Inside Tesla's gigantic Gigafactory | Hacker News Reader

Inside Tesla's gigantic Gigafactory

bbc.co.uk

201 pointsjsingleton94 comments

The amount of publicity associated with this battery pack plant is a bit much. Tesla's battery factory is about the same size as LG's battery factory in Michigan.[1] That's the source of the Chevy Volt battery packs. (LG is adding a fourth production line for an undisclosed customer.) That's just one of LG's plants. They make the batteries for the Chevy Bolt in South Korea.

There are large EV battery plants in operation or under construction all over the place. US, UK, Germany, Korea, China... VW is planning a plant bigger than Tesla's full sized expansion.

Nissan has the largest lithium-ion battery plant in North America right now.[2] They're building a bigger one in the UK.

[1] http://www.greencarreports.com/news/1101606_ford-wants-to-de...

[2] https://www.youtube.com/watch?v=WBboW2Sw3Co

Your first link says nothing about the size of LG Chem's battery factory.

Sorry. Here's a picture.[1]

LG had a problem for a while - more manufacturing capacity than Chevy needed. They shut down the battery plant briefly. But that turned around.

[1] http://www.autoblog.com/2014/09/25/lg-chems-troubled-chevy-v...

That still doesn't lead me to believe that it is a larger battery factory than the Gigafactory, either in square footage(the Gigafactory will be the largest factory in the world when completed, not just battery factory) or in total battery output(the Gigafactory's output is obviously unknown, but there are at least design targets that put it at greater than the entire WORLD's 2014 battery production). It may be larger than the Gigafactory is now(14% complete) or produce more batteries(not very hard, since Gigafactory isn't producing any yet), but those aren't really meaningful comparisons when judging the newsworthiness of the Gigafactory.

If there are actual hard, numbers that put the Gigafactory's to shame out there, please share. As a Tesla investor I want to know :)

"the Gigafactory will be the largest factory in the world when completed"

No way. Not even close. One of the largest single buildings, yes. Largest factory, no. See Ford's River Rouge plant, which once was the largest factory in the world.[1] Today, Foxconn has a whole walled city.[2] (Foxconn makes 40% of the world's consumer electronics. They're getting into auto manufacturing.)

What Tesla has actually built is much smaller. Right now, LG has more battery factory in operation than Tesla does. Looking ahead, VW is talking about building a $15 billion battery factory, bigger than Tesla's.[3] VW can come up with $15 billion.

[1] https://goo.gl/maps/EvoyvV2HB3K2 [2] https://goo.gl/maps/LZKdTUpz8ck [3] http://www.techtimes.com/articles/161873/20160530/volkswagen...

Satellite map pictures are not a measure of size. And yes, I misspoke, it will be the largest building in the world by footprint/square footage, not the largest factory.

VW's plant is still in the conceptual stages, but if it ever happens it will indeed be impressive. But none of that makes the Gigafactory less impressive, especially once completed.

Largest factory building, but according to https://en.m.wikipedia.org/wiki/List_of_largest_buildings_in..., not the largest building in footprint. The Aalsmeer flower auction beats it by the relatively tiny margin of 8,000 square meters

By the way, the BBC recently had a documentary about the flower industry that I think is worth watching if you are interested in manufacturing and logistics. For Brits in England, it is on iPlayer at http://www.bbc.co.uk/programmes/b07czwfd, but Flora Holland seems to have it available for everyone at https://www.royalfloraholland.com/en/speciale-paginas/search....

In particular the parts about an army of marching lilies (starts about 26:50 in) and the way the auction clocks work (about 34:00 in) may be of interest to those interested in automation and efficiency (it seems they manage to make a sale every 5 seconds or so on one clock, with seven of them in the room)

That's a comparison with the Tesla Fremont factory, where they assemble the actual cars. The Gigafactory in Nevada will be 538,837 square meters(if my math is right, 5.8 million square feet for the Gigafactory is the number I have), handily beating the Aalsmeer flower auction at 518,000 sq meters.

Thanks. I guess I was wrong-footed because of limited knowledge of US geography and because Wikipedia (erroneously, I guess) states the Fremont factory is under construction.

Also, the Fremont factory is actually an old GM-Toyota factory.

Your first link made no mention of factory size, and after someone pointed this out, you've provided a picture of a concept drawing.

This does not inspire confidence in your assertions.

The difference is Tesla owns the factory and therefore there isn't a profit layer being skimmed off. That could be passed off, or even just a portion, to the end consumer who buys Tesla products; this has similar characteristics as to why Elon wants Tesla to buy SolarCity.

Actually, Panasonic makes the cells and owns that part of the factory. Tesla just puts them in a box. A fancy box with a titanium bottom, yes, but a box.

There's more to the battery system than cells in a box. The battery system needs to charge quickly without catching on fire, and after an accident that damages cells, needs to minimize the size of a fire.

I imagine that's a bridge for now. Once Tesla starts pumping out 500k vehicles per year they should be able to leverage that for large future expenditures.

Considering Panasonic is funding quite a bit of the factory that would be an unexpected a turn of events. I'm sure it's contracted for the foreseeable future.

Also considering that the manufacturing of these cells is a highly complex and closely guarded proprietary process, which Tesla knows next to nothing about, it's highly unlikely Tesla will ever be making cells.

Think of laptops: Tesla is like Lenovo; they buy CPUs from Intel, RAM from Corsair, SSDs from Samsung etc. Then they put all the parts in a box. Sure, it's a complex and nice looking box made from aluminium or carbon fiber, with quite a bit of custom electronics in it, but it's still a box. Nobody in their right mind thinks Lenovo is going to start manufacturing CPUs any time soon (or ever).

Elon seems to understands economies of scale and the value of owning or controlling a larger percentage of an ecosystem. It could certainly be in their timeline to become experts in batteries, or it is possible the costs by partnering with a third-party is better - at least for a time.

Are we sure Tesla has no hand in the manufacture or know-how of building the actual cells? If that is true, why would they be contracting Jeff Dahn to do battery research for them? It is my understanding that the types of partnerships between companies like the Gigafactory are usually designed to allow both companies to learn from the other as part of the process.

Of course we can't know without seeing the agreement between Panasonic and Tesla but it seems a little misguided to suggest Tesla has no hand in the design of the cells.

Why would Panasonic share their industrial secrets with Tesla? They have nothing to gain and plenty to lose from doing so. There are plenty of other EV manifacturers with equally large (or larger) need for cells, like VW, Nissan, Mitsubishi etc.

I won't speculate as to their motives, just the evidence I see that Tesla is deploying their own battery know-how to the project, so some sharing of some sort must be going on.

Edit: Found some info straight from the source[1].

Panasonic executive vice president Yoshihiko Yamada thought Elon and Co. were crazy initially when they proposed this idea. Eventually he saw the light and now the companies are closer than ever. Here’s how he put it: “Three years ago I thought this gigantic Gigafactory idea was crazy. Because then the production of the factory would exceed production of the industry. I thought it was a crazy idea. But I was crazy, and I was wrong after seeing extreme success of the announcement of the Model 3 and the strong demand…I want to explain to you the relationship with Tesla and Panasonic. I used to be in charge of components five or six years ago. At that time our relationship with Tesla was one of supplier and customer. A conventional business relationship.“But since we started discussion on the Gigafactory that’s completely changed. One example, is production capacity is now two or three times more. Why? Because Tesla and manufacturing people worked together. We are discussing these details. This type of relationship is quite new for business. We are not the simple buyer and supplier relationship.” (emphasis mine)

1. http://electrek.co/2016/07/28/tesla-gigafactory-tour-roundup...

Elon seems pretty smart when he is leveraging his position, and he seems to understand what is fair for what his position is. He might be locked in, or maybe for this location only, or only locked in until a point. I can't imagine in any scenario that he would agree to forever.

Well Panasonic literally owns big parts of the factory (and will be the ones making the batteries!). They are going to be partners for a very long time. In my opinion it's actually a lot more likely that Tesla goes bankrupt and it becomes a solely Panasonic factory than it does a solely Tesla factory.

I think your comparisons are misguided because you got the original claim wrong.

The claim is not about "largest factory". It is supposed to be the "Largest footprint of any building in the World" [1]. That is, the square footage of a single building.

Several buildings in the same area (which is usually what we understand by "factory", or maybe more accurately a "plant") is not the comparison you should be looking for if you want to disprove this.

[1] https://en.wikipedia.org/wiki/List_of_largest_buildings_in_t...

As an aside, I typically associate a single building with the word "factory", and a complex of buildings that work together to manufacture a product a "plant" or "facility". Not sure how common that usage is, or if the distinction really even matters outside these measuring contests.

I found some actual numbers for the LG Chem plant. They currently produce 650MWH worth of battery cells per year, and they expect to expand that to 3GWH within a couple of years.

This is definitely smaller than what is planned for Gigafactory.

http://www.autonews.com/article/20151214/OEM06/312149992/lg-...

Thanks. If that number is right, their top number is less than 1/5th of the originally planned Gigafactory top number.

Keyword: planned

I think you're either just Tesla's target market or align with it in media consumption. They certainly have a PR division putting in work but without HN practically no Tesla information would hit my eyes.

Now here's something interesting. Here's Tesla's Gigafactory and nearby buildings in Google Earth.[1] There are three buildings about 3000 feet northwest of Tesla's that are about the same size as Tesla's building. There's another big one about 400 feet to the west. What are they? This is an industrial park zoned for Really Big Rectangular Buildings, the Reno-Tahoe Industrial Park.

The building to the west is PetSmart's distribution center. It's almost as big as Tesla's building. Just to the west of it is a WalMart distribution center. That's two buildings, but they add up to about the size of Tesla's.

The identically-sized buildings to the northwest are the Eagle Valley Industrial Center.[2] CBRE seems to be building two rows of 500,000 square foot buildings and leasing them to companies that need distribution centers. They have Amazon, Diapers.com (owned by Amazon), chewey.com (not owned by Amazon), and other tenants in the third building. Markings on the ground indicate that buildings #4 and #5 are coming along soon. The distribution centers aren't built as big single buildings because they need lots of outer wall for loading docks. Zoom in and you can see that semitrailers are docked along most of the east and west exterior walls. That's a distribution center - stuff comes in one side, and goes out the other.

So, while Tesla does have the biggest single building in that industrial park, CBRE has more floor space than they do.

[1] https://goo.gl/maps/4rWXYqxa1Eo [2] http://ktrcapital.com/PDF/Brochure/NV%20-%20Eagle%20Valley%2...

They have only built like a tenth of the Gigafactory though.

Serious question: My impression is that when Verizon bought Yahoo (for 5 Billion), this was a pretty standard affair M&A. However, this Gigafactory has the same price tag but is met with the question "Does Elon Musk have more money than sense?" How are M&A purchases different from direct investments in a company?

Verizon currently makes a profit, Tesla doesn't[1].

1. They might be able to run a profit without R&D expenses, but that wouldn't make any sense at this point, so they are bleeding money and need to raise capital constantly.

Bleeding money and need to raise capital constantly?

-> Investing in infrastructure to meet the huge demand for their next generation electric automobiles, using common stock sales.

>Investing in infrastructure to meet the huge demand

Can you explain why Tesla didn't add to the capital raise in May? They couldn't foresee needing more capital for the M3 and GF a month-and-a-half ago? That seems odd.

Likely because Tesla leadership expects the stock price to be significantly higher in a few years, and they can make do on a smaller amount + proceeds from vehicle sales until then.

To be precise, the capital raise in May was for the gigafactory and Model 3. The expected capital raises in the future are for Tesla Energy expansion, plus future vehicles. That plan might change, but this is the latest official message.

Tesla doesn't make a profit, and they're spending billions on a factory that will make a valuable product with good margins.

Verizon makes a profit, and they're spending billions on a failed company.

I don't really see how the profit status of the acquirer is relevant. What matters is the profit potential of what's being acquired.

Tesla /might/ make a valuable product with good margins from the factory. I would rate that chance as greater than 90% personally, but it still isn't certain. Just like it isn't certain that Yahoo is a failed acquisition for Verizon, although you and I might rate the chance of that as quite high.

The profit of the acquirer is relevant in so far as it relates to the ability to finance massive acquisition projects such as these, and the market's willingness to do so. I'd say in both instances the market's willingness has been high, but people have good reasons to be skeptical both ways.

Profit and ability to finance aren't quite the same, though. So far, capital markets have been willing to give Tesla big piles of money with excellent terms (buying issued stock at a price that values the company similarly to major automakers and such) so they seem to have no trouble financing the Gigafactory.

Verizon may well have access to more money, but then that just supports the first half of "more money than sense."

Yahoo makes a profit as well. Verizon may have overpaid (I couldn't say), but they are getting a mature property which makes money.

I don't really know, but could it be that you can buy a company with mainly stock yet building things requires cash?

Panasonic has invested just under $2bn in the plant. I assume that's in cash.

Verizon is also paying cash for Yahoo.

Simple: Verizon earned last quarter $32 billion with profits of $4.31 billion. Even if the Yahoo aquisition turns out to be total worthless it doesn't threaten the company.

In comparison the Gigafactory is a big bet for Musk.

Investments in innovation are a much harder sell. The HN crowd is a minority. Most people alternate between fear and incredulity regarding anything new.

You also have to remember the people championing this deal probably have bonuses and other economic incentives that might run counter to the best interests of the company.

I call to your attention a series of three blogs written by someone who had the opportunity to interview Elon Musk himself before the writing.

I, for one, think that big thoughts, big visions, and big ability to implement them can't help but change the direction of humanity.

http://waitbutwhy.com/2015/05/elon-musk-the-worlds-raddest-m...

If you do the math, best case scenario: Tesla's budget model will have twice the range (~200 miles) of the Nissan Leaf at the same price point ($30k). This assumes that Tesla is capable of a $35k model today. My estimate for a 200 mile battery power pack today is around $20,000 going by what amateur EV enthusiasts have managed, using the same 18650 cells.

It's seems a little unlikely that the Leaf would see 0 progress.

There's also a Bolt in the room. It's pretty likely to ship this year, with a 200 mile range and competitive price point.

The Bolt is probably the closest competitor to what Tesla is doing right now for EVs. They have the range, but they still haven't figured out styling or performance, nor is the charging infrastructure as robust yet.

>The Bolt is probably the closest competitor to what Tesla is doing right now for EVs.

The Model S is 2x the price (and 10x the car). I don't think they are competition.

Or do you mean the Model 3? As of right now it's nothing but a prototype, so I'm not sure that's competition either. Theoretically it will be $30k with 200mile range, but I'm skeptical.

I meant the Model 3. Theoretically it will be $35k for a base model with 200 mile range, and the Bolt is also not much less theoretical than the Model 3 at this point. There are lots of reasons to be skeptical about the time frame of availability and hitting production numbers of the Model 3, but Tesla has been pretty good about hitting the raw performance and price numbers they announce.

Doesn't the Bolt use a fairly standard charging cable? I see those connectors in parking garages and grocery store parking lots.

There are adapters for Teslas to use basically every standardized(or otherwise) EV charging connector. No other brand has access to the long-range and fast-charging supercharger network however, which is a rather large part of the value proposition of owning a Tesla for many people, as it mostly removes the need to rent or own a vehicle for longer trips.

But the supercharger network is woefully inadequate for the number of Model 3 vehicles they intend to produce. Either Tesla is going to have to put a whole lot of money into its supercharger network (more than they've announced, certainly), or they'll have to restrict it to Model S owners.

I recall (at the M3 event) that Elon said what they plan to build (2X current stalls) by the time Model 3 is released. Of course that's inadequate for future demand; it seems that they're scaling the network linearly with the number of cars sold which have access to the network.

Unfortunately, most of the public charging stations out there are pretty pointless due to their slow charging speed.

Without faster charging, you can't do long-distance travel (nobody wants to stop for 4-8 hours in the middle of a road trip to recharge), so the only value they offer is for longer charging sessions while the driver is already in the area for work/shopping/etc. However, even then, you can't really rely on their availability, since it's always possible someone will already be there tying up the station for an 8-hour stretch...so they really aren't terribly useful or convenient for the vast majority of EV owners.

I would argue that the only charging network out there that really makes sense is Tesla's Supercharger network: it actually charges fast enough to make road trips practical, is generally quite reliable, and typically includes at least 6-8 stalls per station to avoid congestion during peak travel times. This gives Tesla a huge competitive advantage over other EV manufacturers, and I think it'll be a long time before anyone catches up to them in this regard.

A slower charging network could be useful for putting a charging station at every parking space at say workplaces, where your car will be left for an extended time Slower chargers are presumably cheaper to build and run, so they might make more sense there. But otherwise completely agreed that superchargers are basically the only type of charger that makes sense for 99% of the rest of use cases that a gas station would normally take care of.

Yeah if there were parking garages or lots where every parking space had a charging port, that would be awesome, and could actually be part of a practical solution in the long term. Unfortunately most garages I've seen have somewhere from zero to two spots in the entire garage :(

Facebook HQ has around 150 L2 chargers. It's just a matter of time.

I'm not exactly sure what facebook having 150 car chargers has to do with the rest of the world getting car chargers.

They aren't the only company with more than 0-2 chargers per parking lot, I could name a bunch of others. The point I'm making is that more chargers are on the way.

In Palo Alto, I can tell that there are a bunch of EVs that charge overnight at the L2 chargers in various city garages -- dunno if their apartment complex won't install chargers or if they're waiting to get a charger installed. Either way, it's extremely useful to a minority of EV owners.

The Bolt will also not be able to ship in nearly the volumes of Tesla Model 3.

Wikipedia has 2 sources that report that the plan is 25k-30k cars per year (https://en.wikipedia.org/wiki/Chevrolet_Bolt). Now that Model 3 received so many reservations, I'd guess that GM is planning on raising that number.

That's a good thing, but just like Tesla, they have to build lots of battery factories to do it. Or their suppliers have to. This process is underway, but unless there are massive projects happening right now that I'm not aware of, their production will be lagging Tesla's by a wide margin.

Tesla's battery costs right now(pre-gigafactory) are closer to $200/kwh, so for a 60kwh battery(gets 200mi range on the model s) that is only $12k.

I'd love to see the progression of costs over the years... Would be useful for extrapolation. Where'd you get the more current figure?

This article is a good resource http://cleantechnica.com/2016/05/15/ev-battery-prices-lookin...

Awesome to see market leaders like Tesla beating all possible projections (they're basically already living in 2020 on the chart at the end of the article).

Given that all forecasts had prices trending downward and eventually hitting the fabled $100/Kwh rate, this bodes well for continued progress as all these large-scale production facilities extract efficiencies...

The Leaf is still very popular despite the range. There are a few near me parked on the road, where there aren't even any charging points.

Range anxiety will probably decrease as charging infrastructure improves, so maybe this is less of an issue than it was. In principle there are already more places you could charge an EV than fill up a liquid fuel vehicle. It might just not be very practical or require a long cable getting in the way.

In principle, sure, in practice, each liquid fueling station has multiple pumps each with a throughput of 1 vehicle every 3 or so minutes. Charging stations on the other hand usually just have one or two plugs with a throughput of 30 minutes/vehicle best case and 8 hours/vehicle worst case. It isn't even close in practice.

It's not quite that simple. Every person has a charging station at their house, almost nobody has a liquid fuel station at home. Every liquid fuel station customer is not necessarily someone that will need remote charging much of the time.

I don't have a charging station where I live. There are no power outlets in the shared garage. There are also no power outlets at any of the street parking spots my neighbors use.

Yes, I was a bit hasty to say every person has a charging station at home. I knew there were cases where it wasn't going to be true, but wasn't thinking it would be as common as many assigned condo/apartment spaces. There are apparently laws popping up in states regarding the ability of tenants to put their own charging infrastructure[1] in place, but we would probably be better served by complexes putting in a few shared community stations (depending on size of complex). IMO, when there are more cheap electric options (maybe now?), this would be a good stimulus target.

1: http://articles.latimes.com/2013/nov/17/business/la-fi-assoc...

Every person has a charging station? There are vast numbers of apartments in the world that have zero parking, it is all street parking. Look at capitol hill in Seattle for an example.

Perhaps the discussion would be better served by attempting to look to see whether your point has already been made before commenting? In this case, it was, and I responded to it appropriately (which is to admit I was wrong).

If you live in one of the most walkable neighborhoods in Seattle[0], why would you have a car at all, much less an electric one?

[0] https://www.walkscore.com/WA/Seattle/Capitol_Hill

Because there are people in my household who commute to work evenings in Issaquah and Everett.

Tesla states that Model 3 will achieve 215 miles of range. I think that the manufacturing section of Elon Musk's recently published master plan was partly intended to prepare investors for not turning a profit in the next few years. Quote:

> A first principles physics analysis of automotive production suggests that somewhere between a 5 to 10 fold improvement is achievable by version 3 on a roughly 2 year iteration cycle. The first Model 3 factory machine should be thought of as version 0.5, with version 1.0 probably in 2018.

How much will the version of the Model 3 with 215 miles of range cost, though? I know that with the Model S, the headline price was for a heavily range-reduced version that Tesla didn't actually end up making - the handful that shipped were actually the next model up with a software restriction added.

According to the Model 3 announcement, the base model will have at least 215 miles of range. The Model S was never announced with any specific range for the base model afaik, it was always marketed that the 40kwh model would have low range. It didn't sell very well, which is why they ended up being software restricted 60kwh models for initial reservations and never sold it after that.

I think their plan has it costing less than 10k. 1 kwh of cells is ~$100 USD. They will likely use a 60 Kwh pack to hit the 200 mile minimum, or possibly a little bit larger with a 220 mile range.

The current best guess is a 55kWh pack in the base Model 3. The Model S 60kWh has a 210 miles range, and the Model 3 will be smaller and should be somewhat more efficient.

And yes, there is no way that the Model 3's battery pack will cost anything close to $20,000. That's more like what the Model S's pack cost originally, years ago. The Model 3's entire existence is based on the Gigafactory cutting costs by a decent multiple from that level.

Range is not solely due to the battery pack; weight, frontal area and drag have a big impact.

Nonetheless $20k is high for a 200 mile pack even today. It should be around half the cost.

The next leaf will have the same range of ~200 miles.

http://www.greencarreports.com/news/1104610_next-nissan-leaf...

Seems clickbait to me! Most of the pictures are _outside_ the gigafactory.

Tesla is dialed into how modern media works. Every press release makes it to HN and other tech sites, who proudly proclaim "No marketing budget!"

It's genius.

There's footage in the video but they seem to be trying to keep most of their production processes secret.

The completed Tesla Gigafactory area will be bigger than the town I live in population 50,000 people.

Comparison overlay https://i.imgur.com/ixHBkyw.png

Each side of the red square is approx. 3.75km

Correction: Total land area, I guess the factory itself is about a quarter of the entire factory complex land area.

Why does it take such a huge factory to make batteries? FWIW, I work with LiPo batteries for my R/C hobby. Granted, these batteries are likely a lot more complex, but the same basic principle remains. I'd imagine most of the assembly could be automated. Anyone know why it takes so many people?

Because they plan to produce as many batteries in it as the world production was in 2014 all together.

If it works, this will be a win-win for Panasonic and Tesla. Panasonic gets to vastly increase its battery sales, and Tesla gets all the batteries it needs and at a much lower price. So it is not surprising they have gotten together in this venture.

If you can't watch the video because you don't want to install Flash then you can just change your user agent.

Guides:

- Firefox: https://unop.uk/how-to-watch-bbc-news-videos-on-a-desktop-wi...

- Safari: https://unop.uk/how-to-watch-bbc-news-videos-on-a-desktop-wi...

Reply on news.ycombinator.com