Analyst: With Gigafactory, Tesla Battery Cost Will Fall Below $100 Per kWh

SEP 30 2015 BY ERIC LOVEDAY 34

Tesla Gigafactory - Drone fly over - September 2015

Tesla Gigafactory – Drone fly over – September 2015

According to Jefferies analyst Dan Dolev, Tesla could lower battery costs substantially by changing cell chemistry and from economies of scale that will come with the completion of the Gigafactory.

Quoting Dolev:

“Lowering battery cost via changes to cell chemistry and Gigafactory scale benefits are critical determinants of Tesla’s ability to sell an affordable Model 3 starting at $35K. Our detailed battery component cost analysis details a path to 50%+ reduction in battery pack cost to $125/kWh by 2020, driving up to a 1,000bps vehicle GM tailwind, placing TSLA at the upper-end of peer OEM profitability levels.”

Street Insider points out that Dolev actually details a potential path for Tesla to get cell cost below $100 per kWh:

“Our analysis details a potential path to a 30% cell-level cost reduction to ~$88/kWh by using a more efficient lithium-rich nickel cobalt manganese cathode (vs. NCA), doubling the percentage of silicon in the synthetic graphene anode, replacing the liquid electrolyte with an ionic gel electrolyte which eliminates the need for a separator, and using a water-based electrode solvent for the cathode.”

With the arrival of the Gigafactory, Dolev suggests that pack costs will fall even more:

“The Gigafactory, which is expected to begin production in early ’16, should drive down pack-level costs by 70% to ~$38/kWh via economies of scale, supply chain optimization, increased automation, and production domestication.”

Source: Street Insider

Categories: Battery Tech, Tesla

Tags: , ,

Leave a Reply

34 Comments on "Analyst: With Gigafactory, Tesla Battery Cost Will Fall Below $100 Per kWh"

newest oldest most voted

I agree.

Bit optimistic I think.

$125/kWh by 2020 is on the reasonable-but-pretty-optimistic side (my estimate is around $145). Elon is regularly screwed over by suppliers who try to raise the prices[1] when they realize that their customer is making money. So the obstacles he might face in getting costs down aren’t purely technical, but rather, greedy jerks.

$125/kWh makes a hypothetical 50kWh* pack for a Model 3 vehicle at $6,250, which is pretty good if you’re putting in into a $35,000 vehicle. If you figure 20% GM and then a third of the cost will be the pack, thats $8,400, so thats what they need to keep the pack cost at. And the good news for Tesla is that they probably wont end up selling those base Model 3 units until 2019/2020 anyways after they sell out of the Model 3 Signature, Model 3 P65D*, Model 3 65D*, Model 3 65*, and Model 3 50D* units, they they can sell the Model 3 50.

[1] http://www.esquire.com/news-politics/a16681/elon-musk-interview-1212/

* Battery pack sizes are WAGs

Anthony, I’m right there with you on that who gets in line for a Model 3 first. I’ve not once believed it would start with the base version.

I do not think ANY manufacturer ships less than the upgraded model of any vehicle first.

If you guys are right about base models taking a while then its going to be a tough guessing game for me on a model 3 purchase. If you wait for a base model, you run the risk of losing out on a portion of the federal tax credit. if you go for a top end car, you pay more.

The only argument I can see for them selling most models up front is that it guarentees them more money in deposits. The deposits on the Model 3 could be a monstrous loan for the company. If they have $3000 deposits and take them for 100,000 cars, that’s a $300M loan. Why would you want to minimize that by only selling a top end car at first?

It’s very likely, in fact I’d say it’s almost certain, that Tesla will get so many reservations for the Model ≡ that they’ll have to quit taking them months or possibly even years before the car actually goes into production. The demand is going to be quite high whether there is a $7500 tax rebate potentially available for buyers, or not.

You can bet money that Tesla will make and sell high-end “Signature” Model ≡’s before the base model, just as they did with the Model S and Model X. They’re certainly not going to cut into demand for the most expensive trim level Model ≡, offered to those willing to pay more to get them early, by offering a low trim level first!

$38/kWh at pack level is really aggressive! Well, *eventually* that will probably happen but I doubt we’ll see that before 2030.

I read it that way too at first, but when you read the linked webpage they have $88 for the cell costs and $38 for the pack costs (ex-cell), with a total cost of around $125/kWh.

No, $38/KWH is only for the pack-level costs WITHOUT the cells. This article has caused a lot of confusion (perhaps intentionally?) all over the net.

But even the $88/KWH at cell level (and $125/KWH at the pack level) would be an AMAZING achievement that would throw the automobile market into disarray. At that price level, I think people would be foolish not to buy plug-in cars (at LEAST PHEVs but preferably BEVs).

@Anthony, Speculawyer
I see, yes that makes sense. That reminds me that I’ve seen this story before somewhere, with the same confusion!

$88 for battery itself plus $38 for pack plus 15%+ Tesla margin comes to around $150. So it is in line with what Elon and JB have been saying that it will be below $200 and most likely well that. Raw materials for 1kwh battery are supposedly around $70. It will probably be less with the gigafactory and the contracts they are signing with suppliers.

The analyst makes a good point about using ionic liquid electrolytes in future batteries for cost reduction. Ionic liquid electrolytes will also lead to higher-voltage batteries. Here is good article about them:

http://chargedevs.com/features/ionic-liquid-electrolytes-may-open-the-door-for-higher-voltage-batteries/

However, the Gigafactory is a risky proposition since it relies on future demand for Li-ion batteries to be used in utility/business/home energy storage. Recent advancements in safe, affordable, and non-toxic flow batteries may make solid state Li-ion batteries an also ran for large-scale and home energy storage. The Achilles heel for Li-ion in such applications is that flow batteries would scale much better, since you would basically just have to use bigger tanks to hold more liquid, which would cost much less than manufacturing additional Li-ion batteries.

http://www.sciencedaily.com/releases/2015/09/150924151401.htm

I’d certainly like to think so, but let’s remember that any such “analysis” released for public consumption is just some investment company trying to promote its stock picks. If Jeffries did have any proprietary or inside info, they certainly wouldn’t be sharing it in a public article. They’d save it for customers willing to pay for an analysis.

Is it $38, $88, $38+88, $38+88+mark up, it’s not clear at all. In case of ~$150/kWh (38+88+markup), that would mean $7500 for 50kWh battery (model 3?). When it comes time to replace the battery in 10 years, how likely are people going to spend that kind of money for $30K-when-new-car? If today’s any indication, not very likely.

I hope they can bring it down to $50 or less. That’ll make 50kWh to be $2500 (with labor ~$3500), making it like transmission replacement on today’s gas cars.

Let us hope that after 10 years, the price of a replacement pack will drop some more. And it may also be possible to find a used pack from a salvage vehicle that may not be new, but will still work for several years. Also being that Tesla uses industry standard cell sizes in their packs, it is far more likely for aftermarket companies to be able to refurbish those packs with new cells.

The Gigafactory won’t be producing industry standard form factor 18650 cells, but rather cells about 10% larger in all dimensions.

However, it may be that the Gigafactory will supply cells in such large numbers that its cells will create a new industry “standard” size.

I would think by then the tech Tesla uses is not going to need replacement in 10 years, but rather 20 years. It’s totally doable.

You know.. It occurs to me that if the price of batteries gets anywhere close to $100 per kWh, then think what that means for other cars. For example, my understanding is that a Prius battery costs (or at least used to cost) around $1,000 to manufacture. So if you can do a Lithium battery pack at $100 per Kwh, that means you could essentially stick in a 10 Kwh battery pack in a regular hybrid for essentially the same cost as the NiMh pack they’ve been using, with the bonus that it could give you 25 to 30 miles of EV driving. Why even bother building a hybrid without a plug?

If they get down to $100/KWH then you’d be a fool to bother with ICE at all. Just a big battery pack and supercharger access. No need for outdated ICE garbage.

Cheers to that!

I wouldn’t post analysis by the banker analysts. They could be very weighted on the sell-side and talk up things beyond actual science. Dolev is one of the crew that “covers” such stocks. They are allowed to be paid to lean positive or negative. I don’t trust their “hard to believe as unbiased” opinions.

And, if batteries get to $100/kWh – that means everyone has them. LG Chem, Samsung and Chinese companies. That is the panacea of EV futures. However, Demand/Supply curves say that demand will ramp with lowering prices and thus manufacturers will not find it as easy to lower their actual parts prices.

$100/kWh doesn’t solve much, though. We need higher density by the Wh/kg rating. All $100/kWh does is lower the price of the Volt by $3000. What would be better is a 50kWh battery for the Bolt at $5000 total and then you don’t need the extra parts of a Volt. Bolts would bolt out the door at $26-28k. Leafs would Leave at under $24k. Without incentives. That seems to be a good future for the EV industry. It would at least allow for doubling if not tripling the annual sales rate of today by 2020.

Bonaire said:

“And, if batteries get to $100/kWh – that means everyone has them. LG Chem, Samsung and Chinese companies.”

If every battery cell manufacturing company could make cells identical to any other company’s, then there wouldn’t be so many companies signing up for LG Chem’s new cells.

Of course, no tech advantage lasts forever. Even patent protection only lasts 20 years, and it’s almost certain that other battery makers will match whatever advance LG Chem has within a few years, sidestepping patent protection.

But certainly it’s possible for a tech breakthrough to let one company out-compete all others for a time, as LG Chem is doing right now.

NO it does not.

Unless everyone ramps up scale, buys custom machinery instead of generic, and signs 5 year contracts with mining companies for below market prices instead of buying on the spot market, and minimizes logistics by creating precursor materials and components in one location.

Well, if EVs are really going to as popular as some say – then there surely will be manufacturers and builders who can repeat what Telsa is doing. Keep in mind Tesla only has built a Gigafactory 1/7 the planned overall size. And that their car manufacture “plan” is about 500,000 a year by 2020 or later to begin with. Talk of millions should wait until they deliver at least 100,000 per year. Scale – that is a silly problem. Companies like GM produce 27-28 Thousand Chevy Cruze per month out of one Ohio plant. Telsa is still in its infancy and not guaranteed to take overall market share during any sort of “EV Revolution”. They would need 15 more Billion just to build a 2nd and 3rd battery plant, 2nd and third Freemont and more – to what, produce just 1.5 Million units a year? Remember that all this talk about batteries and scale is to “show Tesla might”. Might in the form of first-advantage. It’s batteries. In historic terms, they have always been a commodity. They will become a commodity in the future and high speed product construction will occur and economics will drive out high-cost providers. Let… Read more »
He also “brushes over” the details of graphene, which can be extremely expensive in terms of flakes versus oxides. ————– The Quality Of The Graphene Affects The Price The price of graphene is linked to its quality, and not all applications require superb material quality. For example, graphene oxide powder (graphene functionalized with oxygen and hydrogen) is inexpensive and has been used to make a conductive graphene paper, for DNA analysis, and for other advanced composite and biotechnology applications. Graphene oxide in solution sells for 99 euros per 250 mL from Graphenea. However, the electronic properties of graphene oxide at the moment are not sufficiently good for batteries, flexible touch screens, solar cells, LEDs, smart windows, and other advanced opto-electronic applications. Mechanically exfoliated graphene (obtained with the famous “scotch tape” technique) comes in small, high-quality flakes. Exfoliated graphene has so far shown to hold the best physical properties, reaching towards theoretically predicted current conduction, mechanical strength, etc. The coverage of mechanically exfoliated graphene, however, is only on the order of a few small flakes per square centimeter, not nearly enough for applications. In addition, the price of such graphene can be on the order of several thousands of dollars per… Read more »

So by 2020, the price of pack is $6250 less, but the $7500+$2500 tax credit is expired or reduced. That puts Tesla behind where they are right now. But bigger car makers just start to make their competing cars with this new battery technology (because it is profitable then) which reduces margins for Tesla. No matter how you spin this story there is no way Tesla can make the kind of money that their stock price suggest.

We have seen this exact scenario with solar panels. The cost to make them has fallen by 95%, but competition has reduced margins to next to nothing and all solar makers struggling. Some of their stock trade at 1/10th of all time highs. Oh boy if they could sell their panels at same price when manufacturing cost was 10 times higher they could make a killing. Explain that analyst Dan Dolev.

Because the Chinese government is subsidizing the export of solar panels.

Sure they can this with lithium ion cells too. To crash the market price would be orders of magnitude more expensive than subsidizing solar panels.

I’d like to know how this compares with the price trajectory of LG Chem and other, smaller manufacturers. Presumably small manufacturers can take advantages of new technologies first, as they can by newer machines to do small volumes, but that may not be a compelling advantage.

This analyst suggest switching from an NCA chemistry to an NMC chemistry. But even if that would lover costs per kWh, I’m not sure Tesla would go for it — the energy density of NMC is lower than that of NCA.
(Best figures I could find (from 2011) was 150-220 Wh/kg versus 200-260.)

Other than that, this seems like a very sensible report.

How many kWh on board do we really need. I mean REALLY need? How many Tesla owners are driving 200+ miles per day? They utilize the high density batteries in large amounts so as to lower the # of cycles on the batteries over time. To take a Leaf 250 miles, you recharge 3 times (or more) and once for a Tesla. Less battery wear per year driving the same miles. The big battery also allows higher draw and “Insane + Ludicrous” modes. But it doesn’t solve the problem. EV drivers can be highly successful with 35-40 kWh on board. You can drive most of your miles with half the batteries – allowing for double the cars to be built with the same battery plant footprint. That puts more “butts in seats” and doesn’t hoard the batteries into fewer cars. Look at the Volt – people driving 80-90% on electricity with a “minimum” of batteries – 16-18 kWh with only 10.4 to 14 kWh usable (2011 to 2016 models). To say someone needs 85-90 kWh on board is, well, Ludicrous. It may be a gluttonous want – but it surely is not needed. How many Tesla owners drive 20 miles… Read more »

“How many kWh on board do we really need. I mean REALLY need?”

As much as the customer demands. It is apparent that the vast majority of Tesla buyers opt for the 85 kWh pack. You can theorize all you want, in the end it is the customer who decides with his/her wallet.

Or look at me with my Fusion Energi. I’m about 90% electric with my tiny battery (7.6 kWh). I plan my trips around chargers as I know the electric motor is much more efficient to use.

People are said to drive 40-50 miles a day on average.

What people are planning on is the range AWARENESS of occasionally driving more, which is what Tesla is answering with Superchargers. People are aware that gas stations are everywhere and so they have forgotten that gas cars can have ‘range anxiety’ also (as people still run out of gas).