Editor's Note: Scientific American 's George Musser will be chronicling his experiences installing solar panels in Solar at Home (formerly 60-Second Solar). Read his introduction here and see all posts here.
I had a fun talk yesterday afternoon with Bob MacDonald, the CEO of Skyline Solar, makers of a new concentrated photovoltaic (CPV) array. The thing looks rather like a big solar cooker, with a long mirror that focuses sunlight so that you only need a tenth as many solar cells to cover a given area. CPV may become the first photovoltaic technology to reach cost parity with fossil fuels.
The basic idea, which goes back to the '70s, is to use fewer solar cells and shine more light on each one. A cell's rated output is based on straight-on, full-on sunlight, which is about 1,000 watts per square meter at Earth's surface. In light that is hundreds or even thousands of times brighter, the cell will generate proportionally more current and therefore proportionally more power. (The voltage remains fixed by quantum physics.) Indeed, some types of cells, such as those that capture a wider range of the solar spectrum, perform best under intense light.
The main tradeoff is that the cells then absorb more heat. The Skyline array relies on passive cooling -- namely, natural air flow around metal fins mounted on the back of the cells. You also have to factor in the cost of the mirror and, if the panels track the sun, motors. For small residential systems, regular flat panels are still cheaper, but CPV becomes economical for systems bigger than about 50 kW, such as those that are starting to show up on the roofs of shopping malls. For truly gigantic arrays, those bigger than about 50 MW, the cost of silicon begins to add up and solar thermal systems, which heat up a fluid to spin a turbine, become cheaper.
One of CPV's hangups has been the cost of manufacturing and installation, but MacDonald says his company has come up with a simplified mechanical design that can be built on a converted car assembly line. The units are sized to fit into a standard shipping container and are fully preassembled to make them easy to plug in at the site. A 1 MW array will consume about five acres. The company set up a 27 kW demo system in San Jose last May and expects to start shipping its production units later this year.
Initially, MacDonald says, they will cost about 15 cents per kW-hr, about half the cost of regular solar panels. If so, it is already competitive with nuclear power and closing in on fossil power, which runs about 10 cents per kW-hr (varying with location), not counting its environmental costs.
CPV exempifies how the challenge of solar power these days is not the high-tech lab work but the low-tech, nuts-and-bolts cost-cutting. MacDonald's description of how arrays must be tailored for their site also made me appreciate how the two ends of the solar market are moving in opposite directions. For homeowners, systems are becoming more standardized to reduce installation costs, make it easier to get permits, and allow DIYers to put up panels on their own. But large farms of solar panels are becoming less standardized. Though their components may be plug and play, their overall design needs to be customized to squeeze out every last watt. "There's a lot of benefit to be had by purpose engineering," MacDonald says.