I just was coming over here to see if this had been posted. ;-)

Interesting that Fred's source is "someone familiar with the program" - and not a sighting. (Not that I dispute it - he usually has good info and appears to be a trusted source for Tesla "leaks.") So I wonder if this is being tested concealed in a different body. Or, if it's only being tested on their track until after the unveiling.

mntlvr23 |
8 febbraio 2017

Excellent - good to hear about a mule and about some great range. The trickle of news is starting to pick up a little between this, and the battery max size, and the supercharger status, etc. I am excited that it likely will be speeding up more in the near future with the meeting in two weeks and then the unveil perhaps within the next month after that - and hopefully a slew of information to whet our chops.

Efontana |
8 febbraio 2017

Is there a minimum range for supercharger effectiveness - drive the supported routes?

Hmmm... Looks like the Model 3 would have to be extremely efficient to manage 300+ miles with only a 70 kWh battery pack. Unless someone decided to artificially limit Performance to a 60 MPH top speed. I think that rather unlikely.

SamO |
9 febbraio 2017

But I thought Model 3 would only offer 215 miles of range? Sad.

Efontana |
9 febbraio 2017

I wonder if the front motor is the same technology as the rear motor.

It might make sense to make the front super efficient for regen and range and use what works for the power.

Efontana |
9 febbraio 2017

I wonder if the front motor is the same technology as the rear motor.

It might make sense to make the front super efficient for regen and range and use what works for the power.

CoreDump |
9 febbraio 2017

I think the thing that everyone doing estimates is missing is that the M3 battery pack will be made up of 2170 cells which are almost double the energy density, meaning double the KWh per pound. If you look at the difference between older and newer Powerpacks, they are claiming around 2x the energy density from the 2170 cells.

Smaller car + lighter 70KWh pack and you could imagine 300 miles. The current thinking is that the base model will have something like a 50-55KWh pack, which @55KWh, you would be at 215miles or range with 255Wh/Mile. 255Wh/Mile doesn't seem to be a huge stretch for a smaller and lighter car/pack. At 255Wh/Mile it would be 294 miles for a 75KWh pack. If the 215 mile range pack is actually a 50KWh pack, then you are at 233Wh/Mile which would put a 70KWh pack at just a hair over 300 miles of range.

I think that people are under estimating how much lighter the vehicle could be made with smaller wheels, lighter packs and everything about the car being lighter.

The Bolt is at 252Wh/mile with 60KWh pack at ~3500lbs. 60KWh MS is is around 1000lbs heavier and averages under 300Wh/Mile? Makes me think that we will see a 300 mile Model 3 if the pack really is that much higher density, you could shave 600lbs off the Model S with 2170 cells alone. The Model 3 could actually be the same weight as the Bolt.

I am sure we will find out soon enough.

andy.connor.e |
9 febbraio 2017

@SamO

That was the estimated base range, whos range was estimated based on the March 31 2016 prototype. We do not know what the base pack size will be, nor do we know what "upgrade level" the 70kWh is from the base. Model 3 base range estimate could still very well be ~215 miles. 70kWh could be a significant size upgrade.

andy.connor.e |
9 febbraio 2017

@CoreDump

Dont forget that the M3 will be more aerodynamic than Bolt (less air resistance). Its important to understand that Tesla's future depends on the Model 3 success, whereas GM could care less if the car is successful. That alone determines the amount of effort put into the car to make it the "best".

So you can compare 2 apples one grown with pesticides that received no care, and one grown with no pesticides and was constantly weeded and monitored. Consumers will decide which product is successful.

Frank99 |
9 febbraio 2017

CoreDump -
I don't think the 2170's are twice the KWH per pound. The chemistry likely remains similar, so the ingredients are similar, so the batteries are likely to give about the same energy per pound - maybe 10% more due to more refined chemistry.
I think the 2170's biggest benefits are:
1. Dollars per KWH - with fewer cells to create, package, and interconnect for a given battery size, packs will be cheaper to produce.
2. KWH per cubic meter - the cells are likely more volumetrically efficient, meaning for a given amount of space you'll be able to stuff more kwh into it.

Bighorn |
9 febbraio 2017

No way there's a 100% increase in energy density with the 2170. Not remotely close.

Bubba2000 |
9 febbraio 2017

Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860? I know there is packaging efficiency involved. Significant chemistry improvement?

dd.micsol |
9 febbraio 2017

with a tailwind and going 25mph sure it can go 300 miles. Just making a point.
Distance is relevant to wind resistance and energy output.

dsvick |
9 febbraio 2017

"Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860?"

They can't, at least not as has been stated by them or tested by anyone.

Bighorn |
9 febbraio 2017

@Bubba
Step one: Smoke some crack.

JeffreyR |
9 febbraio 2017

The big boost in energy density is at the pack level not the cell level. Not sure it reaches the 2x claim, but I think that's what people are thinking. Basically Tesla has optimized the cell (instead of relying on 18650 popularity which made perfect sense for Roadster) and w/ the new cooling approach in the '100' pack then you get better pack power density.

- Model 3 goal for Cd was low twenties (IIRC .21)
- Model 3 goal is to be 20% lighter than Model S
- Model 3 will have new inverters and motors (better efficiency)
- Model 3 will use 21700 cells for much better pack energy density

I think Elon used a simple formula for estimating range:

- Expected cost of pack sets expected size (total kWh)
- Projected efficiency sets consumption (Wh/mile)
- Do the math to get projected range (not true EPA 5-cycle)

Other Considerations
- Gigafactory production estimates were way off (3x original now)
- Demand was through the roof (600K+ reservations)
- The Great Ramp Up is coming (convert those reservations as fast as you can)
- GM's Bolt hit 238 miles w/ 60 kWh battery

An upgraded M3 battery pack + AWD will likely reach a 300-mile range. A base M3 will almost certainly exceed the '215' Elon quoted at the 1st Unveiling.

Haggy |
9 febbraio 2017

"But I thought Model 3 would only offer 215 miles of range?"

Tesla never said that. They didn't even say that the base range would be 215 miles. They said it wouldn't be less than that. I can't see them going far above it for the base range, but for marketing reasons they might want to continue rounding pack capacity units to increments of 5. If it takes a 52 kWh pack to give 215, then they might make the base range 227 miles and start with a 55, for example. So although the base might be above 215, it won't likely be above 230.

Tesla did say there would be battery options though.

Frank99 |
9 febbraio 2017

At this very moment, the Tesla model 3 page states "Model 3 achieves 215 miles of range per charge". I think that's got to be the base understanding we work from right now.

Bluesday Afternoon |
9 febbraio 2017

@Haggy

The base Model 3 will most likely be over 238 mile range of the BOLT. I'd be shocked if it was less and lose the base range war to those folks. One great advantage Tesla has with its customers is the option for a larger battery and increased range.

Red Sage ca us |
9 febbraio 2017

CoreDump: Please note that there is the observed Wh per mile that EV owners see, and there is the Wh per mile recorded during EPA testing. Those are definitely different things. Also, it is likely some percentage of a battery pack capacity -- perhaps 5% to 10% -- is held 'in reserve' so that a 55 kWh battery pack capacity has a lower amount as a useable quantity.

Where Tesla owners may report as low as 280 Wh/mile, EPA ratings might be as high as 380 Wh/mile for the same car. And it is the EPA rating that the final official determination of range is made. Better to not presume EPA testing results will produce best case results.

55,000 Wh × 0.9 = 49,500 Wh

49,500 Wh ÷ 300 Wh/mile = 165 miles range
49,500 Wh ÷ 240 Wh/mile = 206.25 miles range

49,500 Wh ÷ 215 miles = 230.23255813 Wh/mile
49,500 Wh ÷ 230 miles = 215.2173913 Wh/mile

Badbot |
9 febbraio 2017

Original power packs were 100 Kw using the 18650
Todays power packs are 205 Kw using the 2170.
can anyone math the differential ?

Bighorn |
9 febbraio 2017

"These batteries are steadily improving every single year – maybe around 5% improvement in their energy density their ability to store energy in a given amount of mass. That’s probably one of the key metrics we worry about. And when we went from the Roadster to the Model S, they have improved by about 40% and when we were designing the Model 3, they were about another 30% better. That improvement just continues on every single year in the background.” JB Straubel 10/18/16

bj |
9 febbraio 2017

@Haggy - I think you missed the joke, admittedly it was subtle. SamO was channeling PigeonPDX, and added a Trump-ism.

Red Sage ca us |
9 febbraio 2017

Bighorn: Thank you for the quote.

bj: Thank you for the input. I thought I spied that correctly as a remark aimed at the [IGNORED].

TASANB |
10 febbraio 2017

@Bubba: "Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860?"

The 2170 (21mm diameter x 70mm long) has 76% more volume than the 18650 (18mm diameter x 65mm long).

Assuming both cells have similar wall thickness there's significantly more room inside the 2170, if not quite twice the available volume.

If it's slightly more efficient at storing energy (I'm not claiming it is or is not without more details), it would reasonable to assume it's has twice the energy capacity. Even if it's not more efficient, a 76% increase could be called "about double" the capacity.

Bighorn |
10 febbraio 2017

It seems the definition of density eludes some folks.

dsvick |
10 febbraio 2017

@TASANB - "Even if it's not more efficient, a 76% increase could be called "about double" the capacity."

Capacity is not energy density. Capacity simply means it holds more if it's bigger. Improved density means that it will hold more for the same volume, that is the " more efficient at storing energy " part of what you said.

TASANB |
10 febbraio 2017

I wasn't aware that any specs for the energy density were released. All I've seen was a statement from EM saying it's the "highest energy density cell in the world"

I guess I should have specified that there might be some terminology confusion here too. Even without significantly more energy density (energy per unit volume or mass), the 2170 could very well have twice the energy storage capacity (energy per unit)

Bighorn |
10 febbraio 2017

@TASANB
Show your volume calculation because I get 47%

dsvick |
10 febbraio 2017

@SimplyRed - "The base Model 3 will most likely be over 238 mile range of the BOLT. I'd be shocked if it was less and lose the base range war to those folks."

I think I have to go with Haggy on this one but for a different reason. Range is still a big consideration for people considering BEVs, if Tesla wants to get more and more people into them then I think they need to de-emphasize range as being a factor for the vast majority of people. For those that really do need greater range they have, as you pointed out, the ability to upgrade to a larger battery.

Keep in mind also what you're getting with the Model 3 as compared to the Bolt. If Tesla can keep the base costs down and include more bells and whistles I'm all for it.

TASANB |
10 febbraio 2017

volume of a cylinder: r^2 x L x pi

18650: 9^2 x 65 x pi = 16540 mm^3
2170: 11.5^2 x 70 x pi = 29083 mm^3

29083/16540 = 1.758 or 176% or 76% increase

TASANB |
10 febbraio 2017

just realized my mistake...
10.5 not 11.5

Bighorn |
10 febbraio 2017

And...

TASANB |
10 febbraio 2017

Your calculations are correct.

The point I was trying to make (and not doing a very good job) was that energy density doesn't have to be doubled for there to be twice the energy in a 2170 v 18650 cell. That distinction (density v. capacity) might be a source of confusion.

KP in NPT |
10 febbraio 2017

Bighorn has spoken. ;-)

topher |
10 febbraio 2017

"energy density doesn't have to be doubled for there to be twice the energy in a 2170 v 18650 cell."

But twice the energy doesn't help if you can only fit half the number of batteries into a pack.

The energy density (volumetric) is exactly the metric you want. if you assume that packing density will be analogous from small to large cells. Energy density of cells * packing density * volume equals total energy. Packing cylinders into rectangles is a well studied problem, and in unlikely to present an significant improvements. So without an improvement in (battery) energy density, the total energy is about the same for a given volume.

Tesla has said that the increase in size of the cells gives a increase in energy density (absent any chemistry improvements), and that chemistry improvements are being made at about 5% per year. I heard a _unconfirmed_ report that the 2170 will deliver 50% improvement over gen II batteries (which would be about 10% per year).

Thank you kindly.

topher |
10 febbraio 2017

An aside: 2170s are taller, so conversion from volume of pack from a Model S to a Model 3, will need to account for a bit more volume per floor area (which is the real limitation).

Thank you kindly.

andy.connor.e |
10 febbraio 2017

Remember that volume is Cubed. (^3) So 2x energy density is not half the volume.

topher |
10 febbraio 2017

Yes it is.
Energy density (volumetric) * volume = total energy.
2* energy density (volumetric) * 1/2 * volume = same total energy.

If you are measuring energy density by weight, it might be different, if the density (ρ) varies. But if it is the same, the above still holds.

Thank you kindly.

Bluesday Afternoon |
10 febbraio 2017

@Bighorn

"It seems the definition of density eludes some folks"

I thinks that's an insult! But I may be too dense to know for sure. :)

Red Sage ca us |
11 febbraio 2017

I haven't checked the math due to laziness, but others have calculated the volume of the new 2170 battery cells as having ~146.667% the volume of 18650 battery cells. JB Straubel has on multiple occasions noted that he expects at least a 30% improvement in energy density for Generation III vehicles in 2017 over what was possible with Generation II in 2012. This tells me to expect the Model ≡ to be capable of storing perhaps ~191% of the total capacity for a given volume compared to Model S in 2012. That is just shy of 200%, or 'double the capacity', but is close enough for polite conversation and speculative projections, I think.

So, if one presumes the battery pack of a 2017 Model ≡ will have 80% the volume of a 2012 Model S...
85 kWh × 0.80 = 68 kWh
68 kWh × 1.4666~ = 99.7333288 kWh
99.7333288 kWh × 1.3 = 129.65332744 kWh
129.65332744 ÷ 68 = 1.90666658

Even if you forget the volume and only go by the increase in energy density...
68 kWh × 1.3 = 88.4 kWh

So the battery pack for Model ≡ will likely have an initial maximum capacity somewhere between 85 kWh and 100 kWh. I fully expect that to be improved to as much as 135 kWh within five years from launch.

Bighorn |
11 febbraio 2017

@RedSage
I didn't check your math either, but you don't seem to have accounted for the fact that the bigger cells take up more space so fewer will fit in the battery pack.

topher |
11 febbraio 2017

~146.667% the volume
~191% of the total capacity for a given volume

You can't increase the volume and at the same time add that increase to volume independent calculations.

A 146% increase in volume gives exactly zero (0) increase in capacity for a given volume.

85 kWh x 0.80 = 68 kWh
68 kWh x 1.3 = 88.4 kWh
88.4 kWh / 84 kWh = 1.04
so a 4% increase, not 100%.

Thank you kindly.

Bighorn |
11 febbraio 2017

@topher
Did you mean volume dependent calculations? Packing efficiency may provide for increased capacity for a given volume, but I haven't seen that math. Easy to calculate cylinders filling rectangles, but there's some tweaking to be had sizing the box.

topher |
11 febbraio 2017

No, I meant volume independent calculations. Ones that include phrases like "for a given volume".

To use a simpler example: density is a volume independent calculation. No increase in volume will change the density. Density is mass 'for a given volume'.

Packing cylinders (actually circles suffice) is (approaching) 1/6 * pi * sqrt(3) ~= 0.9068996821
I think we can assume that Tesla took advantage of any small optimizations due to the precise ratio of radius, to length and width.

Thank you kindly.

Bighorn |
11 febbraio 2017

Okay, but it seems that the crux of the issue was how many of the 2170s would fit in the volume of the current battery pack, thereby making it volume-dependent. That was my point with @RS, that the number of cells did not remain the same.

Bluesday Afternoon| 8 febbraio 2017https://electrek.co/2017/02/08/tesla-model-3-test-mule-battery-pack-dual...

roadkill| 8 febbraio 2017that would be freaking awesome!! I'm in!!!!!!!!

akgolf| 8 febbraio 2017Can't wait.......must wait! Arrrgh!

KP in NPT| 8 febbraio 2017I just was coming over here to see if this had been posted. ;-)

Interesting that Fred's source is "someone familiar with the program" - and not a sighting. (Not that I dispute it - he usually has good info and appears to be a trusted source for Tesla "leaks.") So I wonder if this is being tested concealed in a different body. Or, if it's only being tested on their track until after the unveiling.

mntlvr23| 8 febbraio 2017Excellent - good to hear about a mule and about some great range. The trickle of news is starting to pick up a little between this, and the battery max size, and the supercharger status, etc. I am excited that it likely will be speeding up more in the near future with the meeting in two weeks and then the unveil perhaps within the next month after that - and hopefully a slew of information to whet our chops.

Efontana| 8 febbraio 2017Is there a minimum range for supercharger effectiveness - drive the supported routes?

Red Sage ca us| 8 febbraio 2017Napkin math guesstimates...

_70 ___ 253 - 281

70D ___ 259 - 288

_75 ___ 269 - 299

75D ___ 280 - 311

Hmmm... Looks like the Model 3 would have to be extremely efficient to manage 300+ miles with only a 70 kWh battery pack. Unless someone decided to artificially limit Performance to a 60 MPH top speed. I think that rather unlikely.

SamO| 9 febbraio 2017But I thought Model 3 would only offer 215 miles of range? Sad.

Efontana| 9 febbraio 2017I wonder if the front motor is the same technology as the rear motor.

It might make sense to make the front super efficient for regen and range and use what works for the power.

Efontana| 9 febbraio 2017I wonder if the front motor is the same technology as the rear motor.

It might make sense to make the front super efficient for regen and range and use what works for the power.

CoreDump| 9 febbraio 2017I think the thing that everyone doing estimates is missing is that the M3 battery pack will be made up of 2170 cells which are almost double the energy density, meaning double the KWh per pound. If you look at the difference between older and newer Powerpacks, they are claiming around 2x the energy density from the 2170 cells.

Smaller car + lighter 70KWh pack and you could imagine 300 miles. The current thinking is that the base model will have something like a 50-55KWh pack, which @55KWh, you would be at 215miles or range with 255Wh/Mile. 255Wh/Mile doesn't seem to be a huge stretch for a smaller and lighter car/pack. At 255Wh/Mile it would be 294 miles for a 75KWh pack. If the 215 mile range pack is actually a 50KWh pack, then you are at 233Wh/Mile which would put a 70KWh pack at just a hair over 300 miles of range.

I think that people are under estimating how much lighter the vehicle could be made with smaller wheels, lighter packs and everything about the car being lighter.

The Bolt is at 252Wh/mile with 60KWh pack at ~3500lbs. 60KWh MS is is around 1000lbs heavier and averages under 300Wh/Mile? Makes me think that we will see a 300 mile Model 3 if the pack really is that much higher density, you could shave 600lbs off the Model S with 2170 cells alone. The Model 3 could actually be the same weight as the Bolt.

I am sure we will find out soon enough.

andy.connor.e| 9 febbraio 2017@SamO

That was the estimated base range, whos range was estimated based on the March 31 2016 prototype. We do not know what the base pack size will be, nor do we know what "upgrade level" the 70kWh is from the base. Model 3 base range estimate could still very well be ~215 miles. 70kWh could be a significant size upgrade.

andy.connor.e| 9 febbraio 2017@CoreDump

Dont forget that the M3 will be more aerodynamic than Bolt (less air resistance). Its important to understand that Tesla's future depends on the Model 3 success, whereas GM could care less if the car is successful. That alone determines the amount of effort put into the car to make it the "best".

So you can compare 2 apples one grown with pesticides that received no care, and one grown with no pesticides and was constantly weeded and monitored. Consumers will decide which product is successful.

Frank99| 9 febbraio 2017CoreDump -

I don't think the 2170's are twice the KWH per pound. The chemistry likely remains similar, so the ingredients are similar, so the batteries are likely to give about the same energy per pound - maybe 10% more due to more refined chemistry.

I think the 2170's biggest benefits are:

1. Dollars per KWH - with fewer cells to create, package, and interconnect for a given battery size, packs will be cheaper to produce.

2. KWH per cubic meter - the cells are likely more volumetrically efficient, meaning for a given amount of space you'll be able to stuff more kwh into it.

Bighorn| 9 febbraio 2017No way there's a 100% increase in energy density with the 2170. Not remotely close.

Bubba2000| 9 febbraio 2017Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860? I know there is packaging efficiency involved. Significant chemistry improvement?

dd.micsol| 9 febbraio 2017with a tailwind and going 25mph sure it can go 300 miles. Just making a point.

Distance is relevant to wind resistance and energy output.

dsvick| 9 febbraio 2017"Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860?"

They can't, at least not as has been stated by them or tested by anyone.

Bighorn| 9 febbraio 2017@Bubba

Step one: Smoke some crack.

JeffreyR| 9 febbraio 2017The big boost in energy density is at the pack level not the cell level. Not sure it reaches the 2x claim, but I think that's what people are thinking. Basically Tesla has optimized the cell (instead of relying on 18650 popularity which made perfect sense for Roadster) and w/ the new cooling approach in the '100' pack then you get better pack power density.

- Model 3 goal for Cd was low twenties (IIRC .21)

- Model 3 goal is to be 20% lighter than Model S

- Model 3 will have new inverters and motors (better efficiency)

- Model 3 will use 21700 cells for much better pack energy density

I think Elon used a simple formula for estimating range:

- Expected cost of pack sets expected size (total kWh)

- Projected efficiency sets consumption (Wh/mile)

- Do the math to get projected range (not true EPA 5-cycle)

Other Considerations

- Gigafactory production estimates were way off (3x original now)

- Demand was through the roof (600K+ reservations)

- The Great Ramp Up is coming (convert those reservations as fast as you can)

- GM's Bolt hit 238 miles w/ 60 kWh battery

An upgraded M3 battery pack + AWD will likely reach a 300-mile range. A base M3 will almost certainly exceed the '215' Elon quoted at the 1st Unveiling.

Haggy| 9 febbraio 2017"But I thought Model 3 would only offer 215 miles of range?"

Tesla never said that. They didn't even say that the base range would be 215 miles. They said it wouldn't be less than that. I can't see them going far above it for the base range, but for marketing reasons they might want to continue rounding pack capacity units to increments of 5. If it takes a 52 kWh pack to give 215, then they might make the base range 227 miles and start with a 55, for example. So although the base might be above 215, it won't likely be above 230.

Tesla did say there would be battery options though.

Frank99| 9 febbraio 2017At this very moment, the Tesla model 3 page states "Model 3 achieves 215 miles of range per charge". I think that's got to be the base understanding we work from right now.

Bluesday Afternoon| 9 febbraio 2017@Haggy

The base Model 3 will most likely be over 238 mile range of the BOLT. I'd be shocked if it was less and lose the base range war to those folks. One great advantage Tesla has with its customers is the option for a larger battery and increased range.

Red Sage ca us| 9 febbraio 2017CoreDump: Please note that there is the observed Wh per mile that EV owners see, and there is the Wh per mile recorded during EPA testing. Those are definitely different things. Also, it is likely some percentage of a battery pack capacity -- perhaps 5% to 10% -- is held 'in reserve' so that a 55 kWh battery pack capacity has a lower amount as a useable quantity.

Where Tesla owners may report as low as 280 Wh/mile, EPA ratings might be as high as 380 Wh/mile for the same car. And it is the EPA rating that the final official determination of range is made. Better to not presume EPA testing results will produce best case results.

55,000 Wh × 0.9 = 49,500 Wh

49,500 Wh ÷ 300 Wh/mile = 165 miles range

49,500 Wh ÷ 240 Wh/mile = 206.25 miles range

49,500 Wh ÷ 215 miles = 230.23255813 Wh/mile

49,500 Wh ÷ 230 miles = 215.2173913 Wh/mile

Badbot| 9 febbraio 2017Original power packs were 100 Kw using the 18650

Todays power packs are 205 Kw using the 2170.

can anyone math the differential ?

Bighorn| 9 febbraio 2017"These batteries are steadily improving every single year – maybe around 5% improvement in their energy density their ability to store energy in a given amount of mass. That’s probably one of the key metrics we worry about. And when we went from the Roadster to the Model S, they have improved by about 40% and when we were designing the Model 3, they were about another 30% better. That improvement just continues on every single year in the background.” JB Straubel 10/18/16

bj| 9 febbraio 2017@Haggy - I think you missed the joke, admittedly it was subtle. SamO was channeling PigeonPDX, and added a Trump-ism.

Red Sage ca us| 9 febbraio 2017Bighorn: Thank you for the quote.

bj: Thank you for the input. I thought I spied that correctly as a remark aimed at the [IGNORED].

TASANB| 10 febbraio 2017@Bubba: "Can somebody explain how Tesla is able to get 2x energy density with the 2170 cells relative to old 1860?"

The 2170 (21mm diameter x 70mm long) has 76% more volume than the 18650 (18mm diameter x 65mm long).

Assuming both cells have similar wall thickness there's significantly more room inside the 2170, if not quite twice the available volume.

If it's slightly more efficient at storing energy (I'm not claiming it is or is not without more details), it would reasonable to assume it's has twice the energy capacity. Even if it's not more efficient, a 76% increase could be called "about double" the capacity.

Bighorn| 10 febbraio 2017It seems the definition of density eludes some folks.

dsvick| 10 febbraio 2017@TASANB - "Even if it's not more efficient, a 76% increase could be called "about double" the capacity."

Capacity is not energy density. Capacity simply means it holds more if it's bigger. Improved density means that it will hold more for the same volume, that is the " more efficient at storing energy " part of what you said.

TASANB| 10 febbraio 2017I wasn't aware that any specs for the energy density were released. All I've seen was a statement from EM saying it's the "highest energy density cell in the world"

I guess I should have specified that there might be some terminology confusion here too. Even without significantly more energy density (energy per unit volume or mass), the 2170 could very well have twice the energy storage capacity (energy per unit)

Bighorn| 10 febbraio 2017@TASANB

Show your volume calculation because I get 47%

dsvick| 10 febbraio 2017@SimplyRed - "The base Model 3 will most likely be over 238 mile range of the BOLT. I'd be shocked if it was less and lose the base range war to those folks."

I think I have to go with Haggy on this one but for a different reason. Range is still a big consideration for people considering BEVs, if Tesla wants to get more and more people into them then I think they need to de-emphasize range as being a factor for the vast majority of people. For those that really do need greater range they have, as you pointed out, the ability to upgrade to a larger battery.

Keep in mind also what you're getting with the Model 3 as compared to the Bolt. If Tesla can keep the base costs down and include more bells and whistles I'm all for it.

TASANB| 10 febbraio 2017volume of a cylinder: r^2 x L x pi

18650: 9^2 x 65 x pi = 16540 mm^3

2170: 11.5^2 x 70 x pi = 29083 mm^3

29083/16540 = 1.758 or 176% or 76% increase

TASANB| 10 febbraio 2017just realized my mistake...

10.5 not 11.5

Bighorn| 10 febbraio 2017And...

TASANB| 10 febbraio 2017Your calculations are correct.

The point I was trying to make (and not doing a very good job) was that energy density doesn't have to be doubled for there to be twice the energy in a 2170 v 18650 cell. That distinction (density v. capacity) might be a source of confusion.

KP in NPT| 10 febbraio 2017Bighorn has spoken. ;-)

topher| 10 febbraio 2017"energy density doesn't have to be doubled for there to be twice the energy in a 2170 v 18650 cell."

But twice the energy doesn't help if you can only fit half the number of batteries into a pack.

The energy density (volumetric) is exactly the metric you want. if you assume that packing density will be analogous from small to large cells. Energy density of cells * packing density * volume equals total energy. Packing cylinders into rectangles is a well studied problem, and in unlikely to present an significant improvements. So without an improvement in (battery) energy density, the total energy is about the same for a given volume.

Tesla has said that the increase in size of the cells gives a increase in energy density (absent any chemistry improvements), and that chemistry improvements are being made at about 5% per year. I heard a _unconfirmed_ report that the 2170 will deliver 50% improvement over gen II batteries (which would be about 10% per year).

Thank you kindly.

topher| 10 febbraio 2017An aside: 2170s are taller, so conversion from volume of pack from a Model S to a Model 3, will need to account for a bit more volume per floor area (which is the real limitation).

Thank you kindly.

andy.connor.e| 10 febbraio 2017Remember that volume is Cubed. (^3) So 2x energy density is not half the volume.

topher| 10 febbraio 2017Yes it is.

Energy density (volumetric) * volume = total energy.

2* energy density (volumetric) * 1/2 * volume = same total energy.

If you are measuring energy density by weight, it might be different, if the density (ρ) varies. But if it is the same, the above still holds.

Thank you kindly.

Bluesday Afternoon| 10 febbraio 2017@Bighorn

"It seems the definition of density eludes some folks"

I thinks that's an insult! But I may be too dense to know for sure. :)

Red Sage ca us| 11 febbraio 2017I haven't checked the math due to laziness, but others have calculated the volume of the new 2170 battery cells as having ~146.667% the volume of 18650 battery cells. JB Straubel has on multiple occasions noted that he expects at least a 30% improvement in energy density for Generation III vehicles in 2017 over what was possible with Generation II in 2012. This tells me to expect the Model ≡ to be capable of storing perhaps ~191% of the total capacity for a given volume compared to Model S in 2012. That is just shy of 200%, or 'double the capacity', but is close enough for polite conversation and speculative projections, I think.

So, if one presumes the battery pack of a 2017 Model ≡ will have 80% the volume of a 2012 Model S...

85 kWh × 0.80 = 68 kWh

68 kWh × 1.4666~ = 99.7333288 kWh

99.7333288 kWh × 1.3 = 129.65332744 kWh

129.65332744 ÷ 68 = 1.90666658

Even if you forget the volume and only go by the increase in energy density...

68 kWh × 1.3 = 88.4 kWh

So the battery pack for Model ≡ will likely have an initial maximum capacity somewhere between 85 kWh and 100 kWh. I fully expect that to be improved to as much as 135 kWh within five years from launch.

Bighorn| 11 febbraio 2017@RedSage

I didn't check your math either, but you don't seem to have accounted for the fact that the bigger cells take up more space so fewer will fit in the battery pack.

topher| 11 febbraio 2017~146.667% the volume

~191% of the total capacity for a given volume

You can't increase the volume and at the same time add that increase to volume independent calculations.

A 146% increase in volume gives exactly zero (0) increase in capacity for a given volume.

85 kWh x 0.80 = 68 kWh

68 kWh x 1.3 = 88.4 kWh

88.4 kWh / 84 kWh = 1.04

so a 4% increase, not 100%.

Thank you kindly.

Bighorn| 11 febbraio 2017@topher

Did you mean volume dependent calculations? Packing efficiency may provide for increased capacity for a given volume, but I haven't seen that math. Easy to calculate cylinders filling rectangles, but there's some tweaking to be had sizing the box.

topher| 11 febbraio 2017No, I meant volume independent calculations. Ones that include phrases like "for a given volume".

To use a simpler example: density is a volume independent calculation. No increase in volume will change the density. Density is mass 'for a given volume'.

Packing cylinders (actually circles suffice) is (approaching) 1/6 * pi * sqrt(3) ~= 0.9068996821

I think we can assume that Tesla took advantage of any small optimizations due to the precise ratio of radius, to length and width.

Thank you kindly.

Bighorn| 11 febbraio 2017Okay, but it seems that the crux of the issue was how many of the 2170s would fit in the volume of the current battery pack, thereby making it volume-dependent. That was my point with @RS, that the number of cells did not remain the same.

## Pages