4 ms·

In that article, Murphy calculates the difficulty of doing it with lead acid batteries which seems more of a strawman than a serious attempt to estimate require

In that article, Murphy calculates the difficulty of doing it with lead acid batteries which seems more of a strawman than a serious attempt to estimate requirements for a storage technology to match up to renewable power sources.

DennisP 12y ago
He says he picked lead-acid because it's the cheapest and lead is relatively abundant. The situation isn't any better with other technologies. Even the new liquid-metal "grid-scale" batteries that have been in the news don't work out too well. A while back I ran the numbers to see; to run the world on them we'd need thousands of years' worth of our current annual production of the metals they use. Storage with hydro works pretty well but it's geographically limited. Murphy wrote another post on that. No way it can scale far enough. Storage at civilization scale is hard. I think we'll crack fusion first. If not that, then advanced fission (molten salt reactors, fast reactors).
- mohawk 12y ago
  Nobody needs can't-do engineers. So he found out that lead-acid batteries aren't enough. 1) A fleet of electric/plug-in hybrid cars are a pretty big battery. Build more of those. 2) Power-to-gas seems like a good avenue, too. It also has an existing infrastructure. http://www.eon.com/en/media/news/press-releases/2014/9/1/eon-power-to-gas-pilot-unit-falkenhagen.html http://www.eon.com/en/media/news/press-releases/2014/9/1/eon... You can also generate methane instead of hydrogen. Not super-efficient at the moment, i know. But a solvable problem. But this article originally stated we would need fusion to be able to survive without conventional fuels. Solving the storage problem of renewables just seems so much easier than achieving nuclear fusion.
  - DennisP 12y ago
    And using advanced fission seems easier than either. We need engineers attempting to solve all these problems. But what we don't need is people assuming that hard problems are already solved, and that other lines of research therefore aren't important.
    - mohawk 12y ago
      Sure, just insure your advanced fission reactors against desaster with a prudent insurer instead of relying on the government to bail you out in case anything happens. Proving something with long-tail risk is secure takes ages. There are just too many ways to store energy to go through them here, and it doesn't make sense to take any single one and claim it cannot scale to the problem at hand. I guess we'll just have to agree to disagree on the difficulty of storing energy vs. getting nuclear fusion working. I think fusion would be great to have and research is important, but i'm also glad that we don't need to get it working real soon now.
- digikata 12y ago
  I can't help but compare the investment timing & structure of fusion research vs recent renewable development. Fusion has had many years and billions of dollars of investment and it still hasn't been cracked. I hope fusion succeeds wildly. But, grid-scale storage has just recently started being a focus, and I note that it, like renewable investments, can be made on a much smaller, much more distributed scale. Payback starts when a grid-battery economically succeeds in a niche, and small successes will shift more capital and R&D into that area, and there are already some small scale successes in the grid battery market. If either area is at all physically possible, I think just because of the structure of investment/discovery/payback cycle, that grid-batteries will be available quicker. The "search function" for grid battery tech will execute much faster. Recent "small" fusion investments may change that, but I strongly suspect those small fusion projects, even if they succeed at their near term objectives, will still have many fundamental long term problems to solve, and so would be longer iteration time technologies. Regardless of how we individually assess the chances of success, civilization really needs research in both areas and more because our current civilization runs on processes that aren't sustainable long term.

DennisP 12y ago

He says he picked lead-acid because it's the cheapest and lead is relatively abundant. The situation isn't any better with other technologies. Even the new liquid-metal "grid-scale" batteries that have been in the news don't work out too well. A while back I ran the numbers to see; to run the world on them we'd need thousands of years' worth of our current annual production of the metals they use. Storage with hydro works pretty well but it's geographically limited. Murphy wrote another post on that. No way it can scale far enough. Storage at civilization scale is hard. I think we'll crack fusion first. If not that, then advanced fission (molten salt reactors, fast reactors).

mohawk 12y ago
Nobody needs can't-do engineers. So he found out that lead-acid batteries aren't enough. 1) A fleet of electric/plug-in hybrid cars are a pretty big battery. Build more of those. 2) Power-to-gas seems like a good avenue, too. It also has an existing infrastructure. http://www.eon.com/en/media/news/press-releases/2014/9/1/eon-power-to-gas-pilot-unit-falkenhagen.html http://www.eon.com/en/media/news/press-releases/2014/9/1/eon... You can also generate methane instead of hydrogen. Not super-efficient at the moment, i know. But a solvable problem. But this article originally stated we would need fusion to be able to survive without conventional fuels. Solving the storage problem of renewables just seems so much easier than achieving nuclear fusion.
- DennisP 12y ago
  And using advanced fission seems easier than either. We need engineers attempting to solve all these problems. But what we don't need is people assuming that hard problems are already solved, and that other lines of research therefore aren't important.
  - mohawk 12y ago
    Sure, just insure your advanced fission reactors against desaster with a prudent insurer instead of relying on the government to bail you out in case anything happens. Proving something with long-tail risk is secure takes ages. There are just too many ways to store energy to go through them here, and it doesn't make sense to take any single one and claim it cannot scale to the problem at hand. I guess we'll just have to agree to disagree on the difficulty of storing energy vs. getting nuclear fusion working. I think fusion would be great to have and research is important, but i'm also glad that we don't need to get it working real soon now.
digikata 12y ago
I can't help but compare the investment timing & structure of fusion research vs recent renewable development. Fusion has had many years and billions of dollars of investment and it still hasn't been cracked. I hope fusion succeeds wildly. But, grid-scale storage has just recently started being a focus, and I note that it, like renewable investments, can be made on a much smaller, much more distributed scale. Payback starts when a grid-battery economically succeeds in a niche, and small successes will shift more capital and R&D into that area, and there are already some small scale successes in the grid battery market. If either area is at all physically possible, I think just because of the structure of investment/discovery/payback cycle, that grid-batteries will be available quicker. The "search function" for grid battery tech will execute much faster. Recent "small" fusion investments may change that, but I strongly suspect those small fusion projects, even if they succeed at their near term objectives, will still have many fundamental long term problems to solve, and so would be longer iteration time technologies. Regardless of how we individually assess the chances of success, civilization really needs research in both areas and more because our current civilization runs on processes that aren't sustainable long term.