I appreciate folks trying to tak me down--I have no interest in hyping anything--but let's keep an important distinction in mind between cost reduction ($/kWh) and performance in energy density (Wh/kg). Tesla has been quite clear about the gigafactory to achieve a 30% cost reduction without any new technology. This is achievable through better economy of scale and co-location of the supply chain, lots of little gains in process efficiency. No question about this.
What startles me is that JB is talking about performance gains, that energy density specifically is continuing to advance at the rate of 40% every 5 years (= 100% in 10 years). From Roadster to Model S is 5 years. From Model S to Gigafactory (2017) is another 5 years. Will we see another 40% performance gain in this timeframe and in addition to all the blocking and tackling noted above?
What could a 40% gain in energy density do for Tesla? Take today's 60 kWh pack. It gets about 3.5 miles/kWh, or about 210 miles range. For the same weight pack, 40% more density would result in 84 kWh and a range of 294 miles. This would be even better than today's 85 kWh because the miles per kWh should be about 3.5, up from about 3.2. So the higher density pack would improve both range and energy efficiency. Maybe charge times per mile range could also come down a bit.
So much more is at stake with a gain in performance than just a cost reduction. If we could get both a 30% cost reduction through process efficiency gain AND a 40% performance gain through improved technology, then the combined reduction in $/kWh could be as much as 50% (= 100% - 70%/140%). So Musk often says he can get a 30% cost reduction without new technology. That's enough information to sell bonds, but he also says it may be possible to realize a 50% cost reduction. It seems to me that the difference between 30% and 50% could well be a timely advance in technology.
Of course, I'll be delighted just to see the Gigafactory crank out 50 GWh of packs per year at 30% below the cost today. That is enough to disrupt both auto and stationary energy storage markets. Good luck to us all.
Sorry jhm, I have to talk you down one more time. The 7-10% improvement is supposed to be inclusive of everything. So you either get cost or energy density or some mix of those, but not both at the same time multiplying together. Use the example of the Roadster - Roadster batteries cost about 40k when they came out, and they were 53kWh, and 1,000lbs, roughly. Model S batteries, 4-5 years later, cost about 40k, are 85kWh, and 1,000lbs. Roughly. So you got your 40% improvement, but it was in density rather than cost - more power, same weight, same price. The cost improvements for the car came from scale and materials. Now switching over to Model E, they're going to downsize the car, downsize the battery, and focus on saving cost. So maybe you'll end up back at 40kWh or something, costing 13k, and weighing 500lbs. Just pulling those numbers out of nowhere in particular, for example purposes, but they seem like fair rough guesses. That way you're saving ~35% but keeping about the same energy density. I mean I'm sure there will be some mix of this happening, but the battery improvement numbers aren't going to compound in the way you've described. I mean, they might, maybe battery technology will advance faster now that more people are working on it (some report said it improved by 14% over the course of a year, in like 2012 or something?), but I really don't think Tesla is going to be able to push that number so much quicker ahead than the way technology has been moving all along. Plenty of people have had plenty of motivation to make denser and smaller and cheaper battery cells for all sorts of consumer electronics for decades now. Tesla has just had the smart move of leveraging that (by using commodity 18650 cells) rather than starting fresh on their own. For what it's worth JB and Elon both have said numbers ranging from 5-10% in terms of total yearly improvement and these are essentially the numbers they use in their forecasting estimates, and where Elon is getting his "probably a bit better than 30%" forecast.
As for other people's input on this (kenliles, and someone else?), yes, Tesla certainly has a benefit right now in terms of cost and density, mainly because they went with commodity cells instead of trying to reinvent the wheel. And the process and supply chain improvements of the gigafactory will be significant, also savings in terms of supplier profit margin, but not enormous, again because the cells are commodity so the profit margin is not hugely enormous to begin with, esp. since they're partnering with Panasonic who I'm sure will want to get a taste. But basically, put all these things together, and this is why Elon is fairly confident that they can hit a bit more than 30%. Because close to 30% is already baked in by the standard march of technology (inclusive of weight, cost, whatever), which Tesla will be helping to lead of course, and then on top of that there are the additional benefits of having their own factory.
Also, yeah, the wapo article is a bit silly, I just linked it for the graph really. I think it was an old article anyway.
So anyway this battery stuff is a big deal for sure. But it's just a constant grind of technology, it's important to recognize and see how the curves are easily converging on EVs being better than gas, the same way that solar technology is converging in terms of cost per kWh, and when you see that information you are better informed than most people. But there's no real big secret to Tesla's actions here, it's just regular old good business and spotting of trends in your industry.
