My 84-volt Nikola Triumphs, Tribulations, and Failures

[fryman]

I am assuming the glue issue will fix the problem; but the lack of proper overheating monitoring is crazy. The wheel should have overheated and tilted Marty off.

[stephen]

10 minutes ago, fryman said:

I am assuming the glue issue will fix the problem; but the lack of proper overheating monitoring is crazy. The wheel should have overheated and tilted Marty off.

Yeah i think its  just how a pc works but in this case there was nothing monitoring the heat so it shut down (faceplant) and damaged mosfets because of the heat sensor placement ,, correct me if I'm wrong I'm still learning

Can you use a CPU without thermal paste? 

So a bad paste job can do a lot more harm than it does good. Your CPU will run hotterwithout paste than with, but it will still run. All modern CPUs are designed to throttle themselves if they sense they're overheating, to prevent damage

Edited June 22, 2019 by stephen

[mrelwood]

3 hours ago, Michael Tucker said:

This type of problem should be of no surprise as its from a first batch. Every first batch run of every model has problems. The glue was simply used because they were in a hurry to rush their new 16 incher to market

If you were more familiar with electronics, you would understand that this is not a ”whoopsey, missed that because we rushed a bit”. This is someone who’s unaware of how Mosfet cooling works, deciding how the GW electronics are built. Being rushed, or any other excuse in the world, doesn’t begin to explain what GW has done. Car companies don’t leave out seatbelts to get the product out faster. Air

 

3 hours ago, Michael Tucker said:

 The glue is not really the issue anyways, the small MOSFETs are the issue.

That is incorrect. Properly designed and properly built, the smaller Mosfets would’ve been fine. They should be able to handle the currents being used. But if they are not attached to a cooling plate, even bigger Mosfets than on the MSX would not survive.

 

3 hours ago, Michael Tucker said:

Small MOSFETs are a KNOWN problem and again right out of the gate we see a faceplant captured on camera. Marty's hill test only accelerated the burning up of the small MOSFETs before they would have blown later anyways, glue or no glue.

You are jumping to conclusions. The size of the Mosfets used in the Nikola are not a known problem. There have been many high performance wheels on the market with smaller Mosfets without issues. There are ZERO wheels (or any other product) with hot-glued Mosfets of any size in the market without issues.

[Nils]

1 hour ago, Rehab1 said:

This level of professionalism and customer service should be a norm for all EUC manufacturers.  Unfortunately that not the case.

Inmotion USA came through for their customers in the end (and was willing to help me out with a DYI kit, which I didn't end up getting but very much appreciated at any rate). However,  it should be said that it was a long and frustrating wait before we V10F owners got any official information regarding the issue (the history of this whole affair is now documented for future Internet archeologists in the 154 page thread "The "New Inmotion V10 / V10F"). Can't recall seeing much of anything from Inmotion globally for the issue either. My point is that there's still time to improve markedly on Inmotion's handling of the water ingress issue, not that I really expect Gotway though (but sure would be nice!). Remains to be seen how this unfolds.

So far for me, KS seems to lead the customer response league (thanks to @US69 for instance). That's from a from a manufacturer perspective of course, from an general customer perspective eWheel seems to be the leader by a good margin.

[Lilvodka]

Anyone know how widespread this issue is, does it effect the newest Tesla?

[houseofjob]

A late thanks to @Marty Backe for uncovering this issue and @Phil McLaughlin for the confirm and insight.

For those more knowledgeable than I, do people think maybe using thermal paste, ala computer CPU-to-cooling connection, might transmit heat much better? And would this last long-term, or need occasional re-application? Or is there a better solution?

The one time I ripped into an EUC board itself, I don't remember seeing much in the way of some intermediary agent between MOSFETs and heatsink.

[yourtoys7]

Huge thanks for all your work @Marty Backe and others for all your work, A+. I got really excited about your video @Marty Backe and started to think I might want one, after the incident I just hope that the Nikola Plus will use MSX setup. Time will tell, till than MSX is one solid machine to enjoy...

[Hansolo]

57 minutes ago, houseofjob said:

For those more knowledgeable than I, do people think maybe using thermal paste, ala computer CPU-to-cooling connection, might transmit heat much better? And would this last long-term, or need occasional re-application? Or is there a better solution?

No, the metal part of the mosfets must be electrically insulated from the cooler, if not short-circuited. This is the role of the thermal pad in addition to transporting heat.

[Afeez Kay]

I found this video of a modified Nikola with MSX board 

 

Edited June 22, 2019 by Afeez Kay

[MaxLinux]

1 hour ago, mrelwood said:

There are ZERO wheels (or any other product) with hot-glued Mosfets of any size in the market without issues.

I MARVEL at the IGNORANCE of Gotway and/or the outsourced company that built these boards! Doesn't this show a TOTAL lack of understanding about the heat transfer needs of MOSFETs?? I know less about MOSFETs than most people on this forum, but I wouldn't dream of using hot glue!!

[Marty Backe]

3 hours ago, Rehab1 said:

The more I read this thread the angrier I get.

Aside from @Jason McNeil working feverishly on finding a resolution for Nikola’s main board failure has there been any official GW representation on this forum addressing the issue? NO

Has GW issued any official written statement regarding the defective mainboard? NO!

Other than Jason has GW offered any form compensation to it’s customers such as free repairs or parts or shipping? NO!

If you compare the handling of this situation to how Inmotion handled their waterproofing vulnerability issues last year the result is night and day. 

Can you even imagine GW issuing an official  statement like the one IM issued last year?

 

<snip>

As of today (Saturday), Gotway has not responded to Jason's queries, to the best of my knowledge. They appear to be playing Deaf, Dumb, and Blind.

Jason can elaborate further if he wants, but there has been at least one additional catastrophic board failure that looks like my board (same exploded MOSFETs and glue). Presumably this user was not riding up Overheat-Hill.

So this is a real concern folks. Somehow Gotway needs to feel the pressure.

I think it's probably safe to ride mellow-like, but if you have one I wouldn't push it. But that's me.

[Marty Backe]

2 hours ago, Topkek said:

Anyone know how widespread this issue is, does it effect the newest Tesla?

This is conjecture on my part, but I believe every Gotway control board that has a middle row of MOSFETs that's inaccessible from the top, has hot-glue on some of the MOSFETs. Based on the MSX experience (no failures), the boards that utilize the larger MOSFETs seem to be able to overcome the thermal insulation of the hot-glue. I'd still prefer that there wasn't hot-glue.

I can't remember the configuration of the old or new Tesla board, but I believe that it's the older style where the MOSFET's are on top of each other. So there would be no need to use hot-glue during assembly.

[Marty Backe]

7 hours ago, Michael Tucker said:

This type of problem should be of no surprise as its from a first batch. Every first batch run of every model has problems. The glue was simply used because they were in a hurry to rush their new 16 incher to market before KingSong, many elements of the Nicola seem rushed. The glue is not really the issue anyways, the small MOSFETs are the issue. The glue can simply be skipped for future batches, but those small MOSFETs are there in the design of the 84V.

Future buyers may not be able to see if glue was used, but they can see the small MOSFETs. Who will trust them now...will it blow after 100 miles, 500miles, 1000 miles?!!! Small MOSFETs are a KNOWN problem and again right out of the gate we see a faceplant captured on camera. Marty's hill test only accelerated the burning up of the small MOSFETs before they would have blown later anyways, glue or no glue.

Small MOSFETs should be reserved for casual riding only, and with glue as a new possible risk, who would buy/ride a first batch 84V Nicola? Gotway Daredevils of course!!!

I totally agree with what @mrelwood wrote above, responding to what you wrote here. Sorry but I disagree with you on all fronts. I won't repeat what @mrelwood already stated so elegantly. 

[Marty Backe]

3 hours ago, fryman said:

I am assuming the glue issue will fix the problem; but the lack of proper overheating monitoring is crazy. The wheel should have overheated and tilted Marty off.

In a perfect world, yes. But even if their temp sensor was better placed, the board probably would have fried. There are 12 MOSFETs in play here. Only 2 or 3 were burning hot. So it would have been a hot spot among a relatively cool board. Boom!

[Marty Backe]

3 hours ago, stephen said:

Yeah i think its  just how a pc works but in this case there was nothing monitoring the heat so it shut down (faceplant) and damaged mosfets because of the heat sensor placement ,, correct me if I'm wrong I'm still learning

Can you use a CPU without thermal paste? 

So a bad paste job can do a lot more harm than it does good. Your CPU will run hotterwithout paste than with, but it will still run. All modern CPUs are designed to throttle themselves if they sense they're overheating, to prevent damage

Only if there were temperature sensors for each MOSFET would this have been detectable by the controller software. Otherwise, the 2 or 3 MOSFETS with the hot-glue appear as a localized hot-spot on the control board, but the overall control board appears relatively cool.

As has already been stated, thermal paste can't be used in this situation because the MOSFETs must be electrically isolated.

[Marty Backe]

8 hours ago, Nils said:

Oh man, not a one-off then..  Thanks for looking at this and updating us, and I'd be very interested to see a video of your rework! While I don't believe I'll do this myself, it would be very interesting at any rate. Did you happen to shoot any video of your disassembly and inspection? A shame that this should be so hard to inspect without destroying the board, perhaps some tool like @Harold Farrenkopf suggested could work.

So, how widespread are we to assume that this is then? Sub-set of batches? First batch? All 84V batches? 100V batches as well? Gotway need to provide some kind of official response here. Also, as Marty noted it might be possible that same thing is happening with other Gotway wheels, and that it's just the bigger MOSFETs saving them.

Argh, another year, another wheel, another problem, I'm having a bad streak it seems. Those waiting for the KS16X should be lucky I didn't go with that, if I had it would probably be discovered that they built it out of some new cost-saving material that turned out to spontaneously combust when rubbing it wrong. 

"Mo wheels mo problems"

I would bet anything that if we looked under the middle row of MOSFETs on the MSX control board, we would find hot-glue. If only we could find someone with a dead MSX board (for other reasons maybe) and have a look under those MOSFETs.

I'm positive as to why the hot-glue was used, so by inference we know which control boards are affected.

[meepmeepmayer]

37 minutes ago, Marty Backe said:

As of today (Saturday), Gotway has not responded to Jason's queries, to the best of my knowledge. They appear to be playing Deaf, Dumb, and Blind.

Well, it's the weekend, but still...

[Marty Backe]

13 minutes ago, meepmeepmayer said:

Well, it's the weekend, but still...

He's been trying since last weekend

[Phil McLaughlin]

5 hours ago, Harold Farrenkopf said:

I assume you know about the potential time bomb of correcting this and riding it with this board?  Those 6 mosfets have likely been stressed by operating hot and could fail easier than normal. 

Yes, I share your concern. 4 of the MOSFETs were potentially thermally compromised. This is the main reason that I put off rebuild until Monday. I need to decide whether to replace the affected MOSFETs. Not that difficult since I have a full rework facility. My current thinking is that the components have not been critically stressed at this point. I can have no certainty that this is a correct assumption.

[houseofjob]

3 hours ago, Hansolo said:

No, the metal part of the mosfets must be electrically insulated from the cooler, if not short-circuited. This is the role of the thermal pad in addition to transporting heat.

Thanks!

[AtlasP]

All I've read the past year was how Gotway "used to" have worse build quality and quality control issues leading to higher risk riding them, but how they're supposedly getting better now. Even if you can point to a couple examples of Gotway doing somewhat better on one or two recent wheels like the MSX, overall it looks like the old stereotype still remains true of Gotway as a company as a general rule. Meaning every new wheel of theirs should be assumed to fall under the old stereotype unless/until after a couple years if that particular wheel proves to be one to add to the small list of exceptions.

If you like analogies:

• InMotion = Apple = behind on specs but good production values/lots of polish (alternatively: shiny and scratches easy ;-) ), a straightforward product line with only a few clearly-defined/delineated models, (at least North American branch of IM) more accountable than most (like Apple, not immune to hardware fsck-ups, but relatively rarer and they have actually done recalls occasionally when necessary)
• King Song = Samsung = higher specs, overall solid/rugged hardware construction, some weird quirks people put up with for the higher specs and ruggedness (the King Song whine and poor companion app are sort of the equivalent of some of the janky stuff Samsung has done for example with their UI skin over the years), sprawling product lines but with at least some order behind it, more accountable than most other Asian hardware manufacturers but not quite all the way/still dealing with an Asian business mentality which can bite you on occasion
• Gotway = no-name Asian Android OEM's = rush to cram in the highest specs without concern for much else, often shoddy construction, no cohesive product lines or naming conventions, no accountability -- at the end of the day, yes you can get a no-name Asian Android handset with the most cpu cores and/or the highest clock speeds, but you have to be willing to live with a lot of cut corners for it (although at least if such a phone overheats/fails it won't throw you to faceplant on the pavement)

Of course in all cases, the two universal truths remain:

1. While we are all relatively early adopters, you at least don't want to be an "earliest" adopter (what I am hereby declaring the "let Marty Backe fall down first" principle ;-p )
2. Everyone worships ewheels.com and Jason for very good reason, as the only surefire 'insurance'/protection against EUC manufacturer shenanigans.

__

(For reference, here is an excellent write-up on the history of wheel issues by the various manufacturers: 

 )

Edited November 16, 2019 by AtlasP

[Michael Tucker]

5 hours ago, mrelwood said:

If you were more familiar with electronics, you would understand that this is not a ”whoopsey, missed that because we rushed a bit”. This is someone who’s unaware of how Mosfet cooling works, deciding how the GW electronics are built. Being rushed, or any other excuse in the world, doesn’t begin to explain what GW has done. Car companies don’t leave out seatbelts to get the product out faster. Air

It seems Gotway, KingSong, Inmotion, and NineBot all rush to get their first batch to market and make mistakes in that rush. I think Gotway has a long established reputation of low build quality and questionable choices in build materials. Even the mounting of the Nicola board in the cavity seems very hastily designed. Of all the brands who would be suspected of hiding/not caring about first batch problems (even if the problem is design related) Gotway would be the obvious one. Many people who are familiar with Gotway products would use the word AMATEURISH!

That is incorrect. Properly designed and properly built, the smaller Mosfets would’ve been fine. They should be able to handle the currents being used. But if they are not attached to a cooling plate, even bigger Mosfets than on the MSX would not survive.

In the past all the manufacturers used small MOSFETs because the belief was that properly designed and properly built small MOSFETs would be fine, but of course that was not the case and the move to TO-247 is much safer as real life use in MSX/Monster/MCM5/KS18X(L) has shown. The danger with small MOSFETs is that they mostly work...MOSTLY! Every once in a while they don't; hard to trust. Marty believes that the glue use is on MSX type boards as well because of the hidden second row of MOSFETs. 

You are jumping to conclusions. The size of the Mosfets used in the Nikola are not a known problem. There have been many high performance wheels on the market with smaller Mosfets without issues. There are ZERO wheels (or any other product) with hot-glued Mosfets of any size in the market without issues.

I believe the general excitement with the release of the MSX/KS18L type boards was that it was "Over-engineered" with the TO-247. Sometimes jumping to conclusions on the side of safety is better than relying on "sound principles of design and build" with little room for errors like "glue" or fan failures. Recently I have seen posts here about how motor watt rating is irrelevant, small MOSFETs are no problem, and charging your wheel at 10 amps or higher is totally safe. I think EUCs are complex and built with some secrecy and experimentation. These attributes are dangerous and I think promoting safety and over-engineering is better than attempting to apply perfect science to an imperfect industry.

 

[Jason McNeil]

Apologies for the late attendance to the party. Since receiving a 2nd board failure (this case was different, it was powered on from a stand still & didn't respond), including Marty's, within two weeks of shipping out the first batch of Nikolas, I've reached out to Linnea on Thursday to try to come up with an action plan to see what options are available to greatly reduce/eliminate the risk of these MOSFET thermal overload events.

The evidence from both of these failures has marked similarity in outcome (see below), where this glue may be acting as a thermal barrier, not helped by the absence of thermal paste on the underside of the metal heat transfer surface. One option being explored, is to recall the boards. remove this glue, & apply the thermal paste. To date Gotway have shipped 600 units; they say they have not received reports from other regions of this failure—it's likely more will trickle in in time. Putting out an official recall is not something to be taken lightly; in their defense, they're probably looking for a few other reports, before taking this step.

Gotway have a lot of experience with building high power motor vehicles; they had gone through several rough patches, particularly with reliability of earlier generation boards, with the weaker MOSFETS, but at least then they were pretty consistent with applying liberal quantities of thermal paste to this underside. It was baffling to me why they would go back to an inferior TO22x FET package AND now we learn about the lack of thermal paste. From my experience, such oversights are usually not the result of penny-pinching, but execution on the production line, where the factory needs a ruthless eagle-eyed floor manager, to see that every step of assembly is done by the book. 

The silver-lining to all this, is that an early finding in the production cycle is a necessary facet to a small-scale non-ISO production facility, where the manufacturer's financial pain of having to deal with a crisis, means that they will be imbued with a sense that this situation must never occur again. 

For all current Nikola owners, please hang in there, we should have an action plan in the next few days. 

EDIT: To anticipate questions on the 100V Nikolas, they are using the TO247 MOSFET package. The first production of these special edition 1845Wh variants is still underway, so there is assurance that in light of this episode, they will getting the thermal paste treatment!

Edited June 23, 2019 by Jason McNeil

[Han Vo]

27 minutes ago, Jason McNeil said:

Apologies for the late attendance to the party. Since receiving a 2nd board failure (this case was different, it was powered on from a stand still & didn't respond), including Marty's, within two weeks of shipping out the first batch of Nikolas, I've reached out to Linnea on Thursday to try to come up with an action plan to see what options are available to greatly reduce/eliminate the risk of these MOSFET thermal overload events.

The evidence from both of these failures has marked similarity in outcome (see below), where this glue may be acting as a thermal barrier, not helped by the absence of thermal paste on the underside of the metal heat transfer surface. One option being explored, is to recall the boards. remove this glue, & apply the thermal paste. To date Gotway have shipped 600 units; they say they have not received reports from other regions of this failure—it's likely more will trickle in in time. Putting out an official recall is not something to be taken lightly; in their defense, they're probably looking for a few other reports, before taking this step.

Gotway have a lot of experience with building high power motor vehicles; they had gone through several rough patches, particularly with reliability of earlier generation boards, with the weaker MOSFETS, but at least then they were pretty consistent with applying liberal quantities of thermal paste to this underside. It was baffling to me why they would go back to an inferior TO22x FET package AND now we learn about the lack of thermal paste. From my experience, such oversights are usually not the result of penny-pinching, but execution on the production line, where the factory needs a ruthless eagle-eyed floor manager, to see that every step of assembly is done by the book. 

The silver-lining to all this, is that an early finding in the production cycle is a necessary facet to a small-scale non-ISO production facility, where the manufacturer's financial pain of having to deal with a crisis, means that they will be imbued with a sense that this situation must never occur again. 

For all current Nikola owners, please hang in there, we should have an action plan in the next few days. 

EDIT: To anticipate questions on the 100V Nikolas, they are using the TO247 MOSFET package. The first production of these special edition 1845Wh variants is still underway, so there is assurance that in light of this episode, they will getting the thermal paste treatment!

And this is why i chose to buy from @Jason McNeil. It might be my last Gotway but not my last purchase from Ewheels with this level of customer service. Thanks to @Marty Backe for such a thorough analysis of the problem.

Edited June 23, 2019 by Han Vo

[Unventor]

@Jason McNeil thank you for sharing inputs and explaining the situation. I think that is one of the best posts I read in a long time.

Yes it doesn't give an answer, but explaining the situation to buy a little time before an action plan is in place is a smart move. 

Why people buy from AliExpress with the issue popping up in byound me.

Having a dealer like Jason to breach the culture differences and taking discussions with manufacturers like he does is simply but nothing less than priceless.

Now I am not directly impacted by this issue with the Nikola, but as a community it hurts to see QC, design choices and bugs, coming up again and again. 

Thank you for the status report.