INMOTION V12

[Denny Paul]

I’d imagine you’d want to do these tests at a full charge right? Really ensure that those mosfets aren’t going to blow close to their rated voltage?

[Steve Evans]

1 minute ago, Denny Paul said:

I’d imagine you’d want to do these tests at a full charge right? Really ensure that those mosfets aren’t going to blow close to their rated voltage?

yes, wheel was fully charged

 

[nosamplesplease]

23 minutes ago, supercurio said:

It reads like you executed the test in good form... @evans036

Denis Hagov and @eevees both were doing pendulums to make the Abrams fail. Would that test put more load on the motor/MOSFETS and get closer to the limits?

Or does that fall in the category of wheel abuse that few wheels are expected to survive for long?

Also, shouldn't the wheel be at or near full charge to avoid software limiting due to battery state?

[Steve Evans]

1 minute ago, nosamplesplease said:

Denis Hagov and @eevees both were doing pendulums to make the Abrams fail. Would that test put more load on the motor/MOSFETS and get closer to the limits?

Or does that fall in the category of wheel abuse that few wheels are expected to survive for long?

Also, shouldn't the wheel be at or near full charge to avoid software limiting due to battery state?

yeah, i was wondering the same. would those 'pendulums' break a healthy wheel?

[supercurio]

12 minutes ago, nosamplesplease said:

Denis Hagov and @eevees both were doing pendulums to make the Abrams fail. Would that test put more load on the motor/MOSFETS and get closer to the limits?

Or does that fall in the category of wheel abuse that few wheels are expected to survive for long?

Also, shouldn't the wheel be at or near full charge to avoid software limiting due to battery state?

Seeing several attempts, including from professionnals in the field completely unable to reproduce the overpower condition on the Abrams was a valuable indicator: it's not easy to trigger for everyone, and might require skills riders only develop with agressive riding, technical off-road or skate park practice.

Plus if it fails, you fall.

I don't know how Inmotion came up with the test but in my mind it might simulate the triggers reported so far, where wheel rotation speed can change very quickly either because of an obstacle or because it's briefly spinning in the air then slowed down by the ground instantly when regaining grip.

So not only about overpower but also abrupt change in rotation speed (which has a control algorithm component)

Efficacy still TBD.

Edited January 5, 2022 by supercurio

[Tawpie]

Penduluming is hard on the MOSFETs because the the motor is pulling near max current (it's not spinning so there's no back EMF). It'll heat things up quickly, and will of course stress all the parts. I wouldn't expect it to kill a properly designed and operating wheel, but I don't do it to mine (on purpose) because it is an unusual operating condition and... I like riding, not repairing. I have been practicing 3 point turns on the MTen which is kind of like penduluming, so far knocking on wood seems to be working.

Edited January 5, 2022 by Tawpie

[nosamplesplease]

6 minutes ago, Tawpie said:

Penduluming is hard on the MOSFETs because the the motor is pulling near max current (it's not spinning so there's no back EMF). It'll heat things up quickly, and will of course stress all the parts. I wouldn't expect it to kill a properly designed and operating wheel, but I don't do it to mine (on purpose) because it is an unusual operating condition and... I like riding, not repairing. I have been practicing 3 point turns on the MTen which is kind of like penduluming, so far knocking on wood seems to be working.

Gotcha, it seems like Inmotion's test is to try and recreate similar conditions to the failures that occurred with minimal to light usage(descending a curb).

Doesn't really tell us much besides which wheel might fail.

Testing wheels at extreme usage really aught to be the way right?

So pendulums cause extreme current demands. Therefore to determine if the MOSFETS are rated at too low a voltage we need to create high voltage conditions?

Things like regenerative braking downhill? (How could this be done safely?)

[supercurio]

20 minutes ago, evans036 said:

yeah, i was wondering the same. would those 'pendulums' break a healthy wheel?

Note that the Abrams cutout wasn't triggered by pendulum (which resulting stress is mostly heat buildup until overheating) but by briefly overpowering the wheel, which would cutout instead of regaining balance.

For these V12 failures, it could be something else as we can see really well in the video of the gentleman riding slowly onto a moderate curb.

https://youtu.be/eddv3BZw71I

 

Edited January 5, 2022 by supercurio

[.....]

48 minutes ago, nosamplesplease said:

Gotcha, it seems like Inmotion's test is to try and recreate similar conditions to the failures that occurred with minimal to light usage(descending a curb).

Doesn't really tell us much besides which wheel might fail.

Testing wheels at extreme usage really aught to be the way right?

So pendulums cause extreme current demands. Therefore to determine if the MOSFETS are rated at too low a voltage we need to create high voltage conditions?

Things like regenerative braking downhill? (How could this be done safely?)

Breakin periods and extreme conditions testing has been a want of MANY of us for ALL models. Some (non euc)companies can afford to test each device AND run it to help with the bell curve of electronic failure. It takes time and equipment and room, to conduct such testing. It would be a WONDERFUL thing, and would probably cut down on a LOT of user problems. It wont catch ALL failure, as things just happen sometimes. However, it can be used to minimize failures and also track any trends from the equipment that DOESNT pass testing. Testing should include max load, speed changes, temperature saturation and other cycles. Its an expensive endeavor, but some companies bother. I've seen rows of products being made to run for days and weeks, prior to shipping. Companies that do this, typically do this so they can remain a premium name and to lessen losses. If you have GREAT warranty, return service and reputation, failures cost you a LOT more than it would a company that has none of those things. Top tier companies CANNOT afford to lose the quality they are synonomous for. Especially when competition is lurking around every corner. The euc market is a little slim, so the competition is less than fierce and the sales volume is less than stellar. Business models vary and ina  world market, theres a TON of factors in each decision.

Edited January 5, 2022 by ShanesPlanet

[Tawpie]

Like this?

 

That company put 100% of their product on a multi level conveyor for two days—operating and bouncing along. If it still worked, then it went to a customer.

But, it was horribly expensive and one wonders how they could possibly survive. It’s no way to run a successful business, stressing your product like that. Think of the lost revenue while they went round and round. And who paid for the rejects?

[nosamplesplease]

17 minutes ago, ShanesPlanet said:

Breakin periods and extreme conditions testing has been a want of MANY of us for ALL models. Some (non euc)companies can afford to test each device AND run it to help with the bell curve of electronic failure. It takes time and equipment and room, to conduct such testing. It would be a WONDERFUL thing, and would probably cut down on a LOT of user problems. It wont catch ALL failure, as things just happen sometimes. However, it can be used to minimize failures and also track any trends from the equipment that DOESNT pass testing. Testing should include max load, speed changes, temperature saturation and other cycles. Its an expensive endeavor, but some companies bother. I've seen rows of products being made to run for days and weeks, prior to shipping. Companies that do this, typically do this so they can remain a premium name and to lessen losses. If you have GREAT warranty, return service and reputation, failures cost you a LOT more than it would a company that has none of those things. Top tier companies CANNOT afford to lose the quality they are synonomous for. Especially when competition is lurking around every corner. The euc market is a little slim, so the competition is less than fierce and the sales volume is less than stellar. Business models vary and ina  world market, theres a TON of factors in each decision.

Absolutely, we would like THEM to do these tests prior to releasing these wheels 

I was trying to point out that IMO the test that these users have been trying doesn't tell us much about the engineering choice of MOSFETs in question. Just recreates known conditions of failure.

Would like to see the control boards of these wheels though not all component failures are identifiable via visual inspection. 

But the fun ones are.

[Gixxer]

The EE I talked to was shocked (pun intended) that they were not speced for 50% over bus voltage

Although he works on big targeting simulators for the government so some very overspeced mosfets add very little to the cost of that system.

Edited January 5, 2022 by Gixxer

[Patrick Robert]

5 hours ago, supercurio said:

But what we get instead today is that the risk with Gotways was escalated to house fire, and Inmotions are not safer or more reliable than Gotways from the past.

And for the Veteran Abrams, we get a wheel with potentially bad hall sensors to boot. Again. Remember all those E HALL error codes on the early Shermans? (had one myself) 

Can't wait for those detailed S20 reviews! Only KingSong can save us now 🤣

Edited January 5, 2022 by Patrick Robert

[Waulnut]

In addition to the spin test, would there be some sort of data that can be seen through EUCworld or even from the inmotion app? Something to put a number on. 

[lirva]

I don't mind buying first batch wheels and being QC tester if the price better and parts are available. Now I don't know if the replaced board (that took 2 months to arrive) is any better.

Here is my commute, workplace have a curb before main entrance, which didn't kill the wheel, but after hitting it, screen lights and bluetooth died. 

I guess I just go hit curbs with the wheel over and over again.

Edited January 5, 2022 by lirva

[Patrick Robert]

1 hour ago, ShanesPlanet said:

Breakin periods and extreme conditions testing has been a want of MANY of us for ALL models.

True. You want to test overall behavior of the beast. And they already do, to some extent. But there are many levels of testing. 

I've been watching a lot of Tesla (the car) videos over the holidays, and one thing seems to work for them: they design almost everything in house. Down to the batteries, as we all know. 

Other car manufacturers use parts from external suppliers, and often don't bother repeating QA on the individual parts. Hence higher numbers of quality issues.

At least, individual parts should be (re) tested systematically by the integrator (the EUC manufacturer) to ensure reliability. All it takes is a bad hall sensor, a bad mosfet, or a bad cell. 

By "bad", I mean slightly below spec in some circumstances for that individual unit. Perfection is so rare, better to test and filter for it! 

[Khazik]

22 minutes ago, lirva said:

screen lights and bluetooth died. 

If that's all that happened, I'd highly suggest checking (Edit: and re-seating) the Bluetooth antenna connection to the touchscreen board, and that boards 4-pin connector to the circuit board. My issue taught me the V12 is very proprietary - to the point it'll instantly re-lock, or fail to save settings because of said antenna connector, so no telling what other issues may come from loose connections. 

Edited January 5, 2022 by Khazik
and re-seating

[Paul A]

One pseudo solution would be:

Don't be an early adopter/beta tester/guinea pig purchaser of a new model wheel.

 

Despite all the testing, QC, simulations, designing etc, there seems to be always undiscovered problems.

A recurring theme, across many industries/products.

[onizukagto]

7 hours ago, Waulnut said:

Would this mean Inmotion does not do independent test on controllers to make sure they function over certain limits? 

Besides having guinea pig testers, a manufacturer should have some equipment to test parts for failures. 

I'd only feel safer with a revamped board with bigger MOSFETs if that's the case. 

As someone who works in "China" and deal with Chinese companies with products, prototyping and general engineering designs. 

It's very very rare to have an independent body do any type of QC or auditing of said products. It's almost always done "in-house", the only reason a company will do pay for an independent body is because the market they intend to sell to, demands it.

I. E. EU, Japan, S. Korea. 

 

It's just a cost they won't do unless it's compelling. Also people forget, EUC is still a "niches" market. 

Is they were in any size like Lenovo, LG or Samsung, they will invest in proper certification. 

 

One option though, is if the local dealers demand it. 

If the customers make it expensive or make enough noise, dealers can get Chinese manufacturers to do third party certification. 

Until that happens though, manufacturers liek gotway/Begode can just make superficial "safety" changes and no really engineering revision and keep selling faulty designed products to customers who don't care and will keep throwing money at them. 

[onizukagto]

10 hours ago, lirva said:

I don't mind buying first batch wheels and being QC tester if the price better and parts are available. Now I don't know if the replaced board (that took 2 months to arrive) is any better.

Here is my commute, workplace have a curb before main entrance, which didn't kill the wheel, but after hitting it, screen lights and bluetooth died. 

I guess I just go hit curbs with the wheel over and over again.

You know what, if EUC companies advertise 1st and 2nd batch as QC testing, sell for less then planned msrp with no warranty, but will be willing to sell replacement components. 

I'm sure people will be more than happy to buy.

You are informed, you know the risk. You get a new wheel before everyone else and companies get the feedback they need. No bitching, no crybabies, you die? You signed up for it. 

Everyone wins. It's like kickstarter, everyone knows those products are barely prototypes, hell, they are just happy to get something sent their way, no one buys expecting a polished product. Lol

Edited January 5, 2022 by onizukagto

[Djal]

Ok, done for me. Change my V12 order to a V11. Don't want to risk my life or to wait an hypotetical solution from inmotion...

[Timwheel]

3 hours ago, onizukagto said:

You know what, if EUC companies advertise 1st and 2nd batch as QC testing, sell for less then planned msrp with no warranty, but will be willing to sell replacement components. 

I'm sure people will be more than happy to buy.

You are informed, you know the risk. You get a new wheel before everyone else and companies get the feedback they need. No bitching, no crybabies, you die? You signed up for it. 

Everyone wins. It's like kickstarter, everyone knows those products are barely prototypes, hell, they are just happy to get something sent their way, no does that expecting a polished product. Lol

I'm not sure I like the "You die ? You signed up for it". I know some people are daredevil, but if an engineer designs an unsafe product, he doesn't sleep at night.

At least I don't. I'm an engineer, I worked on barely driveable hybrid GT cars, the sheer torque was enough to get you to 200 kph in well under 10 seconds. The rear crossmember was not supposed to be my design, but I had to implement two very powerful pancake motors (with the drivetrain, gearing, driveshaft, ESC...) on the rear end while trusting the design of an other engineer. The power of those engines combined was around 400 kW, absolute madness, with insane amounts of torque ( if you had seen the driveshafts and gearings necessary to hold that much torque...), and there was an ICE engine in the front with a third 200kW electric motor.

Well, I saw that crossmember, and I immediatly knew it would get completely shredded to pieces at 10% of the torque. I didn't think "Hey, that's not my problem, I signed up for the motors". I redesigned said crossmember at night, using the excuse of a very "nice to have" quick mount version that would, supposedly, help me design a tougher drivetrain. I didn't bruise any ego, didn't say "Hey you are a dangerous engineer, did you get your degree in a cereal pack ?". I did what I had to do to ensure the test driver wouldn't die.

He signed up for it right ? Test driving is dangerous, we all know that. But would I be able to live with the death of that man on my mind ? No. Period.

Tim

[supercurio]

4 hours ago, Paul A said:

One pseudo solution would be:

Don't be an early adopter/beta tester/guinea pig purchaser of a new model wheel.

 

Despite all the testing, QC, simulations, designing etc, there seems to be always undiscovered problems.

A recurring theme, across many industries/products.

Don't be an early adopter & guinea pig, let the other guy be the one who crash and get hurt on an early batch.

Problem solved? 🤔

[Paul A]

Pseudo solution.

Problem not solved at all.

 

pseudo

/ˈsjuːdəʊ/

 

derogatory

adjective

adjective: pseudo

1. not genuine; spurious or sham.

   "we are talking about real journalists and not the pseudo kind"

Edited January 5, 2022 by Paul A
pseudo definition added

[supercurio]

When talking about QC testing, I always remember this old video from @Jason McNeil visiting a Kingsong factory. The video was published in June 2016.

At this timestamp in the video, we can see a bunch of wheels free-spinning on a rack.

They don't seem to be stress tested here however, with a predefined program simulating load by flywheel effect as we're talking about now.

Has anybody seen a similar video for Inmotion factory?