Re: [femm] How to measure force on linear system?

[David Meeker <dmeeker@xxxxxxxxxxxxxxxxx>]

"Davis, Richard W (WNS)" wrote:

 

I'm looking at eddy current brakes--linear kind, not rotating.

I'm looking for all the guidance folks can provide, but here are two specific questions...

• I'm using arrays of magnets (wigglers or undulators).  What's the recommended method to define a circuit, given that what I have is a large number of interacting circuits?  (Basically the same problem as a linear motor.)

Generally, femm doesn't treat eddy current problems with motion. 
However, you can fool it into solving this problem.  Perhaps I'll
try to write a short note explaining how to do it in more detail, but here's
a quick outline:

• Geometry is a conductiving plate moving past an array of magnets (could also be an array of magnets moving past a long conducting plate).
• Assume that the geometry is very long in the into-the-page direction compared to the pole pitch of the magnets in the array.
• Assume that the brake is long enough so that entry/exit effects can be neglected.
• The first real "trick" is to represent the permanent magnets as equivalent current sheets that flow on the sides of the permanent magnets.  This distribution of current could be represented in the direction of motion by a Fourier series.  If we consider just the first term, we get a sinusoidally distributed current density, rather than current that is concentrated at discrete points.  However, the field seen by the plate isn't changed that much by considering just this first harmonic.
• Represent the fundamental by a few sections with constant current amplitude, but with different phases.  This way, the field appears to move across the conducting plate.
• Then apply antiperiodic boundary conditions on the sides of the domain so that you only have to model one pole's worth of the brake.  See attached, for example.

Perform the analysis from the point of view of an observer fixed on the

• Is there a way to get FEMM to tell me the linear force (assume the magnets are moving past the stator or conductive plate).  I know the relative velocity of magnet and conductor, and I know the period (wavelength) of the magnetic circuits, so the equivalent frequency is easily calculated.  It ranges from 0 to about 60 Hz.  (P.S.  I think I can guess what the force is at zero Hz.)

After you do the above sort of model, probably the best way to get forces
on the plate is to integrate the Lorentz force over the plate area. 
This is one of the "stock" integrals that you can do in femm.

Dave.
 
 

Attachment: zip00022.zip
Description: Zip compressed data

begin:vcard 
n:Meeker;David
tel;fax:781-890-3489
tel;work:781-684-4070
x-mozilla-html:TRUE
url:http://members.aol.com/_ht_a/dcm3c
org:Foster-Miller, Inc.;Electrical and Electronic Systems Group
version:2.1
email;internet:dmeeker@xxxxxxxxxxxxxxxxx
title:Senior Engineer
adr;quoted-printable:;;350 Second Avenue=0D=0A;Waltham;MA;02451-1196;USA
fn:David Meeker, Ph.D.
end:vcard