Can't offer much as I don't run GM controllers but, some things are common to most controllers.
If you're referring to the BAC-281 external controller, it falls back to sensorless mode in the event of a faulty hall effect input sequence. Nice feature. I didn't see anything in the manual which tells us how long it takes to switch back into hall effect mode if the proper input sequence resumes. This might account for acceleration jitters if there was an intermittent signal break. That might be an indication of faulty wiring. Check pinch and flex points.
I don't have much faith in that solution actually, but it wouldn't hurt to check continuity in whichever wires you can access at both ends. Of course, if it's an intermittent fault, you pretty much have 50:50 odds of not seeing the break.
I'm stuck on the idea of a misfire though. Just sounds that way from your description. If one of your phases wasn't firing or firing intermittently, it would be running at 2/3 power (= lower top end speed) and would certainly feel more jittery in the lower rpm range. The assumption then is, that the hall effect wires are fine and the controller isn't using back EMF (sensorless mode) to time the sequence. Presumably, if the controller were expecting a back EMF pulse and it missed one, it would be smart enough to call a fault right away.
There could be several reasons for the loss of a phase, a fried FET or two perhaps, but all of them will have to be chased down to the board level. You might get lucky and spot something obvious, but without the electrical schematic (proprietary), the job might be best left to an authorized pro (okay, probably cheaper to buy a new one).
Personally, I don't have the time or the inclination to reverse engineer one of GM's boards, but you may.
I don't use 'em anyway.
Jeff
