Tesla Motors CTO talks future batteries and charging protocols

[Zzzz...]

The conclusion I get from all of this, is that while the Model S may use slightly tweaked cells, replacing them with the standard 3.1 mah Panasonic cell wouldn't make much of a difference in longevity or performance. The difference would be tiny at best.

What we know for a fact is that if you will try to use off the shelf Panasonic batteries in Model S, most of Model Ses would be dead before year four end. Yes, this is not much of a difference in YOUR opinion. Up too you is how you define "slightly tweaked cells". But a matter of fact, most of Model S owners would not be happy with replacing battery pack 5 years. Let alone in 4.

Other problem with laptop batteries - if Tesla tried to use laptop batteries is cold weather. Model S with of the shelf Panasonic cells will simply would not start in freezing temperatures.

You can claim all you want that JB and Musk are idiots and morons and that they waste taxpayers money spending millions on R&D instead of simply buying off the shelf Panasonic/LG/Chinese etc cells... But reality is that most drivers do want Teslas last longer then couple years and work in Norway/Canada/Alaska and such...

And this is exactly why Tesla and Panasonic spend a lot to JB QUOTE: "We’ve totally custom-engineered that cell working jointly with Panasonic to create."

"Totally" in citation above mean millions $$$. More likely dozens of millions investments in R&D.

PS. Claiming use of off the shelf Sanyo cells is a total nonsense too.

- - - Updated - - -

Read my post again. I'm not saying they are using NCR18650A specifically

Here is what you have said before. And what I could not agree with:

And Tesla uses a "retail market" cell, specifically, the NCR18650A (the only 3100mAh 18650 battery that Panasonic makes). Tesla is not using some special formulation by Panasonic

The only thing I don't have is Tesla or Panasonic explicitly saying that the 3.1Ah cell in question has a model number of NCR18650A (I don't expect them to EVER do that because no other EV manufacturer have ever done that either nor do they have a reason to), but it's very easy to connect the two.

And this is what JB saying:

At this point we really have heavily customized that cell. We’ve totally custom-engineered that cell working jointly with Panasonic to create. It’s an automotive cell, tested to automotive standards. It doesn’t go into laptops anywhere.

PS. Even word "specifically" was used in your posts...

 

[Doug_G]

This is starting to sound a little argumentative... please tone it down a bit and help keep things civil. Thanks!

 

[JRP3]

What we know for a fact is that if you will try to use off the shelf Panasonic batteries in Model S, most of Model Ses would be dead before year four end. Yes, this is not much of a difference in YOUR opinion. Up too you is how you define "slightly tweaked cells". But a matter of fact, most of Model S owners would not be happy with replacing battery pack 5 years. Let alone in 4.

Why would you make such a statement when we know that Roadster owners are seeing no such degradation after 3-4 years with off the shelf laptop cells, and no issues in cold weather? There is no drastically different chemistry in the Tesla cells than in other Panasonic cells with similar specific energy. The design changes may have been more physical than chemical in nature.

 

[Zzzz...]

Why would you make such a statement when we know that Roadster owners are seeing no such degradation after 3-4 years with off the shelf laptop cells, and no issues in cold weather? There is no drastically different chemistry in the Tesla cells than in other Panasonic cells with similar specific energy. The design changes may have been more physical than chemical in nature.

Because Roadster also uses modified/altered chemistry. Optimized for automotive use. Once again, there are many thousands parameters, trillions of trillions of possible combinations of "same chemistry".

I fought stopcrazypp and qwk trying to explain that Model S do not use NCR18650A, making no sense to use one. Nope you and others are agreed with stopcrazypp&qwk. Then come JB, say that Tesla tottally custom engineered cells... And only reply from qwk is: "huh! Tesla's total custom engineering mean slightly tweaked" and "difference would be tiny"... Like what? That is why Tesla TOTALLY custom engineered cell??? To archive TINY difference?

Nope. And now they are making equally *** claim that Roadster uses off the shelf cells... No point to argue, if direct statement from JB is not a good source for S not using NCR18650A, then trying to explain basics of batteries again with Roadster in mind is pointless too.

But do not worry - Roadster will lasts for 12+ years. Most of them.

 

[qwk]

Because Roadster also uses modified/altered chemistry. Optimized for automotive use.

This is incorrect. The roadster used Sanyo 2.2ah laptop cells.

Look, I see that you took Tesla's statement for the model S cells used as gospel, but look at all of the other statements they have made that turned out to be not true. Examples are too many to list, but include,

-supercharging cannot be added later to non supercharger hardware cars. All cars did end up with supercharging.
-85kwh pack will have different cells than 60kwh and 40 kwh packs. The cells used in all packs are the same.
-40kwh pack is now magically a 60kwh pack with software limitations.

I really wouldn't be surprised if the "performance" package is just different software, even though Tesla stated that the inverter, battery wiring is different. Tesla charges things mid-stroke very often, so taking their statements to heart is silly at best.

 

[vfx]

In the early Roadster days the battery makers were leery of Tesla's scheme. Sanyo, Panasoninc sold Tesla unmodified batteries. At some point (probably during the MS development) Tesla showed the battery makers they had a valid business and would be buying millions and millions of batteries so they could now spec them for automotive application.

 

[JRP3]

Because Roadster also uses modified/altered chemistry. Optimized for automotive use.

Not true.

Once again, there are many thousands parameters, trillions of trillions of possible combinations of "same chemistry".

Also not true. Roadster used standard LiCoO2 cells with standard ingredients. There are a limited number of components in the cells and to get a specific energy, C rate, and cycle life there are limited ways you can combine them.

 

[Zzzz...]

This is incorrect. The roadster used Sanyo 2.2ah laptop cells.

In the early Roadster days the battery makers were leery of Tesla's scheme. Sanyo, Panasoninc sold Tesla unmodified batteries.

Roadster used standard LiCoO2 cells with standard ingredients.

I see since prominent claims that S uses NCR18650A failed miserably, you guys switching to similar Sanyo/Roadster claims... The reason I was right about Tesla not using NCR18650A is some understanding of how battery actually made. I have tried to explain as to why multiple times, and the exact same reasoning apply to Roadster's cells. But I see this is pointless.

There are a limited number of components in the cells and to get a specific energy, C rate, and cycle life there are limited ways you can combine them.

Hmm, such an obvious fact that li-ion batteries are covered by dozens of thousandth of patent is clear to you?*** And that biggest part of patents apply(or could be applied) to LiCoO2? They just cover different stages of cell production. And different formulations. And in most cases those things yield cells with different characteristics, I mean you understand that, correct?

And with 17 millions cells Roadster program, Sanyo was eager to alter chemistry to Tesla's needs, to get contract. Even if Sanyo was betting that Tesla would produce only 1/20th of actual cars being made. Optimizations are not a rocket science, and very common practice in the industry.



*** For example read this: lithium_ion_battery.pdf(sorry for non-English link, check table 2-2 page 6, the most complete coverage I have found quickly. But it counts only applications under Patent Cooperation Treaty, so many are unaccounted for, filed on national level, and many relevant patents describe specific production methods of chemicals, for example to get specific particle size range or specific 3d/nano structures(think cathodes) without mentioning batteries etc etc. It is complicated.)

 

[JRP3]

17 million Roadster cells spread out over 3 years of production is not a large order, it's a drop in the bucket, and I doubt any large company such as Sanyo would do anything special at all. I don't even know if they would deal with Tesla directly for such a small order. I've spoken with a LiCo battery researcher, he posts as MRTTF on SA, and discussed this topic with him a while ago and he flatly said no way would they do anything special for such a small order. As for all the patents on LiCo how many of them were Sanyo's? Patents also aren't proof of a desirable product or component. The Model S volume production probably would be enough to get Panasonic to make some specific cells for Tesla but at this point we have no idea exactly what changes were made.

 

[qwk]

17 million Roadster cells spread out over 3 years of production is not a large order, it's a drop in the bucket, and I doubt any large company such as Sanyo would do anything special at all. I don't even know if they would deal with Tesla directly for such a small order. I've spoken with a LiCo battery researcher, he posts as MRTTF on SA, and discussed this topic with him a while ago and he flatly said no way would they do anything special for such a small order. As for all the patents on LiCo how many of them were Sanyo's? Patents also aren't proof of a desirable product or component. The Model S volume production probably would be enough to get Panasonic to make some specific cells for Tesla but at this point we have no idea exactly what changes were made.

To add further to this, if these custom cells developed for Tesla were as superior as Zzzz claims, it would also be naive to think that Panasonic would only offer this chemistry to Tesla and not sell them over the counter.

 

[Zzzz...]

17 million Roadster cells spread out over 3 years of production is not a large order, it's a drop in the bucket, and I doubt any large company such as Sanyo would do anything special at all. I don't even know if they would deal with Tesla directly for such a small order. I've spoken with a LiCo battery researcher, he posts as MRTTF on SA, and discussed this topic with him a while ago and he flatly said no way would they do anything special for such a small order. As for all the patents on LiCo how many of them were Sanyo's? Patents also aren't proof of a desirable product or component. The Model S volume production probably would be enough to get Panasonic to make some specific cells for Tesla but at this point we have no idea exactly what changes were made.

Not a large order?:scared: Seriously?!!!:cursing:

Year 2007. Number of cells produced that year worldwide is 2.45 billion units. With 17% year over year grows in year 2006 there was 2.09 billion units produced in a whole world.

Ok, in 2007 or around Tesla Motors comes to Sanyo and say, hey guys, we want to order 2500*6831 cells! Around 17.078 million cells. Sanyo surely know how many cells were produced worldwide previous year. Ahh claims Sanyo, cells number you guys ordering is only 0.82% of total cells produced worldwide in 2006... Why should we bother?

Heh, Tesla replies, look at number of kWh we are ordering! Those 2 and a change billion cells are mostly tiny batteries for crappy flip-phones for emerging market... Mostly. In terms of kWh our cells 2x-3x bigger then average cell produced in 2006!

Sanyo managers make sour faces. So your order have a size of only around 2% li-ion capacity produced worldwide in 2006? Nah, we are kinda big, you know Tesla.

Look guys, says Musk, who are you dealing with? Samsung produce it own batteries. LG too. Apple mostly orders from Taiwan(but China is in the mix too). Lenovo prefers mainland China. So who else are bigger then us? Acer, HP, Dell and Nokia?

Khmm, right on spot, reply Sanyo, but this mean that you, Tesla Motors is only fifth biggest customer in the world available to us. Not even third one! Go F yourself, you are too small and too annoying. Stop bothering us. You want our cells - go to our authorized dealer.

But guys, insist JB, look Japan holds less then 50% of the li-ion market... That bring size of our order to 4%+ of all cells produced in Japan in 2006(in kWh). And there is a good chance that we will increase our order 35 times when we will introduce Model S, in comparison to cells we need for Roadster... Please guys, do the math, multiply 4% by 35!!!

YOU DO NOT UNDERSTAND F WORD? yells Sanyo... GET LOST!!! YOU ARE TOO SMALL!!!

Tesla Motors managers leaving in confusion... Later Tesla place an order with Sanyo authorized dealer, who obviously could not sign contract about prolonged warranty, liabilities, specs variation of the cells(Sanyo could change chemistry at will). Tesla do not even have access to specs other then what is posted on internet. Sure, Tesla had to deal with third party, not directly with Sanyo.

Tesla is too small...

 

[JRP3]

Tesla did not order 2500*6831 cells in 2007. What actually happened is that Tesla, a basically unknown startup company looking to use laptop cells in an electric sports car that cost over $100,000, needed maybe 2-5 million cells that year, with no guarantee they would survive to order more. They got no special treatment and received no special cells. Tesla was too small, and completely unknown.

 

[Doug_G]

17 million Roadster cells spread out over 3 years of production is not a large order, it's a drop in the bucket, and I doubt any large company such as Sanyo would do anything special at all. I don't even know if they would deal with Tesla directly for such a small order. I've spoken with a LiCo battery researcher, he posts as MRTTF on SA, and discussed this topic with him a while ago and he flatly said no way would they do anything special for such a small order. As for all the patents on LiCo how many of them were Sanyo's? Patents also aren't proof of a desirable product or component. The Model S volume production probably would be enough to get Panasonic to make some specific cells for Tesla but at this point we have no idea exactly what changes were made.

I think they were standard chemistry, but according to http://webarchive.teslamotors.com/display_data/TeslaRoadsterBatterySystem.pdf the Roadster cells had PTC protective devices on both anode and cathode. That sounds like they may have been somewhat tailored to the application.

 

[vfx]

Tesla did not order 2500*6831 cells in 2007. What actually happened is that Tesla, a basically unknown startup company looking to use laptop cells in an electric sports car that cost over $100,000, needed maybe 2-5 million cells that year, with no guarantee they would survive to order more. They got no special treatment and received no special cells. Tesla was too small, and completely unknown.

Yeah, I remember stories from Tesla (JB?) how they were pretty ignored then.

And Tesla did a lot of testing on the commodity cells in their own "lab" so they would know how they would react to conditions. One story told how they were blowing batteries up in an engineer's back yard.
But when the Model S came along and they did place that multi B order then Panasonic did the initial testing on the new battery.

 

[Zzzz...]

Tesla did not order 2500*6831 cells in 2007. What actually happened is that Tesla, a basically unknown startup company looking to use laptop cells in an electric sports car that cost over $100,000, needed maybe 2-5 million cells that year, with no guarantee they would survive to order more. They got no special treatment and received no special cells. Tesla was too small, and completely unknown.

You see, Tesla by the end of the year 2007 already had contract with with Lotus for 2500 gliders. Not only that, but Tesla was working with Lotus to tailor/modify glider for Tesla's need for few years. Was Tesla unknown startup? More or less. But that did not prevented Lotus from working with them.

Setting glider aside, Tesla had two major components they needed, powertrain and a battery. Sure enough Tesla was working on powertrain for few years too. And it was also tailored for Tesla's needs. Again outside companies were involved, Fukuta Motors etc. Tesla again was not using off the shelf parts.

And now we come to the third major component, and a critical one. The battery. Don't you think it make sense to assume that Tesla also worked with 3d party companies to tailor cells for their needs? And worked for few years? Sure cells suppliers knew that Tesla signed contract with Lotus. For up to 2500 gliders. Would not it make sense for Tesla and cell supplier to sign similar contract, similar to Tesla/Lotus one? And we are talking about contract that at that time represented 1.5%-2% of worldwide yearly production capacity in terms of kWh. Sure, splited over three years, but nevertheless it quite huge one. Huge enough to interest top li-ion manufacturers in the world(as we could see from Sanyo participation).

And now common sense check. Could you try to place laptop outside for overnight stay at minus 25 Celsius? It is not an extreme weather. Quite normal for Alaska/Canada/Norway in winter. Would you be able to turn on laptop in the morning? And Roadsters were sold to abovementioned countries, and had to start up in the morning. Just like an ice vehicles.

And what about finding 5 years old laptop that not only got battery that alive, but also still have more then 50% of total uptime left compared to the time it was working when the laptop was new? And Roadsters do not show signs of such a huge battery degradation.

And sure Tesla got no special treatment from Sanyo and the rest, I completely agree with you. :wink: Cell optimization is a standard thing for the battery industry. Just business as usual. Moreover, while competition in sport gliders and electric powertrains markets virtually nonexistent(still there but weak), li-ion cells are well developed market with quite tough competition. With a size of an order Tesla was looking to fulfill, there were quite a few companies who were very much interested in trying to get Tesla's contract. And Tesla was looking to secure Roadster program, choosing suppliers that could fulfill volume Tesla required and at a competitive price.

PS. I mention number of li-ion patents just so you could get grasp of in how may ways li-ion cells could be tailored... But you should not try to get exponentiation of 2^n, where n is number of patents that could be applied to specific chemistry. If you add for example patented antifreeze agent to electrolyte, you can add more or less by wt%, depends on temperature you need to archive and how specific additive affect longevity of the cell, C rate, self discharge rate etc...

PPS. QUOTE: "needed maybe 2-5 million cells that year, with no guarantee they would survive to order more", exactly the same way contract with Lotus was working. Lotus had no guarantee that Tesla will survive first year. But that did not prevented Tesla from securing right to buy up to 2500 Tesla tailored gliders from Lotus at a specific price(sure that price was affected by quarterly volumes, time of orders etc). And I quite sure Tesla secured itself signing similar contract with Sanyo.

 

[vfx]

You see, Tesla by the end of the year 2007 already had contract with with Lotus for 2500 gliders. ....

Actually it was for 2000 gliders (or 3.4M batteries) The extra 500 were added when Lotus announced they were shutting down the line.

 

[Zzzz...]

Actually it was for 2000 gliders (or 3.4M batteries) The extra 500 were added when Lotus announced they were shutting down the line.

Ok, thanx for correction. I did not knew that. My math calculations should be multiplied by 0.8 then.

Btw, 2000 gliders * 6831 cells in each is equal to 13.7M cells, not 3.4M.

 

[JRP3]

Working in low volumes is standard procedure for Lotus. Not so much for 18650 producers. Your claims for the Roadster cells have no supporting historical evidence, and in fact run counter to the information we do have from Tesla at that time. I think what VFX was pointing out was that spread out over 4 years the order would have been for 3.4M cells per year, a small order. 3.4M cells per year does not get Sanyo to change it's production lines.

 

[vfx]

Actually I was saying that with 500 less cars in Tesla original glider order that would be 3.4 million less batteries in their original battery request. But JRP's math works too.

Since we lived through the Roadster development we heard over and and over that Tesla was using commodity cells from an unknown (then) manufacture. Not sure why it was so secret then but it was.

Tesla always said that they were standard cells and that they had a system for coddling them. (Doug's find is something could be attached to any chemistry cell)

If you can find any documentation that supports you claim I'd like to see it.

 

[ItsNotAboutTheMoney]

The cells used in a Roadster battery pack are referred to as "18650 form-factor" because of their measurements: 18mm in diameter by 65mm length. This form factor is a commodity in the consumer electronics market - over a billion 18650 cells are produced each year. Tesla uses versions of this form factor modified for use in EVs.vfx;322630]Actually I was saying that with 500 less cars in Tesla original glider order that would be 3.4 million less batteries in their original battery request. But JRP's math works too.

Since we lived through the Roadster development we heard over and and over that Tesla was using commodity cells from an unknown (then) manufacture. Not sure why it was so secret then but it was.

Tesla always said that they were standard cells and that they had a system for coddling them. (Doug's find is something could be attached to any chemistry cell)

If you can find any documentation that supports you claim I'd like to see it.

Roadster Technology - Battery | Tesla Motors

The cells used in a Roadster battery pack are referred to as "18650 form-factor" because of their measurements: 18mm in diameter by 65mm length. This form factor is a commodity in the consumer electronics market - over a billion 18650 cells are produced each year. Tesla uses versions of this form factor modified for use in EVs.

What does that mean?

Straubel's statement was unambiguous: the Roadster used a chemistry different to laptops. What they wouldn't have had is a bunch of collaboration to refine the chemistry or something radically different. That doesn't contradict the "commodity cell"statements, because, as Straubel stated it's the commodity form factor and underlying chemistry that allows them to get cell prices and battery designs with prices competitors can't match.