[EcoDrift.ru] Ninebot Z10: A Diagnosis of Common Issues

[Leonardo Deleon]

I disconnected battery 2 that was causing the issue. Will do a range test with just one battery at 35kmh

[Ziiten]

13 hours ago, Leonardo Deleon said:

I disconnected battery 2 that was causing the issue. Will do a range test with just one battery at 35kmh

Interesting to see if the problem still persists. If not, then maybe all that is needed is a new battery.

[Leonardo Deleon]

1 hour ago, Ziiten said:

Interesting to see if the problem still persists. If not, then maybe all that is needed is a new battery.

It acts even more crazy with just one battery, I could ride just 10km, had to reconnect. Now the battery is unbalanced on the opposite direction.
I bought the 16X, cause I rely on a wheel for commuting 13km everyday

[Ziiten]

22 minutes ago, Leonardo Deleon said:

It acts even more crazy with just one battery, I could ride just 10km, had to reconnect. Now the battery is unbalanced on the opposite direction.
I bought the 16X, cause I rely on a wheel for commuting 13km everyday

So, you got the high cell differential warning also with the battery half that hadnt previously shown low cell voltage?

Would be interesting to hear your thoughts of the KS 16X as a previous Z10 rider.

[Leonardo Deleon]

On 8/26/2019 at 9:47 AM, Ziiten said:

So, you got the high cell differential warning also with the battery half that hadnt previously shown low cell voltage?

Would be interesting to hear your thoughts of the KS 16X as a previous Z10 rider.

I got a different error when only battery 1 was connected, when the battery got low. I turned off the wheel and it didn't turn back on, so I connected battery 2 to be able to get to work. I like the z10, is so sad that it has this problem. I'm gonna keep it around for a bit. Let me know if you find what's happening to yours.

[Damarafaka]

After experiencing same issue of unbalanced battery/cells when reaching 1000km on my Z10, I still had the option in the ninebot app so managed to update to BMS v1.1.7. After riding a few kms, alarm has not repeated. I will let you know if it has any definitive effect on battery errors/alarms .

[Ziiten]

4 hours ago, Damarafaka said:

After experiencing same issue of unbalanced battery/cells when reaching 1000km on my Z10, I still had the option in the ninebot app so managed to update to BMS v1.1.7. After riding a few kms, alarm has not repeated. I will let you know if it has any definitive effect on battery errors/alarms .

I've had the bms-version already for quite a while. I cant remember when I updated to 1.1.7 and if the problems started after that.

[Leonardo Deleon]

5 hours ago, Ziiten said:

I've had the bms-version already for quite a while. I cant remember when I updated to 1.1.7 and if the problems started after that.

After a short 10km ride today the wheel seems to not be charging anymore. I think something in the battery just died.

[Damarafaka]

On 8/24/2019 at 3:46 PM, RoberAce said:

 

Sorry to hear that Ace, always saw you as the ultimate tester of the Z10 in order to find the extreme limits of this wheel. Hope you can get it fixed and are back in business as soon as possible. Will miss your great Z10 adventures.

[Planemo]

This is indeed very sad to see. I too saw Ace as one of the hardcore Z10 testers and to see the wheel fail in the same way as Hsaing's and many others just makes identifying what is going on even harder. Both wheels had many miles on them too which makes the cause even harder to pinpoint. I think the Z should have been able to handle those pendulums for that amount of time as well.

Is the only correlation that both were early wheels?

Still seems weird that Ace's wheel can do so many miles (10k?) before doing this. Makes me think that something else is going on.

Thankfully it didnt fail on one of his numerous crazy fast runs.

[KAY GEE]

13 hours ago, Planemo said:

This is indeed very sad to see. I too saw Ace as one of the hardcore Z10 testers and to see the wheel fail in the same way as Hsaing's and many others just makes identifying what is going on even harder. Both wheels had many miles on them too which makes the cause even harder to pinpoint. I think the Z should have been able to handle those pendulums for that amount of time as well.

Is the only correlation that both were early wheels?

Still seems weird that Ace's wheel can do so many miles (10k?) before doing this. Makes me think that something else is going on.

Thankfully it didnt fail on one of his numerous crazy fast runs.

Sorry, to disappoint you, the Z10 is a beautiful designed wheel, but the electronic components are nothing to right home about, I know some hardcore Z10 fanatics that have given up on their beloved Z10 and moved on to something else, Hsaing rides a Gotway Monster now, it's such a shame that Ninebot does not address this issues. For your information, I ride a Z6, but I only ride it once in two weeks so as not to run into issues.

[Planemo]

1 hour ago, KAY GEE said:

Sorry, to disappoint you, the Z10 is a beautiful designed wheel, but the electronic components are nothing to right home about,

I never said the electronic components were something to write home about....

I was sad because I would just like to know what is causing the problems on Z10's, and possibly prevent it from happening to mine and others. The information/data we have so far isn't conclusive as to what is going on.

We had the first early, brand new wheels dying almost out of the box, we have Hsaings 1000km wheel dying whilst riding fairly conservatively with no other warning and now Ace's wheel when doing pendulums after 10,000km of hard riding. I find it all very odd, and just wish we could drill it down to something specific.

It's Ace's wheel which I find the most odd. To be problem-free for that sort of hard KM then suddenly fail seems bizarre.

Averaging it out over the year, I would think I ride mine less than you do. I would still like to know what the cause is.

A helpful piece of data would be build dates for wheels that have failed. Reading between the lines, NB suggest that wheels after April 2019 had changes. I do not think we have had a wheel fail since, so that might be some indicator, it might not.

 

[mrelwood]

I’m not familiar with the other total Z10 failures, but doing a pendulum with relatively strong accelerations like @RoberAce did in the video, is (and has been since day 1, on any self-balancing vehicle) by far the easiest and most certain way to blow a mosfet. The current spikes when the wheel switches direction are a lot higher than one can get with any kind of riding.

Unless the wheel has specific protections against that, the Mosfets can blow from pendulums just as well on a new wheel than on one that has been stressed for thousands of miles.

Data points on wheel mileage or other parameters do not help here. The cause is already 100% clear, and well documented on video.

[Planemo]

2 minutes ago, mrelwood said:

I’m not familiar with the other total Z10 failures, but doing a pendulum with relatively strong accelerations like @RoberAce did in the video, is (and has been since day 1, on any self-balancing vehicle) by far the easiest and most certain way to blow a mosfet.

That's fair enough and I take that on board. I knew that pendulums took a lot of current, I just didn't think the result would be that drastic after such a short amount of time doing them. If you are 100% satisfied that this is what caused Ace's wheel to blow I am somewhat relieved as I can't do pendulums!

In fact, I can't ride backwards either. Maybe I shouldn't learn, as the propensity to engage in pendulums would be far greater!

I might, carefully, just do a few 'gentle' pendulums whilst holding onto something and look at the wheelog data just to see what kind of amps are being pulled. I know what my heaviest amp draw to date is during riding, so it would be interesting to compare. 

[Hadrien Simeray]

Hello everyone, 
First of all, thanks for this topic who is very interesting about knowing better the Z10 and it commons failures.

I've saw that one of the biggest issue is the two "Hall sensor" if I understood,  labeled "U1, U2" on the MOS board.

The soldering of the seems to be and issue and also their cooling from what I've understood.

I was wondering if someone had already tried to make a better solder it again and cooling ig by adding heat radiator on it or maybe a fan (look more complicated because you need to evacuate the hot air).

I'm thinking about buying a Z10 but i'm kind scared of the numbers of failures coming from this issue I read on Facebook/forums.

Any reports ?

Thanks

[mrelwood]

 

1 hour ago, Hadrien Simeray said:

I've saw that one of the biggest issue is the two "Hall sensor" if I understood,  labeled "U1, U2" on the MOS board.

While I haven’t looked the teardown videos, I don’t think there are any hall sensors on the Mosfet board. Perhaps they are transistors? Hall sensors’ only job is to react to magnets closeby, hence there are a few inside the motor.

 

1 hour ago, Hadrien Simeray said:

I'm thinking about buying a Z10

You might want to watch Speedyfeet’s 650km review first.

[Rotciv]

3 hours ago, Hadrien Simeray said:

Hello everyone, 
First of all, thanks for this topic who is very interesting about knowing better the Z10 and it commons failures.

I've saw that one of the biggest issue is the two "Hall sensor" if I understood,  labeled "U1, U2" on the MOS board.

The soldering of the seems to be and issue and also their cooling from what I've understood.

I was wondering if someone had already tried to make a better solder it again and cooling ig by adding heat radiator on it or maybe a fan (look more complicated because you need to evacuate the hot air).

I'm thinking about buying a Z10 but i'm kind scared of the numbers of failures coming from this issue I read on Facebook/forums.

Any reports ?

Thanks

The U1 and U2 component on the driver board are hall-effect-based current sensor. Allegro ACS781 to be exact. When you see the pictures and they look burnt, it’s actually the conformal coating on the board overheating and getting discolored. 

Without properly testing a failed board, we can’t know for sure what component is prone to failure. A shorted motor can result in overheated board and eventually component failure. 

[Hadrien Simeray]

On 9/9/2019 at 3:15 PM, mrelwood said:

 

While I haven’t looked the teardown videos, I don’t think there are any hall sensors on the Mosfet board. Perhaps they are transistors? Hall sensors’ only job is to react to magnets closeby, hence there are a few inside the motor.

 

You might want to watch Speedyfeet’s 650km review first.

Okay, I've just watched the review and it's always a pleasure to watch his videos eheh.
About the behave, I've read about it and tried it once for 10 min in the street. It was ... interesting but was an too short experience  to have a feeling.

On 9/9/2019 at 5:48 PM, Rotciv said:

The U1 and U2 component on the driver board are hall-effect-based current sensor. Allegro ACS781 to be exact. When you see the pictures and they look burnt, it’s actually the conformal coating on the board overheating and getting discolored. 

Thanks for the correction and precision.

On 9/9/2019 at 5:48 PM, Rotciv said:

Without properly testing a failed board, we can’t know for sure what component is prone to failure. A shorted motor can result in overheated board and eventually component failure

Yes I understand and agree.

Anyone tried to change the sensors after failure with sign of burn on the sensors ?

 

[Damarafaka]

On 9/1/2019 at 5:35 PM, Planemo said:

Still seems weird that Ace's wheel can do so many miles (10k?) before doing this. Makes me think that something else is going on.

I found it strange that Roberto's Z10 didn't give any warning before blowing the mosfet because in my experience the wheel starts beeping and tilting back hard after heating up from constant pendulum riding, although rider weight does influence a lot, in my opinion he was giving it a real beating with those "violent" pendulum movement in a very short time span without allowing any breathing time for the Z10 to recover. As you say, I think also something else is going on and the problem is that ninebot seems to have no idea of what it is. None of the software updates or different "batches" have definitively solved anything. From the battery drain, the low battery % problems, low temperature start, etc. ninebot never identifies, admits or solves concrete problems, it just says so and so has been fixed, all very vague; one replaces then, for example, a supposed faulty mother board but the wheel still will not turn on. I believe the battery and the BMS system have a lot to do with what is causing many of the Z10 problems. One thing for sure is that putting many miles on the battery like Roberto seems to keep it healthy but allowing a Z10 to stay idle several days with low battery can lead to all sorts of battery imbalances, nightmares and something else that thanks to ninebot we'll never know.

[Planemo]

46 minutes ago, Damarafaka said:

I found it strange that Roberto's Z10 didn't give any warning before blowing the mosfet because in my experience the wheel starts beeping and tilting back hard after heating up from constant pendulum riding

That's interesting. I didn't know the Z had tiltback/beeps for an overheat scenario. That said, maybe it only works in relation to battery temp rather than board temp. I don't know if the Z has both, only that it measures 'something' because I know it won't let you charge if that 'something' is above a certain temp.

[Planemo]

On ‎9‎/‎11‎/‎2019 at 11:27 AM, Hadrien Simeray said:

Anyone tried to change the sensors after failure with sign of burn on the sensors ?

I have never seen or heard of anyone doing this. It would be good to hear if it solved the problem though. I would give it a go personally, the sensors wont be a lot of money and I can solder. If it still didn't work then a new board it is, which of course it would have needed anyway. Nothing lost for the cost of a couple of sensors.

.....unless when the sensors fail it takes something else out, and replacing the sensors with new then causes an issue with something inside the wheel (hall sensors maybe) which would add a world of pain.... it's only my concern because Hsaing's new board didn't solve the problem...

Edit: sensors are £4.34 each

https://www.digikey.co.uk/product-detail/en/ACS781KLRTR-150B-T/620-1827-1-ND/6189116?utm_campaign=buynow&WT.z_cid=ref_octopart_dkc_buynow&utm_medium=aggregator&curr=gbp&site=us&utm_source=octopart

Edited September 13, 2019 by Planemo

[boat dog]

Hello All

another EUC newbie here and probably an idiot too as I have just got a bust z10....I know.

I have opened up the beast and It shows the familiar short circuit damage to those 2 connections on the power board and I have accordingly ordered a new one (I will be looking to experiment with the original board to see if a repair is possible) 

In the interim I charged the batteries and having read exhaustively about the batteries on this forum I left the charger on over night to allow for cell balancing.  the powerboard issues did not stop the wheel form turning on, connecting to the ninebot app, etc so thought it fine to charge.

In the morning I found the charger block light had become a barely visible red and the wheel was off (no bars showing) and then on physically checking the battery voltages (at the disconnected XT plugs) I was rather shocked to get voltages on both over 70v.....they should be 58.8v max, 14s @ 4.2v max v per cell cluster?  Its now some hours later and monitoring the voltages I have seen some very odd significant variations with both batteries, moving from 58.2 to 71.5v, surely this is wrong? If this is wrong, then its a BMS fault? Could this lie at the heart of the issues with this wheel?

Over voltages, 10v or more above spec would surely stress components, including the battery cells, high light design and construction weaknesses? Especially if the wheel is used soon after disconnection from a full charge? Is this what the new BMS firmware was addressing?

Or am I not knowledgeable enough about lithium battery chemistry and the no load voltages are misleading?

I could fit new BMS boards, 1 for each battery but I would expect the brain needs to "see" the battery info in order to function and that will be apart of the ninebot software environment that would be all but impossible to hack? Does anyone out there have knowledge about the ninebot BMS?

Lastly can anyone tell me about the led indicators?  Firstly the 2 for the batteries on the pack, then the red/orange ones on the brain mother board?

I would be grateful for any inputs and I really hope I can get this wheel going as its truly a magnificent bit of design and flawed engineering.

cheer

 

[Planemo]

29 minutes ago, boat dog said:

In the morning I found the charger block light had become a barely visible red and the wheel was off (no bars showing) and then on physically checking the battery voltages (at the disconnected XT plugs) I was rather shocked to get voltages on both over 70v.....they should be 58.8v max, 14s @ 4.2v max v per cell cluster?  Its now some hours later and monitoring the voltages I have seen some very odd significant variations with both batteries, moving from 58.2 to 71.5v, surely this is wrong? If this is wrong, then its a BMS fault? Could this lie at the heart of the issues with this wheel?

It does seem somewhat uncanny that you had a blown board at just around the same time that your BMS also appears to have failed. I do wonder if the two faults are linked.

The only reason that suggests the problems aren't linked is because there have been more than a few board replacements which have resulted in a perfectly running wheel. However, now both yours and Hsaing's appear to have more than just a problem with the board, although in both cases, at least, the board damage is the same. It's all very odd. We still don't know what the 'other' problem is with Hsaing's - it could be the BMS, the motor or even some dicky cables.

You don't have any problem understanding lithium cell chemistry - you are right in that you should not be seeing 70v at the XT plugs. You also shouldn't have any red light, dim or not, after an overnight charge. This would suggest a serious issue with the charger or BMS which should never happen. If it was indeed putting out 70+ volts I think I would do the lottery because you seriously dodged a bullet by avoiding a wheel fire. As an aside, how well do you trust your voltmeter?

What serial number (or build date if you prefer) is your wheel?

[boat dog]

Hi Mr P

thanks so much for the response, but I am gutted that my worst instincts about the BMS appear to be reinforced...but I will check with another meter and update.  I have removed the battery pack and have it in plain site just in case FIRE! though there is no indication of warmth of the case at least.

I will never be able to get a replacement BMS board which means the wheel is very likely a huge waste of money, I really am a twit. Anyway I will carry on investigating; the motor is going to be checked by Dave/"Electrical repair co." in Romford (utterly brilliant place, lovely old school electrical engineering shop and so rare to find these skills now. They specialise in motor rewinds) and I will take the battery pack to a specialist in Bounds Green to have their input.

Do you know what the tiny LED's on the battery tell us?  Red means?  solid blue means? blinking blue means? etc.

I am happy to let my busto power board be a guinea pig for testing if a repair is possible.

all the best

J

[Planemo]

As you say, obtaining a Ninebot BMS is difficult, if not impossible. If it wasn't for my suspicion that the motherboard needs to communicate with it at a fairly in-depth level, I would be inclined to look at fitting a 3rd party BMS and be done with it.

I was about to say that for batteries and BMS', Jimmy at Insat International is excellent.....then realised he is the guy in Bounds Green you are talking about

Not sure what your motor guy will find....splitting the motor is a fairly involved job but for sure it would be interesting to see if he finds anything untoward, especially around the hall sensors and/or windings.

I can't help re the battery/motherboard diagnostic LED's. I thought I had seen some reference to them somewhere, maybe on this forum but I can't for the life of me find it.

Please do keep us updated, any and all info around this will be greatly appreciated. And if you can find your build date it would be handy.