BLDC Motor watercooling ; weird behaviour

I recently was riding my e-bike (with a bldc hub motor) and decided to cool the motor (it -the stator- was approx 100-150 celcius I think) in the river.

The controller is on the top of the bike and definitely stayed completely dry. The motor has cooling holes drilled in its housing through which the water can flow in.

cooling holes in motor housing:


This method was very effective for cooling but the drive train (motor/throttle/controller/battery started acting up after riding it again after cooling...)

Note: after the motor had dried/warmed up again the issues were gone (not sure what caused the issues so not sure if the water or the sudden change in temperature or something else caused the issues).

Symptoms: -weird wining noises from motor when accelerating (much higher pitched and louder than normal) -when the motor reaches a certain rpm (above this rpm) it appears as if the coil groups are misfiring (motor vibrates/shakes and appears to be braking rather than accelerating). The severity of this braking effect appears to be random, sometimes it's worse but not sure why.. The issue appeared to mostly start occurring when a certain rpm was exceeded but sometimes (less often) the issue also occurred at much lower rpm. -the issue was gone after approx 30 minutes of riding when the motor was approx 60 celcius (motor housing)(stator was probably approx 80 Celcius).

There is no temperature sensor from the motor connected to the bldc motor controller (the controller does not have this feature).

What might cause these issues?

Can the water entering the motor mess up the signals coming from the hall sensors? If so what causes this?
could the sudden temperature change cause issues similar to this? If so please explain how
Is there another explanation which I did not think of?

video: https://www.youtube.com/watch?v=l2dHn89Dvso&feature=youtu.be

I tested it again today but this time instead of submerging the motor I cooled it by pooring water on the motor housing using a bottle, total amount of water was approx 1.5 liter I think.. Same result as previously whne I had cooled the motor by fully submerging it..

Another weird effect that I only noticed now on the second test: it appears as if the motor has very little power at low rpms (much lower than before cooling it with water) and then when I get above a certain rpm acceleratiojn rate returns to normal (how it was before cooling with water), until the motor starts acting up/showing the afore mentioned weird behaviour..

Thank you!

The motor could have been damaged before you cooled it due to the high temperatures. It could have been damaged by the thermal shock of being dunked in cold water while very hot. And more more mundane stuff may have happened like water getting into the bearings, which probably aren't great immersed. The halls probably aren't damaged if it runs at all and you don't have sensorless running. By the windings and magnets or magnet frame interface could be. It's also possible the entire assembly is warped.

To me it sounds like your hall sensor signals aren't completely insulated, so the water is messing with the signals until all the water is dried and things are back to normal.

@addy I was thinking something along those lines yes (hall sensor signal being messed up by the water). Indeed they are not fully insulated but I think this could not be the problem since there is only 5 volt on the hall sensors so The current that would flow through the water would be very minimal/non existant? I will check soon by disconnecting hall sensors and checking if the problem persists..

Even with 5V you might still have problems. Also, there's the possibility of the motor phase voltages finding their way through the hall signals when there's water in the motor.