INMOTION V12

[supercurio]

New thread where I develop the argument that Inmotion knew about V12 board failures and instead of addressing that, they introduced a new wheel with "More durable MOS"

TLDR: I think we need to lobby Inmotion to send revised replacement boards, like the fixed V12 HT board, equipped with the right components.
Instead accepting of a half-assed solution which will leave us in (fully justified) doubt every ride.

I made this thread because after talking to my partner about this V12 story, her reaction was to forbid me to ever ride it.
So essentially if I ever ride my V12 it'll be with the right MOSFETs.

Edited January 6, 2022 by supercurio

[ESB]

BTW did anyone see their odometer reset to 0 miles while doing the tests??? Sometime yesterday my wheel reset from showing 30 miles to showing 0 miles. Seems like a weird/scary bug if its possible to make a used wheel look new or otherwise lie about mileage.

[Tawpie]

1 hour ago, ESB said:

Seems like a weird/scary bug if its possible to make a used wheel look new or otherwise lie about mileage.

Or a loose connection. There was chatter about a loose ¿bluetooth antenna board? causing the wheel to reset—the jerk during testing might be jostling things around.

[Cecily Inmotion]

[aze man]

Hello Cecily,

first I want to really thank you that you - Inmotion - take this all very seriously.

At the moment you are stating that if the test - assuming that it is done correctly - is passing, then the wheel will be fine.

On the other side the cummunity has seen that in V12 from the summer (likely only 1st batch wheels?) the same 100V MOSFETs are used as in other 84V wheels from your company.

So from your engineers side you seem to be confident that you have enough headroom in the electronics for your latest wheel to be safe for years.

If that is all the case, to take the wind out of this whole thing, all you would have to do, is post the technical information why there won`t be any case in operation of the wheel, where the MOSFETs get near or at least under all circumstances not more than the specified 100V on one of the used MOSFETS.

And, or clarify, that in all of the 2nd batch wheels and ongoing, are MOSFETS used whitch have more than the 100V in their specification list.

Then it would be 100% clear for us all, that the recent fails are only because of components which are simply not up to the spec which are printed on it (and likely because of the component shortage you got somehow inferior parts onto your mainboards).

Thanks in advance of answering.

Edited January 7, 2022 by aze man

[supercurio]

2 hours ago, Cecily Inmotion said:

Thank you @Cecily Inmotion for coming back to us with an update.

A few immediate questions following:

Users are failing to run the stress test

A significant amount of V12 customers fail to run the stress test perfectly, for multiple reasons like physical capacity (heavy wheel, strong effort), configuration (anti-spin feature), misunderstanding (some thing it's a free-spin test), or due to the sheer difficulty in timing and strength.

How do you plan to take care of these customers?

Recommendation: you can issue a firmware update which includes an automated XX duration self stress-test, where customers would only need to hold the wheel in the air or hold still on a stand.
Much safer and effective.

Risk with tests

Each stress tests increases the risk of failure by accelerating the degradation of components which might not burn instantly but got very close to during the test.
Is your current statement only a first step, or do you plan on offering replacement of all board, regardless if they failed (yet) or not, when supply allows it?

Which replacement MOS?

Are you planning to try finding a consistent supply of Infineon IPP023N10N5 100V capable of operating reliably beyond their specification, or will you use the same "More durable MOS" as found in the V12 HT? (source: V12 HT release brochure)

Self-repair MOS options

Can you please offer a list of compatible MOS which can be installed to repair a failed board including:

1. The exact identifier and specifications of the current MOS
2. A list of compatible, higher voltage rated MOS, as well as the "More durable MOS" selected for the V12 HT board.

V12 HT board availability

Will customer uncomfortable with running a V12 100V MOSFET board on a 100.8V wheel be able to request a free upgrade to the V12 HT board with more robust components?

Will the V12 HT board be available to buy as replacement otherwise?

 

Thank you.

Edited January 7, 2022 by supercurio

[nosamplesplease]

One other POSSIBLE and UNPROVEN theory is counterfeit components. It does happen where parts are re-labeled and sold under false pretenses.

Was looking for a simple explanation of this.

Best I could find:

In that circumstance Inmotion blaming their supplier MIGHT be a dishonest middleman. So "inferior parts from their supplier" MIGHT not be referring to the manfacturer: Infineon as at fault, but the parts they THOUGHT were genuine Infineon. Remember there is a major language barrier. 

Not being able to explain the problem in English, does not mean they didn't try.

 

Yay, more speculation, but this thread and others like it are becoming quite the witch hunt and I don't believe that will result in BETTER wheels for us. Again everyone is entitled to their opinion but I believe more information will be forthcoming from dealers ONCE they start receiving wheels to repair. Boards sent in MIGHT get more info ahead of the March estimate. Still the dealers have ENOUGH work to do with the paperwork and logistics, lets not put any additional pressure on them. 

I will be giving the whole situation some space, and wait for more info. 

Thank you to everyone who wants to make this better, but I'm not really sure WE, as the users of these products, are the best guessers about WHY. 

[supercurio]

34 minutes ago, nosamplesplease said:

One other POSSIBLE and UNPROVEN theory is counterfeit components. It does happen where parts are re-labeled and sold under false pretenses.

Was looking for a simple explanation of this.

Best I could find:

In that circumstance Inmotion blaming their supplier MIGHT be a dishonest middleman. So "inferior parts from their supplier" MIGHT not be referring to the manfacturer: Infineon as at fault, but the parts they THOUGHT were genuine Infineon. Remember there is a major language barrier. 

Not being able to explain the problem in English, does not mean they didn't try.

 

Yay, more speculation, but this thread and others like it are becoming quite the witch hunt and I don't believe that will result in BETTER wheels for us. Again everyone is entitled to their opinion but I believe more information will be forthcoming from dealers ONCE they start receiving wheels to repair. Boards sent in MIGHT get more info ahead of the March estimate. Still the dealers have ENOUGH work to do with the paperwork and logistics, lets not put any additional pressure on them. 

I will be giving the whole situation some space, and wait for more info. 

Thank you to everyone who wants to make this better, but I'm not really sure WE, as the users of these products, are the best guessers about WHY. 

It looks like this is what Inmotion thinks is the root cause.

It's unfortunate Inmotion didn't add a full-system stress test at the end of assembly and before shipping the units, in the current context of difficulty to acquire quality parts in volume.

By the way this doesn't exclude a design failure being a contributor hypothesis.
Both can be happening at the same time, amplifying each other's severity (lower quality parts on low safety operating margin by design)

Edited January 7, 2022 by supercurio

[Silverfish]

5 hours ago, supercurio said:

Will customer uncomfortable with running a V12 100V MOSFET board on a 100.8V wheel be able to request a free upgrade to the V12 HT board with more robust components?

Will the V12 HT board be available to buy as replacement otherwise?

 

Thank you.

Would there be any other effects of swapping that board in besides better mofset? Would it still function the same, or would it work more like the high torque version?

[supercurio]

15 minutes ago, Silverfish said:

Would there be any other effects of swapping that board in besides better mofset? Would it still function the same, or would it work more like the high torque version?

Traditionally, the only difference between a HS and HT wheels are the motor.

Inmotion claims for the V12 HT both "More durable MOS" as well as "Increased max power output by 23%" which is a lot and certainly not possible with the current board which already operate at or beyond capacity.

It is unclear how accurate these claims will prove or the effect this board would have on the current high speed V12.

Personally, I'd be happy to see this promised 23% increase used to keep the hardware healthy and durable.

[Tawpie]

25 minutes ago, Silverfish said:

Would it still function the same, or would it work more like the high torque version?

They might need to make firmware changes, especially if the existing non-HT motor is rated for less power than the forthcoming HT (I don't know if it is or isn't but 23% more power is a lot more power). We don't want the motor wires melting, that's not good either!

Edited January 7, 2022 by Tawpie

[Paul g]

5 hours ago, supercurio said:

Thank you @Cecily Inmotion for coming back to us with an update.

A few immediate questions following:

Users are failing to run the stress test

A significant amount of V12 customers fail to run the stress test perfectly, for multiple reasons like physical capacity (heavy wheel, strong effort), configuration (anti-spin feature), misunderstanding (some thing it's a free-spin test), or due to the sheer difficulty in timing and strength.

How do you plan to take care of these customers?

Recommendation: you can issue a firmware update which includes an automated XX duration self stress-test, where customers would only need to hold the wheel in the air or hold still on a stand.
Much safer and effective.

Risk with tests

Each stress tests increases the risk of failure by accelerating the degradation of components which might not burn instantly but got very close to during the test.
Is your current statement only a first step, or do you plan on offering replacement of all board, regardless if they failed (yet) or not, when supply allows it?

Which replacement MOS?

Are you planning to try finding a consistent supply of Infineon IPP023N10N5 100V capable of operating reliably beyond their specification, or will you use the same "More durable MOS" as found in the V12 HT? (source: V12 HT release brochure)

Self-repair MOS options

Can you please offer a list of compatible MOS which can be installed to repair a failed board including:

1. The exact identifier and specifications of the current MOS
2. A list of compatible, higher voltage rated MOS, as well as the "More durable MOS" selected for the V12 HT board.

V12 HT board availability

Will customer uncomfortable with running a V12 100V MOSFET board on a 100.8V wheel be able to request a free upgrade to the V12 HT board with more robust components?

Will the V12 HT board be available to buy as replacement otherwise?

 

Thank you.

    If the V12 HT motor is different than the V12 HS motor (and usually it is, HT needs bigger magnets), than this board is not likely to work on the V12 HS variant, as the HT motors electro-magnets usually activate at different times than a HSpeed variant (the optimisation for HT differs that of HS). Now I don’t know how advanced the InMotion team is, but I know that in the best motors available at the moment this is how they optimise for these two different functions. 
 

   I also want to ask InMotion Engineers to let us know which higher V and A mosfets  we can use for replacing the factory ones, so we can safely ride our wheels on which we spent thousands of pounds/dollars/euros/etc. until they will send us a factory board replacement.

Edited January 7, 2022 by Paul g

[EUC Phenomenon]

Why can't Inmotion just replace it with the same board but with higher voltage MOSFET's and whatever other components that may need changing ?

[Tawpie]

9 minutes ago, Tel01 said:

Why can't Inmotion just replace it with the same board but with higher voltage MOSFET's and whatever other components that may need changing ?

⏳💰

 

Just joking. sort of. It does take time and money to build the replacements, and they'll have to figure out how to squeeze it into the existing production line. But it should be possible!

Edited January 8, 2022 by Tawpie

[EUC Phenomenon]

10 minutes ago, Tawpie said:

But it should be possible!

That's what I thought. I did the test yesterday and it passed but dam, that thing is heavy, especially when it has to be put on a 30 degree angle.

[Unventor]

1 hour ago, supercurio said:

Inmotion claims for the V12 HT both "More durable MOS" as well as "Increased max power output by 23%" which is a lot and certainly not possible with the current board which already operate at or beyond capacity.

So how did that statement come to life? That board's operate bound their capacity? 

I know they have an issue with V12 but that is due to some components not delivering specified performance. That is not the same as going beyond capacity if it was functional as design to be. 

[supercurio]

Oops @Silverfish, I forgot to mention firmware as everyone else pointed out.
So motor and firmware are different between HS and HT variants usually, with the same mainboard.

The V12 HT being announced using a different mainboard than the HS is an oddity compared to what we're used to.

I can imagine a V12 HT board running a firmware tuned for the HS motor unless they broke hardware compatibility (connectors, pins)
Visually both boards look very similar so far.

[supercurio]

9 minutes ago, Unventor said:

So how did that statement come to life? That board's operate bound their capacity? 

I know they have an issue with V12 but that is due to some components not delivering specified performance. That is not the same as going beyond capacity if it was functional as design to be. 

Simple:

• MOSFET operating beyond voltage specification according to their datasheet
• >20% failure rate across all batches

 

[Unventor]

11 minutes ago, supercurio said:

Simple:

• MOSFET operating beyond voltage specification according to their datasheet
• >20% failure rate across all batches

 

So the failure rate is not down to component not working as specified? 

Because I must have miss read Inmotion public statements wrong. 

As of the mosfets it is not clear to me that what you actually happens. Datasheets is one thing what goes on in the board is something different. We have seen this with other things on EUCs before. Especially with people thinking they could go higher with charging rates as individual cells should be able to handle it by spec sheets. 

What I am saying is if Inmotion said they made boards with too small mosfets that is one thing. But they didn't. They said some components didn't deliver as suppose to. 

@Cecily InmotionI tagged you to this so you get this into. It might be something to add to your faq document. To avoid incorrect info gets spread. 

Edited January 8, 2022 by Unventor

[Tawpie]

33 minutes ago, Unventor said:

What I am saying is if Inmotion said they made boards with too small mosfets that is one thing. But they didn't.

This is what has my panties in a knot. It certainly appears from the part manufacturer's specification sheet that IM is using the parts incorrectly, and it's generally understood that using them beyond their max Vds will cause premature part failure. IM most likely did receive drive boards that had less than perfect parts installed on them, but in my opinion, it was unwise to spec a 100V part in the first place. Never should have happened.

I am not expecting IM to formally acknowledge that this is a design problem because as I've said before, I don't have a schematic and it may not actually be the issue. But if this is a case of a lack of sufficient design margin, IM acknowledging this and revamping their design review process would go a very long way to restoring my trust that I can climb on one of their wheels and expect to come home with all of my skin still on my body.

Fortunately, the person that was injured probably didn't suffer permanent harm.

Edited January 8, 2022 by Tawpie
add a bit of ranting

[Tawpie]

By the way, I feel the same way about KingSong saying my 16X can go 28+ mph and implying that it's safe. It doesn't have the design margin to be safe at that speed and they shouldn't be implying to people that it is. My 67V MTen shouldn't be going 14 mph at 46% battery either, but it's a GW so I sort of knew the 'speed' number meant operating in the danger zone.

The takeaway from KingSong's lack of margin is "don't trust the max performance number, it'll probably put you on your face". The IM takeaway is "don't ride the thing at all unless you like a 1 in much-greater-than-zero chance it'll cutout at any time".

Edited January 8, 2022 by Tawpie

[Unventor]

21 minutes ago, Tawpie said:

This is what has my panties in a knot. It certainly appears from the part manufacturer's specification sheet that IM is using the parts incorrectly, and it's generally understood that using them beyond their max Vds will cause premature part failure. IM most likely did receive drive boards that had less than perfect parts installed on them, but in my opinion, it was unwise to spec a 100V part in the first place. Never should have happened.

I am not expecting IM to formally acknowledge that this is a design problem because as I've said before, I don't have a schematic and it may not actually be the issue. But if this is a case of a lack of sufficient design margin, IM acknowledging this and revamping their design review process would go a very long way to restoring my trust that I can climb on one of their wheels and expect to come home with all of my skin still on my body.

Fortunately, the person that was injured probably didn't suffer permanent harm.

From what I have seen so far on telegram Inmotion do not this the mosfet are too weak if they performed as specified. 

As of the KS16X it can go at 50kmh but you can't max accelerate to that limit as it is easy to overshoot that limit if you do. It is a fine example of a speed limit pushed up by speed addicts but was not designed for that limit. 

I do not think that is the case with V12. A guy just posted he has been riding around in Paris pushing top speed at hitting bumps without any issues. And while doing this pushing passed 5500+watt on the board. 

[supercurio]

31 minutes ago, Unventor said:

From what I have seen so far on telegram Inmotion do not this the mosfet are too weak if they performed as specified. 

Here is the "Safe operating area" as specified by the V12 MOSFET manufacturer in its datasheet (page 6).
10² = 100V.
As you can see, anything above 100V is outside the safe operating area; any on-time, any current.

Adding to this, each time the wheel slows its rotation quickly due to a bump or going up a curb, the sudden reduction of back-EMF associated results in a peak of current until the controller corrects its output. This might be the mechanism which led to the real-world failures.
The inability to control the output quickly enough could indicate that there is a firmware tuning element to the issues observed.

Inmotion is free to make different claims in their statements, and there are possibly additional problems contributing to the failures, but it's unlikely to sway the opinion of those capable of reading a datasheet.

Edited January 8, 2022 by supercurio

[Unventor]

16 hours ago, supercurio said:

Inmotion is free to make different claims in their statements, and there are possibly additional problems contributing to the failures, but it's unlikely to sway the opinion of those capable of reading a datasheet.

Yet still so you claim to have full knowledge what goes on on the drive board?

We are talking apple and oranges here. I am pretty sure Inmotion are aware some people read data sheets. And that numbers on components are visual. I do admit I am not a design engineer. I acknowledge that someone could have made an error too. But I also know I have no way of understanding what goes on inside the drive board.

Now that said I hope Inmotion can help those effected by the problems being reported asap. I am also pretty sure they will test new boards much more as they want to squash these problems asap. rework on shipped out wheels will be a big cost and none benafit from this. By this I mean riders, resellers and Inmotion. And right now it is not something that can be hidden. the biggest damage is already done. But what will have a huge impact is how fast and easy Inmotion gets on top of this situation and riders can enjoy their wheels.

Edited January 8, 2022 by Unventor
Spell correction

[mrelwood]

1 hour ago, supercurio said:

capable of reading a datasheet

I know it’s not related nor comparable in any way, but for some reason this reminded me of the EUC tire max load capacity discussions… Before the MC tires entered the picture, most EUC tires either specified max load 75kg stamped right on the sidewall, or didn’t care to specify anything.

 

Btw, in my understanding the EUC motor is driven with a PWM signal, where the voltage specifies the speed and the pulse width specifies the amount of torque. Wouldn’t the full battery voltage then be passing the mosfets only at max lift speed (-back-EMF)?