RE: [femm] want high-accuracy measurement of force between 2 magnets

["James Rabchuk" <ja-rabchuk@xxxxxxx>]

A number of possibilities:

First of all, have you downloaded the beta version 3.3a of FEMM? It has a
weighted stress tensor that takes an average of a number of integration
paths around the object of interest. Much improved accuracy and consistency.
But ....

you do need to increase the mesh density around the moving object.

Third, make your problem space large enough, and use the mixed boundary
conditions to mimic an infinite domain.

Fourth, try using the coenergy method. The force is the positive derivative
with respect to the space coordinate.

Maybe you've tried these things already, but these are the thoughts that
come to mind right off the bat.

Jim
-----Original Message-----
From: mikeshonle [mailto:mike@xxxxxxxxxxxxx]
Sent: Thursday, March 06, 2003 1:44 PM
To: femm@xxxxxxxxxxxxxxx
Subject: [femm] want high-accuracy measurement of force between 2
magnets


Hi, I'm trying to measure the force vs. distance between two magnets
over the range of 3 to 3.5 inches. I have a script running, etc.,
but the final result is not monotonic and varies considerably and is
sometimes even negative.
What all can I do to increase the accuracy? Is a finer mesh more
important than the solver accuracy? What is a good minimum mesh
spacing and accuracy limit? What is the practical limit of accuracy?
Thanks,
-Mike







Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/