When do you think the 2170 battery comes to MX and MS?

[LargeHamCollider]

It ARE superior cells. Less weight/power means better car efficiency too. Not putting them in the 'superior cars' seems strange. Maybe it will change again over time...

There is no reason to think that 2170s will have greater power density than the 18650s, there are reasons to think that the 18650s will have greater power density than the 2170s, though the differences should not be significant.

I agree that the 2170s will probably have slightly greater energy density, there will probably be a slightly higher fraction of active material, but it's going to be a small difference.

I don't think there will be a meaningful increase in packing density, maximum circle packing density doesn't change as circle size scales, and given a rectangular container there are greater edge voids for larger circles. Though if coolant lines remain the same diameter it may increase a small amount.

I don't want to seem anti-Tesla or anything here. I have a quarter of my net worth in TSLA, but I'm hearing a lot of "magical thinking" wrt the 2170s, there's nothing fundamentally different about them, they're just 46% bigger and will be made in much higher quantities. The laws of chemistry don't change with cell size, and in fact chemistry typically favors smaller cell sizes from an overall performance standpoint.

 

[Troy]

There is no reason to think that 2170s will have greater power density than the 18650s

Gigafactory cells are expected to have higher energy density based on these data points:

1. That's what Elon and JB have said when answering a question during a conference call. Listen here.
2. Powerpack 1 used Japan cells. Powerpack 2 uses Gigafactory cells. Both have the exact same dimensions but Powerpack 2 stores 100% more energy (200 kWh vs 100 kWh). Source.
3. Powerwall 1 used Japan cells. Powerwall 2 uses Gigafactory cells. The energy density at the pack level for Powerwall 1 was 6400 Wh / 97kg= 66 Wh/kg. For Powerwall 2, it is 13,500 Wh / 19.9 kg= 113 Wh/kg. Source.

I've said this several times before but there is nothing really "next-gen" about the 2170s, the form factor has been around for quite a while and any changes to anode/cathode/electrolyte can be stuck in the 18650s.

I think the point people are making is, the energy density improvements in the 2170 Gigafactory cells which are based on chemistry improvements, have not been transferred over yet to the 18650 cells Tesla uses in the current 100 kWh packs. wk057 has opened up the 100 kWh pack and counted the cells (link). It just has more of the same cells the 90 kWh pack had.

 

[LargeHamCollider]

Gigafactory cells are expected to have higher energy density based on these data points:

1. That's what Elon and JB have said when answering a question during a conference call. Listen here.

This is true but it's irrelevant, cell chemistry is constantly advancing, the 2170s will be better than todays 18650s and tomorrows 18650s will also be better than todays.

2. Powerpack 1 used Japan cells. Powerpack 2 uses Gigafactory cells. Both have the exact same dimensions but Powerpack 2 stores 100% more energy (200 kWh vs 100 kWh). Source.​

Again this is true, but powerpacks use a different chemistry, and are a different product. This is not an indication of a coming step change in NCA energy density.

3. Powerwall 1 used Japan cells. Powerwall 2 uses Gigafactory cells. The energy density at the pack level for Powerwall 1 was 6400 Wh / 97kg= 66 Wh/kg. For Powerwall 2, it is 13,500 Wh / 19.9 kg= 113 Wh/kg. Source.​

True, but again irrelevant, M3 use NCA not NMC, there is nothing here that should lead anyone to believe that an NCA jump similar to the NMC jump is coming.

I think the point people are making is, the energy density improvements in the 2170 Gigafactory cells which are based on chemistry improvements, have not been transferred over yet to the 18650 cells Tesla uses in the current 100 kWh packs. wk057 has opened up the 100 kWh pack and counted the cells (link). It just has more of the same cells the 90 kWh pack had.

This is likely the case, expect the energy density improvements to be no greater than 7% over current cells. Your facts are correct, but your inferences are going to leave yourself and others disappointed.

 

[Xenoilphobe]

If this is true I'm keeping my Model S until the Y comes out.
You never know what the hell is going on with these damn battery games - just get me a 600 mile car - I don't care what freaking battery goes in it.

 

[Saghost]

I don't think there will be a meaningful increase in packing density, maximum circle packing density doesn't change as circle size scales, and given a rectangular container there are greater edge voids for larger circles. Though if coolant lines remain the same diameter it may increase a small amount.

We saw a Tesla patent go by for end of cell cooling by heat pipes several months ago, and another for a flexible circuit board connection to the cell with both contacts on the same end of the cell and built in fusing necks on the board (so the connection to the cell can be less critical.)

I suspect both of those may show up on the 2170 packs, and may throw off the assumptions about packing density...

 

[Canuck]

I'm currently waiting to purchase a 100D Model X in September with a December delivery. I'm hoping that the S and X will get the new batteries and other technologies soon after the model 3 reveal. Being their premium offerings, it makes sense. I know I won't order my car if there technologies in the model 3 that aren't available in a 100K car. I suspect a lot of other buyers will feel the same way. I think that's a possible reason why they've delayed the model 3 reveal. I really hope I'm right

So if the Model 3 has the new 2170s, while X and S continue to use 18650s, will you buy a Model 3 instead, or not buy a Tesla at all?

It seems so odd to me because I can't imagine not driving a Tesla today (and for over three years now), over some new feature that no one knows when it will come out, and then punishing myself by not driving an S because people can buy a cheaper car with better batteries, assuming for the sake of argument that they are better.

 

[Xenoilphobe]

I think he is concerned about depreciation - why would you buy last years battery tech, when you know that a newer standard is available? I understand the comment about get a Tesla today thing - I would buy a used one - and save you $$$ until this gets sorted out. Why take the massive depreciation hit knowing its just a matter of time before they sunset the old 18650 laptop battery cells..

 

[rowdy]

Sad to see this decision from Tesla. I've had a deposit down for the X for several years and was happy to wait til the end of the year for a changeover to the new cells.

I don't currently see the range of ModelX as being suitable for my needs and have been waiting for a 120kwh pack so I can comfortably do day trips into the country with a loaded ModelX.

The model X just doesn't have the efficiency of the S when fully loaded. I feel it needs >100kwh.

I will keep my S and the two 3's currently reserved.

As for the X though I think I'll be asking for my $6,000 back.

 

[LargeHamCollider]

I think he is concerned about depreciation - why would you buy last years battery tech, when you know that a newer standard is available?

Batteries are not microchips, tomorrow's 2170s will not be significantly better than tomorrow's 18650s in any dimension but cost. Only a very small number of people will care about the form factor of the cells in their battery packs.

There will not be a significant depreciation hit.

 

[Xenoilphobe]

Lets assume the cost of the 2170's are 30% less per KW of output over the legacy 18650. You don't think this will impact legacy cars? I really think this is very short sighted by Tesla - what wouldn't they offer a longer range car with better battery density in a redesigned battery tray under the S & X?
Range is king in cars in this class - we don't all live in California in the mild weather - and some of us actually drive these just like our ICE's: Fast, heavily laden, heat or AC blasting.
More range means I will pay Tesla more money for a new car - not so much for the sub 300 mile car that might do that down hill in with a tail wind, going 45 MPH, with AC and heat off, and tires over inflated - who wants to do that - some nerd who wants to claim the Genius book of world records for causing the most accidents?
I want to do 80 MPH plus with jack rabbit starts, with the sunroof open, ac blasting, towing a boat - charge me more for it - I'll pay it.

Give the market what it wants - Audi / Porsche / BMW / MBUSA are you listening? A7 or Q7 with 600 mile range, a hitch - I don't need a charging network then - everything would be destination based charging - cause I'm not driving more than 600 miles in a day - and I have to sleep somewhere... right?

 

[Cowby]

Lets assume the cost of the 2170's are 30% less per KW of output over the legacy 18650. You don't think this will impact legacy cars? I really think this is very short sighted by Tesla - what wouldn't they offer a longer range car with better battery density in a redesigned battery tray under the S & X?
Range is king in cars in this class - we don't all live in California in the mild weather - and some of us actually drive these just like our ICE's: Fast, heavily laden, heat or AC blasting.
More range means I will pay Tesla more money for a new car - not so much for the sub 300 mile car that might do that down hill in with a tail wind, going 45 MPH, with AC and heat off, and tires over inflated - who wants to do that - some nerd who wants to claim the Genius book of world records for causing the most accidents?
I want to do 80 MPH plus with jack rabbit starts, with the sunroof open, ac blasting, towing a boat - charge me more for it - I'll pay it.

Give the market what it wants - Audi / Porsche / BMW / MBUSA are you listening? A7 or Q7 with 600 mile range, a hitch - I don't need a charging network then - everything would be destination based charging - cause I'm not driving more than 600 miles in a day - and I have to sleep somewhere... right?

Totally agree with this, however, we'd be the minoriy buyer as that'll probably costs too much to produce in the near term. I predict we might see this kind of range near 2025-2030 as more manufacturers get in the game and drive the battery costs down and the demand becomes greater for a vehicle with that range.

 

[LargeHamCollider]

Lets assume the cost of the 2170's are 30% less per KW of output over the legacy 18650. You don't think this will impact legacy cars? I really think this is very short sighted by Tesla - what wouldn't they offer a longer range car with better battery density in a redesigned battery tray under the S & X?
Range is king in cars in this class - we don't all live in California in the mild weather - and some of us actually drive these just like our ICE's: Fast, heavily laden, heat or AC blasting.
More range means I will pay Tesla more money for a new car - not so much for the sub 300 mile car that might do that down hill in with a tail wind, going 45 MPH, with AC and heat off, and tires over inflated - who wants to do that - some nerd who wants to claim the Genius book of world records for causing the most accidents?
I want to do 80 MPH plus with jack rabbit starts, with the sunroof open, ac blasting, towing a boat - charge me more for it - I'll pay it.

Give the market what it wants - Audi / Porsche / BMW / MBUSA are you listening? A7 or Q7 with 600 mile range, a hitch - I don't need a charging network then - everything would be destination based charging - cause I'm not driving more than 600 miles in a day - and I have to sleep somewhere... right?

You're talking about cost per kW? Perhaps you mean cost per kWh. Range is proportional to kWh not kW. This will have ZERO impact on "legacy" cars.

2170 cells will not be notably superior to 18650s in any dimension but cost.

2170 cells will not be notably superior to 18650s in any dimension but cost.

18650s will keep the CG slightly lower and will probably have slightly higher power density. Frankly you come across as clueless on this stuff. A good rule of thumb wrt to Tesla's battery decisions is "if you think they're doing something stupid, you don't understand the situation" ...I thought Tesla was wrong for not building parallel hybrids 10 years ago, I was wrong. 7 years ago I thought Tesla should be using LFP cells, I was wrong. S/X probably will go to 2170s eventually, but there is no pressing reason to make it happen soon.

 

[ecarfan]

Batteries are not microchips, tomorrow's 2170s will not be significantly better than tomorrow's 18650s in any dimension but cost. Only a very small number of people will care about the form factor of the cells in their battery packs.

Agreed, and I would go farther: the percentage of the EV-buying car market that even is aware that the Tesla battery pack is composed of thousands of small cylindrical cells is so minute as to be essentially zero, and almost every EV buyer simply thinks of the pack as just "a battery".

Yes, there are some people in this thread who are remarkably concerned about whether their next Tesla has 2170 or 18650 cells. If you want to wait until the S/X has 2170 cells, feel free. But Tesla not making that change this year will have no significant impact on S/X sales, in my opinion.

 

[Saghost]

Agreed, and I would go farther: the percentage of the EV-buying car market that even is aware that the Tesla battery pack is composed of thousands of small cylindrical cells is so minute as to be essentially zero, and almost every EV buyer simply thinks of the pack as just "a battery".

Yes, there are some people in this thread who are remarkably concerned about whether their next Tesla has 2170 or 18650 cells. If you want to wait until the S/X has 2170 cells, feel free. But Tesla not making that change this year will have no significant impact on S/X sales, in my opinion.

That's fair. On the other hand, already having an amazing Tesla to drive every day, this change is actually one of the three I'm waiting to see along with the Xavier SoC for AP (with new sensors?) and the interior refresh (with HUD? ) before I seriously consider upgrading.

The main reason I'd upgrade is FSDC, and Tesla hasn't made much visible progress on that yet anyway, so I've got plenty of time...

 

[vandacca]

I'll agree to disagree with those that claim that the 2170 cells will have less energy density than 18650. The simple analogy would be potato chip bags. You get much more potato chips and less packaging-per-chip in a larger form factor.

However, IMO the greatest potential benefit of the 2170 cells is that they'll likely be engineered to maximize V3 Supercharging, whereas the 18650 are not.

 

[scaesare]

Agreed, and I would go farther: the percentage of the EV-buying car market that even is aware that the Tesla battery pack is composed of thousands of small cylindrical cells is so minute as to be essentially zero, and almost every EV buyer simply thinks of the pack as just "a battery".

Yes, there are some people in this thread who are remarkably concerned about whether their next Tesla has 2170 or 18650 cells. If you want to wait until the S/X has 2170 cells, feel free. But Tesla not making that change this year will have no significant impact on S/X sales, in my opinion.

I agree with you and @LargeHamCollider. The real secret sauce is the cell chemistry. The packaging in to 2170 vs. 18650 likely has a very small impact in terms of cell casing material contributing to overall weight of the pack.. but I'd be surprised if it was even one percent of the overall pack mass.

The larger cells will likely allow for easier and cheaper pack assembling & layout, etc... but there's going to be little noticeable difference in pack made out of 18650's vs 2170's containing the same cell chemistry.

It's much ado about nothing, IMO. Now a major CHMISTRY change... that has my interest.

 

[LargeHamCollider]

I'll agree to disagree with those that claim that the 2170 cells will have less energy density than 18650. The simple analogy would be potato chip bags. You get much more potato chips and less packaging-per-chip in a larger form factor.

However, IMO the greatest potential benefit of the 2170 cells is that they'll likely be engineered to maximize V3 Supercharging, whereas the 18650 are not.

No one is claiming that 2170 will have lower energy density than 18650, lower power density perhaps but not energy density. Don't expect the 2170s to supercharge faster than18650s that were produced at approximately the same time. Power density and charge speed are related.

 

[RobertF]

Never. Believe they will change S to S2 or something else when that occurs. The 2 battery sizes are not compatible without significant remodeling and engineering changes. They really need to "refresh" the S anyway to keep new customers coming.

 

[Xenoilphobe]

Cost is a major issue and the 2170 claims to address this metric - the 18650 not so much.

My clueless background includes too many power related projects over a 22 year career in the military dealing with power distribution , signal propagation and troubleshooting at a depot level reading schematics down to the component on the board (back when they still did that).

Most twelve year olds realize that batteries are made of many cells. I was probably about ten when I tore down a 6 volt battery to pull the 4 - 1.5 volt "D" cells out to use in a different flashlight.

You might want to read this Model S with 2170 cells conceptualized

This could drive a decline in Model S sales or at least a significant cost reduction, if they don't update the 18650.

 

[Troy]

TLDR: The Model S 100D range might increase from 335 to 354 mi EPA rated range if Tesla upgrades the 18650 cell chemistry.

Hi. Check out this video by Tesla CTO JB Straubel from 35:47 to 37:05. JB says the energy density in Model 3 batteries has improved by 30% compared to the original Model S battery. We need to subtract 5.3% from that because that was already materialized with the upgrade from 85 to 90 kWh packs. Those two packs had the same cell count but the actual capacity increased from 81.5 kWh (in the 85 kWh pack) to 85.8 kWh (in the 90 kWh pack) because of chemistry improvements. Therefore going from 90 kWh aka 100 kWh cell chemistry to 2170 cell chemistry should improve capacity by 30-5.3= 24.7%. The 90 and 100 packs have the same cells.

By switching the 18650 chemistry to the latest chemistry they have, Tesla could either release a 125 kWh pack that weighs the same as the 100 kWh pack or they could make a new 100 kWh pack that weighs 25% less and is about 25% cheaper to produce. I think this second option is more likely because of Elon's tweet here. Why would it be cheaper? Because it has 25% fewer materials. Most of the cell cost is tied to material costs. In the video, JB talks about this after 36:33.

Based on my calculation, the cells in the 100 kWh pack weigh 515 kg. This excludes 113 kg for pack housing. Upgrading the chemistry would mean 128 kg weight reduction. This would increase EPA rated range for the Model S 100D by 19 miles, from 335 to 354 miles.

Detailed calculations can be found here. These are some data points I used in calculations:
1. An official document that shows the 85 kWh pack weighs 545 kg. Link
2. wk057's estimate that the pack housing weighs 250 lbs (113 kg). Link
3. The 100 kWh pack has 102.4 kWh actual capacity. Link
4. The actual capacity of the 85 and 90 kWh packs is listed here.