Re: [FEMM] LIM modeling

[David Meeker <dmeeker@xxxxxxxx>]

Nick Voronov wrote:

> I need help in solution of the following task.
> It is required to create a model of an induction
> device, capable to rotate a metal "wheel" of a
> rectangular cross section, about 1 meter in a
> diameter, at a speed of several  turns per second. 
> As far as I know, it can be achieved  by application
> of bow-shaped linear induction motors  with
> longitudinal or transversal closure of a magnetic
> flux.
> One more condition is the symmetrical distribution of
> the elements of an induction system on a perimeter of
> a working solid.
> Is it possible to model the task and  take the
> good-enough solution in a case of transversal closure
> in a 2D approximation?
> Could somebody give me advice, what type of the motor
> most approaches for a solution of this task and how 
> to construct a model by tools FEMM in the best way?
>
>  
I wouldn't recommend analyzing a transverse flux LIM with FEMM, because 
the secondary current paths are pretty 3D.  For this sort of problem, 
I've used FEMM to help me determine reluctances and so on that were 
rolled into an analytical model of the machine that accounts for its 3D 
nature.

FEMM has been used to analyze longitudinal flux LIMs (e.g. by me). Here 
again, however, great care must be taken in accounting for transverse 
edge effects, or very misleading results can be obtained.  One way to do 
this is to increase the resistance of the secondary to take the 
resistance of the "end turn" part of the secondary current path into 
account, but this is sort of a kludge that only partially accounts for 
the transverse edge effects.

> Another question is how  FEMM computes a bulk coil
> resistance. As I understand, it takes in account the
> "wire" length in z-direction but  not the full length
> of the turns.
> Sorry for my English, I am from Russia.
>
>  
The program does not take account of the end turn resistance.

Dave.