3 ms·
This article seems rather hand-wavy. I would have liked to seen more about what the physical and economic limits might be. In particular, how the cost of the so
by skybrian 4y ago
This article seems rather hand-wavy. I would have liked to seen more about what the physical and economic limits might be. In particular, how the cost of the solar panels themselves compare with all the other costs needed to get a working plant.
Is industry going to learn to make cheaper land? How much can labor realistically be reduced?
- philipkglass 4y agoSee section 5.2 of this report: "U.S. Solar Photovoltaic System and Energy Storage Cost Benchmark: Q1 2020" https://www.nrel.gov/docs/fy21osti/77324.pdf https://www.nrel.gov/docs/fy21osti/77324.pdf For a 100 megawatt utility-scale solar farm with one-axis sun tracking hardware built in the US, total costs come to $1.01 per watt-peak of direct current generating capacity. Solar modules account for $0.41 of that. Other hardware (inverters, electrical balance of system, and structural balance of system) accounts for $0.24. Installation labor and equipment is $0.11. The cost of land is tiny, too small to actually see in the stacked bar chart, but below $0.02. Solar module costs can fall by another 70% before labor costs start to become an equally relevant cost element. There are already companies working on automating additional elements of solar farm construction. You can see a brief video of AES's Atlas robot for solar farm module installation here: https://www.youtube.com/watch?v=HRFDhHa3eKY https://www.youtube.com/watch?v=HRFDhHa3eKY AES claims that with this robot, a crew of only 3 people can install a megawatt of panels in a week: https://www.aes.com/reimagining-solar https://www.aes.com/reimagining-solar You could install all the panels for a 100 megawatt solar farm like that analyzed above with 6 months and 12 people.