All hall sensors dying at once

[NeilP]

Have to admit, I only ever pushed the 5304, not one of these newer motors, but 175 is the phase current he is talking about, not battery current..and we are not even sure if that is a real amount, or merely the controller setting.

Until it gets opened up to check to see if it got toasted, it could still just be a shorted, broken phase wire inside or at the entry point of the axle. Any one single phase wire broken and shorted to the case will show continuity to the stator and case on any phase, since they are all connected at the centre junction of the WYE

 

[Cowardlyduck]

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.

Well I did understand...I just didn't think it would be the case.
I just cut the phase wires inside the motor...retested, and sure enough they have shorted to the stator. DAM DAM DAM

I don't understand how it could have happened though...the motor was perfectly fine running 175 phase amps previously, and in fact during the last ride where it failed, I was only pushing 50 battery amps so phase amps couldn't have been that high, or does it not matter what battery amps I'm pushing if the phase is still allowed up to 175A?

My other question was, can the motor still even operate with phases shorted to the stator? I mean, the motor was operating normally right up until the point the hall sensor ripped out of it's slot. It even continued to work albeit with a shudder after that point, and I didn't push it hard.
So if the windings have in fact burned through, it must have just been a massive coincidence that it happened at the same time the hall sensor ripped out of it's slot...hence my skepticism and question about the motor still working with a short.

Cheers

 

[Cowardlyduck]

Have to admit, I only ever pushed the 5304, not one of these newer motors, but 175 is the phase current he is talking about, not battery current..and we are not even sure if that is a real amount, or merely the controller setting.

Until it gets opened up to check to see if it got toasted, it could still just be a shorted, broken phase wire inside or at the entry point of the axle. Any one single phase wire broken and shorted to the case will show continuity to the stator and case on any phase, since they are all connected at the centre junction of the WYE

Just saw this after I posted.
Yes it was only phase current as mentioned, but apparently still too much for a HS4065.
I'm thinking I might get a HS4040 to replace it, since that should not overheat as easily, and will also give more torque. I don't really need the top speed, as fun as it was.
I might also try adding some cells to my pack for 21S - 77.7V nominal, which will also give back some of the lost speed. That will also decrease the amount of amps I need. Win Win.
I'm assuming it isn't a problem to increase voltage to that level on one of these motors?

Cheers

 

[NeilP]

Well more voltage is still more power..and power is heat, so there are limits..but 20s 84 volts is going to be fine.

maybe get yourself a CA V3 with temp limiting..or at very least fit a temp sensor on the re build.

I'd like to get a toasted motor..not toast my own you understand... just to play with winding my own...look at this as an opportunity to learn how to re wind your stator

if you have opened motor and cut wires inside the case..stator side of axle and short still there, well yes, windings are probably toasted..show us pictures.

motors are never voltage limited, (within reason) ..it is always the controller components that are limiting you..The Fets, the capacitors and voltage regulator side that produces the 5volt and drive voltage for the mcu/switching etc. Check your controller and se that it has 4110 FETs and 100volt caps..if it has 60 volt caps and 3077 fets than 70 volts is no go.

 

[Cowardlyduck]

Thanks again NeilP.
I'm extremely time poor, so no go on a re-wind, although I'd love to try.
The last motor I wrecked (HS3548) I sold dirt cheap, and will probably do the same with this one. If you were geographically closer I'd probably sell it to you.

I have a beefed up EB3 18fet controller so no worries there.

CAv3 is one of the many planned upgrades for my Fighter...just gotta get it running smoothly for a while first.

Again, time poor, but will try and get pics if possible.

Cheers

 

[jateureka]

Did all the windings look toasted?
The pic I posted here shows a little Bafang with totally fried windings but that too was still working under a light load http://endless-sphere.com/forums/viewtopic.php?f=3&t=54586&p=820131#p820131

 

[Cowardlyduck]

Did all the windings look toasted?

Impossible to tell since I spray painted the entire hub with a black silicone based high temp spray paint to protect it when I vented the side covers.

I would scrape the paint off to look, but there's a remote chance the hub might still work for someone when I sell it.

It's filthy I know...but that's what happens when you vent a hub in a dusty climate like Canberra.

The pic I posted here shows a little Bafang with totally fried windings but that too was still working under a light load

I think there is a remote chance the motor could still work, however the windings are shorted to the stator, so plugging it into a controller would risk blowing it, and I don't want to risk that.
I've ordered another motor now anyway, so will just sell this one cheap and forget about it...it's been too much hassle and I just sick of screwing around with it tbh. I won't push 6KW through the next one that's for sure.

Cheers

 

[dnmun]

why not cut it apart at the wye, then separate the wires in the wye so the phases do not touch and then take a picture of the ohmeter as you measure from each phase wire to the steel in a spot where the rust is cleaned off so you get good contact. it would be useful to know if just one of the phases is shorted or all three.

 

[Cowardlyduck]

why not cut it apart at the wye, then separate the wires in the wye so the phases do not touch and then take a picture of the ohmeter as you measure from each phase wire to the steel in a spot where the rust is cleaned off so you get good contact. it would be useful to know if just one of the phases is shorted or all three.

I might do that if I find the time...but then again, what would that prove?
If only one phase is shorted, there is still a phase that is shorted...isn't that bad either way?

Cheers

 

[Cowardlyduck]

So thanks to dnmun, curiosity got the better of me and I tested to see which phase was shorted.

Turns out it's just the green phase that shorted to the stator. Does that help my situation at all though...I mean, there may as well be all 3 phases shorted if one of them is since either way nothing can be done without a rewind.
If only one phase was shorted, is it possible my motor has been like this for some time?
What I mean is, can a motor with just one phase shorted to the stator still work anyway? That would explain why the halls kept having intermittent issues, and why they all died at once.

I'm thinking I might hold onto the motor, and do a rewind at some stage. What does the average rewind end up costing though? I know that it takes a lot of time and patience, but if it also costs more than a few hundred $$$ in parts, then I don't think it would really be worth it for me.

Cheers

 

[dnmun]

it is good to have the info out there for others to compare their measurements to. it does seem like it should work if only one phase is shorted if that phase is electrically isolated from the controller. in other words if the stator doesn't make electrical contact to the circuits in the controller besides that one phase it should work. maybe.

if you can separate the individual wires in your green phase, then test each one it might be possible to find just one or two wires shorted so you could isolate them from the other wires in that phase at each end, then isolate an equal number in the other phases so they have equivalent resistance, and put the motor back together at the wye and each phase and then see if it will run again. see how much power is lost. it could be educational.

maybe you would not even have to isolate wires in the other phases. it might not matter. the motor people may have an opinion.

 

[NeilP]

Now that is a good idea, worth a try at least. Split the bundle and test individual strands.

But it could well have been like that for a long time. Yes it will still work, and it should not be a cause for the Halls to get fried..
Hell it could even have been like that since new.

As long as the controller is not earthed to the frame and that is the only short to chassis it is not going to cause an issue. Thinking about it..don't even bother separating the cores

..get a sensorless controller for it, just a little 6 FET, rebuild motor ..use its drilled side panels on your new motor, and put new side panels on this on ..then you will have a spare 'kit' ready to put on any small girlfriend bike..local shops runabout.

4 part kit...throttle, controller, motor, battery..no complications ....no CA/ Brake levers switches etc.


It is rather like me. it could well have been an existing condition that investigation for another problem has revealed..
I broke my left arm in August...pinned with a k-wire...but weird..thumb not working..seemed odd in the plaster..anyway after plaster was taken off...further investigation..
tendon was damaged..so operate on that ...tendon graft .... Few days after op..my leg went dumb..very odd..potential for blood clot after op..
so back off to hospital to get checked out...MRI scan..turns out I have a prolapsed disk in lower back and being sat on my backside continuously for 4 days on hard chair at home on the computer has just bruised a nerve which caused the dead foot.. The bad back issue / prolapsed disk...probably been like that for years since my farming days..would never had known if I had not broken my arm.

 

[Cowardlyduck]

Thanks for the replies guys, much appreciated.

if you can separate the individual wires in your green phase, then test each one it might be possible to find just one or two wires shorted so you could isolate them from the other wires in that phase at each end

Thanks for the tip. I tested for that just now, and sure enough just 2 of the strands are shorted.

maybe you would not even have to isolate wires in the other phases. it might not matter. the motor people may have an opinion.

I might try to resolder and reassemble everything over my Christmas holiday's, so it would be good to know this. Does anyone know?

Hell it could even have been like that since new.

I actually think it might have been...but it would be impossible to prove that.

..get a sensorless controller for it, just a little 6 FET, rebuild motor ..use its drilled side panels on your new motor, and put new side panels on this on ..then you will have a spare 'kit' ready to put on any small girlfriend bike..local shops runabout.

4 part kit...throttle, controller, motor, battery..no complications ....no CA/ Brake levers switches etc.

It's funny you should mention that. I've already got a beefed up 6 FET (sensored), throttle and spare battery ready to go...was just waiting on a motor and a battery bag. Only problem then is none of my existing bike frames will accept a disc brake setup. Time to start looking for a cheap FS frame I think.

I'll try and post progress of the repair as I go...if it's anything like my hall sensor repairs, I'll probably do something stupid in the process.

Cheers

 

[NeilP]

Join the club!!

maybe you could look at turning your own axle ..make thicker one...and feed water cooling in from the other side..Like Methods did

 

[Rodney64]

Maybe worth trying a Sensorless controller as Dunn mentioned. My current setups Sensorless with my 4080 running 100volt. No need to worry about hall wires. Im also using some old cable that I got from work for my phase wires. The crystalyte wires just no good enough to cope with the higher voltage and as soon as it melts any of the 3 phase wires together and they short it's either the controller blowing or damage to the motor. The wires General Electric cable 600/1000v rated to 90degrees and each strand 2.5 mm so far no issues with over heating. Today was 37 degrees and the 15 kilometers that I travelled it held up well. Averaged 30kms an hour.

 

[Rodney64]

Maybe worth trying a Sensorless controller as Dunn mentioned. My current setups Sensorless with my 4080 running 100volt. No need to worry about hall wires. Im also using some old cable that I got from work for my phase wires. The crystalyte wires just no good enough to cope with the higher voltage and as soon as it melts any of the 3 phase wires together and they short it's either the controller blowing or damage to the motor. The wires General Electric cable 600/1000v rated to 90degrees and each strand 2.5 mm so far no issues with over heating. Today was 37 degrees and the 15 kilometers that I travelled it held up well. Averaged 30kms an hour.

 

[brumbrum]

lets cut to the chase... Halls + off roading = failure/ i.e snapping/dislodging, overheating etc..

I hate hall wires

get a sensorless controller. you will loose initial 'oomph' in accerleration but you will have a much more reliable bike for bumpy ground and for off-roading.
Give yourself more room for thick phase wires and do away with hall wires.

Just my opinion for what it is worth. :?

Sorry i cannot add anything more technical.

 

[brumbrum]

double post...... :x

 

[dnmun]

if you had the phase wire shorted to the stator and the hall sensor wires melted enuff to short out to the stator then the phase voltage would have fried the hall sensors and not the heat which is not usually what kills the hall sensors. usually it is a short to the phase wire that kills the hall sensors. this happens in a twist off all the time.

if all three hall sensors were shorted then all three would die. measure continuity from the hall sensor signal wire to the stator to check.

you may have moved them enuff that they are not shorted but if the phase wires got that hot then the insulation on the hall sensor leads would have melted inside the axle imo.

but if you could fix that with new hall sensor leads and then just leave out the two shorted phase windings' wires, then since you still know the order of the phase wires it should run with the same colors on the new halls too.

might be worthwhile working on since your bike is built this way already.

thanks for doing that. good mental exercise.

 

[Cowardlyduck]

Join the club!!

maybe you could look at turning your own axle ..make thicker one...and feed water cooling in from the other side..Like Methods did

:lol: Glad I'm not the only one.
Would love to turn my own axle, but I have no gear to do so. If the motor still works when I put it back together, I'll just leave it as is and push less power through it next time.

Maybe worth trying a Sensorless controller as Dunn mentioned.

I would, had I not already bought a sensored 6 FET a while back. It's no drama's anyway since my plans for re purposing this motor won't involve as much off roading anyway.

I hate hall wires

Don't worry, I'm beginning to as well. That being said, I do like being able to get moving from a slow start and/or uphill. I do plenty of slow technical riding on my Fighter. I think without halls that kind of riding would be somewhat painful.

if you had the phase wire shorted to the stator and the hall sensor wires melted enuff to short out to the stator then the phase voltage would have fried the hall sensors and not the heat which is not usually what kills the hall sensors. usually it is a short to the phase wire that kills the hall sensors.

I'm thinking this also. I can also see the potential for how even if there is no wire contact, the fact that the stator was sitting at 75V+ (Hot off the charger) would probably be enough to jump a spark to the legs of the hall sensors sometimes. That would also explain the intermittent hall sensor issues I saw throughout that motor's life.

Thanks again guy's, were it not for this forum, the motor would probably be in the trash right now.

Cheers

 

[Rodney64]

Which wire are u going to try with the phase wires.

 

[Cowardlyduck]

Which wire are u going to try with the phase wires.

What do you mean exactly?

 

[Rodney64]

Which wire are u going to try with the phase wires.

What do you mean exactly?

I'm not using the crystalyte wire. Just doesn't take the heat that 24S produces. The wire I'm using at the moment is General Electric cable 600/1000v rated to 90degrees and each strand 2.5 mm so far no issues with over heating.

 

[Cowardlyduck]

I'm not using the crystalyte wire. Just doesn't take the heat that 24S produces. The wire I'm using at the moment is General Electric cable 600/1000v rated to 90degrees and each strand 2.5 mm so far no issues with over heating.

I see. I will probably just reuse the wire I've already got for the section through the axle since it's a good fit. Since I've cut the wires I will need to extend them now, so might use some General Electric cable for the extension.
I'm not too fussed really since if I get this motor back up and running I won't be able to push much power through it anyway due to the shorted strands on the green phase.

Cheers

 

[Cowardlyduck]

Gaaaaaaaar...this is me right now!

So it turns out that not only was my motor phase shorted, but my controller is also to blame for the hall sensor failures.
My new motor arrived yesterday (was a different wind than I was expecting but that's a whole other story), and after 3 hours sorting out the disc brake, freewheel cluster and rear tire, I got it on the bike only to find it shuddered...then it worked fine...then it shuddered.
Then it worked and I managed to get in a test ride and went faster than I've ever been before on this bike...now it's shuddering again.

So anyway, I'm pretty sure the controller is at fault...now I'm just hoping it's only a loose wire, but I have a feeling it's not. I did test it, just to be sure and when applying throttle this is what I get on a tester.

It stay's like that and doesn't rotate. I recall it's supposed to rotate to simulate a motor turning correct?

So does anyone know if, apart from loose wires/bad contacts, can a controller screw up it's hall sensor readings? What can cause it besides loose contacts?

Cheers