Jump to content
dantaylr

Recovering after an S18 crash, would love some thoughts

Recommended Posts

Posted (edited)

I'm posting here because I would love some thoughts from the community to help strengthen my relationship with these wheels.

After putting over 4K miles on my Ninebot Z10 and loving it, I purchased an S18 in hopes of not face planting as often when hitting these huge potholes in Seattle. Shortly after falling in love with the S18, I send the thing into an accident by over-accelerating and inducing a strong pedal dip. This destroyed the trolley handle and ripped me to shreds. Diagnostics showed the unit was working fine after this. After a week of bandages I hopped back on to test if it was working. It appeared to drive fine and then suddenly locked up without warning and sent me into a crash at about 15mph. It was a brick afterwards. Huge props to Rev Rides for taking my wheel in, examining the issue and replacing the board with KS's latest one. It turns out the first accident damaged a capacitor on the board which caused it to fry later when I was testing the unit again.

Now (again huge thanks to Rev Rides) I have a beautiful working fixed wheel and I am scared to death to use it. I would love some thoughts on these units so that I can stop worrying about it. Is there any chance something like a lock-up could happen again? Does anyone know if these boards typically have backup or redundant circuits that kick in if something fries? Should I be worrying about any water getting to the main board while riding in Seattle? I'd love to stay safe and sane as I'm tired of nursing injuries and replacing clothing.

Thanks.

Edited by dantaylr

Share this post


Link to post
Share on other sites

as an s18 rider I'm grateful that you posted about this.

I've gone about 300 miles now without cut outs but that's certainly a larger risk with a powerful motor and battery with 40% less power in the batteries than the original 18xl this board was designed with...

Would you give more details about the original cutout? Speed? Incline? Riding weight? Were you passing someone?

More information will help people like me not to crash.

For getting over the initial fear of getting back on, maybe try full gear in an open grassy area? Falls on the grass are not as bad. That's where I try new stuff with my s18.

Share this post


Link to post
Share on other sites
1 hour ago, meepmeepmayer said:

You overleaned the wheel, and the crash damaged a part which made you crash again. So far, no reason to distrust the wheel. Crashes that are pure unexpected hardware failures with no prior history are extremely rare, unlike the typical overlean crashes.

So the only question is: is the S18 good for going 50kph?

Personally, for a fast wheel like this,  I would add a 10kph error margin just to be sure. So if you want to go 50kph, the wheel must be able to go 60kph without any problems or even just alarms or tiltback. So the S18 would be a mostly-40kph wheel for me. Others might disagree with that.

That's the key question here. Were you simply too fast, or were you riding reasonably but the wheel wasn't strong enough for it?

@meepmeepmayer@~~~o~~~ I really appreciate both comments. I overlooked the battery to motor ratio when purchasing this as I assumed it was better than a three-year-old z10 with a slightly damaged tire. 

IMO the Z10 is gutsy. Gutsy enough that I had never learned the terms for pedal dip and cutoff in three years of riding, and constantly throttled the thing. When I got the S18 I gulped down Kuji's acceleration test video and definitely throttled it as hard as I could, which is naive on my part. I weigh about 135 and managed to crash it at around 25mph within a block. It bounced and rolled like runaway tire which ripped the trolley handle out. The fact that it bumped the "crown" suggests the board failure later on. Seems like an unfortunately vulnerable place for a board.

Share this post


Link to post
Share on other sites
2 hours ago, GoGeorgeGo said:

I don't want to turn you off of EUC but you have to always be aware that a crash is a very real possibility on any vehicle.  With an euc this means falling to the ground 100% of the time.  

Machines can fail, yours sounds like a freak circumstance with the board being faulty, but the s18 is also a brand new wheel so errors should be expected.  From a safety standpoint buying the newest wheel on the market is a bad move.  And thats ignoring the motor power to battery saize dilema the machine has (battery is to small for the motor causes to much sag)

The inmotion v11 does have a built in battery redundancy that will allow the wheel to hold the rider up on one battery, but to my knowledge there are no wheels with a redundant control board capabale of replacing a fried board on the fly.

The important thing is to always gear up and to learn the limits of the machine.  The s18 will have lower performance limits than other wheels so you just need to be aware of that and not push the wheel past its capabilities.  

This is the reason people have fallen in love with the huge battery in the newest wheels, the veteran and the monster pro and the EX.  They provide huge power overhead that keeps you safe, its extremely hard to sag a battery that huge enough to cause cutout.  Thats why they can reach speeds of 50 mph.  

For example the MSP has a 1800wh 100v battery and a 2000w motor.  The s18 has a 2200w motor but only a 1040wh 84v battery . Its almost half the power available for a larger motor

Really good to know. I have mixed feelings for sure as these things are basically ground helicopters in terms of points of failure, but they're so dang convenient!

Awhile back when cleaning my Z10 I managed to jam a towel into the wheel while it was on which fried the control board. I've heard this can happen from too much stress. Do you happen to know if most control boards fail at the same place? Would it be unreasonable for a manufacturer to build a sort of fuse-box scenario into the control board? You can probably tell I'm a designer and not an electrical engineer 😬

Share this post


Link to post
Share on other sites
1 hour ago, dantaylr said:

Would it be unreasonable for a manufacturer to build a sort of fuse-box scenario into the control board? You can probably tell I'm a designer and not an electrical engineer 😬

Shouldn't it be possible in the wheel software alone? the wheel knows how many amps are going out and it should know that the speed has stayed close to zero so that should indicate there's an issue. It could then stop the power. I'm sure there's some edge condition that means some rider somewhere will end up flat on there face but hey ho.

Share this post


Link to post
Share on other sites
Posted (edited)
3 hours ago, dantaylr said:

@meepmeepmayer@~~~o~~~ I really appreciate both comments. I overlooked the battery to motor ratio when purchasing this as I assumed it was better than a three-year-old z10 with a slightly damaged tire. 

I'll echo what @meepmeepmayer is saying, 3P isn't enough to sustain a 2200 Watt nominal 84V motor supposedly able to do 31mph, every such 84V 31mph+ wheel has typically had at least a 6P battery (MSX, 16X), so the specs alone show you that you can't push it as hard as say a 100v Gotway/Vet, etc. Combine this with the fact that King Song tends to be on the conservative side with their amp spikes in general, erring on the "safe" side of things.

 

3 hours ago, dantaylr said:

The fact that it bumped the "crown" suggests the board failure later on. Seems like an unfortunately vulnerable place for a board.

Over my going-on 6 years in EUC, I've learned never to trust these companies, doesn't matter which one, on engineering design, even if their last model was flawless (if you can't tell, I've been burned by this in the past). Hence, I don't pay attention to the flashy YouTube clickbait ride review videos, which often tells you very little (all rider experiences differ dramatically with skill set; can't judge what the reviewer says unless they ride Exactly like you do). I make my decisions on an actual test ride, if I can get on a demo wheel, and I also am always waiting for a proper teardown, so my eyeballs can scrutinize the crap out of it.

On eyeing the S18 teardown, the top board encasing did seem like red flag to me IMHO. I'm never a fan when they cram the board in tight like that, a.) because of poor cooling, and as in your case, b.) poor protection. Capacitor damage on an EUC board is very rare, but if the caps are so close against a not so super robust looking plastic top shell, I'm not surprised your's got hit there.

I guess you just can't ride your S18 as hard as you thought you could, as the YouTube promo would have you believe. Your's is far from such damage stories I've heard so far of new S18 owners, locally and on social media.

Edited by houseofjob

Share this post


Link to post
Share on other sites
32 minutes ago, houseofjob said:

3P isn't enough to sustain a 2200 Watt nominal 84V motor supposedly able to do 31mph, every such 84V 31mph+ wheel has typically had at least a 6P battery (MSX, 16X), so the specs alone show you that you can't push it as hard as say a 100v Gotway/Vet, etc.

4p seems to be fine, especially with the new 21700 cells. MSP, RS, Nikola 1800Wh are all that. Wasn't a problem with the Tesla (for example, 4p 18650) either. Of course the bigger battery beasts (Sherman!) are even better. 

So I say 3p seems not enough, and KS firmware might rather limit too early maybe, who knows.

Anyways, this is a first world problem. Ride "normally" and the S18 is a very fine wheel (if you can't tell, I wouldn't mind one, looks nice and is comfortable:efefd0f676:).

Share this post


Link to post
Share on other sites
Posted (edited)
9 minutes ago, meepmeepmayer said:

4p seems to be fine, especially with the new 21700 cells. MSP, RS, Nikola 1800Wh are all that.

They make 84V versions of the RS & MSP(?). Isn't the 84V Nikola still 2100Wh 18650(?) 

9 minutes ago, meepmeepmayer said:

Wasn't a problem with the Tesla (for example, 4p 18650) either.

Ah I always forget the hideous Tesla!

Edited by houseofjob

Share this post


Link to post
Share on other sites
22 minutes ago, houseofjob said:

They make 84V versions of the RS & MSP(?).

No:) But 1800Wh is 100V and 4p (with 21700s) and these wheels have that battery configuration.

Nobody would argue that an MSP e.g. is too easy to overlean. So it looks like 4p is where the KujizoneTM starts. Of course more is always better, but what else is new:whistling:

26 minutes ago, houseofjob said:

Isn't the 84V Nikola still 2100Wh 18650(?) 

Yep, making it 8p. But there's the 1800Wh 100V too.

Share this post


Link to post
Share on other sites
On 10/7/2020 at 7:32 AM, dantaylr said:

It turns out the first accident damaged a capacitor on the board which caused it to fry later when I was testing the unit again.

Cap failure (usually seen as burnt/evaporated legs) is usually a consequence of MOSFET short circuit, so I suspect it didn't failed from first fall.

10 hours ago, mike_bike_kite said:

Shouldn't it be possible in the wheel software alone? the wheel knows how many amps are going out and it should know that the speed has stayed close to zero so that should indicate there's an issue. It could then stop the power.

Fuses are meant as a last resort protection in case of short circuit condition. Normally, software tries to limit current and this is the cause of what we call "cut out", for example when you push your wheel too much on acceleration. But even software protection may fail, especially when short, high current spikes cause MOSFETs to fail. Fuse is here to protect entire circuit from catastrophic failure due to excessive current flowing. This is why many Gotway motherboards "burned to the ground" - they are lacking any fuses. Same was with Shermans, until Veteran added fuse in battery circuit.

Share this post


Link to post
Share on other sites
Posted (edited)
2 hours ago, Seba said:

Cap failure (usually seen as burnt/evaporated legs) is usually a consequence of MOSFET short circuit, so I suspect it didn't failed from first fall.

To be fair it depends on what the cap looked like. Maybe the OP can post a pic. Caps can age, swell and fail with zero other issues on the board. This then leads to a wheel which still appears to work perfectly but can cut out at whatever speed it feels like. My friends wheel did it at 15mph. Everything on the board looked perfect except one of the caps had a swollen end and appeared a little darker. It still balanced perfectly at a standstill before he swopped out the board.

Quote

 until Veteran added fuse in battery circuit.

I am wondering whether it's worth putting a fuse into my MSX. I am certainly reluctant to add any points of failure, but in the event of a catastrophic fault, I am mindful that the batts will keep pumping out power until a circuit is totally broken. By that time the wheel could be well alight :shock2:

Edited by Planemo

Share this post


Link to post
Share on other sites
1 minute ago, Planemo said:

To be fair it depends on what the cap looked like. Maybe the OP can post a pic. Caps can age, swell and fail with zero other issues on the board. This then leads to a wheel which still appears to work perfectly but can cut out at whatever speed it feels like.

OP wrote that wheel locked, so I assume he meant that motor didn't rotated freely - this is almost always caused by MOSFET failure. Failed cap won't cause MOSFET to fail, so IMO shorted MOSFET caused a very high current draw from caps causing legs to overheat and melt.

23 minutes ago, Planemo said:

I am wondering whether it's worth putting a fuse into my MSX. I am certainly reluctant to add any points of failure, but in the event of a catastrophic fault, I am mindful that the batts will keep pumping out power until a circuit is totally broken. By that time the wheel could be well alight :shock2:

Fuses are relatively reliable protection devices. King Song is using fuses for a long time and we don't see reports of accidents caused by fuse failure not caused by overcurrent. So I think that adding correctly-sized fuse to Gotway wheels would be a good idea.

Share this post


Link to post
Share on other sites
1 minute ago, Seba said:

OP wrote that wheel locked, 

Ahh sorry yes I missed that. So, FET failure as you say :)

1 minute ago, Seba said:

Fuses are relatively reliable protection devices. King Song is using fuses for a long time and we don't see reports of accidents caused by fuse failure not caused by overcurrent. So I think that adding correctly-sized fuse to Gotway wheels would be a good idea.

OK cool....so I wonder what size I should go for....maybe the same as the Veteran or maybe that would be too excessive...hmm..I certainly don't wanna go too small!

Share this post


Link to post
Share on other sites
2 minutes ago, Planemo said:

OK cool....so I wonder what size I should go for....maybe the same as the Veteran or maybe that would be too excessive...hmm..I certainly don't wanna go too small!

I think using the same fuse as used in Sherman is a good idea. For sure you shouldn't go less that 80-90 Amps. It's also important to check voltage rating of such fuse, because if fuse voltage rating is too low, it may keep conducting because of plasma arc.

Share this post


Link to post
Share on other sites
Posted (edited)

S18 main board does not include fuses.  Reviewing current market situation S18 is in fact the only wheel on the market with real suspension with damping - it use bike type air shock with proper main air chamber, negative air chamber, rebound adjustment and lock function.  None of the competitors (Inmotion V11 nor just arriving Gotway EX) offer such solution.

In my opinion each of those 3 wheels have major flaws - Inmotion - only pedals suspension and easy to break parts in suspension "saddle" in case of the fall, and EX - huge weight and small suspension travel of the yet unknown damping type of the whole body (but probably similar in construction to V11) and S18 with nice suspension, futuristic look but LACKING the safety of the minimum 20s4p solution.

I was afraid of the 20s3p configuration - so I have found safe solution - it is relatively easy to add "fourth p" to the S18 as the battery construction on this wheel is modular - each pack contains internal electronics (BMS)  and in fact this wheel has 3 independent batteries - 2 pcs of 84V in front build from 20 pcs of 2170 cell and 1 in the back, build from 2 42V packs of 10pcs of 2170 cells connected in series. I have decided to add 2 additional pieces of 42V original batteries connected in series - creating total setup of 20s1p + 20s1p + (10s2p+10s2p)  this in total offer 84V 1440Wh which for my riding style is enough (I have observed that in usual ride type I do not consume more than 1000Wh, which leaves me with safe margin reserve of 440Wh, and my riding speeds are up to 42km/h with occasional 45km moments.  Also - moderate accelleration / breaking characteristics used to work with 3p setup allows addtional safety of the "forth p" - and as those are in fact 2170 cells - unless it is really cold we can slightly compare it to 20s5p setup build from 18540 cells.

S18 does not accelearate nor break as good as MSP does, but additional safety coming from suspension in my opinion is greater than the decrease in performance.

For more about the wheel - and upgrades needed to make it "road ready" please consult topic below, BTW - work is almost done - the last thing is nylon pedals with stainless pins which shall be ready next week.

 

I always ride in full gear BTW

 

 

ready1.jpg

Edited by Lukasz

Share this post


Link to post
Share on other sites
1 hour ago, Lukasz said:

S18 main board does not include fuses.

I suspect that they are present on BMS board. For example in KS-18XL there are fuses on BMS board and also on mainboard. While having two fuses in series is an overkill and design flaw, in case of S18 fuses locates on BMS board would be enough.

Share this post


Link to post
Share on other sites

As the batteries in S18 are separate units each with own BMS  - fuses in each battery - especially with additional battery in the system seems to be OK. 

Short circuits may happen during the major crash - so fuse can protect the wheel from fire. 

Share this post


Link to post
Share on other sites
21 hours ago, Planemo said:

I agree, despite it's age the Z10 is a strong wheel. Never had mine give me a whiff of being underpowered, and it still feels as strong (disregarding top speed limits) as my 100v MSX. IMO I think the Z10 motor is actually capable of quite a bit more speed (the MRN mod proves this to some extent, but I think it could do even more) but I believe that Ninebot are probably the most cautious designers out there, maybe even more so than Kingsong/Inmotion. As such the 'low' 28mph limit is so far within the Zeds comfort zone that even when beeping it's head off you would have to be extremely unlucky to overlean it. I think it's still one of the safest wheels out there.

In terms of product design the thing is just a beautiful TANK. It feels much more clunky to stand on in comparison to the S18 though. Definitely feels like an earlier innovation in terms of ergo.. but damn. I'd like another wheel made like that again.

Share this post


Link to post
Share on other sites
18 hours ago, Planemo said:

To be fair it depends on what the cap looked like. Maybe the OP can post a pic. Caps can age, swell and fail with zero other issues on the board. This then leads to a wheel which still appears to work perfectly but can cut out at whatever speed it feels like. My friends wheel did it at 15mph. Everything on the board looked perfect except one of the caps had a swollen end and appeared a little darker. It still balanced perfectly at a standstill before he swopped out the board.

I am wondering whether it's worth putting a fuse into my MSX. I am certainly reluctant to add any points of failure, but in the event of a catastrophic fault, I am mindful that the batts will keep pumping out power until a circuit is totally broken. By that time the wheel could be well alight :shock2:

Rev Rides took care of the fix so the only thing I saw was a slightly darker spot on the inside of the shell which I had them replace. What you're saying sounds scary. Is this something that can happen over time with normal use? Is there any way to pre-diagnose something like this? I hate the idea of constantly being ready to be bucked off.

18 hours ago, Seba said:

OP wrote that wheel locked, so I assume he meant that motor didn't rotated freely - this is almost always caused by MOSFET failure. Failed cap won't cause MOSFET to fail, so IMO shorted MOSFET caused a very high current draw from caps causing legs to overheat and melt.

Fuses are relatively reliable protection devices. King Song is using fuses for a long time and we don't see reports of accidents caused by fuse failure not caused by overcurrent. So I think that adding correctly-sized fuse to Gotway wheels would be a good idea.

The wheel did lock. It was like I was being tripped. No ragdoll behavior, it was like someone hit the brakes. No beeps or anything. The unit didn't turn on afterwards.

Share this post


Link to post
Share on other sites
1 hour ago, dantaylr said:

What you're saying sounds scary. Is this something that can happen over time with normal use? Is there any way to pre-diagnose something like this?

Electrolytic capacitors dry out over time. How long this takes is dependent on time and use/heat. As the electrolyte gradually reduces the caps become less effective, until they eventually fail/explode. The caps on the MSX I saw hadn't exploded, but one was clearly swollen and it dumped the rider. The worst thing was that after getting back up, the wheel appeared fine and balanced perfectly on a slow (5mph) test ride so he got on it again and then it dumped him again at 15mph. 2 lots of injuries (and fairly nasty ones at that) in the space of 20 mins. The best thing you can do is keep an eye on them every 6 months or so for any signs of bulging/swelling/leakage/dark deposits. The MSX I saw had done around 6k miles (albeit hard riding), so I will either swop out the caps @ 4k and/or replace the board. That said I think @Mike Sacristan has about 12k miles on his MSX? Either way, caps drying out is definitely a 'thing' so it's a matter of when not if, even on a wheel that isn't used hard (maybe that says something about Mike lol).

The Nichicon brand caps used on the MSX are of a good quality, but ALL caps will fail eventually. See here for a little info on what dodgy caps look like:

https://www.robotroom.com/Faulty-Capacitors-1.html

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...