Published on February 3rd, 2019 | by Dr. Maximilian Holland
February 3rd, 2019 by Dr. Maximilian HollandÂ
â€” Elon Musk (@elonmusk) January 31, 2019
Tesla of course has its own huge lithium-ion 2170 cell production facility at the Gigafactory in Nevada, the worldâ€™s biggest battery plant, which is fast ramping up production capacity. Even with this, however, there is so much demand for Teslaâ€™s products (both its EVs and its stationary battery storage products) that Tesla can always put more cells to good use.
Tesla is looking to quickly build out the Gigafactory 3 in Shanghai, and start making Model 3 vehicles in decent volume for local customers by the end of 2019. Rather than supply China Model 3 production solely with the 2170 cells shipped over from Nevada, Tesla is also open to sourcing cells from Japan or, ideally, from locally within China. Or some mix of all three.
Letâ€™s look at the full context of the comment, and then break it down. The question was put by Colin Langan, US Autos analyst at UBS:
â€śYou commented that you expect [local EV production in] China to be online by the end of the year, but thereâ€™s a lot of articles [about] the battery supplier â€” [that] youâ€™re looking at different battery suppliers. But, I mean, do you have a battery supplier? Because it seems kind of close to when production is supposed to start.â€ť
Tesla CEO Elon Musk answered as follows:
â€śWell, thereâ€™s really 3 things: thereâ€™s the cell, the module, and the pack. Weâ€™ll be making the module and the pack, so itâ€™s really just a question of cell supply. We can essentially use any high energy density 2170 chemistry. We expect it to be a combination of: cells produced at our Gigafactory in Nevada; cells produced in Japan; and cells produced locally in China. We feel confident of sufficient supply to hit the 3,000 units a week.â€ť
This was quite a revelation to me. We know that Tesla has significant technological and scientific investment in the chemistry of the cells it produces in partnership with Panasonic at the Nevada Gigafactory. How do we know this? Tesla has an exclusive 5 year research agreement with lithium-ion battery expert Jeff Dahn and his entire lab team at Dalhousie University, in which Tesla has invested tens of millions of dollars.
Tesla also has filed patents on lithium-ion cell chemistry, such as the recent example of a patent disclosure which â€śincludes two-additive electrolyte systems that enhance performance and lifetime of Li-ion batteries, while reducing costs from other systems that rely on more additives.â€ť This one was filed just within the past week.
These kinds of fundamental cell chemistry research efforts are not undertaken lightly, and involve a great deal of effort and experimentation by highly trained scientists. It might, therefore, have seemed reasonable to believe that Teslaâ€™s advanced cell chemistry know-how is a large part of the â€śsecret sauceâ€ť that gives the company an edge in itsÂ market-leading EVs and energy storage products. But Muskâ€™s comment that â€śWe can essentially use any high energy density 2170 chemistryâ€ť suggests that much of Teslaâ€™s technological performance and cost-efficiency edge comes from its module and battery pack design, engineering, and production assembly.
Charles MorrisÂ recently wrote a detailed account of Teslaâ€™s apparent desire to keep its module and pack production technology under wraps. If Tesla is indeed happy to work with alternative high energy 2170 cells for its China EV production, this supports the idea that the module and pack design, and assembly, are where much of the secret sauce is lurking.
This is further supported by Jack Rickardâ€™s recent teardown of the Model 3â€™s 2170 cells and pack, which we also covered this last week. Jack was impressed by the energy density of Teslaâ€™s 2170 cell, coming in at an estimated 247 Wh/kg, a step up from the 240 Wh/kg 18650 cells he analysed from a Model S 85 kWh battery pack. But Jack was even more impressed with the efficiency of the Model 3â€™s module and battery-pack design and engineering.
According to Jackâ€™s analysis, the Model S 85 kWh battery pack has an energy density (at the pack level) of 126.7 Wh/kg, with DC-DC converter, charger, and assorted junction boxes not included in the pack itself, instead being housed elsewhere in the vehicle. The Model 3 pack, on the other hand, does have all these elements built into the pack itself, and yet still manages a pack-level energy density ofÂ 159.5 Wh/kg. Thatâ€™s technological progress.
All in all, it appears that at least a significant portion of Teslaâ€™s technological and price-performance edge may indeed derive from the Model 3â€™s battery module and overall pack design, and packaging (and high-volume assembly). In this case, it begins to make sense that Tesla can work much of its magic withÂ any high energy density 2170 chemistry.Â If there are other aspects of an independent supplierâ€™s chemistry that differs from the chemistry of Teslaâ€™s in-house cells, a small amount of additional over-provisioning of, for example, capacity and/or power density could likely accommodate the difference, whilst being viable in cost terms.
Local Chinese battery suppliers will no doubt be keen to win cell supply contracts for Teslaâ€™s Shanghai Gigafactory. Tesla is quickly targeting local production volumes of 3,000 EVs per week (at relatively large kWh quantities per vehicle). These should combine to make such a cell supply contract likely the highest kWh supply volume of any single model of EV made in China at the end of 2019. It may also mean that â€” at least in the short-term â€” there will be less supply of cells, or battery production capacity, available to legacy vehicle manufacturers in China.
This is coming just at the time when legacy automakers are facing new, tighter local regulations to increase the proportion of EVs they manufacture for the crucial China market. Looks like their decisions to delay the shift to EVs for as long as possible, and not invest effort to build out their own battery production facilities, may again prove to have been unwise. Much of their potential independent cell supply could now belong to Tesla. â€śHa ha ha haâ€¦.â€ťÂ