Hacker News new | past | comments | ask | show | jobs | submit login
Tesla is a Battery Company (jeremyrwelch.com)
322 points by jeremyrwelch on Apr 2, 2015 | hide | past | web | favorite | 168 comments

If Tesla executes, they are going to be the leading company in a number of industries, of which battery technology and manufacturing is one of them. They are also leaders in direct sales, JIT manufacturing, motor design, any one of which is a disruptive innovation channel they could monetize, all in the service of building electric cars. Tesla is an energy company, a battery company, a car company, etc, just as GE or any other innovator ends up in many segments. What they are is a nascent Multi-business, Multi-national company that the market is betting (based on the stock price) is going to be able to be a global force. A lot of publicly traded companies are single-business companies that derive most of their revenues from a domestic operation. The few companies that manage to become multinationals have this outsized influence because they tend to drive massive revenues, earnings, and markets caps. The correlation between operating in multiple business areas and being a massive multinational is a direct relationship, because disruptive businesses always reach a saturation point. Businesses like Tesla that set out to disrupt industries that represent significant portions of GDP will naturally need to move into multiple markets in order to execute their strategies, because they need to disrupt or create a sequence of disruptive or creative strategies in order to reach that saturation point (natural market size).

I love Elon for the fact that he thinks of his businesses on a planetary basis, and if you consider the point, you'll see he's actually executing Tesla in multiple markets simultaneously, and essentially going directly from startup to Multinational, and skipping the domination of a single category. He knows that if he doesn't, he's going to face stiff competition from entrenched companies and it's likely he'll lose his advantage. But by tight control of the supply line businesses, and a few strategic relationships with existing multinationals, he's able to elevate Tesla into a multi-business, multi-national organization in record time.

Sure, they're a battery company, but what they really are is an organization that's shooting for the moon and actually executing at that scale. Simple incredible to watch.

I have a little trouble seeing all the businesses you see, which unsubstantiates your point a bit for me.

For example, being a leader in direct sales is great for their goal of spreadings EVs. But how is this applicable anywhere else outside the car industry? Does apple 'monetize' direct sales as a 'channel'? Does apple 'monetize' JIT manufacturing as a 'channel'? To me they just use it as a tool to better position the company. Similarly, once established it's unlikely other car manufacturers will consistently buy motors from them. Fortunately or not, electric induction motors are beautifully simple, any manufacturer could come up with a design much quicker than an ICE for example.

Now for it's two markets, electric cars and commercial batteries, I do think they're posed to succeed provided pure electric cars actually gain widespread adoption and so do Li-ion batteries (and not other storage methods which may turn out to be cheaper).

The interesting point for me is that all his business currently rely on steady, yet iterative, technological improvement to reach their goals (they're already successful, but haven't reached the ultimate goals yet). Tesla needs better batteries, SolarCity better cells, SpaceX reusable rockets.

It will indeed be a thrill to watch over the next few years.

> Does apple 'monetize' direct sales as a 'channel'? Does apple 'monetize' JIT manufacturing as a 'channel'?

Yes, keeping their stock pool as low as it does has real financial benefits associated with it. Cook's entire empire at Apple was built on the back of his prowess at streamlining this, such is the value of it to the business. In that sense, they certainly did monetise their skills in JIT manufacture.

I'd argue they also did monetise the direct sales aspect, since the whole 'Apple Store' experience is presumably of net financial gain to them, or else they wouldn't do it.

I agree with what you are saying. I'd just like to be pedantic about 300+ HP induction motors: they are deceivingly complex to manufacture. The tolerances required are measured in microns. Fortunately, the motors are inexpensive (relative to ICE engines), easy to swap out of the car, and cheap to teardown and rebuild. Manufacturing, design for manufacturing, and efficient service and rework departments will be important for scaling motor production.

In what sense is Tesla a leader in JIT manufacturing?

Do you know how manufacturing takes place at all those other car companies in Germany, Japan, and South Korea? I can not talk about the Asian-based companies, but I know for sure that the manufacturing plants in Germany are very ahead. Because they produce different models at same time on the same track. Each and every car is different in every aspect. In Europe it is very common that you order your car they way you like it and do not buy the one that dealer has in its parking lot. You order to your specification and the option lists are huge compared to the one from Tesla.

Tesla is having only one Model with limited different features and many same parts from Ford, Audi, Mercedes, Bosch and others and still is unable to keep up with delivery as you can see from their latest IR reports.

BTW, the Tesla plant is build with German technology by German companies. Just to remind you.

The claim about leading JIT does seem a bit ridiculous, given JIT manufactoring grew out of the Toyata Production System[1] which encompases Kaizen, Kanban and similar.

All of which was pioneered decades before Tesla existed.

[1] http://en.wikipedia.org/wiki/Toyota_Production_System

I learned that as Toyotism.

EDIT: Another therm for that is "Lean manufacturing" http://en.wikipedia.org/wiki/Lean_manufacturing

Lean manufacturing works right up to the point where one of your vendors gives you a bad part. Your choice is to either shut down a line or have excess backup supplies at which point there's no need to have been lean.

I'm not sure how it helps to have a backup supplies of bad parts?

It should be obvious that quality control across the entire supply chain is vital to make lean work -- and indeed that is how Toyota and others does it: They work very closely with suppliers.

The Tesla Factory is the NUMMI plant, which I believe was the first factory outside of Japan that began to use Lean. The manager of Tesla's factory is Gilbert Passin who is a "Toyota Manufacturing Expert" according to: http://www.businesswire.com/news/home/20100203005622/en/Tesl...

It's stupid. Tesla manufactures a couple of cars with an insignificant volume in less than a handful of countries.

When they are building at Toyota volumes with their diverse product range then I will be impressed.

When was the last time a United States automobile manufacturer started and lasted for longer than a decade?

Of the major US manufacturers, none were started after 1925 (though some had brands that began after that - like Saturn, which operated from 1990 to 2010); of the minor ones, only Tesla and DeLorean could be considered household names - and only one of those is manufacturing new automobiles.

Most of the minor manufacturers are niche (e.g., racing / supercar; off-road; golf cart; single-part mfg; kit; etc) oriented. Of the modern minor manufacturers that are or are planning to sell actual automobiles to consumers:

* Tesla produces ~700 vehicles per week, and has a global cumulative sales of 46,948 units through Sept '14. * Fisker has delivered ~1,800 vehicles (ca 2012) * Detroit Electric (revived in '08) has yet to deliver a single automobile * Aurica doesn't even have a prototype * AM General is primarily a military outfit, manufacturing HMMMWVs

Frankly, I'm impressed that a non-military-contractor has been able to even begin producing at the levels TSLA has been able to (~37,000 units per anum).

For a new automobile manufacturer, their sales numbers are anything but insignificant: they're larger than any other minor US manufacturer, and larger than a lot of foreign minor manufacturers.

> In Europe it is very common that you order your car they way you like it and do not buy the one that dealer has in its parking lot. You order to your specification and the option lists are huge compared to the one from Tesla.

I can confirm, I bought a Mercedes and they had 10-20 options, which I could customize however I wanted.

Usually, with exceptions, conglomerates (the dreaded c-word) are not desired by investors. There's even an estimated "conglomerate discount" for companies of this kind. With certain word-replacements, your comment can be applied for Hewlett Packard too, until the last part. GE, Berkshire Hathaway, General Dynamics, UTC, Honeywell, DuPont, 3M, etc. are exceptions to the rule, when it comes to the performance of conglomerates! One thing common to these companies is adaptability: they are constantly in churn and constantly seek changes in direction to focus on whatever products are profitable. For every successful conglomerate, there are dozens which failed or were broken up.

Given all this, it is too early to sing praises for Tesla! Musk is a great visionary, but creating the next GE also requires a lot of luck and happenstance.

[1] http://www.economist.com/news/business/21593412-conglomerate... [2] http://www.foxbusiness.com/markets/2011/01/19/dismantling-co...

Those conglomerates developed before the age of computers. They designed their operational structures without the convenience of modern technology, and therefore relied on expanding work forces to grow their company. Labor was cheap, and necessary for growth at the time because humans had to do so much more than they do now.

Facebook is a multinational company with only a few thousand fulltime employees. If GE or Microsoft developed in the same technological context as facebook, they would likely employ far fewer people than they do now. The market increasingly favors agility, and monolithic business models are becoming obsolete (if they are not also decentralized in some capacity, a la Uber).

Tesla can become a powerful multinational player while remaining agile.

The conglomerate structure is kept by choice in the case of Berkshire Hathaway. There's a pretty good outline of the reasoning by Buffett in the 50th anniversary letter.

3M is often used as an example in Strategic Management, particularly as an example of Mintzberg's views (strategies emerge and are not planned). I think the emerged strategy is usually labeled "working with surface material" or something.

I guess one could argue "battery company" is an emergent strategy for Tesla in that sense but I think there's certainly planning to it and it wasn't just discovered. But you can use that "post-branding" that is typical of Mintzberg and I think the author of the blog does a good job.

My point with this post is that their business hinges on Battery production and tech more than any other component, and thus the lead they've created in this space matters more than any of their other innovations.

But you're right, their execution on multiple fronts is truly unreal.

>thus the lead they've created in this space matters more than any of their other innovations

What lead do they have in battery production? Currently, they purchase all cells from Panasonic, if I'm not mistaken. Sure, the Gigafactory is going to be...big. But it's far from finished (and there's speculation that construction is on hold). Meanwhile, it's not like all the other large-scale battery producing conglomerates are just sitting around; they are also expanding production.

I mean it's a hell of a good point. Not trying to take away from the post.

Somewhat related: I remember that in the late 90s a Mitsubishi subsidiary, SUN-A, was tasked with developing transformers for Toyota's new hybrid cars. It'll be interesting to see if building a megacrop will be better than building a market place with various competing companies/contractors for this technology (Not that Tesla will prevent such a market place to develop -- indeed one already exists. I just mean it'll be interesting to see if Tesla will benefit from dominating by themselves, rather than delegating to existing market leaders).

Don't forget the "car charging" business. Tesla announced before that it's willing to license its Supercharger network to other car makers. That business could one day replace most petrol charging stations.

>That business could one day replace most petrol charging stations.

Why wouldn't gas stations merely re-equip with electric charging equipment? They already have the best real estate.

There are also competing standards for DC charging, so I find it highly unlikely that the other manufacturers will license anything from Tesla, rather than settling on another standard.

Tesla has the established network, and also the fastest technology— by far.

I'm not sure existing gas stations really have the best real estate. Pulling off the highway for five minutes to gas up is pretty different from plugging in for an hour, and I'm not sure I'd want to do those two things in the same kind of location.

And the backup battery business for homes and business: http://www.bloomberg.com/news/articles/2014-05-08/musk-revea...

Exactly. I think they had to re-imagine how the big companies doing car production and it resulted in the innovations you are talking about. The reaction Tesla triggered in the car industry in US is rather laughable, screaming for legislation to limit the impact of Tesla on their sales. There are promising changes though, it is good to see that some companies go this far: http://money.cnn.com/2013/03/27/autos/mercedes-tesla-b-class...

If Tesla is going to be a battery company, be it, but I think they have a bright future for most of their business (including selling electric cars).

I remember trying this strategy in Civilization. While trying to dominate the long term macro quickly, short-term micro-specialists obliterated parts of my network which caused the entire long term strategy to fail. (Yes I did just make that comparison :).

I hope Tesla succeed though.

We need someone out there to run this opportunity and encourage competitors in this space.

Did they rush you with an attack?

I simply cannot wait for the biography. I really find this story to be more interesting than any other in similar vein.

I really want to know did he start out with this plan or did he simply see batteries as being the easiest thing to improve on by multiples compared to alternatives. Perhaps better said, there are many parts to a car that could be innovated on to such great improvements but the battery in electric cars really needed to be improved by an order of magnitude to make them competitive with petrol power.

Was it that fact that drove him to move into the global space using batteries or was moving global going to happen with whatever he needed to do to improve over the competition so much?

How are they multi-national?

They sell cars in multiple countries. For example, here in the Netherlands, Tesla has a pretty big presence for the size of the company. I frequently see Model S's on the roads here.

I'm not sure that qualifies as multi-national. I'd say you have to own R&D in multiple countries, not just sell there (through someone probably). But I see your point.

They are going to be building a UK R&D center apparently, so they are on their way. http://www.managementtoday.co.uk/news/1306223/electric-cars-...

Tesla has a distribution-centre near Eindhoven, The Netherlands.

Musk has previously said in an interview that Li-Ion batteries were at $600 / kwh, but when he looked at the component and manufacturing costs they were at $100 / kwh. Somehow the entire supply chain has piked up a lot of overhead, and he realized it would be possible to reduce that. At those prices, Model-S battery pack costs are between $36k and $51k!

When I realized that, I came to conclusion that Tesla Motors could be viewed as merely a stepping stone on the way to building a Gigafactory.

When demand for batteries is low, a low-volume battery factory has to share its suppliers with other clients, and those suppliers will price their risks and profits into the deal, multiple times along the chain. On the other hand if you have enough volume to justify having the entire manufacturing cycle under one roof, from ore to ready packages, all those markups go out the window. It wouldn't surprise me to see overhead going from the staggering 500% to a more moderate 50%, but only as long as there is enough demand for all of those batteries. Which is why you need cars - they consume a lot of energy.

By now it's pretty much certain that Tesla will be the first to build a Gigafactory. The interesting bit is whether Musk figured out a way to make sure no one has the ability to build a competing Gigafactory and commoditize the business. Right now only Tesla itself can generate enough demand, so they are pretty safe. If Musk can convince the other car manufacturers to buy batteries from Tesla, it will kill the incentive for anyone else to build a competing Gigafactory for a long time. So at the least Musk needs to appear non-threatening to other car manufacturers, and perhaps the best move would be to split battery division away from Tesla Motors completely.

You use "Gigafactory" as if it was an established word. It isn't. It it the name of Tesla's battery factory being built in Nevada.

If we couldn't use words until they are established, no word would ever become established.

Think about this - how do words become words? There isn't a government agency that makes decisions on that, it's the people who do, through common use.

It's so named because its planned capacity is measured in gigawatt-hours of batteries per year. "A gigafactory" to refer to any factory of that size is a reasonable term to use.

Couldn't you just measure the capacity in directly in Watts instead of Watt hours / year?

Certainly, and you could measure battery capacity in joules, and... well, for some reason we've standardized on weird units for these things.

Exciting times!

One thing that Musk has and BYD doesn't is the eager attention of the world's best engineers. Musk has an interview where he described SpaceX having designed things in house instead of buying them, and coming up with much better solutions. I imagine that battery and rockets are similar to software in that regard - a top engineer is 10x more productive than an average engineer. Combination of scale and engineering talent might be Tesla's bet to stay ahead.

Musk is a great marketer. I would be reluctant to conclude BYD doesn't have the attention of great engineers though. There's a very large talent pool to pull from in China.

All told, more energy storage production is a good thing, regardless of the producer. There is a massive technological shift underfoot, the next 5 years are going to be fun to watch.

Global industry decisions are a lot to do with ego as well as necessity.

Once one person has a battery factory the same size as all the other battery factories in the world combined, pretty soon someone else will try and double it. I fully expect China to announce the construction of a new lithium battery plant larger than Elon's, shortly after it starts production.

edit, just noticed mstachowiak's comment about BYD's new factory being built. As a side note, lithium mining is going to get interesting - http://investorintel.com/technology-intel/china-may-making-l...

There are some alternative materials that will become viable if Li becomes too expensive. But yes, interesting indeed.

> Tesla Motors could be viewed as a stepping stone on the way to building a Gigafactory.

This is an even better summary than the post's title!

"The Tesla Gigafactory", "A battery factory".

Great points here. I think Tesla's move to open source their patents was for these reasons exactly.

With all those battery innovations(at the research stage) we keep hearing in the news, how can building a huge factory be considered a stable investment ? at some point in the future,won't something come that will need different manufacturing methods ?

In my opinion, something like the Gigafactory only makes sense in the context of (expected) innovations.

The usual way to lower costs is to specialize, and to have a diverse network of suppliers. The Gigafactory is meant to do everything under one roof, so it's the opposite of specialization (and low costs).

The advantage of something like a Gigafactory, however, is the ability to coordinate actions in-house. Transactions costs are lower, and Tesla is able to "upload" innovations rather fast (compared to a diverse specialized network of companies).

I'm sorry, but what is a Gigafactory?

I don't think this is a revelation to anyone who has been paying attention?

This seems to have been the dominant hypothesis for a long, long time, and is always what justified Tesla's early (very) red activities. They were a battery company using luxury automobiles to fund battery R&D.

> I don't think this is a revelation to anyone who has been paying attention?

Maybe so, but for casual observers, this lays out the argument very concisely, in a way that might not have been apparent before. (To the average American, "battery company" means Energizer and Duracell, and doesn't seem like a very future-facing sector, so characterizing Tesla as such seems diminutive and hyperbolic.)

Exactly my goal. Many on Wall street understand it, but many on Main street (and even many in the Bay area) don't quite get the picture yet.

Especially given that they have been selling batteries to Solar City for use by their customers for a while already, and have publicly talked about larger scale production of that. E.g. here's an article from MIT Technology Review from 2013 that touches on Tesla's battery sales to Solar City:


>They were a battery company using luxury automobiles to fund battery R&D.

This seems questionable to me. I just don't see lithium-ion powering electric cars in an affordable or usable fashion. The density isn't great and after 4-5 years the battery pack wears out, making your $30-50k battery investment disappear. There's no resale value on a battery that can't even hold 30% of its original charge. Musk's solar city shenanigans are cute, but not a lot of people are moving to solar when electricity is so cheap.

Tesla just never got the battery breakthrough it needed and it will probably never become a hybrid engine company. All this talk of "just wait until Musk does this or does that" has been the same excuse we've been hearing for years. It hasn't materialized, at least for Tesla.

$300m losses in 12 months is scary. $100m loss in the past quarter is just bad. Musk can build this factory, but he can't improve lithium-ion tech nor can he raise the price of gas. If gas is under $6 or so, there's very little incentive to buy a pure electric. Fanboyism aside, Tesla just looks like a mess and making excuses about how its really not a car company stinks of desperation. Tesla is pretty much a failed company. Unless they pivot to hybrid engines and start selling a lower cost car, they'll just continue to be the toy of the idle rich.

Wow, you're pulling out some numbers that seem very unfounded here. You also seem to be VERY anti-tesla as seen in your other posts in this thread.

Can you show me where you have seen hard evidence of battery packs "wearing out" after 4-5 years? Also, please define wearing out as there are different meanings to that.

Contrary to what you have said in this thread the battery life has actually been significantly better than expected (http://www.csmonitor.com/Business/In-Gear/2015/0219/How-much...)

Also, saying the only incentive of buying an electric car to save money is ignoring a lot of other factors in the word.

In practice we're seeing ~25% loss of capacity at 40k miles in LEAF real world testing. 30% loss at 50k miles is the accepted norm. Capacity goes down annually after that. It seems that some cars are doing better, but play cars owned by the rich that get very few miles like the early Tesla sportscars. Real world daily-drivers like the LEAF experience the exact decay that was predicted. Lithium-ion isnt some new hot tech we don't understand. We understand it very well.


Americans replace their car every 6 years now. You may have 30-50% capacity at year 6, depending on a lot of factors. This is unacceptable considering the anemic range of a new battery. EV means losing almost all your car's value after 3-5 years. High trade-in values we enjoy now will be lost as the majority of the cost of the vehicle will be its battery. These batteries will fill landfills and be an environmental disaster.

The current "fix" is just to build the battery replacement cost into the warranty. So now your car is more expensive. Warranty claims now become a legal struggle over what is faulty or not. Nissan claims 70% battery power is non-faulty. That means your 75m LEAF range is now ~50m. My gas car gets that in a gallon and a half.

>You also seem to be VERY anti-tesla as seen in your other posts in this thread.

Its a 12 year old company that has done nothing but burn money. They have no mature product on the horizon and their promises of making a middle-class cheap EV with great range and durability are just not in the realm of technology. The stock has been on a decline since mid last year and they have a celebrity CEO who is good at PR and gimmicks but not delivering affordable vehicles.

>Also, saying the only incentive of buying an electric car to save money

This is the only motivator for mass market acceptance. The idle rich love the acceleration of that $100k car, but that doesn't matter to someone who makes half or a third that in a year.

> nor can he raise the price of gas.

I think this is one of two fundamental flaws in your reasoning. First, yes, R&D is incredibly, incredibly expensive. I don't think anyone doubts that, and I find it just hysterical that x or y "social startup" can burn through far more money than that and never have to answer questions about the value of their R&D.

We know what Tesla is doing, and we know that it's difficult. It should be no surprise that it's expensive as hell. This isn't to say I'm betting on their success, it's to say that "R&D is expensive" isn't an argument against Tesla or its potential for success.

Quite the opposite, it seems their model has done a decent job of offloading a lot of R&D costs by, as I said, selling automobiles on the side.

Now, the second flaw ("nor can he raise the price of gas") isn't that the statement is false. He can't. But the laws of thermodynamics paired with the laws of economics can, and will, forever. Further, when it does happen, the success of his cars (or "toys of the idle rich") is irrelevant: he owns the infrastructure and he owns the IP necessary for everyone else to leverage the inevitable growth in fuel prices.

Also, to address the battery capacity loss argument:


Independent study on the Roadsters.

126 vehicles (of 2500 produced) reported a total 3,198,749 miles traveled. Tesla's projections said the battery would have 100,000 miles driving range, so at 50,000 miles the projection was to have 70% capacity. In reality, an average Roadster has 80-85% original capacity after 100,000 miles. That said, there was considerable variation between batteries' health, so take that as you will.

I already addressed this:

n practice we're seeing ~25% loss of capacity at 40k miles in LEAF real world testing. 30% loss at 50k miles is the accepted norm. Capacity goes down annually after that. It seems that some cars are doing better, but play cars owned by the rich that get very few miles like the early Tesla sportscars. Real world daily-drivers like the LEAF experience the exact decay that was predicted. Lithium-ion isnt some new hot tech we don't understand. We understand it very well.

Two completely different cars show different rates of deterioration: "Must be because rich people drive one of them."

I just told you the actual numbers, 100,000 miles at 80-85% capacity. You can't just say "get very few miles." I just told you how many miles.

Its the same battery tech. The only difference is that daily driving, which is the typical use case, wears them out faster. Probably due to getting the batteries hotter as they run longer, in traffic, etc.

Its still lithium-ion. It will behave the same way regardless of brand or celebrity CEO worship.

>I just told you the actual numbers

From a non-daily driver two seat sports car. Other sources follow the LEAF's numbers. My buddy's Porsche runs fairly well and its 30 years old. That doesn't mean you can have a 30 year old daily driver.

Here's the actual data for the Tesla Model S. Far superior to the Leaf and inline with the roadster. No need to compre apples to oranges.


Indeed, money managers have been expounding on this thesis for years. "Bond King" Jeff Gundlach remarked to this effect some with Barry Ritholtz on July 2014. https://soundcloud.com/bloombergview/barry-ritholtz-intervie...

Great to see that other are catching on.

It's a revelation to me.

I thought Telsa was a car company and I followed them since the beginning until their initial IPO.

From there and sorta lost interest in follow any automotive news.

Plus it is very easy to see them as a car company, they started out as a car company and Elon came to Tesla, a car company, and took over.

True. I've already seen an article like this. This also gave a reason why they wanted their supercharging stations everywhere, at cheap price, and open patents. I'm not sure, but I doubt he opened his patents on the battery.

Agreed. CTO JB has referred to Tesla as an energy storage company for a while now.

Yes, this has the tone of something revelatory but I agree.

I don't get this article. The battery in a Tesla Model S is literally a huge array of Panasonic flashlight batteries. It is an off-the-shelf catalog item you can order from Panasonic. Admittedly, you should probably have prior arrangements before ordering millions.

The central insight of the Tesla founders was that people really will pay $85000 for an electric car. The other auto makers didn't believe that.

> The battery in a Tesla Model S is literally a huge array of Panasonic flashlight batteries. It is an off-the-shelf catalog item you can order from Panasonic.

Um, no. Viewing the battery of a Tesla Model S as just a big flashlight battery misses the whole point of what Tesla is doing.

A flashlight battery does one thing: it discharges energy until it's empty. Even an ordinary rechargeable household battery only does two things: it recharges, and then it discharges.

The battery in a Tesla Model S, OTOH, is continuously computing whether to charge or discharge, how much to charge or discharge, what its current state of charge is, how to manage charging and discharging to maximize its life, etc. And it has to do this in a constantly changing environment--a car that is traveling at varying speeds over varying types of roads. (Doing it for a home battery is actually easier, in a way, because there is less variability in the load.)

This sort of technology is absolutely critical if batteries are going to become significant in our energy budget on a large scale. The energy storage medium of the battery is only a tiny piece of the whole picture.

> The battery in a Tesla Model S, OTOH, is continuously computing whether to charge or discharge

Is it? I would imagine the software does that. The battery really is a large array of commodity batteries made by panosonic. Tesla is doing the gigafactory for vertical integration and to bolster it's own supply. Obviously, they have done a lot of innovation around the pack in terms of heat distribution and cell construction, as well as the management system. The innovation is really about the macro-cell construction and the software to manage the array. On a micro-level it is just a bunch of commodity batteries wites into small cells, that make up a large cell array.

> Is it? I would imagine the software does that.

It is. JB Straubel (Tesla's CTO) has done all of the hard work with regards to their energy management system.

Sure, babying the pack with a liquid medium for thermal transfer helps (compared to the inferior air cooling on the Nissan Leaf battery pack), but the foundation for both the performance and longevity of the Roadster and the Model S battery packs is Straubel's remarkable work.

> I would imagine the software does that.

Obviously the lithium-ion cells themselves aren't doing the computation; software is. But the software should be considered part of the "battery" as a system. Without the software the lithium-ion cells are useless for this application, and indeed for any application more dynamic than a flashlight. So the software development that Tesla has done is essential for making batteries viable in the sorts of applications that people want to use them for to help solve our energy problems.

would "realtime battery management software company" be a better fit?

Possibly, but as I understand it, the software is tightly coupled to the battery hardware; if you were to swap in a different kind of cell, or even the same kind of cell but from a different manufacturer, you would have to rework much of the software. That's why I think it's better to view the storage cells and the software as a single integrated "battery" system.

The cells themselves are COTS Panasonic NCR18650A without built-in protection circuits.. The management circuits are probably developed by Tesla but are external to the batteries, most likely a circuit per array of parallel cells.

Having a protection circuit per cell would be wasteful and also result in quite a bit of power loss and complexity in wiring them up.

I toured the Tesla factory in Fremont, CA last Monday. During the completely awesome 1 hour tour, our guide answered many detailed questions about their cars and manufacturing processes, including how Tesla instruments battery cells.

We were told that they instrument every single one of the 7,000+ cells for temperature and voltage.

That's an overstatement. They do not have 7000+ temperature and voltage sensors per pack.

Each of the 16 modules is split into 6 blocks where the 70+ cells within a block are all wired in parallel -- so they are going to have the same voltage. The blocks within a module are wired serially and the modules are also wired serially.

Each module has a battery monitoring chip:


It has seven voltage measurement inputs (between each block and before/after the outer blocks) so it can tell the voltage of each block. It can also control a bleed resistor per block (for balancing). There are two temperature sensors per module.

Each module also has an 8051 (C8051F530A) that talks to the battery monitoring chip and to the data network connecting the modules to the "global" battery management system which has a Altera CPLD and (probably) an ARM CPU.


Thanks for clarification and link to the teardown pix. Very interesting.

They aren't really COTS. They were for the Roadster, but for the Model S they use a specialized chemistry as well as removing the protection circuit.


That's standard for NCR18650A cells though - they ship with no protection circuit and I believe the same chemistry that the Model S is using. The protection circuits are added by companies that buy raw cells from Panasonic.

Charge balancing is a common feature, often built in to charge controller ICs. Of course Tesla has to do this on a much larger scale, like with all of its power electronics. But so does a Prius.

I'm not sure how much of a technological lead Tesla has versus major car companies already producing hybrids and electrics, but it can't be that huge, and those guys have way bigger R&D budgets.

The cells used in Tesla's battery pack are a standard form factor, but they are not standard off-the-shelf parts. If you read up on the battery pack you'll find that it uses 18650 cells, and you'll find that you can buy commodity 18650 cells from all sorts of places. But 18650 is just the shape, not the contents. Tesla's 18650s do come from Panasonic, but they're not off-the-shelf items. They have a custom chemistry and they also don't include the standard battery protection circuits found in almost all such batteries (because Tesla provides that circuitry externally).

Sure. Similarly Google is just a bunch of commodity off-the-shelf hard drives.

If some talking head wrote a blog called "Google is a hard drive company" that would be a worthy retort.

Joking aside, one could probably write that article and make a good case based on the same logic.

They are very good at building cheap servers, and running them cheaply.

You say electric car like it's some souped up Prius (I own a Prius).

The Tesla Model-S is an amazing vehicle that's not only electric but amazing to look at and drive.

As an aspirational status symbol it's got a hell of a lot of cachet 3 years after it's introduction. That's a feat of product marketing and engineering.

Absolutely an achievement of marketing worthy of the best marketeers in history. But the article is wrong on it's face when it states "the Battery Tech they Developed While Building EVs" because Tesla has not developed any battery technology of any kind, ever.

And there's some deal with Panasonic for the Gigafactory as well — it sounds very much like it'll be manufacturing stuff that's basically Panasonic IP to start with.

They're also a software company. I was recently shocked to hear that ~60% of Tesla employees are involved in software engineering, versus a normal car company at around 2%.

Source: http://www.businessinsider.com/morgan-stanley-tesla-multiply...

Also consider:

- solar city benefits from increased storage capabilities of batteries. Solar panels (a capital investment) generate higher returns when they also power buildings at night.

- National electric grids benefit from increased battery storage because they can store electricity for hours after it's actually generated.

- tesla is building a network of superchargers all connected to the electric grid

- tesla has arguably more at stake than any company in battery R&D. Few if any companies are as motivated as tesla for battery tech to improve.

My hypothesis is that Elon Musk is architecting Tesla as the "cloud" of electricity generation, storage, and transfer. He is building the ultimate smart grid. Electric cars are a nice excuse to develop the requisite network effects and initial momentum before expanding to other verticals. He's going from a red ocean to a blue ocean. [1]


How on earth can a strategy centered around generating, storing and transmitting electricity be considered blue-ocean?

Well, the innovation is really about storing electricity. We've been generating and transmitting electricity for over a century now.

Once we can store electricity (in batteries, for example), then intermittent power sources, like solar panels (and also wind turbines) become much more valuable.

Energy storage is a very inconsequential market right now, but if smart-grid-connected Teslas really take off, then they become important not only to the automotive industry, but also the electricity and energy industries.

Once solar cells hit a tipping point, storage becomes one of the best businesses to be in.

I'm sort of curious how the economics of this will end up working out from a consumer's perspective. Lithium ion's big advantage was energy density -- you can store a lot of juice for not much mass or volume, which is a really big deal for portable electronics or cars... for a stationary home installation, though, much less so. Having a giant thing sitting in your backyard might be okay if it was high-capacity and cheap, and it seems like there might be other contenders besides Li-ion that might do well given the different set of operating constraints that stationary storage would have. Maybe other battery chemistries, or maybe kinetic energy storage devices like evacuated flywheels, compressed air, etc....

Energy density is still important for home installations, especially if you have ambitions of retro fitting on a wide scale. If you could get to a situation where solar + battery was cheaper than a grid connection, a majority of the population will still not buy it if it's something which will require a half tonne object to be installed on a specially constructed concrete plinth, brought on a lorry with a crane system, and taking up 2m x 1m of space. Far, far better if it can be within the size and weight range of a normal white box utility, for instance the kind of freezer that people put in their garage.

In the rural or suburban US, it is very common for homes to have large propane tanks for heating and cooking, due to them being too far away from supply lines for natural gas. The tank is typically much larger than your example.

Not to mention essentially all new homes have central A/C with one of the heat exchangers "installed on a specially constructed concrete plinth." We have plenty of space; it's simply not an issue.

> Having a giant thing sitting in your backyard might be okay if it was high-capacity and cheap

There are also walls you could embed batteries in. A house has a lot of surface area that could be used to store energy.

Embedding Li-ions in your walls -- it's safer than embedding gasoline tanks in your walls. It's still far from a good idea, though.

Article fails to mention that Tesla has actually stated that it intends to market a home battery and that many believe this is the product that Elon referenced recently.

Interesting to consider --- is there enough lithium in the world to support this stuff?


They don't actually use that much lithium -- and there is /plenty/ of lithium practically everywhere.

Cobalt is much closer to being a problem -- and it is still very far from being one.


Is Eric Eason a good authority on mineral reserves and resources? I don't think so:



The amount of lithium available in known reserves that are profitable to extract at today's prices is very large. We won't be running into constraints anytime soon. And while lithium makes for a good battery anode now, it would be foolish to assume it's the preferred choice for energy storage 15 years down the road.

Can't lithium batteries be recycled?

Thanks for these. The second link notes that the lithium battery recycling is profitable. Do you know of any links to more detailed economics of lithium battery recycling? I'm trying to estimate the cost of Li batteries in a fully recycling system at equilibrium, and I'm not finding any detailed cost/revenue breakdowns. Thank you.

No, sorry. I'd be interested as well.

Sounds like between using lithium in car batteries as well as for household and industrial energy storage, "peak lithium" might be a real possibility.

Of course. But it's not as though Tesla's cars are just useless, uninteresting husks. They are also a very competent car company, especially for being so young. But their core competency is definitely in batteries.

What's wrong with being more than one thing. They are shaking up A LOT about how we currently think about the ``energy'' models. I think batteries are going to do wonder for countries that don't have a ``grid''.

The coolest thing to me about Tesla is the car is becoming decoupled from what powers the car. It's still electricity but if my power provider was to switch to nuclear, solar, people on bicycles, or coal It would be transparent to me.

Don't ignore the other battery companies.


  Foxconn hopes to build cars priced under US$15,000, 
  according to Gou. The contract manufacturer already 
  develops electric car batteries, which he said were in demand.
Look out, Tesla.

Being that the batteries are the expensive part of an electric car, Foxconn could get to the mass market before Tesla, and possibly set back the market if the experience sucks.

Does Foxconn have much expertise in manufacturing batteries?

Maybe not as much, but Apple has been "poaching" Tesla's engineers.

Also, wasn't Musk going to do a phd studying batteries before deciding that the internet was a better move for him at the time?

Didn't know about this. Source?


How did your career progress from there?

I was offered a place at Stanford University to do postgraduate research into high-energy-density capacitors. But then the Internet came along, and I wanted a piece of the action.

thanks! added it to the post

Most of his talks he mentions this. He was doing a PHD to study ultra-capacitors, or at least, was accepted to UPENN with that intention. He left to start zip2. He has also mentioned that a part of the reason he didn't proceed was that his research was very theoretical and it would've been able to find (or disprove) of a method after years pf study. He mentions some of the other breakthroughs on iltra-capacitors having the efficiency of a lead battery, but, while extremely efficient, the prototype used an extremely rare material that only was mined in a few tonnes a year. Thus the solution could never scale.

Musk has been known to embellish his academic record, it was detailed during the court case against Tesla's original founder Martin Eberhard:


According to court documents he has never had any association with Stanford university, and, contrary to public statements, among them SEC filings, has no degrees beyond a Bachelor of Science in Economics from PennU.

Do we know if we can trust Eberhard on this?

Only as much as Apple was a HDD company with the release of the first iPod.

Yes, their underlying hardware was necessary for the product to be revolutionary, and it was also something that no one else could produce at the time.

But, like Tesla, Apple added a layer of design and marketing that made for a revolutionary UX, as well.

So I agree that Tesla is a battery company, but I don't think that that makes them less of a revolution for the car industry.

As Apple has never made hard drives, calling them a hard drive company makes zero sense.

The hard drive in the iPod was never the most interesting component, or the differentiator from the rest of the market. Tesla's battery is. The hard drive in the iPod is like the wheels on a Tesla. It's useless without them, but they're a completely replaceable commodity and has no bearing on the success of the product

The battery in the Model S is an array of commodity Panasonic cells.

What part of "tesla is going to build their own batteries" do you not understand?

Sounds like the headline of this article should have used the future tense.

Apple never manufactured their own iPod hard drives. That's where the similarity ends. Advancing the state of the art in batteries is essential for Tesla to remain competitive. For Apple, it was adding content and services.

Apple was (and still is, in some cases) a leader in miniaturization. They made a product viable that was never viable before.

Other drives were too big, expensive, and power-hungry. Apple made the iPod a common device because it was the first MP3 player with more storage than most people would need, it only cost a few hundred, and its battery lasted a long time.

Apple continued to compete by getting even better at miniaturization, like when they introduced the iPhone.

Eventually we're going to be arguing semantics, but my point still stands. Being the leader at manufacturing the boring internal components was necessary for both companies to provide the product that we perceive to be their core competency.

Edit: I forgot to address your point about manufacturing. They may not have manufactured the drives or had a unique technology, but they had advantages in manufacturing that no one else had. Tesla seems to be the same way, from what I can tell.

> Tesla is a Battery Company

Too narrow. Tesla is a electrical energy storage and distribution company. Tesla has over 400 Supercharger stations.

I think this quote sums it up:

“I see us more as an energy innovation company, at our core, than even as a car company.”

If true, this puts Tesla and Solar City in complementary positions, and perhaps on track to become the same company under Musk. Solar City tech generates the electricity, and Tesla batteries will store it.

And McDonalds is a Real Estate company.

Yep, in a very tight race with CBRE for top spot. I wonder if there's a place to find sick information, like the Forbes richest list.

I've realized that Pepsi owns Taco Bell just to sell more drinks.

Tesla is not a battery company until they actually start manufacturing batteries.

Just like how Apple is a chip fabrication company? I don't think so. Tesla was the first company to introduce OTA updates, something unheard of in the car industry. So to call them a battery company sounds very patronizing to me.

Except Apple is successful and Tesla is not. When articles like these come out for a company bleeding money its something to worry about. I think its obvious the lithium-ion just doesn't have the lifespan and density to replace cars. At least when gas is under $6 a gallon. After 4-5 years the resale value of that $30k battery pack is non-existant as its begining to fail. We're seeing this a lot with hybrid sales where the car's battery pack has a fraction of its capacity. The makes the motor work harder, but with a pure EV car there is no gas motor to take up the slack.

I have very low hopes for EVs. Even if Tesla lowered battery prices by 20%, which would be impressive, they're still extremely expensive and still wear out quickly. They make sense when gas is crazy expensive, but its not, especially with all this investment in shale oil.

I imagine the future will have more efficient hybrids and perhaps Tesla will migrate to a hybrid company if they want to stay alive. 100k for a range anxiety inducing EV with a low resale value after x years due to its $50k battery pack wearing out is not the vision of the future here.

Do they have to be profitable? Yes. Will they move to manufacturing hybrids? Probably. But they're definitely not a battery company. At least not yet.

Wrong. Tesla is a distribution company. Cars and batteries are commodities, more or less. Corporations are now extremely efficient at copying and commoditizing their competitors' innovations - just look at how fast the android ecosystem popped up alongside iPhone - so technological advancements are not the real source of long term value but merely the temporary advantage needed to be first to a market and set up network effects.

EBay, for instance, was there at the right time to take recent technology advancements, set up a network-effect marketplace and milk it for, oh, 2 decades. They stopped being a technology company a long time ago.

So how does this relate to Tesla? Answer: The supercharging stations. If Tesla owns the most charging stations then EV buyers have a good reason to buy Tesla over a hypothetical Ford or Toyota EV, putting Tesla at a major advantage. Eventually it'll be on Ford and Toyota to bet the farm on a massive supercharger network of their own and launch an EV to compete against Tesla's entrenched domination of the marketplace. Right now, Tesla is in a fleeting moment of technological and branding advantage which they are correctly leveraging to quickly set up their charging network.

It's a lot like telcos bragging about their coverage networks. Eventually all the (surviving) car companies will be doing something similar, running TV ads about their charging station maps.

Companies are either A) marketing companies or B) distribution companies or a blend of both.

Nike is a marketing company that happens to sell shoes. Ditto Taco Bell and cheap Mexican food.

Amazon, Akamai, Google and Facebook are distribution companies --- they own the network or marketplace.

Disney (mostly distribution), American Airlines (mostly distributiom) Walmart (mostly distribution) and Apple (mostly marketing/branding, but appstore and itunes are distribution powerhouses) are a little of both.

Edit: Nielsen, Google are distribution comapnies. They have prohibitively large or complicated information ingestion systems set up to and then resell the results.

Newspapers and TV are too, but their distribution monopolies are crumbling.

Interesting. That means solar advancements are very bad for Tesla. If people don't need to stop and charge, then the distribution network is worthless.

I generally agree with your categorization, but I would be more specific that Tesla is both a Battery distribution AND marketing company. EVs existed prior to Tesla, but they were the first company to identify and correctly serve the high end market with the Roadster.

My company Chrg specializes in charging infrastructure. Note that the Superchargers you mentioned actually use specially designed Tesla batteries, which are precursors to the Home battery product Tesla will likely be launching soon.

Solar advancements are a great thing for Tesla. Once power is captured via solar panels, it will need to be stored – preferably in Tesla batteries.

finally, found the business connection.

Tesla's goal is to accelerate the advent of sustainable transport. They want (a) to see as many electric cars on the road as fast as possible and (b) make a lot of money.

B is required to perfect the product -- which serves A better than anything else.

I think the 'Apple way' is most efficient: selling the whole package and making it so good, that people happily pay a lot.

I'm not sure how cars will ever constitute "sustainable transport". Less unsustainable than carbon fuelled cars, maybe. And that's a big maybe, given batteries, power generation etc.

But I suppose there's not that much money in trying to sell bicycles...

>In the NY Metro area, it’s Con Edison

That is a rather unfortunate (or rather appropriate) name for a company named after Edison.

Especially in French.

I can't see Musk trying to build a battery company as his primary focus. That's not exactly a world-changing life goal (well, for him. I'd be happy with that). That's like saying that Apple is a chip company because their chips are central to efficiency.

The battery might currently be the most important single component of an EV, but there is no guarantee that will be true in five/ten/twenty years. The car is more than just a mechanism to increase scale. You don't beat Musk by making a better battery company. You'd better find a way to beat this:


(The word 'battery' is not even mentioned in the article. Tesla could have less battery life but still win.)

Again, similar to Apple, Tesla found a point of advantage and will own it. The landscape will change and so will Tesla's efforts at building or maintaining an advantage against some very powerful companies.

Batteries may appear quite humble, but they are at the core of just about every technological advancement we have at the minute. Cars, laptops, phones, wearable devices.. they all depend on batteries. At the minute battery life is a real bottleneck for us. We are turning into a nation that plans their life around moving from charging station to charging station. Batteries are heavy and they don't last that long. The next truly world changing advancement is going to have to involve making access to energy much less heavy.

That's not exactly a world-changing life goal

An order of magnitude or two improvement of current battery technology will fundamentally change the world in ways we can hardly imagine. A cool electric car is nothing in comparison.

I have no idea if batteries are Musk's primary focus here, but creating very cheap and dense storage for electricity WOULD be as world-changing as you can get: if you solve that problem, you can't just replace combustion engines in cars, you can also use solar and wind power for base load on the grid.

This would be much more world-changing than cheap rockets.

Having a "Tesla battery pack" at your home is more than just handling grid fluctuations. What if the battery pack can act as a supercharger for your car? It's like having a tank of gas in your garage all the time.

>It's like having a tank of gas in your garage all the time.

And how many people care to have a tank of gas in their garage all the time, even though the idea is completely feasible today?

Selling batteries def have helped them with keeping the lights on. Not sure if many remember but Tesla was about to go under water at some point before cutting a deal with this German car company to sell them batteries.

The article assumes that car batteries will feed into the grid. Tesla seems to think that won't be the case, car batteries need to be secured against harsh outdoor conditions and have tougher charge cycle limits.

Tesla deploys a large battery product at all of their Supercharger locations. These batteries are built to withstand harsher weather conditions, and will likely form the basis for a home battery product.

Except we need an exponential leap in battery technology.

These 1% per year improvements are going to take forever to get us better EVs

But 1% a year IS exponential.

Tesla is no more a battery company than Amazon is a shipping company.

Every company needs a platform to attack other markets and industries from. Build fast and break into other accessible applications of your product or service.

Whose battery IP does Tesla use?

And Google is a search engine...

Google is an advertising company with a monopoly on a specific native ad type called "search ads"

Batteries are to devices what rockets are to space...things.

So there's a trend here.

Totally ignorant of the tech here, but could they now do Netflix for Batteries? $9 / mo. gets you all the AA, AAA, D cells you require?

It's all about the tech. Not disposable cells on the "one to a dozen" scale, but rechargeables by the cubic meter: wanna drive a car 300 miles? Tesla batteries. Wanna buffer your house power, daytime power demand at nighttime costs, or nighttime power use from solar? Tesla batteries. Anything you'd put gasoline in for high-power untethered use? Tesla batteries. Remember: the first Tesla car was powered by an enormous number of laptop batteries (basically AAs by the hundreds/thousands).

Current battery technology insufficient (your phone only lasts a day on a charge, or your electric car only gets 100 miles)? now that he's getting demand up for big-money big-volume batteries, Musk can fund battery research and production on a scale way beyond anything we've seen. Remember: Musk's "gigafactory" project is intended to build more batteries than the entire world currently produces.

can I borrow this idea for my startup? how much will I need to pay you?

Aren't they mainly a subsidy company?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact