All hall sensors dying at once

Cowardlyduck

Moderator
Staff member
Joined
Jun 26, 2011
Messages
4,094
Location
Australia
Hi all,

Since noone seemed to notice the post in the Stealth owners thread, and it's not exactly Stealth specific Ill post this here also.

For background info this is on a HS4065. I've had intermittent hall sensor issues for a while now. About 2 months ago, they all died at once. I replaced them, not doing the best job of connecting them up (i.e. without solder and just using heat shrink), but it seemed to work at first. After a while, the intermittent shuddering issues came back. I continued to live with it since I could usually get normal function back by just messing around with the throttle and application of brake together.

Well, I had a hall sensor completely rip out of it's slot the other day whilst riding. It seems my epoxy job wasn't quite up to scratch on one of the sensors. Funnily enough, I was able to keep riding with a very slight shudder most of the way home even though I could see the sensor was loose through the side cover holes.
Anyway today I got around to cracking open the motor for the repairs.

I thought I'd take the opportunity to attempt (again) to solder the hall sensors on properly...like everyone suggested, but to my surprise all but one of the 9 heat shrunk sensor legs were on very tight. I think I'll still chuck a bit of solder on this time anyway though.

Now for the weird part...all 3 of my hall sensors are dead at the same time...again!

Anyone know what can cause that? Could a short somewhere between the pos/neg hall sensor wires cause them all to blow together?
Could it be a controller issue?

Cheers
 
maybe not the halls.. maybe the 5volt feed going in.

does the motor give any repsonse from throttle movement? If it does, then throttle is OK, so therefore the 5 volt is OK..but check the 5 volt feed at halls just in case. If 5 volt is ok at throttle and controller out, could just be a wiring issue of the feed to the halls. or if 5volt ok, could be the negative
 
Thanks NielP.
It's the halls, I tested them directly with a tester and they are all dead.

Throttle is still fine as the motor does move when applied.
Wiring issue was my first though as well, but in that case, I should not see all 3 halls dying at the same time should I?

That is what I can't figure out...why would all 3 hall sensors die together like that...twice?

Cheers
 
How do you know they're dead? Did you test them with a separate 5v feed and a magnet? Are you sure that it's not the 5v to them that's faulty?
 
d8veh said:
How do you know they're dead? Did you test them with a separate 5v feed and a magnet? Are you sure that it's not the 5v to them that's faulty?
Yes. I hooked up a separate ebike tester with it's own 5V feed directly to the legs of the sensors and waved a magnet past. They are all dead.
I have some new sensors ready to install and the same test on them shows them working.

Cheers
 
I wonder if the non soldered connections combined with vibration caused voltage spikes/ on 5 v which zapped them
 
Every time I killed all 3 sensors at once, I roasted them. Parking the bike when already pretty hot causes a large heat spike when you shut off the wind cooling. Not obviously so hot, but hot enough.
 
if he never soldered the legs it is more likely they are just coated with the hot melt glue inside the heat shrink tubing so they don't make contact any more. we will never know.
 
Thanks for all the comments. Appreciate the feedback.

NeilP said:
I wonder if the non soldered connections combined with vibration caused voltage spikes/ on 5 v which zapped them
Anything's possible, however the last time they all died was as they were from the factory, which were soldered on at that point so it's unlikely.
Also, this time around I had Honeywell SS411A hall sensors installed.
CF181144-01.jpg

http://australia.rs-online.com/web/p/hall-effect-sensor-ics/1811463/
So it's also unlikely a spike could have caused it since these can go up to 30V.

dnmun said:
can you detail how you tested the hall sensors to determine that they are dead? do you have a picture?
I used one of these testers.
$T2eC16Z,!)!E9s2fDPrhBR3oZD2-E!~~60_12.JPG


Can't see where I went wrong exactly...testing the legs of the sensors directly connected to the wires of the tester for motor hall sensors then waving a strong magnet passed. As mentioned the exact same test on my new replacement sensors was successful.

dogman said:
Every time I killed all 3 sensors at once, I roasted them. Parking the bike when already pretty hot causes a large heat spike when you shut off the wind cooling. Not obviously so hot, but hot enough.
This is still a possibility as it was still working somewhat at the point of disembarking from the bike, however both this time and last time all the sensors died was not after an overly strenuous ride. I have gotten the motor far hotter without issue in the past. Also, the hall sensors sit at the bottom of the stator, and my covers are vented, so I imagine it would take a massive amount of heat to cause a triple failure like that, but I guess it's still a possibility.

dnmun said:
if he never soldered the legs it is more likely they are just coated with the hot melt glue inside the heat shrink tubing so they don't make contact any more. we will never know.
Who ever said anything about hot glue?
What I did was insert the legs of the sensors inside the 24awg silicone wire sheath, then heat shrink over the top. I still think it's not such a bad idea, and will do it again, however with a small dab of solder next time around just to be sure.

Any other potential things that can cause all the sensors to die like that? Based on NielP's comments I can see a voltage spike would potentially cause this, so does that also mean a short could as well?
Can one faulty/failing hall sensor somehow effect the others and cause them to fail also?

Cheers
 
Bad luck..you keep receiving perfectly matched sets of three halls ..all with the same faults :lol:


Otherwise..really don't know...cumulative effects of heat over a period of time?
 
The times I had all three heat fail, " it was not that hot" . Yeah right.
 
you need solder and shrink.

likely an intermittant short to one of the phase wires. between the red or black wire to the halls and one on the pahses, either in the wiring or in the motor. also . if your winding is nicked near the hall legs, it could have "arced" over to the neg or pos hall wire. since all three blow at once , that is the common wires between them where i would start looking. also right at the wire entrance to the axle, did you ever spin your dropouts?

another place to look would be at connectors and inside the controller possibly.


halls can be a pain, do it once do it right, also learn to solder! :mrgreen: good luck
 
The most likely problem is that you didn't replace the hall wires when you replaced the halls, and a short in the bundle of 5 wires (probably inside the axle where they likely got hottest) is what caused the original and continuing failures.

Upgrade that phase wires to the max size you can get through the axle too, while you're at it. Then you'll have less worry about phase wires heating up and melting the insulation on the hall wires, especially inside the axle where they are bundled together the tightest.
 
I have been using this stuff
30 gauge = 0.8 mm outer diameter. PTFE coated
http://export.rsdelivers.com/product/alpha-wire/5851-3-rd005/red-hook-up-equipment-wire-30awg/1770637.aspx
 
I've been flat out, so only just getting around to trying to fix this now.
Using my multimeter I tested for continuity between each of the phase wires and the stator metal and axle. Each phase has continuity to the motor structure itself. Is this normal? I'm guessing not.

I think I'm going to need to replace all my motor wires; hall sensors and phase wires. I should really do it anyway since I've been meaning to for a while now.

Thanks for the info on the 30 gauge wire NeilP, that sounds like what I need for the halls. :)

What's the largest size phase wires most people manage to fit through the HS series axle? I will go with silicone wire, since I like it's properties a lot more.
I'm thinking 10 gauge would be the easiest and largest to manage.

Cheers
 
dnmun said:
the phase wires should not be continuous to the stator. they will be continuous to each other through the WYE.
Dam, that's what I expected...at least it will force me to replace the phase wires properly this time.

Nothing changed since the motor was last working...I still haven't even removed the side cover with the wires going through it.
So it makes me wonder if the phases have been shorting to the stator for a while.
Would the motor even work with the phases shorted to the stator?

Cheers
 
Should be no continuity to motor structure through phase wires, so that is one issue. But that in itself would not cause halls to fail. So guessing big melt down has melted hall insulation somewhere.

But if that has happened to phase wires,is controller still good or are has that been fried too?

Trouble with most silicon wire is the insulation is soo thick. With 10 gauge you probably won't fit two phase wires. Look at the 30 gauge I linked too and see if they do 10 gauge in the same series if wire. That is very tough but thin walled.

If you have not yet opened a motor, look at my video

http://endless-sphere.com/forums/viewtopic.php?f=2&t=55184


Just found this reply, un posted, on the iPhone. Looks like I hit preview earlier, not SUBMIT
 
NeilP said:
Should be no continuity to motor structure through phase wires, so that is one issue. But that in itself would not cause halls to fail. So guessing big melt down has melted hall insulation somewhere.

But if that has happened to phase wires,is controller still good or are has that been fried too?

Trouble with most silicon wire is the insulation is soo thick. With 10 gauge you probably won't fit two phase wires. Look at the 30 gauge I linked too and see if they do 10 gauge in the same series if wire. That is very tough but thin walled.

If you have not yet opened a motor, look at my video

http://endless-sphere.com/forums/viewtopic.php?f=2&t=55184


Just found this reply, un posted, on the iPhone. Looks like I hit preview earlier, not SUBMIT
Thanks for confirming NeilP.
I think the controller is still good as it was 'working' right up until the point the hall sensor ripped out of it's slot.
Ok, so 10 gauge silicone is out. What about 12 gauge?
Can't seem to find any wire like the one you linked in a larger gauge. What have others used...guess I should search. :)

I'm currently at 175A phase current...is that too much for 12 gauge?

Cheers
 
NeilP said:
Should be no continuity to motor structure through phase wires, so that is one issue. But that in itself would not cause halls to fail. So guessing big melt down has melted hall insulation somewhere.

But if that has happened to phase wires,is controller still good or are has that been fried too?

Trouble with most silicon wire is the insulation is soo thick. With 10 gauge you probably won't fit two phase wires. Look at the 30 gauge I linked too and see if they do 10 gauge in the same series if wire. That is very tough but thin walled.

If you have not yet opened a motor, look at my video

http://endless-sphere.com/forums/viewtopic.php?f=2&t=55184


Just found this reply, un posted, on the iPhone. Looks like I hit preview earlier, not SUBMIT
Thanks for confirming NeilP.
I did see your video, and it was perfect timing as I saw it right before I needed to pull apart the motor. :D

I think the controller is still good as it was 'working' right up until the point the hall sensor ripped out of it's slot.

So if 10 gauge silicone is out, what about 12 gauge?

I can't seem to find any wire like the one you linked in a larger gauge on that site. What wire have others used though...I was sure I had seen others had used silicone wire...guess I should search. :)

I'm currently at 175A phase current...is that too much for 12 gauge?

Cheers
 
you could try Kenny at crystalyte

That is where I got my last lot of wire from..Like the original PTFE insulated, thin walled, and near enough 10 gauge I believe...fractionally under...3.166 mm diameter conductor

depends on which table you believe.

http://www.bulkwire.com/wiregauge.asp

http://www.rbeelectronics.com/wtable.htm

These are is pictures, he e-mailed to me

Crystalyte crystalyte@yahoo.cn

Kenny kenny@crystalyte.com

Ask for
kenny said:
Hi Neil,
Nice to hear you get the order.
The wire called less resistance 4 mm2 ( square ) for motor.
But it will make think simple if we called it for Crystalyte X5 motor's phase. So we will understand.
Kind Regards
Kenny
 

Attachments

  • 4 square mm2 wire.jpg
    4 square mm2 wire.jpg
    48.4 KB · Views: 3,035
  • 4 square mm2 wire 316.jpg
    4 square mm2 wire 316.jpg
    44.3 KB · Views: 3,035
Lol, that's the wire I've already got on there...if that's really the best stuff, I might just reuse it for the section through the axle and join it to some 10g silicone wire outside the axle as many others have done.

Now I just gotta figure out the best way to redo the shrink-wrap section through the axle and put it back in without shredding it.

Any tips on that?

Cheers
 
Just check the size you have...I think it is possible you may have a smaller gauge. A mate has an Xlyte motor..modern one like yours, and it has wire that look identical , but it is one size smaller.



I run my bike with this phase wire at 84 or 100 volts..with battery current peaking at 140 amps..14kW..phase current i think set to 270 amps max..not had an issue in 9000 miles.

BUT it may not be a melting issue,,,I believe some of these newer series motors had issues with dmaging the cable on entry to the axle..or washers/ spacer ares..or something..anyway..check for damage rather than melting..this stuff does not melt...well I am sure it does , but I guess if you had got it that hot, then the windings are likely to be toasted too :shock:



err..what am i saying..you have a short, so cetain to be damage somewhere.

As for tips on threading it through..??? grease/ vaseline? Home made bigger axle?

like mine

file.php
 
Cowardlyduck said:
dnmun said:
the phase wires should not be continuous to the stator. they will be continuous to each other through the WYE.
Dam, that's what I expected...at least it will force me to replace the phase wires properly this time.

Nothing changed since the motor was last working...I still haven't even removed the side cover with the wires going through it.
So it makes me wonder if the phases have been shorting to the stator for a while.
Would the motor even work with the phases shorted to the stator?

Cheers

i don't hink you understand. the phase wires are shorted to the stator in the windings. at 175A the tiny wires overheated and burned the varnish off that keeps them from contcting the stator and it must have melted the plastic shields they put on the stator to prevent the wires from contacting the stator.

you can establish this when, if you do, you open the motor. unsolder the phase wires and check continuity of the windings themselves to verify.
 
Back
Top