Rumor: Model 3 to use new 4416 battery cell

[Cloxxki]

The thing that I have learned is that the Cells are only half the battle. Much of the cost, weight and complexity is actually in the pack. They have made just as much progress in simplifying the pack, packing the cells tighter and more efficient/effective cooling that requires less and less space. A simpler pack is easier to automate mass production. This could account for as much as half of the cost reductions that Elon and JB have spoken about. The other half would come from the continued 7 percent per year improvement to energy densities at the cell level from improved materials and chemistry, which inversely goes directly to cost to produce the same KWh in a finished cell/pack. This is maximized by the scale and shortening of the supply chain from the Gigafactory which allows you to get as close as possible to lowest marginal cost from massive economies of scale and automation from Grohmann. In 3 years there is no reason to think that they wouldnt be able to squeeze another 20% or a 120KWh pack into S/X. It would be the same physical size and weight, just more energy dense. It should actually cost the same or maybe a little less if they can further simply the pack architecture. A new chemistry might allow the batteries to operate at higher and higher temps and still maintain efficiency and long life. This allows Tesla to go with less expensive and less power hungry cooling. In short, energy density is not just at the cell level but also the module/pack level.

There has been 7% progress exactly never so far. The cells from the 2012 Model S got updated all of once, and it got them the 85->90 upgrade. Well, 81 to 86...

People love to throw around hugely low numbers to indicate the scale of Model 3 vs Model S.
Official figures (if such exist) that I've seen state 4% narrower and same height. Where are the other 11% or so coming from?

 

[Reciprocity]

There has been 7% progress exactly never so far. The cells from the 2012 Model S got updated all of once, and it got them the 85->90 upgrade. Well, 81 to 86...

People love to throw around hugely low numbers to indicate the scale of Model 3 vs Model S.
Official figures (if such exist) that I've seen state 4% narrower and same height. Where are the other 11% or so coming from?

7% would be what we call an average. Meaning there could be no change, because the chemistry does not change every year. The 7% is a historical value and the value that Elon and JB have reiterated. Im no battery expert, I trust they are. Maybe they are lying to us all. Or maybe its just an average over a long period of time.

 

[CarlitoDoc]

4 more days people

 

[Zaphod]

I could possibly see these allegedly larger cells used for the Semi, but think it makes no sense for the 3 (or no sense at this point in the game to have such a drastic design change.)

 

[JeffK]

Ben seems like a pretty cool guy but he's a bit gullible sometimes.

 

[Cloxxki]

7% would be what we call an average. Meaning there could be no change, because the chemistry does not change every year. The 7% is a historical value and the value that Elon and JB have reiterated. Im no battery expert, I trust they are. Maybe they are lying to us all. Or maybe its just an average over a long period of time.

Cool, so if in 2012 they got 81.5kWh out of 7,104 cells, what would we have today with the 8,256 cells that fit in the biggest pack? 5 years of average annual 7% progress means +40%.
81.5*(1.07)\5 * (8256/7104) = 132.8kWh.
And we take that 85 figure seriously, 138.5kWh.
Tesla is holding out on us!

I was hopeful also when those numbers were thrown around, but we've been let down.
The 85->90 upgrade of the cells was just from a 81.5 to 85.8 (5.2%). None since that I'm aware of.
About testing of the 2170 chemistries, JB and Ellon said something like 15% improved chemistry density to be observed. That gets us to +21% only. But it's just a claim so far adn only the wildest of fanboys have such expectations.

 

[Yggdrasill]

I agree, the guy is out of his depth.

JB/Elon claimed seeing 15% better density in testing in terms of chemistry alone. And the bigger diameter of the 2170's helps a bit also, let's sy 3%.
Plus, the cells will stay upright, and are 7.7% taller than 18650's.
We're already at 102.4kWh. Above would account for 130.6kWh.
How do you see this?

A 5-seat Model S could use the back row foot well. Fits a serious amount of cells. I know it will never happen.

I don't think you'll see any improvement due to the larger diameter. The current 102.4 kWh pack already has the updated cooling pipes, allowing for more cells in the same area. (I'm assuming the Model 3 has the exact same architecture.) So basically, all you'd get is the 7.7% from increased height and improvements in chemistry.

I don't think the 15% can be taken at face value, (I don't know in which context it was said.) If I'm right about 3840 cells for the Model 3, and the pack is exactly 75.0 kWh, that implies an improvement in chemistry of 5%.

That would allow a 21-70 based pack to be 102.4 x 1.077 x 1.05 = 115.8 kWh.

Now, maybe I'm way off base with my estimate for improvement in chemistry. If I'm still right about 3840, but the pack capacity is 77 kWh (allowing for an *available* capacity of 74.6 kWh), that would be an improvement in chemistry of 7.8%.

That would allow a 21-70 based pack to be 102.4 x 1.077 x 1.078 = 118.9 kWh.

What if I'm wrong about both the number of cells and the capacity? If there's only 3600 cells and 77 kWh, that would be an improvement in chemistry of 15%.

That would allow a 21-70 based pack to be 102.4 x 1.077 x 1.15 = 126.8 kWh.

The reason why I think the last scenario here is unrealistic is that I feel pretty confidently that the Model 3 pack has exactly four times as many cells as the Powerwall. With only 900 cells per powerwall, the implied improvement in chemistry for NMC is a whopping 44% going from the Powerwall 1 to the Powerwall 2. I find that highly implausible. But my first two scenarios seem quite reasonable, i.e. 5-7.8% improvement in chemistry for NCA, and 34.9% improvement in chemistry for the powerwall/NMC. (That's still high, but also a lot more plausible.)

 

[Yggdrasill]

I went looking for the "15%" quote:

25:23 Journalist: On the Gigafactory, is the chemistry going to be the same battery chemistry that you're currently using or is that part of the discussions that are going on with Panasonic?
25:34 Elon Musk: There are improvements to the chemistry, as well as improvements to the geometry of the cell. So we would expect to see an energy density improvement and of course a significant cost improvement. JB, do you want to add anything?
25:53 JB Straubel: Yeah, that's right. The cathode and anode materials themselves are next generation. We're seeing improvements in the maybe 10% to 15% range on the chemistry itself.
26:09 Elon Musk:Yeah, in terms of energy density.

Okay, not 15%, but 10-15%, and this is something that was said back in July 2014, before they started using cells with the partial silicon anode. The partial silicon anode accounted for ~6%, leaving 3.8-8.5% on the table. If they were correct back in 2014, both my scenarios are plausible.

 

[Reciprocity]

Cool, so if in 2012 they got 81.5kWh out of 7,104 cells, what would we have today with the 8,256 cells that fit in the biggest pack? 5 years of average annual 7% progress means +40%.
81.5*(1.07)\5 * (8256/7104) = 132.8kWh.
And we take that 85 figure seriously, 138.5kWh.
Tesla is holding out on us!

I was hopeful also when those numbers were thrown around, but we've been let down.
The 85->90 upgrade of the cells was just from a 81.5 to 85.8 (5.2%). None since that I'm aware of.
About testing of the 2170 chemistries, JB and Ellon said something like 15% improved chemistry density to be observed. That gets us to +21% only. But it's just a claim so far adn only the wildest of fanboys have such expectations.

I dono bro, how bout someone else explains it to you:

The graph that Tesla's J.B. Straubel showed a while back exhibits this behaviour perfectly. While the improvements are fairly erratic from year to year, from this graph we can deduce that the improvement ranges, on average, from 5 to 8% per year. Again, however, we see this long linear stretch which leads us, in the short term, to think that "batteries are not getting that much better"... only to be surprised the next time a "leap improvement" occurs when a new chemistry or energy storage paradigm comes along.

https://www.quora.com/Is-it-true-that-battery-energy-density-improves-5-8-per-year

 

[Yggdrasill]

Okay, not 15%, but 10-15%, and this is something that was said back in July 2014, before they started using cells with the partial silicon anode. The partial silicon anode accounted for ~6%, leaving 3.8-8.5% on the table. If they were correct back in 2014, both my scenarios are plausible.

Also, just to mention it, something is wrong somewhere, with these figures.

- The 85 kWh pack was actually 81 kWh, with 7104 cells, which is 11.40 Wh/cell.
- The 90 kWh pack added 5 kWh, with 7104 cells, which is 12.11 Wh/cell. An improvement of 6.2%.
- The 100 kWh pack is actually 102.4 kWh, with 8256 cells, which is 12.40 Wh/cell. An improvement of 2.4%??

What's happened here? Did Tesla quietly phase in superior cells without telling anyone? Is the 102.4 kWh estimate too high? Is the 86 kWh for the 90 pack too low? I don't have the answers, but it's not impossible there's less on the table than the 3.8-8.5% I indicated.

 

[Yggdrasill]

No problem When reworking your numbers remember that frontal area is proportional to both width and height. So 85% of the frontal area would be an average of 92% the width and 92% the height. Also don't forget that actual battery capacities aren't the same as they report for the pack, there's always some left untouchable so that you're not doing 100% deep cycles. If I recall correct, something like 5%-ish, depending on which model you get.

I redid my calculations, and there wasn't much difference. I used 2.24 m^2 for Model 3, and 2.34 m^2 for Model S. (I've always used the Model S as the benchmark, so with the Model S having a lower front area than I thought, the difference was actually smaller with the new figures.)

I arrived at 298 EPA miles for 75 kWh pack (77 kWh gross, 74.6 kWh net) and 236 EPA miles for 55 kWh pack (57.8 kWh gross, 55.4 kWh net).

 

[KarenRei]

I redid my calculations, and there wasn't much difference. I used 2.24 m^2 for Model 3, and 2.34 m^2 for Model S.

You think Model 3 is going to have 96% of the cross section of Model S?

98% the width and height?

 

[Yggdrasill]

You think Model 3 is going to have 96% of the cross section of Model S?

98% the width and height?

How big are you under the impression that Model 3 is?

I assume the Model 3 is exactly the same height, 1.44 meters. (The Model S is really low, and the Model 3 will be too. Remember, the pack builds up ~8 cm, so 1.44 m for the Model 3 is the same as 1.36 m in a fossil car.) I also assume the Model 3 is 4% narrower, 188 cm vs 196 cm.

This is actually based on the estimates by Motor Trend: Tesla Model 3 In-Depth - Exclusive Photos and Analysis

 

[scaesare]

Have you guys seen this post from Teslarati? Thoughts?

Tesla could be using a "4416" battery cell in Model 3 and next-gen S, X fleet

That would be a cell over 6" tall... not likely in a pack with the cells oriented vertically as they are today. While they could always use a horizontal orientation, I'm dubious... particularly since media event covering the cell production at the gigafactory have published pics of the cells and they are 2170 (at least they appear to match those claims... I haven't seen an exact measurement, but they certainly aren't 6+ inches long...)

 

[Reciprocity]

That would be a cell over 6" tall... not likely in a pack with the cells oriented vertically as they are today. While they could always use a horizontal orientation, I'm dubious... particularly since media event covering the cell production at the gigafactory have published pics of the cells and they are 2170 (at least they appear to match those claims... I haven't seen an exact measurement, but they certainly aren't 6+ inches long...)

It could all be a shell game, but I doubt it. I agree that it does not make sense that the cells would be anything other then what they have shown to date, which is the 21-70. Could it be possible they are working on a different cell, maybe for the Semi? I guess they could be. Certainly the Semi has more room for taller cells. Do larger cells charge faster is is the opposite true. I dont think the Teslanomics guy is making things up, he probably has a source he trusts. It could actually be a trap to ferret out the moles in the organization. Feed different bits of disinformation to people you suspect of leaking and see what specifics shows up in the public domain.

 

[Lloyd]

The downside to this battery is a slower charge rate due to cooling. I noticed on the photo of the model III supercharging that at about 1/3 fill, it registered 163 MPH charge rate. That is about 1/2 of what we would expect in the model x or S! Am I wrong?

 

[Yggdrasill]

The downside to this battery is a slower charge rate due to cooling. I noticed on the photo of the model III supercharging that at about 1/3 fill, it registered 163 MPH charge rate. That is about 1/2 of what we would expect in the model x or S! Am I wrong?

I don't think you can draw any conclusions. I know some pictures were taken of a Model 3 sharing a charger with another Tesla. Was this those pictures? Also if it was an RC, they could have prototype SW, known issues with the cable harness that will be resolved on the production versions, etc. Even if the car was working perfectly, sometimes superchargers don't give the full speed for whatever reason. Also "mph" is fudible based on how it's calculated. If it was kW, it would be easier to draw conclusions.

 

[gavine]

Didn't Elon say that there are no plans to ever make the S/X battery more than 100kWh. My take is that they feel as though 100kWh's is more than enough range with Supercharging and I agree. I mean, there are certain edge cases where more than 100kWh's makes sense, but for the most part, that's a good number to have as a max.

 

[Electroman]

but for the most part, that's a good number to have as a max.

On a blistery cold day, I would suspect the range will drop to 200 odd miles even for a 100D. I think the range on the cold sub-zero days should be a comfortable 250-280 miles before we end the range anxiety thoughts once for all.

 

[PJFW8]

Have you guys seen this post from Teslarati? Thoughts?

Tesla could be using a "4416" battery cell in Model 3 and next-gen S, X fleet

Garsh on the Model 3 owners site postulates that this is not a 4416 cell, but a battery containing 4,416 individual cells. He calculates that this will provide a 77 kWh pack. Thus: "4416 cell battery".