Don,

Point well taken. While we are bouncing around in the vagaries of induction
generators, I believe what you propose is a "perfect" induction generator
(for the load that is). But for an externally driven induction generator it
would seem the scenario is: The 3rd leg is exactly restored by generation
and slips are totally complimentary; then the watt/hour meter would still be
going in the normal direction because of small excitation losses in the
induction machine. Now, if external drive is increased by a small fraction
of rated slip speed, the excitation losses will be overcome and the
induction generator will be driving the grid.

And you wrote: " In practical terms, it's probably easier just to use the
biggest idler
you can find. Bigger is better because Zphase is less so there is
correspondingly less IphaseZphase drop in the driven windings,
correspondingly higher EMF, and correspondingly less IphaseZphase drop
in the third leg."

Yep! This the rationale of those that forget about all that pesky "Fitch
balancing", and just use the biggest honking idler that they can figure out
how to start. Sometimes known as the the "Rozen system"; inelegant but
workable.

Bob Swinney


"Don Foreman" wrote in message
...
On Thu, 15 Sep 2005 12:44:24 -0500, "Robert Swinney"
wrote:

Negative slip is
another point of discrepancy between the two. Many years ago there was a
fascinating series of articles in Live Steam about someone "feeding the
grid" with a steam engine driving an induction generator.


I think an induction motor identical to the load motor might make a
nearly perfect RPC under the following condx:

mains excited on two phases, externally driven to negative slip equal
to the positive slip of the load; e.g., if load is 1725 RPM, RPC is
driven at 1875 RPM.

Now the phase currents in the RPC are all in the "generating"
direction. One phase voltage will be EMF - Iphase*Zphase by
definition. With the third leg of the RPC is connected to the third
leg of the load, third leg voltage will also be EMF - IphaseZphase
-- which is exactly what the third leg voltage should be.

The imperfection here is that if the load motor slows due to load then
the slipspeeds won't be equal anymore so the third leg voltage will be
a little off -- and phase might drift a bit due to slightly different
Iphase Zphase contribution -- but it might be pretty close over a
fairly wide range of loads; probably considerably closer than a
self-driven RPC running at positive slip.

This setup would not run the electric meter backwards because the
drive motor draws at least as much power as the driven motor
generates.