Electromagnet design

**likephysics** · Nov8-09, 07:32 PM

I was trying to get the numbers to design a 1T solenoidal electromagnet with a core of permeability >1000 (electric steel or Permalloy). I think the permeability saturates at some field strength(0.2T?). But I am not sure how to use the formula after saturation. should I just use u0 instead of u0ur?
How exactly is the field strength measured? Say for a solenoid and a button type permanent magnet. At what distance from the surface?
I am using the formula B=u0ur*n*I

**Bob S** · Nov8-09, 11:03 PM

Here in thumbnail is the permeability curve for soft iron. 300 amp-turns per meter should push soft iron above 1 Tesla inside a long solenoid. The equation for the field inside a finite length solenoid is given in Smythe "Static and Dynamic Electricity" 3rd edition page 297.

Bob S

**likephysics** · Nov10-09, 01:23 PM

Bobs, thanks. That makes sense. I just need to grab the static and dynamic electricity book.

**Bob S** · Nov11-09, 01:58 PM

Smythe's exact solution for the on-axis B field in a finite-length solenoid is posted in the thumbnail.

Bob S

**likephysics** · Nov11-09, 08:10 PM

Thanks. Any idea how you measure the field of a permanent magnet?
I bought one of those neodymium magnets. They are supposed to be 1T. But I am not sure how to measure the field.

**Bob S** · Nov11-09, 10:18 PM

I have used three ways. All are difficult.
1) Integrating coil. Wrap 100-200 turns of #40 wire into a coil with ID the same as the magnet OD. Count # turns. Build an integrating circuit with about 1 minute time constant. R series = 100 k, C feedback= 600 uF. Put a 10 meg bleed resistor in parallel with C. choose a low bias current, low voltage offset op amp.
2) Use a Hall Effect Sensor. Make good differential amplifier op amp to amplify signal. Calibration accuracy of sensor is only about +/- 20%
3) Proton resonance NMR. very accurate. Very difficult to set up.
Bob S.

**likephysics** · Nov11-09, 10:57 PM

Proton NMR seems intriguing.
I tried (2), my hall effect sensor (from allegro) saturated at 0.4T when the sensor was close to the magnet (~2 inches). I know mag field drops as u go away from the magnet, but I don't know the relationship(1/r^2?).

**Bob S** · Nov12-09, 12:03 PM

Originally Posted by likephysics

Proton NMR seems intriguing.
I tried (2), my hall effect sensor (from allegro) saturated at 0.4T when the sensor was close to the magnet (~2 inches). I know mag field drops as u go away from the magnet, but I don't know the relationship(1/r^2?).

I am aware that the Allegro units saturate. I attach an Analog Devices datasheet that may not saturate, IF the offset adjustment allows for extending the linear range. The surface mount package is difficult to mount, however.
http://www.analog.com/static/importe...ts/AD22151.pdf
Proton magnetic resonance is difficult in inhomogeneous fields because it requires very small NMR samples, of the order of 1 mm cubed. You also need a grid-dip equivalent circuit (do you remember grid-dip resonance detecting circuits?) to detect resonance.
Bob S
[added] Here is a site that calculates axial field densities for any size cylindrical neodymium magnet.
http://www.magnetsales.com/Design/Ca...tance-Disc.asp
Here is another calculator:
http://www.arnoldmagnetics.com/mtc/calc_gauss_cyl.htm
Here is a third site:
http://www.dextermag.com/Calculation...ldCalcCylinder
Bob S

**Bob S** · Nov12-09, 05:05 PM

The two attached thumbnail drawings outline a voltage integrator circuit for measuring magnetic fields using the Faraday induction law.
Bob S