Help with Bafang Hub Motor

[Salty Souls]

Hi,

I recently purchased a non working ebike and upon diagnosis it would appear that it is a hall sensor issue. However, I also have some other questions about the set up as well.

Overall Set Up
- Battery = 72v, 14.4Ah, BMS 40Ah
- Controller = 60V, 30Ah
- Hub Motor = Bafang RM G0F4.1000w (same or similair to Bafang G063 motor) 48v to 60v?

My questions are:
1. On checking the Bafang website it indicates they only make this motor in 48v to 52v. However, it is marked on the hub wheel that it is 60v (see pictures). Does this matter and if so how can I check?:


2. From different forum sites I've gotten different answers (this could also be just my lack of understanding) about the requirement to match battery v to controller v to motor v - at a minimum battery and controller v has to equal. Is this right or does it have to been within a range.

3. Related to question 2, should this set up work in terms of battery, controller and motor specs? Or does the battery have to much V for the controller and in turn the controller for the motor (assuming I can confirm if the motor is 60v or not)?

4. I've tested the motor and it would appear that the hall sensors are faulty. I ran a continuity test using a multi meter from the motor cable connection to the PCB board and it all appeared fine - so I ruled out frayed or cut motor cable. Then connected power to the motor and got different voltage responses using the multi meter from each of the hall sensors as I slowly turned the wheel.

I've tried to find a replacement PCB board but can't find it anywhere - thought it would be easier to replace the PCB board than the hall sensors. If someone knows where I could purchase the PCB board that would be great (I've looked on aliexpress, etc and Ive contacted Bafang directly)


Thanks

 

[amberwolf]

Overall Set Up
- Battery = 72v, 14.4Ah, BMS 40Ah
- Controller = 60V, 30Ah
- Hub Motor = Bafang RM G0F4.1000w (same or similair to Bafang G063 motor) 48v to 60v?

Controllers shouldn't have an Ah rating, that is only for batteries. Could you recheck the label to be sure what it says?

BMS don't have an Ah rating either, so I'd also recommend rechecking it's label to find out what that really is.


If the battery is 72v that means it's 84v fully charged.

A 60v controller might be ok with that, but it depends on the actual parts used in it's LVPS (that makes 5v/etc from battery voltage) and it's capacitors and FETs. You might want to look up the ratings for all those parts to be sure they are well above that 84v the battery will charge to. If they're not above it, or only a little above it, you might have problems from that at some point.

Motor doesnt really have a voltage--it has an RPM/volt, which you can guesstimate from the voltage it's "rated" for and the speed it is advertised to go at that voltage. Markings on motors (and other parts) are commonly made custom for specific resellers, or changed by those resellers for whatever reasons they have (whether they are true or not).


So...you don't need an exact match between controller and battery--the controller has to be rated at least as high as the voltage of the battery; normally these are all "nominal" ratings, meaning the half-charged voltage of the battery. But contorllers have an LVC, which means they'll shut off motor power when the battery drops below that, and may reduce power as it's approached. They may aklso have an HVC, meanign a voltage above which they won't operate, or where some functions (regen, etc) won't oprate. Those are not usuallly marked on them.

Some controlelrs are programmable for some functions like KT, Lishui, etc., but the available paramenters and values vary a lot between models, brands, and even sellers. Some only program via their displays, and some only do it via a phone app or a computer program.

4. I've tested the motor and it would appear that the hall sensors are faulty. I ran a continuity test using a multi meter from the motor cable connection to the PCB board and it all appeared fine - so I ruled out frayed or cut motor cable. Then connected power to the motor and got different voltage responses using the multi meter from each of the hall sensors as I slowly turned the wheel.

The halls should toggle on and off (around 0v when on, around 5v when off) as long as you are measuring with the controller connected to the motor, and the controller on but no throttle applied. This gives the halls pullup voltages so they operate correctly. If there's no pulup on the signal lines from the controller, or your own test setup, then the outputs are not valid signals as the sensors don't output anything, they only ground the voltage supplied by the controller on each signal line.

If htey all toggle fully off and fully on when tested this way, then they're not faulty. If any don't, they may be faulty; depends on the specific measurements you get. Some readings indicate a controller or cable fault.

Regarding cable issues, there are damages internal to the cable from various issues that can leave wires shorted together, which won't show up in a continuity test from end to end, only from wire to wire. (motor phases will always show shorted, so don't worry about those--you can't test for a phase short problem with a regular ohmmeter).

 

[Salty Souls]

Thanks for the reply amberwolf.

You are correct my bad and typo - the controller is 30A and the BMS is 40A. Below is a picture of the controller:


Let me double check to see if I can get the specs of the controller.

Re motor, on the Bafang website it has a rated voltage of 48 / 52 and a max speed of 45 / 45+ - does this equate to the speed it is advertised to go at that voltage?

Is there a specific way for me to test the motor to determine if it is a cable or hall sensor issue?

Thanks

 

[E-HP]

What did the seller say happened to the bike? What were the symptoms when you bought it?
Testing/trouble shooting:

Testing BLDC motor's Phase Wiring - Hall Sensors and Wiring. - Electricbike.com Ebike Forum

Testing electrical components of a typical hub motor with a digital multi meter. First take a few minuets and look over the wiring going into the motor for cuts, scrapes, missing insulation and other wiring issues or poor connections that should be resolved first. Smell for...

electricbike.com

 

[Salty Souls]

Thanks for the reply E-HP.

Everything turns on and then when you press the throttle the motor starts to move but then quickly stops, almost like a jolt. The seller basically explained that as the sympton and said it just stopped one day!

Thanks for the link I'll take a look and see how I go.

Cheers

 

[amberwolf]

Everything turns on and then when you press the throttle the motor starts to move but then quickly stops, almost like a jolt. The seller basically explained that as the sympton and said it just stopped one day!

That is usually hall sensor or cable damage, and usually happens from a crash or an axle that wasn't secured that then twisted in the dropouts, causing itnernal wire damage to the cable.

Actual hall sensor damage can happen from the cable damage (shorting phase wires to hall signals/power), or from overheating of the motor. They rarely "just fail". Corrosion can also do it, but I see no signs on the PCB of water intrusion.

 

[Salty Souls]

Amberwolf thanks for the reply - very helpful

The axle was twisted in the drop out and I had to realign to get the hub motor out of the frame. While i can‘t see visible damage to the motor cable it could obviously be internal. Sounds as if i could be dealing with both a damaged motor cable and hall sensors.

You are right, the internals of the motor look very clean.

 

[amberwolf]

The axle was twisted in the drop out and I had to realign to get the hub motor out of the frame.

Unforutnately that's not uncommon. If used correctly with the right dropout design, the "torque washers" that come on many hubmotors can be sufficient for the original rated controller / system that comes with the motor, but it's pretty common for users to not know how to install them (few kits have instructions of any kind, and few of the ones that do have them are useful guides), so motors then have nothing to prevent the axle from rotating counter to the torque the motor provides, instead of transfering the torque to the frame and spinning the wheel like they're supposed to.

Even if the hardware is correctly installed, it's not infrequently made with an axle hole that is too large (even a teensy bit too large is bad), so the axle can rock back and forth in it, and cut into the metal utnil it doesn't do anything to hold it in place.

So the axle can spin, and tear up the wiring.

Good, properly-installed tight-fitting torque arms on each axle end can prevent this. If you don't have any, I'd recommend getting the v7 Grin Tech ones from ebikes.ca (not any of the clones out there, which won't be properly machined to fit).


if the motor axle did spin and damage the cable, and there's enough undamaged length outside the motor past the axle, you can pull the cable thru the axle from the inside of the motor, cut the damaged area off, then reconnect each wire in the cable matching the colors on the remaining portion still inside the motor. It's important that the wiring match, or else you'll end up having to figure out which ones go where between mtoro and cotnroller.

That's assuming the controller and motor worked together before, and that the controller still works.

If there was a short created between phase and hall wires in wiring damage, it can destroy the halls invovled, and the controllers hall signal inputs, or the 5v supply, or the MCU itself.

 

[E-HP]

The seller basically explained that as the sympton and said it just stopped one day!

The axle was twisted in the drop out and I had to realign to get the hub motor out of the frame.

I guess the seller didn’t tell you the whole story.

 

[Salty Souls]

No, it would appear he didn't.

 

[E-HP]

There are a couple of solutions, once you determine the extent of the damage. My gut feel is that the motor is fine, and there’s a good chance the controller is ok too. If you’re getting a jolt of movement, and no smoke, I think the phase wires probably escaped damage. I don’t think the hall sensor conductors were that lucky.
You can test the phase and hall conductors to identify issues via the motor connector.

Testing BLDC motor's Phase Wiring - Hall Sensors and Wiring. - Electricbike.com Ebike Forum

Testing electrical components of a typical hub motor with a digital multi meter. First take a few minuets and look over the wiring going into the motor for cuts, scrapes, missing insulation and other wiring issues or poor connections that should be resolved first. Smell for...

electricbike.com

If the phase wires aren’t shorted, but there’s issues with the halls, you could reopen the motor and check continuity directly. So, if the phase wires are good, but some of the hall conductors are severed, but not shorted, you could either fix the cable (might be challenging, but doable), or switch to using a sensorless controller.

Check the 5v red wire of the controllers hall sensor connector with the controller on. If you’re getting 5v, you probably didn’t damage the controller.

 

[Salty Souls]

Thanks for the reply E-HP.

I'm planning to test thoroughly this weekend based on the link you've provide above.

I'll report back and let you know how I go.

Thanks again

 

[Salty Souls]

So ive started to test the hub motor and have the following results (apologies if i get the unit of measures wrong).

Hub Motor
Tested continuity between all combinations of the phase wires at the connector - all combinations had continuity.

I tried running a resistance test across the 3 phase wires but couldn’t get a reading, multi meter was just indicating low. Not sure if this is an issue or just the fact I have a cheap multi meter that doesn’t read that low? Also tested for any shorts in the phase wires and all appeared ok.

i then tested the hub motor with the battery and controller connected to the hub motor. I have a molded connector on the hub motor so thought this would be the best way.

Tested each hall sensor directly on the points on the pcb board.
Each hall sensor had a reading of ~4.5v. When i slowly turned the hub motor two of the hall sensors voltage moved between ~4.5 v and ~0v, however, the third was just stuck at 4.5v (im assuming this is a faulty hall sensor?)

Controller
i did a resistance test by connecting the black lead of the multi meter to the negative of the battery connection and then each of the points on the female end of the motor connector.

The phase wires all gave a reading of 12.05 ohms, one was slightly lower at 12.02 ohms but I’m assuming that is ok?

I then tested the hall sensor points. Three gave a reading of 3.67 ohms and one that was materially lower at 1.947 ohms. Im assuming this potentially indicates an issue in the controller as well?

Please let me know any thoughts

 

[fechter]

Tested each hall sensor directly on the points on the pcb board.
Each hall sensor had a reading of ~4.5v. When i slowly turned the hub motor two of the hall sensors voltage moved between ~4.5 v and ~0v, however, the third was just stuck at 4.5v (im assuming this is a faulty hall sensor?)

Yes. Bad hall sensor or signal wire is shorted to power.

 

[Salty Souls]

Can someone help tell me identify what is RT, located on the underside of the PCB board on my bafang hub motor:

This (RT) is attached to the negative side of the last hall sensor in the series and it appears to be a resistor? I've tried a google search but get different answers - surge protector, temperature resistor?

I dont have continuity between the negative connection on the PCB board to the negative side of this hall sensor - I'm assuming the connection is broken at this point, that is RT?

Thanks for any help.

 

[amberwolf]

RT is probably a thermal sensor (most often a 10K NTC). Center pin of halls is usually ground, so it's just sharing that hall sensor's ground line at one end of the sensor; the other end probably goes to a separate wire out to the world.

 

[Salty Souls]

thanks amberwolf

Yes, the other end just goes out to a separate end.

im assuming it’s damaged as I don’t get a continuity signal from the negative point on the PCB board to this hall sensors ground line - where as I do for all the other hall sensor ground lines.

im assuming this damaged sensor has broken the circuit and is giving me a hall sensor failure - I’ve tested that specific hall sensor individually and it appears to be working

Have I got this right, everything else appears to be working and I need to replace this sensor?

Thanks

 

[amberwolf]

If you don't get continuity from the ground to the hall, then the wire or trace to the hall is broken, not hte temperature sensor.


The tepmerature sensor doesn't do any grounding so hasn't anything to do with the hall ground. you could physically remove the senosr and it wouldn't change anything about the halls ground.

The senosr will read a resistance acrsso it based on temperature if you set your meter to 20kohm range. most of these are 10k ntc so at ambient about 72f they read 10kohm, and drop as temp is above that, and rise as temp drops below that.

 

[Salty Souls]

Thanks again amberwolf

yep, just tested it and it’s working fine. The circuit actually goes to the speed sensor and I get continuity from the negative point on the PCB board to the ground pin of the speed sensor.

I just can’t see where the trace to the ground is broken?

thanks again, I’ll keep searching.

 

[amberwolf]

yep, just tested it and it’s working fine. The circuit actually goes to the speed sensor

Ah--then it's one of those that uses the combined (multiplexed) speed and temperature signals, which look like this:
from the baserunner manual. If a controller doesn't have circuitry to deal with this type of signal and expects just one of those on the line, it may read speed incorrectly when temperature is high or low enough to drop the signal level of the speed pulses low enough, and may never read temperature correctly.

and I get continuity from the negative point on the PCB board to the ground pin of the speed sensor.

I just can’t see where the trace to the ground is broken?

If you get continuity from the main ground point of the PCB to the grounds of all the hall sensors (including hte one that you're not getting a valid signal for) then there are no breaks in the ground traces.

If you don't get continuity from that main ground point (where the (usually) black "hall ground" wire from the main motor cable is soldered to the PCB) to one of the hall grounds, *then* there is a break in the trace somewhere. Usually that is at the thru-holes called vias, where the trace stops on one side of the PCB and goes thru a plated barrel in the hole to a continuation of the trace on the other side. The barrel can separate from the trace. Easy to fix by putting a wire thru the hole and soldering it to the trace pads on either side, the same way a component leg would be.

 

[Salty Souls]

Ok, after testing all the trace lines and the vias points, it turned out to be a bad solder joint of the ground point of the hall sensor itself!!!

It appears to have been fixed by taking out the old solder and re-soldering the point of the hall sensor ground and the PCB board.

Thanks to everyone for the input / advice, much appreciated.

 

[fechter]

Thanks for reporting the results.

Time for a test ride!