Tesla Employees Say Gigafactory Problems Are Worse Than Known (cnbc.com)

An anonymous reader quotes a report from CNBC: Tesla's problems with battery production at the company's Gigafactory in Sparks, Nevada, are worse than the company has acknowledged and could cause further delays and quality issues for the new Model 3, according to a number of current and former Tesla employees. These problems include Tesla needing to make some of the batteries by hand and borrowing scores of employees from one of its suppliers to help with this manual assembly, said these people. Tesla's future as a mass-market carmaker hinges on automated production of the Model 3, which more than 400,000 people have already reserved, paying $1,000 refundable fees to do so. The company has already delayed production, citing problems at the Gigafactory. On Nov. 1, 2017, CEO Elon Musk assured investors in an earnings call that Tesla was making strides to correct its manufacturing issues and get the Model 3 out. But more than a month later, in mid-December, Tesla was still making its Model 3 batteries partly by hand, according to current engineers and ex-Tesla employees who worked at the Gigafactory in recent months. They say Tesla had to "borrow" scores of employees from Panasonic, which is a partner in the Gigafactory and supplies lithium-ion battery cells, to help with this manual assembly. Tesla is still not close to mass producing batteries for the basic $35,000 model of this electric sedan, sources say.

Comments

It's a 10^9 factory

[Hognoxious]

It's a 10^9 factory, but they were expecting 2^30?

How is this different ...

[pablo_max]

How is that different than every other company in the world? I have worked with and for at a lot of places over the years and one thing is universal, most of the people have no idea at all what they are doing.
It is amazing to me that some companies are even able to put products on the market at all. I am not talking only about the small guys either.
I was once testing a wireless product for one of the largest companies in Europe for global radio certification (FCC/ISED/CE and many others). Once I got the devices I told them.. hey, thanks a lot for sending these samples, but it would be great if you could send them with a SMA connector so we could test the radios as well.
What is a SMA connector, was the response. After explaining it a couple of days went by and they called me up and explained that the guy who knows how to do that quit the company so it would be better if we changed the design for them to make it work.
Of course this kind of shit happens ever every company every single day. These are not things which people know about it.
So, you can say that 100% of companies are shittier than people on the outside know about.

Re:How is this different ...

[tomtomtom]

It's different because most companies realise this and try to underpromise and overdeliver publicly. Tesla has a track record of doing the opposite. And the stakes are also a lot higher than usual - the company's survival could depend on this single product.

Re:How is this different ...

[swb]

I think it's getting worse. We seem to have a growth in the amount of complexity of things individually and an increasing number of them, combined with a corresponding lack of investment in training.

As one kind of an example, an IT department 15 years ago had simpler networks, servers and software to manage. Now each of those things is much more complex than it used to be but the number of people managing it is the same and they probably don't know any more details than they did when it was simpler. High level management (virtualization, etc) may have made managing larger breadth easier, but I think the individual complexity has been addressed at all and lots of it is essentially not understood.

Re:How is this different ...

[burtosis]

I came to the same startling conclusion after talking to a "PhD" mechanical engineer at a medium sized contract manufacturer and discovering they couldn't explain the technical side how a hammer works.

Re:How is this different ...

[drinkypoo]

It's different because most companies realise this and try to underpromise and overdeliver publicly.

It's different because Tesla is operating on hype. Every other car company teases designs, but they only make actual claims about performance and scheduling when they are absolutely sure they can meet them. Tesla can lose money and the stock goes up, traditional automakers have to turn a profit just to keep it flat. But Tesla will die just as soon as the buzz does. If all the unpaid Tesla promoters like Rei stopped doing the work they're doing on behalf of Tesla, the company would fold up like a cheap magazine, because Wall Street would stop rewarding them for losing money.

Re: How is this different ...

[Anonymous Coward]

The difference is that Tesla always delivers, just slower than anticipated. All companies miss their deadlines. Most companies deliver over hyped products. Tesla always delivers what it promises. Reviews are consistently positive on their products. That's the difference

Re:How is this different ...

[Socguy]

Meh. A handful of fired employees saying bad things about their former employer... A bunch of short sellers on Wall Street currently in line to lose BILLIONS after shorting Tesla stock for months only to see the stock shoot up... not surprised at all to see a hatchet job like this.

Re: How is this different ...

[Solandri]

Tesla is deliberately delaying the Model 3. CARB (California Air Resources Board) created a ZEV mandate. A certain percentage of each car company's sales have to be ZEV - zero emissions vehicles. Right now that's almost entirely EVs (Toyota has a hydrogen fuel cell vehicle on the market). The percentage increases each year - the details are a bit complex but bottom line it's about 2% for 2018. If an automaker fails to reach the required percentage, they have to buy ZEV credits from an automaker which exceeded it. If they fail that too, they are banned from selling cars in California and the approx dozen states which automatically adopt CARB's guidelines. That's about 1/3 of the U.S. by population.

Since Tesla produces only EVs, they always have excess ZEV credits. Part of their finances is selling those ZEV credits. But the closer the other automakers come to meeting their ZEV requirement in a year, the lower the price for ZEV credits. So if Tesla produces too many EVs in a year in which other car companies sold enough of their own EVs, they get little to nothing for their ZEV credits, and they have to bear a larger fraction of the Tesla 3 production cost themselves.

You can tell how well EVs are selling by how good the discounts are at the end of the year. 2015, sales were really poor (relative to the ZEV mandate that year) and there were incredible discounts on EVs (in California - the only state where CARB counts sales/leases). Dec 2015 I almost picked up a 3-year lease on an e-Golf for $79/mo, no money down (there was also a $49/mo with $1500 down offer, but that's more money overall). The EV deals in late 2017 were close to nonexistent, which is a pretty good indicator that the automakers were hitting their ZEV mandate percentages. That means there wasn't much of a market for ZEV credits in 2017, which meant Tesla had to delay Model 3 production to try to push some of those credits into 2018. And that's exactly what they did.

The problem for Tesla is that they set the pre-order price of their EVs based on assumptions for how much they'll receive for selling the ZEV credits. If the other automakers consistently hit their ZEV percentage every year (or come close to it), Tesla is in a world of trouble - all those Tesla 3 pre-orders could have been "sold" for less than what it cost to manufacture because they'd assumed selling the ZEV credit would've made up the difference. So paradoxically, the better EVs sell, the worse off Tesla is financially.

Creating a product only half the battle

[Anonymous Coward]

I think maybe the worst part of Tesla is that nobody knows how to make a lot of vehicles efficiently and be profitable. Critics have said all along that Tesla needed someone in manufacturing that knew how to build cars. Instead Musk rejected this ideal and went it alone and it shows. Obviously critics said the real test for Tesla would be how it handle's the Model 3 production schedule. Its very clear from reports that they bit off more then they could chew.

Re:Creating a product only half the battle

[K. S. Kyosuke]

And how would employing more people who know how to build cars in particular help overcome problems with building a battery factory? I mean, besides their generic manufacturing experience.

Re:How many factories do not have any problem?

[Rei]

Rather...

This is an extremely misinformed and misleading article. To be absolutely clear, we are on track with the previous projections for achieving increased Model 3 production rates that we provided earlier this month. As has been well documented, until we reach full production, by definition some elements of the production process will be more manual. This is something Elon and JB discussed extensively on our Q3 earnings call, and it has no impact on the quality or safety of the batteries we’re producing. As noted in our Q4 deliveries release, during the fourth quarter, “we made major progress addressing Model 3 production bottlenecks, with our production rate increasing significantly towards the end of the quarter.”

Furthermore, as is often the case in manufacturing, some parts of the production process require the expertise of employees with engineering or manufacturing experience, and others don’t. We’ve created thousands of new high-quality jobs in Nevada in recent years. As we continue to expand Gigafactory 1 and ramp Model 3 production, we’ve been able to teach new skills to thousands of new employees, many of whom had no manufacturing experience prior to joining Tesla. New hires on the module line receive extensive training, including safety training, and learn about the importance of proper cell-to-cell spacing so they can identify such issues in the production process. More broadly, battery production – and the module line in particular – is overseen by our top engineering talent, and many of Tesla’s most senior leadership.

Finally, the implication that Tesla would ever deliver a car with a hazardous battery is absolutely inaccurate, contrary to all evidence, and detached from reality. It is irresponsible to suggest as much based on unnamed, anonymous sources who have provided no such evidence and who obviously do not have a complete understanding of the extensive testing that all batteries in Tesla vehicles are subjected to. As with Model S and Model X, which have well demonstrated safety records, we maintain a rigorous approach to quality and process control for the Model 3 battery. Even more importantly, to our knowledge, there has not been a single safety concern in the field related to Model 3 batteries at any point over the six months of Model 3 production.

As for the assertion about cells touching in Model 3 batteries, this is extremely misleading and displays a complete lack of basic knowledge about how our batteries work. Every battery in a Tesla vehicle has thousands of cells, the vast majority of which are at the same voltage potential as neighboring cells. Hypothetically, even if two cells of the same voltage potential were touching, there would be absolutely zero impact, safety or otherwise – it would be as if two neutral pieces of metal touched. Despite this fact, all Model 3 battery modules’ cell positions are measured twice in manufacturing to verify process control and quality of outgoing parts. Conversely, if at any point in the production process cells are touching at different voltage potentials, they cannot be electrically interconnected. Over the course of the production process, we conduct three different tests to ensure the right number of cells are electrically connected in Model 3 modules. Additionally, the long term reliability of cell position is something validated through testing, including shock and vibration, and high temperature and humidity testing, as well as thermal cycling endurance testing throughout design and via sampling in production. All of this testing is designed to prevent touching cells from being installed in any of our vehicles, including Model 3. Finally, the safety aspects of our module design would continue to function even in the presence of touching cells, so the concerns raised are further unfounded.

These false claims are being made even tho

Re:How many factories do not have any problem?

[Hognoxious]

What can I, AC, do?

You could try to rectify the situation.

Doesn't sound so bad...

[BobSteinVisiBone]

...as long as Gigafactory batteries are not composed *of* Panasonic employees.

Re:3.7 volt 18650's

[K. S. Kyosuke]

The funny thing is, the cells are 18650's

No, they're not. Chances are that the switch to a cell size that's more efficient in the long term has created problems for them in the short term.

Re:The first one is always the hardest.

[drinkypoo]

Building the first factory is the hardest part about building factories. Once you've built it, you can build 200 more just like it in a fraction of the time.

Except nobody builds lots of factories all the same. Automakers for example build different factories with different lines to produce different vehicles. The building is not the interesting part, the production line is. And the production lines are different for each vehicle. Also, by the time you've got the first factory completed, new techniques have been developed, and new equipment has hit the market. Maybe you've been just welding all your cars together, and now you're starting to use structural adhesives. Now you're going to change the line again.

Hatchet job of a story

[aaarrrgggh]

Those things are already well known; apparently they didn’t hit their stride until the end of December, where they were at a rate of 1000 model 3’s per week in the last three days. Timing now seems designed to hit the stock before earnings.

Also, the base $35k model is a random reference... of course the lowest margin version will be last.

Based on the fact that I have seen a few model 3’s on the road this past week (first ones for me), I am guessing production is consistent now and possibly accelerating beyond 1,000/week.

Re:Hatchet job of a story

[zurmikopa]

I agree, I'm pretty sure things are accelerating pretty rapidly.

I took delivery of my model 3 last week. When I was taking delivery in Bellevue I met a couple employees who were flown in from Texas to help out with the increased load. (They are moving from West to East in deliveries.)

There were at least 5 or 6 other Model 3s in the delivery area waiting for pickup.

I have 2 friends who have also gotten their notifications to do configuration of their orders. (It took about 2.5 weeks from when I got the notification until when I took delivery)

Not a surprise...

[QuietLagoon]

Companies always have internal problems that are not known outside the company. Companies also have management in place to address and resolve those problems. In a start-up situation, it is one problem after the other, sometimes many at once. If it weren't Tesla, it'd be a non-issue.

Re:How many factories do not have any problem?

[Anonymous Coward]

On track is PR BS. They are way behind waht they promised. They might eventually get there many years late if ever. In four years Tesla will have literally millions of competitive vehicles from manufacturers that know what they are doing.

Re:Anybody surprised?

[DNS-and-BIND]

LOL someone's never seen a factory start up before. I love these oh-so-wise internet commenters who come in with harsh rhetoric for other people, with zero knowledge of the topic at hand or how things usually work. I mean, seriously, "Musk and his band of semiskilled dimwits"? This is some kind of emotional release going on here.

Re:3.7 volt 18650's

[K. S. Kyosuke]

So you're saying that the Chinese have simply avoided high power density. Well in that case it's rather obvious that many of those things that Tesla is doing don't worry the Chinese. But I'm not sure why you have to contradict yourself on a space of two paragraphs ("small cells all require a lot of cooling ... bigger cells will require more cooling ... huge brick cells are air cooled"). And why would copper conductors be necessarily better when copper is heavy per unit of conductivity? And why do you think Tesla doesn't use massive internal conductors already? And how come that cathode swelling is suddenly such an issue for lifetime but suddenly thermal control isn't? Mind you, data shows that battery lifetime already isn't an issue even for the older and presumably worse Tesla packs, why should it be a worry in the future? And that's with the smaller cylindrical cells already, which you said are worse, not with the bigger ones.

Re:Anybody surprised?

[Balthisar]

I have seen factory startups before. It's what I've been doing for living since 1996. Automotive factory startups, that is. And, yeah, Tesla is a dysfunctional organization. This is organizational incompetence. That's not to say that they won't recover, but this is not normal startup pains. This is a collosal fuckup.

I'd suggest that you're correct about the semiskilled dimwits; Tesla has hired a lot of smart people. But without skilled leadership, you end up with the manufacturing capabilities of Tesla. (I mean operational leadership; this isn't a dig at Elon.)

Competition is scared of Tesla. Very scared.

[Qbertino]

I wouldn't be surprised if this weren't some hidden PR bullshit being spread by the competition. Do you remember the blatant lies about the first tests on the model s that were quickly debunked by the data provided by the test models? This has very much the same smell. There are reports of paid goons renting Teslas and deliberately mistreating them to put them out of service. This article is along these lines IMHO.

I'd trust Tesla and Musk more than I'd trust any news outlet, that's for sure.

Re:Competition is scared of Tesla. Very scared.

[DCFusor]

Well, my Volt refuses to catch fire despite all the propaganda about that before they came out. Yeah, they totalled one, stuffed into a junkyard upside down, and 3 weeks later it caught fire - somehow I think if I were trapped in a totalled car upside down for 3 weeks, starvation would be more of an issue...and they do drain gas tanks of normal cars before crash testing them.

Yet all this crap came out. ...I put some flame decals on mine just for humors sake.

Re:How many factories do not have any problem?

[Hognoxious]

would it have been much different if the IBM PC has been introduced a year later?

Possibly. In that time an alternate standard could have become entrenched.

Is this really news?

[sootman]

I think a bigger story would be "Company is pushing the envelope and nothing goes wrong at all."

GM

[JBMcB]

GM built it's own battery factory. Practically nobody knows about it. They make all of their own battery packs for their hybrid and pure EV vehicles. It came on-line on time and roughly at capacity.

GM hasn't run a large-scale battery operation like this but it managed to figure it out. Building the factory in an area already saturated with large factory operations probably helped out a bit. Building a factory in the middle of the desert, where the nearest, largest factory builds slot machines, probably is a hindrance.

Re:How many factories do not have any problem?

[vtcodger]

The problem isn't that the factory has teething problems. It is that based, on its quarterly reports and other public data, Tesla is on its way to running out of money. It really looks from outside like Tesla needs to start delivering a lot of Model 3s and making a reasonable profit on each if it expects to stay out of bankruptcy court.

Conventional wisdom seems to be that without some significant revenue stream, Tesla doesn't have enough cash and locked in credit to make it through 2018. Google turns up a plethora of articles on this. Are they accurate? How the hell would **I** know?

Re:Anybody surprised?

[hey!]

I suggest that there's a difference between bringing up the first factory of its kind and bringing up a factory which is just a variation on what's been done many times before. Even if an entire vehicle is fundamentally more complex than an enormous battery back, the number of novel solutions you need to come up with is probably a truer measure of engineering risk.

Re:Manufactory - industrial production

[Guy Smiley]

Electric motors are much more likely to last a long time than internal combustion engines. There are virtually no moving parts in contact with each other in an electric motor (I've read reports saying 18 moving parts in a Tesla drivetrain), and the motor(s) are directly driving the wheels. Compare that to an ICE with hundreds of moving/wearing parts (valves, pistons, seals, crankshaft, spark plugs, transmission, etc) that need to withstand high temperatures and low tolerances to seal against burning fuel, then convert the explosive force into rotational energy in a different part of the vehicle at varying speeds.

There are already reports of Tesla taxis hitting 250k miles and 300k miles with minimal service.

Highly biased article...

[denzacar]

Two main sources for the story are people who either "worked at the Gigafactory in recent months"... Past tense...

But more than a month later, in mid-December, Tesla was still making its Model 3 batteries partly by hand, according to current engineers and ex-Tesla employees who worked at the Gigafactory in recent months.

...aaaaand a guy with a huge "shorting" investment, standing to win millions from perceived losses by Tesla.

Stanphyl Capital's Mark B. Spiegel, who has a significant short position in the company, told CNBC:
"While I've no doubt that Tesla will eventually work out its Model 3 production problems, the base model will cost Tesla at least mid-$40,000s to build.
The company will never deliver more than a token few for less than the current $49,000 lowest-cost offering.
Sales will hugely disappoint relative to expectations of over 400,000 a year.
And even at those higher prices Tesla will never come anywhere close to its promised [profitability]."

Also, article is reeeeeaaalyyyy trying to paint a picture of doom and gloom.
It takes a line from a Tesla engineer about how workers were "slapping bandoliers together as fast as they possibly could" back in December - and presents it as a doom&gloom subtitle:
'Slapping bandoliers together'

Hell, it even manages to paint higher test standards as bad, by omission of the fact that test standards are higher than expected not simply "[not] the same kind".

The two engineers also said that Tesla doesn't do the same kind of "stress tests" of its Model 3 batteries which would be expected of other electronics or carmakers.

And then there's that thing where I can't seem to find a single article by that author, about Tesla, which isn't a story about how VERY DOUBLEPLUS BAD Tesla really is.

Feds to investigate Tesla crash driver blamed on Autopilot
Tesla factory workers have filed a lawsuit claiming widespread racism, unsafe conditions
https://www.cnbc.com/2017/11/01/elon-musk-tesla-fired-700-people.html
https://www.cnbc.com/2017/10/17/tesla-firings-former-and-current-employees-allege-layoffs.htmlTesla employees detail how they were fired, claim dismissals were not performance related
Tesla employees detail how and why they were fired
Tesla cites performance reviews as it fires SolarCity employees, though workers say reviews never took place
Tesla fires hundreds of employees while trying to ramp up vehicle production

German report calls Tesla's Autopilot a "hazard"
Senate committee calls out Elon Musk, wants answers on Tesla Autopilot
Tesla under investigation for possible breach of securities law, WSJ reports
What the NTSB know

Re:How many factories do not have any problem?

[Rei]

, Tesla is on its way to running out of money.

TTAC called, they'd like you to write a column for their Tesla Deathwach 7 years ago. ;) The "Tesla is going to run out of money" nonsense that shorts at Seeking Alpha love is tiring (but hey, if you buy into it, by all means short them!). It's premised on the concept of no additional cash streams (Semi and Roadster reservations are basically no-interest several-year loans), no improvements in the Tesla Energy division orders (which by all measures seem to be taking off after the success of the Australia battery), no revenue from the solar gigafactory (which is now starting deliveries), and they pretend that there's no revenue from Model 3, despite the fact that they're now up to 1k per week, reporting (as of today) to be on track to 2,5k by the end of Q1 and 5k by the end of Q2, and that Model 3 purchase prices / margins are frontended by delivering only the premium versions first.

The shorts' math is laughably bad. Which is why they keep losing money over and over on TSLA. They first really started hyping that argument in early November, when the stock was down to ~$300. It's at $343 right now. Care to lose your money like they have?