King Song S22 motor stator slippage issue: more severe than expected

[level9]

Sounds like ewheels is ahead of the game and not waiting on KS. Just received this. Nice ! I really appreciate this company 

 

Quote

 

Here's another group update since last week's communication: we received a load of S22s at the service center on Wednesday for preshipping testing. 

 

The stress tests involves spinning up the motor & reversing direction at maximum speed. While this procedure is effective at loading the hardware components to [and above] the design limits, the results were mixed, & after discussing the matter with King Song, it seemed a surer bet to replace all the motors in these machines, while we have them on site.

 

Plan of Action:

1) King Song has provide us tracking numbers for 100x replacement motors, these will be arriving next week. 

2) We've hired a couple additional Techs to help take on this project, with a target of completing between 10-15 Wheels/day 

3) If all goes according to plan, these S22 orders will be shipping out over the next two weeks, with the new motors installed. 

 

 

 

[oolong]

That's exactly what I wanted to see... "replace all motors".   Unfortunately Ewheels already shipped mine because I couldn't wait 😅.   Maybe I can get them to ship me one of the new motors.  

[supercurio]

Thank you @level9 for relaying @Jason McNeil's post.

What prompted this originally is a high failure rate of their shipment to the spin test as demonstrated in the original post.
King Song did not share their motor validation method with anybody I know so far, and it might be defunct now anyway.

Then something important was highlighted during one of eWheel's spin test:

• First, the motor failed after a few seconds
• 1 nasty clunk or worse sound on a rotation change
• no scraping whatsoever
• motor failing and taking the mainboard on one of the next rotation change

What does that mean?

1. The previous assumption that scraping sounds can be heard before a short caused by motor slippage is wrong.
2. None of the S22 affected is safe to ride before passing the stress test without any abnormal sound.
3. Or more conservatively, none of the S22 affected is safe to ride before replacing the motor to a July+ production one.

I will update the first post to reflect this new understanding thanks to eWheels testing.

Edited September 9, 2022 by supercurio

[supercurio]

Following eWheels recommendation, owners who asked wheels to be shipped directly to them are running spin stress tests and wheels are failing, like James Jamroz on Facebook, quoting:

Quote

So wheel died on me today. While doing ewheels updated stress test. Full spin, stop. Seems like the mobo.

I was riding it home, and I tried to brake more aggressively, and a few time it felt like something was skipping. So I didn’t trust it anymore. Did the spin test a few more times, then suddenly just died. Resistance in wheel now.

Happy to see that the message is reaching riders and safety recommendations saves them.
I'm hoping for more dealers recommending the spin test now.

[chanman]

If we're seeing a lot of failures seems somewhat wasteful to do the test before the motor is replaced, no need to fry a board when the motor is what needs replacing.

Edited September 9, 2022 by chanman

[techyiam]

7 minutes ago, chanman said:

If we're seeing a lot of failures seems somewhat wasteful to do the test before the motor is replaced, no need to fry a board when the motor is what needs replacing.

If Kingsong's reporting of the motor failure rate falling between 5-15% was accurate, it may not be wasteful. But it doesn't look that way right now.

[chanman]

I'm not sure if EWheels released how many of the wheels they tested and how many failed, but they seem to plan to replace all motors, in which case it definitely doesn't make sense to continue to test with destructive failures beforehand, (to clarify I definitely support testing in general, after replacement)

If you and your dealer are subscribing to the idea that 80%+ of the motors are perfectly fine and that the test rules out any issues developing the future, then it makes sense to try it to avoiding needing to replace the motor I suppose. Possibly that's true but I think I'd prefer EWheels' approach.

Edited September 9, 2022 by chanman

[supercurio]

@chanman agree that some tests will be wasteful, depending on how they are run.

My recommendation is to:

• setup a camera, at least a phone recording the sound
• run the test until hearing a bad or.. really awful sound like I've seen in some example. Then stop here: test failed no need to go further
• if continuing, there's a risk to burn the board.
• send the test results, sound should be sufficient

However even with this approach some boards will fail and that's fair enough, because it might actually be the board and not the motor failing as we've seen for some riders.
In this case the motor was not tested as it should and should be tested again with the new board, if not replaced at the same time depending on its manufacturing date.

@techyiam 5-15%: nowhere near that, not even in the ballpark.

[Jamroxorz]

Hey I signed up to recount my experience for the forum. 

I have about 500km on my s22. Feb Motor.
First time I felt something strange was climbing a hill, I felt a skip.

Started doing stress test. Did it about 6 times. When going back and fourth, I heard the same clunking as @supercuriodid in his stress test. Nothing after that, no scraping, no clunking while riding. 
Then a few times I felt that strange clunk/skip again while torquing, but no issues.

Ran the back and fourth stress test for another 10 times or so. After one of the test I got an error and the wheel stopped balancing. App said hall sensor error, something like that. I don’t remember. After cycling, the wheel was fine, still running.

I took the motor apart, to check if the stator moved. It all seemed in perfect place. No moving of the silicon either.

Today while almost home, I did some harder braking. I felt that strange clunking/skipping. I started to think maybe the tire is slipping on the rim? I decided to stress test again. Did back and forth a few times, same clunk, no issue. Then I did full speed forward and pull back. I didn’t hear any clunks when doing this. After about 4 times the wheel just died. No sounds, nothing. Wheel is hard to push, so mobo fried I guess. 

[Tawpie]

Keep in mind that eWheels allowed all pre-orders to wait until the dust clears so there are a lot of refunds on the line... and I can't be the only one that was/is ready to cancel if the motor thing isn't attacked aggressively.

Blanket replacement of pre-July motors is a good PR move and a cost measure, it's essentially a recall of unshipped wheels so eWheels saves F'dExing back and forth. If I were eWheels, I'd install the replacement motors and still stress test them all if for no other reason than to assure customers that the wheels are "good"er.

The next step is: what happens with the wheels in the field? Stress test by the owner? A new KS originated test? Return to reseller for testing? New motors for all? Replace your motor every year?

 

Edited September 9, 2022 by Tawpie

[oolong]

I plan on doing a "light" test to check for a major issue, check for sounds, and then just ride it lightly while waiting on a new motor.  I don't see a point in running an aggressive test that might destroy the motor and the board when a permanent fix from ewheels is on the way in a few weeks.

[supercurio]

32 minutes ago, oolong said:

I plan on doing a "light" test to check for a major issue, check for sounds, and then just ride it lightly while waiting on a new motor.  I don't see a point in running an aggressive test that might destroy the motor and the board when a permanent fix from ewheels is on the way in a few weeks.

It might sound like a sensible compromise when priorities are driven by the excitement of finally riding a new toy that you've waited or so long, but IMO:

Instead of putting yourself at risk on a wheel of unproven or sketchy reliability, I would definitely recommend testing the shit out of it. If breaks then it didn't deserve to have you on it, if it holds: great!
For anybody owning more than 1 wheel: ride the other one(s). After following closely the V12 mainboard saga: It's never worth the risk. It's just another wheel.

Edited September 9, 2022 by supercurio

[NErider]

Just want to add, Ewheels has always supported me and my wheel with nothing but the BEST service I could wish for. I assume someone from there reads this thread. So I would like to publicly say: 

 

Thank you Ewheels for being awesome! You are are company made of the right people! Great job!

Also, I did give the go ahead to ship mine a day or so earlier than thier notice. I'm not sure if they held it back regardless or not. I'm happy either way. 

[oolong]

1 hour ago, supercurio said:

It might sound like a sensible compromise when priorities are driven by the excitement of finally riding a new toy that you've waited or so long, but IMO:

Instead of putting yourself at risk on a wheel of unproven or sketchy reliability, I would definitely recommend testing the shit out of it. If breaks then it didn't deserve to have you on it, if it holds: great!
For anybody owning more than 1 wheel: ride the other one(s). After following closely the V12 mainboard saga: It's never worth the risk. It's just another wheel.

The thing I fear is "passing the test" but then having the vigorous testing itself causing a small slippage that manifests into a bigger issue after continuous riding.   That was really my mindset for doing a light test with light riding.   Especially when you report that you may not even hear a clunk or scraping noise.

[onkeldanuel]

+ the next Problem with this testing : when is the test considered passed? Since if it passes 3 rounds and wouldve failed at the 4. or 5.?

At this point it would be really helpful what exactly is the root of this to know how much Motors from Batch 1 are affected....all? 50%? Was it a Problem of the assembly itself or a bad part of it....did only some Motors hold barely up and if adding 5% more force on it they would fail too , or if motor is affected it will fail at one of first hard brakings for example??

Edited September 10, 2022 by onkeldanuel

[mikko]

Not sure how much of an effect it has, but consider thermal expansion of the aluminium hub and the steel sheets. If not mistaken aluminium both expands and contracts more than steel, thus on a cold day the problem could become more pronounced maybe? And it might worsen after lots of thermal cycles? Passing a stress test now wouldn't  necessarily mean anything.

Also if we do use heat guns to open glued together parts then having these two parts rely on glue (and maybe a flimsy plastic ring) to hold them in place sounds like a bad idea.... Imho, just replace or offer to replace *all* the motors with ones that have real solid physical cuts between the parts. It's the only right thing to do.

This will be a problem for a very long time once new people start buying second hand "in good condition" S22s without doing any research and then hurt themselves badly because a recall wasn't done properly because it cost too much or something.

[Forwardnbak]

53 minutes ago, mikko said:

This will be a problem for a very long time once new people start buying second hand "in good condition" S22s without doing any research and then hurt themselves badly because a recall wasn't done properly because it cost too much or something.

As it stands anyone that knows will already be expecting cheaper ks22’s in the used market, and more people are only going to become aware.

It’s a bad outcome for those invested in Ks22 hoping for some return on resale to upgrade into their next wheel.

 

 

[onkeldanuel]

KS will hopefully deliver Motors to distributors to solve this...assuming the Problem is solved for Sure july+ (but this needs prove tho  , since there are nearly no Juli Motors tested or?)

[supercurio]

9 hours ago, Jamroxorz said:

Hey I signed up to recount my experience for the forum. 

Thanks a lot for contributing with your experiments!

9 hours ago, Jamroxorz said:

I have about 500km on my s22. Feb Motor.
First time I felt something strange was climbing a hill, I felt a skip.

Started doing stress test. Did it about 6 times. When going back and fourth, I heard the same clunking as @supercuriodid in his stress test. Nothing after that, no scraping, no clunking while riding. 
Then a few times I felt that strange clunk/skip again while torquing, but no issues.

Okay good to know as a starting point.
Were these clunk/skip comparable to Adam's S22 when climbing stairs?
https://youtube.com/clip/Ugkx4ryKlODDx8nZ7P8aQVJ6zfcKVJzHh0o_
https://youtube.com/clip/UgkxlP6V7s7TFEMb7KP_xXSAbcRZDZ7hwI8G

9 hours ago, Jamroxorz said:

Ran the back and fourth stress test for another 10 times or so. After one of the test I got an error and the wheel stopped balancing. App said hall sensor error, something like that. I don’t remember. After cycling, the wheel was fine, still running.

I took the motor apart, to check if the stator moved. It all seemed in perfect place. No moving of the silicon either.

Great methodology here 👏
You're the first to report on this!

What was your point of reference to check for stator rotation: the position of black frame on the side of the phase wires vs the inner spokes? I suppose you didn't open before to put a mark yourself. Interested if you have any picture.

9 hours ago, Jamroxorz said:

Today while almost home, I did some harder braking. I felt that strange clunking/skipping. I started to think maybe the tire is slipping on the rim? I decided to stress test again. Did back and forth a few times, same clunk, no issue. Then I did full speed forward and pull back. I didn’t hear any clunks when doing this.

Good call to test instead of continue riding with uncertainty after observing the clunk/skip again.

I think you highlighted a difference between the test I ran back then and showed on video, where I focused on quicker rotation change that make the clunk appear, versus spinning full speed for maximum inertia and changing rotation direction like Jason has been doing to test his shipment, and that's probably a much tougher variant of the spin test.

9 hours ago, Jamroxorz said:

After about 4 times the wheel just died. No sounds, nothing. Wheel is hard to push, so mobo fried I guess. 

If there was no clunk prior, it is possible that the motor is still fine, but not guaranteed.
I think you might have found a valid stress test for the mainboard failure which is a separate issue, very problematic as well.

One question BTW, at what battery voltage did you run this test?
Let us know what you find when opening the motor. With the elements I have, I can't predict for sure if you'll see disconnected phase wires that shorted the board or not. 50/50 IMO.

Did you film any of this by the way?

[supercurio]

21 minutes ago, onkeldanuel said:

KS will hopefully deliver Motors to distributors to solve this...assuming the Problem is solved for Sure july+ (but this needs prove tho  , since there are nearly no Juli Motors tested or?)

There has been July+ motors tested with the spin test as I've shown, but not yet with the spin test @Jason McNeil and @Jamroxorz completed which lets the motor accelerate to max rpm before changing direction.

So that's still up in the air, I'll ask the dealers if they can reproduce that one.

[Jamroxorz]

5 hours ago, supercurio said:

Thanks a lot for contributing with your experiments!

Okay good to know as a starting point.
Were these clunk/skip comparable to Adam's S22 when climbing stairs?
https://youtube.com/clip/Ugkx4ryKlODDx8nZ7P8aQVJ6zfcKVJzHh0o_
https://youtube.com/clip/UgkxlP6V7s7TFEMb7KP_xXSAbcRZDZ7hwI8G

Yes, I believe it was similar to this, but while climbing a hill, not going up stairs.

5 hours ago, supercurio said:

Great methodology here 👏
You're the first to report on this!

What was your point of reference to check for stator rotation: the position of black frame on the side of the phase wires vs the inner spokes? I suppose you didn't open before to put a mark yourself. Interested if you have any picture.

I just used the black frame yes, and just looked at the wires in general to see if they were tight. Also checked to see if there were any loose wires in the silicon. Everything looked to be in place properly. 

5 hours ago, supercurio said:

If there was no clunk prior, it is possible that the motor is still fine, but not guaranteed.

Yeah, I want a new motor after all the sounds, clunking, and skilled sensation haha

5 hours ago, supercurio said:

One question BTW, at what battery voltage did you run this test?

 

Not sure the voltage exactly, but it would probably be around 80%. Just finished my commute home, so a total of 16km going 50km/hr or so from full charge.

5 hours ago, supercurio said:

Let us know what you find when opening the motor. With the elements I have, I can't predict for sure if you'll see disconnected phase wires that shorted the board or not. 50/50 IMO.

Did you film any of this by the way?

I know I should open the motor, but it’s kind of a PIA. Just want to have a working s22 cause I love the wheel otherwise.

Unfortunately I didn’t film anything. There are pictures of the inside of the motor on the Facebook post. 

[oolong]

Tested my S22 from eWheels today.  Got the clunk.  Motor from May 31st.

[NErider]

9 hours ago, oolong said:

Tested my S22 from eWheels today.  Got the clunk.  Motor from May 31st.

Did they ship your wheel with the original tracking number sent in an email? Or did they issue a new tracking number? 

 

I'm just wondering if my wheel is on its way with a new track number, or if they held it back? 

I'm holding off on messaging them directly as I'm sure they are busy with all this mess. 

[pedroMDK]

Received a 05-28 s22 build from Jason and team at ewheels yesterday, almost 70 miles with no issues so far. 35mph max and only a few curb drops and some easy stairs. Initial thoughts, build quality is better than my gen3 s18. No clicking, buzzing, whirring or other odd noises. Just FYI, YMMV.

[oolong]

5 hours ago, NErider said:

Did they ship your wheel with the original tracking number sent in an email? Or did they issue a new tracking number? 

Original tracking number.  Actually I got two tracking numbers initially and one of them started updating.

5 hours ago, pedroMDK said:

Received a 05-28 s22 build from Jason and team at ewheels yesterday, almost 70 miles with no issues so far. 35mph max and only a few curb drops and some easy stairs. Initial thoughts, build quality is better than my gen3 s18. No clicking, buzzing, whirring or other odd noises. Just FYI, YMMV.

Did you do the test from @supercurio to check for motor slippage?

I don't plan on riding mine extensively until I get a new motor.  I might do some short light riding to get a feel for different pads and accessories while I'm waiting, but nothing beyond that.