Panasonic cells for Model S

[Eberhard]

Tesla has only one qualified Source for the 18650 cell, Panasonic.
The attached link gives more information for the current 2200mA cells as well as the new 18650A cells with 3100mA used for the 300miles pack

http://industrial.panasonic.com/www-data/pdf/ACI4000/ACI4000CE2.pdf

 

[stopcrazypp]

Tesla has only one qualified Source for the 18650 cell, Panasonic.
The attached link gives more information for the current 2200mA cells as well as the new 18650A cells with 3100mA used for the 300miles pack

http://industrial.panasonic.com/www-data/pdf/ACI4000/ACI4000CE2.pdf

How were you able to find that datasheet (what page linked to it)? This one is more informative than the "official" datasheets on the other parts of Panasonic's website because it actually goes to 500 cycles instead of just 300.

 

[Norbert]

My understanding was that Tesla will be using custom cells (from Panasonic) in the Model S, with different chemistry and topology, based on input from Tesla itself. I wouldn't expect data about these cells to be available on Panasonic's website.

 

[stopcrazypp]

My understanding was that Tesla will be using custom cells (from Panasonic) in the Model S, with different chemistry and topology, based on input from Tesla itself. I wouldn't expect data about these cells to be available on Panasonic's website.

Tesla is using cells with new chemistry (Panasonic calls it Nickel Oxide based New Platform or NNP), but they are still 18650s. A previous Tesla/Panasonic press release already revealed it will be 3.1Ah 18650 cells, which are obviously the NCR-18650A described in the document (it's the only 3.1Ah cell Panasonic makes and as far as I know also the only 3.1Ah cells made by any manufacturer).
http://www.engadget.com/2010/04/23/panasonics-3-1ah-batteries-to-be-used-in-the-tesla-model-s-hav/

 

[Norbert]

Tesla is using cells with new chemistry (Panasonic calls it Nickel Oxide based New Platform or NNP), but they are still 18650s. A previous Tesla/Panasonic press release already revealed it will be 3.1Ah 18650 cells, which are obviously the NCR-18650A described in the document (it's the only 3.1Ah cell Panasonic makes and as far as I know also the only 3.1Ah cells made by any manufacturer).
http://www.engadget.com/2010/04/23/panasonics-3-1ah-batteries-to-be-used-in-the-tesla-model-s-hav/

Yes, they use the 18650 format, but a proprietary (not just new) chemistry and internal geometry. Elon Musk described this in the Q4 2010 webcast available here (main section and a bit more in the Q&A):

Tesla - Events Presentations

The batteries in the Roadster are a standard 18650 variety, but the Model S batteries are heavily modified. My understanding is that they won't be available to other customers of Panasonic.

 

[Eberhard]

From the annual paper:

Panasonic
Panasonic is the supplier of cells for one of our current battery packs. In January 2010, we announced that we were collaborating with Panasonic on the development of next-generation electric vehicle cells based on the 18650 form factor and nickel-based lithium ion chemistry.

The charts shows for the 18650A cells a quite quick degradation within the first 300 cycles but then gets stable even beyond 500 cycle.

 

[tomanik]

Yes, they use the 18650 format, but a proprietary (not just new) chemistry and internal geometry. Elon Musk described this in the Q4 2010 webcast available here (main section and a bit more in the Q&A):

Tesla - Events Presentations

The batteries in the Roadster are a standard 18650 variety, but the Model S batteries are heavily modified. My understanding is that they won't be available to other customers of Panasonic.

So is this only 'true' for batteries in the 300 mile pack since it is being assumed the 3.1AH cell are only used in this pack? Meaning the 160 adn 230 mile packs are likely just existing roadster cells

 

[Eberhard]

The 160 and 230 mile will have the 2200mAh pack. These shows a quite linear degradation over 500cycles.

 

[Doug_G]

The charts shows for the 18650A cells a quite quick degradation within the first 300 cycles but then gets stable even beyond 500 cycle.

How much degradation? Does this mean that a 300 mile pack will turn into a 230 mile pack after 300 cycles?

 

[bolosky]

How much degradation? Does this mean that a 300 mile pack will turn into a 230 mile pack after 300 cycles?

Eyeballing the chart, it looks like the knee in the curve is around 2400 mAh at 300 cycles. If 3100 mAh is the nominal capacity that gives 300 mile range, then this would be about 230 miles. The degradation up to there looks pretty linear, and after that it seems nearly flat (as in it doesn't get worse for the next 200 cycles).

Their definition of a cycle is running from 4.2V to 2.5V, which I'm guessing is (way) more than Tesla will do given how the Roadster works. So, I'd expect that 300 miles on the S would be maybe 90% of a cycle on the chart, so 300 (chart) cycles would be 333 S cycles, or almost exactly 100,000 miles. Furthermore, I'd guess that keeping the battery in the middle of the range will reduce the degradation (probably by a decent amount). So, what this chart says, roughly, is that the 300 mile pack should be good for 230 miles after 100,000 miles of driving, and that it will stay at 230 miles for at least the next 65K miles (and probably more, that's just the end of the Panasonic measured).

This doesn't take into account the effect of calendar age on the batteries, so if it takes you 15 years to get to 100,000 miles your results will be (much) worse.

 

[Doug_G]

Good analysis - thanks.

 

[Norbert]

There was some other text where Tesla described their priorities for batteries, and although I don't recall where this was mentioned, I'm quite sure that cycle life was very high on this list. So there is a chance they have been trying and perhaps succeeded in improving cycle life in their custom version.

 

[AndrewBissell]

I'm honestly not convinced there is a custom Tesla version of the Panasonic battery. My reading was that they input their requirements and suggestions strongly to Panasonic and that these had influenced Panasonic's design for the 3100mAh cells.

Does anyone have specific quotes to show otherwise? (Here's hoping....)

 

[stopcrazypp]

I have attached the relevant graphs, annotated them with the "Model S miles" on the left, and also noted the intersection points in terms of cycles. The 2250mAh version has a basically linear/constant drop. The 3100mAh version has a steeper almost linear drop first and then after ~300 cycles the drop levels out. Interestingly the 2250mAh starts out at about 2200mAh instead, and the 3100mAh drops to 3000mAh within 5 cycles.

As pointed out before, the conditions are not the same. The 2250mAh had a cutoff of 3.0V and charging rate of 0.7C while the 3100mAh had lower cutoff of 2.5V and lower charging rate of 0.5C.

 

[Lloyd]

Note also the with the 18650A's they charged them to a higher state, and discharged them to a lower state of charge for each cycle. It seems like they are a more robust battery. It would be intersting to see how they respond to DC fast charging cycles!

 

[Norbert]

I'm honestly not convinced there is a custom Tesla version of the Panasonic battery. My reading was that they input their requirements and suggestions strongly to Panasonic and that these had influenced Panasonic's design for the 3100mAh cells.

Does anyone have specific quotes to show otherwise? (Here's hoping....)

Did you follow the link to the webcast, and listen to Elon Musk's comments on the batteries, in his first section, and later in the Q&A session? The posted link to a transcript on that part isn't very readable and doesn't reflect very well what is said in the webcast, but it does include the word "proprietary". Why would Tesla use the term "proprietary"? (My interpretation is not just based on this one word, though.)

 

[stopcrazypp]

Besides from the webcast (which I haven't watched since it requires registering, so I'm not sure what exact quote points to custom cells for Tesla), the rest of the evidence points to high probability of the 160 & 230 packs using cells similar in capacity to the Roadster ones, and the 300 pack using the NCR-18650A. And the NCR-18650A certainly uses "proprietary" Panasonic technology: the aforementioned NNP, which gives them the highest energy density in the industry, and also a HRL (Heat Resistant Layer) for safety.

The press release I posted before confirms Tesla will be using NCR-18650A; unless they changed their minds recently. And the press release says Panasonic plans 300 million cells per year, so obviously it won't be exclusive to Tesla, since that's enough for 37,500 300 mile Model S's. Tesla might be able to get a contract to get exclusive use in terms of cars, but those cells definitely will also be used in the consumer electronics and laptop market.

Of course, so far we are just guessing the 160/230 packs will use ~2200mAh cells, but no details on what manufacturer (no press release so far, although for Roadster cells I've seen battery reports that says Sanyo is a confirmed supplier for the Roadster and Smart ed packs). It is possible those cells will be custom made for Tesla, since Tesla will use a lot more of those than the NCR-18650A (making the reasonable assumption the 300 mile Model S will sell a lot less than the other two). However, I'm fairly certain NCR-18650A will be used by Tesla.

 

[Norbert]

Besides from the webcast (which I haven't watched since it requires registering, so I'm not sure what exact quote points to custom cells for Tesla), the rest of the evidence points to high probability of the 160 & 230 packs using cells similar in capacity to the Roadster ones, and the 300 pack using the NCR-18650A.

I wouldn't expect the Tesla battery to have a significantly different energy density. However I had the impression they were aiming to improve the cycle life.

"Registering" for the webcast only requires entering your name and email. You don't have to create a complete account IIRC.

Unfortunately I don't have time today to listen to it again and extract quotes.

And the NCR-18650A certainly uses "proprietary" Panasonic technology: the aforementioned NNP, which gives them the highest energy density in the industry, and also a HRL (Heat Resistant Layer) for safety.

No, Elon Musk indicated it is not vendor specific and they could transfer it to other vendors. Panasonic is the primary (and currently only) supplier for Model S batteries, but not exclusively so.

The press release I posted before confirms Tesla will be using NCR-18650A; unless they changed their minds recently. And the press release says Panasonic plans 300 million cells per year, so obviously it won't be exclusive to Tesla, since that's enough for 37,500 300 mile Model S's. Tesla might be able to get a contract to get exclusive use in terms of cars, but those cells definitely will also be used in the consumer electronics and laptop market.

Of course, so far we are just guessing the 160/230 packs will use ~2200mAh cells, but no details on what manufacturer (no press release so far, although for Roadster cells I've seen battery reports that says Sanyo is a confirmed supplier for the Roadster and Smart ed packs). It is possible those cells will be custom made for Tesla, since Tesla will use a lot more of those than the NCR-18650A (making the reasonable assumption the 300 mile Model S will sell a lot less than the other two). However, I'm fairly certain NCR-18650A will be used by Tesla.

I'm not sure if it was said in the webcast or elsewhere, but Tesla can take advantage of economies of scale by using the same 18650 format which Panasonic uses for their "standard" batteries.

 

[Tommy]

.....evidence points to high probability of the 160 & 230 packs using cells similar in capacity to the Roadster ones, and the 300 pack using the NCR-18650A. And the NCR-18650A certainly uses "proprietary" Panasonic technology: the aforementioned NNP, which gives them the highest energy density in the industry, and also a HRL (Heat Resistant Layer) for safety.....

I would be very disappointed if the 160/230 version batteries did not include the HRL as well as have very similar performance characteristics of the higher energy density battery; the main difference being lower energy density than the 300 mile version. Tesla has made a point about their obsession with safety and quality built into the vehicle; using battery versions (160/230) that fall short in terms of specifications to the 300 mile version (safety and battery life) is going to raise questions about their commitment to safety and quality.

 

[stopcrazypp]

I would be very disappointed if the 160/230 version batteries did not include the HRL as well as have very similar performance characteristics of the higher energy density battery; the main difference being lower energy density than the 300 mile version.

If we are only looking at the Panasonic cells, the possible 2250mAh cells for the 160/230 uses a more stable chemistry called "PSS" (Panasonic Solid Solution) that is supposedly comparable in safety to lithium manganese batteries (like in the Leaf, which doesn't even need liquid cooling). That is why it doesn't need the HRL.

As for performance, a smaller pack with similar chemistry will have inherently less performance (in proportion to capacity): that's just unavoidable. That shouldn't be an issue as long as it can handle the demands of the Model S. And from the other talk about the Model S, there might be a bottle neck at the PEM or the motor which hides any performance advantage the larger pack might have.