>>> >> >> Fairly straight forward. Sample away from the PWM edges. Try and >> avoid motor commutation edges also. > > The motors have three poles. Is there a straight forward technique > to use to determine where the commutation edges will occur? Probably not, alas. Or, not totally simply. Looking at current waveforms, or voltage waveforms, will allow you to find a repetitive pattern. > Does the commutation take place at a predictable amount of time > after the "turn on" edge of the PWM signal once the motor is > running? That's easier. No :-(. Even at constant motor speed the commutator transitions are not synchronised with the PWM signal so could occur anywhere. How much disturbance they cause in a given case is probably best determined by experiment. You can probably get a reasonable result by performing several samples and rejecting outliers. A motor changes state rather slowly compared to a uPs A2D sampling rate. eg at 6000 rpm that's 100 rps or 10,000 uS /revolution. If there are 6 commutator segments that's about 1600 uS between commutation events. Long enough to do some fairly intelligent sampling. RM -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist