Magnetic field from a permanent magnet

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

User avatar
Minerva
Posts: 946
Joined: Sat Nov 11, 2006 2:58 pm UTC
Location: Australia
Contact:

Magnetic field from a permanent magnet

Postby Minerva » Wed Mar 01, 2017 5:00 am UTC

Consider a small cylindrical neodymium permanent magnet, suppose it is say 5mm diameter and 5mm high for the sake of example.
The magnet engineering data from real-world vendors typically provides the remanence, coercivity, max energy product, etc.

http://www.arnoldmagnetics.com/en-us/Pr ... um-Magnets
See above for typical example data.

Suppose we have a magnetometer sensor some distance away from the magnet, on-axis, and centered on the magnet, say 10mm away.
The only variable is that on-axis distance in one dimension.

So we want to calculate B as a function of x.

B(x) = mu0 * 2 * mu / (4 * pi * x^3) ... should be right.

But that requires the magnetic moment mu, and I want to substitute this in terms of the known parameters from the magnet data.

Can anyone jog my memory as to what I'm looking for to do that?
...suffer from the computer disease that anybody who works with computers now knows about. It's a very serious disease and it interferes completely with the work. The trouble with computers is you play with them. They are so wonderful. - Richard Feynman

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7312
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Magnetic field from a permanent magnet

Postby Zamfir » Fri Mar 03, 2017 1:34 pm UTC

It's often useful to model a (uniformly magnetized) permanent magnet as a current sheet on its surface. That is, the magnetization of the magnet consists of microscopic current loops in the magnet that cancel each other out everywhere, except at the boundary of the magnet. In addition, you model the interior of the magnet as a permeable material (which can simply have the permeability of vacuum, for example in rare earth magnets).

So you basically replace a cylindrical magnet with a very thin-wired solenoid with the same length and radius, and with its turns around the cylinder. The coercivity of the permanent magnet becomes the current density (in Ampere-turns per axial meter of length) in the current sheet.

The above is enough information to accurately model the magnet in a program like FEMM, and find the flux density at any point around it. A closed form will be complicated, though perhaps it's doable if you;re only interested in the axis. I am not sure.

You can do a simplification: model the magnet as a current loop with the same radius but no length, and a current I = Hc*L. That gives a simple expression that will be more exact the further away you get from the magnet.

B(z) is flux density on the axis of the current loop, at distance z from the middle of the loop.

B(z) = muzero / (4*pi) * (2 * pi * R^2 * I)/ (z^2 + R^2)^(3/2)

Now, if z is much larger than R, then that formula simplifies to

B(z) = muzero / (4*pi) * (2 * pi * R^2 * I)/ (z^3)

Which is is the same as your formula for an infinitely small magnet, if we take magnetic moment equal to pi * R^2 * I = pi R^2 * Hc * L = Hc * volume

So at large distances in vacuum (or air), a permanent magnet behaves as a magnetic moment equal to its volume times its coercivity. But close to the magnet (like the distances you mention), the field behaves different depending on the shape of the magnet.

_------------
EDIT: I realized I skipped a complication. When we simplify the magnet to a current loop, we should base that on the total magnetisation of the magnet. That's the coercivity, plus the extra magnetisation induced by the flux if the material has a permeability above vacuum. So the magnetic moment is magnetization times volume, not just coercivity times volume.

Low grade neodymium magnets are easy, they have basically constant magnetisation. So magnetization is roughly coercivity, and remanence is coercivity times muzero. The magnetisation of higher grade neo magnets increases a bit with flux, but they saturate quickly and operate mostly in the saturated area all the way to the remanence point. So the magnetisation is still mostly constant for practical purposes, and can be calculated from the remanence as remanence/muzero. It will be a tad higher than only coercivity.

Some magnets, like Alnico magnets, are unsaturated when you keep them in free space ("open circuit"). Then you need to find out the operating point in free space first, before you can calculate the magnetization.


Return to “Science”

Who is online

Users browsing this forum: No registered users and 13 guests