On Tue, 18 Mar 2008 22:28:04 -0500, "Phil Kangas"
wrote:


So, many experimentors start thinking, that if we build it,
and spin
it up, we can add a load to it, and it won't take power to
keep it
spinning, since it won't react against the motor driving it.

Experiments show anomolous results, and building a model to
play with
is truely trivial.
I challenge any of you to build a model to prove that it IS
confusing
to think about.

Here is a simple experiment to try..
http://jnaudin.free.fr/html/farhom.htm

The cited experiment sez it generates 0.5 volts, 1.5 mA or 0.75 mW at
2600 RPM. This power level at this speed represents a torque of about
..0004 ozf-in (.00275 Newton-mm to y'all unit purists) regardless of
how the current was generated. It ain't zero, but it's easy to see
why an observer might think so. Compared to countertorque due to
parasitic losses like windage, bearing losses and brush drag it's a
fleafart at a rock concert.

I suggest the following modification to the experiment:

Rather than continuously powering rotation with an electric drill, use
a flywheel. With brushes in place but external wires disconnected,
spin up the flywheel (perhaps with a tangential air jet), then
release it and note the time it takes to slow to a given rate as
measured with a photo tachometer. Could use a hole in the disc, LED
and photosensor from surplus store or discarded mouse and oscilloscope
-- or perhaps just a pattern of radial lines on the disc viewed with
fluorescent or neon light and watch for the "strobe stop".
Now repeat the experiment with external leads connected.

Parasitic losses are the same in both cases if both experiments start
at the same speed.

If the flywheel slows more quickly when the external leads are
connected, there is clearly countertorque due to the rather small
generated current.