Shorted out hub by short on phase wires while riding?

[Skedgy Sky]

I was pedaling with rare of use throttle, just strolling around the park. Motor cool to touch as I haven't gone hard on it yet yesterday. Suddenly: I go WOT and my motor boggles down and grumbles.

What I thought the cause and result was: My (not very insulated) phase wires shorted either against the frame or each other and blew FETs in my controller.
http://i.imgur.com/t4Wi3ml.jpgc
What must have happened: My phase wires shorted and instead of controller getting blown, motor phase wires or windings now have a short (how's that even possible with a cold motor?)


Resistance on controller phases: high and consistent... FETs look healthy
http://i.imgur.com/PNdnXpl.jpg
http://i.imgur.com/Vpsg6eI.jpg

Resistance on hub motor phases: 0 ohms
But windings look good:
http://i.imgur.com/YTZMrnO.jpg
http://i.imgur.com/n83QR7X.jpg

Take note of the wires at the axle. They look slightly worn.
I wiggled the cable around at the axle, with multimeter hooked up to phase wires. I saw intermittent jumps of high resistance, which may mean my phase wires are shorted within the axle?

 

[Spicerack]

Direct drive?

Yes you could have shorted phase wires in the axle. Common enough. Still check tho:

Lift wheel with phase wires unplugged.

Spin wheel. Now connect any two phase wires together (from the motor). Resistance to spinning should be alot greater. Now repeat so that all phases have been tested against each other. If there is no difference between manually shorting the phase wires and none connected together, you have a wiring issue somewhere.

Next test is halls. Easiest to get a tester.

Obviously you've checked all connections as they sometimes wiggle loose especially hall ones.

Many more tests from here on but these are the first, most basic and usually the problem.

 

[Skedgy Sky]

There was already some resistance on the hub motor with power wires unplugged.

I didn't feel like assembling the motor again to test if there would be more resistance by shorting them out manually.

Instead I decided to isolate the axle from the rest of the hub so I cut the three phase wires and tested resistance again.




Very little resistance on the windings. It's the same for each one.
I couldn't have possibly burned my windings with a cold motor, right?

Edit: I put the hub together with the phase wires cut from the axle and the resistance/cogging effect I'm getting from the wheel feels like a dead short across all phases.
Again, I don't see how I could have shorted my windings and not my controller when the phase connectors came in contact of one another near the bike frame yesterday (let alone this all happening with a cold motor). (http://i.imgur.com/t4Wi3ml.jpgc)

 

[dnmun]

the windings are normally shorted through the wye.

 

[Spicerack]

A normal multimeter ain't much chop for measuring motor windings- that is a normal result you have there. Without using an ebike/brushless motor tester (which would tell you instantly where the problem may be- phases or halls and associated wiring), the only way is to turn the wheel by hand. You'll get a little bit of cogging with no phase wires joined together, a lot more cogging with any two wires connected and a slightly different cogging with all three connected together (but still a lot more than with no phases connected together).

The difference in cogging between no phases connected and some connected is massive.

 

[Skedgy Sky]

A normal multimeter ain't much chop for measuring motor windings- that is a normal result you have there. Without using an ebike/brushless motor tester (which would tell you instantly where the problem may be- phases or halls and associated wiring), the only way is to turn the wheel by hand. You'll get a little bit of cogging with no phase wires joined together, a lot more cogging with any two wires connected and a slightly different cogging with all three connected together (but still a lot more than with no phases connected together).

The difference in cogging between no phases connected and some connected is massive.

Yes, this "cogging effect" is very extreme with this motor, as I have said in a previous post.
I know the difference between none, two, and all three phases being shorted.
All three phases are shorted out with each other and it appears to be coming from the motor windings (not anywhere along the cable).

Hall sensors appear fine, and wouldn't cause what I am experiencing.

I hooked up the stator alone to the controller and turned up the throttle to find considerable heat building up from one part of the windings while the rest was cool.

Unless there's some easy way to find and/or replace the bad section of the windings, I think my motor is useless/junk in the state it's in... and I don't know how it came to this by running it when it was ambient temperature. :|

 

[John in CR]

Motor doesn't look burned or measure shorted at the raw phases. Test the halls while you have it opened, and give it a new and improved wire harness. Maybe you just got lucky and it was a bad solder connection on a phase wire and it melted the solder and came loose when you pushed more current. Otherwise the short probably blew the controller.

 

[dnmun]

the phases are normally shorted together at the wye. that is not causing the cogging. cogging is caused by shorted mosfets in the controller.

 

[Skedgy Sky]

the phases are normally shorted together at the wye. that is not causing the cogging. cogging is caused by shorted mosfets in the controller.

The motor is unplugged from the controller when I turn the axle (I used a wrench to grip and turn the axle). I feel strong intermittent stops of resistance when I spin it. The same if I lock the axle down on something and spin the outside diameter of the wheel.

I physically feel the sign of what must be shorted windings without the motor visually looking like the windings are burnt.

 

[dnmun]

something may be rubbing in the motor or the bearings may have bad spots. did you test to see if the phases are shorted to the stator? that is the only thing i can think would allow them to short besides the wye since they don't get wound on the same pole and the windings don't come too close as they go around the stator from end to end.

 

[John in CR]

the phases are normally shorted together at the wye. that is not causing the cogging. cogging is caused by shorted mosfets in the controller.

The motor is unplugged from the controller when I turn the axle (I used a wrench to grip and turn the axle). I feel strong intermittent stops of resistance when I spin it. The same if I lock the axle down on something and spin the outside diameter of the wheel.

I physically feel the sign of what must be shorted windings without the motor visually looking like the windings are burnt.

Does it get stronger if you try to turn it faster? Turning at the axle is somewhat hard, and you can feel a notched cogging too. Are you sure two of the phase wires outside aren't touching each other. If it is kinda hard to get started but gets easier, then that's normal. As someone stated before, you can alternate shorting 2 phase wires together. Each shorting should make it more difficult to turn. If a pair feels the same with the ends shorted or not, then those are the pair that have a short somewhere else other than where the other ends of the phases are terminated to each other.

 

[Skedgy Sky]

the phases are normally shorted together at the wye. that is not causing the cogging. cogging is caused by shorted mosfets in the controller.

The motor is unplugged from the controller when I turn the axle (I used a wrench to grip and turn the axle). I feel strong intermittent stops of resistance when I spin it. The same if I lock the axle down on something and spin the outside diameter of the wheel.

I physically feel the sign of what must be shorted windings without the motor visually looking like the windings are burnt.

Does it get stronger if you try to turn it faster? Turning at the axle is somewhat hard, and you can feel a notched cogging too. Are you sure two of the phase wires outside aren't touching each other. If it is kinda hard to get started but gets easier, then that's normal. As someone stated before, you can alternate shorting 2 phase wires together. Each shorting should make it more difficult to turn. If a pair feels the same with the ends shorted or not, then those are the pair that have a short somewhere else other than where the other ends of the phases are terminated to each other.

It gets stronger as I turn it faster.
Yes, phase wires outside were not touching each other. I even cut them off on the inside to isolate the wire coming out of the axle as a potential problem (I resoldered the wires on the inside of the motors afterwards).
The cogging is more significant when I manually short the connectors, but it is already pretty bad even when the connectors are seperated. It was not like this before. Again, feels like the windings are slightly shorting somewhere.

 

[John in CR]

Again, feels like the windings are slightly shorting somewhere.

If turning the axle gets harder and it's a very notched effect then 2 are definitely shorted. You can figure out which by shorting the ends of 2 until you find which don't change the effect because they're already shorted. If it's in the windings like you suggested, then it's toast without a rewind unless you can find the strands with the varnish scraped off to short and put some coating to insulate. What makes you think it's not in the wiring harness, which would be far more common? Maybe you just got a motor winder noob who left some unprotected corners and scraped the varnish pulling the wire through and it just took some time before the short completed. Have you checked for continuity between each phase and the motor itself?

 

[dnmun]

did you check to see if the phase wire is shorted to the stator. you can do it while the motor is still together by looking for continuity between the case or axle and the phase wires.

 

[Skedgy Sky]

What makes you think it's not in the wiring harness, which would be far more common?

I said I isolated the cable going through the axle by cutting it on the inside of the hub at all the 3 phase wires. This means that there was no connection on the inside from the wiring harness to the windings. I put the hub back together and the extreme cogging problem persisted. I have since soldered the three wires back to how they were.

did you check to see if the phase wire is shorted to the stator. you can do it while the motor is still together by looking for continuity between the case or axle and the phase wires.

I did check, there is no connection between the stator and any of the phase wires.

I carefully felt the force the wheel was giving back to me when I rotate it by hand by checking the three pairs and found that: shorting yellow and blue, and yellow and green together gives back most spinning resistance.

In short, the least to greatest spinning resistance/cogging effect combinations is as follows (but all are pretty severe and the differences in how it feels is minimal so the order in the middle of the 4 may be wrong):
1. Wheel with phase wires connections separated
2. Any color combinations except the following two
3. yellow & blue, and yellow & green
4. All phase wires connected together.



The question here that caused all of this really is: why did my motor get damaged and not my motor controller?
(Again, the day it happened: I went to apply throttle and right after I did my motor stopped and grumbled when throttle was applied. I discovered a mark on my frame indicating that two phase wire connections touched each other at the point of connection between hub motor and motor controller, yet motor controller is not damaged but hub motor is).

 

[gusty]

I'm kinda in the same position as you just that its a new motor and I haven't used it except for a test ride I did hear a whip while tying everything down to the frame (which must be the phase wires touching the frame somehow and shorting out). I get good readings from the hall sensors and throttle and wiring seems intact. Probably the plastic bit over my anderson connectors slipped off while I messed around with everything. Clogging while spinning the wheel is similar to your results where anything that is coupled with the yellow wire seems to have strong clogging. Any updates or suggestions on how to tackle this thing?

 

[dogman dan]

Does it cog a lot when all the fat wires are disconnected from the controller, and they are not touching each other? If disconnecting removes the cogging, the short is in your controller.

Shouldn't have had any current at the controller side phase wires though, even if there was power stored in the caps. Phase wires should have been 0v unless the throttle was on.

On the battery input side, you can short those wires and get a fat spark, from the power stored in the controller. Disconnect the battery, then open the throttle, to drain caps before fooling with controllers. That works with functioning stuff.

 

[Subsa]

When you repair the controler, you should change the burned transistor (No1) and its parallel transistor. That is enought to work properly. No need to change all transistors, if they are not burned.
As i see from the pictures - first 2 transistors.
http://i.imgur.com/Vpsg6eI.jpg
You should measure the wires with Henry-meter to determine short circuit.

 

[gusty]

Does it cog a lot when all the fat wires are disconnected from the controller, and they are not touching each other? If disconnecting removes the cogging, the short is in your controller.

Shouldn't have had any current at the controller side phase wires though, even if there was power stored in the caps. Phase wires should have been 0v unless the throttle was on.

On the battery input side, you can short those wires and get a fat spark, from the power stored in the controller. Disconnect the battery, then open the throttle, to drain caps before fooling with controllers. That works with functioning stuff.

nope, no cogging at all with the wires disconnected from the controller and isolated. However, if I short the disconnected phase wires together, the hub shows a lot of resistance while spinning....

The battery input is all intact as I took extra care to make sure all the wires are well insulated while I hooked up switches and a pre charge resistor to it. However, my throttle signal wire was out of place and exposed. Also the motor didn't stop working all of a sudden it died out after whining for 10secs while I tested it on stand. Whining as though I had put a lot of load on it but nothing was on the bike except batteries... The reason why I tested it is I had transported the bike laying on its side the night before. I heard the spark before I tested the bike.... Any clues?

 

[dnmun]

that is normal for cogging when you short the phase wires together. if the mosfets are shorted in the controller then it will cause the same thing.

 

[gusty]

Yea man i just checked the phase wires from the controllers for continuity and the yellow wire beeped.. Its a fried mosfet I guess.. Just need to open up box to confirm..

 

[dnmun]

when you say checked, what does that mean? do you know what you are doing? you will need to use the diode tester function on the digital multimeter and then test each of the mosfets but you do that without opening the controller.

 

[gusty]

well to be honest not really.. just following instructions from an online post... I hooked up the red probe of my multimeter to the positive pin of the power input at the controller end and the negative probe on each of the phase wires. Two of the phase wires show 1 (infinite resistance I believe) while the yellow one showed 0.01. that shows a blown mosfet doesn't it?

 

[gusty]

when you say checked, what does that mean? do you know what you are doing? you will need to use the diode tester function on the digital multimeter and then test each of the mosfets but you do that without opening the controller.

yep yep... didn't open the controller... adn yes used the diode function =D

 

[dnmun]

yes. the drain leg of the mosfet is attached to the positive lead and the source to the phase wire so if it conducts when there is positive voltage on the drain then the mosfet is shorted. you should test the other three sets too. in that case you put the red lead on the phase wire and the black on the ground, black wire to the controller from the battery.