Hey, I'm bench testing a cheapo Chinese controller for a potential scrap build. Using a hoverboard hub motor, 48v variable bench supply that lets me see current draw.

It's one of those learning wire type controllers. I've used the same/similar model on a separate build, no issues. For the bench test, I've connected the battery + and -, phase cables, hall sensors, and a throttle, plus connecting the red wire to battery positive to turn the controller on.
First turn on, no response from the throttle. No problem. Turn off, connect self-learning wires. Turn on, motor spins slowly (for my bench test, forward direction doesn't matter, but just to check, disconnecting and reconnecting the learning wires does indeed reverse direction). While controller is still on, I disconnect learning wires. Motor stops. Now, throttle works without issue. Motor spins up nice and fast, good throttle response. The three-speed switch works as advertised.
I turn off the controller by disconnecting the red "door lock" wire from battery positive. Controller turns off. I reconnect the red wire to battery positive, controller turns on. Now, the throttle doesn't do anything. I switch throttles, still nothing. Turn off and on, still nothing.
I turn controller off, reconnect learning wires, turn back on. Motor spins slowly. I disconnect learning wires, and use the throttle, and the motor works normally again, spins fast with throttle response. I turn off controller again, and turn back on, and motor again doesn't respond to throttle.
So the behavior is that the controller works once, then doesn't work after being turned off and on, unless the learning function is repeated.
Any hints for troubleshooting? Perhaps I just got a dud controller. No big deal if that's the case, it was just a $40 controller, but perhaps someone here has seen this behavior before.
Thanks!

It's one of those learning wire type controllers. I've used the same/similar model on a separate build, no issues. For the bench test, I've connected the battery + and -, phase cables, hall sensors, and a throttle, plus connecting the red wire to battery positive to turn the controller on.
First turn on, no response from the throttle. No problem. Turn off, connect self-learning wires. Turn on, motor spins slowly (for my bench test, forward direction doesn't matter, but just to check, disconnecting and reconnecting the learning wires does indeed reverse direction). While controller is still on, I disconnect learning wires. Motor stops. Now, throttle works without issue. Motor spins up nice and fast, good throttle response. The three-speed switch works as advertised.
I turn off the controller by disconnecting the red "door lock" wire from battery positive. Controller turns off. I reconnect the red wire to battery positive, controller turns on. Now, the throttle doesn't do anything. I switch throttles, still nothing. Turn off and on, still nothing.
I turn controller off, reconnect learning wires, turn back on. Motor spins slowly. I disconnect learning wires, and use the throttle, and the motor works normally again, spins fast with throttle response. I turn off controller again, and turn back on, and motor again doesn't respond to throttle.
So the behavior is that the controller works once, then doesn't work after being turned off and on, unless the learning function is repeated.
Any hints for troubleshooting? Perhaps I just got a dud controller. No big deal if that's the case, it was just a $40 controller, but perhaps someone here has seen this behavior before.
Thanks!