Ummm. . . Not quite. power goes with the square of current, and current is linear with voltage into a constant resistance. (P= I^2 * R)
put another way:
- W/220V = approx 3 amps. (660W, 220V is a common heater size in the US)
- V/3A = approx 75 ohms.
- V/75 ohms = approx 1.3 Amps
- V *1.3A = about 130 watts
- /660 = about 1/5 the amount of power absorbtion capability when running on a little under 1/2 the rated volts. which checks; if you double the volts, power goes up 4 times.
this of course ignores the whole question of the element resistance changing with temperature, etc. . . also ignores the variablility of the volts/current power coming from the mill and any strangeness due to odd wave-forms and wierd duty cycles. (not that I'd expect those to crop up in a resistive circuit like this one)
Watlow makes lots of heaters; they might have a heater that would work for your application. (or maybe they'd be willing to make one; they've made small runs (<10 units) for us.)
as other posters have said, 1 phase heaters are standard; because heaters are resistive devices, there is no problem running physically separate heating elements from legs of a single 3 phase feed, so long as the currents into those heaters are fairly close in value.