Problems with QS Motor when in hall sensor 5 position

[TommyCat]

Just in case you haven't seen this...

QSMOTOR Motor Manual

Careful with static electricity around these hall sensors.

The readings are consistent with the secondary hall connector which for some reason causes a “lack voltage” error on the controller

What error code or number of flashes is this?
What are the hall sensors supply voltage readings, and do they remain that way for both sets after each of the hall sensor cable are connected? (I.G. 5vdc)

What are the halls signal voltage readings after each hall sensor is activated? (I.G. 0vdc)

The halls seem to be moving in a consistent pattern too with one on, then two, then one. Would this be correct for bipolar hall sensors?

This is correct for a 120-degree output sequence which your motor does use.


But curious how you came to these conclusions...

Using a hall diagnostic tool I have been able to ascertain that the hall positions correspond with the phase colourings below:
1 blue
2 blue/yellow
3 blue/green
4 green/yellow
5 green
6 yellow

There are no shorts across any of the cables and the connectivity from the hub to the end connector all seems to be fine.

Did you check between cable wires and metal stator frame? Occasionally slightly pulling or moving the cables a bit whist testing?

Yellow hall sensor signal voltages.
5vdc coming from the controller.
6.1vdc when connected to the motor.
When connected and with the main phase wires disconnected... anything change?


Give ALL bench test voltages when available.

 

[C4m3Ltoe]

But curious how you came to these conclusions...

C4m3Ltoe said:
Using a hall diagnostic tool I have been able to ascertain that the hall positions correspond with the phase colourings below:
1 blue
2 blue/yellow
3 blue/green
4 green/yellow
5 green
6 yellow

Thanks for the reply, I will test some voltages later today. In regards to the hall position/colouring I first connected the controller to the hall sensor, powered on and got the hall reading from the MQCON app. Then powered off, and connected the hall sensor to the hall diagnostic tool to see what

Just to clarify on the readings I have done so far:

Hall reading from connector with controller powered on and three phase cable connected

ground/blue - 3.6v
ground/green - 3.6v
ground/yellow - 6.2v

(I now realise that a good test here would have been to remove the 3phase cable from the controller and repeat)

Hall reading from the controller connector (this correctly initiates a "hall fail" error)

ground/blue - 3v
ground/green - 3v
ground/yellow - 5v

While the wheel was off yesterday, I did check the voltage using power from the hall diagnostic tool (4v) between the ground/yellow which did report back as 4v.

I will attach the wheel back on today, but with a cover removed so I can get further readings from there if needed. The readings I plan to get are

- Repeat the above
- Hall readings with the controller powered but no phase wires
- Confirmation of the power sent on the ground/red rail

 

[TommyCat]

Any switched signal values yet?

Why not test both sets whilst you're at it... I always go after the obvious first as it may influence what you
are focused on.

I’d probably take the stator out from the ring magnet to further test at this time, leaving it apart. And change the status of the hall sensor by using an individual hand held magnet. First one pole, then the other.

But that’s just me…

Your electrician friend have a megger?

 

[C4m3Ltoe]

So I have just been out to run some voltage tests and the results are a lot different to how they were previously being tested.

First I tried testing using the 5v from the diagnostic tool, everything checked out. So I hooked the 3 phase supply on the controller and powered it while still using the diagnostic tool (just so I can confirm using multiple power sources).

Red/Black - 4.95v
Ground/All hall phase wires - 3.36

Then I connected the hall to the controller and again powered on

Red/Black - 4.91v
Ground/All hall phase wires - 2.91v

Both results are a lot better than when the cover was on where I was seeing 6+v on the ground/yellow hall phase.

I do want to run a hall test with the cover off but will need to properly mount the motor if I am to have the wheel moving so I will give this ago tomorrow when I have better light.

I do know people with Megger's that we could use to further determine where a possible short could be occurring

 

[C4m3Ltoe]

Motor casing back on and remounted to the wheel, the issue seems to have got worse and the motor seems to stall/grind even with no load with the bike lifted.

The hall voltage values have changed slightly:

(results are the same with the 3 phases from the wheel>motor attached or removed. The hall connector is connected to the controller for power)

Red/Black - 4.91v
Ground/green - 2.91v
Ground/blue - 2.91v
Ground/Yellow - 4.91

As you can see from yesterdays tests, the voltages between the ground/hall phase wires were consistent when the plate was removed so something is defiantly changing once the plate is fixed back in place.

Using the diagnostic tool, I see the halls appear to be still functioning properly and in sequence still so they appear to be working.

I see that there is a hall plug converter which I can get for £10 which would mean I could hook the wheel to my working sabvoton controller and test. This will at least confirm the operation status of the motor. The person who repaired the halls informed me that the motor was 100% working fine when tested on a 48v bench controller he has.

This issue first started after the hall wire was replaced (I don't think the phase wires were replace). The bike was making the grinding sound after the cable was replaced (the replacement work blew two of the halls, presumably due to static so we did a lot of troubleshooting on the bike and assumed the grinding was part of the overall problem) so I am inclined to think this is still a motor/wiring issue rather than a failed controller which hasn't really been touched.

 

[AGS]

Try using the multimeter set on volts to check that each hall output switches by rotating the wheel very slowly by hand instead of using your diagnostic tool?

The diagnostic tool may be giving a false result.

The controller needs to be connected because the halls need a pull up resistor to switch on and it also provides the 5 volt supply.

If they are all bipolar they should switch on and stay on until they see an opposite magnet pole and then they should switch off. If one or more is a unipolar then they should switch on but not latch on. You should see a difference on the multimeter between a unipolar switching and a bipolar switching. Unipolars will switch off quicker than bipolars as the wheel is rotated.

Both sets of halls can be tested this way.

 

[C4m3Ltoe]

Try using the multimeter set on volts to check that each hall output switches by rotating the wheel very slowly by hand instead of using your diagnostic tool?

The diagnostic tool may be giving a false result.

The controller needs to be connected because the halls need a pull up resistor to switch on and it also provides the 5 volt supply.

If they are all bipolar they should switch on and stay on until they see an opposite magnet pole and then they should switch off. If one or more is a unipolar then they should switch on but not latch on. You should see a difference on the multimeter between a unipolar switching and a bipolar switching. Unipolars will switch off quicker than bipolars as the wheel is rotated.

Both sets of halls can be tested this way.

Hi, those voltages were taken with the multimeter and with the hall connector & phase wires from the hub, connected to the controller. So essentially its the setup that should be working on the bike. I am taking these readings from the hall/motor connector where I can slide the probes down the back and get a good reading.

The halls all fire up and close down while the wheel is rotated so the results match the diagnostic tool.

I did reach out to the person who replaced the halls who has assured me that they are bipolar, and that the bike was working on his 48v controller

 

[AGS]

So are all three hall sensors switching between 2.91v to 4.91v when the wheel is slowly rotated by hand?

 

[C4m3Ltoe]

So are all three hall sensors switching between 2.91v to 4.91v when the wheel is slowly rotated by hand?

No, the blue and green are switching from 0v-2.91v, and the yellow is from 0v-4.91v. Hope that makes sense

 

[AGS]

Blue and green are broken, they aren’t switching off properly.

 

[C4m3Ltoe]

They seem to be switching fine, the voltage drops from 2.91v to zero as the wheel moves. It looks like the higher voltage on the yellow is due to the controller sending out 5v on the ground/yellow, while it sends 3v on the blue/green. Looking at some post history, it seems other Sabvotons have been found to do the same.

I have just run another test using the diagnostic machine plugged in and I get 3.35v on all phases (dropping on and off as the wheel rotates).

My friend came round earlier and advised me that he has already ordered a new motor, he acknowledges he was quite rough when taking off the motor cover the first time and may have inadvertently damaged it (although visually, everything looks fine). I do need to order a new converter lead for the new motor, so to fully rule out the controller I can hook it up to my Sabvoton and test before the motor is removed from the rim. We will attempt to get the motor running again so he can keep it as a spare for another project.

I will update once its been tested with a working controller.

 

[AGS]

Yellow is switching on and off correctly. Blue and green are not switching off properly. When they are switched off the voltage on the output should be at 5v and when they are switched on the voltage at the output should be at 0v.

This is because these hall sensors are open collector and there is a pull up resistor inside the controller that is connected between the 5 volt supply and the output of the hall sensor. So when the hall sensors are off the resistor pulls the output of the hall sensor up to 5 volts.

If you are only getting a reading of 2.91 volts on the output of the blue and green hall sensors it means they aren’t fully switching off and the controller may think they are still switched on, depending upon the threshold level of the controller switching circuit.

There are 2 reasons for this, either the hall sensors are damaged or the magnetic field isn’t strong enough to switch them off completely.

Are you sure they have been inserted into the slots properly?

I clean out the slots carefully to make sure there isn’t any debris inside them and then put a spot of super glue into the slot before inserting them fully home and hold them for a few seconds to make sure they don’t move.

Here is a diagram that may help you understand how they work. If you can get hold of some 10k resistors you can test the hall sensors independently of the controller.

 

[C4m3Ltoe]

I may have confused things because of the voltage difference on the controller. All sensors are dropping to 0, and then opening at a similar voltage from the source (either controller or diagnostic).

The voltages I have provided are through me moving the wheel slightly and reading the voltage as the hall becomes active.

 

[TommyCat]

I am taking these readings from the hall/motor connector where I can slide the probes down the back and get a good reading.

Which side of the connector are you testing From? I.E. Controller side, or motor side. Can you get to both sides easily?

Red/Black - 4.91v
Ground/green - 2.91v
Ground/blue - 2.91v
Ground/Yellow - 4.91

As you can see from yesterday’s tests, the voltages between the ground/hall phase wires were consistent when the plate was removed so something is defiantly changing once the plate is fixed back in place.

I find this the most intriguing, and would personally explore this abnormality further.
Have plenty of testing suggestions if you’re interested. Just let me know.

They seem to be switching fine, the voltage drops from 2.91v to zero as the wheel moves. It looks like the higher voltage on the yellow is due to the controller sending out 5v on the ground/yellow, while it sends 3v on the blue/green. Looking at some post history, it seems other Sabvotons have been found to do the same.

I have noticed this also and wondered why? EMF issues?

Good luck with the new motor.

 

[C4m3Ltoe]

So my friend has purchased a new QS 273 V3 and has spend sometime lacing it to the wheel. We went to get it setup on the controller today and guess what... exactly the same symptoms.

Positioning the wheel into hall sensor position 5 doesn't fire the motor up at all (but a whine can be heard from the motor so something is energising).
A hall test doesn't complete, at best it moves 8 notches and then makes a loud and erratic movement.
Controller settings are all ok, the motor is set to a 16 pair pole, hall angle is at 241 for now.

Looking at the Sabvoton documentation, it mentions trying swapping phase wiring (presumably) on the controller. A quick google found this post which seems to suggest the phase wiring order may not correspond to the colours indicated. Sabvoton wiring confusion

Happy to try swapping over phase wires to see if this helps, but unsure how safe this is? Is there potential to damage anything long term? I can drop the amps down on the controller while testing if that helps.