Learn more physics!
A friend recently brought me an old gas pump (formerly used in a hospital probably pumped oxygen or air). He asked how it worked. We looked in side and it consisted of a pwer supply that energized two electromagnets. These electromagnets are facing each other with a thin non-magnetic, metal disk between them. When the power is turned on the disk spins and a cam operates a small piston to pump the gas. But why does the metal disk spin? Why is there a net force in one direction? It clearly creates eddy currents in the disk because it becomes quite warm after running a few moments, but these currents would be circular and symmetric and it seems that the forces on the eddy currents would be equal and opposite. The does not appear to be any electrical connection to the metal disk. Can you explain how a time varying magnetic field would cause the metal disk to spin?
- Kimball (age 55)
Gallipolis, OH, USA
This is a fascinating question, but unfortunately mere text files don't do a great job of conveying the complete description of this old pump's construction. Maybe with a few exchanges we can figure this out.
Is it possible that the electromagnets are not centered on the same part of the disk? If, for example, one stirred up circular eddy currents centered toward one side of the disk, and the other made a field centered on a spot rotated a bit from that, the induced magnet in the disk could be pulled around toward the other magnet's field. If the two electromagnets were in phase, symmetry would give another opposite torque. I think that by staggering the phases of the two electromagnets by 90°, one can get a net torque on the disk.
Maybe Lee has a better idea. Anyway, the exact construction details may be crucial. Also, if you can measure the phases of the currents in the two electromagnets, that could be important.
(published on 10/20/2011)
Follow-Up #1: magnetically spinning can
I searched the internet and found a youtube that if explained, I beleive would explain the eddy current motor. The link is below.
I can reproduce the effect and make a pop can spin. So the effect is real and not a hoax. I believe it may have to do with the complex geometry of the electromagnetic field (Fringing effect and possibly near field geometry??) but this taxes my understanding.
If you can explain what is happening to make the popcan spin and why adding the aluminum ring makes it spin faster, then I think the answer to the eddycurrent motor will follow.
- Kimball Clark (age 55)
Gallipolis, OH, USA
At least one of the features in that nice video makes sense with exactly the sort of explanation we gave. The aluminum ring has ac induction currents generated in it, about 90° out-of-phase with the ac magnetic field from the coil. (I assume there's a coil inside the main central tube.) The ring is centered somewhat away from the center of the coil. So there are two ac magnetic fields, spatially shifted from each other, and 90° phase-shifted. They will generate a torque on the can, just as we described before.
So that leaves the question of why there's a little torque even without the aluminum ring. It's hard to say without seeing what the rest of the magnetic source is. Generally speaking, any phase-shifted sources at different locations will generate a torque. Maybe there's a magnetic core, responding a little out-of-phase to the electromagnet, and not quite centered at the same place.
(published on 11/08/11)
Follow-up on this answer.