Part 1: Tesla 4680 Teardown // Cell Disassembly // 4 hours in 1 hour
ฝัง
- เผยแพร่เมื่อ 1 มิ.ย. 2024
- Part 1 of the Tesla 4680 Teardown where the cell is disassembled. The teardown took 4 hours and this video compresses that down to 1 hour. The data/characterization will be coming soon. The video is timestamped so you can jump around! Who loves you? 😘
**Thanks to to Teardown Team at the LESC at UC San Diego!**
Weikang Li (me) (caption)
Yijie (Jackie) Yin (chief officer)
Varun Gupta (2nd officer)
And of course Shirley Meng for making it happen 😀
***Thanks to COREY CODDINGTON for driving 40 hours round trip to deliver the cell from @HyperChangeTV to UC San Diego!***
Patreon: / thelimitingfactor
Paypal: www.paypal.com/paypalme/limit...
Bitcoin: 3B3UXCZsnoPZxG6vYQ6npsF4TW8f5kK4LH
Teespring: teespring.com/stores/the-limi...
Reddit: / tlfbatteries
Twitter: / limitingthe
Timeline
00:00 Introduction
03:02 Prep
04:19 Voltage Check (OCV)
05:17 Creating a Cut
07:04 Comments on Toughness
07:50 Prying Access to the Copper/Anode
13:17 Copper/Anode Finally Opens
16:31 Removing the Tabless Cover (Flower/Hubcap thing)
19:21 Tabless Electrode Exposed (Copper)
23:33 Cutting the Side Wall
27:57 Jelly Roll Removed and Tabless Electrode (Aluminum)
30:15 Removing Tape Wrap
31:45 No Tabless on Outer Wraps
33:45 Cleaning Up the Glovebox and Collecting Electrolyte Crystals
36:49 A Quick Look at the Aluminum ‘Flower’
37:48 Collecting Electrolyte, Cleaning, Unwrapping Jelly Roll
41:34 Notes on the Cathode
42:30 Folding Cathode
42:56 More Notes on Cathode
43:48 Folding and Measuring Anode
47:44 Cathode and Anode Lengths
47:59 Measuring Width (roughly 70mm)
49:25 Decision to Cut a Piece and Measure Outside Glovebox
49:31 Cutting a Piece
50:42 Discussion on Length and Cutting Coin Cells
52:33 Closing Notes on Day 1
53:37 Day 2 Start - General Discussion About Structure
57:43 Cutting Open the Sidewall to Access the Aluminum Top
01:01:17 Discussion of the Button, Plastic, Aluminum ‘Flower’
01:02:37 Attempting to Deconstruct Aluminum Top Further
01:04:34 More Discussion of the Button, Plastic, Aluminum ‘Flower’
01:06:45 Cutting a Piece of the Sidewall
01:07:29 Sidewall Thickness
01:08:24 Measuring the Thickness of Other Pieces
01:10:40 Comparing the Aluminum and Copper ‘Flower’
01:11:29 The Cell Can is Structural
01:12:19 When Can We Expect Results?
Long Road - Futuremono (YT Free Music)
Dance of the Mammoths - The Whole Other (YT Free Music)
Arp Ascent - DivKid (YT Free Music)
Pizzicato Traveling (Premium Beat)
Chilltronic - RKVC (YT Free Music)
Sad Past - Silent Partner (YT Free Music)
Metamorphism (Premium Beat)
Secret-universe (Premium Beat)
Ocean View - Dyalla (YT Free Music)
Dark Planets (Premium Beat)
#BattChat #BatteryTwitter
Intro Music by Dyalla: Homer Said - วิทยาศาสตร์และเทคโนโลยี
Notes:
1) I originally wasn't going to release this video to the public. That's why it was shot in portrait. It was only for me and suited my purposes. However, I would like to say thanks to the team for holding the phone for 4 hours!
2) Hindsight is 2020. Before tearing it down, we didn't know the shell would be 600 microns thick. The tools used in this teardown work okay for a 2170 that has a ~100-200 micron shell. In the future, we'll use a pipecutter. Dremel? No. Sparks and dust contaminants.
3) When I said that "yeah that's the first time I've seen it" on the cathode/aluminum foil end, I was referring to the fact that I hadn't seen the tabless electrode in its full glory on the cathode side. We've seen them unrolled and seen them teased in footage, but not under full compression straight out of the can (at least from my memory).
Were you able to determine a capacity for the cell?
Thanks for the added explanations. Any chance we'll see links to the photos in the video description?
makes me wonder what happened to the NIKOLA flooded cells they were Allegedilly working on.
Regarding the copper flower on the tabless anode, isn't that to create pressure between the copper tabs so there's less resistance and maybe also better heat transfer? What do you think?
@@alemed01 A kind of spring to keep contact during heating and cooling, a "expansion joint" so to speak.
I used to teach high school math, and students were so bored by the content they would sleep through class. Now I work for Tesla in cell engineering, and fans make entire video series detailing our product.
It means a great deal to me that you and many other TH-camrs are so passionate about and interested in our products. I’ve never worked for a company with such loyal fans and customers.
Ah well, back to work!
Thanks for stopping by Richard! And nice work with the battery cells 🤠
As a very happy owner of a model 3 with a Y on the way: thank you!
@@thelimitingfactor, Richard Anderson Thank you both.
I honestly want to thank all Tesla employees for their hard work. Whether current or past, fired or new hire you all are important to a movement more important than any one individual.
I am very thankful "it works" isn't good enough at Tesla. It's hard to convince people that Tesla's batteries are far more advanced then what they have experienced with cell phones. Not to mention Tesla's BMS. You are doing awsome work at Tesla! Love my Model Y.
Can we give some appreciation for the person holding the camera for all that time?
No, because he was too stupid to turn it 90 degrees.
Thanks Jordan! As a chemical engineer (and having worked in the polymer industry), I cringe when you ask, "is that plastic or rubber."😁 It will be interesting to see/hear the composition of those super important polymer insulators. Are they polyamides (nylon), or polydimethylsiloxanes (silicones), or other, or custom designed from the base monomer units? The reliability engineering on those insulators is also interesting. Simulating the toll of miles and time. Getting them to last hundreds of thousand of miles, in all of the heating and cooling cycles, mechanical stresses of a structural pack, in the presence of the electrolyte...truly amazing work. This must be what Maxwell contributed to the puzzle.
the fact that tesla allow these student to audit their batteries mean that they're confident with their tech. this is a good sign of an honest company and elon goals of pushing EV tech to masses. even if the battery tech shared, it will takes lots of effort to copy it because the secret is not the battery chemistry but the machine that build the machine that lower the production cost and increase the profit margin.
Jordon looks SO happy! I love seeing this excitement. It’s weird that a retired non-technical woman like me is glued to this tear down…. But I know I’m watching the future unfold. Or maybe I should say ‘unroll’ :-)
😂😁
Learning is its own joy. Don't let anyone gatekeep the passion of learning about any topic prevent you from experiencing the things that are important to you.
Great insight to the engineering.
Looks like the 6 arms of the petals form 6 controlled impedance paths to ensure equal current distribution from tabs to buttons.
Well hot damn, that's phrased beautifully!
And the difference in how the “petals” attach to the flower is due to the fact that the Cu electrode is welded to the can along its circumference, while the Al electrode is welded to the nipple in its centre. In both cases the conduction paths are of roughly equal length. Initially I thought this configuration might be due intended to accommodate the thermal expansion during high-C cycles.
I was also trying to see where the point of highest thermal stress might be, i.e chemistry isolated, safety fuse mechanism.
Hi, what branch of engineering should know about the stuff you just said? Chemical? materials? electrical? mechanical ?
600 micron sidewall converts to +/- 16 gauge sheet metal.
That is way larger than I thought.
Most definitely structure-capable
Its like honeycomb structure. torsional rigidity and bend inertia gona be insane.
What's that as a percentage of the whole cell, compared to a 2170?
@@SilentSalad ao normally thickness is about 200um?
What % weight savings from a frame structure holding thin walled cells? I suppose the difference could be calculated between 4680 & 2170 MYs?
@@Crunch_dGH @gary itano i guess only tesla know and has stiffness and fem calculation to support any claims. Holistically speaking. Before only battery case provided to the overall stiffness and structural integrity. So battery cell was total dead weight. Theoretically now the part of cells becomes part of structure. I postulate that means at least that 200 kg goes into using as pure strength factor (yep napkin calulations if stainless steel is used for 800 cells) .
I bet production cell will have less thickness. Or same thickness and just using aluminium. Weight of stainless with 0.2 thick shell is same like 0.6mm thick Aluminium one. Witch is about 75kg. Seems much sensible numbers for both cases.
Arguably stainless steel could eliminate any form of fire retardation potentially. So maybe you can live with poor-ish performance due weaker thermal managements of cells and you can made lighter and cheaper pack because some features can be eliminated further , not only because u got structural benefits in first place.
Driving back from a battery conference. Limiting factor pops up… so I pull over to services for a burger so I can watch on my headphones 😂 cheers Jordan 🤩
Ok, some things I observed:
-The non-conductive white plastic parts are probably PTFE
-The welds on the flower 'tab holders' look like laser welds to me
-The anode 'tab holder' looks to be welded to the anode current collector via resistance spot welding? So they probably need an electrode coming in through the centre of the jelly roll.
So the assembly of the can goes something like this (I think):
1). Jelly roll is wound around mandrel and taped. Two colours of tape to indicate which side is the anode and cathode.
2). The anode flower 'tab holder' is laser welded onto the anode tabs right after winding
3). Somewhere else the anode is crimped to the empty can (with a PTFE spacer inbetween).
4). The jelly roll with the welded holder is inserted into the can, anode side first
5). On the outer anode side a copper electrode is placed, and an electrode is also inserted through the centre of the jelly roll, to press the anode 'flower tab holder' and the anode current collector together and weld them via resistance spot welding (at least that's what it looks like to me) Once the weld is completed the electrode is removed from the centre of the jelly roll.
6). Only then is the cathode flower 'tab holder' welded on via laser welding.
7). electrolite is pumped in (*somehow*)??
8). Lid is placed on and crimped on.
I could be totally off with the electrode spot welding. It's also possible that the mandrel IS the spot-welding electrode? There is a hole in the cathode 'flower tab holder' (see 1:02:11), possibly for a mandrel or for an electrode, but I think it's just way too small for an electrode. Plus you don't want to risk sending current through anything other than the anode and anode flower tab holder. Some problems I can see with my order of assembly is that the cathode flower 'tab holder' is welded on at a late stage, and I don't thing you want loose tabs when your jelly roll is flying through the factory at 120mph. Maybe they have a holder cap to hold everything together?
I also don't fully understand why the cathode makes contact with the side wall of the can rather than the bottom lid? That's weird.
This teardown definitely left me with more questions :D
EDIT: See comments below, probably isn't PTFE, cathode flower is welded in step 2. Thanks @BlueToad and @Eigen Value
Amateur observer here but is it possible that the cathode makes contact with the side wall to better dissipate heat? Remember 4680 uses side cooling ribbons and not bottom cooling ones.
@@BigBoneESB Yeah, that is definitely possible, even for such a small strip of copper. Some back of the envelope calculation gives me that the conductance is something like 5W/°C (assuming the distance to bridge is like 3mm orso). That's pretty good, cause I expect the cell to only produce something in the order of 9W of heat when charging, assuming 98% efficiency. Though then I would also have expected the lobes of the flower tab holder to start from the outside rather than the inside in that case, but maybe all those copper tabs packed together is already more than enough conduction?? Hard to say without simulating. My gut says the 4680 has a lot of headroom for fast charging, as the existance of tabs all around hugely shortens the electrical path compared to a single tab cell. but I could be wrong (as I am an amateur observer myself).
I have a longer comment as it's own thread, but as to step 7: I assume the electrolyte is pumped in at the end after placing and sealing the lid! On the lid there was this little copper nub which I am very sure to be a blind rivet. It would make a lot of sense for the lid to be placed and sealed, then the electrolyte is pumped through the small hole in the middle of the lid after which that hole is plugged with the copper blind rivet that we saw.
PTFE: Pretty Tough, Fucking Expensive
Non-conductive white plastic parts: As an alternative to PTFE, how about polyester film, aka Mylar or Melinex? These are used to electrically insulate within electric motors.
Very impressive credentials of your team seen on their LinkedIn pages. And I am glad to see that Shirley Meng is still involved after moving to University of Chicago.
Oh my gosh. That was incredible to watch. The design and engineering is so exciting to see. Tesla has both designed the casing for its structural strength and as a battery component. But I never thought the can strength would be as high as it is. A lot of tradeoffs in that can to get it to fit both roles. And the lack of any liquid in the coating that was applied to the foil surprised me. I thought it would be a heavy paste but it was dry it seems. Don't know how they roll that out to that exact a film while it's that dry. This shows just what Tesla had to do to design this battery. And they are making these at very high speeds which is hard to believe. No wonder the other battery companies are having a hard time making these batteries. They are totally different from anything else being made.
There was some liquid in there but it was sparse! Some of it appeared to dry out (possibly due to damage), but still pretty thin.
@@thelimitingfactor So were the crystals just the desiccated liquid electrolyte? Or is the electrolyte somewhat solid and crystalline? If there is very little liquid electrolyte then could this be considered a sort of quasi-solid state cell?
It is fun to see the person doing the disassembly have his own onsite audience. That takes a lot of stamina to work through gloves in an isolation chamber for so long. Thank you all for the effort and work.
Oh what a wonderful present for this, the last day, of Q2. Thank you for your work!
Here here. Thank you Jordan
Firing those cannisters out like bullets is impressive on top of amazing. Looking back at battery construction techniques with new respect.
amazing , since tools were ment for 100 - 200 microns of 2170 , and 4680 being more Robust at 600 microns.
@@markplott4820 yes, it's all that momentum in the shifting material.
At this point Jordon I think you could make a video about paint drying and I would be in. I am really appreciative of you sharing this with us. I can hardly believe I'm this invested in this battery and it's tear down.
JORDAN should wear a BOB Ross wig and do the JOY of Painting , I would watch that !!!!!!.
🤔
Depends...
Gloss, or Matt paint?
😉
Okay I must admit the lab is slightly more safe than me with a hammer in my garage. Good job with the disassembly!
Don't sell yourself short. Do you have a high-quality hammer?
Using the ''Master Mechanic'' simply works. Did you wear safety glasses?
@@CiaranMcHale he could ask bigclive for his XRay device..
@@LosZonga - " these Goggles do Nothing, my eyes Burn " - Radioactive Man.
@@CiaranMcHale - Captain Hammer to Rescue.
Spends hours watching a battery teardown then says "I'm going to go watch Star Wars". 😂 I found that amusing. Thanks for the video Jordan.
I hope you guys do the coin cell thing. I think that would be interesting.
Amazing work Jordan!
Congrats to you and the whole team for pulling this off.
Tesla nerds rejoice! :))
So BYD got a tear-down but still unable to figure it out what they are looking at. Tesla is so alien, love that.
Wow Jordon! Another one hit out of the ball park. So much so that I just upped my patreon support. As an arm chair quarterback I would suggest next time a non-sparking oscillating multi-tool equipped with a carbide blade. I use this setup to great success in all of the stuff I manage to intentionally destroy. I’m sure Sandy Munro took great interest in this tear down . Great job!
A good suggestion!
The episode we've all been waiting for. Amazing. Thanks Jordan and all involved for coordinating this! So cool.
That's one clean and unform cathode coating 😲
Just think, this is the product design and manufacturing outcome of an 'auto' company. What other OEM is even close to this level of engineering prowess in terms of battery design - let alone doing it in house, and not outsourced? It blows my mind that Tesla is just on another level in every arm of their business. This was fascinating and educational all at once. Thank you so much Jordon and the lab team!
Lol……….
Great timing. Munro Live will also do a teardown of a brand new Texas model Y with 4680
awesome, and its not easy to work in a glovebox, cuz u have like 3 gloves between you and the item, usually (cotton, rubber, here: insulation/penetration protection gloves), all done in Argon or Nitrogen inert atmosphere
also, if you can get me a giga berlin sample, i can do the same SEM/EDX stuff here in dresden 😉
"Made by humans. On Earth". Yeah right.
Flower comparison at 1:10:50. Copper one with a connection to help electrons flow inward and Aluminum one with a connection to help the electrons flow outward. That's efficiency thinking to the max.
Also average relative relation current vs thickness is in same direction of "flow" of electrons. Each layer is parallel connection essentially.
ive never seen someone look so excited when opening a lid.
very entertaining to watch
Looks like some flower shaped mini bus bars on each side within the cell. Absolutely impressive and elegant design!
Jordan this was just tremendous to watch. Thank you for bringing us along! And kudos to you for reaching out to Tesla before you published. You are a class act and really great at what you do, I love sharing your enthusiasm. Weikang Li was great to watch and his teaser comments at the end of the video have me feeling like a kid waiting for Christmas!
Thank you!
Yeah, this is once in a life time opportunity on tear down an assembly. Thank for sharing.
Thanks to sharing tear down in video.
As far as I can see, the 46800 stands out from conventional by the following features besides the patented outline.
- no apparent pressure sensitive gas vent as seen in 18650 and 21700
- no anode / cathode tabs as seen in 18650 and 21700
Other features not deviate from patent are
- patented anode bristle wafer is off
- instead, a Cu wafer is employed, wafer bridging - anode flaps and cylinder base.
- that Cu wafer has 6 flaps ultra sound welded to a the anode flaps only.
- spokes on Cu wafer not welded at cylinder base.
Concern: Cu wafer touching cylinder bass:-
Anode heat can pass on to Cu wafer directly, but the heated wafer can’t pass anode directly because it isn’t welded to the cylinder base. Therefore creating a significant thermal gradient between battery core and cylinder base, which is an issue on high rate charge and discharge applications.
Remedy?
1) Bypass and omit the 6 spoke wafer. Ultra sound weld the anode flaps directly to cylinder base. Control cylinder base thickness to enable ultra welding onto the anode flaps. This is second best compare to soldering.
2) for best thermal and electrical conduction , use solder bond bridge those anode flaps to cylinder base. Soldering heat by burst magnetic induction at cylinder base to reflow pre deposited paste solder in cylinder base. So solder will flow between anode and cylinder by capillary force. Burst time width for precision heat control is needed to avoid melting of insulation and ion membrane.
Thank you very much for sharing this video. As a person works in the battery industry, seeing this is exciting. The floor plates are called transfer tabs, their function is ensuring a solid electrical connection to the rolls as well as the outer terminals. We have that on cylindrical and prismatic cells. The only type I do not see those are pouch cells as they are directly welded to the outer tabs. The white part on the cathode is called AT9. It is designed to provide electrical isolation, a common design on all cathodes of cells. It is specially preventing touch of Al cathode to Anode graphite as it is the most likely case to cause a catastropic TR.
Questions:
1. Would you mind sharing the photos to us?
2. For the SEM, EDS and other test results, when you would make a video about that or share some data? I cannot wait.
For sure man! So long as Tesla legal doesn't come down on me, I'll be sharing everything...SEM images, chemical composition, and all.
As soon as I have the complete data set, I'll start releasing it 🤠
Thanks so much for your comments!
The 600 micron can will certainly contribute to the structural aspect of the pack. Could it be that it was also necessary to contain the pressures that will be experienced as the contents heat and expand?
This was my first thought, too! Silicon?
The way I look at it, they need some thickness for the structural function within the car, and some other thickness to safely contain the contents of the cell. I gather that the thickness needed for the cell is less, but either way, whatever the lower number, that's how much weight you've saved with this design by having a structural cell.
So if a "normal" cell is 200 micrometers and the "structural" requirement is 600 micrometers, then you've eliminated/saved 200 micrometers from the design.
Then there's also the structural benefit of having pressure within the can. The same way a filled soda can is stronger than an empty one. Not even sure how to speculate on savings from that.
The researcher first says 0.0025", but Jordan hears 0.025" for his calculation. If it's 0.0025", then it's really 63.5 microns, not 635.
EDIT: after watching further, I think the researcher probably added an extra zero and 635 um is probably correct.
Speaking of pressure. I don't see where would it vent in case of overpressure?
Seeing you with a 4680 cell teardown is Christmas!
I'd guess the flower 'tab holders' are there to support thermal expansion/contraction. The one oriented to connect in the middle and bend on the outside, the other oriented to connect on the outside and bend in the middle.
On the Anode side (copper flower side) there is this little copper nub on the lid - I am very sure this was a blind rivet that would be used to essentially plug a hole that was there before in the lid. Also it would be quite likely to me that this rivet would be put in as one of the last steps (after putting on the lid and sealing the battery around the lid). Main reason/assumptions why: 1) why would you have a hole and plug it with a rivet? you need that hole as access or for manufacturing purposes - what would that be?: 2) this could be the hole where electrolyte is pumped into the battery at the end. Essentially the whole thing is assembled and sealed except for this hole where electrolyte can be pumped in and sealed super easily with the blind rivet that has a super predictable displacement on the inside.
What I am surprised by though is that the rivet is made from copper and not aluminum/steel as they usually are. You would usually only use copper if you want the benefits of great electrical or thermal conductivity. Possibly copper rivets have better sealing capabilities though and it's just that.
My thoughts exactly. I also assumed it was for electrolyte filling, but couldn't puzzle out why it was copper. Maybe something to do with coefficients of thermal expansion and maintaining sealing? Or maybe it can act as a vent if the cell starts gassing and builds too much pressure? Neither seem like good answers, so I'm still scratching my head.
so I did some more digging and copper rivets apparently can be much stronger than aluminum and even the hybrid aluminum/steel ones.
Possibly a huge reason though is galvanic corrosion: copper has one of the highest measures of nobility, meaning it will not galvanically corrode due to nobility differences to surrounding (touching) metals (aluminum in contrast could much easier). This comes even more in play when considering that apparently that lid and rivet were electrically isolated from the rest of the battery, meaning it's not "grounded" and one of the metals would (if significantly enough less noble to the other) essentially act as a sacrificial anode to that one and corrode away over time. The stainless steel / copper combination apparently is generally a good combination to prevent this from happening.
(EDIT: depends on the type of stainless! some are good, some are not)
@@janik6501 awesome stuff! The galvanic corrosion factor had actually occurred to me, but I dismissed it since copper/stainless were still dissimilar, but thats great insight and two really good reasons! Thanks for sharing!
@@NickGobin thanks for the idea ping pong! always great to check in whether a thought process makes sense :))
Wow, that's an amazing effort, thank you Jordan an all involved!
It most likely will take me considerably more than the length of the video(s) to completely absorb all the information contained within.
Now send the scraps to Redwood Materials and have them show us how to recycle it and how much material they can reclaim.
To open the can, I'd put the battery in lathe, mount a Drexel tool on the tool turret with thin diamond disc and slowly turn the can by hand (so won't do any interna damage with centrifugal forces). The roll lip would away clean.
I'm not sure how they'd get that in the argon glovebox, but it certainly would be clean
@@thelimitingfactor A mini lathe is only a 7" x 12"
@@thelimitingfactor I'd recommend a wheeled pipe cutter
I've got to intern last year at Kato Rd in cell and seeing this and how much the public appreciates the engineering going through making the cells makes the Tesla engineers happy. Thank you for providing content for everyone limiting factor!
😊 Happy to hear it!
Awesome video. The background music during the teardown made it feel like one of those science fiction movies where we uncovered some mysterious alien technology, later to be followed by an invasion of some sort. Cool!
Lol! Kind of the mood I was going for. Thanks!
Physically there are so many innovative engineering ideas in this battery cell. Having the anode and cathode concentric with one another simplifies battery pack fabrication and is genius in its simplicity. The contacting ‘flowers’ at each end of the jelly roll appear spring loaded accommodating contraction and expansion movement within the can and is another simple yet clever idea. I was bowled over by the tabs extending from the anode AND cathode layers, top and bottom of the jelly roll. I knew about the bottom but not the top too. Like being shown the solution to a Chess ‘mate in four moves’ problem, does one say ‘well of course, that’s the answer!’. We have to wait and see what chemistries are being used but I would guess that there is a lot of clever stuff there too..
Fabulous! Can hardly wait for the final details! (BTW, could have tested for continuity between the stainless steel outside case and the aluminum 'button' using the OEM meter.) Great stuff!!
wow watched every minute. cell structure was more complex than I envisioned. the thickness of the can was a surprise. The two different structures of the flower was interesting but the construction was way more complex and how that's done would be an eye opening video. Also each wrap must be slightly longer than the previous one. Thanks amazing breakdown video.
Jordan says he cut 3 hours out.... pretty sure all three hours was them trying to open the end of the can! 😬
The flower parts are likely different shapes simply to prevent them from being swapped/mistaken at any point in the process. This is especially important given that they have different size holes in the center, and different outer diameters with the larger one connecting to the outside of the can. It is possible that the difference in the flowers could even be recognized by a robotic vision system that may not nessesarily be able to distinguish aluminum and copper. You can "read" them as arrows: the aluminum ones point "inward" to the button connection, and the copper ones point "outward" to the outer casing connections. Also, the designs are similar enough that they can use the same identical laser welding setup on each end, but different in every other way to prevent screwups and easy identification of which end of the cell is which.
Amazing. super like this teardown. We can concretely conclude now 4680 is superior than its brethren. Thank you, Jorden
You beat Sandy on the finish line... 😎
Sandy is at a different finish line this week.
The steel cans, when grouped into a pack, are going to be akin to a hexagonal grid of material. As a unit, they are going to be tough.
I agree. With the cell walls that thick and their grouping pattern. It is almost like a honeycomb.
This feels like I am watching an alien (technology) autopsy. Very cool!
It's possible that the can was super-thick while they were experimenting with the cells. It could be as the production process advances that they thin out the cell casing thickness to a more realistic strength to suit the structural pack requirements. These cells may be very early builds that were pretty much completely depleted for safe loading on conveyor systems etc. Given the time elapsed I would not be surprised if the current generation of cells were more refined / optimised. Sandy might have some current cells pretty soon, he's buying a "Y" from Texas ATM. This assumes that he ends up with 4680's rather than 2170's.
Thank you Jordan, the team and the patreons for giving us all this beautiful insight. This feels like exotic sci-fi tech
thanks for sharing and to all those who worked on this
“Don’t Bust dont bust dont bust” *compressed tabless jelly roll comes out of the tube* “Mission failed, we’ll get ‘em next time”
Seeing the details of the cell design helps to reinforce the benefits of the structural pack design:
1) All of these thick cell casings together basically act as a giant lattice structure
2) The thickness of the casing also helps to reduce the risk of puncture of the cell in case of an accident. This is significantly thicker than what you’d ever use if it was just a cell on its own
🤓💪💪
Absolutely Astounding!!!! Thank you Jordan for making this happen and for sharing this with us👍👍👍👍👍
I would be surprised if they added a significant amount of silicone in the early batteries but I can't wait to find out. I'm sure they did the dry electrode as well. I did not realize they would make the case so thick, definitely makes the batteries much safer from punctures on top of the structural aspect. The energy density has been a bit disappointing imo, but I can kind of see why (and it will probably improve)... would love to see a test on the power density with the tabless design though!!
A very rough calculation.
The concencus seems to be that "the jellyroll of the 2170 is about 32" (812mm) long"
.
If we assume the measured length for this cell to be 3400mm, that rounds up to 4.2 times.
Multiply that by the ratio of the cell heights (80÷70= 1.14) and the total area is +/- 4.8 times.
That's pretty close to the "5x capacity" from battery day?
This might suggest that "5x" was for a "base" 4680 cell using essentially 2170 chemistry.
The "6x power" would then be attributed to the tabless construction giving a 20% performance advantage. (Less wasted through resistive heating?)
A four hour teardown tells me one thing. This structural battery is very, very tough. Using the Maxwell dry ink process and solid crystalline electrolyte makes this a "Solid State" battery to me. If I am wrong, please tell me why.
There is still liquid electrolyte in it. The white separator is porous and contains the electrolyte. So it cannot specify as a solid state battery.
I couldn't see clearly how the bottom of the cell was electrically separated from the electrode but if that's the case then it's also thermally separated and the final nail in the coffin of "bottom cooling".
One end has insulation at the edges. In this video that end was opened first. The other end had a plastic insulating plate with a hole in the middle to contact the button. That was opened last. I don't think any path is vastly thermally superior to the others.
Top vs bottom is.. very fuzzy.
@@rhamph It's not top vs bottom it's bottom vs side. Some speculated that it was using bottom cooling, even though there are obviously side cooling ribbons being used, just as in all of Tesla's other packs. The lack of a thermal path to the bottom puts an end to any further speculation, it's side cooling only.
Bottom cooling is used on Lucid. Side cooling is so much ''cooler'' because it has a bigger contact surface. This batteries are covered in thermal paste and have top to bottom sideways cooling channels. It can draw much more current and it can charge much more faster without overheating.
@@JRP3 Agreed. We even saw side cooling ribbons on display at the Cyber rodeo, I this one is pretty much settled.
Stainless steel (if that is what the can is made of) is generally a poor thermal conductor. I not sure side cooling would be very effective (speaking as an elderly couch potato who imagines he knows anything). I'm guessing the cooling will have to come from the copper end cap.
Making Starship from stainless steel was just a scheme to develop battery casing technology :D
Tesla has developed something very special here and I get the feeling it’s going to put them even farther ahead of the competition than they already are.
TESLA uses SpaceX technology to get job Done.
the Model Y was Design with help from SpaceX.
@@markplott4820 no shit Sherlock!!!! Why do you always have to bring spacex as if they are the only ones developing alloys!? You realise that other companies also have alloys right!? From a lot of those materials have been developed/improved in racings for decades from f1 to lmp, super cars…… heck Porsche even had a racing car that had a tubular chassis from a thin alloy filled with gas to make their cars even lighter.
@@moineaux9173 - gm & FUD motors Aint got SHIT. MARY & JIM are both INCOMPETENT.
engineering is art
Thank you Jordon so much.
I love and appreciate your content, you are one of the reasons why the EV industry will boom because of your well-done content.
Thank you again and greetings from Egypt 🇪🇬
Sweet! Welcome!
So BYD got its hands on one 4680? Interesting ;)
Thanks for the great content, also cant wait for Monroes teardown!
I'm sure Tesla has a blade cell in a lab somewhere.....
It's "good business practice"
This'll become a future archeological find, the new Baghdad battery.
Thanks Jordan for making this available to everyone to see. Quite incredible feat of engineering. Looking forward to the analysis video.
I bet the "flower" cover is meant to hold the tabs in place on the assembly line until it's placed in the can and top put on
It ensures a uniform wide area contact with good pressure.
Would also prevent random movement of the tabs.
Flower is likely part of assembly process to keep tabs in place and also acting as a current collector bussbar
Oh yes this is the solid state battery that actually exist in large quantity.
The hell.. micrometer but measurements is in inches 🤯
Thanks Jordan. We may not know exactly how a 4680 looks like for a long long time if not for you and everyone else behind the scene.
Kudos!
Jordan it is absolutely incredible that you got this. Amazing work and big congratulations to you on the break!
If you ever get another one of the 4680s can you do some strength (mechanical) tests on it? See what loads it can take. Maybe compaire that to a 2170 or 18650.
Munro is selling “keep sakes” from his Model Y tear down for $800/ea. Maybe we could crowd fund one, so we’d have the latest configuration?
@@Crunch_dGH - MUNRO cells are sealed in Acryllic , unless you pick up in Person.
Tricky, since if you have an intact cell, you also have a live cell.
Load testing may lead to fire
Dang, 1 cell alone looks like it could take half a ton of force LOL, the shell seem to be built like a tank!!!
Another excellent video Jordan. Thanks for sharing!
Really the cell isn't tabless but rather tabfull. Thanks for showing the inside.
No, those are tabs, they're flags. There's a difference.
Now Tesla just needs to make 40 of these per second per factory.
Damn. This is awesome/ amazing. Thanks for sharing this with the public :)
Wow this was great. The UC San Diego people were very fun to watch and listen to.
Holy shit... This is Alien Tech.😳😳😳
2170 cells, Aluminum foil (cathode current collector) and copper foil (anode current collector) are about 7 microns thick each. The foil is coated on both sides, and the active material is about 30 microns thick per side. Very little 'free electrolyte', most is absorbed by the anode and cathode and the separator. To get electrolyte liquid out, you have to put the cell in a centrifuge and then at best you may get less than a milliliter out. Could anyone see if the inner 'flower' endcaps were laser welded?
Awesome video! Looking forward to your video on results and thoughts!
You understandably looked a little tired after 4 hours :D
I loved the can opener
Thanks Jordon! The team (& you) put together an amazing show!
Now we’re talking! Nicely done, Jordon.
Peeps - let the ads roll!! 👌
Jordan, I heard Elon say that the copper hole in the bottom is designed to vent any over pressure events downwards so it prevents thermal run away and it faces down to minimise over pressure blasting up into the cabin
Correct!
@@thelimitingfactor Geez apart from the basic way a battery works, they have redesign and engineered that 4680 into the 21st century. It's fascinating how they designed that pressure valve with the plastic and then that precision bit of alloy or something is a complete redesign of a battery
I just realised that the casing is probably so thick to contain silicone swelling when they load them up maybe
Very cool! Thanks for the work everyone!
Amazing video! Length is totally justified!
Just started watching. Right off the bat I'd be getting out my oscillating multi-tool with the steel cutting bit. 😆
lol! Yeah, popping it could have been smoother
Operator: “It’s not crystal, it is salt”.
Is salt crystal? Yes.
Me: “I wonder what they’ll use to disassemble it?”
Science: Glove box? Check. Garden shears?!? Check.
When I was a kid, I used to take a hammer to dead batteries to see what was inside. Messed around with the powder till I was bored. In retrospect, I've done so many things as a child that were incredibly dangerous. I learnt how to make pure chlorine gas in a jar but nearly poisoned myself when it burst in the middle of the night in my room. Fun times....
lol! Been there
This was freaking awsome Jorden!!! Thank you so very much!!!
Excellent video thanks 🙏
Did they check for continuity between the top button and the bottom plate? The "flower" disks are probably like durable springs to keep the tabs in contact with the collector plates and the anode/cathode terminals. The thin foil would likely lose its "springiness" after a few hot/cold cycles, leading it to lose contact. The thicker plates will retain that contact even after 1000s of cycles.
There was no connectivity between the copper flower as they called it and the bottom plate. That is, I don't think it performs a spring function. Particularly with copper that work hardens
Great idea through observation.
this is amazing! thank you so much for sharing
Astonishing. AND very much enjoying the sounds. Nice.
He said “It’s really smart”, well yeah! Tesla has the most brilliant minds working for them!
Great work, thanks
Stunning.
Thank you guys!
Very interessting. Thank you very much. 👍🏻