GlideOutside

Pull up a chair, I've got some tea☕🔋 A reliable anonymous source has told me some interesting information about the Panasonic 2170 produced at Giga Nevada: 1) Tesla is working with Panasonic on a new version of the 2170. a) In conjunction with the cell can supplier, they intend to thin out the cell can of the 2170 in the same way that they did with the 4680. For the 4680 thinning out the cell can added about 20 Wh/kg, or about 8% of the 11% energy density increase we see between the Gen 1 and Gen 2 4680. I doubt the same will happen with the 2170, but a conservative guess would be a 5% energy density increase, bringing the 2170 up from about 273 Wh/kg 287 Wh/kg If they did manage an 8% increase, it would mean about 295 Wh/kg. I think this is unlikely, but worth mentioning. Either way, thinning out the cell can seems reasonable and plausible. The Gen 1 4680 had the same hoop stress capacity as the current Panasonic 2170. They put the Gen 1 on a diet to create the Gen 2 4680, and so it only makes sense to put the 2170 on a diet too. b) They are looking at using carbon nanotubes in the new version of the 2170. As I said in my video on the topic, the effect on energy density and charge speed of using carbon nanotubes in the 2170 is heavily dependent on where they are used, the quantity, and the type (single- vs multi-wall carbon nanotubes) At minimum I'd expect slightly faster charging speeds (maybe 10% faster) and maybe a 1% energy density increase. Possible, but less likely is that the carbon nanotubes, combined with silicon, add another 5% energy density increase and much faster charge rates or higher charge rates (325 kW vs 250kW?) BYD is already using carbon nanotubes in their batteries, so using carbon nanotubes and I think the whole industry is moving in this direction. They're expensive, but the fact that they provide such huge benefits for a given weight, adding them will at most cost about $1/kWh. That is, carbon nanotubes make sense and aren't far fetched. c) These new cells are currently being cycle tested and they're working on tooling in Japan. So, it's probably at least 1 year away, probably more like 2 years before we'd see these in a vehicle? Either way, if all of that is true, the next version of the Panasonic 2170 could be around 300 Wh/kg and provide 5-10% (8% being likely all things considered) or additional range. That would mean 20-40 miles of additional range (depending on the trim, etc). 2) Giga Nevada has stabilized the past few years at a run rate of around 6.25 million cells per day, which is an annual run rate of around 40 GWh/yr. Given the ups and downs due to factors like vehicle refreshes in the past couple of years, they've accumulated a stockpile of hundreds of millions of cells. That is, probably at least 2-3 weeks' worth of backlog, maybe more than 4 weeks? So currently, based on that information, they're definitely not cell constrained for EV batteries cells in the U.S. I'll be interested to see how that changes later this year (soon hopefully) as they ramp "more affordable models." 3) As a final note, this source has been reliable in the past. That doesn't guarantee the information above is all correct or will come to pass. But as I said, it all sounds reasonable and likely from a technical perspective. It's very rare, maybe once every 1-2 years, that I get 'inside' information, and when I do, I share it as quickly as possible. 🤠