Tesla Gigafactory Production Process Is Futuristic And Fascinating

SEP 24 2018 BY EVANNEX 36


Everybody’s got an opinion about Tesla’s prospects for profitable production of Model 3. Some of these are, shall we say, more informed than others. Of particular interest are the reports of financial analysts who’ve actually visited the company’s factories for a first-hand look at the production process.

*This article comes to us courtesy of EVANNEX (which also makes aftermarket Tesla accessories). Authored by Charles Morris. The opinions expressed in these articles are not necessarily our own at InsideEVs.

Above: Inside the lobby of the Tesla Gigafactory (Image: Motor Trend)

George Galliers, an analyst at Evercore ISI who visited Tesla’s Fremont factory in August, said that he “did not see anything to suggest that Model 3 cannot reach 6k units per week and 7k to 8k with very little incremental capital expenditure.”

More recently, investment management firm Worm Capital sent a team to visit Tesla’s Nevada Gigafactory. Worm analysts Eric Markowitz and Dan Crowley were treated to behind-the-scenes tours of each production wing of the massive battery factory, which is ultimately expected to become the largest building in the world.

Markowitz and Crowley believe that the all-important question of profitability hinges on how efficiently the company is able to manufacture its battery packs. After touring the facility, they feel “highly confident in Tesla’s production process. Previous bottlenecks appear to have been remedied, and we’re increasingly optimistic in Tesla’s ability to hit – and sustain – weekly production rates of 6,000 Model 3 battery units per week, and with new Grohmann machines, scale to [around] 8,000 per week with minimal additional capital investment.”

Contrary to the frequent media depictions of Tesla as a disorganized and unhappy ship, Worm’s team found “a positive energy and high degree of organization” at the Gigafactory.

Markowitz and Crowley believe that Tesla’s 2170 NCA battery cells, produced in partnership with Panasonic, give Tesla an advantage over its would-be competitors. “We also continue to believe that small cells are superior for several reasons, making it perplexing that many legacy OEMs continue down prismatic or pouch paths. Tesla is also proactive – not reactive – in its refinement of battery process, looking ahead many years for potential shortfalls in commodities. We believe this cannot be said for its competitors.”

Above: A look at 2170 battery cell production at the Gigafactory (Youtube: Tesla Garage via Tesla)

Messrs. M and C note that Tesla’s cells need to be both cost-efficient and energy-dense, and believe the process by which they are assembled and placed into battery modules and packs must ultimately be automated. The two analysts got a first-hand look at the automation Tesla is using to construct and integrate battery modules into Model 3 packs, and called it “a fascinating and futuristic system.”

“Cells from the Panasonic wing of the Gigafactory are carted over to assembly lines on self-driving vehicles. From there, cells are loaded into ‘Zones 0-4,’ where the cells are placed into modules and packs and tested for quality,” write the analysts.

Martin Viecha, head of Tesla investor relations, told the men from Worm that new equipment from Tesla subsidiary Grohmann Automation will help module production become three times faster and three times cheaper. Three machines will be sent to the Gigafactory around the end of Q3 or beginning of Q4. The new process was designed to alleviate a previous bottleneck in module production which delayed Model 3 production significantly.

Viecha also said that the Gigafactory is around 90% automated. Eventually, the company will strive to fully automate battery cell production, pack assembly, and drivetrain unit production.

Above: Model X overlooking the Tesla Gigafactory during the early days of construction (Image: Teslarati)

According to Viecha, Tesla will likely achieve a battery cell cost of $100 per kWh by the end of this year, assuming commodity prices remain stable.


Written by: Charles Morris. This article originally appeared in Charged. Source: Worm Capital

*Editor’s Note: EVANNEX, which also sells aftermarket gear for Teslas, has kindly allowed us to share some of its content with our readers, free of charge. Our thanks go out to EVANNEX. Check out the site here.

Categories: Battery Tech, Tesla

Tags: ,

Leave a Reply

36 Comments on "Tesla Gigafactory Production Process Is Futuristic And Fascinating"

newest oldest most voted

So they still haven’t fully addressed the production bottleneck as these three more machines are required. I guess since they have already achieved 5k packs per week, these will push capability to a much higher level, 3 x 5k = 15k / wk and that would be enough for the 10k/wk goal for Model 3 and 5k for S & X. Maybe we will see 2170 cells for S & X starting early 2019?

It’s not a bottleneck if it matches the production rate at other stages. Even without a bottleneck, further improvements are needed to continue increasing capacity, in line with ramping of capacity in Fremont.

AIUI the machines being three times faster doesn’t imply that they will produce three times as many packs — but rather, a similar number with a third of the effort…

Actually, the statement “Martin Viecha, head of Tesla investor relations, told the men from Worm that new equipment from Tesla subsidiary Grohmann Automation will help module production become three times faster and three times cheaper.” definitely means “three times faster”

“So they still haven’t fully addressed the production bottleneck as these three more machines are required.”

Do you understand that production of battery cells and battery packs is continually increasing, as Tesla continues to ramp up production of the Model 3? Steady growth in ramping up production, as Tesla is doing, which includes preventing bottlenecks from developing, isn’t something you do once and then stop; it’s an ongoing process.

Bottlenecks only happen when there is a failure to predict what places in the assembly line or production process won’t be able to handle an increase in production speed. The goal should be to avoid bottlenecks by building out capacity at these points before production increases too far.

According to the article: “Worm’s team found ‘a positive energy and high degree of organization’ at the Gigafactory.” If that’s true, then it seems most likely they are on top of things and are preventing any bottlenecks from developing.

2170 cells are 70mm high instead the 1865 which are 65mm. So they will not fit in the existing battery casings S and X.
Maybe a Model-S 2.0 with redesigned battery pack can accommodate 2170 some day in the future.

Currently Model S/X production is limited by the supply of 18650 cells coming in from Japan. But it’s not a good idea to redesign and switch S/X to 21700 cells at a time when Model 3 production is constrained by battery packs and drive trains derived from them. It is understandable that Tesla wants to improve productivity as much as possible before scaling by increasing capex.

Elon explicitly claimed that they can fit the 21700 cells in the same form factor.

The new design of the Model 3 modules saves some vertical space — so that’s perfectly plausible.

Of course they might choose to go for a more significant redesign anyway, to maximise benefits…

From other things I’ve read on this website I believe that the S and X would need a new battery pack to accommodate the 2170 cells which are taller than the older ones. so probably not until the S and X get a major update.

Interesting that the price comment referred to cells and not packs, but no mention of improvements to cell part of the business. I think Panasonic is expanding the cell manufacturing to increase production so maybe the price reduction anticipated is not due to further refinements as simply increasing scale. That would mean they must be very close to that target now like maybe $105.

I guess the cell production wasn’t covered, since it’s mostly Panasonic’s business rather than Tesla’s… I’m sure they are further refining the processes — but the fundamental situation is probably not different than any other producer of cylindrical cells across the industry.

It’s a good thing if they can even hold cell prices where they are considering the increasing prices of cobalt and lithium.

The Panasonic/Tesla batteries don’t contain Cobalt. Transfer material used instead is Aluminum.

NCA is nickel, cobalt, aluminum.

Tesla talks about eliminating cobalt in the future. For now they use 110-120g/kWh.

My question is how many battery packs do we have to build till we see gasoline demand go down in some cities?

What also will devour a lot of batteries are all these utility scale storage projects along with homes.

By my estimates, global residental/grid storage needs will only be a fraction of what’s needed for cars. Storage batteries usually only cover a fraction of one day’s total electricity demand; while EVs usually store several days’ worth of energy.

(Seasonal storage needs might very possibly be larger — but that will likely require entirely different technology.)

It’s possible though that Tesla might grab a larger chunk of the storage market than of the automotive market, thus increasing the relative capacity needs of their storage business…

Demand response with EVs involved and V2G might also reduce grid storage needs.

I’m very sceptical about V2G, since automotive batteries aren’t really optimised for this extra cycling — installing dedicated storage batteries is probably more economical.

Demand response through smart EV charging however definitely is among the things that should keep demand for grid storage down 🙂

I saw an article or two about how it’s already having an noticeable effect in Norway, so whatever the current % of EVs to ICE is there. (10-20%?)

Yes, storage will devour a lot of batteries, but since they’ll often be backing up renewables or replacing gas peaker plants, then they help reduce emissions. And once used (non_tesla) EV batteries start to become available, those can be used for storage applications too.

Loads…it has barely affected Norway yet.

Norway hasn’t been into EVs in a big way for long at all, and V2G hasn’t started yet. It will start to happen soon and will come fast.

What does gasoline demand have to do with V2G?…

Norway actually has been into EVs for a long time — it’s just that only fairly recently they started becoming attractive to mainstream customers…

“My question is how many battery packs do we have to build till we see gasoline demand go down in some cities?”

So long as EVs are less than 2% of the total automobile market, they are not going to have a noticeable impact. The year-to-year fluctuation in automobile sales, as the economy improves or declines, creates more variance than that. Also, the U.S. population at last count was growing by 0.7% per year, so that will cause a year-on-year increase in overall auto sales which would have to be overcome by increasing EV sales.

Give it a few more years until we start to see EV sales have a measurable impact on U.S. demand for gasoline. Worldwide it will take somewhat longer, as world population — and industrialization in high-population countries such as India and China — are increasing at a much faster rate than here in the USA.

On the other hand, big Chinese cities are much more serious about pushing EVs than most other places in the world… So I wouldn’t be surprised for them to actually see an effect sooner.

BEVs are (very roughly) 2% of new car sales worldwide, but this has been growing quickly. If you assume a steady 36% growth every year, then BEVs should be about 4% of new car sales in 2020, 8% in 2022, and 16% in 2024 — and one in six new cars being BEV should create a noticeable effect. By 2030 all new vehicles sold should be BEVs. This is going to happen quickly.

Also, note that ICE vehicles will continue to be on the roads after 2030, but they will disappear rapidly — as demand for fuel drops, so will the supply of fuel. These are very approximate not-based-in-concrete-research numbers — expect the growth of BEVs to be a bit slower at the beginning and MUCH faster at the end. Expect the price of fuel to skyrocket starting sometime after 2030, becoming virtually unavailable at any price by 2040.

There is NO WAY that all new vehicles will be EVs by 2030 or by 2050 for that matter. They will continue growing certainly, especially world wide but exponential growth at 36% per year is not maintainable.

Why not?

A slightly different retelling of the same story has run here before — but there are some bits that I didn’t pay that much attention to before, and already had forgotten about…

The bit about different cell formats is particularly interesting IMHO 🙂

Wonder if next Model S comes with the 2170 cells and what the cars´ range will be in that case.

I expect the Models S and X will be switched over to the new 2170 cells either late this year or early next year. But I wouldn’t expect any dramatic improvement in range. I expect a slight improvement due to increased efficiency thru less wiring, but not that much.

Keep in mind that 90 or 100 kWh of batteries is still 90 or 100 kWh of batteries, even if the pack is made at a lower cost and designed/assembled more efficiently.

I see the main improvement with a newer 2170 battery design is that the charging performance will be improved. Today when you compare the M3 with the MS/MX in Supercharging capabilities, the Model3 wins. This would greatly increase the speed on road trips, the best I could obtain (including charging) was about 55 mph.

Charging speed improvement could be due to improved coming.

There’s no reason for 18650 chemistry to be different than 2170 at this point.

We all do 🙂

I would expect the next major iteration of the S and X (2022?) would update to 2170 cells. By that time I’d expect the module-level price to be substantially below $100/kWh, and cars will likely come with more kWh at the entry level price (its hard to guess this far out, but I’d say 90 and 120kWh packs).

I don’t know if they want to re-engineer the cars any earlier since the cells are taller and would likely need lots of changes around the pack and cooling.

It would make sense for Tesla to use their Model 3 rear engine (or its derivatives) as much as possible as they are easier to manufacture and have lower raw material cost (less copper and rare earth). They could put two M3 switched reluctance batteries in the rear and the induction motor in the front like for Roadster. Roadster will still retain performance edge due to lower weight

I can’t imagine they will wait that long. Panasonic is apparently not scaling production of 18650 cells for Tesla further. Unless Tesla expects Model S and X sales to stagnate at the current level, it just wouldn’t makes sense.

Strange how when analysts actually take the time to see a company with their own eyes and do a little research/observation that they form an accurate view of that which they’re trying to ‘analyze.’