You're talking about battery cells, but you need to look at the battery _pack_ energy density. The model 3 battery pack, which uses the same cells as semi, is about 170Wh/kg. The numbers in TFA are very reasonable estimates.
I already addressed that by including mass overhead for the overall battery. And sure, it’s always possible to assume more conservative numbers for a whole battery and assume Tesla DIDN’T do any engineering work to reduce overall battery mass, but that’s not something you should put in a headline.
(Tesla has typically done better than their peers in overall battery specific energy as well as energy per mile.)
You're using a 16-17% figure for packaging overhead? That's low. Even with relatively expensive construction techniques (for example if you're doing EVTOL) you see 20-25% packaging overhead.
I agree, but a long haul, full payload electric Semi is a very challenging project comparable to an electric aircraft since you’re limited to near the same maximum total weight. I’m merely pointing out that such a battery pack is one possible solution, one that cannot be ruled out based on public information.
The in-use 2170 batteries are apparently 250Wh/kg. With that figure and a more realistic packaging overhead of 30% I'm seeing something like 190Wh/kg, which is more in line with your observation.
From the numbers I can find, the Model S 85kWh has 7104 18650 cells - if we use 49g for mass of cell, that's 767lb of cell vs the pack weight of 1200lb~, or a very rough packaging overhead of ~36%
This is a naïve analysis but it helps assuage my ego, as I've worked on designs for 26650 based packs where we would have been quite happy to hit 30% overhead :) That being said, I realize people can, and do in fact make more mass efficient packs.
