Re: [femm] Force calculation on an axisymmetric model

[David Meeker <dmeeker@xxxxxxxxxxxxxxxxx>]

Agnaldo Souza Pereira wrote:

> The greater forces arised from femm calculations, with
> a greater number of nodes.

There's probably no way to tell a priori whether the predicted force result
is too high or too low, compared to an exact solution.

Anyhow, the example in my previous message was for two magnets in
attraction, rather than in repulsion as in your problem. That is, the
forces quoted are in a direction that tends to pull the magnets together.
To treat the problem where the magnet are in repulsion, you just flip one of
the magnets. The analytical solution predicts exactly the same magnitude of
force, but with the force acting in repulsion rather than attraction..

To see what happens, I re-ran the same case as the previous message with the
magnets in repulsion. In this case, the force (in a direction so that the
magnets tend to repel each other) is:

Analytical solution: 0.02486 N (same as before)
Force from femm 3.0 using a mesh of about 6000 nodes: 0.02479 N
Force from Quickfield 3.4a using a 499 node mesh: 0.011737 N

The same integration contours were used as in the "magnets in attraction"
case, to keep the comparison consistent. With the magnets changed so as to
be in repulsion, the quickfield result is about 53% low, as compared to the
analytical solution, whereas the femm solution (with a much higher mesh
density) is still within 1% of the analytical result. This seems consistent
with your result.

Again, I guess the point is that if you want accurate force calculations,
use a lot of elements. You can run multiple cases with increasingly fine
mesh densities to check for convergence of the force result.

Dave.


begin:vcard 
n:Meeker;David
tel;fax:781-890-3489
tel;work:781-684-4070
x-mozilla-html:TRUE
url:http://members.aol.com/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