Phase sequence shift

[phil b]

Is there a simple way to take ac power from a 5 phase 500 watt generator and shift all phases into one single ac phase without distroying the sine wave form?

[Flux]

NO

You can take a single phase output but the rating will be low.

The only other way is to rectify to dc and use a pwm inverter to generate the single phase sine wave. Not simple.

Flux

[Ungrounded Lightning Rod]

You could do it with a rotary converter - a free-spinning motor with both three-phase and five-phase windings.

Start with a motor rated for at least twice the horsepower you'd get from your mill if it were spinning at a speed to produce the motor's rated frequency (and ignore the fact that running it that way would fry your genny in minutes - you're just sizing the motor's laminations and windings).  Tear out the existing windings and count the turns.  Rewind it with half the copper allocated for the three-phase winding, half for the five phase winding.  To compute the size of the three phase winding again assume the genny is spinning to produce the motor's rated frequency and compute the under-load voltage.  Start by assuming the same number of turns in the motor but with a wire of half the cross-section, then multiply the turns by the ratio of your genny's output voltage to the motor's rated voltage and resize the wire to divide the cross-section by the same ratio, ending up with the same total cross-section of wire for the three phase winding.  The five-phase winding also has that cross-section so you'll need another factor of 3/5ths for that winding's wire size.

Your slot count will probably come out wrong so you'll have to wind some of your five-phase windnigs with turns distributed between adjacent slots so the composite pole comes out in the right place - and you may get an efficiency hit from this.

(You can get a similar effect by appropriately winding and interconnecting transformers.  But without the mass of the motor's rotor to transfer energy between phases you end up with voltage droop if your three-phase load isn't phase-balanced).

But why in the world would you want to do this?

If you're trying to run a motor it will still change speed and power as the wind changes and the mill speeds and slows.  AC appliances don't like lower than rated frequencies.  Resistive heaters (including incandescent lights) don't care about waveform and frequency - just RMS voltages and currents.  And if you're rectifying it to charge batteries or the like, five-phase is better than three-phase for charging smoothly.

Meanwhile, any conversion is going to drop a nontrivial percentage of your power as heat in the converter.

[phil b]

Thanks for the comments. This is just a 'what if' type question I had while reading about the voltage and current phase relationships in ac when applied to RC circuits. I wondered what if the phases could be matched close enough to be combined simply. That would mean 2 wires from the genny instead of 5. I realize I know just enough about power distribution to be VERY dangerious. It's much safer to ask first than to try it. Thanks.

[Ungrounded Lightning Rod]

This is just a 'what if' type question I had while reading about the voltage and current phase relationships in ac when applied to RC circuits.

Only if you want to throw most of the energy away heating the R.

I wondered what if the phases could be matched close enough to be combined simply. That would mean 2 wires from the genny instead of 5.

You need at least 3 (for two phase or three phase) or a BIG capacitor or inductor sitting beside the genny to shift the phase, to avoid having a generator that varies the load force on the shaft depedning on its position, creating major-league vibration at twice the generation frequency.  Force on the shaft comes from current in the coils, and you need the total force to be constant throughout the rotation - which means more than one phase for an AC machine.

Also:  Phase shifters (other than rotating machinery or vector-sum transformer interconnects) are frequency selective, and windmills have shaft speeds that vary wildly.  You won't get them to work across your mill's operating range.

[Ungrounded Lightning Rod]

By the way:  If you really want only two wires from the genny, you can mount the rectifiers up the tower.  (The two DC wires would have to be a bit heavier than the 3 or five AC wires, but the total copper content could be slightly lower.)

IMHO it's better to bring the phases down, at least to the tower base.  That way you can use electric braking without popping your diodes - and replace your diodes if they ever do pop without climbing the tower.  An extra conductor is a small price to pay to avoid even one trip up the tower - doubly so with the mill spinning.  B-)

[RP]

ULR,

Are you sure about the total copper content being lower with DC?  I thought one of the advantages of polyphase was less copper to carry the same current.

[phil b]

"IMHO it's better to bring the phases down, at least to the tower base.That way you can use electric braking without popping your diodes..."  

I agree. I have the rectifiers on the generator now. It would be nice to hook my meter to one phase to read frequency. Kinda hard to do when it's dc on the ground. So that's change #1. I have shorted the stator out a few times in 10 mph winds. It has slowed down the mill to maybe 10 rpm, but never stopped it completely. This couldn't be good on the diodes.

RP,

There is less copper content needed, ac or dc, if the amps are distributed evenly between 3 wires instead of 2. The amps, IMHO, determines the wire size. Please correct me if I'm wrong.

[Ungrounded Lightning Rod]

Amps SQUARED determines the required cross-section.  For AC that's higher for a given average current than for DC.

For a resistive load amps squared also determines the power delivered to the load so it's a wash.  But for a charging load the average current determines the power delivered so DC beats AC for a given amount of copper.

When you're distributing the current over three or two wires you can also distribute the copper ditto, using fatter wire for DC and achieving the same total amount of copper.  And the insulation and other overheads makes it cost somewhat less to use two fat wires rather than three with a given amount of copper - another win.

But it's not a BIG win.

AC has big advantages for cross-country power transmission:  You can step it up and down with transformers, jacking it 'way up for the long haul (so you don't have to use copper bars the size of train tunnels) down to a dozen or so kilovolts for cross-town (so you can still put a half-megawatt into a neighborhood through wire that wouldn't serve an electric oven at 220), and then to something that's not going to jump out of the wall and fry you in your bed for running to and around the house.

Once you're committed to shipping it as AC, three-phase gets the most power through a given amount of copper at a given voltage.

For generation you need more than one phase to keep vibes down.  Three phase is good.  (I'd have to run the numbers to see if five phase is better or worse for battery charging.  It ends up with more even current and thus less spiking to invoke the square-law penalty, so it might beat three-phase.  But I haven't examined how it does for resistive loss with a given amount of room for copper coils in the stator.)