A new 2x battery technology seems unlikely. There's nothing that good coming on line within three years. Panasonic is already setting up Tesla's battery cell plant, so that decision has been made. (Panasonic makes the cells, Tesla packages them; many of the employees at Tesla's "superfactory" work for Panasonic.)
Quoted from http://waitbutwhy.com/2015/06/how-tesla-will-change-your-lif... :
Right now, Musk says Tesla could make their cars with a 500 mile range—they don’t do it because it would increase the cost of the car.
How can they pack more cells into the already-crammed, heavy underbody battery pack, unless they also raise battery efficiency?...
Perhaps Tesla is transitioning into a battery manufacturer that makes cars on the side.
Aerodynamic resistance is key, and it goes up with the square of your speed, so driving slowly lets you go much farther. If you want to set a distance record in an EV, you want to drive around 25-30MPH.
The current record in a Tesla is over 500 miles, while the official range is under 300. Musk's comment was about how ongoing small battery improvements will allow the record distance, driven at 25-30MPH, to break 1000km in the next couple of years.
But all the idiot news organizations out there either misunderstood it entirely or saw it as an opportunity to drum up page views, and are either outright presenting it as a practical range, or are at least not trying very hard to present the real context, as is happening with this article.
i'd say giving the amount of money on the table, not only 2x, more like 10x is very likely. If you look at the actual amount of the lithium ions and actually participating oxidizer in the battery - it is 5% at best out of the total battery weight. So even modest progress at the decreasing of the weight of the "non-active" components would lead to multiple "x" of capacity/weight improve. And this is without even getting into metal-air territory.
http://batteryuniversity.com/learn/article/battery_statistic...
(I'd be interested to see a graph that includes the last ten years....)
I'd call that a considerable improvement, albeit nowhere near 10x.
It would be fantastic if this was a problem where solutions are accelerated significantly (ie. 10x jump in a few years) by throwing more money at it, but I am not sure it's the case.
Here is a relevant Quora thread:
https://www.quora.com/Is-it-true-that-battery-energy-density...
Compare to a similarly magic CPU. That would be cool, but it wouldn't change the world. (Until somebody figures out how to run superhuman AI on it, anyway.)
1MWh is 3600000000 or 3.6GJ. 3.6GJ/L is twice the energy density of Uranium, and 3.6GJ/kg is 40 times the specific energy (energy/mass). We wouldn't need a power grid anymore because we could just charge batteries on the moon and lob them to earth.
If we're in dreamland, we can just think about what'd happen with instantaneous subspace energy transmission. We could get rid of batteries and electrical grids altogether and we'd just turn the moon, mars and mercury into giant solar arrays. I'd wager that's more likely to happen than a 1L/1kg 3.6GJ battery too.
>1MWh is 3600000000 or 3.6GJ. 3.6GJ/L is twice the energy density of Uranium, and 3.6GJ/kg is 40 times the specific energy (energy/mass). We wouldn't need a power grid anymore because we could just charge batteries on the moon and lob them to earth.
well, these densities is fusion energy. No need to charge at all. Just scoop up some He-3 on the moon :)
The problem with reducing the mass of the non-Li portion of the battery is that all of the practical candidates for the anode are metal oxides (e.g. Co, Fe, Mn oxides). Unless the whole metal oxide motif is replaced, you can't really achieve significant mass savings. There are research groups and startups working on alternate anode and cathode materials but there are tradeoffs in power density and safety. Also given that these approaches are still in the research phase, it'll take at least a couple of years before they reach commercial testing.
On my EV, I can drastically change the battery's range (from 80 to 120 predicted miles) based on driving behavior, much of which could be "shaped" by software. "Softwareing" 50% more distance out of that stored energy matters.
There's more potential for various optimization in an internal combustion engine.
EVs have been designed from the get go with most of the possible software optimizations...
Reducing weight is a very difficult issue here, all manufacturers try to reduce the weight of their cars and lugging a thousand pounds of batteries doesn't help. Frankly I think all electrics are not ready for prime time, improved REX solutions are better. Traditional engines perform badly accelerating from a stop but can be very efficient one speed improves. So some combination similar to the Volt but with more base EV is probably going to be a sweet spot. (figure 100 EV/300REX).
Plus this doesn't even touch the elephant in the room, long distance trucking where weight is always an issue and range. Neither of which favor batteries.
I would much prefer Elon solve a more constant problem that would start the next generation off on the right track, Tesla needs to lead the industry in electrifying school buses.