We've been riding the "powertrain mule" whenever we can this winter. It's got a 4500w motor and a 300 amp controller set at 75% current limit. It's been stone reliable, in spite of being held together with duct tape and tie straps.
We finally got the first prototype with our own chassis put together. We put on a new 5000w motor to test, and another 300 amp controller. I threw the plastic back on it and hurried off to ride it while the weather was good. It's pretty zippy, I had a big smile on my face and I was testing top speed (53 mph on GPS) a couple miles from home when..... it died. No smoke, no pop, just a slight lurch and then coasting to a stop. Wow, I thought, that's the first time one of these electric bikes has let me down, really pretty surprising that it hasn't happened sooner.
Then I noticed the bike was really really hard to push. Uh, Oh, resetting the circuit breaker isn't going to fix this one! I parked it in a safe spot, found out that the steering lock works, and set off for a nice hike back to the "works" through the Saginaw Forest. It was almost as good a day for hiking as for riding, I didn't mind too much.
The postmortem took days, and involved lots of headscratching as well as bad noises and smoke coming out of the controller. It finally turned out that in our haste to ride, nobody programmed the controller, so it went out with the default 300 amp current limit. We had a datalogger on the bike, and we saw a few seconds over 150 amps, which shouldn't have damaged the controller. But, it means we were putting over 10kw through a motor rated for 5kw. We melted the insulation on some wires inside the motor, which caused a short and killed the hall sensors in the motor, and then killed the controller. Our mistake.
But now I have seen the inside of the motor and the controller, and I have a much better idea of what is going on in there, including thermal issues. We're putting the bike back together with a new controller (same model, but programmed correctly), huge new heatsinks, and a new motor. We've also accelerated our plans to build a dynamometer, so we can test the powertrain in a controlled environment at whatever temperature and load we want.
