Of That

Brandt Redd on Education, Technology, Energy, and Trust

15 April 2011

Update: The Cost of Solar Energy

Nearly a year ago I wrote a three-part series on energy. At the time, I calculated a cost of $83.33 per gigajoule for solar power. That compares to $1.42 per gigajoule from nuclear power.

Google is investing $168 million in the Ivanpah Solar Farm in the California Desert. As far as I can tell, the total investment will be approximately $2.068 billion. It will be capable of generating 392 gross megawatts of electricity and should last at least 25 years.

In order to convert these numbers to a cost per gigajoule, we have to make some assumptions. I'll use some very generous ones. The solar array cannot generate energy at night and will only generate peak output for part of the day. Not surprisingly, the California desert location chosen for the Ivanpah project happens to be the most favorable in the entire United States. The approach used with solarvoltaics is to multiply peak output by 6 hours per day in that region. Lower numbers are used in other regions. I'll assume that the Ivanpah project is engineered to collect excess solar energy compared to its peak output and so I'm using an 8 hour multiplier instead of 6. Since a net megawatt figure isn't offered, I'll use assume 100% delivery efficiency and use the gross figure. These, of course, are unrealistically favorable assumptions.

It works out to 392 megawatts * 8 hours * 365 days *  3,600 joules/watt-hour = 4,120,704,000 megajoules/year or 4,120,704 gigajoules per year.

Assuming a lifetime of 25 years, construction cost of $2.068 billion and no maintenance costs we get $2.068 billion / (25 years * 4,120,704 gigajoules / year)  = $20.07 per gigajoule. That's an improvement of four times over my previous calculation for solar power. It starts to approach the $13.89 per gigajoule cost of wind power.

It's a huge improvement over solar panels but this still remains the most expensive way in the world to generate electricity. It's an order of magnitude more expensive than conventional energy sources which have the added advantage of delivering power 24 hours a day regardless of the weather.

I'm glad to see this happening but it won't spark a revolution in energy production.

1 comment:

  1. It frustrates me that so many energy decisions seem to be made in a political and emotional way rather than an informed scientific way. Thanks for doing the math!