The announcement came as part of President Obama's little tour of the west to promote solar energy and other environmentally desirable projects. Google themselves have announced:
Today at an event with President Obama, we announced the Little Box Challenge, a $1 million prize to develop the next generation of power inverters.
Their site for this project is just a holding page at present so that's all the detail we really have to go on.
However, it is well known that this solar power inverter is the next part of the solar power puzzle that needs to be attacked. The inverter is the part of the system that converts the DC power from the cells themselves into the AC power that is compatible with the grid and thus all of the various electrical and electronic machinery we already have compatible with that grid power. It's a lot cheaper to use an inverter than it is to try and redesign everything else to run on DC power. There's a number of more detailed technical aspects: we'd probably like to be using a pure sine wave design for example but we can leave that to the engineers themselves.
It's the economics that really matter here. The past 10 years has seen the price of solar cells themselves plummet. This is partly because people have worked out better and cheaper ways of making the silicon ingot itself (prices have come down from $450 a kg to $20 a kg today) and partly because of something akin to Moore's Law, we just get better at tracing circuits onto silicon with each iteration of the technology. However, inverters have had nothing like that change in price: thus the cost of the inverter is becoming an ever larger part of the entire solar installation. To use entirely made up numbers just as an example, say that a solar system cost $100 a decade ago. $50 for the cells and $50 for the inverter. Now the cells cost perhaps $10 and the inverter still $50. Meaning that our whole system now costs only $60: but to gain any great further price reductions we need to get the cost of the inverter down. What happens to the price of solar cells isn't going to make much difference. Please do note that those numbers are entirely made up just to make the logical point.
So, concentrating minds on the costs of the inverters sounds like a good idea. But if Google's willing to pay a million in a prize for this then why doesn't Google just go spend a million on research? They're known for having some pretty good engineers around the place after all. The argument here is that offering a million as a prize encourages many groups to go out and spend a hundred thousand, or anything up to a million if they're being rational, on trying to win the prize. Thus the amount of capital that will be spent on trying to win the prize, overall, will be much greater than the amount of the prize itself. It wouldn't be a surprise at all to find that $10 million, or $20 million, was addressed to trying to solve this problem in the hopes of winning that $1 million prize. Sure, if any one group spent $10 million then they're guaranteeing themselves a $9 million loss. But the cumulative spending of many teams is highly likely to be a large multiple of the prize itself.
The economic effect of running this as a competition is therefore the leverage it gives to the original sum of money offered as the prize.
Given that I know absolutely nothing about the details of electrics or electronics (I'm hazy on the difference between an amp, joule and watt for example) what technology is likely to win here I'm not willing to speculate upon. Unless it's something either quantum or graphene, one of those two seems to be the solution to just about everything else at present.
