10" Hub Wheel Stuttering

simonen

1 mW
Joined
Jul 8, 2020
Messages
17
Location
Carlsbad, CA
Hi everyone, I am new to the forum and looking for some help with an electric scooter build using an ebike controller.

I seem to be running into an error code 7 on the controller, which I assumed meant that it was over voltage. I built a 16S6P battery for this which is at around 67 volts fully charged. Since I was getting this error, I tried a 52v battery from my ebike, but I am still getting the error code.

The controller is connected to a 10" 52v 1000w hub wheel without a hall sensor.

To make matters worse, this controller came with zero documentation on how to setup via the lcd screen. I found some material on this forum, and I was able to set the voltage to 60 with my 16s battery or 52 with my 52v battery, but error persists.

Does anyone have any experience with this controller that may be able to offer some guidance?
 

Attachments

  • IMG_7266.jpeg
    IMG_7266.jpeg
    197.2 KB · Views: 580
  • IMG_7269.jpeg
    IMG_7269.jpeg
    153.5 KB · Views: 580
simonen said:
Well, I guess this could be the problem
Looks pretty likely. Did it come that way from the seller? Or has it been mounted and run at all?


The former is common with damaged boxes cuz there's usuqally no packing around the axle.

The latter can damage the wires when the axle isn't clamped down securely and it spins under power, twisting up the wires.
 
This was my first attempt to run the motor having just received the controller. So it appears to have arrived this way. The seller has offered to provide a partial refund, but I will need to take responsibility to get it fixed and I do not feel like waiting another month for a replacement.

On a side note, it is absolutely irritating that there is no documentation provided with the controller describing what all the connections are for or how to program it. I have found diagrams online that come close enough that I can determine the essential connections, but still irritating. To make matters worse, it seems to have only specific wheel sizes to choose from in the setup which are all too big for my wheel. My wheel is 10 inch, and the smallest choice is 16 inch. Of course, that is assuming I have identified the correct selection to modify that.
 
After rebuilding the phase wires to the hub motor, everything seems to be working except I still have the error 7. Since I do not have a Hall connector coming out of the motor, could this be the issue? I was thinking about adding an external speed sensor and attach the signal wire to the white wire on the controller hall connector. Hopefully that will solve the error.
 
simonen said:
This was my first attempt to run the motor having just received the controller. So it appears to have arrived this way. The seller has offered to provide a partial refund, but I will need to take responsibility to get it fixed and I do not feel like waiting another month for a replacement.
I would recommend removing the side cover, pulling the cable into the motor thru the axle, then cut out the damaged section (it's probably half inch or more either side of that spot, just in case) and reconnect and reinsulate the wires.

This is a common damage on shipping motor wheels, as they rarely put much packing in there, and sometimes none. Usually nothing to protect the axle. A very few do, but they have to charge a lot more for shipping, as they use wooden crates or very heavy duty cardboard boxes, often multilayer, with packing shaped to fill the whole box.


On a side note, it is absolutely irritating that there is no documentation provided with the controller describing what all the connections are for or how to program it.
This is completely normal, unfortunately. I wish I had more positive information for you, however:

There are *very rare* cases of companies, like Grin Tech, that provide such information for the things they sell.

Most sellers, and even manufacturers, have, at best, a half-translated connection diagram, with wires that look somewhat similar to the wires actually on the controller you received (which generally doesn't look quite like what they advertised, because it's probably not the same kind). But more commonly, they have zero information at all, and don't care to respond to inquiries of any kind, at least not *after* they've sold you something.

To make matters worse, it seems to have only specific wheel sizes to choose from in the setup which are all too big for my wheel. My wheel is 10 inch, and the smallest choice is 16 inch.

Also very common, rather than simply following the "cycle computer' convention of inputting the *actual circumference* in mm. There is a way around it, but it is very DIY:

You could make something with a small MCU, maybe atttiny, etc., that you program with the ratio of your actual tire diameter vs the calculated diameter it's using (the difference between your actual wheel speed and what it thinks it is), and it applies that to the pulses coming in from the speed sensor and calculates the pulses to send to the controller to give the corrected speed reading. It would probably take a few to a few dozen pulses to do it, so it may not be quite realtime, but it will be close enough except at really low speeds. It can be made in hardware instead, but it's harder to change when you change tires (which changes the circumference).


simonen said:
After rebuilding the phase wires to the hub motor, everything seems to be working except I still have the error 7. Since I do not have a Hall connector coming out of the motor, could this be the issue? I was thinking about adding an external speed sensor and attach the signal wire to the white wire on the controller hall connector. Hopefully that will solve the error.
Depends on what the error actually means.

If it means the speed sensor is not connected, then connecting one will fix it. Most geared hubmotors that don't have hall sensors (and some that do) have a speed sensor built in, and a 3-wire plug for the purpose.

If it means the hall sensors aren't connected, then you'll have to connect them; if the mtoor doesnt' have any and the controller requires them, then it is not compatible with that motor. You may be able to add them to the motor, but it is much easier to use a different controller, or different motor.


Did the motor come with the controller? Or were they purchased separately from different places (or the same one)?

If they came together, they should work together, but many sellers will sell things that don't work together to the same person in the same order without bothering to tell the buyer that, even if the buyer thinks to ask. :(
 
The hub motor only has 3 phase wires, there is no hall sensor. This is a little 10" hub motor for a scooter, no gears.

The controller was purchased separately, but they told me it would work with motors without hall sensors.

I was able to take it out for a short spin around the block, and I think as soon as I hit 3 mph, it started stuttering. That being said,I do not have the external speed sensor hooked up to the white hall wire on the controller yet.
 
simonen said:
The controller was purchased separately, but they told me it would work with motors without hall sensors.
Then the error shouldn't be for those.

I was able to take it out for a short spin around the block, and I think as soon as I hit 3 mph, it started stuttering. That being said,I do not have the external speed sensor hooked up to the white hall wire on the controller yet.
Then my best guess is that the error is a speed monitor wire...but, it shouldn't cause any problems other than it wont' be able read speed.

Motor stuttering is usually from a wrong phase/hall combo, but that can't be the case here if it's sensorless.

Addiitionally, it wouldn't be from the motor being sensorless while controller required sensors (if it did), beause that would be the opposite problem--working ok at faster speeds, but terrible at the startup and low speeds.

Maybe there is more wire damage than was obvious? (though again, taht's usually a stop or low speed issue, not as you get faster).
 
I just took it for another spin, and I found that I could reduce the stuttering by setting the magnet count lower. I think I counted 20 magnets in the motor when I had it apart. Setting it to 5 helped, but it threw off the speed reading. Also, the motor is wicked hot after about 5 minutes doing figure eights in front of the house.
 
If you get a speed reading without a speed sensor, then this means the number of magnets (poles) is used for speed reading, directly off a phase wire (so a speed sensor shoudln't be needed). If it's wrong number, of poles, then it would show wrong speed.


But it could also be used by the controller to determine motor positioning...so if it's nto right, it would affect motor operation.

Many hubmotors of larger diameter are 23 poles, but small ones could be a lot less. If you open up the motor again, count the number of magnets to verify, and divide by two, and that's the number of poles.

If your'e sure it's 20, then that's 10 poles (pole pairs, really).


IF it's gettng hot, that could be from too high a power level at too low a speed, or from a poor sensorless algorithm that causes bad timing of power to the motor, etc.
 
I wondering if the stuttering and heat build up is related to a timing issue since there is no hall sensor from the motor. I know the seller on ebay said the controller works without the hall sensor, but they could have simply lied or said what I wanted to hear.
 
If the motor spins under load, it's a sensorless capable controller. As noted in the previous post, however:

IF it's gettng hot, that could be from too high a power level at too low a speed, or from a poor sensorless algorithm that causes bad timing of power to the motor, etc.

As well as that it needs to have the right number of pole pairs set in it so it knows what speed it is at. If there's 20 magnets, then that's 10pp.

Ifyou're not sure of the number of magnets, then you need to verify that first.
 
amberwolf said:
If the motor spins under load, it's a sensorless capable controller. As noted in the previous post, however:

IF it's gettng hot, that could be from too high a power level at too low a speed, or from a poor sensorless algorithm that causes bad timing of power to the motor, etc.

As well as that it needs to have the right number of pole pairs set in it so it knows what speed it is at. If there's 20 magnets, then that's 10pp.

Ifyou're not sure of the number of magnets, then you need to verify that first.

I double checked the motor, and it has 20 magnets. The controller setup asks specifically for the number of "steel magnets", I take this to mean that I should enter in the actual number of magnets instead of dividing the value by 2.

Some other observations and items worth rementioning:
1.) Battery voltage drops significantly on heavy throttle up, but doesn't go below 2 bars. I don't believe it is hitting the LVC. The battery pack is 16s6p made with Samsung 26F cells. If I am not mistaken, they are each rated for 5A output.
2.) The controller lets me set a maximum of 15Amps when set at 60v. I have tried reducing it to 5Amps, but stuttering persists under load when throttling up.
3.) When stutter occurs, I will see the speed reading jump wildly up to 60-70 mph even though I may only be going 5 mph in reality.
4.) Phase wires will get pretty warm, and the hub motor will get very hot after a few minutes testing under load.
5.) Hub Motor is 52V1000W (sensorless), controller is 60V1000W (Sensored, but claimed not to be necessary for operation), and battery is 60V (16s6p).
6.) All issues occur when under load.

The only other thing that I have not tried yet is to rearrange the phase wires, currently they are color matching. I had assumed the colors were standardized, but after some research, that appears not to be the case.

Update: Moving the phase wires around had no effect in the stuttering, it remained the same no matter the arrangement. Battery bottoms out at 15Amps, but no more than 50% at 5Amps. Either way, stuttering remains the same when throttling up.
 
simonen said:
I double checked the motor, and it has 20 magnets. The controller setup asks specifically for the number of "steel magnets", I take this to mean that I should enter in the actual number of magnets instead of dividing the value by 2.
It might want that, but almost always they are asking for pole pairs, regardless of terminology they use. Either way, if it doesn't work with 20, try 10. Can't hurt. ;)

Another possibility is that it is using a sensor with it's *own* magnet(s) embedded in the side cover, as some of the geared hubmotors I have here do. On one of them, it's only 1 magnet. On two others, it's six magnets. So you might just have to try different numbers of magnets until it reads correctly.

Some other observations and items worth rementioning:
1.) Battery voltage drops significantly on heavy throttle up, but doesn't go below 2 bars. I don't believe it is hitting the LVC. The battery pack is 16s6p made with Samsung 26F cells. If I am not mistaken, they are each rated for 5A output.
"rated for" and actually being capable of it without severe voltage sag are often completely different things. Some cells I have had were rated for up to 10C+, some even claim 20C (which means 10x or 20x their capacity in Ah as A, so a 5Ah cell at 10C is 50A, or at 20C is 100A), but they sag in voltage by more than a volt each cell, and rapidly heat up, which means they're not really capable of that.

I recommend using cells to a capability where there is minimal or no voltage sag on them when under the load you need them to support--that usually means using more parallel cells. This way you have several benefits: The cells aren't stressed, dont' heat up, and age less quickly so they last longer, and the system has a higher voltage at the same current delivery, meaning you get more watts at the wheel, and it also doesn't drop in voltage to trigger LVC under load as the battery runs low.


2.) The controller lets me set a maximum of 15Amps when set at 60v. I have tried reducing it to 5Amps, but stuttering persists under load when throttling up.

Is it lessened at all, with less current?



3.) When stutter occurs, I will see the speed reading jump wildly up to 60-70 mph even though I may only be going 5 mph in reality.
If it's trying to read the speed from the phase wires, that's probably due to the "misfire" / stuttering; the stuttering could be a timing problem or it could be battery dropout.

If it's trying to read the speed from a separate sensor inside the motor, or whose wires run parallel to the phase wires at any point along the route, then the stuttering whcih causes current surges could be inducing noise into the sensor wire. Often, routing the sensor wire somewhere else fixes this.


4.) Phase wires will get pretty warm, and the hub motor will get very hot after a few minutes testing under load.
5.) Hub Motor is 52V1000W (sensorless), controller is 60V1000W (Sensored, but claimed not to be necessary for operation), and battery is 60V (16s6p).
6.) All issues occur when under load.
That is usually a sign of a wrong phase/hall combination, but you have sensorless operation. Other than that, thsi type of issue generally leads to a battery voltage sag / LVC problem, where it may dip down below LVC so fast and rebound so fast you may not even see it on a digital voltmeter, much less a controller's battery meter whcih may ahve a much slower response. If you have an analog voltmeter, with a needle type display, you *might* see it on that, where it dips in time with the stutter.


The only other thing that I have not tried yet is to rearrange the phase wires, currently they are color matching.
It makes no difference what pattern you use when running sensorless, except that swapping any pair of them will reverse the motor direction.

It would make a difference if you were using hall sensors. (phase / hall combo or wiring combination)
 
I took it for a spin with a 52v 20Ah pack that I use for the Bafang BBSHD, and it stuttered as well.

To answer you question if lowering the amps when the 60v battery was hooked up, no it made no difference in the stuttering.

While I totally understand that the C rating for these cells is probably not high enough, it is nice to know the battery is likely not the cause of this issue.

I have another controller on the way that specifically states the hall connections are not required. So we'll see if that makes a difference.

I let my son take it for a spin yesterday. He weighs about 80 pounds and it stuttered for him as well. The best way I can describe the sound is like when you are driving a manual shift car and you miss the gear and it makes the grinding/fast ratcheting sound. And it is not just the sound, you can feel it in the frame...it is quite startling.
 
simonen said:
I took it for a spin with a 52v 20Ah pack that I use for the Bafang BBSHD, and it stuttered as well.

To answer you question if lowering the amps when the 60v battery was hooked up, no it made no difference in the stuttering.

While I totally understand that the C rating for these cells is probably not high enough, it is nice to know the battery is likely not the cause of this issue.

Probably not, based on that.

I have another controller on the way that specifically states the hall connections are not required. So we'll see if that makes a difference.
It shouldn't, because if your controller required hall sensors, it woudln't spin the motor at all.

But it might just be really crappy at sensorless operation--many are not good at it, some worse than others.

The best way I can describe the sound is like when you are driving a manual shift car and you miss the gear and it makes the grinding/fast ratcheting sound. And it is not just the sound, you can feel it in the frame...it is quite startling.
That's common for hall sensor problems, and battery cutout problems, but can also happen with crappy sensorless controllers. Not many are all that great at starting up under load, but with high power and a small wheel they're usually at least "ok" at it.

It's also a symptom of poor phase wire connection, where one phase has much higher resistance than the others, like a wire that's been broken (like what may have happened at your axle) or a connector that's got poor contact (bad crimp, poor mating of contact surfaces, etc).

Very occasionally, I have seen a controller with damaged FETs (like from a phase-to-phase short) that didnt' actually fail outright, have symptomes like this. If that's the case, a different controller would fix it, as long as the motor or cable damage that caused it is fixed.
 
Well, I am happy to report that the issue was the controller. The new one I got today works perfectly!

Thank you to everyone for your help in diagnosing the issue!
 
Back
Top