All hall sensors dying at once

[dnmun]

you can test the output mosfets on the controller to see if they are functional. use the diode tester function on the voltmeter.

test for ths source drain body diode in the output mosfets. the drain is always higher in voltage in the controller so you wanna look at the source-drain body diode by testing with the diode tester between the RED power wire and each of the phases and the BLK ground wire and each of the phases. test both directions. 12 measurements.

it should conduct, when the body diode is forward biased, and it should show open circuit when the body diode is reversed.

 

[Cowardlyduck]

you can test the output mosfets on the controller to see if they are functional. use the diode tester function on the voltmeter.

test for ths source drain body diode in the output mosfets. the drain is always higher in voltage in the controller so you wanna look at the source-drain body diode by testing with the diode tester between the RED power wire and each of the phases and the BLK ground wire and each of the phases. test both directions. 12 measurements.

it should conduct, when the body diode is forward biased, and it should show open circuit when the body diode is reversed.

Bit over my head, but a while back when troubleshooting this intermittent hall sensor issue I entirely pulled apart the controller to see if anything obvious was wrong with it. Nothing appeared to be wrong, and on putting it back together it's been continuing to cause the issue intermittently ever since. At that time I did do a similar test though, and others in the Stealth thread were saying it meant my controller was faulty...but then it continued to work...albeit intermittently.
I posted about it here: http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=52695&hilit=+phase
But noone ever responded.

I don't understand the correlation between the hall sensors intermittently failing and potentially bad mosfets? Why would a bad mosfet cause hall sensor readings to be off?

Cheers

 

[dnmun]

instead of using the ohmeter to measure between the red wire and the phases, try using the diode tester, each way. on both the red and black to the phases. in one direction it should read open circuit and in the other direction it should read about 400-500mV i think.

i don't know if you pushed in new phase wires along with the hall sensor wires, but the hall sensor wires can tear on the inside of the axle and if the sensor lead is touching the axle when the windings shorted to the stator then it would destroy the hall sensor. that was what i was talking about earlier. the phase wire does not have to touch the hall sensor wire where the hall sensor wire is cut.

so the shorted winding could have damaged both. but you would expect just the green phase mosfets to be affected. imo

will know when you can test the body diode and see if they are all identical.

 

[Cowardlyduck]

I still haven't got around to doing the test, but I'm not sure I need to.

Would the above test prove anything considering the controller works sometimes?
What I mean is, the other night I went for a ride, and the controller did it's shuddering thing probably about 3 times all up. Each time it was just a matter of waiting it out, or playing around with light throttle and it would work again. Usually it doesn't take longer than 3-5min of waiting/messing around to get it going again and then it will work anywhere from hours to minutes, but almost always work for longer than it doesn't.

I'm not sure if it means anything, but one time I got it working by crawling along below the controller cutout threshold. What I mean by that is, when the shuddering happens the controller will do the typical 1 second of throttle then cut out to protect itself until you throttle off and on again. However if I only apply a very minimal amount of throttle, I can actually feel/hear the phases clicking on and off, and it won't cut out. When I did this the other night with a load (i.e. me sitting on the bike), after about 30 seconds the bike started working again. Not sure if it's repeatable though.

I kind of almost seems to me like the controller forgets it's positioning and needs a nudge to remember it.

Anyway, just some thoughts...probably way off.

Cheers

 

[Cowardlyduck]

Ok, so I finally got around to reconnecting this motor all back up. I excluded the 2 strands on one phase that were shorted. The motor spins up fine and smooth (apart from a squeak I'm yet to fix).
So if anyone else has a similar issue where only a few strands on one phase are damaged or shorted, there is hope still, and the motor should still work without those strands connected.

I'm also doing some cooling mods with this motor. Checkout my other thread specifically for that.
http://endless-sphere.com/forums/viewtopic.php?f=30&t=56965

Cheers

 

[Dauntless]

Wait a minute. Is it running NO halls, one more more missing, what?

I have what will probably turn out to be much the same problem when I get around to tearing that one apart. Just curious what made your resurrection work.

What I think mine will be if I get it working again.

http://i.huffpost.com/gadgets/slideshows/233709/slide_233709_1864271_free.jpg

 

[Cowardlyduck]

Sorry should have specified.

I replaced all the halls. So it's running as it was pretty much.
Just without 2 copper strands on one phase.

Cheers

 

[Cowardlyduck]

Update on this motor.
I've put about 500km on it since repairing it now, and I've noticed a few things I think are related to removing the shorted strands from the phase.
At very low speeds, or when moving from stopped, the growl and roughness of the motor is noticeably more than it was previously. If I go WOT from standstill, the motor will judder quite violently, and loudly for the first few meters, before smoothing out.

The other issue, which is very much related is Regen is not as smooth as previously, but the roughness is nowhere near what I get from acceleration under WOT.

The roughness as low speeds seems directly proportional to the amount of current fed to the motor, which makes sense.

Both issues are not show stoppers, and I will continue using the motor indefinitely as it's not so bad to make me concerned at this stage.
I might eventually get a sinewave controller like an Adaptto which may help if it can detect one phase is a little weaker than the others.

Cheers