On 3 Jan 2006 13:58:38 -0800, jim rozen
wrote:

In article , Christopher Tidy says...

Hi all,

I'm trying to figure out if there is any benefit in adding a flywheel to
a rotary phase convertor.

There is no experimental data on this subject as far as I can see.

I have seen coherent, cogent arguments from respected folks
that support both views - one that it will help, the other that it
will hinder.

Those who suggest a flywheel is bad say that rotary converters
deliver transient power to the generated phase by allowing the rotor to
slip, and a flywheel prevents this.

Those who suggest a flywheel is good say that that flywheels store
rotational energy and will this is made available to transient loads.

Those two preceeding statements are pure paraphrase on my part, and
I of course apologize if I have mis-represented anyones comments.
But there's no empirical data out there as far as I can tell.

It wouldn't be that hard to instrument and measure.

Jim

Some kinetic energy is necessary for the thing to work, but my bet is
that the rotor has more than enough and more would not help.

Kinetic energy is necessary for the idler to produce power in the
third leg during parts of the cycle when less or none is being drawn
from the mains. Energy is also stored in the magnetic field, but its
ebb and flow is in quadrature with third leg power. This is a
cycle-by-cycle event: it accelerates (accumulates energy) during
part of each cycle and decelerates (gives up energy) during other
parts of each cycle. The result is speed ripple, which would be
greater for rotors with small moments of inertia.

The power levels drawn and delivered are a function of slip speed
which governs both stator current and induced emf -- back emf in the
case of the driven windings and generated emf in the case of the third
leg. As the third leg produces more countertorque from higher
current flow thru it, the rotor will slow until slipspeed has
increased to the point where enough power is drawn from the mains to
regain equilibrium.

Observers (Jerry and Fitch) have said they didn't note much change in
idler slipspeed with varying loads. However, resolution of 1% or
better would be necessary to see speed variations because the slip
speed range from no load to full load in most induction motors is
only a few percent at most.