BionX Stuttering / Dead hall effect sensor?

Hello Everyone

I have successfully replaced the bearings in my BionX 250HT motor two weeks ago. After this I've done about 200km on it without any issues, it was running perfectly smooth, so I don't think that it is related.

Anyway, yesterday - I'm not 100% sure, but I think while braking (recup). The thing suddenly started to stutter. When I pedaled again, it stuttered even more. The support by the motor went on and off. I have switched to PAS level 1 which was a lot smoother, but I guess it is due to less power involved.

The speedo showed incorrect values, and when I set PAS to 0, the speedo showed 0kph.

So my assumption is at least one of the hall sensors died, but I wonder what you guys think?
Does anyone know what kind of hall sensors I need to buy? I'd like to order them ASAP, before I have the time to disassemble the motor again.

Cheers from Switzerland
Nico

The first thing I suggest checking is the phase and halls connectors as well as possible. Unplug it all, look for bent or loose pins, corrosion and so on.

If that didn't solve it, have you tried contacting where you purchased the motor from about purchasing replacement halls sensors?

Hi bowlofsalad

Thanks for the suggestions.
The BionX system has the controller built into the motor, so it's not possible to actually check this. There are almost no replacement parts officially available.

Cheers, Nico

nioko said:

Hi bowlofsalad

Thanks for the suggestions.
The BionX system has the controller built into the motor, so it's not possible to actually check this. There are almost no replacement parts officially available.

Cheers, Nico

Click to expand...

why can't you open it? it just sounds like bad connections if it was working up to that point. most of the parts can be replaced with other sources. if one of the hall sensors has a bad connection then that could account for it but you may have a bad connection on something else like the torque sensor.

Of course I can open it, I just didn't have the time. In fact I have opened it two weeks ago to replace the bearings, but I don't have the necessary tools, that's why I haven't reopened it.

What I wanted to find out is:
Does my assumption make sense, that it's most likely the halls? Maybe I'm sure that I'm not the first one who has these symptoms on a BionX.
If so, what hall effect sensors should I order? If I could order them now, I might have them when I have the time to open the motor.

there is no evidence the hall sensors have failed so why would you replace them?

Well, something is wrong with the system, it's stuttering, and the speedo is not working anymore. I'm pretty sure that the BionX speedo is based on the halls, so I assume it's the halls, but I don't know for sure.
That's why I opened this thread, and asked about what halls, and if my assumption makes sense.

but you have not tested the halls so it is impossible to say that they are bad.

Kiriakos GR said:

Any ambitious ebike repair man should have one of these.

http://www.ebay.com/sch/m.html?_odk...rushless+Motor+Controller+Tester+New&_sacat=0

Click to expand...

Poor advice. How is this supposed to test an integrated motor/controller? By the time you disassemble the hub in order to connect a "tester" you can perform needed tests with equipment most people already have laying around.

Any ebike repairman worth his salt doesn't need more than a DVM, few clip leads, 10k resistor and 5-8V battery to test any and every function of a controller, motor, Hall sensors, throttle, etc.

Troubleshooting and test guides found here:

http://www.ebikes.ca/learn/troubleshooting.html

Haha! Shows how much I know about bionx motors, integrated controllers.

Oh, that is an interesting guide. http://www.ebikes.ca/documents/HallSensorTestingFinal.pdf

Seems like a good next step to try and duplicate.

If bionx isn't willing to sell you an individual hall sensor for you to replace yourself, you might be able to find out which one to match it with (I don't even know if this is necessary) by looking at whatever letters and numbers you find on the hall sensor. Keep us posted on your progress and questions!

Hello everyone, I appreciate the help. Yes it's a different system, and I know many moved away from it because of the high costs, and lets say complicated service by BionX. I for myself am not a huge fan of it, but I do like the torque sensing in their system + you can get these systems used in Switzerland, most likely with a worn Battery, that I'm able to rebuild, which makes it interesting for me.


Anyway I don't even need to try getting parts through BionX as I know the answer will be an offer for a replacement motor.

As soon as possiblr, I will disassemble it, and then I will find out about the halls. i will publish the info here + on my personal Website: www.nicobaumgartner.ch

Still looking forward to suggestions.

Kiriakos GR said:

You need a true digital Oscilloscope if you need to deal seriously with ebikes.

Click to expand...

Maybe if your'e designing them, or figuring out communications problems between parts of "advanced" bikes, but almost never is one needed just to troubleshoot a problem with the typical ebike kits or components most people come here to troubleshoot.

Most troubleshooting can be easily accomplished with just a multimeter (digital or analog), and someitmes some resistors for testing halls and such (cuz you may need pullups or pulldowns to emulate what is inside a controller, when testing a motor separately).

nioko said:

Hello everyone, I appreciate the help. Yes it's a different system, and I know many moved away from it because of the high costs, and lets say complicated service by BionX. I for myself am not a huge fan of it, but I do like the torque sensing in their system + you can get these systems used in Switzerland, most likely with a worn Battery, that I'm able to rebuild, which makes it interesting for me.


Anyway I don't even need to try getting parts through BionX as I know the answer will be an offer for a replacement motor.

As soon as possiblr, I will disassemble it, and then I will find out about the halls. i will publish the info here + on my personal Website: http://www.nicobaumgartner.ch

Still looking forward to suggestions.

Click to expand...

you should test the halls to see if they are functional and switch states.

power up the controller and move the hub magnets by the hall sensors and measure the voltage on the signal line to determine if it functions. you can find the pin out for the hall sensor by googling the data sheet for the honeywell SS41.

the signal should swing from about .3V to 4.5-4.7V.

I have narrowed down the problem to the battery. The BionX battery has it's own PCB which communicates through can bus with the motor + handlebar display. The battery PCB seems to have gone bad, as I know my battery is good, and my system runs fine when i use a friend's battery.

I'll have a look at the PCB ASAP, don't know if I'll be able to repair it, I'm looking for a dead battery to get a replacement PCB, but no luck so far.

Cheers Nioko

nioko said:

I have narrowed down the problem to the battery. The BionX battery has it's own PCB which communicates through can bus with the motor + handlebar display. The battery PCB seems to have gone bad, as I know my battery is good, and my system runs fine when i use a friend's battery.

I'll have a look at the PCB ASAP, don't know if I'll be able to repair it, I'm looking for a dead battery to get a replacement PCB, but no luck so far.

Cheers Nioko

Click to expand...

Hello Nioko,

Thanks for keeping us posted on your situation. To say the PCB has gone bad is pretty vague. After reading your post I did a little searching for replacement parts and found a place selling fuses for the battery, any chance replacing that might do the trick? http://www.nycewheels.com/bionx-parts-fuse.html#tab1

Maybe you could spell out exactly how you are so certain the 'pcb has gone bad', it is smashed into little pieces? If one component on the PCB is broken, buying a new one and replacing it may be a far easier and cheaper task then getting a replacement PCB.

Are you familiar with performing a continuity test using a multimeter?

Kiriakos GR said:

Ykick said:

Kiriakos GR said:

Poor advice. How is this supposed to test an integrated motor/controller?

Click to expand...

Click to expand...

Since when you become chef of the army? so to qualify about having an opinion over other people offered advices

I do not have hub based motor with hall, but even so I did buy this gismo checker so to help me even with it throttle test function.
DVM & resistors ?

You need a true digital Oscilloscope if you need to deal seriously with ebikes.
But this is not what I am going to suggest to young people when they ask a technical question in a forum.

Click to expand...

You don't what you're talking about. Welcome to my iggy list....

I knew it must be the PCB within the battery, because I know my battery is good (I run a custom battery with the original PCB). When I tested with my friends battery (had to use my wheel in his bike), it worked, so it had to be the PCB.

In the meantime, I have disassembled my case that holds the PCB:




I can't read whats written on the FET's, but I do have a second (also defective) PCB, I'll try to move these fets...

While those FETs look bad, you still haven't ruled out the cables and connectors and display-controler. If you have the older Bionx bus, intermittency was an issue with the bus.

alan said:

While those FETs look bad, you still haven't ruled out the cables and connectors and display-controler. If you have the older Bionx bus, intermittency was an issue with the bus.

Click to expand...

It's a can Bus System - not an even older I2C.
The FET in the middle is blown, into pieces, so I'm 100% sure that this is the problem.

I'm now looking for a replacement battery PCB + trying to move the FETs. If I still have problems with that, I'll try another console + cabling.

you do not have to remove the mosfet that burned up to determine if the that BMS will function.

just cut the source leg open so the mosfet is disconnected from the circuit and then test the remaining mosfets in parallel with that mosfet by using the diode tester on your voltmeter to evaluate the body diode of the remaining mosfets.

they will most probably be ok and you can reinsert the BMS in the circuit to see if it allows the bike to function again.

if not then we can look further to see what else may have happened. eventually you can buy another mosfet identical to those adjacent in parallel if that is the problem. looks like it is. imo

nioko said:

It's a can Bus System - not an even older I2C.
The FET in the middle is blown, into pieces, so I'm 100% sure that this is the problem.

I'm now looking for a replacement battery PCB + trying to move the FETs. If I still have problems with that, I'll try another console + cabling.

Click to expand...

Just a thought --- but.

The power connection to the motor is only 2-wire. None of the wires in the "comm" cable are heavy enough to carry any real current. SO... my best guess is those aren't transistors, they are diodes for rectifying the return voltage during regenerative braking. And you did say this started during regen braking.

Hold the PCB up to a BRIGHT light and try to see the associated board traces. If you can, can you determine if they are wired in parallel - three cheap diodes in parallel instead of one expensive high current diode.

IF they are diodes, then replacing them should probably fix the problem. You can maybe save time and effort by checking the two undamaged diodes to be sure they are still good.

All the above could easily be nonsense, but I don't understand what FETS would be doing in the battery circuit board. I think I have seen pictures of the motor interior and the phase MOSFETS are inside the motor. I am also not sure why failure of regen diodes would cause cogging when the motor is driving ... more and more convinced that this might all be nonsense....but maybe worth a thought.

In any case, good luck and please keep us posted. I can use whatever info I can get on Bionx internals, I am keeping an old I2C PL350 running on my Terratrike with 10s LiPo pack in the battery box in place of the old dead LiMn pack. I just LOVE the simplicity of the torque-based PAS, but I HATE the closed system.

If you know of anywhere there is even a minimal schematic of the Bionx electronics I would appreciate a hint on where to find it.

no, they are not diodes, they are mosfets. if you cut open the source leg of the one that shorted then you can test the body diode of the parallel mosfets.if you can figure out what the label is under the black goop then you can order a replacement.

otherwise you would have to swap out all three mosfets so they are identical.

icerider said:

The power connection to the motor is only 2-wire. None of the wires in the "comm" cable are heavy enough to carry any real current. SO... my best guess is those aren't transistors, they are diodes for rectifying the return voltage during regenerative braking. And you did say this started during regen braking.

Click to expand...

All the above could easily be nonsense, but I don't understand what FETS would be doing in the battery circuit board.

Click to expand...

FETs are used in BMS boards like that to turn the battery input (charge) or output (discharge) on or off depending on the battery condition, and/or the current flow or voltage coming in / going out, depending on how it's designed.

So one way they coudl fail during regen is if somehow the regen current or voltage was higher than the FETs' rated current or voltage. Probably wasn't current, but a voltage spike that's high enough could do it. (the battery should act as a capacitor to smooth such spikes, but that does not always appear to happen).

Regen is basically jsut charging, albeit in small bursts rather than a constant charge to full like plugging into the wall woudl do. But it does the same basic thing, and those FETs would be able to block regen teh same way, too, if they detected a condition that required that.



Speculation:

If there is no limitation inside the motor itself (where the controlelr is) then if the FETs in the BMS turned off to protect the battery for whatever reason, the now-open-circuit from the motor/controller would spike in voltage possibly quite a lot.

If that happened, and it was still applied across teh FETs, it could've exceeded the BMS FETs' voltage rating and POP.


(it could also do the same thing to the FETs inside the controller, I suppose).

Hello Everyone, and thanks for all the thoughts / ideas.

I like the Diode idea, but I'm pretty sure that these are FET's that actually turn on the "Power-Lines" to the motor. I was also thinking about why it happens in both ways (acceleration and regen).

Anyway, I have replaced the FET's or whatever they are, from the second BMS (I actually don't like calling it a BMS, as it does not Balance or monitor individual cells). The Problem persists, so there must be some other thing that went wrong.

I'm looking for a replacement BMS, but it's difficult to find this old model. What I have found is 5 dead batteries at a bike store. But they are all newer, 13s models. I was able to build a working battery from it, as one had good cells and a bad BMS, the others have bad cells.

My BionX however has an older Firmware version, and I think that's why my "new" battery is not working with it. I do however have a newer BionX motor + Controller that seem to work with it, but I have only done a little bench test.

I'll try to find someone who can update my Motor + Console. And then retry.

Replacing the FET's on the Battery PCB did not solve the problem, but made it worse. I have fried my motor, it does not respond on the can bus anymore. (not to the console, not to my BigXionFlasher, and not to the official BBi). This was tested with another battery.

Well' that's obviously not what I wanted to achieve, I'm now trying to build a 13s Battery to use with my other BionX motor, and then sell what's left after this works.

I will not invest more time on the fried motor, perhaps I'll use it with an external controller, but what I like about the BionX is the torque sensor, and interfacing this will be difficult.

Cheers, Nico