[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [femm] simple femm, feed back appreciated.

At 16:16 03/02/2003 -0500, you wrote:
Hi Keith:

This is exactly the kind of feedback I wanted, Thank you.

You applied a periodic boundary condition and enlarged the area boundary for
the air space.

I can understand enlarging the area boundary. I can understand changing the
boundary to a circle. After careful reading of Apendix C.3 in the manual I
believe I understand applying the periodic boundary condition to the area
boundaries.

Where to but the boundary can be a bit of a problem if you are trying to simulate an open boundary, i.e. infinite (usually non-magnetic space). In the old days it was done by making A=0 on the boundaries and putting them a long way away from the problem (something like 5 times the size of the region of interest is usually recommended), but that makes for big problems with the majority unwanted. The Asymptotic Boundary is better but it still needs to be quite a distance from the problem. The Kelvin Transformation form though I think could be put anywhere because it represents the "far field" explicitly (am I right here Dave?). It has to be circular though, I think.


Making the mesh the same size in both areas is just common sense since the
problem is so easy the number of nodes in the air space will not bog down
the computer.

Well it just seems sensible for what you are trying to do. You may eventually have to grade the mesh size depending on the approach you use to calculating forces on the magnets; It would need to be fine around the magnetics if you are going to use a stress tensor approach. Most computers these days will eat large problems so mesh size tends not to cause as much trouble as it once did.


Sometimes it takes a very simple example to get the point across.

Do you have a good reference source for understanding the boundary
conditions for magnetic analysis and they way they are referenced in FEMM?

I'm not particularly well read on the subject and I can't say that I've ever seen a straightforward description of all boundary conditions. The references in the manual would be good place to start.


Thanks

John D. Ayer

-----Original Message-----
From: Keith Gregory [mailto:k.gregory@xxxxxxxxxxx]
Sent: Monday, February 03, 2003 2:36 PM
To: femm@xxxxxxxxxxxxxxx
Subject: Re: [femm] simple femm, feed back appreciated.


John,

A few points:

You have no boundary conditions set. The boundary is a bit close to the
problem if it was intended to be open. The mesh density set inside the
magnets was a bit fine and the density outside the magnets a bit course for
what you are trying to do.

I have modified your file using the Kelvin Transformation type of open
boundary and made the mesh densities the same inside and outside the
magnets. If you compare this to your original I think you will see the
difference.

Keith.

At 07:50 03/02/2003 -0500, you wrote:
>Hi:
>
>This is a very simple femm model of 2 magnets with opposing forces. I
would
>like to know if I set this model up correctly to understand the interaction
>between 2 magnets with opposing forces.
>
>The goal is just to better understand how to use the FEMM tool. From doing
>the tutorials, reading the manual and "lurking" on the news group I believe
>the model to be properly set up.
>
>Being a manufacturing engineer and designer my experience with analysis is
>limited to simple static FEA for stress and this model is designed to test
>my understanding of basic magnetic analysis. I can swap different
materials
>in and out, change boundary conditions and do simple experiments and see
how
>the results of the FEMM relate to my simple experiments.
>
>John D. Ayer
>
>
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>



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







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

Dr. Keith Gregory
Senior Lecturer
Department of Electronic & Electrical Engineering
Loughborough University
Loughborough
Leicestershire
LE11 3TU
UK

Phone: +44 (0) 1509 227025 Fax: +44 (0) 1509 227014
Department web: http://www.lboro.ac.uk/departments/el/