On 2010-05-27, Robert Swinney wrote:
Gunner,

Save your left one and your cash. Juice from a properly designed,
capacitor balanced, rotary phase convertor is indistinguishable from
"real 3 phase". The current circulating within a RPC and load is real 3
phase.

Not quite. The balance is a function of the load. Tune it
for proper balance with one machine (load) and if you turn on a
different one with a different horsepower, the balance will be off a
bit, as will it be when you turn on a second machine while the first is
running. (Each extra machine adds to the rotary transformer capability,
but does not add to the tuning capacitance.)

Though I guess that you could tune for a no-load balance, and
add tuning capacitors to each load machine's motor and keep things a bit
closer.

Certainly under certain circumstances, you can lose seriously
using a rotary converter. An example would be with the old Bridgeport
Series-1 CNC mill with the BOSS-3 through BOSS-5 controllers and stepper
motors. The stepper motor voltages are controlled by a saturable
reactor to tune for either slow step and stationary (lower voltage and
current) or fast step (higher voltage to overcome the switching
inductance of the motor windings). Each axis presents a load on only
one phase, so the balance shifts with X-axis, Y-axis and Z-axis moves --
with load on one often causing the other two to be higher. The problem
is that the motor windings are stepped with power transistors which just
barely had enough maximum voltage rating, and an unbalanced rotary
converter can start popping transistors -- ones which are now hard to
find and thus expensive.

And this is neglecting what imbalance might be introduced by
other loads connected to the same rotary converter.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---