Extent of Tesla's battery advantage

A lot of information about the battery is in this thread:
Model S Battery Pack - Cost Per kWh Estimate

The talk I generally read about is that the large prismatic cells are coming down in price and eroding Tesla's cost advantage. However, they really can't ever be cheaper than small cells because small cells have better yield and can be effectively cooled so they can run "hotter" so to speak and have a higher energy density and produce more power.

The Chevy Bolt is probably legitimate but they are targeting way less than Tesla for production at 20,000-30,000 vehicles. So they aren't really competing unless the market is smaller than Tesla thinks it is but it really doesn't matter in that case because Tesla will pretty much go broke if they don't get Fremont up to 500,000 cars/year by the end of the decade.

stealthology said:

Appreciate it. Do you (or anyone else) know of any legacy OEM who is using Tesla's approach of liquid cooled small cells, or do they have that under patent?

Click to expand...

If anyone is using them they are doing so in secret. As far as the patent goes Tesla does have it but they have made it available for anyone to use.

Extent of the Tesla's battery advantage is very significant. There was an excellent battery pack density chart posted by 32no in the Long Term thread. Given that the historical average speed of advances in battery technology leads to doubling of the density in 10 years (7.18% increase yearly), the competitors are years behind.

Somehow all the other EV manufacturers have convinced themselves that the large pouch cells are the way to go, even in the face of Tesla's success with their cells. Personally, I just don't understand why they cling to the pouch cells. Yeah, I know engineers are notorious for clinging to their pet ideas, but across all other EVs?

Cosmacelf said:

Somehow all the other EV manufacturers have convinced themselves that the large pouch cells are the way to go, even in the face of Tesla's success with their cells. Personally, I just don't understand why they cling to the pouch cells. Yeah, I know engineers are notorious for clinging to their pet ideas, but across all other EVs?

Click to expand...

Larger format cells allow them to reduce the amount of interconnects and enclosure overhead, given more volume versus external surface area. Pouch cells allow them to pack the cells more closely than cylindrical.

However, the reason why it hasn't worked out as they plan is because chemistries that are safe enough to be put in large format cells have lower density in the first place, so any advantage in reducing overhead is eliminated; and given the need for cooling for the high density chemistry, the gaps created by a cylindrical format can still be taken advantage of.

What they are banking on is a chemistry that is high density, safe, and doesn't need liquid cooling. NMC is poised to be that chemistry. However, it is still less dense than the NCA Tesla is using, but if they can build a pack using it that has no cooling (similar to Leaf pack), it may get pretty close to a Tesla pack after all the overhead is accounted for.

I do worry about how Tesla's e pluribus unum battery approach will scale to the 100,000x level. It appears that there is a great deal of labor involved in battery assembly. That area is not on public tours, so I don't have a clear picture of how much automation could be, or has been, achieved, but the idea of having floors full of workers soldering cells together is out of step with delivering "millions" of cars annually.

Robert.Boston said:

I do worry about how Tesla's e pluribus unum battery approach will scale to the 100,000x level. It appears that there is a great deal of labor involved in battery assembly. That area is not on public tours, so I don't have a clear picture of how much automation could be, or has been, achieved, but the idea of having floors full of workers soldering cells together is out of step with delivering "millions" of cars annually.

Click to expand...

I'm not sure how much battery assembly is automated. I assume cell into casing is automated because they've shown the process, but I'm not sure about the soldering. I know that battery installation involves manual labor, but they're working on automating that as well (which is also necessary for effective swap).

Given that the benefit of automation grows with volume, I simply assume that Tesla wants to automate everything it can.

Several of us did tour the battery assembly area at Christmas.
It is almost entirely automated - all the cell unpacking, testing, assembly and connections etc. are fully automatic. As far as I recall the only activity that seemed to involve humans was inserting and positioning the cooling tubes.

Cosmacelf said:

Somehow all the other EV manufacturers have convinced themselves that the large pouch cells are the way to go, even in the face of Tesla's success with their cells. Personally, I just don't understand why they cling to the pouch cells. Yeah, I know engineers are notorious for clinging to their pet ideas, but across all other EVs?

Click to expand...

The idea was that in the long run the larger-format cells would be easier to assemble into batteries. JB Straubel has noted that a large-format cell simply hides some of the complexity from the battery manufacturer.

It's similar to the preference of some manufacturers, particularly the Japanese, to avoid a TMS, because it potentially has the lowest cost and greatest efficiency. US manufacturers focused on TMS because they saw it as a route to lowering battery costs, because a TMS's ability to control temperature makes it easier to manufacture a suitable cell. For now the TMS approach is winning, but that doesn't mean that skipping the TMS won't be better in the longer term.

JB Straubel said:

For the immediate future we see 18650 as the most compelling. Believe me, we challenge it constantly. It always ends up being very controversial, for reasons I don’t totally understand. Nobody gives a damn about the shape and size of your fuel tank! But for some reason the shape and size of what you put your chemicals in to carry your energy in an EV is super controversial. What people should really debate is what are the nature of the chemicals inside; they’re what determines the cost and performance.

...
What keeps us in that general shape and size is the production and cost efficiency. We’re seeing price points that none of the larger-format cells are able to meet.

Click to expand...

It's worth noting that LG Chem has said that they're at 99% yield with their large-format cells, so the yield advantage might not be that great.

RichardL said:

Several of us did tour the battery assembly area at Christmas.
It is almost entirely automated - all the cell unpacking, testing, assembly and connections etc. are fully automatic. As far as I recall the only activity that seemed to involve humans was inserting and positioning the cooling tubes.

Click to expand...

Great (and comforting) info; thanks, Richard.

I don't see how GM can have the Bolt ready by 2017. Unless LG already has the breakthrough ready today. I suspect they'll just delay - or sell it at a big loss. It's still annoying that they might essentially beat Tesla to the 3.

ItsNotAboutTheMoney said:

It's worth noting that LG Chem has said that they're at 99% yield with their large-format cells, so the yield advantage might not be that great.

Click to expand...

That is a valid point, but does it require more expensive equipment or more downtime for calibration or extra purification processes to achieve that yield? Either way even if large format gets to cost parity will they be able to achieve the same energy density and power output than one can achieve with the small cells?

Beckler said:

I don't see how GM can have the Bolt ready by 2017. Unless LG already has the breakthrough ready today. I suspect they'll just delay - or sell it at a big loss. It's still annoying that they might essentially beat Tesla to the 3.

Click to expand...

If the Bolt is ever actually produced, I would assume Chevy would launch it as soon as possible for the halo effect irrespective of profitability, assuming not horrible and given low volumes and materiality to EPS. I also hope Tesla will launch the 3 at its goal price even if the GF economics are not yet achieved, with consideration to sales and battery cost projections over 2017-2024. The cost and msrp comparison will be interesting with the Bolt getting the full federal tax credit, and the Tesla 3 less to none. Will most US buyers choose 3 if within $8k or so after tax due to Superchargers and vehicle specs, even though Chevy brand position is improving with Volt and Cruze?

It will be interesting to see if the tax credit approach is ever converted into a point-of-sale incentive as per Obama's current budget proposal (which is DOA).

CHGolferJim said:

If the Bolt is ever actually produced,

Click to expand...

Is GM lying? The Bolt (or whatever they decide to call it), is going to be produced. Their CEO introduced it as a concept and less than 1 month later they announce production of it. Chances are good that the Bolt will beat the M3 to market. That being said, would I ever buy a Bolt over a M3? No way. But to many, GM is a known whereas Tesla is an unknown.

Agree with those last points, but seems likely GM is exaggerating most details. As a Volt owner, you may have a better feel.

Interesting about the perceived lack of a temperature monitoring system (is that what TMS means?). Arguably, Tesla actually could use MORE cooling, not less.For ordinary driving in all weather conditions it works great. But it hits a limit if you are driving at a track, or going full out on the autobahn.

Beckler said:

I don't see how GM can have the Bolt ready by 2017. Unless LG already has the breakthrough ready today. I suspect they'll just delay - or sell it at a big loss. It's still annoying that they might essentially beat Tesla to the 3.

Click to expand...

LG is aiming to have the next generation ready at the right price for some time in 2016, I believe, which is why they're getting manufacturers to sign an MOU that they'll build a "200 mile" BEV. If GM's aiming for 2016 with the Bolt, it implies that they're already testing the cell, and LG's issue will be about production cost. The idea, as I see it, is that both sides will be ready together so that LG Chem would be confidently able to invest in production capacity. It's similar to Tesla and Panasonic, but with more car manufacturers and low initial targets.

And they could well beat the 3 to market, but so have the Leaf, Volt, i3 and soon the Volt 2. Doesn't mean that the those cars will be competition for a mid-sized, aerodynamic long-range BEV with a low-cost, coherent fast-charging network. For BEVs long-range != long-distance, and EREV != BEV.

I'm not annoyed about it. I'm happy. The next generation of PEVs could be the one that loses the price subsidies and creates a self-sustaining plug-in market. That'd still be a huge milestone.

GM may very well beat Tesla to market with the "affordable BEV" in the Bolt. But the real wild card in this scenario is the responsiveness by the dealers to sell the car. Will the salespeople know all there is to know about BEVs? Will they offer a free or discounted J1772 for home charging? Most people don't understand charging speed differential between a 110/15 and a 240/30. If the dealers' past is prologue, they will try to persuade their customers to buy an ICE, and thousands of Bolts will sit at the back of the dealers' lots gathering dust.

Unless GM provides a sizable incentive for the dealer to sell the Bolt, they will languish unsold, I believe. Perhaps a smarter approach for GM would be to try to sell the Bolt to municipalities and other fleet purchasers for starters. That will get the cars into the public awareness through visibility, and the employees who run their errands in them will have first-hand knowledge of their benefits.