Re: [femm] IEE ring problem

[David Meeker <dmeeker@xxxxxxxx>]

ahmad_m_eid wrote:

Dear all

I used FEMM in solving the attached ring problem.
the attraction force I get is about 20 mN while the actual attraction
force is 76.5 mN.

I used Lorentz and Maxwell tensor they give nearly the same answer
around 20 mN.

I attached the problem and my solving in another email.
Please comment on this problem.

the problem found in :
A. Benhama et all.," force and torque computation from 2D and 3D
finite element field solutions", IEE Proc.-Electr. Power Appl., vol.
146, No. 1, 1999.

thanks a lot
ahmad

From your attached picture, you made one "circuit" property called "c" and applied it to both wires.  The way that the "circuit properties" currently work, all regions driven with one circuit property are driven in parallel.  You must have defined a current of 3A/mm^2 * 10 mm * 10 mm = 300 A.  When you apply it to both wires, the current is split such that each wire carries 150 A.  This would reduce the calculated force by exactly a factor of 4 from what it "ought" to be.

The boundaries are also drawn too close to your region of interest--the 5X rule-of-thumb would indicate that you'd want to draw your boundaries at a radius of about 300.  It's a lot more efficient to use an "asymptotic boundary condition" for this case, like in the attached example.  The boundary can then drawn fairly close in, and the results are good--the result of the "Lorentz Force" integral is -7.678973e-002 N in this case--pretty close to your analytical result.

Dave.

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

Attachment: zip00052.zip
Description: Zip compressed data