3 ms·

The world uses about 556.63 exajoules per year. Say your nuclear reactor produces 3GW. That's more than 5000 reactors to power the world. You can probably safel

The world uses about 556.63 exajoules per year. Say your nuclear reactor produces 3GW. That's more than 5000 reactors to power the world. You can probably safely cut that in half because electricity is a better source of energy for transportation and heating that fossil fuels, but still. Do you think we can scale reactor production sufficiently to build thousands of reactors in two or three decades?

orthecreedence 5y ago
> Do you think we can scale reactor production sufficiently to build thousands of reactors in two or three decades? Humans can do some cool things when we decide to. And scaling solar isn't as easy as it sounds. What's the cost, in energy and fossil fuels, to mine the materials, build and ship not just the panels but also the massive batteries required for solar? Not to mention having to constantly replace them? That's something I don't really see solar advocates addressing: if we have to burn every last gallon of oil to build our marvelous solar and battery arrays, then the planet is screwed anyway. Nuclear has much higher energy density. The problem is our economic system isn't accounting for energy output vs input...it uses prices, which are incredibly useless in this particular instance because prices do not internalize costs. Price and cost have significantly diverged, so you can't look at "what's cheaper in price" and assume that it's the optimal outcome. Why not attack from all angles? Build solar, but also crank out as many nuclear reactors as humanly possible.
- adrianN 5y ago
  I don't know how much more material you need for solar panels and wind turbines compared to nuclear reactors (there's a lot of concrete in a nuclear reactor too). But as you build more renewables the amount of fossil fuels you need for manufacturing goes down. It would of course be nice if externalities were better reflected in prices.
- jhgb 5y ago
  > Humans can do some cool things when we decide to. And scaling solar isn't as easy as it sounds. Weirdly enough, we're already currently annually installing so much solar power (in terms of actual energy output, not just nameplate) that it equates the progress accomplished in the nuclear industry in the past fifteen years or so. In other words, the nuclear industry would have to accelerate by more than an order to magnitude just to catch up with today's solar power growth. But of course solar power growth itself increases by around 30% every year on average... To me that almost sounds like it's scaling nuclear power that isn't as easy as it sounds. > but also the massive batteries required for solar Yeah, there's no such thing as "massive batteries required for solar" in a proper high-renewable grid. Simulations show (https://doi.org/10.1016/j.euroecorev.2018.07.004 https://doi.org/10.1016/j.euroecorev.2018.07.004) that in a cost-optimal grid mix, pumped hydro is economically preferred over batteries.
  - orthecreedence 5y ago
    > pumped hydro is economically preferred over batteries. Which is a massive battery.
    - jhgb 5y ago
      If by "battery", you mean "a primary or secondary electrochemical cell or cells", as is its usual meaning, then it's not.
panick21_ 5y ago
Yes. 5000 reactors is not actually that much. Think about how many planes we produce. Imagine 5 Gigafactory each producing 100s reactor cores every year. That not really hard to imagine is it? Those would more likely be 500MW. So you need a few 10000+. The rest of a nuclear plant without water cooling is none nuclear and building a concrete hole, and a concrete bunker a whole bunch of electronics, pipes and the same turbine used for gas plants. And we have tons of thorium all over the place for basically free in every country. Look at fast France managed to make their nuclear, and that was with 60s technology.
- adrianN 5y ago
  France installed less than 60 reactors over 15 years, I think they aimed for 80. Even just constructing "a few 10k+" fossil fuel plants over say 30 years seems really ambitious to me. Each is a fairly large construction site that binds specialized construction skill for several years. In many countries you would also face prolonged legal battles before you can even start construction.
- jhgb 5y ago
  For comparison, assuming a "business as usual" situation in the solar power industry, the electricity output of five thousand 3GWt nuclear units would be reached around 2040 (maybe 2035 if the long-term annual growth factor of 30% were to be maintained, but that's unlikely in the future).
NumberWangMan 5y ago
Good question - note that the ThorCon project is not a nuclear power plant, but a factory for producing nuclear power plants (class NuclearPlantFactory? :) ). From their site: "A single shipyard can produce 20 GW’s of ThorConIsle power per year. In terms of resource requirements, one gigawatt of ThorCon power is not a big deal. The scale up rate will not be limited by shipyard capacity, but by the rate at which the turbogenerators can be built." So let's say 18000 GW (converting 556 exajoules/year and rounding up). Divide by 20 (GW per shipyard per year) to get 900 ThorCon-shipyard-years. If we say two decades to get it done, that's 45 shipyards, which seems doable? There are 285 active shipyards worldwide right now. But anyway, like you said, we can maybe cut it in half, and not everything has to be powered by nuclear, just enough to replace fossil fuels. On the other side, many areas of the world are energy poor and impoverished because of that. It would be great if we could use a lot of this energy to improve lives of people living in these areas, so we would probably want to build substantially more.

orthecreedence 5y ago

> Do you think we can scale reactor production sufficiently to build thousands of reactors in two or three decades? Humans can do some cool things when we decide to. And scaling solar isn't as easy as it sounds. What's the cost, in energy and fossil fuels, to mine the materials, build and ship not just the panels but also the massive batteries required for solar? Not to mention having to constantly replace them? That's something I don't really see solar advocates addressing: if we have to burn every last gallon of oil to build our marvelous solar and battery arrays, then the planet is screwed anyway. Nuclear has much higher energy density. The problem is our economic system isn't accounting for energy output vs input...it uses prices, which are incredibly useless in this particular instance because prices do not internalize costs. Price and cost have significantly diverged, so you can't look at "what's cheaper in price" and assume that it's the optimal outcome. Why not attack from all angles? Build solar, but also crank out as many nuclear reactors as humanly possible.

adrianN 5y ago
I don't know how much more material you need for solar panels and wind turbines compared to nuclear reactors (there's a lot of concrete in a nuclear reactor too). But as you build more renewables the amount of fossil fuels you need for manufacturing goes down. It would of course be nice if externalities were better reflected in prices.
jhgb 5y ago
> Humans can do some cool things when we decide to. And scaling solar isn't as easy as it sounds. Weirdly enough, we're already currently annually installing so much solar power (in terms of actual energy output, not just nameplate) that it equates the progress accomplished in the nuclear industry in the past fifteen years or so. In other words, the nuclear industry would have to accelerate by more than an order to magnitude just to catch up with today's solar power growth. But of course solar power growth itself increases by around 30% every year on average... To me that almost sounds like it's scaling nuclear power that isn't as easy as it sounds. > but also the massive batteries required for solar Yeah, there's no such thing as "massive batteries required for solar" in a proper high-renewable grid. Simulations show (https://doi.org/10.1016/j.euroecorev.2018.07.004 https://doi.org/10.1016/j.euroecorev.2018.07.004) that in a cost-optimal grid mix, pumped hydro is economically preferred over batteries.
- orthecreedence 5y ago
  > pumped hydro is economically preferred over batteries. Which is a massive battery.
  - jhgb 5y ago
    If by "battery", you mean "a primary or secondary electrochemical cell or cells", as is its usual meaning, then it's not.

panick21_ 5y ago

Yes. 5000 reactors is not actually that much. Think about how many planes we produce. Imagine 5 Gigafactory each producing 100s reactor cores every year. That not really hard to imagine is it? Those would more likely be 500MW. So you need a few 10000+. The rest of a nuclear plant without water cooling is none nuclear and building a concrete hole, and a concrete bunker a whole bunch of electronics, pipes and the same turbine used for gas plants. And we have tons of thorium all over the place for basically free in every country. Look at fast France managed to make their nuclear, and that was with 60s technology.

adrianN 5y ago
France installed less than 60 reactors over 15 years, I think they aimed for 80. Even just constructing "a few 10k+" fossil fuel plants over say 30 years seems really ambitious to me. Each is a fairly large construction site that binds specialized construction skill for several years. In many countries you would also face prolonged legal battles before you can even start construction.
jhgb 5y ago
For comparison, assuming a "business as usual" situation in the solar power industry, the electricity output of five thousand 3GWt nuclear units would be reached around 2040 (maybe 2035 if the long-term annual growth factor of 30% were to be maintained, but that's unlikely in the future).

NumberWangMan 5y ago

Good question - note that the ThorCon project is not a nuclear power plant, but a factory for producing nuclear power plants (class NuclearPlantFactory? :) ). From their site: "A single shipyard can produce 20 GW’s of ThorConIsle power per year. In terms of resource requirements, one gigawatt of ThorCon power is not a big deal. The scale up rate will not be limited by shipyard capacity, but by the rate at which the turbogenerators can be built." So let's say 18000 GW (converting 556 exajoules/year and rounding up). Divide by 20 (GW per shipyard per year) to get 900 ThorCon-shipyard-years. If we say two decades to get it done, that's 45 shipyards, which seems doable? There are 285 active shipyards worldwide right now. But anyway, like you said, we can maybe cut it in half, and not everything has to be powered by nuclear, just enough to replace fossil fuels. On the other side, many areas of the world are energy poor and impoverished because of that. It would be great if we could use a lot of this energy to improve lives of people living in these areas, so we would probably want to build substantially more.