Fusion, Shmusion
December 8, 2021 9:11 PM Subscribe
Read the papers and it seems like a fusion power plant is imminent—we’re just a few years away from our first fusion generator and then to widespread commercialization. After all, that’s what the fusion companies themselves are saying.
Tokamak Energy: A working power plant connected to the grid by 2030 • General Fusion: Demonstration power plant beginning operations in 2025 • Helion Energy: We’ll do it in 2024 • First Light Fusion: Yeah, 2024 • Zap Energy: 2023—so there!
But if we’ve learned anything from the Theranos debacle, it’s that we can’t take any company’s claim at face value when “fake it till you make it” is a standard corporate motto. (Slate)
I have a friend who made a bet twenty years ago that fusion reactors will not be deployed by 2030. It's looking like a pretty strong bet.
There are reasons other than reputation that nuclear plants aren't attractive (their use in building nuclear weapons is a big issue); while some of them are solvable, the biggest problem in 2021 is time. Nuclear plants take years to build, the ones that more efficiently use spent uranium or fresh thorium even more so, and we need to decarbonise faster than that.
posted by Merus at 10:02 PM on December 8, 2021 [8 favorites]
There are reasons other than reputation that nuclear plants aren't attractive (their use in building nuclear weapons is a big issue); while some of them are solvable, the biggest problem in 2021 is time. Nuclear plants take years to build, the ones that more efficiently use spent uranium or fresh thorium even more so, and we need to decarbonise faster than that.
posted by Merus at 10:02 PM on December 8, 2021 [8 favorites]
Incidentally, "D. T. and the Hot Neutrons" is the name of my new metal-fusion band. Our hit single is "Radiator Of Love".
"I'm A-Goin' Nucular Fishin'" is the B-side.
posted by Greg_Ace at 10:11 PM on December 8, 2021 [23 favorites]
"I'm A-Goin' Nucular Fishin'" is the B-side.
posted by Greg_Ace at 10:11 PM on December 8, 2021 [23 favorites]
Even the most promising fusion experiments that claim near “break-even” yields have no story for how that energy could be practically extracted. We’re very far from feasible fusion power.
posted by sjswitzer at 10:15 PM on December 8, 2021 [4 favorites]
posted by sjswitzer at 10:15 PM on December 8, 2021 [4 favorites]
Helion's plan is interesting - deuterium and helium-3 fusion. From their FAQ:
• "Helion produces helium-3 by fusing deuterium in its plasma accelerator utilizing a patented high-efficiency closed-fuel cycle"
• "Our device directly recaptures electricity; it does not use heat to create steam to turn a turbine, nor does it require the immense energy input of cryogenic superconducting magnets... This change in magnetic flux interacts with the magnets around the machine, increasing their magnetic flux, initiating a flow of newly generated electricity through the coils."
posted by ShooBoo at 10:24 PM on December 8, 2021 [1 favorite]
• "Helion produces helium-3 by fusing deuterium in its plasma accelerator utilizing a patented high-efficiency closed-fuel cycle"
• "Our device directly recaptures electricity; it does not use heat to create steam to turn a turbine, nor does it require the immense energy input of cryogenic superconducting magnets... This change in magnetic flux interacts with the magnets around the machine, increasing their magnetic flux, initiating a flow of newly generated electricity through the coils."
posted by ShooBoo at 10:24 PM on December 8, 2021 [1 favorite]
no story for how that energy could be practically extracted.
Yeah, how do they heat the water up on those designs?
On post: oh, I see, harnessing magnetic flux is interesting
posted by DeepSeaHaggis at 11:04 PM on December 8, 2021 [1 favorite]
Yeah, how do they heat the water up on those designs?
On post: oh, I see, harnessing magnetic flux is interesting
posted by DeepSeaHaggis at 11:04 PM on December 8, 2021 [1 favorite]
the biggest problem in 2021 is time
I hear this all the time, and yes ideally we would have started building 20 years ago and now short-term decarbonisation will have to be done with solar and wind, batteries and hydrogen (and don't think there aren't resource and manufacturing limits and environmental harms associated with those technologies). We're going to be hair-shirting it for a while if we want to meet our climate change goals, but we can do more than one thing at once. Nuclear we build (or even research) now will be useful in the 2040-2060+ time-frame as we start to work on extremely energy intensive projects like atmospheric carbon capture (and as hopefully increased global energy equality drives substantially increase base demand).
Deciding we are going to just limit ourselves forever to essentially one energy source (ie solar: wind, hydro, and (currently practical) geothermal don't scale to much more than current total demand) is inherently fragile.
That said, there is another limit at around 1000x current usage, where waste heat just from energy consumption (as opposed to greenhouse forcing) becomes significant in the climate, but even on current trends that's far enough away that I'm not worrying about it :)
posted by memetoclast at 12:32 AM on December 9, 2021 [7 favorites]
I hear this all the time, and yes ideally we would have started building 20 years ago and now short-term decarbonisation will have to be done with solar and wind, batteries and hydrogen (and don't think there aren't resource and manufacturing limits and environmental harms associated with those technologies). We're going to be hair-shirting it for a while if we want to meet our climate change goals, but we can do more than one thing at once. Nuclear we build (or even research) now will be useful in the 2040-2060+ time-frame as we start to work on extremely energy intensive projects like atmospheric carbon capture (and as hopefully increased global energy equality drives substantially increase base demand).
Deciding we are going to just limit ourselves forever to essentially one energy source (ie solar: wind, hydro, and (currently practical) geothermal don't scale to much more than current total demand) is inherently fragile.
That said, there is another limit at around 1000x current usage, where waste heat just from energy consumption (as opposed to greenhouse forcing) becomes significant in the climate, but even on current trends that's far enough away that I'm not worrying about it :)
posted by memetoclast at 12:32 AM on December 9, 2021 [7 favorites]
radioactive waste in the form of irradiated infrastructure
Neutron activated steel is much less of a hazard than many of the things produced by fission reactors, being less mobile. Plus you have the exact same issue with the fission reactor vessel itself and all the parts inside it.
posted by wierdo at 12:46 AM on December 9, 2021 [4 favorites]
Neutron activated steel is much less of a hazard than many of the things produced by fission reactors, being less mobile. Plus you have the exact same issue with the fission reactor vessel itself and all the parts inside it.
posted by wierdo at 12:46 AM on December 9, 2021 [4 favorites]
Fusion reactors: Not what they're cracked up to be
posted by eye of newt at 1:04 AM on December 9, 2021 [1 favorite]
posted by eye of newt at 1:04 AM on December 9, 2021 [1 favorite]
we’re just a few years away from our first fusion generator
It's reassuring to know that, after all those years, this is still the case.
posted by acb at 1:24 AM on December 9, 2021 [33 favorites]
It's reassuring to know that, after all those years, this is still the case.
posted by acb at 1:24 AM on December 9, 2021 [33 favorites]
My joke about these has always been something along the lines of "Fusion was twenty years away twenty five years ago, now it's only ten years away. Soon it will be only five."
Have we ever done neutron-neutron fusion? I'm not looking for break even or anything, just the possibility of doing it on earth. Part of the appeal /promise of fusion was that it would eliminate nuclear waste associated with fission power plants.
posted by Hactar at 1:36 AM on December 9, 2021
Have we ever done neutron-neutron fusion? I'm not looking for break even or anything, just the possibility of doing it on earth. Part of the appeal /promise of fusion was that it would eliminate nuclear waste associated with fission power plants.
posted by Hactar at 1:36 AM on December 9, 2021
Neutron activated steel is much less of a hazard than many of the things produced by fission reactors, being less mobile.
You also have the irradiation-activated heat-transfer loop fluid, the irradiation-activated (and potentially tritiated) confinement lining, and the irradiation-activated interstitial fluids (air, probably) inbetween. With the ITER model the idea is to breed tritium from lithium blankets outside the confinement, so the neutron flux is clearly substantial...
Other models (polywell, helicon as mentioned above) are neat except that there's exactly no evidence they'll work (and here we're back to the corporate-mendacity thread, which is maybe more interesting that I thought). I'm not against fusion _research_, I'm against the idea that fusion is some magical non-nuclear nuclear and if we just wait for it all our energy problems will be solved. Breeder reactors have some problems (proliferation (although I think that's overblown - states that want nuclear weapons seem to be quite able to acquire them regardless), waste (although look where fossil-fuel waste has got us)) , but the dominant anthropogenic source of atmospheric radionuclides is currently coal-fired power stations.
posted by memetoclast at 1:48 AM on December 9, 2021 [4 favorites]
You also have the irradiation-activated heat-transfer loop fluid, the irradiation-activated (and potentially tritiated) confinement lining, and the irradiation-activated interstitial fluids (air, probably) inbetween. With the ITER model the idea is to breed tritium from lithium blankets outside the confinement, so the neutron flux is clearly substantial...
Other models (polywell, helicon as mentioned above) are neat except that there's exactly no evidence they'll work (and here we're back to the corporate-mendacity thread, which is maybe more interesting that I thought). I'm not against fusion _research_, I'm against the idea that fusion is some magical non-nuclear nuclear and if we just wait for it all our energy problems will be solved. Breeder reactors have some problems (proliferation (although I think that's overblown - states that want nuclear weapons seem to be quite able to acquire them regardless), waste (although look where fossil-fuel waste has got us)) , but the dominant anthropogenic source of atmospheric radionuclides is currently coal-fired power stations.
posted by memetoclast at 1:48 AM on December 9, 2021 [4 favorites]
Radioactive waste from fusion reactors is not a problem but damage to the first wall, structure, magnets, and other parts of the reactor is, given the capital intensity of the reactor a serious obstacle.
I'm not sure why the author has included SPARC which is a conventional Tokamak design with superconducting magnets coming out of an academic lab in a list of ideas which are a lot more conceptual.
posted by atrazine at 2:21 AM on December 9, 2021
I'm not sure why the author has included SPARC which is a conventional Tokamak design with superconducting magnets coming out of an academic lab in a list of ideas which are a lot more conceptual.
posted by atrazine at 2:21 AM on December 9, 2021
It's reassuring to know that, after all those years, this is still the case.
I am pretty sure that commercial energy from fusion plants has been “just 10 years away” for my entire life, and one of my high school chemistry teachers said something similar to my teenaged self, so that covers the entire history of fusion research.
posted by GenjiandProust at 2:35 AM on December 9, 2021 [2 favorites]
I am pretty sure that commercial energy from fusion plants has been “just 10 years away” for my entire life, and one of my high school chemistry teachers said something similar to my teenaged self, so that covers the entire history of fusion research.
posted by GenjiandProust at 2:35 AM on December 9, 2021 [2 favorites]
sjswitzer: Even the most promising fusion experiments that claim near “break-even” yields
Sabine Hossenfelder made an interesting video recently about what "break even" means and what it actually means. Short version: When projects report that they're near or at "break even", they're talking strictly about the power that goes into the plasma and the power that comes out of it. The power needed to run the whole plant is an order of magnitude greater than that, which means that they're an order of magnitude away from true break-even.
posted by clawsoon at 4:34 AM on December 9, 2021 [10 favorites]
Sabine Hossenfelder made an interesting video recently about what "break even" means and what it actually means. Short version: When projects report that they're near or at "break even", they're talking strictly about the power that goes into the plasma and the power that comes out of it. The power needed to run the whole plant is an order of magnitude greater than that, which means that they're an order of magnitude away from true break-even.
posted by clawsoon at 4:34 AM on December 9, 2021 [10 favorites]
The funny thing about that perennial joke is that way back when the US Government did a study looking at how long it would take to develop a workable fusion reactor. It said it could be done in 20 years for $x per year, in 40 for $y per year, and that with less than $z a year it would never get done. We chose the latter, so it's never happened.
In some ways that's actually a good thing. Materials science has improved dramatically since the 70s. You can buy superconductors that would have seemed impossible back then off the shelf. Advances in computation have vastly improved our ability to model plasma instabilities. Together, that means that if we choose to invest the necessary money the thing we'll get will be far better than what we could have had. More easily constructable, more efficient, and more reliable.
It's just too bad ITER was designed so long ago that no matter how well it proves the basic science people are going to see it as a boondoggle. Never mind that our governments all made the active choice to make it take over 20 years to go from design to operation. The fact that it can't generate enough energy from the plasma to generate a significant amount of power is already being used to shit on the entire concept of fusion power despite the fact that we already know how to do much better.
Ultimately, fusion power's biggest handicap is that it doesn't have any obvious military purpose. Fission became widespread because you can run a submarine or an aircraft carrier with it, so the money was spent to make it work. And then, having been proven, it was a lot easier to get money to scale it up. Until the last decade or so, fusion could only work at the same scale as the largest fission plants, so it never got that kind of push.
posted by wierdo at 4:35 AM on December 9, 2021 [2 favorites]
In some ways that's actually a good thing. Materials science has improved dramatically since the 70s. You can buy superconductors that would have seemed impossible back then off the shelf. Advances in computation have vastly improved our ability to model plasma instabilities. Together, that means that if we choose to invest the necessary money the thing we'll get will be far better than what we could have had. More easily constructable, more efficient, and more reliable.
It's just too bad ITER was designed so long ago that no matter how well it proves the basic science people are going to see it as a boondoggle. Never mind that our governments all made the active choice to make it take over 20 years to go from design to operation. The fact that it can't generate enough energy from the plasma to generate a significant amount of power is already being used to shit on the entire concept of fusion power despite the fact that we already know how to do much better.
Ultimately, fusion power's biggest handicap is that it doesn't have any obvious military purpose. Fission became widespread because you can run a submarine or an aircraft carrier with it, so the money was spent to make it work. And then, having been proven, it was a lot easier to get money to scale it up. Until the last decade or so, fusion could only work at the same scale as the largest fission plants, so it never got that kind of push.
posted by wierdo at 4:35 AM on December 9, 2021 [2 favorites]
what if we use blockchain
posted by glonous keming at 4:59 AM on December 9, 2021 [51 favorites]
posted by glonous keming at 4:59 AM on December 9, 2021 [51 favorites]
In the middle of The Star Builders by Arthur Turrell, a physicist that is not currently in the fusion industry so at least seemingly on first read, fairly balanced about the issues. So far he's digging in to both the tech and personalities of key scientists. He differentiates between "scientists" and "engineers", one underlying point is scientists goal is, well, science. Getting published.
The engineer wants to make something work. And with non-gov/university projects, getting the devices to work as a critical goal may make all the difference. So not always 20 years away anymore...
(unfortunately it seems like even tho' we're now 15 years away, that could remain "always 15")
posted by sammyo at 5:05 AM on December 9, 2021
The engineer wants to make something work. And with non-gov/university projects, getting the devices to work as a critical goal may make all the difference. So not always 20 years away anymore...
(unfortunately it seems like even tho' we're now 15 years away, that could remain "always 15")
posted by sammyo at 5:05 AM on December 9, 2021
what if we use blockchain
we've tried to harness the power of the sun, but have we ever tried to harness the power of the Dunning Krueger field?
posted by condour75 at 5:10 AM on December 9, 2021 [15 favorites]
we've tried to harness the power of the sun, but have we ever tried to harness the power of the Dunning Krueger field?
posted by condour75 at 5:10 AM on December 9, 2021 [15 favorites]
this is so 20 years ago. the new zeitgeist is, you can live the exact same lifestyle, but with green energy. does anyone else remember when environmentalists were allowed to actually advocate for conservation, or am I just hallucinating?
posted by AlbertCalavicci at 5:16 AM on December 9, 2021 [7 favorites]
posted by AlbertCalavicci at 5:16 AM on December 9, 2021 [7 favorites]
8 don't really understand when people bring up that we could be doing nuclear in some form, even with less radioactivity, that for many of us it's not that we don't trust the science/technology it's that we don't trust the companies that would be running it
posted by kokaku at 5:22 AM on December 9, 2021 [5 favorites]
posted by kokaku at 5:22 AM on December 9, 2021 [5 favorites]
Does anyone else remember when environmentalists were allowed to actually advocate for conservation, or am I just hallucinating?
Science tells us it should theoretically be possible to do this, but engineers always tell us it's 20 years away.
posted by condour75 at 5:32 AM on December 9, 2021 [1 favorite]
Science tells us it should theoretically be possible to do this, but engineers always tell us it's 20 years away.
posted by condour75 at 5:32 AM on December 9, 2021 [1 favorite]
At "break-even" you're talking about 1% "extra" power generated with 99% going back into maintaining the reaction. That power still needs to be generated (ie with steam off the reaction). It may not generate as much radioactive waste or carbon pollution, but the waste HEAT is 99 times that of carbon power. That quickly becomes an issue. An economist once figured out that fusion power would raise the earth to boiling temperature within a few hundred years. Boiling. Forget fusion
posted by sexyrobot at 5:41 AM on December 9, 2021 [1 favorite]
posted by sexyrobot at 5:41 AM on December 9, 2021 [1 favorite]
We’re very far from feasible fusion power.
About 93 million miles, although it varies a bit.
posted by notoriety public at 5:58 AM on December 9, 2021 [36 favorites]
About 93 million miles, although it varies a bit.
posted by notoriety public at 5:58 AM on December 9, 2021 [36 favorites]
but the waste HEAT is 99 times that of carbon power
Yeah, I don't think that's how it works - if you have large amounts of 'waste' heat you can always convert it to lower-entropy forms at (or near) the Carnot efficiency (see combined-cycle fossil-fuel plants). Fusion people generally talk about 'break-even' meaning plasma Q=1, which as mentioned above is not whole-system break-even but does mean that the plasma is self-heating so the rest is just (theoretically) a matter of scale. Clearly a situation with Q>10 would be desirable for commercial deployment (and that's a way away, hence my position advocating breeder fission where the equivalent ratio is in the thousands)
posted by memetoclast at 5:58 AM on December 9, 2021 [2 favorites]
Yeah, I don't think that's how it works - if you have large amounts of 'waste' heat you can always convert it to lower-entropy forms at (or near) the Carnot efficiency (see combined-cycle fossil-fuel plants). Fusion people generally talk about 'break-even' meaning plasma Q=1, which as mentioned above is not whole-system break-even but does mean that the plasma is self-heating so the rest is just (theoretically) a matter of scale. Clearly a situation with Q>10 would be desirable for commercial deployment (and that's a way away, hence my position advocating breeder fission where the equivalent ratio is in the thousands)
posted by memetoclast at 5:58 AM on December 9, 2021 [2 favorites]
The funny thing about that perennial joke is that way back when the US Government did a study looking at how long it would take to develop a workable fusion reactor. It said it could be done in 20 years for $x per year, in 40 for $y per year, and that with less than $z a year it would never get done. We chose the latter, so it's never happened.
Here's that graph.
posted by paper chromatographologist at 7:08 AM on December 9, 2021 [9 favorites]
Here's that graph.
posted by paper chromatographologist at 7:08 AM on December 9, 2021 [9 favorites]
For everyone convinced that we should just build more fission reactors, my dad worked for Duke Power for 40 years, starting out in nuclear. I want to believe in nuclear because I heard about its benefits at the dinner table throughout my childhood. But the reality is that Georgia Power has been building the new units at Plant Vogtle for 13 years now, and it's still nowhere near done, and we the customers have paid huge amounts for it and will continue to do so. The completion of Plant Vogtle at this point feels exactly as likely as the development of workable fusion reactors.
Yes, part of the delay was Westinghouse nuclear going bankrupt in 2017 and part of it was COVID, but all of that money that they have collected from their customers could have been used to cover every strip mall and properly oriented home roof in Georgia with solar panels and to install ground geothermal heatpumps (which work great in our climate) instead.
posted by hydropsyche at 7:11 AM on December 9, 2021 [13 favorites]
Yes, part of the delay was Westinghouse nuclear going bankrupt in 2017 and part of it was COVID, but all of that money that they have collected from their customers could have been used to cover every strip mall and properly oriented home roof in Georgia with solar panels and to install ground geothermal heatpumps (which work great in our climate) instead.
posted by hydropsyche at 7:11 AM on December 9, 2021 [13 favorites]
what if we use blockchain
wait, no, hear me out. NFT energy credits. you want to be a player in crypto, you need proof of work
energy = the ability to do work
work = force x distance
you guys thinking what i'm thinking? time to disrupt the first law of thermodynamics
posted by Mayor West at 7:39 AM on December 9, 2021 [9 favorites]
wait, no, hear me out. NFT energy credits. you want to be a player in crypto, you need proof of work
energy = the ability to do work
work = force x distance
you guys thinking what i'm thinking? time to disrupt the first law of thermodynamics
posted by Mayor West at 7:39 AM on December 9, 2021 [9 favorites]
The other "just around the corner" technology I have been anticipating since the mid late 80's is the hydrogen or methane fuel cell. I remember we were supposed to get one in every basement, providing electricity for our needs and the grid, with the waste heat going for hot water.
posted by Pembquist at 8:07 AM on December 9, 2021 [1 favorite]
posted by Pembquist at 8:07 AM on December 9, 2021 [1 favorite]
what if we use blockchain
have we ever tried to harness the power of the Dunning Krueger field?
New cryptocurrency: Dunning-Krugerrands.
posted by biffa at 8:29 AM on December 9, 2021 [10 favorites]
have we ever tried to harness the power of the Dunning Krueger field?
New cryptocurrency: Dunning-Krugerrands.
posted by biffa at 8:29 AM on December 9, 2021 [10 favorites]
At least, unlike Theranos, everybody here seems to still be forecasting for what could happen in the future rather than saying it works now and that they put it in a bunch of US Army attack helicopters and copy and pasted the DOE logo all over their reports to lend credibility to an ongoing fraud.
Sorry, I'm following the Theranos trial with immense enthusiasm and couldn't help but dig into it a bit here.
posted by mmcg at 9:39 AM on December 9, 2021 [1 favorite]
Sorry, I'm following the Theranos trial with immense enthusiasm and couldn't help but dig into it a bit here.
posted by mmcg at 9:39 AM on December 9, 2021 [1 favorite]
At least Zap Energy addresses one of my great puzzlements. Magnetic confinement is problematic because maintaining a cryogenic superconductor near a scorching hot plasma doesn’t seem very practical, and removing that heat for useful work even less so. Moreover extracting energy via magnetic flux is going to be hard when the plasma is surrounded by flux-fixing superconductors. Maybe people have worked that out but I’ve never seen it explained. Zap Energy’s solution is to just not use magnets! But I don’t understand it much beyond that.
posted by sjswitzer at 10:49 AM on December 9, 2021
posted by sjswitzer at 10:49 AM on December 9, 2021
you guys thinking what i'm thinking? time to disrupt the first law of thermodynamics
Quick, somebody mock up a photo of five disaffected hipsters standing in front of a brick wall for the marketing brochure and Twitter press release!
posted by darkstar at 11:48 AM on December 9, 2021
Quick, somebody mock up a photo of five disaffected hipsters standing in front of a brick wall for the marketing brochure and Twitter press release!
posted by darkstar at 11:48 AM on December 9, 2021
Ultimately, fusion power's biggest handicap is that it doesn't have any obvious military purpose.
Or rather we found the military purpose 70 years ago (as a booster to fission) and it doesn’t depend on solving any of the problems we’d have to solve to build a power station.
posted by atoxyl at 12:11 PM on December 9, 2021 [2 favorites]
Or rather we found the military purpose 70 years ago (as a booster to fission) and it doesn’t depend on solving any of the problems we’d have to solve to build a power station.
posted by atoxyl at 12:11 PM on December 9, 2021 [2 favorites]
Or maybe sustained fusion is really hard at less than an octillion tons or thereabouts?
posted by clawsoon at 12:50 PM on December 9, 2021
posted by clawsoon at 12:50 PM on December 9, 2021
what if we use blockchain
Then we're gonna need even more power.
posted by straight at 12:54 PM on December 9, 2021 [1 favorite]
Then we're gonna need even more power.
posted by straight at 12:54 PM on December 9, 2021 [1 favorite]
Love the cryptosnark but actually that one last bitcoin pretty much will require a fully dedicated fusion plant. Win win.
posted by sammyo at 1:14 PM on December 9, 2021
posted by sammyo at 1:14 PM on December 9, 2021
NFT: Nuclear Fusion Token.
posted by storybored at 1:54 PM on December 9, 2021 [2 favorites]
posted by storybored at 1:54 PM on December 9, 2021 [2 favorites]
all of that money that they have collected from their customers could have been used to cover every strip mall and properly oriented home roof in Georgia with solar panels and to install ground geothermal heatpumps (which work great in our climate) instead
This is the main issue for nuclear that I'm seeing. You aren't going to be building nuke plants without massive government subsidies. A nuke plant that can generate 2-3 gigawatts will cost you upwards of $5 billion, IIRC - if things go smoothly. It's going to take 5-7 years to build, again, if all goes smoothly. (Nuclear can be built safely but it takes a lot of money and time to do so.) If you're looking to invest $5 billion into power generation, are you going to put that money down on a nuke plant that will not collect a dime for possibly close to a decade? Or are you going to put in into solar and wind, along with a couple of natural gas plants, which can be online and generating ROI in far, far less time? The market conditions do not support building nuke plants, so the government would have to be supplying most if not all the money to build reactors.
But fusion? Yeah, let me know if someone ever figures it out. Hey, it looks good on paper...
posted by azpenguin at 2:25 PM on December 9, 2021 [2 favorites]
This is the main issue for nuclear that I'm seeing. You aren't going to be building nuke plants without massive government subsidies. A nuke plant that can generate 2-3 gigawatts will cost you upwards of $5 billion, IIRC - if things go smoothly. It's going to take 5-7 years to build, again, if all goes smoothly. (Nuclear can be built safely but it takes a lot of money and time to do so.) If you're looking to invest $5 billion into power generation, are you going to put that money down on a nuke plant that will not collect a dime for possibly close to a decade? Or are you going to put in into solar and wind, along with a couple of natural gas plants, which can be online and generating ROI in far, far less time? The market conditions do not support building nuke plants, so the government would have to be supplying most if not all the money to build reactors.
But fusion? Yeah, let me know if someone ever figures it out. Hey, it looks good on paper...
posted by azpenguin at 2:25 PM on December 9, 2021 [2 favorites]
What a deeply negative article that was, so focused on the financial fraud aspect.
I liked this New Yorker article by Rivka Galchen in October 2021 better - she actually looked at the technology, rather than the financial engineering.
posted by RedOrGreen at 2:27 PM on December 9, 2021 [2 favorites]
I liked this New Yorker article by Rivka Galchen in October 2021 better - she actually looked at the technology, rather than the financial engineering.
Over decades, in the pursuit of nuclear fusion, scientists and engineers built giant metal doughnuts and Gehryesque twisted coils, they “pinched” plasmas with lasers, and they constructed fusion devices in garages. For thirty-six years, they have been planning and building an experimental fusion device in Provence. And yet commercially viable nuclear-fusion energy has always remained just a bit farther on. As the White Queen, in “Through the Looking Glass,” said to Alice, it is never jam today, it is always jam tomorrow.But maybe SPARC?
... The allure of fusion has attracted brilliant, imaginative minds; it has also attracted a crowd of shysters, cranks, and false messiahs. ...
Commonwealth Fusion Systems, a seven-person private fusion-energy company with an ongoing relationship with M.I.T. (C.F.S. funds research at M.I.T., which shares its intellectual resources and some lab space with C.F.S.; patents are filed jointly.)I thought it was a pretty balanced and sober take, but it left me optimistic. Consider reading, as they say, the whole thing.
... “Most of the criticism we hear is not about the science but about the timeline,” Mumgaard said. The magnets inside ITER took thirty years to develop. “It took us three years.” He could barely repress a grin; it was the one moment of boyish bullishness and ego that I saw in him.
... many fusion scientists take seriously C.F.S.’s claims that by 2025 it will be demonstrating a fusion device that gives out considerably more energy than it takes to run. But many, many technological challenges remain before fusion will turn on the lights in your kitchen.
posted by RedOrGreen at 2:27 PM on December 9, 2021 [2 favorites]
Comrades, blockchain requires huge amounts of energy to create.
But what if we run the machines in reverse and turn bitcoins back into energy?!
Feed cryptobro into hopper and power city for a week.
posted by fallingbadgers at 3:22 PM on December 9, 2021 [5 favorites]
But what if we run the machines in reverse and turn bitcoins back into energy?!
Feed cryptobro into hopper and power city for a week.
posted by fallingbadgers at 3:22 PM on December 9, 2021 [5 favorites]
Or are you going to put in into solar and wind, along with a couple of natural gas plants
Sounds like a good way to get undercut (in both money and response time) by the battery next door. Deliberately building a NG plant to run purely as a peaker is already not financially competitive.
posted by rhamphorhynchus at 5:41 PM on December 9, 2021
Sounds like a good way to get undercut (in both money and response time) by the battery next door. Deliberately building a NG plant to run purely as a peaker is already not financially competitive.
posted by rhamphorhynchus at 5:41 PM on December 9, 2021
she actually looked at the technology
I would point out there is very little about the technology. It is primarily written, as much lay science writing is, as a bunch of anecdotes about scientists. There's a little bit about science, and even less about actual engineering and development.
For example, there's an account of an IEEE meeting where some scientists say 30 billion dollars and 20 years to a prototype. The primary sources for the story--the protagonists, the POV characters--say they can do it in a few years and a few percent of the cost. It's narratively central to the story but I can't say I came away with any particular technical understanding of the differences, or any confidence the reporter had an understanding either.
This line is great and typical:
I'm poo-pooing this not just for the heck of it. It's fine to read profiles of engineers. It's just in this context, it's worth remembering that this is essentially what happened with Theranos. One thing that struck me reading Bad Blood was how easy it was for people to detect Theranos was a bad company with bad tech, but there was basically no market for those people to convince others until after it hit it big. (Here's the New Yorker profile of Holmes, if you're interested to compare tone to the one you recommend on fusion.)
For fusion to work economically, they have to come through simultaneously on multiple technologies *and* basically see wind/solar and storage development stagnate over the next two decades. I'm all in favor of funding research, because maybe those two things happen. But it's relevance is a long shot--and thinking it will matter is a sign of pessimism, not optimism, in the larger picture, because it presupposes the failure of all this other stuff.
[1]These science stories are a cousin to a staple of the arts section: One of the profiles of an actor or director while they're in middle of filming a well funded movie that might make them famous. No one's seen the movie; in fact, it doesn't exist yet. It might be horrible. But the reporter got access and the subject was interesting, they're not going to dwell on that.
posted by mark k at 6:05 PM on December 9, 2021 [6 favorites]
I would point out there is very little about the technology. It is primarily written, as much lay science writing is, as a bunch of anecdotes about scientists. There's a little bit about science, and even less about actual engineering and development.
For example, there's an account of an IEEE meeting where some scientists say 30 billion dollars and 20 years to a prototype. The primary sources for the story--the protagonists, the POV characters--say they can do it in a few years and a few percent of the cost. It's narratively central to the story but I can't say I came away with any particular technical understanding of the differences, or any confidence the reporter had an understanding either.
This line is great and typical:
Fusion requires no major advances in batteries, it would be available on demand, it wouldn’t cause the next Fukushima, and it wouldn’t be too priceyFusion needs things that are lot more complicated than "batteries" and how predicting the commercial pricing scheme for inventions that haven't been invented yet is unclear. But that's not the point; it's a profile of scientists, not a technology assessment, so you need to frame the work as likely to succeed[1].
I'm poo-pooing this not just for the heck of it. It's fine to read profiles of engineers. It's just in this context, it's worth remembering that this is essentially what happened with Theranos. One thing that struck me reading Bad Blood was how easy it was for people to detect Theranos was a bad company with bad tech, but there was basically no market for those people to convince others until after it hit it big. (Here's the New Yorker profile of Holmes, if you're interested to compare tone to the one you recommend on fusion.)
For fusion to work economically, they have to come through simultaneously on multiple technologies *and* basically see wind/solar and storage development stagnate over the next two decades. I'm all in favor of funding research, because maybe those two things happen. But it's relevance is a long shot--and thinking it will matter is a sign of pessimism, not optimism, in the larger picture, because it presupposes the failure of all this other stuff.
[1]These science stories are a cousin to a staple of the arts section: One of the profiles of an actor or director while they're in middle of filming a well funded movie that might make them famous. No one's seen the movie; in fact, it doesn't exist yet. It might be horrible. But the reporter got access and the subject was interesting, they're not going to dwell on that.
posted by mark k at 6:05 PM on December 9, 2021 [6 favorites]
The problem for all these technologies is the continued advance of battery and solar in terms of cost and performance. Even if they meet the most optimistic timelines and get a working reactor by 2025; it probably will take at least 10 years to build the commercial scale reactors and bring them online and it’s billions of dollars to build that kind of facility. Meanwhile solar panels and batteries keep getting cheaper.
posted by interogative mood at 7:07 PM on December 9, 2021 [3 favorites]
posted by interogative mood at 7:07 PM on December 9, 2021 [3 favorites]
Ultimately, the thing that's going to kill commercial fusion power is that the entire project was conceived in an era before it became apparent that mass production is the key to making power too cheap to meter, and that countless decentralized small-scale grid inputs will inevitably outcompete every large-scale centralized technology on economic grounds decades before new designs for the latter become technically ready to deploy, just as they're already doing to coal.
Moving nuclear fusion 150 million km closer to my home than it needs to be is essentially just tech bros showing off, at this point. It has no practical purpose and never will.
posted by flabdablet at 8:21 PM on December 9, 2021 [3 favorites]
Moving nuclear fusion 150 million km closer to my home than it needs to be is essentially just tech bros showing off, at this point. It has no practical purpose and never will.
posted by flabdablet at 8:21 PM on December 9, 2021 [3 favorites]
what if we run the machines in reverse and turn bitcoins back into energy?!
As any Doctor Who afficionado would instantly be able to tell you, that doesn't even involve running the machines in reverse. All we'd actually need to do is reverse their polarity.
posted by flabdablet at 8:28 PM on December 9, 2021 [1 favorite]
As any Doctor Who afficionado would instantly be able to tell you, that doesn't even involve running the machines in reverse. All we'd actually need to do is reverse their polarity.
posted by flabdablet at 8:28 PM on December 9, 2021 [1 favorite]
Oh that's easy, just flip the power plug the other way up in the wall socket!
posted by Greg_Ace at 8:39 PM on December 9, 2021
posted by Greg_Ace at 8:39 PM on December 9, 2021
Anybody who still harbours doubts about the feasibility of powering a huge industrial infrastructure via a grid fed from an endless sea of kilowatt-scale rooftop solar and household battery storage needs to think hard about how crazy it once seemed to try to design the world's fastest production cars around AA cells.
Scale is scale whether you get it via huge things or numerous things. But going the numerous route gets you way more resilience and is faster and cheaper too.
posted by flabdablet at 8:41 PM on December 9, 2021
Scale is scale whether you get it via huge things or numerous things. But going the numerous route gets you way more resilience and is faster and cheaper too.
posted by flabdablet at 8:41 PM on December 9, 2021
just flip the power plug the other way up in the wall socket
No, it's AC so that won't work. You need to swap the black and green wires.
posted by flabdablet at 8:43 PM on December 9, 2021 [1 favorite]
No, it's AC so that won't work. You need to swap the black and green wires.
posted by flabdablet at 8:43 PM on December 9, 2021 [1 favorite]
In the long run, numerous things are at a significant efficiency disadvantage relative to larger things. It simply requires more material to make a bunch of 18650s than it does a smaller number of larger cells (which is one reason why Tesla is moving to 4680s).
That said, there is a significant advantage in redundancy that outweighs the loss of material efficiency for certain applications. There is also the advantage of being able to iterate more rapidly when the cost of each unit is lower. On the other hand, large things certainly can be produced relatively quickly and cheaply given sufficient demand. Large boilers and even nuclear reactor pressure vessels used to be made reasonably inexpensively and reliably because there was a time when we were making a lot of them along with a vast number of things similar enough that the experience and tooling translated easily. We left most of that behind a generation ago, so we're back to the beginning of the cost curve. Building a small number of solar panels would be completely uneconomical, and was for a long time, but billions of dollars and decades worth of scaling have made them cheap.
In short, it's a mistake to say that one way is definitely better than the other. It's a matter of weighing advantages and disadvantages, which change dramatically depending on when you're doing the analysis and the state of related industry.
posted by wierdo at 9:54 PM on December 9, 2021 [4 favorites]
That said, there is a significant advantage in redundancy that outweighs the loss of material efficiency for certain applications. There is also the advantage of being able to iterate more rapidly when the cost of each unit is lower. On the other hand, large things certainly can be produced relatively quickly and cheaply given sufficient demand. Large boilers and even nuclear reactor pressure vessels used to be made reasonably inexpensively and reliably because there was a time when we were making a lot of them along with a vast number of things similar enough that the experience and tooling translated easily. We left most of that behind a generation ago, so we're back to the beginning of the cost curve. Building a small number of solar panels would be completely uneconomical, and was for a long time, but billions of dollars and decades worth of scaling have made them cheap.
In short, it's a mistake to say that one way is definitely better than the other. It's a matter of weighing advantages and disadvantages, which change dramatically depending on when you're doing the analysis and the state of related industry.
posted by wierdo at 9:54 PM on December 9, 2021 [4 favorites]
No, it's AC so that won't work. You need to swap the black and green wires.
Is that "g" as in "grancy"?
/oblique Archer reference
posted by Greg_Ace at 9:58 PM on December 9, 2021 [2 favorites]
Is that "g" as in "grancy"?
/oblique Archer reference
posted by Greg_Ace at 9:58 PM on December 9, 2021 [2 favorites]
At least, unlike Theranos, everybody here seems to still be forecasting for what could happen in the future rather than saying it works now and that they put it in a bunch of US Army attack helicopters and copy and pasted the DOE logo all over their reports to lend credibility to an ongoing fraud.
This is what made Holmes an amateur. You are allowed to "lie" i.e. be ridiculously optimistic to your investors about the future, as long as you are rigorously accurate about the past and present. Whatever ultra hype comes out of Elon Musk, every quarter Tesla just reports actual quarterly results.
You also can't really compare the costs of all energy technologies on a strict LCOE basis (which would just be all solar PV and some wind) because you need the energy *system* to work as a whole. The problem is that what you really want as a complement to high capital cost / zero marginal cost VREs is a low capital cost / high marginal cost form of generation. Neither fusion nor fission is in that category.
What's interesting is that in my day job, none of this matters because I'm mostly concerned with getting projects built *now* to get our emissions down to about 20%-30% of what they are currently. My academic research interest though is in precisely that last few % and that's where the rogue's gallery of hypothetical technologies comes in, including fusion. It is right to spend a few % of our anticipated adaptation budget on technologies that might see 2040s deployment, as long as we do actually execute on the main programme to get all our "easy" emissions done by then.
posted by atrazine at 2:21 AM on December 10, 2021 [5 favorites]
This is what made Holmes an amateur. You are allowed to "lie" i.e. be ridiculously optimistic to your investors about the future, as long as you are rigorously accurate about the past and present. Whatever ultra hype comes out of Elon Musk, every quarter Tesla just reports actual quarterly results.
You also can't really compare the costs of all energy technologies on a strict LCOE basis (which would just be all solar PV and some wind) because you need the energy *system* to work as a whole. The problem is that what you really want as a complement to high capital cost / zero marginal cost VREs is a low capital cost / high marginal cost form of generation. Neither fusion nor fission is in that category.
What's interesting is that in my day job, none of this matters because I'm mostly concerned with getting projects built *now* to get our emissions down to about 20%-30% of what they are currently. My academic research interest though is in precisely that last few % and that's where the rogue's gallery of hypothetical technologies comes in, including fusion. It is right to spend a few % of our anticipated adaptation budget on technologies that might see 2040s deployment, as long as we do actually execute on the main programme to get all our "easy" emissions done by then.
posted by atrazine at 2:21 AM on December 10, 2021 [5 favorites]
No, it's AC so that won't work. You need to swap the black and green wires.
But whatever you do, don’t cut the blue wire!
Or is it the red…
posted by Big Al 8000 at 4:36 AM on December 10, 2021
But whatever you do, don’t cut the blue wire!
Or is it the red…
posted by Big Al 8000 at 4:36 AM on December 10, 2021
does anyone else remember when environmentalists were allowed to actually advocate for conservation, or am I just hallucinating?
It doesn't really feel like it, but that argument was won by the conservationists, helped along by the economic reality that it's cheaper to not build more power plants.
From cars to light bulbs to home appliances to air conditioning and more, energy efficiency has doubled in my lifetime. Better, in many cases. Energy use has not been increasing at anywhere near what one would expect from population growth, even as we have spread out, bought bigger houses, and otherwise used up some of those gains in energy efficiency.
posted by wierdo at 2:30 PM on December 10, 2021 [2 favorites]
It doesn't really feel like it, but that argument was won by the conservationists, helped along by the economic reality that it's cheaper to not build more power plants.
From cars to light bulbs to home appliances to air conditioning and more, energy efficiency has doubled in my lifetime. Better, in many cases. Energy use has not been increasing at anywhere near what one would expect from population growth, even as we have spread out, bought bigger houses, and otherwise used up some of those gains in energy efficiency.
posted by wierdo at 2:30 PM on December 10, 2021 [2 favorites]
I have yet to see anything in here that convinces me selling mintng and selling FusionCoins ($Tokamakens, 'cause Tokamak Token) is a bad idea. After all, fusion is closer than ever, and a portion of the money raised will be donated to fusion research. Make fusion power a reality sooner by fusing your money to my shitcoin.
Donate to our kickstarter to help make the FusionCoins project possible. We can achieve a brighter future together.
(Note: this is sarcasm. None of this is real. Don't buy any crypto, especially not from me, and I don't have a kickstarter. But hey, who's gonna stop me?)
(What if, instead, this comment was part of a marketing campaign for a novel I wrote and self-published on Amazon about two epic cryptobros who had the idea for fusion coins at the same time and then battled it out to see who is the one fusion coin to rule them all, and then I had a kick-started to write that novel, and then I published the novel concurrently with the launch of two actual fusion coins, named after the novel of course, and then I sold each word in the novel as an NFT?)
(This has already happened hasn't it?)
posted by saysthis at 7:29 PM on December 11, 2021
Donate to our kickstarter to help make the FusionCoins project possible. We can achieve a brighter future together.
(Note: this is sarcasm. None of this is real. Don't buy any crypto, especially not from me, and I don't have a kickstarter. But hey, who's gonna stop me?)
(What if, instead, this comment was part of a marketing campaign for a novel I wrote and self-published on Amazon about two epic cryptobros who had the idea for fusion coins at the same time and then battled it out to see who is the one fusion coin to rule them all, and then I had a kick-started to write that novel, and then I published the novel concurrently with the launch of two actual fusion coins, named after the novel of course, and then I sold each word in the novel as an NFT?)
(This has already happened hasn't it?)
posted by saysthis at 7:29 PM on December 11, 2021
« Older Lawless loggers | RIP Robbie Shakespeare Newer »
This thread has been archived and is closed to new comments
I get that fusion is a cool idea, but mag-confinement D/T fusion just doesn't seem like a viable power source. If we could work out some way of doing aneutronic reactions that didn't require breeding unstable isotopes (Tritium) for fuel (P+/B is often touted) then maybe, but this is not that.
(Sorry for the derail - I just find the technical issues more interesting than the corporate mendacity angle :)
posted by memetoclast at 9:49 PM on December 8, 2021 [32 favorites]