5 ms·
Article is 15 years old, energy and labor prices have increased a lot. We need a new cost estimate.
by labster 3y ago
Article is 15 years old, energy and labor prices have increased a lot. We need a new cost estimate.
- pstuart 3y agoSolar prices have plummeted, and that use case can skip storage issues (just run when there's power available). Obviously land, labor, and infra cost are not so simple.
- nerdawson 3y agoIf you use solar to generate electricity then the cost of that electricity is whatever you’re missing out on by not selling it to the grid.
- fbdab103 3y agoAs a consumer, falling solar prices have not led to a noticeable decrease in my electrical bill.
- cronix 3y agoAs a recent convert to solar, we no longer have any electric bill, period.
- ourmandave 3y agoThey haven't tried to bill you for "maintaining the grid" cost?
- markdown 3y agoNo, but hopefully they do soon, or tax me somehow. Having a grid is important.
- dpierce9 3y agoWhat do you think the price would be without solar? The answer: higher.
- redeeman 3y agosure, but only because they purposely exclude those that would be cheaper :)
- etimberg 3y agoAt some point the price of the bill is dominated by the fixed transmission costs which won't go away as long as there is a power grid.
- ehnto 3y agoDesal plants need to be running constantly and incur untennable startup costs otherwise, if I am not mistaken. Perhaps state of the art has fixed that issue.
- scythe 3y agoDesalination seems to be well-positioned to take advantage of variable energy prices, particularly in dry-summer climates where demand for water increases at the same time as solar energy availability. If you run the desalination system when energy is cheapest, your energy storage problem becomes a water storage problem, but a simpler one than with pumped-hydroelectric storage. I haven't seen a detailed analysis, but this could look good for California, notwithstanding the other problems with desalination (ecological).
- 10u152 3y agoAbsolutely, but it's a scale problem. Say you want to produce 1000m3 a day. Well you can build a small plant that makes 41m3 ish an hour and run it 24/7. If you want to produce 1000m3 a day in the three or four hours where solar energy is very cheap you need a plant that does 250m3 an hour - or about 5x the size. So your energy is cheaper but your capital cost is 5x or so.
- scythe 3y agoIf you save enough on energy, you can switch your system to save on capital. RO has a fancy nanostructured membrane but electrodialysis or MSF might be easier to build at the cost of 20-50% worse efficiency.
- gonzo41 3y agoIf you build a few batteries next to the plant and that problem is solved.
- pstuart 3y agoAnother thing I'm curious about -- Dean Kamen's "Slingshot". Seems like we'd be well served if his vision was realized: https://www.slingshotdoc.com/the-film https://www.slingshotdoc.com/the-film