Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Poll: How is Tesla getting to 100kWh? and some history.

How is Tesla getting to 100kwh?

  • Case 1

    Votes: 3 15.0%
  • Case 2

    Votes: 0 0.0%
  • Case 3

    Votes: 3 15.0%
  • Case 4

    Votes: 6 30.0%
  • SRSLY Why do we care about this?

    Votes: 8 40.0%

  • Total voters
    20
This site may earn commission on affiliate links.

LargeHamCollider

Battery cells != scalable
Jan 10, 2015
990
1,885
United States
Tesla's battery marketing labels ("85kwh" etc.) aren't precise descriptions of pack capacity making this a bit trickier, for background here's a brief history of Tesla's batteries.

Some time after the car went on sale 85 owners noted that their cars never showed anywhere near 85kwh used. This was explained by appealing to a 9kwh "brick protection" portion of the battery. Later wk057 showed (conclusively imo) that the "brick protection" consists of only 4kwh and the 85kwh pack only holds ~80.7kwh.

A teardown (I think by wk057 again) of a 60 pack showed that 60's have 14 modules of 384 cells each. This showed the 60 pack was actually slightly under rated, its actual capacity being ~80.7kwh*[14*384]/[16*444] which is about 61kwh.

The new 70kwh pack has 14 full 444 cell modules for ~80.7kwh*14/16 or about 71kwh, the 70kwh pack gets you 87.5% the range of an 85 and is a great deal imo. The 70kwh pack is closer to the 85 than the numbers 85 and 70 suggest because the 85 is actually closer to 80.7 and the 70 is actually closer to 71.

The new 90kwh pack is 6% better than the 85 according to Tesla, and range numbers bear this out. But 6% better than 80.7kwh is only 85.5kwh...

Which brings us to the new 75 and 100kwh packs. The 75kwh model X pack gets 17 more miles of range than the 70 putting it at about 76kwh. Given that the 90 pack is actually only 85.5kwh the difference in range between the 75 and 90 of only 20 miles makes sense if we assume the extra mass of the 90 costs it 2-3% in the efficiency department.

But how is Tesla getting to 75kwh? and more interestingly, how are they getting to 100kwh?

There are many possibilities, here are a few of the possible cases arranged sort of in order of simplicity.

Case 1
A dramatic 15-20% increase in cell energy density has occurred, 100kwh is achieved by dropping the new cells into the existing 16 module packs.
The 75kwh pack is achieved by using 12 full modules or by using 14 384 cell modules as the old 60 pack used to, probably the latter to keep the voltage up.

This theory is attractive but I have doubts about such a large advancement in cell chemistry.

Case 2
75kwh is achieved by taking the same 14 modules in the current 70kwh packand upgrading the cells to the latest chemistry. If the latest chemistry is about 2% better than what's in the 90kwh pack, which seems reasonable, this would get you the ~76 kwh required.

But Case 2 has difficulty explaining the 100kwh pack, to get it we'd have to add 2 modules on top of the 2% advancement in chemistry, this would get us to about 98kwh. The problem with adding modules is it would increase pack voltage or require a redesign of each module to keep pack voltage the same. Increasing pack voltage would almost certainly require a redesign of the motor controller, and possibly a motor redesign. Also adding modules means the pack probably cannot be retro-fitted to existing cars. Its also possible that they just take an 87kwh (85.5kwh plus 2%) pack and call it a 100, but that would be disappointing...

Case 3
Cell geometry change in 1 dimension.
If there's a cell geometry change in one dimension it will probably be to lengthen the cell, Tesla is constrained by the height of their modules and a .4 inch increase in module height would correspond to a ~15% increase in pack capacity, which sounds worth it to me. Additionally, lengthening cells would allow for minimal changes in cell and pack assembly equipment while also minimizing the decrease in gravimetric energy density that accompanies increasing cell size.
The 75kwh pack would be achieved by using 12 full modules or by using 14 384 cell modules as the old 60 pack used to, probably the latter to keep the voltage up.

It's likely that this pack would not be retro-fittable to older vehicles but it's *possible* that it would be. Some benefits of this scenario would be that increased power and charge speed would be more or less guaranteed, which wouldn't be the case for a pure chemistry improvement scenario.

Case 4
Cell geometry change in 2 dimension (or 3 dimensions if you want to use cartesian coordinates on a cylinder).
If the cells change in length and radius it would be difficult to anticipate what might happen, 18750s would be the obvious candidate for Case 3 but in Case 4 here it could be 20700s or 20750s or 22700s or numerous other combinations of height and radius. For this reason i won't even try to speculate as to how it might play out.
Case 4 is possible but I think it's the most unlikely of the possibilities at this time due to the number of engineering changes it would require.


Right now I'm leaning towards 1 and 3 as the most likely cases but my level of confidence is not high. Cast a vote, the collective wisdom of TMC has been correct before.

Also if the new 100 pack is a true 100kwh pack, the Model S 100D should have a range of about 340 miles:), and the Model X 100D about 300.
 
  • Informative
Reactions: Johan
Elon and JB stated on an earnings call a while back that the battery would change in both height and width. Any new chemistry advances plus the volumetric density by cell size can take the packs beyond 100kw. Tesla can make the battery any size they wish, no need to conform to existing standards for consumer electronics.

Gigafactory will produce larger cells that the Typ 18650

Yes, but I doubt these cells will be from the gigafactory.
 
Where else would they come from? If you're investing in new tooling you can dictate the specs you need. Panasonic can spit out 18650's as they currently are for use in powerwalls, 70kw and 75kw packs. Until they build out the rest of the Gigafactory they may be limited on how many cells they can produce. It's possible the new cells would only be used in the 100kw packs.

IMO, this needs to happen way before the Model 3 is released, they need to prove they can pump these new cells out at the economies of scale to meet the price target of $35K on the Model 3.
 
Excellent post and interesting speculation.

The new "gigafactory cells" will be 20700 size according to several sources. It is not likely that high volume manufacturing equipment a third format, with change in only one dimension, would also be developed for an interim period. This would be a good question for people in the cell manufacturing business, of which I am not.

I speculate that Panasonic would never consider making the first production of a new cell design overseas in Nevada. It would be more efficient to install the first production line at home in Japan, and debug it locally. Since the gigafactory is supposed to be making cells by the second half of this year, I expect Panasonic to be making 20700 cells now. There is no reason these cells could not be used in Tesla's 75 and 100 kWh packs.

Note that the gigafactory plan was to make 35 GWh cells per year and import 15 GWh, for a total 50 GWh pack production, so Panasonic will need long term Japan production of the new cells.

I am voting for case 4, since Tesla plans to redesign their packs for the 20700 format cells, and it makes sense to get that done before the gigafactory starts up. That way they have a proven design with a new factory, instead of both new design and new factory.

GSP
 
Last edited:
  • Like
Reactions: LargeHamCollider
@GSP, care to share these sources? 20700 definitely seems reasonable but I can't think of a time this geometry was specifically mentioned.

@shokunin, Panasonic has factories in Osaka and Hyogo that could do it (and one other one in Japan that I'm forgetting) I believe they still have two factories operating in China as well.
 
Dr. Alderman of Advanced Automotive Battery Conference (AABC) revealed the 20700 size in his Tesla Battery report. It was included in the free extract of the report on the AABC website. I consider Dr. A dermal an authoritative source, since he has contacts with every company in the battery industry, including manufacturing equipment suppliers, end of line test equipment suppliers, cell material suppliers, and all of the cell manufacturers and their automaker customers.

Also, Telsa exec(s) have mentioned the 20700 size. IIRC J.B. Straubel mentioned it in a battery presentation he made last year. I think someone also mentioned it during a Tesla earnings conference call.

GSP
 
  • Like
Reactions: LargeHamCollider
It appears that autocorrect butchered Dr. Menahem Anderman's name in my post. I did not catch it soon enough to correct it. Darn new TMC host software with its 60 min limit for editing.

GSP