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

Re: [femm] variable inductance motor



bhaskar bollapragada wrote:
Hello,

My name is BHASKAR. I am trying to realise the
following in femm but facing problem. I wanted to
realise a variable inductance motor which is actually
a pretty good model for a stepper motor. When the iron
piece is moved away from the gap, the cross sectional
area of the core will decrease, there by decreasing
the inductance.For a constant flux the internal energy
will increase and a force will be exerted on the iron
piece to attempt to restore it to its equilibrium
position. The force F is given by

F = m0 w N2 I2 / 2 lg 

Where m0 is the permeability of free space , w is the
width of the air gap, N is the number of turns and I
is the current and  lg is the length of the air gap.

For a force of 0.2 N, current of 1 amp, air gap of 2
mm and width of 10mm  we get  the number of turns as
252. I tried to realise this in femm but Iam getting a
force very less than this in the x-direction where did
I went wrong can any one give suggestions.I am sending
the fem file attachment.Please send me your
suggestions.

BBHASKAR

First, you've set the length units to millimeters when, from the description of your geometry, things need to be in units of centimeters. Second, I think that you meant to draw your blocks as 2 cm by 0.5 cm blocks so that you'd get a total of  252 Amp*Turns with a current density defined as 2.52 A/mm^2.  However, the coil cross-section is actually drawn as 2 cm * 0.4 cm, giving you less total amp*turns than you thought.  Lastly, the mesh that you are using around the block is a little bit too coarse for really accurate force computations.  You need to refine the mesh around the block a bit, e.g. by imposing a finer mesh density on the lines that make up the pole and block surfaces.

When I make these changes and run the analysis, I get about a 16 N/m force on the block.  Multiplying by 1 cm in the into-the-page direction, yields a force of 0.16 N.  This is in the same ballpark as the 0.2 N result from the F=(m0*w*(N*I)^2)/(2*lg) formula that one would derive from magnetic circuit theory under no leakage / no fringing assumptions.

Anyhow, I've attached a modified version of your problem with the changes that I'd noted.

Dave.
--
David Meeker

Attachment: zip00033.zip
Description: Zip compressed data