Original poster: Jim Lux <jimlux@xxxxxxxxxxxxx> At 11:07 PM 1/9/2007, Tesla list wrote:
Original poster: Bert Hickman <bert.hickman@xxxxxxxxxx> Tesla list wrote:Original poster: <davep@xxxxxxxx> > Actually, with a suitable zigzag transformer, apparently > synchronization is no worse than just paralleling them. > > Frequency/phase takes care of itself (the generator that wants to run a > bit fast takes more load, slowing it down). Which means it takes more of the load, so there is less than 3X power available. Also the underspeed.voltage units will tend to _absorb_ power, robbing the external load. > There IS an issue when throwing the switch to connect them. > Amplitude is a matter of regulators and throttles. An current sruging baack and forth as the throttles run the units up and down. And the fact the voltage regulation is less important than frequency regulation when paralleling AC sources. And the fact that 'identical' regulators AND governors on three 'identical' units won't be. cf any text book on power system stability. best dwpThe above may be true when attempting to synchronize AC outputs. However, both of the above problems should go away if each generator's output is first rectified before being combined to supply a common DC bus... especially if the bus also has a substantial energy storage cap.
The idea was to make three phase power to run into things like HV power supplies that have 6 or 12 pulse rectifiers (so the ripple is lower)
I suppose you're right. If you had three single phase HV transformers and the rectifier, the generators only need to need to be "close" to 120 degrees apart, and, the actual voltage regulation isn't as critical. Might well be that the load impedances seen by the generators would tend to pull them into approximate phasing. (i.e. if two generators were exactly in phase, the current pulses through the rectifiers would be simulataneous, and lower in amplitude, causing BOTH generators to speed up, but they'd be different.)
Also, I'm not sure how a standard three leg transformer (E core with a set of windings on each leg) would work with imbalanced phases. But that's a pretty standard problem.