EEVblog #1340 - New Tesla 4680 Battery Cell EXPLAINED

[EEVblog]

Dave explains why Tesla have switched from a 2170 cell to a bigger 4680 cell announced at Battery Day. What is the new tabless technology and what are the thermal cell and battery pack implications?
Discussion on Lithium Ion battery cell construction, chemistry, manufacture, thermal design, internal resistance, heat sinking, and how cylindrical cells compare to large pouch cell construction.
Tesla Patent: https://patents.google.com/patent/US20200144676A1/en?oq=20200144676

[Schwuuuuup]

Maybe I'm overseeing a very simple thing, but stacking smaller cylinders or stacking bigger ones of same length, should result in the same density... or the at least in the same ratio of battery vs. void in between: Big cylinders -> big voids, but just a few, small cylinders -> small voids, but many of them.

I guess that one benefit of the bigger batteries is that the inner core can be the same small size. There must be a limit how small you can make the innermost layer. So making a bigger roll, means you get pure additional usable volume.

On the other hand.. using big cylinders with therefore bigger voids, you could fill this voids with smaller cylinders, thus archiving a better battery to void ratio ... but I don't think that this is practical.

[Bud]

@3:18

"Longer and thicker is better...Obviously!"

 

[rs20]

I agree with Schwump, the volumetric efficiency of packing cylinders is always 90.69%, regardless of dimensions. So not sure what's being referring to by "improved volumetric efficiency"??

[bristpi176]

I just did some 8th grade science on the subject of packing density with a compass and paper.  If the new cells are packed squarely an 18650 or maybe even a 2170 will fit in the space in between.  I know nothing about connecting them together with the larger cell and assume the two sizes would have to be separate circuits, but your ultimate density would be greater than any single size cell.

[wraper]

The reason Tesla uses cylindrical cells is because they are easier and cheaper to mass produce. Companies using prismatic or pouch cells usually make no profit or even sell their cars at loss. While tesla makes about 25% gross profit margin. By increasing battery size they can make them faster and cheaper. Old technology did not allow to increase battery size due to inability to transfer the heat away.
As already mentioned in other eevblog thread and youtube comments there are many other improvements besides the form factor.
https://www.eevblog.com/forum/renewable-energy/why-tesla-cars-are-the-best-evs-and-tesla-is-the-best-car-company/msg3246970/#msg3246970

[wraper]

I just did some 8th grade science on the subject of packing density with a compass and paper.  If the new cells are packed squarely an 18650 or maybe even a 2170 will fit in the space in between.  I know nothing about connecting them together with the larger cell and assume the two sizes would have to be separate circuits, but your ultimate density would be greater than any single size cell.

They don't pack squarely. And using different cells is a big no from manufacturing standpoint.

[tszaboo]

The wall thickness is about 0.15mm. On the top of the cell,there is about 3-4mm lost for PTC, tab connection, vent holes, insulator, cathode cover. The internal 1-1.5mm is lost, because you cannot roll it up. Back of the napkin calculations:
A=9*9*65*3.14 #volume
B=2*9*9*3.14*0.15+9*9*3.14*4 # lost on top and bottom
C=65*3.14*18*0.15 # lost on sides
D = 1.5*1.5*65 # lost in the middle
E=(A-B-C-D)/A # total efficiency
E = 89%
The 18650 cell has about 89% of the volume left for the guts.
The same would be 93% for the, if the top part would stay the same. I expect, it isnt. I really think Tesla might have done away with the PTC. That would place us at 98% of the cell volume. That is still only like 10% improvement. The rest could come from the smaller ESR losses, or they use thinner material.
Here is a link for a 73430 battery, so everyone can chill out, and see how much improvement was made:
https://nl.aliexpress.com/i/32919620579.html
The investment/GWH reduction graph doesnt really make sense. Somehow the cathode material made the factory cheaper?
They showed the inside of the factory for some 10s clip, that was indeed impressive. It is around 1:27:00 in their video. I've designed machines like this. We had individual machines, that are hundreds of KGs, but you could walk in between them, tough they were automated.  I dont want to go into details, probably I would break my NDA. They made some multi stock monstrosity.

But then solid state lithium comes along, integrates the anode and cathode into one sheet, makes ie 48V packs with only two terminals, and makes everything else obsolete.

[bristpi176]

My thought was that if your goal is storage density and the space between cells could be better used with a smaller cell than a staggered configuration, given the life span and cost of these, would the added complexity of manufacturing be worth it?  It's not like it can't be done. 

[wraper]

But then solid state lithium comes along, integrates the anode and cathode into one sheet, makes ie 48V packs with only two terminals, and makes everything else obsolete.

The question is can you affordably mass produce it and if technology is anywhere near to be ready for mass production to begin with.

[pickle9000]

Don't forget about Tesla integrating the battery holder into the frame to lower the overall weight of the vehicle. Marketing may imply things like power density but the overall design of the vehicle counts, cooling included.

[Doctorandus_P]

You forgot to count squares.
Assume 45mm height for the guts, that's  800/45 = 17.78 squares.
So the resistance goes from 17.8 squares in series to the same number parallel, which is a factor of 316, which is much more then "an order of magnitude".

I'd also find it quite sad if something like this is patent-able. Capacitors have been made this way for some 30+ years.
But these days you can get a patent for any fart in a plastic bag, and if you need to fight or defend it, then the lawyers stand in line waiting for you to fill their wallets.
Fair? No. Profitable? Yes, Yes, Yes...

I do wonder if any use can be made by making the cells pressure resistant, which is much easier with the round cells then with the flat packs.
Since the can is only open on one side maybe something can be done with venting in a safe direction, or when a faulty cell blows, it ejects itself from the battery pack.
On the other hand. Li-Ion chemistry seems to be pretty well under control, and most cells that die are due to gross abuse, such as the deliberate overcharge video's on youtube.

[wraper]

I'd also find it quite sad if something like this is patent-able. Capacitors have been made this way for some 30+ years.

They were not. It's a completely different process. There is no laser cutting or folding while producing capacitors.

[Poe]

Can't believe Telsa's marketing is so effective that people are talking about such marginal process improvements. 

[wraper]

Can't believe Telsa's marketing is so effective that people are talking about such marginal process improvements.

It's not a marginal improvement. It's being able to produce battery with 5x higher capacity and better thermals. There were no larger cylindrical cells because it wasn't possible to cool them. Not to say way reduced machinery and improved manufacturing speed.

[sibeen]

Is the tabless design actually used for low ESR electrolytic capacitors?

[wraper]

Is the tabless design actually used for low ESR electrolytic capacitors?

No, they still have usual tabs.

[Fungus]

My thought was that if your goal is storage density and the space between cells could be better used with a smaller cell than a staggered configuration, given the life span and cost of these, would the added complexity of manufacturing be worth it?  It's not like it can't be done.

That's only if you pack them in a square grid. In reality you'd pack them in a hex grid.

I'd have thought the biggest advantage of this "tabless" design would be better thermals for rapid charging and discharging.

(I believe their current cars have to wait around and cool off a bit if you do too many launches)

[sandalcandal]

People seem to be confused with the meaning of the improvements and their implications. I agree with everything wraper has said.

Cylinderical cells far outperform prismatic cells with the only slight disadvantage of increased costs in pack integration due to the larger number required to make up an equivalent capacity; power density as well as weight and volumetric energy efficiency has always been better using cylindrical cells because the jelly roll internal structure is most efficient in a hard cylindrical casing. Cylindrical cells have previously not been able to maintain the performance required for high power applications with larger sizes due to thermal and ESR limitations. The tab less design implemented by Tesla solves these size limiting problems elegantly and practically. The tab less design should also see general improvement in battery life and max charge/discharge speed due to: the lowered ESR, improved thermal homogeneity and improved current density homogeneity.

The implications of a larger cylindrical cell (ignoring the tabless implementation) are improved manufacturing efficiency and improved energy density + specific energy. Manufacturing efficiency is improved due to the lower number of cells which must be managed and integrated for the battery pack manufacturing process. Slight improvements in energy density and specific energy are due to the fact internal void (at the centre of the jelly roll) and casing thickness remains roughly equivalent with increased size/diameter thus the ratio of active jelly roll material vs inactive void + casing material is improved. Again, neither of these benefits can be realised by simply increasing the size of a conventional cylindrical cell due to typical trade off size increases bring with ESR and thermal limitations worsening with increased size. For the previous conventional cell construction, the optimal trade off for size vs thermal + ESR limitations was the 2170 in most high power applications but it appears the optimal point for Tesla's application is now a 4680 size using tabless construction.

What the larger form factor + tabless design DOES NOT MEAN is improvements in packing efficiency of the cylinders themselves, as others have pointed out this is a geometric constant for any single cylinder size. Some people speculate smaller cells could be packed in interstitial spaces between larger cells but this is not going to happen because A. Tesla shows their planned battery pack construction using only a single cell size and B its a bad idea. Mixing cell sizes hugely complicates the manufacturing process due to the need to now manage another entire line feeding and integrating the second cell size. Safety and performance is also compromised by the second cell which will either need some heavy design and control to match performance characteristics of ESR and thermal of the larger cells or otherwise cause internal mismatching issues. Perhaps in the far future where competition is strong and the absolute limits of all other improvement avenues have been exhausted will multi cell-size packs be a sensible idea for the challenges it introduces.

I've honestly wondered why no one else has done tabless/continuous tab cells before Tesla's announcement myself. I have heard hearsay that other people have also been working on continous tab designs before Tesla. My guess is that the impetus has not been present or sufficient to surplant the conventional design but here Tesla and Elon Musk is again pushing things towards their fundamental first principal limits.

[Per Hansson]

Is the tabless design actually used for low ESR electrolytic capacitors?

No, they still have usual tabs.

Also remember that radial caps used on all modern PCB's have the tabs on the same side of the capacitor.
You would need an axial capacitor to be able to make it tabless, but it sure is an interesting thought why they have not been made?
I don't see any reason we could not just take the same calculations and end up with a capacitor with an order of magnitude lower ESR!

[sandalcandal]

You would need an axial capacitor to be able to make it tabless, but it sure is an interesting thought why they have not been made?
I don't see any reason we could not just take the same calculations and end up with a capacitor with an order of magnitude lower ESR!

I believe in the case of most (older) electrolytic capacitors, the ESR is limited by charge carrier mobility in the electrolyte used to convey the electric field into intimate contact with the electrodes. Some newer solid/polymer caps seem to be pushing past this limit towards the ESR of MLCC and foil caps which do use full endcap type connection so I wonder what these advanced electrolytic capacitors use?

Edit: Aside, while I was reviewing my knowledge of electrolytic caps I came across this https://en.wikipedia.org/wiki/Electrolytic_capacitor#Water_based_electrolytes

A stolen recipe for such a water-based electrolyte, in which important stabilizing substances[47][48] were absent,[49] led in the years 1999 through at least 2010 to the widespread problem of "bad caps" (failing electrolytic capacitors), leaking or occasionally bursting in computers, power supplies, and other electronic equipment, which became known as the "capacitor plague". In these e-caps the water reacts quite aggressively with aluminum, accompanied by strong heat and gas development in the capacitor, resulting in premature equipment failure—and development of a cottage repair industry.[21]

Apparently capacitor plague was due to a bad stolen recipe? That was interesting and new to me.

[Siwastaja]

Note that classically cylindrical cells dissipate heat through the cylinder walls, not the ends, because there are voids at both ends. 18650 cools very well because it's only 18mm in diameter; heat flow needs to travel only a few mm on average; surface-to-volume ratio is good; hence this was the choice for Tesla. 25650 was available, but the cooling was inadequate. Tesla's first micro-optimization was to find the sweet spot where the cooling is still good enough, but casing and assembly costs are reduced, and that was just at 21mm diameter, between standard 18650 and 25650 size. All this makes sense and, as an anecdote, I actually happened to predict their exact 21700 cell size before they published it.

But this, I couldn't predict. It's funny how trivial idea this is, yet it never crossed my mind!

The new idea is to change the primary cooling path from the cylinder walls, to the endcaps, by removing the void and replacing it with solid connection to the electrodes. This is actually quite a big fundamental change. As a result, the diameter can be arbitrary; increasing the d increases the cooling area at the endcaps. I suppose, this means a big design change in their liquid cooling system, as well.

Much of the Teslas original choice to go for cylindrical cells was their off-the-shelf availability, superior cost ($/kWh) and superior energy density (Wh/kg) back when they were designing the Roadster. They had 200Wh/kg cells available from the big manufacturers like Panasonic while the other EV pioneers were struggling with in-house custom-specified pouch or prismatic cells, using new weird chemistries, somewhere around 90 - 150 Wh/kg depending on case. Tesla's initial success is because of all the development seen in laptops at the time; they used predominantly 18650 cells. BTW, this market situation was still similar while designing Model S.

The history shows their choice of cylindrical cell was clearly a correct one, and it seems that even when the market situation has changed so that the original main driving force, using a superior COTS product, has become completely irrelevant as they can design anything they want now, they are still not going to abandon it; clearly cylindrical cells work well.

And, finally,

Can't believe Telsa's marketing is so effective that people are talking about such marginal process improvements. 

Claimed 15% energy density gain is far from marginal. It's one level up from micro-optimization; and even micro-optimization makes sense at such scales. And they didn't even talk about the actual numbers in process cost savings.

The optimizations need to be done everywhere. As a result of dozens of different small improvements, li-ion cells are approaching 300Wh/kg as we speak, being some 100Wh/kg originally when introduced in late 90's. There is no single big breakthrough in the history.


PTCs were mentioned earlier, note that since Model S, Tesla has custom-ordered their Panasonic cells without the PTC, replacing it with their fusible link wires instead. This is likely a significant cost saving, and also improves performance.

[Fungus]

But this, I couldn't predict. It's funny how trivial idea this is, yet it never crossed my mind!

https://en.wikipedia.org/wiki/Egg_of_Columbus

[Fungus]

Can't believe Telsa's marketing is so effective that people are talking about such marginal process improvements.

I bet one of these can be manufactured for approx the same cost as a standard 18650 but Tesla will need to manufacture 5x less of them per car. That's not trivial.

I also assume that this construction could have a massive impact on charge times. If a supercharger can charge your car 6x faster than before then it's a huge win for Tesla. If their power-grid batteries need an order of magnitude less cells and electronics to control them then it could have an enormous impact, too.

[Siwastaja]

Obviously it's not going to be a 6x improvement in power density. This is marketing; maybe it is 6x improvement in power per cell, but since only one fifth of the cells are used, the actual improvement in total power is 6/5x. Which still is a remarkable optimization.

But charge current especially isn't limited by the metal resistance, it's limited by the anode lithium intercalation reaction rate (that's why charge and discharge rates are asymmetric; anode can't accept lithium ions as easily as cathode can).

Dave's video is a very good explanation about how the resistance of the metal sheet goes down by an order of magnitude, but the total cell ESR isn't dominated by the metal resistance; it's the ion transfer itself. Dave completely ignores this biggest part of the ESR, as does Tesla in their presentation. This is OK, they are deliberately talking about one aspect in separation from the complete picture. But be careful this isn't about the cell current capability or cell resistance, only about the metal sheet capability.

Thermally, it's really a Big Deal. ESR wise, it's a good small optimization among others but not the game changer. Capacity wise, it may be actually hurting their energy density; they have more copper and aluminium adding weight. At their 2C charge and ~3C peak discharge rates though, reducing losses in the bonding may pay pack for the "lost" energy density. This would be different if these were non-performance cars with just huge range and slow charging. OTOH, their design of laser-cutting or stamping the sheets allows precise optimization; they can calculate the optimum amount of metal to be cut out. The photo shown shows quite a lot of copper, with just small cutouts that allow folding. I'm sure they will fine-tune the amount of removed metal to maximize the product-level energy density where the efficiency is also a factor.


Say, a typical energy 18650 cell has DC ESR of approx. 50mOhms at room temperature. My guesstimate is that maybe 5-10mOhms comes from the PTC (which Tesla got rid of for good reasons IMHO) and some 5-10mohms from the metal, but rest is chemistry. The fact that increasing temperature significantly reduces the DC ESR, even halving it down to some 25mOhms, proves this; the resistance contributed by the PTC and the metal is increased slighly, so the chemistry-related ESR must go down even more.