V12 Cutout tracking

[supercurio]

2 minutes ago, Planemo said:

I would deffo do the test, break that damn board. Do the test 100 times. Everything you can in order to screw that wheel right up. Then you will get another board

Yes and the right to wait for months instead of using the wheel you've been waiting for 6+ months.

It seems to be a reason why some who decide to run the stress test do it kind of half way because in the back of their mind they don't really want to break their wheel.

One more reason why the end customer should never be relied on to run destructive QC.

[Rawnei]

36 minutes ago, terlikaa said:

Well.. There's already one wheel that passed the test and now, a week later, mosfets are blown. 

https://m.facebook.com/groups/inmotionv12/permalink/502406191191640/

"it passed the free spin test just last week" the test is not to test free spin.. might be bad wording might not be, it's problematic that the test itself seems difficult to execute properly, maybe doing pendelums is a good test as well? But not everyone can do that either.

[skedra]

4 minutes ago, supercurio said:

While I fully understand the snark here (well deserved), some people thought that the test was merely lifting the wheel and letting it free-spin.
One person did that and uploaded it on YouTube so commenters warned him about the problematic execution (and second test is not that much better unfortunately)

Knowing this is a possibility, what is the ratio of actually stress test vs nothing test, and car we tell if this report of later cut-out disqualifies the test entirely or not.

I gave some material with good and bad tests to the distributors where I and a few friends order a V12 from. He sent me a video with the best stress test I've seen so far, 4+ minutes long!
Hopefully, most distributors will do effective stress tests before sending out wheels, at least.

But doe it mean that the synthetic stress test is nothing compared to real-world stress or that this person couldn't reproduce the stress test as intended (and really, no blame here)?

Let's investigate.

The distributor I bought from had their demo fail a day after passing the stress test twice (and they did have wheels fail that stress test so I'm pretty sure they did it right). So to me this test only filters out the absolute worst wheels but plenty will "just" pass it and probably fail in the real world.

[Paul A]

If an owner does not conduct the stress test correctly.......they could then potentially have a cutout whilst riding and suffer extensive injuries.

 

[supercurio]

9 minutes ago, skedra said:

The distributor I bought from had their demo fail a day after passing the stress test twice (and they did have wheels fail that stress test so I'm pretty sure they did it right). So to me this test only filters out the absolute worst wheels but plenty will "just" pass it and probably fail in the real world.

London. Was that erides?
I'm asking because they shared on their YouTube channel a very well executed stress test (which results in a fail)

Thing is I would not believe a stress test result unless I see the video myself, at this point. There's so many ways it can go wrong, including by doing well enough but not enough times.

[Planemo]

12 minutes ago, supercurio said:

One more reason why the end customer should never be relied on to run destructive QC.

This 1000%. And the fact that the 'test' can be performed to different degrees doesn't help matters.

[taiguy]

Another dead V12.

I'm posting for a buddy who rides with me. Batch 2 from eWheels, received in November 2021. 126 miles on the odometer. Battery was at 80% charge and did a slow ride down 2 stair steps. It cut out at the bottom. Wheel power button does not respond. If you plug in the charger the top display will turn on. After removing the charger, the power button will prompt the lock keypad. After entering the unlock PIN, the wheel shuts down, and the power button is unresponsive again.

Comparing the broken V12 to my still functioning V12 (same batch from ewheels, received a day apart) his wheel does not freespin while off. There is a definite sticking point every 3 magnets/clicks? Would seem to indicate a MOSFET failure.

We're preparing ourselves to disassemble his V12 so we can send the control board back to eWheels for repair.

[supercurio]

1 minute ago, taiguy said:

Another dead V12.

I'm posting for a buddy who rides with me. Batch 2 from eWheels, received in November 2021. 126 miles on the odometer. Battery was at 80% charge and did a slow ride down 2 stair steps. It cut out at the bottom. Wheel power button does not respond. If you plug in the charger the top display will turn on. After removing the charger, the power button will prompt the lock keypad. After entering the unlock PIN, the wheel shuts down, and the power button is unresponsive again.

Comparing the broken V12 to my still functioning V12 (same batch from ewheels, received a day apart) his wheel does not freespin while off. There is a definite sticking point every 3 magnets/clicks? Would seem to indicate a MOSFET failure.

We're preparing ourselves to disassemble his V12 so we can send the control board back to eWheels for repair.

Thanks for the report, I hope your buddy didn't get injured in the process.

Since you didn't mention it I presume his wheel wasn't stress tested with the recommended procedure?

[Planemo]

20 minutes ago, skedra said:

The distributor I bought from had their demo fail a day after passing the stress test twice (and they did have wheels fail that stress test so I'm pretty sure they did it right). So to me this test only filters out the absolute worst wheels but plenty will "just" pass it and probably fail in the real world.

Bugger me gently. Anyone for some Russian Roulette whilst we're here?

Another one to put the nail in the coffin for this ludicrous 'test'.

[supercurio]

After exchanging a few private messages (thanks for reaching out), it appears that at least one cut-out & burned MOSFET instance in real-world usage happened after a well executed stress test.

Sorry no quotes , but the sources seem credible enough. I hope we'll get official statements soon.

At this point, I see no other recourse than:

• Full recall of Batch 2
• Batch 1: inconclusive at this point
• "More durable MOS" of V12 HT vs higher quality supply of the same: inconclusive at this point

@Cecily Inmotion, please advise on recall options.
The stress test, even when ran by distributors is not enough to keep your customers safe, as demonstrated in only a few short days.
Running the same stress test in factory will also be insufficient.

[kiitick]

I'm not sure if this counts, since mine shuts down each and every time I've run the pendulum test.  It becomes unresponsive, and the wheel still spins. I plug it in, and it will power back on and self balance.  I run the pendulum test, and it shuts down every time on the tilt back portion. (hopefully I'm doing it right.)  

This is a wheel that I bought from a friend, who got it from ewheels.  I bought it from him because the first time I rode it, I wiped out on an emergency brake.  At the time, I just assumed it was my own error, but now, I'm not quite so sure.  It had about 50 miles on it, it was cold, below freezing, and we'd ridden about 4 miles taking it easy while I learned the wheel. I took it up to 37, got some wobble, so decided to brake...down I went. The lights stayed on, but it wouldn't balance, and the screen wouldn't work. 

I'm a bigger rider (101Kg, 224 lbs). I haven't pushed the wheel since then, setting the speed limit down to 25, and really taking it easy since. 

Edited January 12, 2022 by kiitick

[Unventor]

13 minutes ago, kiitick said:

I'm not sure if this counts, since mine shuts down each and every time I've run the pendulum test.  It becomes unresponsive, and the wheel still spins. I plug it in, and it will power back on and self balance.  I run the pendulum test, and it shuts down every time on the tilt back portion. (hopefully I'm doing it right.)  

This is a wheel that I bought from a friend, who got it from ewheels.  I bought it from him because the first time I rode it, I wiped out on an emergency brake.  At the time, I just assumed it was my own error, but now, I'm not quite so sure.  It had about 50 miles on it, it was cold, below freezing, and we'd ridden about 4 miles taking it easy while I learned the wheel. I took it up to 37, got some wobble, so decided to brake...down I went. The lights stayed on, but it wouldn't balance, and the screen wouldn't work. 

I'm a bigger rider (101Kg, 224 lbs). I haven't pushed the wheel since then, setting the speed limit down to 25, and really taking it easy since. 

Someone had something similar on telegram. It turned out to be an overload due to watts (or it is was wh) used at or close to 0kmh. So it might be the balancing load is high yet the wheel register only very low ride speed. 

Edited January 12, 2022 by Unventor

[supercurio]

8 minutes ago, kiitick said:

I'm not sure if this counts, since mine shuts down each and every time I've run the pendulum test.  It becomes unresponsive, and the wheel still spins. I plug it in, and it will power back on and self balance.  I run the pendulum test, and it shuts down every time on the tilt back portion. (hopefully I'm doing it right.)  

This is a wheel that I bought from a friend, who got it from ewheels.  I bought it from him because the first time I rode it, I wiped out on an emergency brake.  At the time, I just assumed it was my own error, but now, I'm not quite so sure.  It had about 50 miles on it, it was cold, below freezing, and we'd ridden about 4 miles taking it easy while I learned the wheel.

I'm a bigger rider (101Kg, 224 lbs). I haven't pushed the wheel since then, setting the speed limit down to 25, and really taking it easy since. 

Your story is uncommon: deciding to buy a wheel after determining it could not handle your weight in an emergency situation is counterintuitive.

Regarding the stress test, if you can it would be interesting to review a video of it. Make sure to look at a few available videos to see what works first.

[supercurio]

1 minute ago, Unventor said:

Someone had something similar on telegram. It turned out to be an overload due to watts (or it is was wh) used at or close to 0kmh. So it might be the balancing load is high yet the wheel register only very low ride speed. 

Yeah high speed motor, (somewhat) heavy rider and hard braking, we can only ask for so much at the end when getting close to 0 speed.

[kiitick]

I bought it from him because I'd crashed his new wheel. As I said, at the time, I was pretty certain it was user error. Kind of like car insurance, if I hit someone, I should pay for the damages.

[supercurio]

1 minute ago, kiitick said:

I bought it from him because I'd crashed his new wheel. As I said, at the time, I was pretty certain it was user error. Kind of like car insurance, if I hit someone, I should pay for the damages.

Yes a pellicular aspect here is that it's an item in very short supply and high demand (although maybe not as much right now).

Let us take a look at the test 😉

Edited January 12, 2022 by supercurio

[taiguy]

9 hours ago, supercurio said:

Thanks for the report, I hope your buddy didn't get injured in the process.

Since you didn't mention it I presume his wheel wasn't stress tested with the recommended procedure?

the facebook post linked above is mine. his wheel was stress tested as per inmotion's video. no injury as he was going down 2 stairs at 5mph. he managed to hop off as it was falling over.

[rolis]

10 hours ago, Paul A said:

If an owner does not conduct the stress test correctly.......they could then potentially have a cutout whilst riding and suffer extensive injuries.

 

And even if done correctly, you can never be sure. There is still a chance of creating a voltage spike that will destroy a mosfet. I had to change a few entries for passed tests to failed because the owner did the test again later. No batch 2 is safe until inmotion sends out improved boards.

[Paul A]

https://youtu.be/KLz9yK1OkNs

From the comments section:

 

Pinned by Zen Lee

Zen Lee

6 hours ago

 

Looks like eWheels is recalling the V12! Glad they're leading the way on this one.

 

[Rawnei]

40 minutes ago, Paul A said:

Looks like eWheels is recalling the V12! Glad they're leading the way on this one.

Eh lets please wait for an official statement before spreading rumors, the situation is already infected enough.

[Paul A]

Could save owners from injury, conducting a heavy test unnecessarily.

Wait for official statement, confirmation.

[pico]

I have a very different take on all of this.

 

If I had a talk to the Inmotion Engineers: Transient Core Saturation.

 

In layman terms momentarily over driving(or braking) a motor beyond its intrinsic capabilities.

In car terms Momentarily flooring an engine without a rev limiter.

 

I remember how I lost quite a few weeks of my life designing a switching Power Supply.

 

I would use higher and higher rated switching components until I realized I was saturating the transformer core.

As soon as I increased the power demand, poof! Before I knew it, the MOSFET would be toast.

Changing the transformer or diminishing the power demand solved the problem.

 

I also rewound quite a few BLDC motors. The noise I hear when those wheels would jam and fail is fairly reminiscent of the result I would get when I used a too aggressive (less turns) winding with quick throttle changes.

 

Just my $.02 

Edited January 12, 2022 by pico

[RagingGrandpa]

6 hours ago, pico said:

Core Saturation

my life designing a switching Power Supply

That's interesting...

My thoughts about context:

• For me, magnetic saturation in a motor results in the problem where increasing the electric current does not give an increase in output force (torque). Input power goes up; output power does not.
  In an EUC, this could be problematic because of the self-balancing feedback control. During normal operation, a forward torque (increased current) is commanded in response to a disturbance in pedal angle, and the torque should then act to move the pedal angle back to a level position. If the torque is insufficient to move the pedals to a level position, the torque command is increased, until the desired motion (leveling) is achieved. Classical feedback control.
  If the motor reached saturation during this control, the control logic would 'wind-up' and request more and more current, without ever achieving its desired level position. It would end up commanding 100% effort (max possible current) and be stuck there. Such currents could be so high or so long-lasting that they damage components.
   
• Fortunately, EUC's do have current-limiting logic in their controls, including direct measurement of motor current. So I think it is not possible to quickly melt a controller simply from the weakening effects of the motor (like saturation), because if the controller commands 100% effort, current would reach the limit threshold and separate current-limiting regulation will begin. Control will not be successful... an overlean condition happens, and often the rider falls; but without electrical damage.

(My thoughts apply equally to transients and steady-state operation)

You mentioned switch-mode supply design...
A power supply is another sort of feedback control (modifying an output current to achieve a target voltage), and would be subject to the control wind-up issue, if acting through a component that had saturated and cannot produce more output.

But your comment seemed focused on transients- could you elaborate? Are you suggesting an overcurrent problem? Overvoltage somehow? Something else?

Thx

Edited January 12, 2022 by RagingGrandpa

[Paul A]

On 1/8/2022 at 12:11 AM, evans036 said:

i am 64 years old, and trying to do that stress test on my wheel really did a number on my back. 

still not sure i did the test correctly. i think i'll wait a few weeks to let my back recover then try again.

that wheel is freakin heavy.

steve

 

 

People could be injuring themselves by conducting a heavy, awkward test.

A test that is now in doubt as being reliable, as there are reports of wheels failing despite having earlier passed the stress test.
 

Even if the Zen Lee comment that Ewheels will be conducting a recall does not eventuate, it seems pointless to do the test.

[pico]

43 minutes ago, RagingGrandpa said:

But your comment seemed focused on transients- could you elaborate?

I would rather say Surge rather than Transient.

My first though when I rode the wheel was, Geez the V12 is very aggressive in shutting down the motor when the wheel falls off.

(Hey, this is what I do, my wheel when doing tricks falls more than the casual rider) The V10F is a real fiasco in that respect. Takes forever to stop the wheel spin. This was noted in the V10F thread.

I do understand feedback loops. I also know that you can also control the gain of the loop.

My first thought was that WOW tons of gain (VERY GOOD for tricks) lets tone it down.

My theory from what I have observed is that it is very easy for the software to be OVER aggressive when SLOWING the wheel and that the ferro magnetic is saturated by pulses width way too large to be absorbed by this 16 inch wheel material. I suspect also variations in the over current detection and variation in quality of the windings would explain different wheels behaviours...

I could be wrong... 

In the mean time I am doing single legged tic-tocs but in COMMUTER mode. I am staying away from the OFF ROAD....

Edited January 12, 2022 by pico