How fast is lithium-ion energy density improving? What will be the energy density if 5 years vs current energy density?
Elon and JB said it many times... 8% per year on average! but will not be introduced continually but in steps.
Current cell is 2008 technology mass produced 4 years later in 2012. 2013 technology should be mass produced in 2017. 2013 cell technology should have 40% higher energy density then 2008 technology.
In other words prototype cells driving around today in test cars...
but is that cost, or cell density?
And is cell density measured per volume or per weight?
Very good questions that I would also like the answers to.
The volume of the 18650 battery cell is the same. It is an industry standard form factor. What Kleist spoke about above was 'energy density'. That refers to the amount of energy that can be stored within each individual battery cell. According to JB Straubel, lithium-ion storage capacity increases by a factor of two, in general, every ten years. He also stated that he personally witnessed an improvement of 40% in only four years, between the release of the Tesla Roadster and the Tesla Model S.
Do you remember floppy disks? No? Of course not... No one does...
Howzabout CD-ROMs? Yes? OK. The physical format of those discs is to the naked eye, identical to that of DVD-ROMs. Yet the typical CD-ROM could hold around 700 MB of data, while a DVD-ROM could hold 4.2 GB of data. Now there are BluRay discs, that to the naked eye, still look like either a CD-ROM or DVD-ROM at a glance. But those BluRay discs can hold upward of 30 GB of data. But they are all optical storage media.
It's a similar progression with lithium-ion battery cell technology. Same form factor, more storage. But of more energy, as opposed to data.
Perhaps not everyone realizes that these factors are multiplied, not added.
For example, if something's price decreases 20% every year, in five years is it free? No, it's
80% times 80% times 80% times 80% times 80%
in other words, about 33% of the original.
lithium-ion has a chemical energy density limit, it hasn't increased very much since Sony introduced the fist lithium-ion batteries
it's entirely possible that there will be a new electric technology to power EV cars much more effectively, but it's unrealistic to expect it will be lithium-ion
We understand what energy density is and that it is getting better...we're not idiots. The questions, or at least my questions, are:
How is energy density judged in EV batteries? Is it Kwh per pound? Is it Kwh per cubic foot? Is it Whr per cell?
How is cost judged in EV batteries? Is Whr per dollar? Is it cost per cell?
Let's say I have an 18650 that is rated at 12 Whr and it costs $10. If I can make next year's 18650 be rated at 24 Whr at the same cost of $10, did I just double energy density AND cut my cost in half (Whr per dollar just doubled)? Or could I only say that if I made next years 18650 24 Whr at $5?
There are several ways of increasing EFFECTIVE energy density of Model S/X. Besides the stepwise increase of energy per cell, it may be much easier and cheaper to reduce the weight and drag of the autos, and achieve the same results.
Model S weighs 4,647 lbs, and even with the 1,200 lbs battery pack, that is quite heavy. It is quite a bit over engineered and they may not have had time to optimize design. Tesla could use stronger alloys of Al and steel. Use hollow tubing and honeycombed structures to reduce weight, while meeting safety requirements. May be cut the weight by 500-1,000 lbs. The design could be further optimized to reduce drag - it does present a bulky profile.
$KW-hr of Li-ion do fall, but slowly. However, it is much easier to cut TOTAL cost of Model S, X by optimizing design, increased automation, economies of scale. Supply chains costs can drop significantly with use of commodity parts like tires, rims, glass, electronics, etc.
Tesla could cut the cost of manufacturing Model S, X by $20,000 and increase EPA range of the Model S85 by 25% without adding extra cells. Then offer an 125 KW-hr version. There are folks who will pay.
Not a fan of GF. Rather see Tesla source the manufacturing of low cost cars to somebody else. Or at the very least, buy cells from Samsung, LG Chem, etc for the low cost Model 3.
Strawberrylove declared, "lithium-ion has a chemical energy density limit, it hasn't increased very much since Sony introduced the fist lithium-ion batteries"
JB Straubel | Energy@Stanford & SLAC 2013 (30:45)
@Bubba2000 - the battery is the key technology. Tesla is well advised to be at the forefront of the technology AND controls the volume manufacturing. Batteries you can not buy in quality and quantity today, everything else you can write a PO.
You can come up with a better technology but I you lack the factories to actually make the product is just an academic exercise.
Two key numbers
(a) specific energy density measured in Wh per kg ( important for the design )
- Models S is 160 Wh / kg on pack level ( >=245 on cell level )
- typical ICE is in the low 200 Wh / kg
- e.g. i3 REX extender is only 140 Wh / kg
(b) $ per kWh ( that determines how "cheap" the car can be )
- Model S is $320 per kWh retail price
- i3 REX is $150 per kWh retail price
In order to make Model3 work two things have to happen
(1) energy density increase by at least 30%... then energy density would be about 210 Wh / kg
(2) cell cost reduction by 30%... multiply 1and 2 and you end up at 50% retail per kWh or $160 per kWh
GF is the key stone in the plan... critical to cost and critical to control of technology.
And Elon and JB laid it all out in the open many times since 2008 - for anyone willing to listen...
With increased energy density per weight, you can have a lower weight battery pack and a car with more range
@Red Sage | JULY 26, 2014:
"He also stated that he personally witnessed an improvement of 40% in only four years, between the release of the Tesla Roadster and the Tesla Model S."
"... the typical CD-ROM could hold around 700 MB of data, while a DVD-ROM could hold 4.2 GB of data. Now there are BluRay discs, that to the naked eye, still look like either a CD-ROM or DVD-ROM at a glance. But those BluRay discs can hold upward of 30 GB of data. But they are all optical storage media."
"It's a similar progression with lithium-ion battery cell technology. Same form factor, more storage. But of more energy, as opposed to data."
Tesla's progress in advancing Li-Ion battery technology is noteworthy. And I see no reason to discount their reports of continuing this advance since the MS battery details were finalized.
However, I think your analogy with optical storage media suggests a wildly optimistic view of what we should expect from Tesla's batteries during the next 10-20 years. Significant gains are likely, but I see no reason to expect any order of magnitude improvements.
On the other hand, as I see this evolving, dramatic battery improvements aren't needed for Tesla to demonstrate compelling, mass-market BEV cars in the next 10 years.
P.S. Possibly a new and better type of BEV battery may be developed in the next decade or two, making Tesla's current batteries obsolete. If that happens, I'd expect Tesla to embrace the new battery, accelerating the conversion from ICE cars to BEV cars.
There seems to be another side to the equation. That is as the potential density (storage capacity) increases, what is the time on recharging.
For example: If a Model S take 10 hours to fully charge today at 40 amps. If you double the capacity 300 to 600 miles, with the same 40 amps, it may take 20 hours to charge. Assuming everything is linear. That would leave a mere 4 hours of usage.
The math may not be 100% accurate (so the engineers don't have a field day).
The point is the law a diminishing return. There is a limit to charge capacity and time to charge vs. time to use the power stored.
That's may monkey wrench.
Here is a different perspective...
I use every day 10 kWh on average...
I can recharge at 10 kW for about 6 hrs ( TOU window )
Therefor daily usage no sweat
Going from the Bay Area to San Diego takes about 150 kWh... leaving 15 kWh as reserve
- 85 kWh battery.... I need to charge 80 kWh on the road
- 125 kWh battery... I need to pick up 40 kWh on the road
- 170 kWh battery... no need to charge on the road at all, supercharging is optional at my convenience.
A larger battery gives you more flexibility when you need to charge.
Grinnin' Ron chastised, "However, I think your analogy with optical storage media suggests a wildly optimistic view of what we should expect from Tesla's batteries during the next 10-20 years. Significant gains are likely, but I see no reason to expect any order of magnitude improvements."
I was not saying that the battery capacities will increase at the same rates as optical media storage increased. I was merely giving an example of a standardized physical shape of media does not remain static in its storage capacity as technology evolves. This was in answer to the claim that there has been no 'significant increase' in storage capacity for lithium-ion batteries since like, ever, that people keep putting into the ether. I also hope to combat the notion that weight will be an issue for a given capacity of a battery pack. And I also want to point out that Tesla Motors' costs for battery cells used per vehicle will specifically have an effect on how profitable the Tesla Model ☰ cars will be at each trim level.
I have already written at length in terms of my own predictions, extrapolations, calculations, and [WHISKEY ALPHA GOLF] regarding how I personally believe, or hope, the improvement in 18650 battery cell energy density will evolve over the next decade or so. I simply chose not to include it here, once again, because I have posted those number quite a few times already. And since I am now posting both here and at Tesla Motors Club, I may get a bit confused at times about what I posted where, when, or why... So I'm trying to employ the K.I.S.S. principle wherever I can.
So, I will now reiterate perhaps the one thing I should have written on this thread without further comment:
"According to JB Straubel, lithium-ion storage capacity increases by a factor of two, in general, every ten years."
The bottleneck would be connections and supply. As capacity increases, so does the "C" factor, in linear proportion, so that % charge per hour remains constant, while absolute kWh charge per hour increases. Any diminishing return is due to challenges with external delivery.
at the cell level panasonic is pretty much close to the theoretical limit.
at the pack level is where tesla is breaking ground and gives their 8% per year estimate.
That's cobolt -based chemistry. There are (a lot of) other. Silicon anodes for example.
@cantcurecancer, SI-based units: Wh/kg, Wh/Liter. No fuzzy definition foots, lbs etc.
I don't know if this is real or not (as it is manufactured in China and they can put anything on their labels) but I recently bought a Cree flashlight that has an 18650 li-on battery in it rated at 3.7V 6800mAh. That's like twice the capacity of the cells in the Tesla.
What hpjtv said, Panasonic already produces higher mAh 18650 cells, but Tesla is not, as far as I'm aware, using them yet. What sort of red tape is involved when Tesla upgrades the cells in their packs? New crash/fire testing and the like?
hpjtv: Websites like CandlePowerForums that review flashlights and batteries say that claims of generic lithium ion battery cells having in excess of 5,000 mAh capacity are bogus. Typically, in practice, when measured through charging and discharge, they have maybe 1/10th the capacity as claimed.
5000mAh is too much, but not that much too much. 4.0Ah is 4000mAh.
@Red Sage like I said, it's made in China and I generally don't believe in their claims. Thanks for the heads up though.