[MPaulHolmes]

I think all the testing with sometimes really badly behaved voltages and currents and overshoots because of too big of P's or I's and the micro-controller sometimes going into latchup may have caused a problem with my control section. It already was a "walking wounded" because it was in the exploded controller. I can't do any other testing until it's fixed or replaced. boohoo... I did learn the following:

Probably the best performance was one of the PI loops, but I forgot which one, and the controller is acting weird.

Another disturbing fact: I would compile a program, run it in the simulator, and it would have completely different behavior in the car. This is all very disturbing.