4 ms·

Actually they have raised (a round lead by Sam Altman) $500M to demo net electricity by 2024. Then their plans are to have a factory to build 20 50MW-generator

by joak 4y ago

Actually they have raised (a round lead by Sam Altman) $500M to demo net electricity by 2024. Then their plans are to have a factory to build 20 50MW-generators per day. Each generator is shipping-container size. Link to the interview in local press here: https://news.ycombinator.com/item?id=32088113 https://news.ycombinator.com/item?id=32088113

onlyrealcuzzo 4y ago
2GW of energy per day? Isn't that like a medium-sized nuclear reactor every day?
- zdragnar 4y ago
  A significant cost driver of most energy sources is the work involved in dealing with the high temperatures and high pressures- cooling towers, steam chambers and turbines that can withstand superheated, high pressure steam, etc. It's a lot of concrete, thick steel walls and safety regulations as each one is basically a small bomb should the containment fail. If they can actually produce a 50 megawatt generator the size of a shipping container and requiring inexpensive, easy to source fuel, with sane emissions, every power producer in the world will be chomping at the bit to buy them. My state is currently in the process of putting up solar panel fields on the sites of coal power plants that are being decommissioned, because it is getting harder to find suitable locations with cooperative local governments. Many see them as eyesores, along with the added trouble of putting in new HVDC lines out to the fields. If instead you could get significantly more power, and equally clean, from the same site, power companies will ditch solar in a heartbeat. No need to buy more land or run new lines, just decommission your coal and nat gas power plants and put up a few of these.
ncmncm 4y ago
Their big problem is not manufacturing, not personnel, not extraction of useful energy from kinetic motion of charged particles, and not even physics of light-nucleus fusion. Those are, in principle, probably all solvable. Their big problem is that they depend for fuel on 3He, and there isn't any. Not on Earth, and not even, usefully, in the lunar regolith. (By the way, there should be a better name for 3He as there is for 3H, tritium. "Helionium"?) There is one source for 3He, and that is decay of 3H, tritium. Tritium can be made in high-neutron-flux processes such as in a CANDU reactor, slowing the neutrons with heavy water until they fuse with the 2H deuterium atoms in it. Then, keep it for 12+ years waiting for it to decay. We actually do manufacture a noticeable amount of tritium, and then store it in missile silos for the requisite 12 years, purifying out the 3He half periodically and topping up the tanks. Tritium is stored there ready to be funneled into the middle of a hydrogen bomb shortly before lofting to its unlucky recipients. Operating these things is supposed to produce a little bit of 3H, which in principle could be collected, banked, and harvested. But each 3He they burn is very unlikely to make the 3H they will need to get another 3He some years on. They have to extract, anyway, the tritium they make, to avoid fusing those by accident, producing useless and harmful hot neutrons. There is a competing technology, that has been worked on at Princeton and generically labeled "field reversed configuration", FRC, also burning D-3He, optimized for spacecraft propulsion. They hope to loft a test article by 2035. So there are a few practical uses for this technology, if it can be made to work at all: to power spacecraft exploring the outer solar system, and maybe even for a moon base; but not for practical energy here on Earth. For energy here, solar + wind + storage are the key to forestalling civilizational collapse. It might still be possible to prevent worldwide collapse if we radically increase the investment rate in these known-practical technologies. But not if we don't, and we haven't, enough. If civilization can be saved from collapsing, this tech could become interesting someday. Otherwise it will stall when civilization collapses out from under it. It is safe to say that would be a shame.
- jamesfisher 4y ago
  Surely "Threlium"?
  - ncmncm 4y ago
    Trilium. Android autocorrect thinks there is a word "trillium" already. So, maybe trelium or terlium. Or live with it, as we do lama and llama. Dreilium? Internet would probably go all in for threelium. Or heliumy-mc-heliumface.
    - avmich 4y ago
      Just re-use the same word. With different meaning. Big deal.
- joak 4y ago
  Actually their only primary fuel is deuterium. The fusion reactions are 2/3 of D+D and 1/3 of D+He3 D+D fusion reaction yields 50% of T+p (tritium and a proton) and 50% of He3 + neutron So in total their fusion reaction does not consume He3. He3 is here like a catalyzer. They still need some He3. The D-D reaction produces tritium, and the tritium decays into He3. Half life of T is 12.5 years. Done, no issues with the helium 3. Actually they will probably produce T in excess and could sell it.
  - ncmncm 4y ago
    Running deuterium-rich enough to net significant 3He out would seem to emit enough neutrons to render the reactor vessel strongly radioactive, and require the sort of cooling and shielding that makes the pure D-D reactors being pursued by others grossly impractical.
    - joak 4y ago
      As I understand the neutrons of the D+D->He3+n reaction are slow neutrons (2.45MeV) that won't damage too much the materials. These are not the fast neutrons of the DT fusion (14.06MeV) They claim it won't be worse than waste neutrons of radiation therapy medical devices. To be seen... But yes, agree this is not exactly aneutronic fusion. Around 10% of energy is lost as slow neutrons and this is definitely a problem to address.

by joak 4y ago

onlyrealcuzzo 4y ago

2GW of energy per day? Isn't that like a medium-sized nuclear reactor every day?

zdragnar 4y ago
A significant cost driver of most energy sources is the work involved in dealing with the high temperatures and high pressures- cooling towers, steam chambers and turbines that can withstand superheated, high pressure steam, etc. It's a lot of concrete, thick steel walls and safety regulations as each one is basically a small bomb should the containment fail. If they can actually produce a 50 megawatt generator the size of a shipping container and requiring inexpensive, easy to source fuel, with sane emissions, every power producer in the world will be chomping at the bit to buy them. My state is currently in the process of putting up solar panel fields on the sites of coal power plants that are being decommissioned, because it is getting harder to find suitable locations with cooperative local governments. Many see them as eyesores, along with the added trouble of putting in new HVDC lines out to the fields. If instead you could get significantly more power, and equally clean, from the same site, power companies will ditch solar in a heartbeat. No need to buy more land or run new lines, just decommission your coal and nat gas power plants and put up a few of these.

ncmncm 4y ago

Their big problem is not manufacturing, not personnel, not extraction of useful energy from kinetic motion of charged particles, and not even physics of light-nucleus fusion. Those are, in principle, probably all solvable. Their big problem is that they depend for fuel on 3He, and there isn't any. Not on Earth, and not even, usefully, in the lunar regolith. (By the way, there should be a better name for 3He as there is for 3H, tritium. "Helionium"?) There is one source for 3He, and that is decay of 3H, tritium. Tritium can be made in high-neutron-flux processes such as in a CANDU reactor, slowing the neutrons with heavy water until they fuse with the 2H deuterium atoms in it. Then, keep it for 12+ years waiting for it to decay. We actually do manufacture a noticeable amount of tritium, and then store it in missile silos for the requisite 12 years, purifying out the 3He half periodically and topping up the tanks. Tritium is stored there ready to be funneled into the middle of a hydrogen bomb shortly before lofting to its unlucky recipients. Operating these things is supposed to produce a little bit of 3H, which in principle could be collected, banked, and harvested. But each 3He they burn is very unlikely to make the 3H they will need to get another 3He some years on. They have to extract, anyway, the tritium they make, to avoid fusing those by accident, producing useless and harmful hot neutrons. There is a competing technology, that has been worked on at Princeton and generically labeled "field reversed configuration", FRC, also burning D-3He, optimized for spacecraft propulsion. They hope to loft a test article by 2035. So there are a few practical uses for this technology, if it can be made to work at all: to power spacecraft exploring the outer solar system, and maybe even for a moon base; but not for practical energy here on Earth. For energy here, solar + wind + storage are the key to forestalling civilizational collapse. It might still be possible to prevent worldwide collapse if we radically increase the investment rate in these known-practical technologies. But not if we don't, and we haven't, enough. If civilization can be saved from collapsing, this tech could become interesting someday. Otherwise it will stall when civilization collapses out from under it. It is safe to say that would be a shame.

jamesfisher 4y ago
Surely "Threlium"?
- ncmncm 4y ago
  Trilium. Android autocorrect thinks there is a word "trillium" already. So, maybe trelium or terlium. Or live with it, as we do lama and llama. Dreilium? Internet would probably go all in for threelium. Or heliumy-mc-heliumface.
  - avmich 4y ago
    Just re-use the same word. With different meaning. Big deal.
joak 4y ago
Actually their only primary fuel is deuterium. The fusion reactions are 2/3 of D+D and 1/3 of D+He3 D+D fusion reaction yields 50% of T+p (tritium and a proton) and 50% of He3 + neutron So in total their fusion reaction does not consume He3. He3 is here like a catalyzer. They still need some He3. The D-D reaction produces tritium, and the tritium decays into He3. Half life of T is 12.5 years. Done, no issues with the helium 3. Actually they will probably produce T in excess and could sell it.
- ncmncm 4y ago
  Running deuterium-rich enough to net significant 3He out would seem to emit enough neutrons to render the reactor vessel strongly radioactive, and require the sort of cooling and shielding that makes the pure D-D reactors being pursued by others grossly impractical.
  - joak 4y ago
    As I understand the neutrons of the D+D->He3+n reaction are slow neutrons (2.45MeV) that won't damage too much the materials. These are not the fast neutrons of the DT fusion (14.06MeV) They claim it won't be worse than waste neutrons of radiation therapy medical devices. To be seen... But yes, agree this is not exactly aneutronic fusion. Around 10% of energy is lost as slow neutrons and this is definitely a problem to address.