magnet. You'd
then perform the analysis, prescribing w as the frequency.

Otherwise, you probably have two options. Either beat on the problem
with a more elaborate finite element code, or do it the "old fashioned"
way, by making some sort of a lumped parameter/magnetic circuit model of
your geometry, possibly using a series of finite element analyses to
help deduce parameters for your model if the geometry is complicated.
It's often good to try a model like this first, because it can give you
a sense of what mechanisms/geometric parameters are driving the behavior
of interest (because they are clearly in evidence in the equations).

Dave.
--
David Meeker
email: dmeeker@xxxxxxxx
www: http://femm.berlios.de/dmeeker