4680 cell design, chassis integration & factory discussion for investors

[bxr140]

Great for robots who do the assembly but how about when the pack needs to be serviced by humans?

Remember that the service model is what keeps traditional automakers in business, so its in their best interest to condition the consumer base to applaud serviceability. That model doesn't exist with Tesla; rather they're clearly taking more first principals approach of trading the the fleet level upside of the more cost and mass efficient construction method against the fleet level downside to more difficult (including potentially impossible) servicing.

Its unfortunate that this scenario has the potential to screw over an individual consumer, but the bottom line is that if the total consumer dollars saved on initial purchase price and incremental efficiency gains outweigh the total consumer dollars "wasted" on scrapping what would otherwise be a serviceable vehicle, then its a win for us consumers. And if that scenario can also result in additional per unit profit for Tesla, then its a win for Tesla (and TSLA).

Intuitively, its going to be a huge win for everyone.

Its also worth addressing the statistical relevance of "what if the battery needs servicing?"
--Rarely do packs actually need servicing, and its pretty fair to assume that as technology evolves/matures, that will continue to decrease. (Note that you're talking to someone who drove around with a loaner pack for almost two years) One can also safely assume that Tesla is addressing the root cause(s) of the historical "need to service the pack" issues, which will necessarily drive down the rate of 'need to service' even lower than it is today.
--The inability to remove a pack from a car does not necessarily mean the pack is not serviceable; certainly a significant portion of historical issues will still be serviceable...just in a potentially more inconvenient way.
--From a collision perspective this is a low risk evolution, as the write-off criteria isn't going to change relative to structural vs cosmetic damage.

So bottom line, there are going to be very few scenarios where a vehicle [with an integrated pack] will have to be scrapped were a similar vehicle [with a removable pack] would be serviceable without first salvaging.

 

[Buckminster]

First 45 mins has a great explanation on how the cells are manufactured.

 

[MC3OZ]

First 45 mins has a great explanation on how the cells are manufactured.

The following comment in that YouTube video is interesting:-

Just another hypothesis about the anode and cathode caps is that it allows for expansion and contraction of the cell with high silicone content?

This is an interesting idea, but we need to be sure the expansion isn't sufficient to short anything out..

 

[MC3OZ]

I've been thinking more about the little plug in the bottom cell.... following on form the discussions in this video,,,
I think the rest of the pack can be glued to the bottom face,sheet, whether or not that acts as cooling depends on whether or not the fire retarding glue is also thermally conductive.
But I think they might have small holes/indentations in the bottom face sheet lining up with these plugs... sufficient to let the plug escape all the gas to escape and IMO to allow the electrolyte to escape, of the electrolyte escapes the faulty cell might quickly die...
IMO an indentation is the most likely possibly as it provides some method of containing all of the stuff ejected from the faulty cell... the 2nd likely option is simply a hole corresponding to the plug and bottom tray.

 

[petit_bateau]

If it is a completely dry electrolyte (is it ?) then it won't run out the bottom. I'm not sure if it is so cannot comment.

I think the anode/cathode end arrangements with the various cuts and laser welds serve the following purposes:
- conduction of current (obviously !);
- accomodation of tolerances in mechanical assembly;
- create flow path for high temperature fused plug, and for assembly tooling;
- accomodation of thermal expansion/contraction in main cell without onwards transfer of those stresses to the relatively weak welds;
- possibly also act as high current fuses, as the slits act to define the cross sectional area of the active conductor at some bottleneck points.

Just imho based on visual observation without doing any sums.

 

[MP3Mike]

If it is a completely dry electrolyte (is it ?) then it won't run out the bottom.

No, they are using a dry electrode process. They still use a liquid electrolyte. Though I think most of it is absorbed by the separators, so not much of it would drain out. But I think it would evaporate as it heated up and escape as gasses.

 

[petit_bateau]

No, they are using a dry electrode process. They still use a liquid electrolyte. Though I think most of it is absorbed by the separators, so not much of it would drain out. But I think it would evaporate as it heated up and escape as gasses.

Thanks. At what point does that get added ? Is there a vapour space in the very top of the can or is it fully liquid-filled ?

 

[MP3Mike]

Thanks. At what point does that get added ? Is there a vapour space in the very top of the can or is it fully liquid-filled ?

The electrolyte gets injected through the hole at the bottom after the whole cell has been assembled. Then the plug/rivet is put in to seal the cell. (This is a different design/process than how the 18650/2170 cells are currently manufactured.) They inject a very specific amount of electrolyte to appropriately soak the separators, it is not fully liquid filled. (There should be no liquid sloshing around if you tilt the cell.)

 

[petit_bateau]

The electrolyte gets injected through the hole at the bottom after the whole cell has been assembled. Then the plug/rivet is put in to seal the cell. (This is a different design/process than how the 18650/2170 cells are currently manufactured.) They inject a very specific amount of electrolyte to appropriately soak the separators, it is not fully liquid filled. (There should be no liquid sloshing around if you tilt the cell.)

Thanks, helpful.
If it is not fiully liquid filled then it could slosh around. If it is fully liquid filled then it could not slosh around. In shipping terminology the latter is called "tanks pressed full".
If fully filled then there would be a much faster pressure response if there was any gassing off, however minor. They may prefer to have a small gas cap (air, nitrogen, whatever) above the liquid (very small, at the height of the upper white insulator band) if they want a slower pressure response. The slower pressure response would tend to minimise most external can deformation, i.e. the gas phase acts as a shock absorber). It all depends what they want.
Is it known if the plug/vent actuates due to pressure, or temperature, or some determined combination ?
Is the electrolyte gelled in any way ?

 

[MC3OZ]

Is it known if the plug/vent actuates due to pressure, or temperature, or some determined combination ?

It isn't actually known it is speculation,... but the cell needs some sort of safety mechanism,.... and the suggestion is a build up of gas pressure forces the plug out.. but temperature might also melt it... I was wondering if they were using any metal with a relatively low melting point anywhere... if so, that could be part of the safety mechanism...

If the electrolyte can escape in some form liquid/gas and as a result the cell dies, then that might be timely enough to mean they don't need some other form of electrical disconnection,

My other thought was the process of plug being ejected out the bottom might somehow trigger an electrical disconnection, this is a bit far fetched, but if they have done that,, it would be very interesting to find out how it works.

 

[petit_bateau]

It isn't actually known it is speculation,... but the cell needs some sort of safety mechanism,.... and the suggestion is a build up of gas pressure forces the plug out.. but temperature might also melt it... I was wondering if they were using any metal with a relatively low melting point anywhere... if so, that could be part of the safety mechanism...

If the electrolyte can escape in some form liquid/gas and as a result the cell dies, then that might be timely enough to mean they don't need some other form of electrical disconnection,

My other thought was the process of plug being ejected out the bottom might somehow trigger an electrical disconnection, this is a bit far fetched, but if they have done that,, it would be very interesting to find out how it works.

Do we know if the electrolyte fill goes in through the same plug/vent ?
Do we know if the electrolyte is gelled ?
Do we know if it is a one-shot electrolyte fill, or if there can be multiple shots during the conditioning process ?

 

[MC3OZ]

Do we know if the electrolyte fill goes in through the same plug/vent ?
Do we know if the electrolyte is gelled ?
Do we know if it is a one-shot electrolyte fill, or if there can be multiple shots during the conditioning process ?

My impression based on the speculation.
Yes electroyte fill is via the same hole prior to plug insertion.
The electroyte is liquid with a 1 shot fill.
Conditioning is via the formation process, the one-shot probably contains extra lithium for the SEI layer.

 

[MC3OZ]

Just kicking off some speculation about the volumes of cell production at Austin, the mix of chemistry and the volumes of various models supported.
It is highly speculative, but gies and idea of the options, in particular how vehicles can be made with a mix of chemistries, using a different chemistry for the shorter range model.

Assumptions:-

• Austin eventually makes at least 200 GWh of cells annually
• Split is 60 GWh High Nickel (HN), 80 GWh Nickel Manganese (NM), 60 GWh Iron (LFP)
• Semi LR 1 MWh (HN) SR 600 kWh (NM)
• Cybertruck LR 200 kWh (HN) MR 150 kWh (NM) SR 120 kWh (NM)
• Model Y LR 85 kWh (NM) ,SR 65 kWh (LFP)
• Model 3 LR 80 kWh (NM), SR 55 kWh (LFP
• Model C (2) - LR 45 kWh (LFP), SR 35 kWh (LFP)

HN - 30 GWh Semi LR 30,000 per year, 30 GWh Cybertruck (LR) 150,000 per year.
NM - 400,000 Model Y LR = 34 GWh, 12 GWh = 80,000 Cybertruck MR 12 GWh = 100,000 Cyber truck SR,
12 GWh = 20,000 SR Semi,.. 10 GWh = 125,000 Model 3 LR
LFP - 100,000 Model Y SR = 6.5 GWh, 400,000 Model 3 SR = 22 GWh --- approx 30 GWh for Model C.
15 GWh = 330,000 Model C LR, 15 GWh = 430,00 Model C SR

Totals Semi = 50,000, Cybertruck = 330,000 Model Y = 500,000, Model 3 = 525,000, Model C = 760,000

 

[petit_bateau]

Just kicking off some speculation about the volumes of cell production at Austin,.........

Where are you getting your info re the multirollers from ?

Are they trying to run them in parallel or in series or both ?

(I guess it is OK if they line the roller ends up on the white stripes)

 

[MC3OZ]

Where are you getting your info re the multirollers from ?

Are they trying to run them in parallel or in series or both ?

(I guess it is OK if they line the roller ends up on the white stripes)

Gali (Hyperchange YouTube) mentioned the small rollers in a video.
There was some discussion about it on TMC. 7-10 days ago?

Another YouTube video mentioned that the DBE material wasn't drenitng the big rollet but was pushing it out of position. They also mentioned thst tje pilot line was seperate and was running fine.

This isn't 100% verified, but there seems to be leaks from within the team.

Multiple people including Elon have mentioned that the pilot line is going ok.

 

[heltok]

Where are you getting your info re the multirollers from ?

Are they trying to run them in parallel or in series or both ?

(I guess it is OK if they line the roller ends up on the white stripes)

 

[Buckminster]

00:00 Intro
00:50 Help finding a 4680
01:41 Annual Meeting // Fremont 50% Boost
02:47 Annual Meeting // Supply Chain & Battery Shortages
04:35 Annual Meeting // Shortages & Vehicle Delays
06:22 Annual Meeting // Final Thoughts on Supply Chain Shortages
07:10 Annual Meeting // 4680 Production and Vehicle Ramps
08:41 Annual Meeting // Recycling Shortfall
09:49 Annual Meeting // Lithium Supply and Alternatives (Sodium)
11:58 Annual Meeting // LFP Cathode and Nevada Lithium
12:18 Annual Meeting // Spooking the Horses and Nevada Lithium
13:44 Annual Meeting // Electric Planes (eSuperVTOL)
15:33 Annual Meeting // Megapack Factory and Battery Suppliers
17:17 Giga Berlin // Where will Berlin’s initial cells come from?
18:07 Giga Berlin // The Battery Packs
21:31 Giga Berlin // 12km Anode Roll
23:04 Giga Berlin // Cell ‘Wiring’
23:47 Tabless Cell Leak // What is this thing?
28:16 Plaid Model S Pack Teardown // Significant Improvements
30:49 Quarterly Report // LFP Standard Range & 4680 Pack Progress
32:03 Q3 Earnings Call // When will 4680s go into vehicles?
35:19 Q3 Earnings Call // $25,000 Tesla
36:00 Q3 Earnings Call // Localisation of LFP & LFP Suppliers
38:20 Q3 Earnings Call // Commodity Prices Question
38:34 Q3 Earnings Call // Elon not on the Call - Thoughts
39:29 Q3 Earnings Call // Commodity Prices – Gigacasting Riff
41:18 Q3 Earnings Call // Commodity Prices and Contracts
42:28 Q3 Earnings Call // Reducing vehicle cost despite high material prices
44:01 Q3 Earnings Call // Diversification
44:44 Q3 Earnings Call // Strategy for Anode Materials
47:56 Q3 Earnings Call // Closing thoughts – News Tidal Wave

 

[Buckminster]

 

[Buckminster]

It's side cooling tubes, just like all other Tesla vehicles.

View attachment 791328

Better views of cooling separators.

https://twitter.com/x/status/1512406145847209988

View attachment 791329

 

[Buckminster]

https://twitter.com/x/status/1557917516789387264

Interesting Sandy says the canister is thinner than they thought, The Limiting Factor tear down said it was really thick. So likely that was where Sandy got his expectation from and he now has a newer version of the cell.

Also, sounds like Munro did power/energy testing of the cells from their tear down with higher numbers than they expected, can´t wait to see the numbers!