KS16 Folded twice under me during "standing" take off

[OliverH]

13 hours ago, xoltri said:

I have the KS16 with 518km according to the app so far (although this is inflated by about 24% due to a bug that calculates mileage incorrectly) and I haven't experienced this yet.  However I mount the wheel by putting my right foot on and then kind of skateboarding with my left leg then hopping on.  

That's may be the reason. I do it the same way, my wife hopes on. But both ways are normal usage.

I just told her that others had the same issue. She got very angry and I don't repeat her words here.

Anyway she'll get an ACM to try with. We need to replace the Ninebot to manage tours (distance) so we need a new EUC. 

[MaxLinux]

On ‎7‎/‎25‎/‎2016 at 7:01 PM, OliverH said:

I don't need a description what to do or how long I need to wait to use it. A normal user switch it on and likes to drive. As there is no "be cautious, Kingsong prepares your next ride" voice note this user likes to ride immediately without any difficulties. Don't search for work arounds at the user, investigate on the EUC firmware.

Absolutely correct. It seems KS released the product with insufficient testing.

[MaxLinux]

19 hours ago, rolauk said:

I was about to buy a KS-16 or a KS-18, but now I don't know if I should anymore.

If you want a nice 16 inch wheel that is very reliable, try IPS191 (Lhotz) or IPS121.

[TremF]

9 minutes ago, MaxLinux said:

Absolutely correct. It seems KS released the product with insufficient testing.

They aren't the only ones to do that. Look at the failure that was the Ninebot One P, various Ninebot One firmwares and the issues people have had with the Gotway ACM. (these are just ones that I am aware of)

It might be that they are unable to test every eventuality and they don't have issues during their test rides but at the price of these things people don't want to be beta testers and risk their EUC getting damaged or worse themselves getting injured. 

Maybe they should all initially have a beta program where they sell discounted EUC's to those willing to test them. Once it seems like most if not all bugs are worked out they then sell at the full price?

[OliverH]

24 minutes ago, TremF said:

They aren't the only ones to do that. Look at the failure that was the Ninebot One P, various Ninebot One firmwares and the issues people have had with the Gotway ACM. (these are just ones that I am aware of)

It might be that they are unable to test every eventuality and they don't have issues during their test rides but at the price of these things people don't want to be beta testers and risk their EUC getting damaged or worse themselves getting injured. 

Maybe they should all initially have a beta program where they sell discounted EUC's to those willing to test them. Once it seems like most if not all bugs are worked out they then sell at the full price?

Maybe Kingsong is not the only one having this problem. I don't need to accept this untested products. With real regulations in place (not this tolerating state)  there's no way to pass a certification with this culture of development and driving the business.

I've freezed our Ninebot FW at 1.23 or so to live with the failures of this FW version as they seems to be controllable in opposite to the power cuts, destroyed main boards later.

I got no answer regarding FW testing from Kingsong so I freezed the KS16 also. At the point when Kingsong pulled back the 1.20 FW I felt good with my decision.

Also the MSuper has behaviours where balancing don't come back when you struggle on an urban track uphill.

I you design/ develope Software you've having test procedures. The company I work for has a department only dealing with testing (frameworks, tools, support). In the developing process we continuilly test software and categorise them that we know if we can launch at the release date and to deal with the right resources for bug fixing. If we go back to EUCs which are driven by humans I've really no understanding that, it looks like, there's no test procedure to test again use cases with different driver weights and run tests (documented) on the bench going through the behaviours/ maps of the firmware. 

[MaxLinux]

30 minutes ago, TremF said:

It might be that they are unable to test every eventuality and they don't have issues during their test rides

They're not doing enough test rides. The people writing these terrible experiences in the forums had their wheels for how long? Sometimes just a few weeks. This tells me Gotway, KS, etc. are not trying very hard to be sure their products are ready to sell to customers.

Reading these horrible stories about Gotway, KS, Ninebot, etc. makes me very thankful for my IPS121 and IPS191 which have been rock solid. If you're tired of quality issues, consider switching to IPS.

 

[TremF]

23 minutes ago, MaxLinux said:

If you're tired of quality issues, consider switching to IPS.

 

The big thing for me is battery size. The small battery and limited mileage are what made me switch from the Ninebot One. After the distance I can travel with the ACM (at speed too) I don't think I could drop back down to the IPS   

The new MSuper V3 is meant to have a 1640wh version, according to Jane, so I may even be tempted by that - once it's been tested by others and any bugs/design flaws worked out.

Edit - I've just seen that IPS do do bigger batteries (I initially looked at their Asia site) so may look into those.

[OliverH]

34 minutes ago, TremF said:

The big thing for me is battery size. The small battery and limited mileage are what made me switch from the Ninebot One. After the distance I can travel with the ACM (at speed too) I don't think I could drop back down to the IPS   

The new MSuper V3 is meant to have a 1640wh version, according to Jane, so I may even be tempted by that - once it's been tested by others and any bugs/design flaws worked out.

Edit - I've just seen that IPS do do bigger batteries (I initially looked at their Asia site) so may look into those.

Don't make the fault to read the big battery from two IPS build together like a balance board. They count both battery sizes to on battery.

[lizardmech]

I don't know how much of the firmware the mid sized EUC companies write themselves. The motor control section looks like a recycled design from an old ebike controller. If they outsource motor control code or use standard ST software they may not have much influence on how it is performing. The same applies to the motor control layout, if they didn't design it there's always the possibility it has flaws that prevent it being reliable enough for EUC.

[OliverH]

26 minutes ago, lizardmech said:

I don't know how much of the firmware the mid sized EUC companies write themselves. The motor control section looks like a recycled design from an old ebike controller. If they outsource motor control code or use standard ST software they may not have much influence on how it is performing. The same applies to the motor control layout, if they didn't design it there's always the possibility it has flaws that prevent it being reliable enough for EUC.

I thinks they stick with STM32 F103 because of the setup tool for the motor configuration.

The old eBike controller modified/ advaced with Gyro was the Gen1 boards I think.

But you're right there's no real improvement it's always work arounds.

[esaj]

On 26.7.2016 at 2:01 AM, OliverH said:

As I rode the KS16 again on the last tour as my wife again had problems with her feets on the KS16 (she rode my MSuper instead again :-( ) I was aware of the problem. I powered the KS16 on, I got all the lights indicating it's on but there was no balancing. After 2-3 seconds I got balancing. Is this the time the firmware needs to boot up with all services/ daemons? Can I disable this fancy audio pass through payback to maybe have more safety?

I've probably said this before, but I don't think they're running a realtime OS there, so no daemons/processes, just a bunch of code running on the ST32M F3. Also, in my opinion, it would make no sense to try to put something like an embedded Linux there, the F3 K6's that Firewheel uses have 32 kilobytes of flash for program and 16 kilobytes of RAM, and a single core running at max frequency of 72MHz. Even as a small OS probably can be fitted there, I don't see the reason for such (it would probably just make the boot-time longer, eat up more resources and give very little to no advantage over "purpose built" software).

On a more "general purpose" computer that does a lot of tasks, like a smart phone, it makes sense to use RT-os, plus modern smart-phones have multi-core processors running at over gigahertz frequencies (4 cores at 1.2GHz for example), a gigabyte of RAM and gigabytes of storage space for OS and software. (1 gigabyte = 1 million kilobytes, 1 gigahertz, GHz = 1000 megahertz, MHz).

The wheel hardware and software is purpose-built for only that device and use.

But, one thing that comes to mind that can cause a delay at start up is that I noticed on the MPU6050's that using the internal sensor fusion, it takes a few seconds before the values stabilize, and even up to around 10 seconds before they're completely stabilized.

[jesi]

Sadly, i join this club also

I will chat a bit with Tina first, before i write here more what i experienced just a moment ago and what was outcome of that...........

[HEC]

Just now, jesi said:

Sadly, i join this club also

I will chat a bit with Tina first, before i write here more what i experienced just a moment ago and what was outcome of that...........

I'm sorry to hear that. Please do share the both outcome as well as actual incident details once ready.

[HEC]

So it happened again today morning while going to post office to pickup a parcel (100x 40A fuses for KS). As always the wheel was charged to 87% the previous evening after leisure afternoon ride using Charger Doctor. At the morning I've carried the wheel down to lobby, switched it on and pushed it powered outside of the house where I've switched the lights to "On" mode suffering again from loud, annoying and most of all unavoidable "Hello Kingsong" shout. Then I've pushed it across the street to my usual take off point (smooth flat tarmac at bicycle lane next to traffic sign post I use to lean against during take-offs). Put on my headband, attached the safety / training belt, put on the bell, zipped up the jacket and tested that the wheel is balancing by pushing it to front and back a few times. Climbed on the wheel - again tested it's balancing by rocking twice forth and back - and bam! Wheel folded during take off  At the point wheel was switched on for several minutes so there is no chance that it was used "too early" after powering up. Battery level was more than sufficient and "acceleration" / lean forward was very mild.

I'm sorry to say so but I don't trust this piece of crap anymore!! @tinawong this is a SERIUOS issue which needs immediate attention! I know that @Jason McNeil is doing his best to chase King Song to address this as well as other known / pending issue but sadly KS response is less than impressive on this front. This could lead to serious injuries of unaware riders and even if you're aware of the problem you can't clearly avoid it 100% of time (I have now again serious pain in my left foot being hit by pedal due to this incident). And most of all - you should not be using any "workarounds" on product worth over $/£ 1000 of your hard earned money. I urge all King Song dealers present in this forum to pressure KS for prompt solution of this serious issue and most of all to be honest and warn their customer about this problem.

I'm considering getting rid of KS-16, recommending anyone NOT to buy any of the KS wheels and looking for a replacement from some more serious manufacturer like Inmotion. Although I'd miss the excellent range of KS-16 I'd prefer my health and safety to unreliable product like this.

[US69]

@HEC

You should really check for the capacitor leg's....(you or your Distributor)

That at all does not Sound anymore like the fault what others are experiencing!

I can understand your Frustration!!!

 

Btw.: You ordered Fuses? You have had a Fuse blow before? Because that would lead/Show to the capacitor leg Problem, too!

[HEC]

1 minute ago, KingSong69 said:

@HEC

You should really check for the capacitor leg's....(you or your Distributor)

That at all does not Sound anymore like the fault what others are experiencing!

I can understand your Frustration!!!

 

Btw.: You ordered Fuses? You have had a Fuse blow before? Because that would lead/Show to the capacitor leg Problem, too!

I'll most likely strip the wheel over the weekend and check the soldering on the large capacitors although this still falls under warranty as it's basically brand new wheel (under 2 months old) and while I'm proficient with soldering and have all the necessary tools I should not be "fixing" problems like this. I'll check with @Jason McNeil what's the best way to proceed on this.

And no - I did not have any issues or fuse blown. I've just order those from aliexpress "in advance" considering the delivery time and also to use the fuses in some other of my electronic projects. Hopefully sett of 100 will last for a while  

[DS]

57 minutes ago, HEC said:

Climbed on the wheel - again tested it's balancing by rocking twice forth and back - and bam!

Since you tested the wheel for a presence of balance power and the "fold" happened AFTER that....oh my...IMO, means the fold could happen any time after! 

Hard to trust such a wheel You may need control board replacement

I understand very well your frustration, cause I was in similar situation with KS14 back in March this year...a happy start of the thread "About to join KS club" turned out in bitter frustration with a final good ending.

I think with such a highly reputable dealer as our, your story will have the same end...but it's not easy for anybody to accept that he is randomly chosen for shit to happen.

[US69]

35 minutes ago, HEC said:

I'll most likely strip the wheel over the weekend and check the soldering on the large capacitors although this still falls under warranty as it's basically brand new wheel (under 2 months old) and while I'm proficient with soldering and have all the necessary tools I should not be "fixing" problems like this. I'll check with @Jason McNeil what's the best way to proceed on this.

And no - I did not have any issues or fuse blown. I've just order those from aliexpress "in advance" considering the delivery time and also to use the fuses in some other of my electronic projects. Hopefully sett of 100 will last for a while  

When you have bought it from Jason you are in best Hands!

Like @DS said: Your Problem sounds more like this can happen to you anytime.....

[Jason McNeil]

@HEC, completely agree with you. I'll ship you out a replacement board ASAP or we can replace for you if you wish, you should not have to be doing PCB repair.

It was the first reported incident of this kind that I had heard, the diagnoses info from KS has to be defective based on your evidence.

[HEC]

27 minutes ago, Jason McNeil said:

@HEC, completely agree with you. I'll ship you out a replacement board ASAP or we can replace for you if you wish, you should not have to be doing PCB repair.

It was the first reported incident of this kind that I had heard, the diagnoses info from KS has to be defective based on your evidence.

@Jason McNeil I don't mind replacing the board myself (I have both the tools needed as well as the skills) though I'll be unable to say if it's really bad soldering on the capacitors or anything else discoverable by visual inspection or by basing measuring (I have several digital multimeters, basic oscilloscope and frequency counters and generators) until I open the wheel itself and if you'd send the board I'd prefer to open it only once during replacement instead of twice - once now to check the board and then again to replace it - unless indeed you need first some visible proof of failure at which case I'd probably take it apart and not use it / leave it open until the replacement board would arrive. I'd not even mind to re-solder the capacitors if that would be visibly verifiable fault though I'd not be able to see if other "hidden" faults are present in the board.

[SeabirdHH]

@HECTwo days ago I was in similar situation. I switched on the wheel (without "Hello K."). All green LEDs were on ... fully charged. Took my iPhone to open the app "Road Bike", then put it in my pocket. About 30 seconds already had passed. Took the wheel, pushed it - what I always do - back and forth, right foot on the pedal and in the moment when I put my left foot on the left pedal, the wheel tilted forward, so that I nearly fell on the ground.

I did about 600 Km and it never happened before. This morning I started without any problems. But neverthless I'm very concerned!

[HEC]

16 minutes ago, SeabirdHH said:

@HECTwo days ago I was in similar situation. I switched on the wheel (without "Hello K."). All green LEDs were on ... fully charged. Took my iPhone to open the app "Road Bike", then put it in my pocket. About 30 seconds already had passed. Took the wheel, pushed it - what I always do - back and forth, right foot on the pedal and in the moment when I put my left foot on the left pedal, the wheel tilted forward, so that I nearly fell on the ground.

I did about 600 Km and it never happened before. This morning I started without any problems. But neverthless I'm very concerned!

I'm sorry to hear that. Was it a "running" / push off start or have you been leaning against something and performing the "standing" take off? From your brief description it looks more like former but just to be sure as so far this problems seems to be demonstrating from "standing" take-offs only.

So far we have about 4 or 5 reports of this or similar behaviour just from the forum members here so I'm afraid it's more widespread / common than KS claims or wants it to be. While I can't exclude the possibility that in my case (or yours or anyone's else with the same issue) it's really a hardware issue (poor soldering and affixing of the large controller board capacitors as already mentioned in this thread) rather than FW issue (which I think is more likely) I simply refuse to accept this. Be it HW (poor assembly and QA) or SW (bad coding, insufficient testing) issue that's not something what customer shall see on few weeks or months old product of considerable cost.

[SeabirdHH]

@HEC I have not been leaning against something but nevertheless it should have been a so called "standing" take-off. 

[HEC]

4 minutes ago, SeabirdHH said:

@HEC I have not been leaning against something but nevertheless it should have been a so called "standing" take-off. 

OK - so you perform "bunny" hop on wheel while standing still (without holding or leaning onto anything) and after that tilt forward to start moving, correct? So there is no (even minimal) wheel movement before you put second feet up ... so yes that would technically be similar to our "standing" take-offs.

[SeabirdHH]

6 minutes ago, HEC said:

OK - so you perform "bunny" hop on wheel while standing still (without holding or leaning onto anything) and after that tilt forward to start moving, correct? So there is no (even minimal) wheel movement before you put second feet up ... so yes that would technically be similar to our "standing" take-offs.

Yes, that's correct! But I'm not sure if there was a minimal movement before I put the second foot up. What's a minimal movement? ... 1mm? That's not easy to assess .