The Oil Drum: Europe

Pondlife, A number of points:-
Photovoltaic generation has been around a long time. I remember reading a book about it as a boy 45 years ago. Most of the basic patents will have long since expired. However as in all complex industrial processes there are always a host of patents on minor aspects. Photovoltaic production is fairly simple technology compared to semiconductors. Doing it cheaply is the difficult part. It is no more poisonous and energy intensive that many other industries

We don't have to go to Japan, there is plenty of production in Germany which is now installing more capacity than the rest of the world put together, over 850MW in 2005 out of a world installation of over 1480MW from crystalline silicon and another 176MW by other means (the UK managed 4MW). European producers have about 26% of the market and the Americans 12%

We cannot produce tonnes of the stuff? According to this report in 2005 10,000 tonnes of silicon was produced for photovoltaic use (out of 30,000 tonnes of semiconductor silicon) and another 7,000 tonnes was used from stockpiles from previous years. This year production will increase 30% to 13,000 tonnes but there is nothing left in stock. This will reduce the amount of silicon available and increase the price of polysilicon still further. It has a already increased it from $30/kg to $60/kg. Fortunately cells are being made with progressively less silicon and output is expected to rise this year to 1738MW rated capacity.

However it is estimated that soon the production squeeze will have eased and by 2009 photovoltaic use will have overtaken electronics and be using 27,000 tonnes of polysilicon a year to produce 3400MW of rated generation which with other photovoltaic will bring the total for the year to 4100MW. This will bring total world installed rated capacity to about 15.6GW.

Tracking the sun ideally  gives only  a 38% increase over an optimally angled fixed panel at UK latitudes and given that the UK tends to be more cloudy morning and evening it is probably less than that. A multi-section tracker on a roof with sections paced to avoid self shadowing will not give more than a fixed panel covering the whole of same roof area. The only advantage is that the total area of the spaced out trackers will be less than the continuous fixed panels. I doubt the gain is worth the cost of the scanners and the reliability problems they might cause. Interestingly the book that I mentioned above and that has stuck in mind suggested building houses that float in a circular pond so that the whole house could turn to track the sun. I wondered then and I wonder still how the sewerage would connect in such a system.

I strongly suspect Chris is right when he suggests that the 20 year live of photovoltaic panels is a cautious under estimate. I have had my system for over 2 years and have monitored it closely. There has been no drop in output at all and I think I could detect a 1% drop off. Space borne systems have survived many years in a vastly more hostile environment than UK roofs.

You say that gas central heating is a problem in the UK. I don't know of many houses that fell down because of such installations. Gas heating is still a lot cheaper than electricity and likely to remain so for many years. As long as most electricity is generated from fossil fuels (and this is not going to change much in the next 15 years) the price of gas and electricity are going to go up together. Domestic combined heat and power is the way to go if you are going to use fossil fuel.