Everything I've seen talks of a production defect (actually two). Misaligned separator combined with a torn anode resulting in a future short.
Misaligned factory robot may have sparked Chevy Bolt battery fires
960 works out nicely to 96s10p, but that might be a little oversized. 96s9p is 864 cells. Depends how close they got to 5x energy per cell, and what their range goal is.
If they tweak pack voltage, they can adjust by 9 or 10 cell increments.
I'm not sure we've got the whole story on the table yet regarding the LG pouch cells.
What's more this is not just about the GM (Chevy Bolt), it is also about Hyundai (Kona), Renault (Zoe), VAG (ID3, some ID4, presumably ? the badge cloned Skodas and Seats as well) maybe others I missed and also LG residential battery storage. It may also affect some of the Li drop-in replacements for lead-acid 12V batteries made by LG (e.g. in Peugot e208/e2008).
(When I was figuring out the global battery supply picture I posted up over the last few days one of the more helpful googling sessions was searching for the number of affected recalled cars from each brand, as that set a lower bound on each OEM's take from LG).
So what's not yet being clearly stated ? How about : is this a manufacturing defect, or a design defect, or multiple defects ? Originally the accounts were fingering just one misaligned production machine in just one LG factory in China. But then it became apparet that an LG plant in S Korea was also producing affected cells. Now you can argue the toss whether the misaligned separator and torn anode are mechanical alignment issues traceable to manufacture or design ("we made it the way you designed it" - vs - "you didn't make it the way we designed it"), but it is not obvious that the immediate action of voltage limiting (keeping SoC above x% and below y%) are really addressing that issue. Instead most people driving a nickel-rich car are used to SoC management in order to stop dendrite growth causing premature cell failure. Yet the press accounts don't seem to be discussing that issue at all. Are these in fact even all the same chemistry ? Combined with the physical stresses of flexible pouches expanding and contracting as they charge and discharge and get hotter and colder, all stuffed together into a big tub shoehorned into whatever nook and cranny of the ICE-conversion-du-jour is in play. With whatever BMS the cost-cutters thought they could get away with, set to deliver the range that the marketeers insisted on, irrespective of the consequences. And when did it start ....... because for sure it doesn't just neatly only affect things that were just made during one calendar year and are still in-warranty.
The known recall costs and apportionments that I noted down in my research so far are:
Hyundai / LG = $900m combined writedown @ 30/70 split, i.e. $630m to LG
GM / LG = $1900m combined, $1200 of which to LG
Frankly the press are not paying enough attention here, either the motoring press or the financial press. Do they have the skills to join the dots ? And where are the regulatory authorities in all this - highly conspicuous by their absence and lack of curiousity imho.
Despite all this LG doubled cell production, from 34 GWh in 2021 to 62 GWh in 2022, enough to go from half-a-million-cars to one-million-cars at the typical pack sizes we have in play here.
Replacing 34 GWh of cells at (say) $100/kWh is a $3,400-million problem. Just the known Hyundai and GM contributions together add to $1,830-million.
Oh yeah, and all in the same period of time when LG Chem was spinning out LG Energy Solutions (i.e. the battery business) into a separately quoted (listed) stock with an IPO. Whilst desperately sweeping all this under the carpet.
Like I said, I'm not sure all this story is on the table yet.
/cloudfront-us-east-2.images.arcpublishing.com/reuters/H4HFE3EQTVIJRKG7DUQUOOWXGU.jpg)
agree, there will not be a giga in france. 
