Hi Ruairihev & Flux,
My business imports and distributes Aurora inverters, and I've been doing that many years, so I know 'm well...
Power-One specs a minimum of 200V DC for full output power, for the 3.6kW wind inverter. So I would shoot for at least that much. Cut-in can be anywhere above 50V.
For a 12' diameter turbine, assuming reasonable efficiencies, you can expect 3.6kW out to the grid at about 12.5 m/s wind speed (if your alternator can handle that). Cut-in for such a turbine would be around 3 m/s, with about 50W out to the grid at 3.5 m/s wind speed. Since voltage is approx. linear with wind speed; if 12.5 m/s corresponds to 200V DC, then 3 m/s works out to 55V as the starting voltage.
That gives you the lower boundary. Should you go that low? Personally I strongly believe in a fool-proof mechanism to keep RPM under control, not just by furling, but by adding dump-load resistors at a certain voltage and possibly by disconnecting the inverter if a set value is reached (an Omron measurement relay plus contactor will do so very reliably). Even if the maximum working voltage is 'only' 200V DC. I've had to deal with too many broken inverters that were fried due to overvoltage (not covered under warranty by the way), and a factor of 3x is not much for an unloaded rotor to speed up. Keep in mind that the 200V working voltage is loaded. When unloaded that becomes closer to 250V (or more, depending on the stator resistance), so your margin of safety may not be all that great.
In short, rather than shooting for the lowest voltage I would target a voltage where stator losses are reasonable. Heating goes with the square of the current for the stator! Even if that means a higher voltage. In practice many turbines shoot for around 300 - 350V DC at full power, and incorporate good overvoltage protection. Low or high, either way will work fine as far as the inverter is concerned.
I don't know what the design TSR is for your blades/turbine. Assuming it's around 7, then 3 m/s would correspond to 110 RPM. Of course, it's always possible to program the inverter to run the turbine at a slight overspeed in the lower wind speed region, and go towards optimal for mid and higher wind speeds. That's the beauty of inverters where you can program your own curve.
What works best in terms of voltage/wire for the stator would need a few calculations: If you can get twice the windings (double the voltage) onto stator coils, and double the resistance in doing so, the effect would be half the stator current, and half the heat/losses in the stator (losses are I^2 * R, with I half and R double the original values that makes 1/4 * 2 = 1/2). From that perspective going with a somewhat higher voltage than the absolute minimum can have advantage.
Hope this helps!
Shoot if you have any other questions.
-Rob-