No voltage when connecting Hall sensors

[snowzoor]

Hello everyone . I have an issue with my ebike and would be really happy if someone with more experience could help me.

When I connect my hall sensors to controller, suddenly the voltage (from controller) drops from 5V to around 0,8V all around the circuit. I have already replaced my controller and the issue persists. 48V power is OK, it's only 5V that causes problems.

If there is someone who could give me some guidance on what could that be, I would be very thankful?

 

[E-HP]

Hello everyone . I have an issue with my ebike and would be really happy if someone with more experience could help me.

When I connect my hall sensors to controller, suddenly the voltage (from controller) drops from 5V to around 0,8V all around the circuit. I have already replaced my controller and the issue persists. 48V power is OK, it's only 5V that causes problems.

If there is someone who could give me some guidance on what could that be, I would be very thankful?

Can you take a picture of the hall sensor connector, connected to the controller's hall sensor output? If you can take them at an angle so the wiring is visible, that would be helpful.

 

[amberwolf]

When I connect my hall sensors to controller, suddenly the voltage (from controller) drops from 5V to around 0,8V all around the circuit. I have already replaced my controller and the issue persists. 48V power is OK, it's only 5V that causes problems.

This means that the 5v wire on the hall connector from the controller is being shorted out, probably to ground or one of the hall sensor signal lines, at the motor.

Did the system ever work correctly?

Or is this a new problem?

If it never worked, it is probably that the motor is wired at the connector in a different order than the controller.

If it did work and now doesn't, what specifically happened between the time it worked and the time it didn't? This will help us help you figure out the specific problem causing it.

 

[snowzoor]

When I connect my hall sensors to controller, suddenly the voltage (from controller) drops from 5V to around 0,8V all around the circuit. I have already replaced my controller and the issue persists. 48V power is OK, it's only 5V that causes problems.

This means that the 5v wire on the hall connector from the controller is being shorted out, probably to ground or one of the hall sensor signal lines, at the motor.

Did the system ever work correctly?

Or is this a new problem?

If it never worked, it is probably that the motor is wired at the connector in a different order than the controller.

If it did work and now doesn't, what specifically happened between the time it worked and the time it didn't? This will help us help you figure out the specific problem causing it.

Yes, ebike worked fine before. I don't know what happened to it since I actually sold the bike, but the buyer called me the next day stating that the bike is not working anymore. He claimed all he did was driving around the house for around 20km and after that it stopped working. He said he tried to recharge it but it would not work regardless. I just gave him new bike and decided I would try to solve the problem myself.

Someone asked for the picture of hall sensor connection, so here it is (only 5V is connected due to solving the problem).

 

[amberwolf]

c
Yes, ebike worked fine before. I don't know what happened to it since I actually sold the bike, but the buyer called me the next day stating that the bike is not working anymore. He claimed all he did was driving around the house for around 20km and after that it stopped working. He said he tried to recharge it but it would not work regardless. I just gave him new bike and decided I would try to solve the problem myself.

Someone asked for the picture of hall sensor connection, so here it is (only 5V is connected due to solving the problem).
1661335900678.jpg

Well, the most likely thing is that the buyer did something to damage the bike that they wont' admit to (this has happened with people here enough over the years). The most likely two are:

--Overheating the motor causing either failed hall sensors in a way that shorts out the 5v, or melting insulation in the motor cable in the axle that shorted the 5v to something else (most likely a hall sensor signal or the hall sensor ground; if it was shorted to a phase it would almost certainly have destroyed the hall sensors and the MCU in the controller, and the throttle / PAS sensor / etc (anything that connects to the 5v line).

--Dropping or crashing the bike onto the motor wire, usually where it exits the axle, which caused the same kind of problem melting the insulation in the cable would.


Either one can be tested for with a multimeter. Disconnect the motor cable entirely from the controller, both phase and halls, etc. Set meter to 200ohms or continuity. Place red meter lead on motor cable's hall 5V, and use black meter lead to test connection to every other motor wire. Note down all readings and what wire they reference.

You should get no continuity, and "OL" or "1." or whatever reading high resistance is for your meter, for every wire. Any wire that gets continuity or an ohms reading at 200ohm setting is shorted to the hall power wire.

 

[snowzoor]

Well, the most likely thing is that the buyer did something to damage the bike that they wont' admit to (this has happened with people here enough over the years). The most likely two are:

--Overheating the motor causing either failed hall sensors in a way that shorts out the 5v, or melting insulation in the motor cable in the axle that shorted the 5v to something else (most likely a hall sensor signal or the hall sensor ground; if it was shorted to a phase it would almost certainly have destroyed the hall sensors and the MCU in the controller, and the throttle / PAS sensor / etc (anything that connects to the 5v line).

--Dropping or crashing the bike onto the motor wire, usually where it exits the axle, which caused the same kind of problem melting the insulation in the cable would.


Either one can be tested for with a multimeter. Disconnect the motor cable entirely from the controller, both phase and halls, etc. Set meter to 200ohms or continuity. Place red meter lead on motor cable's hall 5V, and use black meter lead to test connection to every other motor wire. Note down all readings and what wire they reference.

You should get no continuity, and "OL" or "1." or whatever reading high resistance is for your meter, for every wire. Any wire that gets continuity or an ohms reading at 200ohm setting is shorted to the hall power wire.

Could be that he overheated the motor, as he stated he was driving it for longer period of time continuously and he is located in the hills, so there was probably more work for the motor to do uphill. I have checked the wire for visible damage (in case he would drop the bike) and could not find any.

I have checked with multimeter as you suggested and I have continuity on black and yellow wire. I have also tried to get inside of the motor to see what can be done, but unfortunately I don't have the right screw to open up cassette, will have to wait until it arrives from China. Thank you for everything so far! Hopefully I will be able to fix it after I receive the tool.

 

[amberwolf]

I have checked with multimeter as you suggested and I have continuity on black and yellow wire. I have also tried to get inside of the motor to see what can be done, but unfortunately I don't have the right screw to open up cassette, will have to wait until it arrives from China. Thank you for everything so far! Hopefully I will be able to fix it after I receive the tool.

Black wire is typically ground, and yellow (if it's a thin wire) would typically be a hall signal. Shorting those together should not short out your 5v. (it would probably make the system not run the motor correctly, if at all, but not affect the 5v itself). There is a pullup resistor between 5v and the signal inside the controller, and the halls ground that during normal operation, so grounding the signal with the wire would not affect 5v. The hall outputs are open-collector, so no 5v to them except from the controller's pullup, so grounding the hall output would not affect 5v.

If it's a thick yellow wire, that's probably a motor phase, and shorting that to the hall ground would probably damage the FETs on that phase, possibly blowing them up and causing the controller to not run the motor at all, but still not affecting the 5v in the way you see (it could well blow up the MCU or a gate driver and *that* could take out the 5v regulator, but that would be a complete failure and not just when the motor hall power and ground is connected.


So...unless they're using different wiring colors, or there's something else wrong not detected by a multimeter test, or is intermittent, I am not sure how that specific short could cause the failure you see.

 

[snowzoor]

Black wire is typically ground, and yellow (if it's a thin wire) would typically be a hall signal. Shorting those together should not short out your 5v. (it would probably make the system not run the motor correctly, if at all, but not affect the 5v itself). There is a pullup resistor between 5v and the signal inside the controller, and the halls ground that during normal operation, so grounding the signal with the wire would not affect 5v. The hall outputs are open-collector, so no 5v to them except from the controller's pullup, so grounding the hall output would not affect 5v.

If it's a thick yellow wire, that's probably a motor phase, and shorting that to the hall ground would probably damage the FETs on that phase, possibly blowing them up and causing the controller to not run the motor at all, but still not affecting the 5v in the way you see (it could well blow up the MCU or a gate driver and *that* could take out the 5v regulator, but that would be a complete failure and not just when the motor hall power and ground is connected.


So...unless they're using different wiring colors, or there's something else wrong not detected by a multimeter test, or is intermittent, I am not sure how that specific short could cause the failure you see.

Sorry, maybe I have written it a bit sloppy :? . I have checked with multimeter as you suggested to do, with the red probe on the red wire and then went on other wires with black probe of the multimeter.

The continuity was between red and thin yellow wire. And the continuity was also between red and black wire. This suggests that there is a short somewhere on the yellow wire if I understand correctly?

 

[E-HP]

Sorry, maybe I have written it a bit sloppy :? . I have checked with multimeter as you suggested to do, with the red probe on the red wire and then went on other wires with black probe of the multimeter.

The continuity was between red and thin yellow wire. And the continuity was also between red and black wire. This suggests that there is a short somewhere on the yellow wire if I understand correctly?

Just for completeness, can you confirm you also have continuity between the black and thin yellow?

 

[snowzoor]

Just for completeness, can you confirm you also have continuity between the black and thin yellow?

Yes, there is also continuity between the black and thin yellow.

 

[E-HP]

I think you're going to have to open up the motor.

 

[amberwolf]

Sorry, maybe I have written it a bit sloppy :? . I have checked with multimeter as you suggested to do, with the red probe on the red wire and then went on other wires with black probe of the multimeter.

The continuity was between red and thin yellow wire. And the continuity was also between red and black wire. This suggests that there is a short somewhere on the yellow wire if I understand correctly?

Ah, that makes more sense. Yes, that indicates all three wires are shorted together.

It could be the actual wires are shorted together, anywhere along the wiring between connector and sensor(s) (though most likely at the axle, or right there near the sensor inside the motor), or that the hall sensor for the yellow wire has failed in a way that shorts it's power, ground and signal wires together.

 

[E-HP]

It could be the actual wires are shorted together, anywhere along the wiring between connector and sensor(s) (though most likely at the axle, or right there near the sensor inside the motor), or that the hall sensor for the yellow wire has failed in a way that shorts it's power, ground and signal wires together.

Out of those possibilities, I'm guessing the latter, due to the odd voltage drop when the connector is in place, instead of pulling the voltage down to zero.

 

[snowzoor]

I have finally opened up the motor and I have found out there are actually only two haal sensors in there (photos) and out of them one is constantly outputing 5V and the other one does not output anything even when rotating the wheel.
Any suggestions?

 

[AGS]

It’s not beyond the realm of possibilities that the person you sold the bike to took it home and swapped out the inner of the motor and brought it back to you saying it’s broken.

The two black cable ties on the hall sensor pcb does not look like they were fitted at the factory.
And the position of the hall sensor pcb does not look like it was done at the factory. Especially with one of your two remaining hall sensors being out of position. Or did you prise the pcb away from the glue when investigating?

 

[snowzoor]

---

---

When I opened it up it actually looked like this (photo), but I just soldered the wires and added cable ties. I will never know if it was the buyer or just a shitty Chinese vendor, but I just don't care anymore, I just want to fix it so I can use or possibly sell it. Do you suggest I just buy a new PCB for it?

 

[AGS]

Let’s assume the buyer hasn’t messed with it.

Have you checked down the side of the magnet for the missing hall sensor? It may have broken off the pcb and still in location in the crack in between two magnets. I think I can see what looks like a broken off hall sensor in between two magnets in the photograph.

So when you opened it, the blue wire was disconnected and you resoldered it?
And did you pull the pcb away from the glue or did you find it like that?

That could be a dry joint from the factory.

So on the face of it it just needs a new set of hall sensors and the pcb needs to be re glued in place, providing the controller isn’t damaged.

 

[amberwolf]

I have finally opened up the motor and I have found out there are actually only two haal sensors in there (photos) and out of them one is constantly outputing 5V and the other one does not output anything even when rotating the wheel.

When you tested the sensors, was the motor connnected to the controller? If not, you need to use a separate pullup resistor (5k-10kohm) from each sensor signal wire to 5v, or you don't get a valid signal output (motor hall sensors are usually open-collector so they have no output, they only ground the voltage supplied by the controller on the internal pullup resistors on the signal pins).

If sensors are failed, and for the missing one, you can get Honeywell / Allegro SS41 or SS411 in whatever variation fits in your motor and works with the voltages you have there. There's also a bunch of clones of those, which are likely what are actually in the motor (not sure I've ever seen real ones in a motor except when someone has replaced the originals for whatever reason).