[identity profile] anton-p-nym.livejournal.com 2010-08-16 02:51 pm (UTC)(link)
It would be nice to think we could power everything with windmills and solar panels but that's about 30 years down the road.

I don't think that's down the road at all, honestly, if we're going to maintain a civilization. The problem with wind and solar is their unpredictability and uncontrollability... you could get a reasonably-accurate average over a big enough area, if you can solve the transport problem, or maybe store excess if superbatteries or other storage media finally prove out, but you can't turn up the Sun if you suddenly need more juice.

The absolute maximum you can get out of solar power is ~1kW/m^2 anyway, which assumes perfect conversion of all incoming sunlight into electricity and ignores cloud cover and dust. That's a hard limit; that's how brightly the Sun shines, less the amount soaked up by the atmosphere itself. If you knock off about 25% for inefficiencies (which is absurdly optimistic; current consumer-grade photovoltaics lose about 80%, while the best commercial and military pVs lose "only" about 75%) we can call it 1 horsepower per square meter.

A Kawasaki Ninja ZX-14 motorcycle has a 190hp engine. To power it with our magic photocells we need a square solar power array about 14m to a side. The bike itself is only 2m long... so that's one heckuva big parasol, and remember to never, every go into the shade and keep up with the wash-and-wax.

Just an illustration of the limits of the method that technology cannot surpass; solar needs huge land areas to generate power at the levels we take for granted today. Relying on it (and wind, though the math on that is far more fraught) would lead to a huge drop in standard of living even if we increase efficiencies to as-yet unattained levels.

-- Steve thinks we'll need throttleable power supplies to make up the gap for a lot longer than 30 years, and maybe indefinitely.