# How to find the poles on a spherical magnet?

To any nonmagnet, the whole sphere is a magnet. To another spherical magnet though, there is a rough area on the surface where it is strongly repelled.

Given a spherical magnet, how should the poles be found?

My crude attempts were as follows

• Grip one sphere in a forceps
• Bring another sphere close to the forceps
• Rotate/roll the spheres until the most repulsion is sensed
• Mark the facing surface of the sphere in the forceps using a permanent marker

The trouble with the above approach is that I rely upon tactile memory to determine maximum repulsion.

Is there a better, inexpensive way to do this?

-
I'm a bit confused. Why don't you just stick 2 bucky balls together and mark the half spheres facing away from each other? – Michael Sep 18 '11 at 16:21
That would work except they would stick over a larger portion of the sphere than they would repel ; I could conceivably mark the wrong position instead of the region nearest/over the pole – Everyone Sep 18 '11 at 16:51
Ahh I see, you want the exact North and South pole? – Michael Sep 18 '11 at 16:52
As close as possible (+: exact would be ideal – Everyone Sep 18 '11 at 16:52
Can some one provide a pointer to a paper describing this phenomena? Some effect related to the bonds? Nuclear spins? I'm just not seeing it. – dmckee Sep 18 '11 at 16:54

If the magnet can support its own weight, I'd stick it on the underside of a piece of well leveled, (ferromagnetic but not magnetized!) sheet metal. It should hang from one pole or the other, as the center of mass of the magnet should end up vertically in line with the strongest part of the pole.

-
Going with this answer, with a modification that there'll need to be two sheet-metal pieces for two spheres at a time to determine the un/like poles. I didn't realize a magnet would hang from the pole ... wouldn't it hang from any part of the sphere that was magnetized? – Everyone Sep 19 '11 at 4:35
assume the sphere is in contact at some point other than where the flux density is max (i.e., the pole). The distribution of the field around that contact point will be asymmetric. In particular, the region of maximal flux density should attract the plate more than other points at the same angle away from the contact point. Most of this force would be upward, but there would be a slight tangential component (going as sine of the angle of contact-center-pole) that should torque the sphere so as to roll until the pole produces no such torque, as happens when the poll IS the point of contact. – JustJeff Sep 19 '11 at 10:45
the torque being proportional to the sine of the angle is problematic, though, in that it wouldn't take much to influence where the equilibrium point lands. the plate should be smooth, as should the magnet, to reduce any kind of friction that would prevent rolling. Any kind of field in the plate would hugely skew the result. – JustJeff Sep 19 '11 at 10:48

Use a tiny compass. It works good.

-

Stick a piece of paper to the magnet over the approximate pole (found from sticking the magnet to a fridge door etc), then stick the magnet back and spin it around on the pole, it will mark the paper with a dot where it was touching the metal.

-

Magnetic field viewing film will show the exact borders of each magnetic domain. So, you rotate the magnet under the film until the equator is shown on the film and shows up perfectly vertical. Remove the film, and now your left hand side and right hand side are the poles.

-

Stick it to fridge, it'll stick by the pole (approximately). For more accurate results do that on top of the fridge, or under a piece of sheet metal. Alternatively you can stick it to a flat magnet.

-

Here's an approach. You'll need:

1)A cup that can hold the magnet, and allow it to roll freely on a flat bottom. A pyrex bowl would be ideal 2)another permanent magnet, with a flat top 3)some spacers. Paper would work fine 4)the spherical magnet, obviously

Place the spherical magnet in the cup. Place the permanent magnet underneath the cup. Add enough spacing so that the spherical magnet can just barely roll Wait for the spherical magnet to settle. It should end up with the appropriate pole pointing toward the permanent magnet.

-

## protected by Qmechanic♦Apr 11 '14 at 7:01

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site.