Hall sensor problem?

[Memran]

Last week I got cought out in a downpour. Me and the bike got absolutely soaked through.

Today, I wanted to take the bike out for a spin, but it just felt wrong. Very jerky, noisy, and didn't feel as powerful as it used to. At slower speeds, I can feel the motor pulsing.

So I suspect there is something wrong with the hall sensors, or that there is water trapped inside the hub.
The 3 signal wires are sort of working:
1) 0v - 12v -0v , seems ok
2) 0v - 10v - 12v - 0v , dunno, kind of weird
3) 0v - 12v - 10v - 0v , also kind of weird

Would I be correct in thinking that if everything was working correctly that all 3 halls should be acting like number 1? My reseatch so far seems to say that when halls are faulty they read 0v - 6v or something like that... so is 0v - 10v ok?

I probably should add that the motor is a rear Crystalyte 408, and the controller is a 20A Crystalyte Journey. Battery is 26v, 9Ah, Li-Ion.

 

[Grizzlybear]

Hi
those voltages look a bit high to me, I would expect 0v 5v 0v, I recently had a bafang motor do simular things. Are you using the correct voltage controller? because the resister R1 in the controller(infineon) steps down the battery voltage to 5 volts for the control circuit. The fact that you are seeing voltage then no voltage as you turn the wheel indicates that the halls are actually working.

 

[HighTekBikes]

Check your throttle output voltage, make sure it is steady when on, not jumping around. Although most motor kits use 5V for the sensors, some controllers might use 12V. Check the voltage going to the sensors from the controller. Open the controller and look for moisture and dry it out with hair dryer.

 

[dogman dan]

We are assuming, of course, that you already dried out the hall sensor connector plugs? The same symptoms could happen from wet, or otherwise bad phase connections too. So check and dry everything, and look for the one that got me, phase wires rubbed on a tire.

 

[Memran]

Thanks!
Yes, my controller's sensor wires run at 12v.
I have not opened the hub motor to peek inside yet, but I did bring the bike inside te house, instead of leaving it in the garage, so should be a little drier! Somehoe I doubt that will make much difference.
All cables are visibly intact, so I guess the next thing is to open the hub and check for wetness inside.
Is there any online guid to opening these things up? Looks simple enough, but a guide would be handy

Thanks

 

[mwkeefer]

Memran,

I would suspect the controller before the motor.. That's an analog controller if I am not mistaken.

For the hall readings... based on the way you display the leading and trailing edge of detection (10v position) I would say that the halls are positioned internally:

#2 is actually first in rotational order, likely phase A for the controller (Yellow on most, not sure on yours)
#1 Is most likely in the center and unless I am mistaken, you should have seen a leading and trailing edge of 10v on that one
#3 is last in rotation.

Some of those controllers (actually it seems about 1/2 of all digital and analog have this) have a learning function or mode selection method... where you short two wires and keep momentarially shorting them until the wheel spins smoothly and forward, it is possible that got moisture and shorted... causing your controller to jump to a mode of operation not compatible with your motor (60vs 120 degrees, hall sensor sequence, pole count, etc)...

I am assuming the motor will spin freely with the phase power wires not connected - ruling any sort of actual coil shorts out...
I am also assuming (as you seem to be) that the downpour and subsequent malfunction are not just coincidental happenstance

Do you by chance have another controller to test with (plug and play) or perhaps a buddy who could help ya test it using his controller?

Hope it helps!

Regards,
Mike

 

[Memran]

That makes a lot of sense about the hall sensor position and order.

The controller, as far as I am aware is not programable, but I could be wrong on this. I have no manual or anything for it. I'll have to doa little research on this.
I do have a friend with an identical motor and controller, so I'll try to run with his controller and see what happens.

Thanks for info

 

[Memran]

I've taken everything off the bike, and tried to let everything dry.
Inside the controller there is no visible sign of moisture, and also I can't see any jumper to short out for mode selection/programming.

While testing I've noticed that if I hold the wheel in certain positions and use the throttle, the wheel does not want to spin. It makes a klunky noise, and a whine which goes up in pitch if I apply more throttle. If I move the wheel slightly, it will then spin up, but still roughly.

I also tried to open the hub to see if there was any water damage inside, but was unable to pull it apart. I think the magnets are holding it, and I will need a pulling tool of some sort to get inside.

 

[mwkeefer]

Okay... well holding the wheel at a given angle and different results, thats kinda a give away...

1.) Brake of conductor in the hall wires outside the motor
2.) Loose or broken hall connector inside the motor

Can't be anything else..

If it's #1, cutting the wiring at about 4-6" out from the motor and testing there should determine if the issue is internal...

I could be mistaken but after removing the hex bolts on the cover, you simply need to work around with a putty knife and work it up... no puller required.

Regards,
Mike

 

[dogman dan]

Yes, you should be able to get the cover off. The other method some use is to put a block of wood on the floor, and rap the axle end down on it, to pop the cover up enough to continue to carefully pry off the cover. Don't stick tools in too far, and pry on motor windings. It could be just wet inside still.

 

[Memran]

This is confusing me.
I can't understand how there can be a break in the hall wires, because I tested the voltages inside the controller (with phase wires disconnected).

I'll have a go at the block of wood method when I get home from work.
Oh, which side do I actually want to get off? drive side or disk break side?

Thanks

 

[mwkeefer]

Sorry it was late and I wasn't clear...

If you have a wire break inside the insulation tubing, I have seen before where connections become intermittent at best. Unless your wiring was being held absolutely still while you moved the motor and acheived different results in various positions it could indicate that the wires for one of the halls is damaged in the actual cable (I've seen it in about 1/2 of the cases) and normally that's caused by Zip ties being too tight (odd as it sounds) but it leaves the wires so sometimes they touch and somtimes they don't... = Intermittent issues related to position (bent one way the wiring will work as the broken wire tips will touch).

With regard to connectoins in the motor.. I've seen halls with shrink tube on each leg, where the connection (soldered) broke for one reason or another and yet the shrink managed to keep it connected - most of the time, leading to similar issues as you seem to have experienced.

All that said - Dogman is probably right, it's prb just a little bit of water collected inside and yes knocking the opposite axle against a wood block while holding the hub part of the motor loosly in yuor hands (not the rim) will give you a beginning on removing the covers

-Mike

 

[scriewy]

Okay... well holding the wheel at a given angle and different results, thats kinda a give away...

1.) Brake of conductor in the hall wires outside the motor
2.) Loose or broken hall connector inside the motor

Can't be anything else..

If it's #1, cutting the wiring at about 4-6" out from the motor and testing there should determine if the issue is internal...

I could be mistaken but after removing the hex bolts on the cover, you simply need to work around with a putty knife and work it up... no puller required.

Regards,
Mike

this is a give away ? it's not the zero crossing or something ?
or sensored motors when functioning properly can start spining from every position to the milimeter ?
because in my motor it's anough to turn the motor few milimeters and there's a reaction (starts spinning) or wont, or i can simply hear a beee.... sound from the controller when slowly turning the motor and every few milimiteres it's making the sound or geting silent.

 

[Memran]

[EDIT] Not really success... it failed again [/EDIT]

Success!

When opened up the hub water ran out of it. The sensors looked fine, and there wasn't much sign of corrosion, so I dried it out with a hair dryer, and left it open for a few days, before reassembling.

First test on my bench seemed to be ok, so I chucked it back onto the bike and gave it a proper test ride. All is well!

Thanks for your help guys

 

[HighTekBikes]

Intersting. The water must have been interferring with the hall sensors, throwing off the timing. You would have had to check the outputs with an oscillocope to check the rise and fall times and any irregular pulse shape.

 

[Memran]

Continuing problems

OK so I though it was resolved but it is not.
I have used the bike for a few days to commute to work and back and it seemed to run just fine. Then this morning, after a weekend rest, as soon as I turned it on I knew something was wrong :/

With power to the controller, but no throttle, when pushing the bike I can feel the wheel 'cogging', as if there is one coil powered. When I apply throttle to spin the wheel, it is extremely rough, and I can feel the same kind of symptoms as when it had water in it.

Frustrated with the whole thing right now

 

[Memran]

Ok some more investigating today revealed that I have a new hall sensor pattern!

1) 0-12-0
2) 0-12-0
3) 12

I also double checked that I have continuity on the phase wires between motor and the plug that connects to the controller. Seems ok.

In certain positions the wheel won't go, and in others it does, but jerky.

I guess I have to open the motor again!

 

[mwkeefer]

With all the power wires removed from the controller (disconnected) does the wheel spin freely or still cogging effects?

Your having a lot of issues with this - it seems to have krept up all the sudden a little while back, any idea what changed?

-Mike

 

[Memran]

With the power from the battery disconnected and the phase wires connected there is no cogging, and the wheel spins free.
With the phase wires disconnected the wheel also spins free.

I dunno, maybe there is still water inside? Worked ok for a few days then after a weekend break it stopped again. Perhaps having been left standing, the little bit of water still inside was able to trickle down and pool at the bottom, on a hall? Just guessing... Could it really just be water causing all these problems?

I'll open it up tomorrow after work. No time till then

 

[mwkeefer]

If it spins freely without the phase wires connected -- then I doubt it's water in your hub.. It's also not internally shorting the phases (good thing)

I'd be looking at the wiring loom between the controller and the motor - if that seems ok, i'd check the controller internally to ensure the phase wires are soldered properly.

-Mike

 

[Memran]

I found a super easy way to open the motor up! The trusty gear puller! ;D

Ok so I opened up the hub again, and checked all wires from inside controller to inside motor, and all checkout.

Then I started testing the hall sensors again but with a magnet, and multimeter.
Two of the halls returned expected outputs when I toggled the direction of my magnet. The 3rd did not, but instead the voltage on its output gradually faded away to nothing, and now no longer reads anything at all! I thought the bettery was flat, but no, the 1st and 2nd halls still read as expected.

At this point I'm pretty sure the hall sensor is to blame, so I dug it out of the epoxy, ready to replace it as soon as I find where I left my spares!

Hopefully this will solve the problem once and for all!

 

[Memran]

I finished replacing the hall sensor today, and the set up the wheel in my vice and connected it all up. It spun up smoothly on the throttle, and also spins freely without throttle once again.

Could this finally be the end of the problems? Here's hoping!

Thanks for all the input guys, much appreciated!

 

[HighTekBikes]

Did it look like the sensor was shorting out? Can you test the one you removed?

 

[Memran]

No it wasn't shorting out. Each of the three legs was inside heatshrink, which I had already removed in the photo.

Pretty sure, it was faulty, but havn't tested it off of the motor yet. I'll build a test circuit after I have reassembled the bike tomorrow.

 

[Bubba]

No it wasn't shorting out. Each of the three legs was inside heatshrink, which I had already removed in the photo.

Pretty sure, it was faulty, but havn't tested it off of the motor yet. I'll build a test circuit after I have reassembled the bike tomorrow.

Yes, I know this thread is a year old, but I thought I would add my experience to this issue, for others searching for a solution.

I am having a similiar problem, but when checking the hall sensor (on the yellow wire), I found that it had failed in the closed position. The other sensors, when measured at the controller, were showing voltages of 13.5 to zero to 13.5, etc. This is with the Journey 36v20a controller and 408 motor. The bad sensor showed a constant voltage of about 9v while turning the wheel. It wasn't toggling like the other sensors. Wires looked fine while wringing them all out. No continuity problems. Also checked (with power disconnected) continuity from the yellow wire to black, then red. Closed circuit no matter what the wheel position. Bad sensor for sure!

I'm currently waiting for a new one to arrive from Canada. I'll post more if it doesn't work.

EDIT: Forgot to add one thing. Thanks for the great pic of the gear puller! I used a similiar one to take the motor apart. One word of warning to all:

DO NOT put your fingertips in between the stator and rotor (hub)!!!!! These are some very strong magnets!! I had a brain fart during this process and now I'm expecting to lose a fingernail or two because of it. Just wait until the two parts fall apart and then KEEP them apart until it's time to reassemble! At that time, wear heavy gloves and be very careful.