4 ms·

I'll take that bet - say, $500 in 2021 dollars, that a fusion power plant is selling energy to the grid? I'll even make it easier and halve the time you suggest

I'll take that bet - say, $500 in 2021 dollars, that a fusion power plant is selling energy to the grid? I'll even make it easier and halve the time you suggested to 25 years, so we can settle the bet in 2046.

jetbooster 5y ago
https://longbets.org https://longbets.org ?
- Tossrock 5y ago
  That would be my favored platform!
  - billiam 5y ago
    I like what Tim Bray's doing with his time. https://longbets.org/863/ https://longbets.org/863/
    - ad404b8a372f2b9 5y ago
      I'd bet a bitcoin against him.
leephillips 5y ago
I’m interested, but the bet needs another condition, to exclude toy demonstration projects. The reactor will have to generate at least 100MW (far less than existing coal plants) and be in operation for an integrated time of at least 90 days over the course of any one year on or before 2046. Accept?
- Tossrock 5y ago
  100MW seems like a substantial moving of the goalposts, given your earlier statement that "there is no reasonably foreseeable future with fusion as part of the electricity grid" and that I've already cut the timetable by 25 years :) That said, I'll still accept - I'm emailing you at your profile address for the details!
  - maxerickson 5y ago
    "Practical" is in there. A price competitive 10 MW generator probably meets that standard though (Islands, small towns, isolated mills, etc).
    - fasdf23967 5y ago
      ...very fast spaceships for the solar system, interstellar big ones probably. With continous thrust.
      - maxerickson 5y ago
        That would likely not demonstrate grid comparable cost though.
  - leephillips 5y ago
    I didn’t mean to move the goalposts. I just want to exclude demonstration projects that might produce some net energy but not be serious commercial sources of electricity. But thanks for accepting anyway.
    - Tossrock 5y ago
      I've sent an message to the address listed in your profile, it's coming from a nonstandard domain though, so if you don't see it, it may be in spam. Also, I now realize that the longbet page is still under review, so you might not be able to see that either until the staff approve it.
      - kbenson 5y ago
        I wonder how many of the bets on the longbets site stem from HN discussions. Probably not a significant number, but it would be deeply interesting to go back and read the discussions that spawned them.
        5y ago
- DennisP 5y ago
  Seems like any profitable plant should count. Some designs work best at smaller scales, but if they worked out they'd be cheap and for more power you just build a lot of them. Even in fission, there's a big push now to build reactors small enough to mass-produce in factories. Maybe say at least 100MW total power?
  - jollybean 5y ago
    Energy has the most externalized costs of any industry. The 5th fleet is in the Gulf to protect the flow of Oil. The USD is backed to some extent by petrodollar, and that is a geopolitical hammer the Americans like to use at least to some extent. So what does 'profitable' mean? If Climate Change gets really problematic quickly, then guess what, all Nuclear Plants become considerably more profitable because the government will socialize the losses in case of catastrophic failure meaning owners don't pay for massive insurance costs which are a problem for profitability today give the possibility of $100B payouts in the case of failure. I'm wary of the commentator's cynicism. If we can make demo plants operating at some scale, close to break even in 25 years ... then that's a strong hint there's material progress, and that those plants could be breaking even another 25 years later. It also easily justifies a number of scientists working on it now even if only pans out in 50 years. The long term surpluses are potentially ginormous, like, to the point where they existentially shape the future, much like carbon fuels triggered the industrial revolution.
    - DennisP 5y ago
      Yeah maybe leave out "profitable," it's too fuzzy and hard to verify. One of the bettors simply expressed skepticism that there would be "fusion as part of the electricity grid," period, so just leave it at that. At least, that's what I would want if I were betting. (I do think it's entirely possible that fusion will be solidly profitable, especially with carbon pricing.)
ludsan 5y ago
I think this is a safe bet. Between Commonwealth Fusion's Arc/Sparc, or General Fusion's spinning glob of hot metal, or TAE systems, or any of the others, I think you have a better than average chance of settling this bet within 20 years.
- pfdietz 5y ago
  ARC has a power density 40x worse than a LWR's primary reactor vessel. General Fusion abandoned their first scheme because of at least three showstoppers (vaporization of the liquid metal wall, Richtmyer-Meshkov instability turning the implosion into jets of metal, and stochastic magnetic field lines in the spheromak causing unacceptable loss of energy via electrons to the metal). The new scheme has extremely serious engineering problems (the central pillar will be in a radiation/thermal environment orders of magnitude worse than the walls of ITER, and subject to extreme JxB forces). And they've never produced a neutron, as far as I know. Rostoker et al. were told 20+ years ago that their p-11B concept couldn't work, for at least eight different reasons. https://www.researchgate.net/publication/235032059_Comments_on_the_Colliding_Beam_Fusion_Reactor_Proposed_by_Rostoker_Binderbauer_and_Monkhorst_for_Use_with_the_p-11B_Fusion_Reaction https://www.researchgate.net/publication/235032059_Comments_... If I had to bet on any current private fusion effort I'd choose either Zap Energy or Helion.
  - nathanathan 5y ago
    Why is the power density of the ARC/SPARC reactor problematic?
    - pfdietz 5y ago
      Lower power density implies larger size, which implies higher cost. Since the parts outside the reactors themselves (the turbines, generators, heat sink, etc.) will be similar, this means a fusion power plant will cost more to build that a fission power plant. The other problem is that size is also related to reliability. A fusion reactor will have many more parts than a fission reactor. It will also be much more complex, and operate at higher radiation and thermal stresses. This is particularly important because it is very difficult to repair something that is so radioactive that hands-on access is not possible.

jetbooster 5y ago

https://longbets.org https://longbets.org ?

Tossrock 5y ago
That would be my favored platform!
- billiam 5y ago
  I like what Tim Bray's doing with his time. https://longbets.org/863/ https://longbets.org/863/
  - ad404b8a372f2b9 5y ago
    I'd bet a bitcoin against him.

leephillips 5y ago

I’m interested, but the bet needs another condition, to exclude toy demonstration projects. The reactor will have to generate at least 100MW (far less than existing coal plants) and be in operation for an integrated time of at least 90 days over the course of any one year on or before 2046. Accept?

Tossrock 5y ago
100MW seems like a substantial moving of the goalposts, given your earlier statement that "there is no reasonably foreseeable future with fusion as part of the electricity grid" and that I've already cut the timetable by 25 years :) That said, I'll still accept - I'm emailing you at your profile address for the details!
- maxerickson 5y ago
  "Practical" is in there. A price competitive 10 MW generator probably meets that standard though (Islands, small towns, isolated mills, etc).
  - fasdf23967 5y ago
    ...very fast spaceships for the solar system, interstellar big ones probably. With continous thrust.
    - maxerickson 5y ago
      That would likely not demonstrate grid comparable cost though.
- leephillips 5y ago
  I didn’t mean to move the goalposts. I just want to exclude demonstration projects that might produce some net energy but not be serious commercial sources of electricity. But thanks for accepting anyway.
  - Tossrock 5y ago
    I've sent an message to the address listed in your profile, it's coming from a nonstandard domain though, so if you don't see it, it may be in spam. Also, I now realize that the longbet page is still under review, so you might not be able to see that either until the staff approve it.
    - kbenson 5y ago
      I wonder how many of the bets on the longbets site stem from HN discussions. Probably not a significant number, but it would be deeply interesting to go back and read the discussions that spawned them.
      - 5y ago
DennisP 5y ago
Seems like any profitable plant should count. Some designs work best at smaller scales, but if they worked out they'd be cheap and for more power you just build a lot of them. Even in fission, there's a big push now to build reactors small enough to mass-produce in factories. Maybe say at least 100MW total power?
- jollybean 5y ago
  Energy has the most externalized costs of any industry. The 5th fleet is in the Gulf to protect the flow of Oil. The USD is backed to some extent by petrodollar, and that is a geopolitical hammer the Americans like to use at least to some extent. So what does 'profitable' mean? If Climate Change gets really problematic quickly, then guess what, all Nuclear Plants become considerably more profitable because the government will socialize the losses in case of catastrophic failure meaning owners don't pay for massive insurance costs which are a problem for profitability today give the possibility of $100B payouts in the case of failure. I'm wary of the commentator's cynicism. If we can make demo plants operating at some scale, close to break even in 25 years ... then that's a strong hint there's material progress, and that those plants could be breaking even another 25 years later. It also easily justifies a number of scientists working on it now even if only pans out in 50 years. The long term surpluses are potentially ginormous, like, to the point where they existentially shape the future, much like carbon fuels triggered the industrial revolution.
  - DennisP 5y ago
    Yeah maybe leave out "profitable," it's too fuzzy and hard to verify. One of the bettors simply expressed skepticism that there would be "fusion as part of the electricity grid," period, so just leave it at that. At least, that's what I would want if I were betting. (I do think it's entirely possible that fusion will be solidly profitable, especially with carbon pricing.)

ludsan 5y ago

I think this is a safe bet. Between Commonwealth Fusion's Arc/Sparc, or General Fusion's spinning glob of hot metal, or TAE systems, or any of the others, I think you have a better than average chance of settling this bet within 20 years.

pfdietz 5y ago
ARC has a power density 40x worse than a LWR's primary reactor vessel. General Fusion abandoned their first scheme because of at least three showstoppers (vaporization of the liquid metal wall, Richtmyer-Meshkov instability turning the implosion into jets of metal, and stochastic magnetic field lines in the spheromak causing unacceptable loss of energy via electrons to the metal). The new scheme has extremely serious engineering problems (the central pillar will be in a radiation/thermal environment orders of magnitude worse than the walls of ITER, and subject to extreme JxB forces). And they've never produced a neutron, as far as I know. Rostoker et al. were told 20+ years ago that their p-11B concept couldn't work, for at least eight different reasons. https://www.researchgate.net/publication/235032059_Comments_on_the_Colliding_Beam_Fusion_Reactor_Proposed_by_Rostoker_Binderbauer_and_Monkhorst_for_Use_with_the_p-11B_Fusion_Reaction https://www.researchgate.net/publication/235032059_Comments_... If I had to bet on any current private fusion effort I'd choose either Zap Energy or Helion.
- nathanathan 5y ago
  Why is the power density of the ARC/SPARC reactor problematic?
  - pfdietz 5y ago
    Lower power density implies larger size, which implies higher cost. Since the parts outside the reactors themselves (the turbines, generators, heat sink, etc.) will be similar, this means a fusion power plant will cost more to build that a fission power plant. The other problem is that size is also related to reliability. A fusion reactor will have many more parts than a fission reactor. It will also be much more complex, and operate at higher radiation and thermal stresses. This is particularly important because it is very difficult to repair something that is so radioactive that hands-on access is not possible.