Re: "Faraday's Disk" which started it all

[William Beaty <billb@ESKIMO.COM>]

On Mon, 5 Jul 1999, brian whatcott wrote:

> Bill seems to see a categorical difference between sensors of the
> electrostatic field and ordinary voltmeters which are current driven
> - or used to be at least.
>  These days, considering a FET input DVM, one sees a direct
> electric field sensing mechanism, in my view!

Yep, the FET input of a DVM does detect P.D. directly.  But the 10^7 ohm
input impedance of a DVM causes it to resemble the old coil-driven meters.
DVMs draw current when measuring voltage, and this prevents us from using
them to detect constant e-fields.  Crude ballpark figures: if the
capacitance across a DVM is around 10^-11 F, and the parallel resistance
is 10^7 Ohm, then a DVM can measure AC e-fields where the frequency is
significantly higher than 1/[(10^-11)(10^7)] = 10 KHz. This estimate might
be off by an order of magnitude or two depending on actual capacitance.
I've noticed that DVMs do seem to respond slightly to 60Hz fields.  Set
your DVM to 200 mV AC range, then ground one lead while waving the other
near a "hot" lead connected to a 120V line.  A small voltage appears.  An
oscilloscope probe held near the output lead of an audio oscillator
demonstrates this clearly, and allows us to see what happens to the
measured e-field as frequency is varied.

Or turning it around, if we want our e-field measuring device to have a
time-constant of 1000 seconds, and if the capacitance across the leads is
around 10^-11 Farads, then the resistance must be above 1000/(10^-11)
ohms, or 10^14.  Electrometers and CMOS op-amp inputs are rated this high,
but DVMs are not.


> At any rate, the experimental evidence was less clear cut for a
> radial electric field.
>  Using Faraday's approach to measuring an EMF at a spinning disk,
> a field of 1 V/M would amount to 15 millivolts over the 15 mm of a
> ferrite magnet radius. An average DVM hits its minimum reading at 100
> microvolts.

Won't people say that this is NOT due to the e-field, but is due to the
EMF caused by VxB?  If the magnet acts as a conductor, then we create a
setup just like Faraday's Disk and Homopolar Generators:  charges moving
in a b-field.  Nobody questions whether a Homopolar generator can
create an output voltage.


> The bulk resistivity of the ferrite specimen was very high,
> but the measured voltage was not entirely consistent with a motional EMF.
>
>  There appeared to be a considerable triboelectric component - and the
> abrasive quality of the surface at the considerable peripheral speed
> involved was hard on the probes. I saw values hovering around
> 30 millivolts in one run.
> But I took no care to neutralize the geomagnetic field, which itself
> provides the basis for a conventional style Faraday generator.

Idea: if you run the magnet CW and then CCW, the difference in absolute
P.D. would indicate a homopolar-generator effect, while the average of the
readings would indicate some other phenomenon such as the Tolman effect
(free electrons pushed to the rim by centrifugal action.)  If you flip the
entire experiment around by 180 degrees while running the magnet in the
same direction, any changes in P.D. would probably be due to the earth's
field.


I don't think I understand the situation entirely, but maybe the magnet
would have to be a good insulator in order for the field to be caused by
Lorentz-transformation of the magnet's b-field.  If the magnet is
conductive, then in the lab frame we would see charges moving in a
b-field.  If we manage to detect a static e-field with an electrometer, we
would ascribe it down to (VxB).  I guess the magnet would have to be a
very good insulator.  If a rotating magnet can cause a potential via
(VxB), but rotating magnets truely cannot create E, we could only discover
this effect by comparing a conductive magnet with a magnet which was a
very good insulator.

How good?  Good enough that charges would not move much during the
duration of the experiment.  Polyethelene in a low-humidity environment
has a sufficient resistivity.  I was under the impression that ferrite
magnets, being ceramic, were extremely good insulators.  If they are not,
then there's no easy way to answer my original question.  Maybe by
distributing Neodymium magnet powder in molten PE plastic...


Well... rats!  I totally missed this part of the problem: the motion of a
rotating cyclotron MUST affect stationary electrons: in the
non-rotating lab frame the moving metal pole-pieces aquire a radial P.D.
as VxB pushes their free electrons around.  This is the e-field from
normal "homopolar generator" action, and not something arising from a
moving magnet.   A "rotating cyclotron" would deflect an e-beam for the
same reasons that a Homopolar Generator would.



((((((((((((((((((((( ( (  (   (    (O)    )   )  ) ) )))))))))))))))))))))
William J. Beaty                                  SCIENCE HOBBYIST website
billb@eskimo.com                                  http://www.amasci.com
EE/programmer/sci-exhibits          science projects, tesla, weird science
Seattle, WA   206-781-3320          freenrg-L taoshum-L vortex-L webhead-L