INMOTION V12

[Cress]

2 hours ago, Wolverine said:

A Facebook group called Inmotion V11 Owners has an interesting new post. According to the post and comments, Inmotion V11 has problems with power mosfets as it is with V12. The author of the post seems to have two V11 unicycles, and both have broken for the same reason. As it seemed suspicious, I decided to investigate. I contacted a couple of resellers and gave them a link to this specific Facebook post. I got the answer that resellers don't know anything about this situation (no one hasn’t contacted with them regarding to this problem). They also contacted with Inmotion and manufacturer said that the V11 has no problems with power mosfets. At this point, I would like to remind everyone that everything on Facebook may not be true. If it seems suspicious, it is worth investigating, just as I did. Inmotion V11 is still a very good choice to buy.

V11 failures could be nothing more than 'background noise' but . . . I haven’t seen relevant facts, I'm always interested.  Digital media forums spread lots of information and it’s tedious to sort through.  Social media could make the EUC community recognize infrequent failures as EUCs age, then inform manufacturing choices

It's fair to say Inmotion EUCs are good value for their price.  BUT . . . If V11 motor controllers begin to fail the EUC community should track failures, catalog information for each failure and inform Inmotion.  Inmotion used the same switching transistors, MOSFETs, in motor controllers for several EUC models while motors got bigger.  Switching transistors have limited life expectancy in all applications and the safest choice is to maximize operating ‘headroom’ so that the switching transistor outlives the product it is serving.

Some choices in EUC manufacturing are made by the manufacturer and not by engineering staff.  The EUCs we're buying make EUC riders the product development team.  Let's remind manufacturers that we appreciate conservative manufacturing choices that limit our risk. 

 

[enaon]

6 minutes ago, Boostnsvt said:

I dont understand what you mean when you said that "an electrical connection is very possible when using just paste".

no part of the fet must touch the heatsink, else a short will happen. But on the v11 there is no pad really, it is an electric insulator, it depends on paste for hear transfer. 

[Richardo]

@Boostnsvt

on mine, they made a sandwich of white thermal paste to stick a rubber insulator to the wheel’s heatsink, and then more thermal paste to stick the mosfets to the rubber insulator.
 

The insulator is to prevent electrical connection between the mosfet’s heatsink/screw tab and the wheel’s heatsink. 
 

you can see pics from my teardown here

i assumed the “rubber” was actually a silicone thermal pad but didnt test it. 

[RagingGrandpa]

1 hour ago, enaon said:

v11 do not use a pad for the mosfets

Really?

I think V11 used a grey-color, non-electrically-conductive heat transfer pad:

... and V12 used a pink pad (plus unnecessary paste):

[enaon]

2 hours ago, RagingGrandpa said:

Really?

I think V11 used a grey-color, non-electrically-conductive heat transfer pad:

I will take some photos during daytime so that they are clear, my phone is not very good, but they have two types.

1) A very thin, elastic material that is for fets insulation, maybe a pad, but nothing like I have ever seen before, so I do not think it is really a pad,

2) and a pad, a very thick one, for the phase resistors, they are right bellow the fets. 

 

it looks a bit like in your photo, only it is very thin and very elastic to be a pad, a sub mm pad is very easy to break, this is not like that. Maybe it is a type I have not seen before and I rush to conclusions. But there was paste as well for sure, common heat paste, not liquid rubber as Boostnsvt saw on his. 

ps. the solder job on the fets on your photo is a lot better than the one I have here, mine is sub pro. 

ps2, I have this board because on a sudden stop, two fets were gone, the two fuses under the paper near the capacitors are gone, and one leg of the small capacitor next to the logic board (glued with gorilla glue type -polyurethane?- for some mysterious reason) connector has vaporized.  

Edited January 26, 2022 by enaon

[Boostnsvt]

43 minutes ago, Richardo said:

@Boostnsvt

on mine, they made a sandwich of white thermal paste to stick a rubber insulator to the wheel’s heatsink, and then more thermal paste to stick the mosfets to the rubber insulator.
 

The insulator is to prevent electrical connection between the mosfet’s heatsink/screw tab and the wheel’s heatsink. 
 

you can see pics from my teardown here

i assumed the “rubber” was actually a silicone thermal pad but didnt test it. 

I had a thick blotch of the white stuff and can confirm it isnt. It is Liquid Electrical Tape. I have a can of star brite liquid electrical tape and I was able to replicate the same thing on the pad. 

After searching for the best application to use as a heat dissipator; pads, paste, or those mica discs, I found that none of them are supposed to be used together, so my first statement was correct. The rubber substance (liquid electrical tape) that covered over 50% of the pad drastically increases heat accumulation and retainability. Im almost certain if this method was used on more sensitive electric device like a CPU it would be working on borrowed time.

Edited January 25, 2022 by Boostnsvt

[Paul A]

Maybe Cecily needs to be pinged...

[Shane Dougherty]

5 hours ago, Boostnsvt said:

I just pulled my thermal pad off the mosfets and found the rubber glue was used DIRECTLY on the thermal pad or mosfets. If the purpose of the pad is to conduct and disperse heat, a rubber barrier would inhibit that drastically. I am in the process of replacing the pad for paste.

Any possibility of showing a photo of what you mean.  Maybe a before and after the repair? 

[Richardo]

@Boostnsvt Did it look different than what I had? 

 

insulator tape

 

what the insulator tape looked like with thermal paste apple

 

 

[Nostris]

On 1/25/2022 at 7:13 AM, RArtem said:

Jus't don't disassemble the wheel if you don't know what your are doing. You have at least directional thread pattern on your tire and nipple for pumping from only one side on wheel. Wiring is also different from two sides. It has only 3 phase wires (one from the side, and two from the others). So just make photos during disassembly or mark everything with stickers or marker.

Agreed. I also wondered if Chooch had now got the tyre mounted backwards after reversing the motor.  Always best to mark everything clearly when dismantling so that you know you are assembling it correctly. 

[Funky]

On 1/25/2022 at 8:13 AM, RArtem said:

Jus't don't disassemble the wheel if you don't know what your are doing. You have at least directional thread pattern on your tire and nipple for pumping from only one side on wheel. Wiring is also different from two sides. It has only 3 phase wires (one from the side, and two from the others). So just make photos during disassembly or mark everything with stickers or marker.

The thing is, he connected the "wires" correctly, just the motor was other way around.. True about tire direction doh. Then again some tires don't have direction.

I could have made same mistake, if wheel don't show specific way to put back together. Having simple "L" "R" would help a lot.

Edited January 26, 2022 by Funky

[jimjam.nyc]

When i changed my tire. I took some pics of how things were hooked up. With the v12 its a bit weird because i noticed the motor cables actually crossed each other. I actually questioned myself a few times when putting it back together if i was doing it correctly. 

Has there been any other word on this meeting with inmotion and the outcome of the issues with the mosfets? Or any other reported cutouts? It kind seems like it quieted down quite a bit..

[beveik]

26 minutes ago, jimjam.nyc said:

Has there been any other word on this meeting with inmotion and the outcome of the issues with the mosfets? Or any other reported cutouts? It kind seems like it quieted down quite a bit..

Chinese New Year is about to start and i guess we will not hear much from them for about 16 days.
Happy New Year, i guess:)

edit: as for myself, I think i will wait for a Chinese New Year to finish and inmotion coming back with a second solution to this crisis. My main issue: returning the trust of the wheel that it won't cut out at 60kmh. The first solution doesn't help in this regard.

Edited January 26, 2022 by beveik

[Jeff Earl]

On 1/20/2022 at 7:30 AM, supercurio said:

A lot was discussed so it'll take a good amount of time to write a recap here, I'll do that later today or tomorrow.

 

Hey @supercurio, have you had a chance to summarize your thoughts on the Inmotion meeting with @Cecily Inmotion & team? Curious minds wanna know...

[supercurio]

19 hours ago, Jeff Earl said:

Hey @supercurio, have you had a chance to summarize your thoughts on the Inmotion meeting with @Cecily Inmotion & team? Curious minds wanna know...

I ended taking a bit of time off this after focusing a bit excessively on this matter, in the meantime I was on a podcast recorded with the UK distributor PET but it has not been published yet https://www.youtube.com/c/PersonalelectrictransportUk/

No news since but MADpack @Marco Domingos just released an interesting video reflecting Inmotion's public views in their answers.
I'm not always sure how to take MADpack's content since he is an Inmotion influencer disclosing it only rarely, but the Q&A itself is objective.

In this video, I both completely agree with some of his opinions (like Inmotion should offer to replace all boards), completely disagrees with others (recommending to avoid testing a wheel which already has some distance on), and some of Inmotion's answers here contradict what came out in our meeting.

 

So thanks for the reminder @Jeff Earl, I'll do this writeup today after work.

Edited January 27, 2022 by supercurio

[Steve Evans]

i think my preference at this point is to have inmotion send me a new board with the beefed-up mosfets so i can install it myself. that may not be everyones fav option esp if you are not a DIYer.

[Marc Roberts]

Still not clear to me from the various videos etc. if they actually *are* beefing up the MOSFETs on the new boards/HT version...

[Richardo]

I wanna transcribe that MADpak Q&A because a lot of those questions don't seem like they were answered, but maybe we could piece together something from what we have, if it were in textual format

[conecones]

That video from MADpack is great that it gets straight to the questions that has been asked multiple times.

I disagree with the thinking that repeating the spin tests is going to weaken an otherwise "good" board. In fact I think the proposed test is not aggressive enough. In the ideal situation, we should be able to stress test the machine until it cuts out. This cut out should be due to a designed limitation of the board (over amp/voltage/temp) therefore cutting the power. If the cut out occurs as a result of component failure (can't turn back on), then that's how the defects are truly weeded out. From the official spin test video, the wheel is stressed but it does not shut off which means it is impossible to know just how hard you've actually stressed the components because it never hit that limit/cut out.

If this was a Gotway I can understand the oversight/lack of design, and you can probably stress test the thing until it starts smoking, but we all expect Inmotion to be an industry leader in safety.

I'm sure Inmotion knows this already and this is why they are developing the APP function to do this test properly. The spin test should only be used as a general guideline to weed out obviously defective boards, but it likely isn't accurate enough to weed out those boards that are just shy of meeting specs. 

I am riding the repaired V12 (new mobo) with confidence but am also treating it gently until at least the APP test comes out - but realistically I will only be 100% confident once the new board is installed and tested. I think for the Batch 1/2 customers who pass the APP-run spin test, they should get at least a discount on a new board - a free replacement is a bit much if the existing board can be proven to be safe. 

[Funky]

13 minutes ago, conecones said:

a free replacement is a bit much if the existing board can be proven to be safe. 

Latter in news: Rider dies, because of broken board that "passed" so called test..

Test is test, it don't reflect real life situation, with real rider.. Are test made for 70kg rider, or for 130kg rider. There's many things that test can miss..

Edited January 27, 2022 by Funky

[conecones]

6 minutes ago, Funky said:

Latter in news: Rider dies, because of broken board that "passed" so called test..

Test is test, it don't reflect real life situation, with real rider.. Are test made for 70kg rider, or for 130kg rider. There's many things that test can miss..

I don't think you understand my point. There is no test in any field that can account for all variables. Passing a test just means there is proof that it can withstand the predetermined set of variables that are being tested. Of course this threshold for what is "safe" may change depending on individual but for the current EUC design, having a test where the wheel is stressed until cut out due to over amp/volt/temp and NOT as a result of something burning is a realistic and IMO all we can ask for realistically. This is already better than what other companies are doing. If a batch 1/2 board can demonstrate this, then I would feel safe riding it because the cut out would be due to surpassing a designed limit (aka rider fault) and not as a result of component failure (QC problem).

Ex. Veteran Sherman is known to fry the board if you keep riding it uphill because of a poorly placed temperature sensor not kicking in soon enough. Doesn't seem like anyone is foreshadowing deaths because of this. Same with the Gotway boards burning up as a result of allowing too much power (ex., RS 1st gen green boards vs black boards where they limited power draw.)

It might be useful to step back and look at how Inmotion is so actively engaged in improving safety through this mistake instead of doing what other companies are doing which is sweeping it under the rug. Rather than imposing unrealistic expectations on them such as demanding a kind of test that can prevent all rider deaths due to board failure, we need to accept what is realistic. Boards can fail due to component age as well, and this is a risk we all accept in the current EUC design with no-backup boards. 

[Steve Evans]

1 hour ago, Marc Roberts said:

Still not clear to me from the various videos etc. if they actually *are* beefing up the MOSFETs on the new boards/HT version...

i may have misunderstood madpacks q&a's but i thought inmotion did say they will be using more robust mosfets (dont remember the exact words they used)

[Paradox]

53 minutes ago, evans036 said:

i may have misunderstood madpacks q&a's but i thought inmotion did say they will be using more robust mosfets (dont remember the exact words they used)

In the ad for the V12 HT it states "more durable MOS"

[meepmeepmayer]

"More durable" and "bigger" are not the same. Could just be a different supplier with different specs, but the same size. If this is a real picture of the new board, the MOSFETs are the same size as before.

Nobody here knows if it's a real picture.

[Tawpie]

"More durable" has zero meaning to me... it's more 'durable' than what, a crepe?

I have to make allowances for not-first-language writers, but to an English as a first language person (me) all it -says- is that they've done something to make it less prone wearing out due to something that they didn't state. It's new! Improved! It could well have a different epoxy body formulation that isn't as prone to degrading under intense UV light, or uses laser etched part marking so the specifics of the actual part don't rub off as easily (markings are pretty faint on some pictures). It might be able to withstand higher soldering temperatures, or have better heat transfer characteristics.

Or it might have a safe operating region that's more appropriate for the designed use.