Given that California and Germany, most notably, have offered large-scale subsidies for a few years now, I thought it was worth a look at how these cost reductions were coming along.

The graph below shows the cost of power from a very large (500 kw peak output) solar panel array in the United States, in July 2000 US dollars:

Unsubsidized cost of large scale (~500 kw peak) solar installations in the USA, in constant July 2000 USD. Source: Solarbuzz.com, with US inflation data from the Bureau of Labor Statistics

Large scale solar PV is likely to be cost-competitive much earlier than domestic roof-scale installations, the USA tends to be one of the cheaper places in the developed world to do any construction work (thank you, underclass), and California (where most of the US installs are happening) is a sunny place, so this is about as favourable as you’re going to get for solar panels.

I’m no expert in analyzing time series data, so I calculated a simple exponential trendline, which gives a rather nice fit (so does a linear one, given the data). The conclusion? Solar power (at least in the circumstances identified) is coming down in price at about 3.1% a year, in real terms. It’s not too far off the answer that the US DOE got, with reductions of around 3.5% per year over the period 1998-2007.

Of course, perhaps the biggest mistake in predicting the future is relying on simple extrapolation from the recent past. But, if that rate is maintained, it’ll be quite a while before the costs approach that of the alternatives.

Meanwhile, in the renewables world, one of the major limitations of wind turbines is that they only operate efficiently in a very narrow range of wind speeds, meaning much of the time they produce very little power, requiring expensive energy storage systems or backup generation to support them if they become a substantial fraction of an energy grid. There are already a number of ways to reduce this impact – turbine blades that adjust their pitch, mechanical gearboxes, and, most recently, fitting multiple generators to wind turbines, and only engaging some subset of them when the wind is weak. MIT Technology Review reports on a company developing a new approach – an a new generator design. Essentially, generators use the rotating shaft to move magnets past coils of wire to produce electricity. The new design has many independent coils, which can be switched on or off by electronic switches. If the wind is low, only a few coils are connected, allowing the blades to keep turning and generating at least some electriity. As wind speed increases, more coils can be enabled, requiring more energy to turn the blades, and thus generating more energy while the blades continue to operate at similar speeds. According to the company’s simulations, the new generators could increase annual power output by 50 to 100%, depending on the site – obviously, the more variable the winds at a site, the bigger the improvement. Now all they need is to actually deploy a prototype, scheduled for “early next year”. Sounds very promising, but as usual be prepared for a big gap between the press release and the reality.

Finally, the BP Solar cell factory in Sydney is closing because it’s too small to compete against plants overseas, and it’s too far away from its suppliers of things like ultrapure silicon. The government isn’t raising a finger to save this manufacturing facility. You might very well ask why the car industry deserves support and the solar cell industry doesn’t.