The switching action of the Xylotex controller plus
the inductance of the stepper motor have a sort of "transformer" effect.
Suppose the power supply is 24 V, but the motors are need 2.4 V at 2 amps
for full torque. If the Xylotex is set to deliver 2 A to the motor, it does
deliver 2 A average. But the power supply only has to deliver 0.2 A at 24
V, or a little more. 0.2 A at 24 V (from the power supply) is the same
amount of electrical energy as 2.0 A at 2.4 V (delivered to the motor), so
there are no energy conservation laws being broken. But the power supply
current is far less than the motor current.

What actually happens is that the Xylotex draws current from the power
supply about 10% of the time, and uses nothing the other 90% of the time.
The peak current from the power supply is in fact about 2.0 A, but because
of the duty cycle the average current is only 0.2 A. Average current
determines the size of power supply needed. But the motor current averages
2.0 A all the time, so the motor develops full torque.

The place where you need more current from the power supply is at higher
motor speeds. As the motor runs faster, the winding current stays constant
(thanks to the Xylotex) but the voltage goes up, which requires more power.
Since the power supply voltage remains constant, the power supply *current*
increases in step with the motor *voltage*. So to run motors fast,
particularly with more than one motor moving at once, you need more current.

(Dave Martindale)

from

http://groups.yahoo.com/group/Xylotex/message/2384