1. Unable to detect a faulty slave board or easily test and confirm they are functioning correctly. Peter has put led's that give an indication of what they are doing, but 8 of my cells are tucked under the rear seat where the fuel tank was located.
2. My current method of shutting down the controller in a under voltage or over voltage (think regen, full pack and starting from the top of a hill, all of which I can achieve.
) is possibly not safe.
So, solution to the first problem. Peters idea of sending the low voltage warning cascading through all boards provides a method of testing the functioning of the system. I can easily implement this with my controller as a automatic test when the ignition is switched on or the battery charger is connected. If the test fails, then no go. All that is left was to figure out how to test the high voltage cut-off signal. It occurred to me that I can reprogram the slaves so that they signal both a high or low voltage warning on the same output. If we get a warning while charging then it is a high voltage fault and we can implement charger control. If we get a warning while driving, then just determine weather or not regen is active and take appropriate action.
The second problem has to do with the fact that I was feeding the idle and brake signals into my module and then sending them on to the controller. I never did like this approach and actually did not implement it on the car. Just had the low voltage warning only. My new approach will be to instead shut down the controller by turning off the IGN (or KSI as most controllers label it) terminal. This will be much safer because as long as the idle switch is closed the controller will not activate, no matter what throttle signal it recieves and stepping on the brakes overrides the idle switch and throttle signal. That is why it is better to keep them connected directly to the controller but monitor their status instead.
Greg
