9 ms·

What % chance do we feel that we will have fusion power within the next 50 years?

weberer 4y ago
Construction of the DEMO plant is currently set to begin in 2040. It is made to follow ITER, so any delay in ITER will propagate to DEMO. https://en.wikipedia.org/wiki/DEMOnstration_Power_Plant#Timeline https://en.wikipedia.org/wiki/DEMOnstration_Power_Plant#Time... https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/ITER
- DennisP 4y ago
  Yes but ITER is the world's slowest fusion project. Now we can build much smaller reactors, much more quickly, with the same output because we have better superconductors. Several startups are tackling it, and a lot of fusion scientists think CFS will beat ITER to net power by a decade.
  - myrmidon 4y ago
    It sure might be faster to experiment with new tech if your reactor is toy-sized, but building the whole thing at a scale that's somewhat realistic for economically generating power seems an un-avoidable step to me. ITER is doing that NOW and NONE of headline-generating startups have even started... So it seems to me that ITER is both contributing more engineering insights and closer towards a full-fledged fusion power-plant, and unlikely to be overtaken in these regards anytime soon, if ever.
    - DennisP 4y ago
      Tokamak output scales with the square of plasma volume, but the fourth power of magnetic field strength. Double the field, 16X the output. We have superconductors now that can support much stronger magnetic fields than ITER uses. That lets us build smaller tokamaks with the same output as ITER. Several startups are using them, led by MIT spinoff CFS. Their SPARC reactor should have output similar to ITER's, in a reactor half the size of JET, which was built in a year. They expect to attempt net power in 2025 and a lot of researchers think they'll succeed. Assuming that works, as similar reactor the size of JET gets them to commercial output levels.
      - myrmidon 4y ago
        My point is that ITER is designed for 500MW (thermal) output. SPARC aims for something around 100MW (and is even further from continuous operation than ITER by design), and even the ARC successor (which is currently basically a pipe dream) is going to catch up with ITER in output power at best. So even IF the whole SPARC thing goes according to play, it's going to be miles away from anything resembling a competitive power plant, while ITER is potentially something like a quarter way there (assuming that operating at 2GW thermal begins being somewhat attractive/feasible for a fusion-powerplant). Personally, I strongly doubt that fusion power is EVER going to be an option because of simple economics; I just don't really see how a vacuum chamber surrounded by superconducting magnets, cooling systems, turbines and turbogenerators is ever going to compete (financially) with just slapping PV panels on a roof, putting some battery banks in the basement, hooking it up with an inverter and just repeating as necessary until power demands are met...
        DennisP 4y ago
        Bigger is better only if it gives you a better ratio of capital expense to output. Due to the magnetic scaling I mentioned, SPARC/ARC probably does significantly better on that front.
        myrmidon 4y ago
        My point is that any commercial plant MUST be big enough to supply a generic steam turbine setup with heat; every fusion reactor design that cant achieve that is just a glorified toy, because it CANT operate economically. A bigger research reactor is simply better in that it enables you to investigate/solve problems that are related to scale, and all those MUST be solved before a commercial plant can be built. Until ARC or an even further removed successor catches up to ITER in size/power, the SPARC project is just yet another toy reactor IMO.
        DennisP 4y ago
        The US has 37 coal plants below 10MW, 73 plants from 50 to 100MW, and 85 plants between 100 and 250MW: https://www.gem.wiki/Existing_U.S._Coal_Plants#Size_comparison_of_coal_plants https://www.gem.wiki/Existing_U.S._Coal_Plants#Size_comparis... Coal plants of course use steam turbines: https://www.tva.com/energy/our-power-system/coal/how-a-coal-plant-works https://www.tva.com/energy/our-power-system/coal/how-a-coal-...
  - foolfoolz 4y ago
    the U. S. has spent more on ukraine in the last year than on ITER since it’s inception. ITER isn’t slow because it’s big it’s slow because of how it’s funded
    - DennisP 4y ago
      But the US alone has spent more on ITER than CFS will spend on SPARC. This isn't entirely ITER's fault. The superconductor technology SPARC is using didn't exist when ITER was designed.
anigbrowl 4y ago
73%
sien 4y ago
On Metaculus : Most likely - fusion ignition in the 2030s. Fusion supplying 10% of global electricity in the 2050s. https://www.metaculus.com/questions/?order_by=-rank&main-feed&search=fusion https://www.metaculus.com/questions/?order_by=-rank&main-fee...
- zardo 4y ago
  Those two predictions don't seem to fit together.
  - kortex 4y ago
    Why? Even once you figure out ignition, you still have a bunch of engineering issues to tackle. Energy extraction, tritium breeding, reactor longevity, reducing the build time and cost. Hopefully most of that can be figured out in parallel while we inch closer to ignition, but it's not like once we get ignition we can just copy-paste our way to actual commercial fusion power production.
    - zardo 4y ago
      Because going from the science project has finally worked to supplying 10% of the worlds electricity in just 20 years would be a truly incredible pace progress.
      - ForHackernews 4y ago
        Mobile phones went from 0.001% to 12.0% adoption from 1980 to 2000[0]. Things can happen quickly if there is very obviously lots of money to be made. [0] http://stats.areppim.com/stats/stats_mobilexpenetr.htm http://stats.areppim.com/stats/stats_mobilexpenetr.htm
        zardo 4y ago
        > Things can happen quickly if there is very obviously lots of money to be made. There is not obviously lots of money to be made in rapidly building out fusion power plants.
mkaic 4y ago
90%. I'm not qualified in the slightest to make that prediction but I'm making it anyways!
fibonacc 4y ago
99% but petrodollar will still be in power. Access to fusion power will be strictly controlled.
idontpost 4y ago
0% chance.
ncmncm 4y ago
Exactly 0%. Achieving stable fusion does not make practical power from it possible. It would necessarily cost >10x fission, and fission is not competitive. We can be absolutely ,confident that by 2100, there will have been zero kWh of commercial power from Tokamak fusion. It is just barely possible that D-3He fusion might work by then, but it would still struggle to compete. It might find use in outer-solar-system spacecraft.

Construction of the DEMO plant is currently set to begin in 2040. It is made to follow ITER, so any delay in ITER will propagate to DEMO. https://en.wikipedia.org/wiki/DEMOnstration_Power_Plant#Timeline https://en.wikipedia.org/wiki/DEMOnstration_Power_Plant#Time... https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/ITER

DennisP 4y ago
Yes but ITER is the world's slowest fusion project. Now we can build much smaller reactors, much more quickly, with the same output because we have better superconductors. Several startups are tackling it, and a lot of fusion scientists think CFS will beat ITER to net power by a decade.
- myrmidon 4y ago
  It sure might be faster to experiment with new tech if your reactor is toy-sized, but building the whole thing at a scale that's somewhat realistic for economically generating power seems an un-avoidable step to me. ITER is doing that NOW and NONE of headline-generating startups have even started... So it seems to me that ITER is both contributing more engineering insights and closer towards a full-fledged fusion power-plant, and unlikely to be overtaken in these regards anytime soon, if ever.
  - DennisP 4y ago
    Tokamak output scales with the square of plasma volume, but the fourth power of magnetic field strength. Double the field, 16X the output. We have superconductors now that can support much stronger magnetic fields than ITER uses. That lets us build smaller tokamaks with the same output as ITER. Several startups are using them, led by MIT spinoff CFS. Their SPARC reactor should have output similar to ITER's, in a reactor half the size of JET, which was built in a year. They expect to attempt net power in 2025 and a lot of researchers think they'll succeed. Assuming that works, as similar reactor the size of JET gets them to commercial output levels.
    - myrmidon 4y ago
      My point is that ITER is designed for 500MW (thermal) output. SPARC aims for something around 100MW (and is even further from continuous operation than ITER by design), and even the ARC successor (which is currently basically a pipe dream) is going to catch up with ITER in output power at best. So even IF the whole SPARC thing goes according to play, it's going to be miles away from anything resembling a competitive power plant, while ITER is potentially something like a quarter way there (assuming that operating at 2GW thermal begins being somewhat attractive/feasible for a fusion-powerplant). Personally, I strongly doubt that fusion power is EVER going to be an option because of simple economics; I just don't really see how a vacuum chamber surrounded by superconducting magnets, cooling systems, turbines and turbogenerators is ever going to compete (financially) with just slapping PV panels on a roof, putting some battery banks in the basement, hooking it up with an inverter and just repeating as necessary until power demands are met...
      - DennisP 4y ago
        Bigger is better only if it gives you a better ratio of capital expense to output. Due to the magnetic scaling I mentioned, SPARC/ARC probably does significantly better on that front.
        myrmidon 4y ago
        My point is that any commercial plant MUST be big enough to supply a generic steam turbine setup with heat; every fusion reactor design that cant achieve that is just a glorified toy, because it CANT operate economically. A bigger research reactor is simply better in that it enables you to investigate/solve problems that are related to scale, and all those MUST be solved before a commercial plant can be built. Until ARC or an even further removed successor catches up to ITER in size/power, the SPARC project is just yet another toy reactor IMO.
        DennisP 4y ago
        The US has 37 coal plants below 10MW, 73 plants from 50 to 100MW, and 85 plants between 100 and 250MW: https://www.gem.wiki/Existing_U.S._Coal_Plants#Size_comparison_of_coal_plants https://www.gem.wiki/Existing_U.S._Coal_Plants#Size_comparis... Coal plants of course use steam turbines: https://www.tva.com/energy/our-power-system/coal/how-a-coal-plant-works https://www.tva.com/energy/our-power-system/coal/how-a-coal-...
- foolfoolz 4y ago
  the U. S. has spent more on ukraine in the last year than on ITER since it’s inception. ITER isn’t slow because it’s big it’s slow because of how it’s funded
  - DennisP 4y ago
    But the US alone has spent more on ITER than CFS will spend on SPARC. This isn't entirely ITER's fault. The superconductor technology SPARC is using didn't exist when ITER was designed.

anigbrowl 4y ago

73%

sien 4y ago

On Metaculus : Most likely - fusion ignition in the 2030s. Fusion supplying 10% of global electricity in the 2050s. https://www.metaculus.com/questions/?order_by=-rank&main-feed&search=fusion https://www.metaculus.com/questions/?order_by=-rank&main-fee...

zardo 4y ago
Those two predictions don't seem to fit together.
- kortex 4y ago
  Why? Even once you figure out ignition, you still have a bunch of engineering issues to tackle. Energy extraction, tritium breeding, reactor longevity, reducing the build time and cost. Hopefully most of that can be figured out in parallel while we inch closer to ignition, but it's not like once we get ignition we can just copy-paste our way to actual commercial fusion power production.
  - zardo 4y ago
    Because going from the science project has finally worked to supplying 10% of the worlds electricity in just 20 years would be a truly incredible pace progress.
    - ForHackernews 4y ago
      Mobile phones went from 0.001% to 12.0% adoption from 1980 to 2000[0]. Things can happen quickly if there is very obviously lots of money to be made. [0] http://stats.areppim.com/stats/stats_mobilexpenetr.htm http://stats.areppim.com/stats/stats_mobilexpenetr.htm
      - zardo 4y ago
        > Things can happen quickly if there is very obviously lots of money to be made. There is not obviously lots of money to be made in rapidly building out fusion power plants.

mkaic 4y ago

90%. I'm not qualified in the slightest to make that prediction but I'm making it anyways!

fibonacc 4y ago

99% but petrodollar will still be in power. Access to fusion power will be strictly controlled.

idontpost 4y ago

0% chance.

ncmncm 4y ago

Exactly 0%. Achieving stable fusion does not make practical power from it possible. It would necessarily cost >10x fission, and fission is not competitive. We can be absolutely ,confident that by 2100, there will have been zero kWh of commercial power from Tokamak fusion. It is just barely possible that D-3He fusion might work by then, but it would still struggle to compete. It might find use in outer-solar-system spacecraft.