Investor Engineering Discussions

[nativewolf]

Just thinking logically it seems that they have placed the equivalent of the floor pan on top of the pack. That should save assembly time but probably at least a couple of robots that don't need to place seats. Now I wonder how much else might be done to increase preassembly.

Much of what people have been noticing in Grüneheide does suggest there si even more efficiency than we had imagined. That does add color to the Elon statement last year that 'Model Y built in Germany will be a different car than the one built in California'. I cannot find that quote right now thus '...'.

Thus far maybe none of us actually imagine just how consequential having two huge casting plus the third piece of the structural pack. The weight reduction alone should yield a long series of virtuous consequences.

Do any of us have the capacity to make logical estimates of those things?

On the surface it seems they will have:
-reduced Capex for robots and line,
-reduced labor costs,
-increased speeds,
-far increased body rigidity,
-reduction in warranty costs,
-and more.

The engineers among us could maybe do that,
by that I mean almost all of us in thread except me,
I only counted beans and a few other trivial things.

Right I was just now thinking that the human effort related to the interior assembly probably causes quite a bit of physical stress/trauma that a more comfortable operating position alone would result in less workers comp type lost productivity. Plus fewer robots, etc etc. Virtuous cycle should extend from space utilization to efficiency of labor related to well being not just working faster. etc etc. Anyway, your point exactly I suppose. Neat

It's going to be a hoot to dry to duplicate this reinvention of auto manufacturing to sustainable hardwood production. Elon has given a speech when he discussed the low utilization rate of factories and in our field it is certainly believable. He quoted something like 2-3% of space was dedicated to production. I'd like to watch all of that section of his original speech, some outdoor event that was clipped in the VW concerns youtube bit ( a feel good story for Tesla folks because the evil empires finally are concerned validation story sort of thing). I found the presentation boring but I liked that few seconds when EM was discussing factory utilization.

 

[mongo]

Has anyone determined - either from this shot or any of the others emanating from the tour - how and where the seats are indeed fastened to the pack? I do see something looking like a fastener - bolt-head, perhaps? - in the tan riser below the air ducts in the center, affixed to the raised corrugation of the pack, but there must be many more.

Seats bolt to the risers which are welded to the rails which are welded to a separate piece of metal which spans the full width and wraps over the pack. That assembly is placed on top of the pack and probably [possibly] glued. [I expect at least a foam piece for BSR, but may be able to be separated]
I'd guess the entire seat sub assembly is done separately then mated to the pack + drive unit assembly.

 

[mongo]

IMO most surprising was the seats apparently mounted on the battery pack before it is married with the car body, no more robots reaching inside the car...

Other notable mentions:-

• 2 new paint colors?
• A look at the 4680 structural pack... (some debates about the cooling mechanism)

More generally the amount of information and access that was available was surprising... more being gradually revealed as people trawl through video footage would not be a surprise..

AFAIK, only the initial Model 3 line used automated robots for these tasks, later lines used humans operating lift devices.
These will still be needed for some interior pieces like the dash/IP and rear seat.

 

[MC3OZ]

AFAIK, only the initial Model 3 line used automated robots for these tasks, later lines used humans operating lift devices.
These will still be needed for some interior pieces like the dash/IP and rear seat.

Yes, I did wonder if they could install the dash before the pack.

We haven't seen the actual process.

It seems this is part of speeding up the line, it will be interesting to see the actual process running.

 

[MC3OZ]

I wonder if inserting the seats on the battery pack would allow the $25K model to have a metal roof and if that would be a significant saving.

They would need to install a headliner.

The metal roof could be a cast frame with stamped panels.

 

[mongo]

I wonder if inserting the seats on the battery pack would allow the $25K model to have a metal roof and if that would be a significant saving.

They would need to install a headliner.

The metal roof could be a cast frame with stamped panels.

I don't think anything gets installed through the roof currently, so seats on pack is a nonfactor.

 

[MC3OZ]

I don't think anything gets installed through the roof currently, so seats on pack is a nonfactor.

It is probably a question of how much clearance the robots need...

I seem to remember some connection between the glass roof and the installation of the seats, but that could be a faulty memory..

I think the main reason for seats on pack is speeding up the line... processes can occur in parallel.

 

[scaesare]

Side cooling??? And I'm not seeing any adhesive in that picture? Maybe it was left out because this is a display pack?

The side cooling surprises me.

I had expected that cell-to-cell adhesive bonding would be preferred for structural rigidity. On one teardown of previous packs I recall seeing, the cooling ribbon didn't seem particularly robust and rather easily bent. It would seem that having the between every other row of cells would compromise the stucture.

Perhaps:

1- The new cooling ribbon is sturdier
2- The adhesive backfill compensates for this
3- I'm making much ado about nothing and it's not really an issue

 

[MC3OZ]

The side cooling surprises me.

I had expected that cell-to-cell adhesive bonding would be preferred for structural rigidity. On one teardown of previous packs I recall seeing, the cooling ribbon didn't seem particularly robust and rather easily bent. It would seem that having the between every other row of cells would compromise the stucture.

Perhaps:

1- The new cooling ribbon is sturdier
2- The adhesive backfill compensates for this
3- I'm making much ado about nothing and it's not really an issue

Technically what Elon said was cells bonded to the top and bottom face sheets, perhaps the cooling ribbons don't prevent that, and are not part of the structural strength.

 

[scaesare]

Technically what Elon said was cells bonded to the top and bottom face sheets, perhaps the cooling ribbons don't prevent that, and are not part of the structural strength.

However he also likened it to a honeycomb structure. In a honeycomb, each "cell" is not independent of the adjacent cells. It's that bond that gives the structure torsional rigidity in addition to shear strength.

 

[MC3OZ]

However he also likened it to a honeycomb structure. In a honeycomb, each "cell" is not independent of the adjacent cells. It's that bond that gives the structure torsional rigidity in addition to shear strength.

I understand your point, however, it did look like the cooling ribbon was not full height, meaning the cells could be bonded to adjacent cells on top of the ribbon.

But the real answer is, I assume Tesla has modelled and tested this.... I was just as surprised as you were to see the ribbon..

 

[mongo]

It is probably a question of how much clearance the robots need...

I seem to remember some connection between the glass roof and the installation of the seats, but that could be a faulty memory..

I think the main reason for seats on pack is speeding up the line... processes can occur in parallel.

There was much conjecture here at TMC regarding inserting things through the roof, but none came to pass. IIRC, the SR version was also originally going to have a metal roof as a differentiator.

 

[mongo]

The side cooling surprises me.

I had expected that cell-to-cell adhesive bonding would be preferred for structural rigidity. On one teardown of previous packs I recall seeing, the cooling ribbon didn't seem particularly robust and rather easily bent. It would seem that having the between every other row of cells would compromise the stucture.

Perhaps:

1- The new cooling ribbon is sturdier
2- The adhesive backfill compensates for this
3- I'm making much ado about nothing and it's not really an issue

Bottom cooling would put the thermal structure between the cells and pack, making it a critical load path and requiring bonding on both sides. It would also increase pack height, increase ambient thermal loss, and increase risk of leak due to pack damage.

With side cooling, the thermal conduction path is a bit worse (bottom of ribbon does most of the transfer), but the cell tops and bottoms can directly bond to the pack structure. With 100% epoxy fill, if you concider the ribbon as non structural, there are still multiple dual cell row ribs spanning width wise and the top and bottom of the sheet along with epoxy act as tension and compression members lengthwise. Sort of like a series of solid I beams flange welded together. Also, the ribbon zig-zags by half a cell diameter, which reduces the weakening effect.

Would no ribbon be stiffer? Yeah, probably; however, this is much stiffer than typical rocker panels + stamped floor + pack with modules provides.

 

[mongo]

Yeah, less modules allows finer control on pack voltage. Assuming all modules are the same, the total number of bricks would now only need to be a multiple of 5 instead of 16. Previous packs were 96s overall with each module having 6 bricks in series. Granularity of 4.2*16= 67.2V.
Five modules gives granularity of 4.2*5= 21V. Setting these up with 20 bricks each gets 100s for 4% more voltage.

Design constraint then becomes area or cell count. 100kWh old pack was 8,256 cells. Starting at 100s, each cell added/ removed from a brick is a delta total cell count of 100 cells, or about 1.2kWh.
Much easier to adjust size/weight/ capacity.

100s83p = 8,300 cells
100s82p = 8,200
105s78p = 8,192
105s79p = 8,295

Or 110s72p=7,920 cells, 99kWh, 462 V (@4.2V per cell)

 

[RobStark]

 

[MC3OZ]

• After my conversation with a source familiar with the exciting details of the Giga Berlin press, I'm more confident than ever that the manufacturer's true competitive advantage is yet to come

The VW board is in a panic, rushing from one crisis meeting and wake-up call to the next, and the first top maneuverer may be to fired, Wöllenstein CEO VW China who received a clear warning. If he doesn't hit the low end of 80-100k IDs in China in 2021, he will likely be fired.

View attachment 721456

According to my source, Tesla is in talks to take a next step with a 12,000 ton Giga Casting Machine for a full body casting. The 8,000 ton casting machine for CT is already a done deal.

If you want to hear more, listen to my video

This is interesting, I think Alex would have a good source, so this is accurate in some way...

Elon was asked a pointed question on full body casting at Berlin, I wonder if the questioner heard the same rumour..


Elon ruled out full body casting or said it doesn;t make a lot of sense, Elon is careful about how he answers questions...

Mulling this over a 12,000 ton casting machine might make sense for vehicle like a Sprinter van.
IMO the floor, roof and the 2 rear sides could be cast,, 2 A-Pillars could be cast, and the whole structure simply bolted together via locating pins..

One question would be the finish on the cast roof and sides:

• I don't know if it makes a lot of sense to mount stamped panels on top of the frame.
• I'm not sure how painting the cast panels would go...
• A wrap is one option that would work, especially for commercial vans, which might need a company color scheme,.

A cast roof makes some sense for a vehicle without a glass roof.. the rear cargo section of a van would never have a glass roof.

 

[JRP3]

12V lithium battery opened up, CATL small format prismatic NMC cells with some impressive charge/discharge specs

 

[MC3OZ]

Tesla confirms acquisition of battery startup in new patent

Tesla confirmed that it quietly acquired Colorado-based battery startup SiILion, Inc., in a new patent for a silicon-based battery anode....

electrek.co

“Large-scale anodes containing high weight percentages of silicon suitable for use in lithium-ion energy storage devices and batteries, and methods of manufacturing the same, are described. The anode material described herein can include a film cast on a current collector substrate, with the film including a plurality of active material particles and a conductive polymer membrane coated over the active material particles. In some embodiments, the conductive polymer membrane comprises polyacrylonitrile (PAN). The method of manufacturing the anode material can include preparation of a slurry including the active material particles and the conductive polymer material, casting the slurry on a current collector substrate, and subjecting the composite material to drying and heat treatments.”

Tesla may have quietly acquired a new lithium-ion battery cell startup in CO

Recent observations by active members of the Tesla community suggest that the electric car maker may have acquired a lithium-ion battery cell specialist startup from Louisville, Colorado. If these speculations prove accurate, Tesla may very well be on the cusp of extending its lead in the...

www.teslarati.com

Tesla's SilLion acquisition and Roadrunner batteries will make EVs inevitable

Tesla’s Project Roadrunner is expected to usher in a new generation of batteries that would widen the gap between the electric car maker and its rivals in the EV industry. To accomplish this goal, Tesla has been acquiring companies whose technologies would allow it to develop and produce the...

www.teslarati.com

The Highest Energy Li-ion Battery: Unlocking the Potential of the Silicon Anode and Nickel-rich NMC Cathode • scientia.global

www.scientia.global

With this philosophy in mind, the team at SiILion has undertaken the task of redesigning the old lithium-ion battery system, and their approach has already revealed some impressive achievements. SiILion has shown that when integrated into their unique system, state-of-the-art ‘nickel-rich’ cathodes and silicon anodes demonstrate a much-improved structural stability and safety, even at high temperatures. Dr Molina Piper also says that ‘SiILion has created the first viable 80% (by weight) silicon lithium-ion battery anode, capable of integration into standard electrode manufacturing processes’.

I have a long standing hunch that Tesla migh t be intending to use more Silicon than most assume...

The slurry method is probably to allow partners like LG, CATL and Panasonic to make the high nickel / high silicon cells..

So Tesla would be sharing Silicon, and Tabless technology, but not necessarily sharing DBE. or maybe the patent just covers the slurry method

 

[mongo]

Tesla confirms acquisition of battery startup in new patent

Tesla confirmed that it quietly acquired Colorado-based battery startup SiILion, Inc., in a new patent for a silicon-based battery anode....

electrek.co

Tesla may have quietly acquired a new lithium-ion battery cell startup in CO

Recent observations by active members of the Tesla community suggest that the electric car maker may have acquired a lithium-ion battery cell specialist startup from Louisville, Colorado. If these speculations prove accurate, Tesla may very well be on the cusp of extending its lead in the...

www.teslarati.com

Tesla's SilLion acquisition and Roadrunner batteries will make EVs inevitable

Tesla’s Project Roadrunner is expected to usher in a new generation of batteries that would widen the gap between the electric car maker and its rivals in the EV industry. To accomplish this goal, Tesla has been acquiring companies whose technologies would allow it to develop and produce the...

www.teslarati.com

The Highest Energy Li-ion Battery: Unlocking the Potential of the Silicon Anode and Nickel-rich NMC Cathode • scientia.global

www.scientia.global

I have a long standing hunch that Tesla migh t be intending to use more Silicon than most assume...

The slurry method is probably to allow partners like LG, CATL and Panasonic to make the high nickel / high silicon cells..

So Tesla would be sharing Silicon, and Tabless technology, but not necessarily sharing DBE. or maybe the patent just covers the slurry method

Yah, Battery day showed them moving to an all Silicon anode:

 

[MC3OZ]

Yah, Battery day showed them moving to an all Silicon anode:
View attachment 731001

Yes, the amount of Silicon in the anode is not 100% clear. but battery day flags a clear increase.

When we saw the anode foil at Berlin, that looked dark, indicating at least some Graphite...

When in doubt I always take battery day literally, often I find people have a bias towards a more 'normal' rather than 'literal' interpretation of battery day.