Geared Hub Motor Over Current

[chuyskywalker]

I've had this MXUS XF19FAT motor for a while now but it was only on my bikes for a little bit for lots of various component switching around reasons. I pulled it out recently for a cargo bike build where I really wanted the torque it ought to provide over a DD hub.



Well, something is amiss with it and I'm stumped. I've run it through two phaserunners, two Trampa 100/250 VESCs, and even picked up a Sabvoton SVMC7245 and all 5 of them give me the same problem trying to run the motor; over current cut outs.

I can tune the motor with each controller, and I can get it to run, but eventually (either with a quick full throttle blip, or at max speed) it always goes into whatever that controller's "over current" protection is. I put the stock motor back on the bike and it has no such problem.

TBH, I'm throwing the towel in on this motor, but I'm curious if this kind of behavior has some kind of mechanical/wiring explanation, or is maybe resolveable.

 

[amberwolf]

Does it only happen if using the hall sensors, or does it happen even if in sensorless with no other wiring besides the phase wires connected to the controller from the motor?

Does it happen only under load, or offground as well?

Does it happen at any throttle amount, or only above some limit?

If it's a phase overcurrent, then it probably means a short between one of the phases and something else--either another phase, or the stator laminations.

A fault that happens under *any* conditions, if caused by a short, means the short is detectable with a multimeter in resistance mode. Since all phases are connected to each other inside the motor, you can't test for a short between them or the phase wires (without disconnecting the WYE connection), but you can test for a short to the stator, since it's also connected to the axle mechanically and almost certainly electrically.

To test for the inter-phase short, you have to open the motor and disconnect the point where all three phases are connected to each other, and then you can measure resistance from each phase to the others (which should be infinite this way).


Or, maybe, spin the motor backwards by another motor friction driving the tire at a constant speed, measuring the voltage between all three phases (with three multimeters set to AC volts, if you have them). If the voltage is identical between all phases, it's either not an interphase short or it's enough resistance that it's not loading that voltage down.


If there's a short to the laminations, then you might be able to locate and fix it by reinsulating that (like with CoronaDope or similar, soaked down into the windings and let to cure).

If it's a phase-tophase short, then if it's in the phase wires from motor to controller, it's almost certainly in the part that goes thru the axle, probably right at the axle exit.

If it's in the windings, it's probably in a place that gets mechanically vibrated, like the long stretches of wire that go from stator tooth to stator tooth in some motors.


FWIW, sometimes a motor with a short like this works "perfectly fine" on a generic non-FOC controller that doesn't bother to measure phase current,