I've been trying to get my head around the interactions of moving magnets, toroidal magnets, and toroidal cores. As I understand it, we have a situation like this:
1) Toroidal magnetic core has a non-toroidal field when current is off. Another magnet will be attracted to the core. 2) Current on => field of core is *rotated* so that it becomes entirely toroidal. At this point, the field outside the toroidal coil/core combination *vanishes*. The external magnet is no longer attracted to the core, and can be moved away. 3) Current turned off again => field of core rotates back, and the externally visible field returns. Magnets are once again attracted to the coil/core combination. So exactly what happens when we throw an external moving magnet into the mix? Here's an approach to visualizing it: Real magnets are complicated but we can imagine something much simpler which may clarify this. Imagine that the magnetic core consists of a myriad of tiny magnetic dipoles mounted on gimbals. A lot of little springs hold them in the orientation which produces the externally visible field. Furthermore, the gimbals can be locked and unlocked, using negligible energy. Now let's look at some interactions. * * * First, no external magnet: A-1) In field-on position, springs are "relaxed". A-2) Turn on the coil. Dipoles rotate against the spring force until they are parallel to the applied field; at that point the external field vanishes. Springs being conservative, if we don't want our dipoles oscillating, we need to add some friction, which damps the motion; that results in nearly all the energy we just pumped into the system turning into heat. The rest of the energy went into the springs, which are now tense. There was back EMF on the coil during this step, and it is caused by the rotating fields of the dipoles. That's where the energy comes from to turn the dipoles. A-3) Turn off the coil. The energy of the springs comes back out and turns mostly into heat (through friction) as the dipoles rotate back into their "field on" positions. There's more induced EMF in step (3), as the energy of the springs turns partly into electrical energy. * * * Now let's add an external magnet to the mix. B-1) In field-on position, with gimbals locked so the dipoles don't jiggle, bring an external magnet up to the toroid. We get useful energy out as it's attracted to the toroids. B-2) With the external magnet *stationary*, turn on the coil (and unlock the gimbals). The dipoles rotate to line up with the coil's field. (Assume they line up *essentially* exactly with the field of the coil, which is assumed to be far stronger than the field of the external magnet. Alternatively we can run the experiment other-way-around, with field turned off in step (1) and magnets repelling each other in step (3), and avoid the misalignment problem entirely.) The external field vanishes. What's the back EMF in this step? It's the SAME AS IT WAS IN STEP A-2. That's because the back EMF is caused by the *change* in the B field inside the coil, and that's the same in B-2 as it was in A-2, due to linear superposition. Current the same, voltage the same, means we the energy dumped into the coil is THE SAME in B-2 as in A-2. B-3) Pull the external magnet away (against *zero* resistance from the toroid, which has no visible field at this point). Then, with the external magnet far away, turn off the coil. The induced EMF in B-3 will be, once again, identical to that in A-3. * * * We got energy out in B-1 which we didn't get out in A-1, yet we put the same amount of electrical energy into the system in steps B1-B3 as we did in steps A1-A3. Where'd that energy come from? This is the Steorn Mystery. Here's what I think is the answer, in the Gedanken I just described: The effect of the external magnet's field in step B-2 is to reduce the total torque on the dipoles. Consequently they gain less energy as they swing into "field-off" position, and as a result less energy is turned into heat in that step. (They move a little more slowly, by the way.) So, the magnets warm up *less* when the motor is running than they do when the motor is not running. But, the back EMF is identical in both cases. Essentially, when the motor is running, some energy which would have been wasted as heat goes to mechanical work instead. Of course, this is an analysis of a gedanken experiment, which may or may not apply to Steorn's motor. None the less it's a gedanken which was bugging me, so whether the result applies to Steorn or not I still found it interesting :-) .
