Solar Power

I started typing this in reply to an e-mail question, but realized that it was going to be too long and interesting to use it for such a temporary message. This is partly out of inspiration, partly out of curiosity, and partly out of sharing the message with others. This began in response to a question about using solar power to melt metal, perhaps for recycling.

Why go to the trouble of building a solar array (in the megawatt range, of course; might as well be economical about it!) just for melting metal, when you can install a steam turbine and generate and sell electricity? But even then, few such arrays have been built.

Regardless of the use, two things matter to utilizing solar energy: how much power can you grab, and at what temperature (and pertinent to any heat engine, what temperature difference)?

Power is simple: a perfect black body can pick up about 1kW/m^2, so you need as much area as you want power. You do, of course, want a lattitude where the sun shines substantially all year round, and a climate where clouds will not interfere. The U.S. southwest, the Sahara desert (at least where dunes don't roam) and much of Australia are prominent possibilities, which if utilized, could together generate more power than the human race currently knows what to do with at the moment.

Temperature is somewhat harder. A black sheet laying on the ground will only reach such-and-such a temperature, because the heat input is balanced by radiation to the atmosphere or space (which averages substantially cooler than the sun), conduction to the ground below it (which had already been in equilibrium with its now absent radiation input) and convection into the air, especially if the wind is blowing. To reach higher temperatures necessary to generate steam or melt salts or metal, some focusing is needed. Lenses are the archetypal "ant burner", but very expensive when several kilometers are needed. They also cost some efficiency due to reflection and absorbtion; why waste power? So, reflectors are required.

I'm partial to the design where black pipe, carrying oil, sits in the focus of a cylindrical reflector. The oil is pumped through the array where it is heated to a substantial temperature (perhaps near its boiling point; not too hot, or else decomposition will ensue, or else much more expensive silicone oil must be bought and used), then cooled into some water, which is made to boil. A multistage steam turbine then captures the power as usual. This is a typical layout for coal, oil or nuclear power plants. Practical considerations are the power used to pump the liquids, cleaning for the reflectors when desert dust builds up too much (a major concern for the two rovers on Mars), movement to track the sun during the day and year, and of course, materials.

Solar sounds great, but is it? Consider a square kilometer area for the collectors. I would estimate a reflector, pipe, actuator and support system might weigh 25 pounds per foot. This is about the weight of a refrigerator, but much deeper -- the reflectors might be, say, four or six feet across. They will, of course, be tightly spaced, giving little room for precious sunlight to reach the ground. Therefore, it will weigh about 80 pounds per meter, or 80k = 40 tons per kilometer. If the reflectors are two meters across, it will take 500 rows, or 500 km total length, weighing 20,000 tons. Including fabrication, the materials (mostly mild steel mill products like angle iron and pipe) will probably cost around $1 to $10/lb, or $40 to $400 megabucks. One square kilometer is a million square meters, so it might produce about a mega-kilowatt, or a gigawatt. This, so far, is a pretty reasonable price for a power plant, and incurs no apparent fuel costs, making for very cheap electricity (or excellent profit, which is more likely, as hydroelectric installations have shown...).

I can't help but think there must be some horrible maintainance cost, or power cost (perhaps in pumping the fluids), but I'm not seeing it. In my estimation, solar power is competetive with other current technologies.



Return to Index


Web page maintained by Tim Williams. All rights reserved.