Tesla Battery Investor Day

[JRP3]

The above scenario is highly unlikely. More reasonable B Day speculation can be found here:

 

[Buckminster]

The above scenario is highly unlikely. More reasonable B Day speculation can be found here:

Agree that this is good and I have great respect for anyone trying to pull all the various threads together and make whole cloth, but:

1. he plays fast and loose with percentages, for instance the 10% increase in cost for a single crystal cathode is 10% of what? [not the $88/kWh he applies it to almost certainly].

2. if the cells are increased to 4070 they are almost 4 times the power so the same cell production rate gives 4x the GWh line capacity. He then suggests 2x or 4x the cell production rate which then would lead to 8x to 16x the line capacity, not the 4x he then uses.

3. Tesla already reportedly use 5% silicon in their anode, an increase to 20% gives most of the advantage of a completely silicon anode. I think it is likely that the cells will have more silicon and hence a greater capacity increase. Panasonic cells may not use the higher silicon content that is consistent with their 20% overall capacity increase.

4. tabless electrodes do not just allow heat to be extracted faster (perhaps with plate cooling), but also cause less heat to be generated due to lower resistance.

5. cell filling with electrolyte must be part of the picture, else why would Tesla buy a specialist in cell filling equipment, they could just have bought on the open market if the cell filling was similar to existing lines.

6. roadster is an opportunity to start with low volume production of the advanced cell to pack line (building cells directly in the pack). A high scrapage rate does not no matter so much in an expensive performance above all sports car.

7. Advanced cell to pack is I think the ultimate goal, this could enable whole battery packs to be made at machine gun rates and in less volume than at present, and at lower cost. Advanced cell to pack is very difficult, verging on the impossible. If they can demo that then my mind really would be blown.

 

[MC3OZ]

if the cells are increased to 4070 they are almost 4 times the power.

Maxwell DBE possibly allows bigger cells, but 4070 isn't confirmed.

Overall Jordan (TLF) has done a good job, it is important to not over-hype the event and set unrealistic expectations... but I guess Elon saying "mind blowing" doesn't help..

I'm still expecting Roadrunner cells in Plaid Model S, as are most people..

Model Y made in Berlin might have Roadrunner cells...

 

[Mich]

Something in my head:
There are the cells, modules, packs. We all know that modules are somehow EoL'd and we expect not much between the cell and the pack.
There is also "revolutionary changes" in the way cars are build in the talk. (Model Y at Giga Berlin).

What if the structural lower part of the car frame is replaced by a single molded part and the cells are just plugged in? The molded part has 3000 (or whatever the amount of cells is) holes for cells and pipes for cooling liquid. For sure there's more to think of like how is each cell added to the electric circuit; but I like the idea.

Musk wants to eliminate complex structural parts with molded parts.
Musk wants to eliminate modules in the cell-module-pack assambly.

 

[JRP3]

What if the structural lower part of the car frame is replaced by a single molded part and the cells are just plugged in

That leaves cells in direct contact with structural members with no added protections. Plus you can't remove a pack for repair or replacement. Unlikely.

 

[Doggydogworld]

Interesting schematic of a li-ion cell factory in this white paper by Sila who develops silicon anodes. It's not to scale, but gives a sense of the equipment involved. Maxwell's DBE eliminates the solvent evaporation ovens which are part of the coating stage.

 

[ZOE5IVE]

I think they will move away from the cylinder cell design as "The limiting factor" suggested. Their reliability is certainly higher and I think the main reason they went cylindrical was related to them being smaller. Cylindrical batteries would limit them scaling at this point. And whatever they choose should be easier to recycle.

 

[Buckminster]

Anyone think that Tesla could licence battery tech for mobile phones etc.?

 

[heltok]

It’s late here in Europe so I don’t have time to watch all of it tonight, but this was just posted:

Battery day will not only be about chemistry and manufacturing but also about the future of energy and this video is a good talk about that. https://youtu.be/kd-SzFv4XxQ

 

[Buckminster]

Something in my head:
There are the cells, modules, packs. We all know that modules are somehow EoL'd and we expect not much between the cell and the pack.
There is also "revolutionary changes" in the way cars are build in the talk. (Model Y at Giga Berlin).

What if the structural lower part of the car frame is replaced by a single molded part and the cells are just plugged in? The molded part has 3000 (or whatever the amount of cells is) holes for cells and pipes for cooling liquid. For sure there's more to think of like how is each cell added to the electric circuit; but I like the idea.

Musk wants to eliminate complex structural parts with molded parts.
Musk wants to eliminate modules in the cell-module-pack assambly.

Crazy idea:
Could Tesla remove the metal cell casing in addition to modules?

This would involve a new plastic insulating case (or several layers of the separator) and use a gigacast pocketed floorpan to replace some of the metal case functionality (such as fire resistance). The pockets would need to be for 2 or more cells so that main busbars are all on the top. You would then have to cool the floorpan rather than the cells so that the cells stay contained. You would have to fill the pocket with electrolyte.

This would be in-line with first principle thinking plus:

“It is going to be the first time that there will be a transformation of the core structural design of the vehicle. It’s quite a big thing.”

 

[jeewee3000]

Anyone think that Tesla could licence battery tech for mobile phones etc.?

That's a completely different use case. Mobile phone batteries are optimized for fast charging and have way faster degradation (lower cycle life).

Not saying Tesla couldn't innovate on that front, but it would require a completely seperate design and production effort (i.e. from scratch). As they already have enough on their plate (and bigger fish to fry), I don't feel the need for it.

 

[JRP3]

You would then have to cool the floorpan rather than the cells so that the cells stay contained. You would have to fill the pocket with electrolyte.

That leaves cells in direct contact with structural members with no added protections. Plus you can't remove a pack for repair or replacement. Unlikely.

Also plastic is an insulator which would make temperature control much more difficult.

 

[Todd Burch]

Surprised there’s been no discussion of this yet:

https://twitter.com/fredericlambert/status/1305937320114352128?s=21

Could this be Tesla’s new “tabless” battery cell?

 

[Buckminster]

Surprised there’s been no discussion of this yet:

https://twitter.com/fredericlambert/status/1305937320114352128?s=21

Could this be Tesla’s new “tabless” battery cell?

4070 is included in my summary of BD expectations.

 

[Todd Burch]

Allegedly confirmed according to Fwed at Ewektwek:

First look at Tesla's new battery cell produced in-house with Roadrunner - Electrek

 

[Doggydogworld]

4070 is included in my summary of BD expectations.

Looks closer to 4090, but the pic could be distorted.

 

[Buckminster]

Looks closer to 4090, but the pic could be distorted.

I am reliably informed things can look larger in smaller hands.

 

[JRP3]

It does look larger than 4070. Someone suggested 4269, because Elon

 

[JRP3]

For local reference:

 

[Thomas Edison]

If this cell is the new one and they are going cell to pack, would that free up 20mm in height and the pack wouldn’t be any taller? That’s my thinking.