S22 Motherboard Failures

[buffs]

1 minute ago, Tawpie said:

Kind of different circumstances though. The second test was not really the stress test, he was trying to get video of the motor noise by free spinning it. Not that I find any comfort at all in that.

yes, and i'd argue that this stress test caused the wheel to fail without him riding, which is a win.

[techyiam]

11 minutes ago, buffs said:

There was another cutout reported today, and I'm trying to do anything to avoid that. We know the OG board has issues. We have no idea if the "new" board is any better. That's my dilemma.

According to your communication with your distributor, in light of recent events related to cutouts, Kingsong has updated the controller. You don't have the updated controller yet, and you also don't want to risk getting hurt. I don't see how you can avoid risking getting hurt if you don't stop riding your S22.

When I rode my V12 before getting the updated board, I was fully aware that I could get hurt. Of course, I mitigated the risks. But should it cutout, I knew I would have a bad day.

[techyiam]

5 minutes ago, Tawpie said:

Kind of different circumstances though. The second test was not really the stress test, he was trying to get video of the motor noise by free spinning it. Not that I find any comfort at all in that.

Then it is even worse. Once passed the stress test, no cutout should occur period, if it was a valid test. Mosfets did failed.

[buffs]

2 minutes ago, techyiam said:

According to your communication with your distributor, in light of recent events related to cutouts, Kingsong has updated the controller. You don't have the updated controller yet, and you also don't want to risk getting hurt. I don't see how you can avoid risking getting hurt if you don't stop riding your S22.

When I rode my V12 before getting the updated board, I was fully aware that I could get hurt. Of course, I mitigated the risks. But should it cutout, I knew I would have a bad day.

yeah, i rode it for 3 days and haven't used it again for 2 weeks.

[techyiam]

1 hour ago, buffs said:

yes, and i'd argue that this stress test caused the wheel to fail without him riding, which is a win.

Depending on how you look at it.

He could have simply stop riding it. And wait for a response from his dealer/distributor. His S22 is still under warranty. 

1 hour ago, buffs said:

yeah, i rode it for 3 days and haven't used it again for 2 weeks.

Your intuition served you well.

Edited August 18, 2022 by techyiam

[Tawpie]

2 minutes ago, techyiam said:

Once passed the stress test, no cutout should occur period, if it was a valid test. Mosfets did failed.

Fail they did. But if the motor locked up due to something mechanical inside (which it sounded like it either did or was about to do), then it's not that surprising that the MOSFETs go poof.

[chanman]

On my S18 I artificially locked up the motor (by connecting the hall sensors incorrectly) the result I got from that was the controller full sending current for a bit, and then getting cut off by firmware. No poofing of any kind.

I guess I'm somewhat conflicted on the best practice here. I value my intact body more than the intact control board, so if I need to squeeze a few more amps to not fall over and risk burning up the board, I guess that's fine, but also I don't want my board frying doing simple stuff later.

Edited August 18, 2022 by chanman

[techyiam]

15 minutes ago, Tawpie said:

Fail they did. But if the motor locked up due to something mechanical inside (which it sounded like it either did or was about to do), then it's not that surprising that the MOSFETs go poof.

Nope, the motor did not appear to be locked up in the video. 

Also, with all that racket the motor is making, does it make sense to do a spin test?

Furthermore, his wheel is still under warranty. 

[Tawpie]

1 minute ago, techyiam said:

Nope, the motor did not appear to be locked up in the video. 

Also, with all that racket the motor is making, does it make sense to do a spin test?

ok. uncle. whatever.

[NSFW]

1 hour ago, techyiam said:

Additionally, this is not an official test from Kingsong.

I'd be shocked if King Song ever shared an "official test" with the public.

That would require acknowledging that there is a problem.  

And we're talking about a company that deletes unflattering-but-accurate posts from their Facebook group. 

If you want to avoid having your S22 fail while you're riding... well, we're on our own to figure out how to prevent that.

 

Edited August 18, 2022 by NSFW

[techyiam]

5 minutes ago, NSFW said:

That would require acknowledging that there is a problem.

Well, maybe not directly.  

 

1 hour ago, buffs said:

I have no details other than what the distributor told me: "we have communicated these concerns to King Song, and King Song has updated the controller hardware to be more robust."

 

[techyiam]

37 minutes ago, Tawpie said:

ok. uncle. whatever.

Are you going to perform the @supercurio special test after you have unboxed it?

Or, are you going to ask your dealer/distributor to delay your order until a later batch which comes with the updated controller? 

[Tawpie]

14 hours ago, techyiam said:

Are you going to perform the @supercurio special test after you have unboxed it?

Nope. Well, probably not. That thing is heavy. I'm old and don't want to wrench my back, defects in my physical plant don't "just get all better" anymore. But I do plan to take it up a very steep hill, stall it, and step off since that seems to be a common failure mode. And remember, the V12 stress test was to try to suss out MOSFETs that either weren't stored properly (IM's analysis) or were a very poor design choice (my unsubstantiated personal opinion, possibly supported by the recall but the recall might have been a PR necessity or something else entirely).

14 hours ago, techyiam said:

Or, are you going to ask your dealer/distributor to delay your order until a later batch which comes with the updated controller?

I'm not sure what I'm going to do, but tomorrow is kind of a drop dead date for getting a tracking number. I have a road trip planned early next month and FedEx is horrible at delivery schedules—if it doesn't ship tomorrow I may not be here to receive it, much less run it around for a week to build at least a smidgen of trust. I only have room to bring along 1 wheel so it has to be trustworthy—and no wheel is that trustworthy until it has earned it. 16X or S18 will do if need be.

As to waiting for a new "more robust" controller, it's tempting, but my distributor is eWheels and based on my past dealings with them I believe they'll have my back either way. So that's not really a problem for me.

But if they can't turn it over to FedEx tomorrow, I will most likely wait until spring to buy something new. T4 will have shown its plusses and minuses (although I can already judge it to be fugly). KS will be at S22 batch3 or so. I don't want a faster 16X (looking at you V12), or a 100 lb behemoth I can't get into a car, or a wheel that requires me to clothe it in a 4-500USD 3D printed shell substitute and may still fall apart the first time it cartwheels down a hillside.

Or I might just buy nothing and happily ride my existing wheels... for the most part, they serve my needs. A new wheel is a want at this point.

(note: that dang S22 I saw the other day went by me again this evening—no clicking, motor turning just fine, rider sporting a cat-ate-the-canary grin... I do still want my oompaloompa)

UPDATE: had a quick discussion with eWheels and they have no problem delaying shipping the oompaloompa—I asked for the delay only because I won't be able to receive it due to my road trip (to be taken with the S18). They're the best. This will still be from the eWheels batch1 so if anyone has extra good karma please feel free to send it my way.

Edited August 18, 2022 by Tawpie
add update

[wstuart]

2 hours ago, Tawpie said:

Nope. Well, probably not. That thing is heavy. I'm old and don't want to wrench my back, defects in my physical plant don't "just get all better" anymore. But I do plan to take it up a very steep hill, stall it, and step off since that seems to be a common failure mode. And remember, the V12 stress test was to try to suss out MOSFETs that either weren't stored properly (IM's analysis) or were a very poor design choice (my unsubstantiated personal opinion, possibly supported by the recall but the recall might have been a PR necessity or something else entirely).

I'm not sure what I'm going to do, but tomorrow is kind of a drop dead date for getting a tracking number. I have a road trip planned early next month and FedEx is horrible at delivery schedules—if it doesn't ship tomorrow I may not be here to receive it, much less run it around for a week to build at least a smidgen of trust. I only have room to bring along 1 wheel so it has to be trustworthy—and no wheel is that trustworthy until it has earned it. 16X or S18 will do if need be.

As to waiting for a new "more robust" controller, it's tempting, but my distributor is eWheels and based on my past dealings with them I believe they'll have my back either way. So that's not really a problem for me.

But if they can't turn it over to FedEx tomorrow, I will most likely wait until spring to buy something new. T4 will have shown its plusses and minuses (although I can already judge it to be fugly). KS will be at S22 batch3 or so. I don't want a faster 16X (looking at you V12), or a 100 lb behemoth I can't get into a car, or a wheel that requires me to clothe it in a 4-500USD 3D printed shell substitute and may still fall apart the first time it cartwheels down a hillside.

Or I might just buy nothing and happily ride my existing wheels... for the most part, they serve my needs. A new wheel is a want at this point.

(note: that dang S22 I saw the other day went by me again this evening—no clicking, motor turning just fine, rider sporting a cat-ate-the-canary grin... I do still want my oompaloompa)

I'm in the exact same boat as you!!!

What do I do everyone? - do I pay the balance on my S22 pre order, or do I wait?

I'm only going to use this for lower speed offroad trail riding, so I feel my risk is a little less.  I won't be going 30mph on this any time soon. 

Please give me an answer so I can blame you all later :).....

[techyiam]

8 minutes ago, Tawpie said:

I do plan to take it up a very steep hill, stall it, and step off since that seems to be a common failure mode.

This makes more sense than performing an artificial test that had not been approved by Kingsong.

I believe stalling is very stressful for the euc. I personally won't go that far.

 

7 hours ago, Tawpie said:

I only have room to bring along 1 wheel so it has to be trustworthy—and no wheel is that trustworthy until it has earned it. 16X or S18 will do if need be.

As much as it is so exciting to be able to ride your S22 during your trip, especially after you have waited so long and have spent a pretty penny, logically, it won't be prudent to bring the S22, IMO.

I see two big negatives that can ruin your trip. Firstly, is that you can get seriously hurt due to a faulty S22. Secondly, if it breaks down, you have no other euc to ride. In light of recent reported failures, IMO, the odds are too high for your specific circumstances. If you don't bring the S22, you won't have to burden yourself with this unnecessary worry. 

IMO, it would already be bad if the failure reports were from the wild, but it is worse than that. Some of the reports are from the regular members here, who have experienced the failures first hand themselves. 

 

7 hours ago, Tawpie said:

Or I might just buy nothing and happily ride my existing wheels... for the most part, they serve my needs. A new wheel is a want at this point.

(note: that dang S22 I saw the other day went by me again this evening—no clicking, motor turning just fine, rider sporting a cat-ate-the-canary grin... I do still want my oompaloompa)

After my personal experience with my V12, I would rather not keep a dead or untrustworthy euc in my house. If the wait for new parts takes too long, the self draining battery packs can get either too depleted or too out of balance. Just a pain. And you don't get to ride it anyways. 

If you can tough it out, cancel your order if you are allowed, and if not, delay your order until the next batch with the updated board.

But then it is just me. But if you don't mind the hassle, are a risk taker and can risk injuries, then ....

[supercurio]

On 8/14/2022 at 9:08 PM, supercurio said:

the root cause is the motor self-destructing, but the result is still a motherboard failure

Maybe it was not clear enough so,

13 hours ago, techyiam said:

I don't follow your logic.

I can understand if everyone who performed and passed your recommended V12-style stress test on their S22 don't subsequently cutout in the field. Then it would make some sense to perform this test. But this isn't the case.

You are the only person who performed this test. It passed the first time. But at a later date, when you performed the test again, it promptly killed your board.

You haven't demonstrated this is a valid test in any capacity. 

Additionally, this is not an official test from Kingsong.

Okay I see what you mean.
Since there is some misunderstanding, I will try to clarify.

The wheel passed 3 series of spin stress test out of the box, and once riding it, I've been very impressed by the mainboard flawless power delivery: good safety margin beyond beeps, handled hours of riding single tracks over roots, jumps, drops, stairs without flinching. I was able to pedal dip briefly it only with extreme braking test and over-torquing beyond 45 kph after a maximum acceleration... on an incline - mind you.
What gave me the confidence progressively ride it aggressively was that it passed the stress tests first.

Now, the motor failing is another issue. I'll give more details from latest todays learnings but it appears to be confirmed that some motors have hall sensor issues.

So I want to be super clear: my video captured a motor failure, recorded in order to demonstrate the scraping sound issue without me risking harm. There would be no point 
This motor failed so bad (we will know how once it's opened) that it managed to kill the mainboard.
This is not the stress test that killed my board, it is a defect in the motor.

I hope that clears this up.

 

16 hours ago, buffs said:

I've asked my distributor about getting a replacement, and they will provide me with a "more robust" controller at cost.

One of the update I have today is that a friend in China received a replacement mainboard, part of the ongoing domestic recall. He inspected closely the board and compared to the original he had in his wheel. Findings are surprising:

The replacement board: 

• is visually identical to the old board
• EDIT: is v2.3 2022-01-05 instead of v2.1 2021-11-25
• components checked one by one: no difference found

Until a new mainboard revision is seen in the wild, I find unlikely that a "more robust", updated, different controller is available.
There's no proof it doesn't exist either, granted.

EDIT: My S22 unit with 2205 (May) dead motor already came with v2.3.

@buffs what's the revision of your S22 mainboard?

My friend thinks that some boards might have been identified to use bad components, or at least that it could be KS's working theory for the board failures.

Edited August 18, 2022 by supercurio

[buffs]

8 hours ago, Tawpie said:

That thing is heavy. I'm old and don't want to wrench my back, defects in my physical plant don't "just get all better" anymore. But I do plan to take it up a very steep hill, stall it, and step off since that seems to be a common failure mode.

I'm in the same boat. I don't want to risk aggravating  back issues (the reason I want a great suspension wheel), so I was going to ask if anyone knew a way to safely stress test the controller but not my back. Your idea sounds like the next best thing. I'll try to find a steep grassy hill.

[techyiam]

1 hour ago, supercurio said:

I hope that clears this up.

Thank you for detailed explanation. 

Do you know of anyone else that has a motor as damaged as yours? Do you think it is widespread thing, or just an outlier? 

[wstuart]

I wonder if the motor damage is something that grows over time?  "Forwardnback" had similar sounds in his motor but stopped riding it.  If he kept riding bit or performed a stress test maybe he would have a result similar to supercurio

12 minutes ago, techyiam said:

Thank you for detailed explanation. 

Do you know of anyone else that has a motor as damaged as yours? Do you think it is widespread thing, or just an outlier? 

[supercurio]

16 minutes ago, techyiam said:

Thank you for detailed explanation. 

Do you know of anyone else that has a motor as damaged as yours? Do you think it is widespread thing, or just an outlier? 

👍
Yes there are are several classes of motor issues discussed between dealers, it's an ongoing issue developing day by day, seems to get more frequent than mainboard.
BMS are giving troubles too.
I'm writing a post about S22 motor issues details now, it's a complex topic.

By the way I only hope to avoid people getting injured and recommend the spin test, with absolutely 0 authority.
My thinking is that since my unit passed 3 strenuous ones, any another unit should pass as well and not passing would highlight a defect that would later result in a crash like we've seen.
But there's no doubt that it requires some general strength and fitness and I respect each individual choice. I found it much easier to execute than on V12 thanks to the good grabbing points the S22 have, offering superior mechanical advantage.

Edited August 18, 2022 by supercurio

[conecones]

21 minutes ago, buffs said:

I'm in the same boat. I don't want to risk aggravating  back issues (the reason I want a great suspension wheel), so I was going to ask if anyone knew a way to safely stress test the controller but not my back. Your idea sounds like the next best thing. I'll try to find a steep grassy hill.

If you want to do a spin test without lifting, you can just put the wheel in a stand where it's suspended by the pedal hangers - turn the unit on and rock it back and forth as it's spinning.

But understand that no test is definitive, so I suggest doing what the local riders here have all done (which originally started as a joke), and that is to buy one or more spare controllers right away. It's assumed that if you intend to ride this wheel aggressively, the board will likely fail at some point since the replacement boards that shipped out a month ago look exactly the same as the originals. One of the local wheels that failed back in July was already abused for a few weeks before the board died, so there's a lot of uncertainty for any reliable testing to weed out bad boards. At least the spares are not crazy expensive.

[supercurio]

Correction on what I wrote earlier regarding mainboard versions:

Further look to the mainboard showed that my friend's replacement board revision number changed from:
- v2.1 from 2021-11-25
- v2.3 from 2022-01-05

My S22 with 2205 (May) motor already had v2.3. Board very felt strong, unlike the weaknesses described in earlier reviews.

My questions to anyone with a dead board: was it a v2.1 revision?

[techyiam]

1 hour ago, wstuart said:

I wonder if the motor damage is something that grows over time?  "Forwardnback" had similar sounds in his motor but stopped riding it.  If he kept riding bit or performed a stress test maybe he would have a result similar to supercurio

I see what you are saying. But it is not easy to answer that question because @supercurio is suggesting that there could be more than one type of failures occurring with the motor. 

 

Edited August 18, 2022 by techyiam

[Rawnei]

1 hour ago, supercurio said:

Correction on what I wrote earlier regarding mainboard versions:

Further look to the mainboard showed that my friend's replacement board revision number changed from:
- v2.1 from 2021-11-25
- v2.3 from 2022-01-05

My S22 with 2205 (May) motor already had v2.3. Board very felt strong, unlike the weaknesses described in earlier reviews.

My questions to anyone with a dead board: was it a v2.1 revision?

Wouldn't be surprised if those are the boards that was also in the pre-prod units and which got some kind of redesign during the delay period but if so puzzled why they would sell those at all.

[FunTech4Real]

My controller died from putting the wheel down on its side too fast which resulted in the tire spinning when hitting the grass which caused the wheel to do about a half rotation... crazy thing is the charger started smoking after plugging it in to see if it would do anything. I made a video of it (caught the charger smoking on video):

https://www.facebook.com/75401032/videos/1455728894950625/

 

The controller is version 2.3 (2202-01-05) and the replacement I received is the same date and version, both from ewheels

Edited October 1, 2022 by sevin7