Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Is Tesla making progress in improving battery energy density?

This site may earn commission on affiliate links.
The new 2170 cell, which is now being produced at the Gigafactory, is slightly larger – 21 mm by 70 mm. More importantly, it can store a lot more energy. According to Elon Musk, it's “the highest energy density cell in the world, and also the cheapest.” The 2170 cell is around 50% larger by volumethan the 18650, but it can deliver almost double the current (the 18650 delivers 3,000 mA, and the 2170 has been tested at 5,750-6,000 mA).

Tesla 2170 Battery Cells: Greater Power At Comparable Cost

google will find many other similar articles quoting Elon or JB.
Why does the Model 3 not seem to express ANY of that increased energy density? 4416 cells which are 50% bigger than before and total around 80.5kWh. If you put that math in a calculator, it's worse density than a 18650 on sale for years.
It may well be the CHEAPEST celll on the market, as boasted many times by Tesla, but they go in a not-so-cheap car, and many are needed for the range it gets at the economy it's proven to have.

Anyone have links to tests of a Tesla 18650 from this decade performing as little as 3000 MAh (the correct unit as far as I know), let alone a 2170 that achieves 6000mAh, of any brand?

The highest density has been claimed, but Model 3 so far is proof to the contrary.
 
4416 cells which are 50% bigger than before and total around 80.5kWh. If you put that math in a calculator, it's worse density than a 18650 on sale for years.

Model 3 uses 2170 cells, not 4416.
http://ir.tesla.com/common/download...673CC&filename=TSLA_Update_Letter_2017-2Q.pdf

Edit: or are we having ambiguity issues due to 4,416 2170 cells?
Model 3 80

Actual capacity: 80.5 kWh (350V*230Ah= 80,500 Wh. Source for 350V and 230 Ah, see page 3 here)

Usable capacity: 78.27 kWh (Source: see page 6 footer here)

Cell count: 4,416 (source)
 
Last edited:
Why does the Model 3 not seem to express ANY of that increased energy density? 4416 cells which are 50% bigger than before and total around 80.5kWh. If you put that math in a calculator, it's worse density than a 18650 on sale for years.
It may well be the CHEAPEST celll on the market, as boasted many times by Tesla, but they go in a not-so-cheap car, and many are needed for the range it gets at the economy it's proven to have.

Anyone have links to tests of a Tesla 18650 from this decade performing as little as 3000 MAh (the correct unit as far as I know), let alone a 2170 that achieves 6000mAh, of any brand?

The highest density has been claimed, but Model 3 so far is proof to the contrary.

Original 18650 cell from 85D are 11.36 Wh Tear down of 85 kWh Tesla battery pack shows it could actually only be a 81 kWh pack [Updated]
If the cells are only 50% better then the 2170 it would be 17.04 Wh × 4,416 = 75,249 Wh. Off by about 10% from the 80.5 kWh pack. So they need to be 18.23 Wh each, an increase of 60% from 18650.
 
Original 18650 cell from 85D are 11.36 Wh Tear down of 85 kWh Tesla battery pack shows it could actually only be a 81 kWh pack [Updated]
If the cells are only 50% better then the 2170 it would be 17.04 Wh × 4,416 = 75,249 Wh. Off by about 10% from the 80.5 kWh pack. So they need to be 18.23 Wh each, an increase of 60% from 18650.
Fairer to compare to a more recent 90, 75 or 100 pack, don't you think? There was the 85->90 upgrade to consider, same number of cells, higher claimed capacity. Although that one infamously has since resulted in reduced charging speeds for many owners. Model 3 actually starts out charging a little bit slower than the 85 pack you reference. Pretty much same actual capacity of 80-81 kWh, 108 kW in stead of 120 kW peak charge oberved thusfar at least.

Teardown of new 100 kWh Tesla battery pack reveals new cooling system and 102 kWh capacity
102.4kWh / 8256 = 12.403 Wh per cell.

+47% in nominal size for 2170 = 18.235 Wh per cell.
4416 18.235 Wh = 80.51 kWh.
Increase in volumetric cell density: zero.

They could be a bit lighter than + 47%, of course, for a better gravimetric density. but we are talking mostly about volume as in Roadster, those cells need to be put somewhere and no car weight has been mentioned by Tesla.
 
Last edited:
Fairer to compare to a more recent 90, 75 or 100 pack, don't you think? There was the 85->90 upgrade to consider, same number of cells, higher claimed capacity. Although that one infamously has since resulted in reduced charging speeds for many owners. Model 3 actually starts out charging a little bit slower than the 85 pack you reference. Pretty much same actual capacity of 80-81 kWh, 108 kW in stead of 120 kW peak charge oberved thusfar at least.

Teardown of new 100 kWh Tesla battery pack reveals new cooling system and 102 kWh capacity
102.4kWh / 8256 = 12.403 Wh per cell.

+47% in nominal size for 2170 = 18.235 Wh per cell.
4416 18.235 Wh = 80.51 kWh.
Increase in volumetric cell density: zero.

They could be a bit lighter than + 47%, of course, for a better gravimetric density. but we are talking mostly about volume as in Roadster, those cells need to be put somewhere and no car weight has been mentioned by Tesla.

That all makes sense. I would expect the 3 to have the same formula as the latest 18650 cells, so in term of energy per volume, the line up is reasonable to me.
Looking back at the Elon quote, the following article was talking about discharge rate, not capacity. The new cells should be better in power per cell though due to increased active area. Perhaps he meant to say highest power density (per volume).
Big question I come away a from this with: has anyone done a 2170 tear down to see what the actual mass, wall dimensions, and packing density are? Or are we all just using exterior size? Big if, the cell has increased empty space, the chemistry could be an improvement.
 
  • Like
Reactions: transpondster
@Cloxxi
@mongo @Brando

In order to calculate the energy content of the 2170 cell you need to go to the detailes EPA report
http://www.teslarati.com/wp-content/uploads/2017/09/Tesla-Model-3-EPA-CSI-HTSLV00.0L13.pdf

you need "integrated amp hours" and average system voltage. The EPA also calculates the number. It is 78.3 kwh usable. That usable energy goes along with the longer tested range of 344 miles. Remember Tesla has derated the Model 3 range.
Tesla Model 3 actually has 334 miles of range according to EPA data

We also know the usable energy of the P100 pack and the cell count from Jason hughes.
Teardown of new 100 kWh Tesla battery pack reveals new cooling system and 102 kWh capacity

So we can compare the volumetric energy density of the 2170 vs the 18650 in the P100 pack.
Tesla's Model 3 2170 cells=same energy desity as 18650's

The volumetric energy density is basically the same.
So,I agree with Cloxxi.

The only catch could be if Tesla has held even more in reserve that they are not telling us about.
 
  • Like
Reactions: GSP and Cloxxki
That all makes sense. I would expect the 3 to have the same formula as the latest 18650 cells, so in term of energy per volume, the line up is reasonable to me.
Looking back at the Elon quote, the following article was talking about discharge rate, not capacity. The new cells should be better in power per cell though due to increased active area. Perhaps he meant to say highest power density (per volume).
Big question I come away a from this with: has anyone done a 2170 tear down to see what the actual mass, wall dimensions, and packing density are? Or are we all just using exterior size? Big if, the cell has increased empty space, the chemistry could be an improvement.
That's hindsight.
The larger formfactor promised to be more efficient, giving a higher volumetric energy density even with equal chemistry. No such thing has been observed.
JB and Elon were speaking about a 15% more energy dense chemistry, and this doesn't seem to have been implemented. Perhaps it was not ready for production, perhaps it lacked durability, perhaps it was too slow with charging, perhaps it broughtt up the kWh cost. Or a combination of factors.

You say bigger cells should offer more power, but this is contrary to what battery experts say. The smaller outer surface (per volume unit) makes heat dissipation slower, effectively reducing botht he discharge and charging rates. The "current" TM3LR isn't an all-out drag racer such as a ludicrous plus product, so we don't know how much can be drawn from the 4416 cells present. Numbers of 165kW and 200kW have been floated, both can be expected well below what will eventually be squeezed out in a PDL car powered by a similar size pack with (perhaps) identical cells. Peak charging rate observed seems to be 1.34 C, similar to the 75 and 90 cars (subject to factory imposed charge throttling?), but still slower than the original 85's 1.44 C.

Makes us wonder, WHY are the new cells not more dense? The new form factor might have brough 2-3 in itself, and you'd expect SOMETHING after 2-3 years of 5-8% annual incremental improvements. Yet after 5 years of Model S, Model 3 is just level with the first 90 cars for cell volume density. Let's hope they will not require charge throtting, tthere will be a lot of cars coming in for a charge after TM3 production really gets going.
 
That's hindsight.
The larger formfactor promised to be more efficient, giving a higher volumetric energy density even with equal chemistry. No such thing has been observed.
JB and Elon were speaking about a 15% more energy dense chemistry, and this doesn't seem to have been implemented. Perhaps it was not ready for production, perhaps it lacked durability, perhaps it was too slow with charging, perhaps it broughtt up the kWh cost. Or a combination of factors.

You say bigger cells should offer more power, but this is contrary to what battery experts say. The smaller outer surface (per volume unit) makes heat dissipation slower, effectively reducing botht he discharge and charging rates. The "current" TM3LR isn't an all-out drag racer such as a ludicrous plus product, so we don't know how much can be drawn from the 4416 cells present. Numbers of 165kW and 200kW have been floated, both can be expected well below what will eventually be squeezed out in a PDL car powered by a similar size pack with (perhaps) identical cells. Peak charging rate observed seems to be 1.34 C, similar to the 75 and 90 cars (subject to factory imposed charge throttling?), but still slower than the original 85's 1.44 C.

Makes us wonder, WHY are the new cells not more dense? The new form factor might have brough 2-3 in itself, and you'd expect SOMETHING after 2-3 years of 5-8% annual incremental improvements. Yet after 5 years of Model S, Model 3 is just level with the first 90 cars for cell volume density. Let's hope they will not require charge throtting, tthere will be a lot of cars coming in for a charge after TM3 production really gets going.

I thought we just showed that the 2170's line up volume to energy wise to the latest 100 cells.

I agree, one would expect some improvement. Although, one could also make the case that it is better to start with a known and proven chemistry, especially if it meets the requirements. Has anyone done a tear down of a 2170 to see what the internal utilization is?

My thought regarding power is that increase in surface area allows higher currents for the same unit area limit. I understand what you are saying regarding volume vs cooling area, but there is also the thermal impact of the number of layers heat travels through before reaching the other surface.

Pure speculation due to not finding any 2170 tear down:
Tesla may not be using the entire interior volume. Say the new cells have the same layer thickness, and 20% more energy density.
18650 cell -> subtract 1 mm for can walls 5mm for cap and bottom net 17mmx60mm interior space. Layer width of .1 mm (made up for relative comparison). Center pin diameter of 2 mm. Result: 150 layers with a total surface area of 270,000 mm^2.

Now with a 2170 -> same package loss so net 20mmx65mm, to achieve the same surface area/ lithium volume requires only 84 layers with a core size of 11.7 mm for the same chemistry. The additional 50% area for max fill requires another 96 layers. So the thermal path for those middle/inner layers is much worse (~181 total)
In the mix is 10% better, then it has 138 layers with a 6.3 internal diameter. Less layers and each layer has more surface area, so cooling is better than the original 18650.
If the mix is 20% better, then only 117 layers are needed with an 8.4mm inner diameter.

So maybe, they have a better chemistry, and are underpacking the cells with a a larger center pin to improve thermal performance.

Anyone want to sacrifice a 3 or Powerwall?
 
I thought we just showed that the 2170's line up volume to energy wise to the latest 100 cells.

I agree, one would expect some improvement. Although, one could also make the case that it is better to start with a known and proven chemistry, especially if it meets the requirements. Has anyone done a tear down of a 2170 to see what the internal utilization is?

My thought regarding power is that increase in surface area allows higher currents for the same unit area limit. I understand what you are saying regarding volume vs cooling area, but there is also the thermal impact of the number of layers heat travels through before reaching the other surface.

Pure speculation due to not finding any 2170 tear down:
Tesla may not be using the entire interior volume. Say the new cells have the same layer thickness, and 20% more energy density.
18650 cell -> subtract 1 mm for can walls 5mm for cap and bottom net 17mmx60mm interior space. Layer width of .1 mm (made up for relative comparison). Center pin diameter of 2 mm. Result: 150 layers with a total surface area of 270,000 mm^2.

Now with a 2170 -> same package loss so net 20mmx65mm, to achieve the same surface area/ lithium volume requires only 84 layers with a core size of 11.7 mm for the same chemistry. The additional 50% area for max fill requires another 96 layers. So the thermal path for those middle/inner layers is much worse (~181 total)
In the mix is 10% better, then it has 138 layers with a 6.3 internal diameter. Less layers and each layer has more surface area, so cooling is better than the original 18650.
If the mix is 20% better, then only 117 layers are needed with an 8.4mm inner diameter.

So maybe, they have a better chemistry, and are underpacking the cells with a a larger center pin to improve thermal performance.

Anyone want to sacrifice a 3 or Powerwall?
That would explain the lack of improvement in volumetric efficiency. It would also give hope that they have improved gravimetric.

Great hypothesis!
 
That would explain the lack of improvement in volumetric efficiency. It would also give hope that they have improved gravimetric.

Great hypothesis!

Or I'm cray cray;)...

If I were really going off the deep end, I'd suggest doing a deep draw to make a center feature of the cell case (cross section of a U) to pull heat even more... But that would be an assembly nightmare, and can be achieved with a top-hat center pin to pull heat to the bottom. Sewing bobbin style?
 
  • Like
Reactions: kbM3
@mongo
@Cloxxki

I had previously assumed that energy storage (kwh) in the cell would be a linear scale with volume. In other words, if you increase the volume of the cell 10% you will get 10% increase in kwh.

Maybe that isn't true. Maybe a small increment in volume added at the outer radius is worth more than that same incremental volume added at the inner radius of the cell.

Just a wild thought. Probably all wrong.
 
Has anyone even peaked inside a new Powerwall to see how many cells there are? Of course, different chemistry to the cars, although supposedly not different for the Semi.
If someone buys a new Powerwall and sells it for parts, surely there is a profit to be made? How much would fans pay for a 2170 cell? :)
@mongo
@Cloxxki

I had previously assumed that energy storage (kwh) in the cell would be a linear scale with volume. In other words, if you increase the volume of the cell 10% you will get 10% increase in kwh.

Maybe that isn't true. Maybe a small increment in volume added at the outer radius is worth more than that same incremental volume added at the inner radius of the cell.

Just a wild thought. Probably all wrong.
Slight, SLIGHT improvement to be expected from a bigger cell. In terms of volumetric density. Say, 50% bigger cell, 55% more energy.
Tesla explained this at length, without numbers.
Within the limits of their cell design and manufacturing equipment options, they chose a size that can be done without thickening up the cell casing. So, a lower percentage of the cell weight and volume will be metal. Overall, a liter of cells would lose some weight and pack more energy. In theory. Marginally. So far, we can't assume this has been successful, due to the Model 3 having so many of the big cells.
 
Slight, SLIGHT improvement to be expected from a bigger cell. In terms of volumetric density. Say, 50% bigger cell, 55% more energy.
Tesla explained this at length, without numbers.

Can you find where they said that?? I'd like to read it.

So far, we can't assume this has been successful,

As I said earlier, the energy density calcs I did that show the 2170's as the same volumetric energy density as the 18650's are all based on a usable energy that Tesla sets with software.

It's possible that there is more energy in these cells than Tesla is leading us to believe. It's competition sensitive so no need for them to show more than required.

Come the truck and the roadster and they just unlock the rest of the cells capability.
 
Or I'm cray cray;)...

If I were really going off the deep end, I'd suggest doing a deep draw to make a center feature of the cell case (cross section of a U) to pull heat even more... But that would be an assembly nightmare, and can be achieved with a top-hat center pin to pull heat to the bottom. Sewing bobbin style?
It does seem weird that volumetric density has not been improved. Maybe you are correct, or maybe the number of cells (4416) is actually incorrect. Tesla did a false flag operation to determine who is leaking data (the old catch a spy trick). Only half joking:)
 
It does seem weird that volumetric density has not been improved. Maybe you are correct, or maybe the number of cells (4416) is actually incorrect. Tesla did a false flag operation to determine who is leaking data (the old catch a spy trick). Only half joking:)
It only takes one person to open uit the pack to uncover the truth. Not a very clever long term strategy, obviously.
And after boasting best densitty on ttop of lowest cost, why make a product that could be cheaper or better still at zero change in cost? Tesla is well used to charging people for somethig they supplied extra (60/75 packs software restricted). Since they could be thought to keep 20% of energy from buyers, a 95-100 kWh Model 3 pack would in fact be possible, despite Elon claiming75 would be all that would fit. Guess how many customers would LOVE to pay extra for a 100 kWh Model 3 which was already built and sold to them?

Best guess I have, is that a yet cheaper way to make a decent cell was found, and it made the cut to production. So perhaps indeed they have more in store, but are just postponing it. First to put in S and X to keep margins up amidst general market cost and eventually consumer prices to drop. Say, they change the 100D's to 125's at the same price, but in fact the 125's cost them less than tthe 100's anyway. Model 3 was promised at this price, and reservation holders will buy it until Audi or Jaguar make something with a bettter value proposition. No need to discount those cars just yet. Long range cars, those will come under price pressure when the rest of the market gets involved around 2019.
Tesla will want to get cost of Model S and X down asap. So, when both cell production is up there and they've figured out how to let robots assemble the packs to save YET more cost. The $30K marginal price mentioned a good while ago (by Tesla) for a RWD Model S of course will have since crept down, but with 2170's, even of first gen, will be closer to $25K if not less. Remember when batteries were the most expensive part of the BEV? That may still be true, but its share in total car cost has been plummeting. Exhibit A, the Chevy Bolt 63 kWh pack is $15k retail. And car makers are not known to subsidize spare parts, let alone batteries which are in short supply. Every proper cell made on Earth, has a buyer or will spend vvery short time on a shelf. Not such thing as Li-Ion overstock in meaningful quantities. If the product is good and price fair, it's (pre-)sold.

Back to the Roadster. Since basic 5-year-old Tesla motors seem up for the job of powering this car using 3 of them, unless they want exceptional efficiency and/or long burst performance, batteries will be a tiny factor only towards total car cost. For sure by 2020, 2170's will cost Tesla under $20K for the whole 200 kWh car, 10% or less of sale price. As also in fully loaded Model 3's, by the way, $8K in cells towards a $80K car). Let alone Model S/X P100D if that's possible with 2170's (I'm not sure the same power can be drawn for the same amount of time), $10K in cells for a $150K car.
Batteries becoming such a relative afterthought in cost management, if by 2020 Tesla can acquire some batteries of higher gravimetric energy density (say, double), at triple the cost, that'd be TOTALLY worth it. Absolutely no need for bottom of the barrel ultra cheap ass 2170's to make up the 200 kWh.
 
Moderator note: All the posts in this thread before this one were moved from a thread in the 2020 Roadster forum that was a dicussion of how Tesla was managing to get 200kWh of energy in the new Roadster battery pack. The thread discussion veered off into a much more general discussion of Tesla battery energy density and that is why these posts were moved to this forum in their own thread.
 
  • Like
Reactions: GSP and Cloxxki
To use two complete S/X packs isn't possible given the geometry of the car and packs - for one thing, remember the 100 is a 16 module pack - with a double height section up front.

I wouldn't be at all surprised if Tesla used a double height design and existing modules to power the prototype, but they'd need a new pack structure if nothing else.

I keep seeing the reference about double height design. remember that roadster 1.0 did not have batteries underneath the passenger section, so comparisons to the floor height of 1.0 are probably not valid.

https://www.tesla.com/sites/default..._2.5owners_manual_-_web_version_-_ja_en_0.pdf
 
  • Like
Reactions: Cloxxki
I keep seeing the reference about double height design. remember that roadster 1.0 did not have batteries underneath the passenger section, so comparisons to the floor height of 1.0 are probably not valid.

https://www.tesla.com/sites/default..._2.5owners_manual_-_web_version_-_ja_en_0.pdf

This is true. However, I thought folks were comparing it to their experience with the S and 3, instead.

My point in the quoted post was simply that the form of the S pack wouldn't allow simple stacking.
 
  • Like
Reactions: Cloxxki