Re: simple magnets question

[Leigh Palmer <palmer@SFU.CA>]

Chuck Britton suggests:

> Let's say that the pole piece is actually made of two equal sized disks.
> One covered with an (unknown) positive charge distribution and the other
> with an identical. but negative, distribution.
>
> When the two disks are are at rest relative to each other we have the
> required charge neutrality. If we spin the positive disk while the other
> disk remains stationary (or vice versa)we generate a B field with the
> circulating charges on the MOVING disk. OK?)
>
> If we 'ride the rim' of the rotating disk we will see the OTHER disk as
> contracted (locally) and it must be this contraction of the 'background'
> disk that produces the required E field. (I'm sure there may be some land
> mines strategically placed here, but I'm not going beyond the locallity of
> GR)

I, too, had tried this model. It is called an "Amperean current model".
I can't analyze it using special relativity because when I go to an
inertial frame moving (instantaneously) with the charged particle, I
find that my source is a wheel rolling past me*. It is not constant, and
I must take time delays into account to analyze the situation. I can go
into a frame in which the source terms do not vary in time, and in which
the particle is always at rest. It is, however, not an inertial frame,
and SR is not simply applicable. Calculating the radial dependence
sigma(r) of the charge distributions in this model is also a problem to
which I see no easy solution.

Leigh

*In the general case, with slipping!