[dead]
That doesn't sound right. You can already get high-quality, prismatic LFP cells for ~50 US$/kWh [1] from European distributors at low-volumes, so bulk from China should be even cheaper. And there's no way that balancing and BMS cost an additional 150-250 US$/kWh at scale, not even a tenth of that.
[1] https://www.nkon.nl/novat/eve-lf230-prismatic-230ah-230a-lif...
We don't have a lot of car makers and we do not have a huge amount of EV models.
Either you are doing this on small scale, than you can select one or two specific EV models and use these batterie packs or you do it in big scale, than you can also easily adapt the most x batterie packs and only use these across your system.
If you go down to the cell level, its more effort true but then you probably only handel cell types of a handfull of manufacturer again.
This is only practical if you reuse the whole pack, or at least the modules. And for that to work well you also need a lot of complex software to keep the packs working well with each other (like balancing power levels between the packs).
BMS software is no joke, it is already hard and complex enough when using brand new battery packs and cells of the same chemistry and manufacturer and wear level. Any kind of mixing massively increases the complexity and safety concerns.
Create a MetaBMS. The BMS of the cell pack doesn't care if a car requests energy between 0 and max and a BMS doesn't care either if it gets 0-100 kWh.
My car can easily charge from a household plug up to fast charger. My car consumes very random amounts of energy while i drive.
Complex, by t very much doable. Toyota implemented such a system with enough packs to power a Mazda factory [1]
> However, you need to factor in significant integration costs (~$50-75/kWh) to build compatible BMS systems and thermal management.
Shouldn't this be a once-off cost per battery-pack version? Or was this your armortized cost for your deployment capacity. If you've written the BMS for a 2019 Model 3 Panasonic battey pack once, won't you reuse it for all the subsequent battery packs of the same model/SKU?
A BMS isn't just software. It requires a µC, voltage and temperature sensors for cell monitoring, and power electronics for cell balancing. However, those are all comparatively cheap, especially at scale, and the quoted cost of 50–75 US$/kWh looks ridiculously overpriced to me.
Isn't that exactly what the article is about? Have you actually read the article?
In what sense? I'm a newbie, but curious because I'm working on stuff related to https://mesastandards.org/mesa-der-std/.