7 ms·

Firstly, your undersea cable is gonna cost a lot to build and maintain. Next, nuclear is expensive because the economies of scale are basically zero. Make a nu

Firstly, your undersea cable is gonna cost a lot to build and maintain. Next, nuclear is expensive because the economies of scale are basically zero. Make a nuclear power plant factory and you can get those numbers way down. Finally, cost probably isn't the right way to look at it. The issue at hand is mostly efficiency, where obviously the nuclear option wins handily.

ben_w 6y ago
> Make a nuclear power plant factory and you can get those numbers way down. Except we’ve never done that so we don’t know how big a scale we’d need to get costs down. In the absence of evidence, it’s entirely conceivable there isn’t enough economically recoverable uranium in the world for mass production of reactors to be cost effective. Therefore the only cost estimate we can rely on is what we have actually managed thus far. > Finally, cost probably isn't the right way to look at it. The issue at hand is mostly efficiency, where obviously the nuclear option wins handily. I disagree with both of these claims. For the first: for most of my life, the argument against green power, the reason nothing changed, was always the cost. Make the solution cheap and the problem goes away all by itself. For the second: so long as it is energy positive (EROI>>1, true for both), and the cost per unit is low (which you’re already dismissing), and you’re not resource constrained (we aren’t), I can’t see what other measure of efficiency matters much. The light-flux to electricity conversion efficiency? No more important than thermodynamic limits in a reactor heat engine. Resistive losses in transmission? I’ve just demonstrated that is irrelevant.
- fsflover 6y ago
  > Make the solution cheap and the problem goes away all by itself. If we also made the energy storage cheap then you would be right. But it's not cheap.
  - ben_w 6y ago
    Batteries combined with PV are already cheaper than nuclear, and grids of this scale and in these locations reduce the need for storage in the first place.
    - acidburnNSA 6y ago
      On the margin, PV and wind are cheaper than nuclear. At deep decarbonization, which implies sufficient long-term storage (not just 4 hours) and/or other backup options, it is not cheaper than nuclear. Furthermore only Germany has seen significant anti-renewable NIMBY groups so far (I believe they have over 300 anti-wind groups). At significant scale, these may proliferate even more and find ways to drive up costs (just like they did with nuclear). For some models showing these scenarios, see https://doi.org/10.1016/j.joule.2018.08.006 https://doi.org/10.1016/j.joule.2018.08.006
      - ben_w 6y ago
        Last time I ran the numbers, I was assuming 16 hours of battery (~sunset to sunrise, winter solstice, London), and those batteries needing replacement every 1000 nights. It still worked out in favour of PV, even without the suggestion — up in this thread — of connecting to the longer sunlight hours of the Sahara, which reduces need due to each of latitude, covering multiple time zones, and reduced cloud cover.
        VBprogrammer 6y ago
        Running the numbers from sunset to sunrise is extremely optimistic if you want 100% renewables. I've seen estimates of more like 4-7 days. However, that's not the biggest problem, current world wide lithium battery production is around 300GWh per year. Typical demand in the UK is around 30GW, even your 16 hours is around 1.5x the current battery manufacturing capabilities. You'd be competing against every car, laptop, tablet and phone manufacturer for that capacity.
        ben_w 6y ago
        Given battery factories are being built as fast as possible, and given the batteries in the cars are often suggested as overnight storage (and that transport is such a major part of overall power use), I disagree that this a problem. I would instead describe it as an economic opportunity. Also, the cost of grid storage batteries is dominated by the short lifespan rather than by the number of days or hours it can replace supply. If the battery is enough for a week of zero power from all sources in your entire grid, the wear from normal use takes longer to become a significant cost.
- sir_bearington 6y ago
  > Therefore the only cost estimate we can rely on is what we have actually managed thus far. Correct. And we've managed to bring the cost of nuclear down to 1-2 billion dollars per GW when building plants at scale [1]. This was done during the late 60s and early 70s when we built lots of nuclear plants at once [2]. 1. https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151... 2. https://ars.els-cdn.com/content/image/1-s2.0-S0301421516300106-gr2.jpg https://ars.els-cdn.com/content/image/1-s2.0-S03014215163001...
  - ben_w 6y ago
    > These results show that there is no single or intrinsic learning rate that we should expect for nuclear power technology, nor an expected cost trend. Your first link. Second link is invalid for me.
- freeflight 6y ago
  > In the absence of evidence, it’s entirely conceivable there isn’t enough economically recoverable uranium in the world for mass production of reactors to be cost effective. That was actually the official outlook of the US Atomic Energy Commission back in the early 1960: They feared the advent of the nuclear age would see world uranium supplies rapidly exhausted. This potential outlook is what triggered research into early attempts at thorium fuel cycles [0] [0] https://thebulletin.org/2014/05/thorium-the-wonder-fuel-that-wasnt/ https://thebulletin.org/2014/05/thorium-the-wonder-fuel-that...
maxerickson 6y ago
The cost is a measure of efficiency.
- fastball 6y ago
  In an ideal world, yes. In this case, not really.
Retric 6y ago
Nuclear’s construction costs aren’t the actual issue. Nuclear power plants need 24/7/365 security and highly trained staff, that’s not going to change if you start mass producing them. Raw fuel cost is low, but enrichment is expensive. Further decommissioning is again extremely expensive as is insurance etc. Dig down into the numbers and a 50% drop in construction costs wouldn’t make unsubsidized nuclear even close to cost competitive with unsubsidized wind or solar.
- fsflover 6y ago
  Care to show some sources?
  - Retric 6y ago
    If you want to do a deep dive into subsidized historical numbers this is a good starting place. https://www.nrc.gov/docs/ML0400/ML040070512.pdf https://www.nrc.gov/docs/ML0400/ML040070512.pdf You can get actual construction costs for comparison from various sources like Wikipedia. More recent US numbers are inflated due to the age of these reactors, but looking at costs over time shows they where never cheap to operate.
    - fsflover 6y ago
      2003. Are you sure it's up to date?
      - Retric 6y ago
        Yea, the US hasn’t been building nuclear reactors for the last 20 years. Here is the list: https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_States#Nuclear_power_plants https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_St... Excluding Watts Bar which is a very odd situation: The plant, construction of which began in 1973, has two Westinghouse pressurized water reactor units: Unit 1, completed in 1996, and Unit 2, completed in 2015. Unit 1 has a winter net dependable generating capacity of 1,167 megawatts. Unit 2 has a capacity of 1,165 megawatts. Both units are the newest operating civilian reactors to come online in the United States, and Unit 2 is the first and only new reactor to enter service in the 21st century in the US as of 2021. https://en.wikipedia.org/wiki/Watts_Bar_Nuclear_Plant https://en.wikipedia.org/wiki/Watts_Bar_Nuclear_Plant Their lifetime capacity factor is also 73.45% which makes generalizing based on them a poor choice.
fennecfoxen 6y ago
> Make a nuclear power plant factory and you can get those numbers way down. That's exactly what modern nuclear proposals look like. You make a nuclear power plant factory, mass-produce smallish mostly-sealed reactor units of a few hundred megawatts each, truck a few of the units to a power plant for installation, and truck them back to the factory occasionally for service. Look at what the DoE has to say about things like "Advanced Small Modular Reactors". Or https://en.wikipedia.org/wiki/Small_modular_reactor https://en.wikipedia.org/wiki/Small_modular_reactor
- acidburnNSA 6y ago
  You can also do large nuclear reactors in a factory if you use shipyard construction. In the 1970s Westinghouse and Newport News had a joint venture called Offshore Power Systems to do this. They employed 1000 people at their peak and did serious design and construction work of their nuclear plant manufacturing facility in Jacksonville Florida. In fact they actually got a manufacturing license to build eight gigawatt-class floating nuclear power plants in it from the NRC in 1982. https://www.sciencefriday.com/segments/floating-nuclear-power-plants/ https://www.sciencefriday.com/segments/floating-nuclear-powe... https://www.youtube.com/watch?v=0awiL0BeZ-k https://www.youtube.com/watch?v=0awiL0BeZ-k https://whatisnuclear.com/blog/2020-01-26-offshore-power-systems.html https://whatisnuclear.com/blog/2020-01-26-offshore-power-sys...
sir_bearington 6y ago
Nuclear's economy of scale is not at all "basically zero" even with traditional PWR reactors. Here's a graph of nuclear plant cost relative to construction year: https://ars.els-cdn.com/content/image/1-s2.0-S0301421516300106-gr2.jpg https://ars.els-cdn.com/content/image/1-s2.0-S03014215163001... [1] There's a big cluster of plants built for 1-2 billion per GW during the late 60s and early 70s. That's nuclear's economy of scale. 1. https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151...
- fastball 6y ago
  That is definitely not what I would call "scale".
  - sir_bearington 6y ago
    What do you mean? Over the course of a decade and a half we brought the share of nuclear generation to 20%. That's way more impressive than what intermittent sources have achieved.
    - ben_w 6y ago
      Persistently near-exponential growth over multiple decades isn’t impressive to you? If this carries on, peak PV output alone will exceed 100% of current total power demand by 2030. (Naturally this means I think the growth seen since 1992 won’t last until 2030, but the point remains)
      - sir_bearington 6y ago
        Assuming that exponential growth will be steady is a very risky assumption. Where's my single-core 100 GHz CPU? Moore's law says we should have that by now.
        ben_w 6y ago
        Hence second paragraph

ben_w 6y ago

> Make a nuclear power plant factory and you can get those numbers way down. Except we’ve never done that so we don’t know how big a scale we’d need to get costs down. In the absence of evidence, it’s entirely conceivable there isn’t enough economically recoverable uranium in the world for mass production of reactors to be cost effective. Therefore the only cost estimate we can rely on is what we have actually managed thus far. > Finally, cost probably isn't the right way to look at it. The issue at hand is mostly efficiency, where obviously the nuclear option wins handily. I disagree with both of these claims. For the first: for most of my life, the argument against green power, the reason nothing changed, was always the cost. Make the solution cheap and the problem goes away all by itself. For the second: so long as it is energy positive (EROI>>1, true for both), and the cost per unit is low (which you’re already dismissing), and you’re not resource constrained (we aren’t), I can’t see what other measure of efficiency matters much. The light-flux to electricity conversion efficiency? No more important than thermodynamic limits in a reactor heat engine. Resistive losses in transmission? I’ve just demonstrated that is irrelevant.

fsflover 6y ago
> Make the solution cheap and the problem goes away all by itself. If we also made the energy storage cheap then you would be right. But it's not cheap.
- ben_w 6y ago
  Batteries combined with PV are already cheaper than nuclear, and grids of this scale and in these locations reduce the need for storage in the first place.
  - acidburnNSA 6y ago
    On the margin, PV and wind are cheaper than nuclear. At deep decarbonization, which implies sufficient long-term storage (not just 4 hours) and/or other backup options, it is not cheaper than nuclear. Furthermore only Germany has seen significant anti-renewable NIMBY groups so far (I believe they have over 300 anti-wind groups). At significant scale, these may proliferate even more and find ways to drive up costs (just like they did with nuclear). For some models showing these scenarios, see https://doi.org/10.1016/j.joule.2018.08.006 https://doi.org/10.1016/j.joule.2018.08.006
    - ben_w 6y ago
      Last time I ran the numbers, I was assuming 16 hours of battery (~sunset to sunrise, winter solstice, London), and those batteries needing replacement every 1000 nights. It still worked out in favour of PV, even without the suggestion — up in this thread — of connecting to the longer sunlight hours of the Sahara, which reduces need due to each of latitude, covering multiple time zones, and reduced cloud cover.
      - VBprogrammer 6y ago
        Running the numbers from sunset to sunrise is extremely optimistic if you want 100% renewables. I've seen estimates of more like 4-7 days. However, that's not the biggest problem, current world wide lithium battery production is around 300GWh per year. Typical demand in the UK is around 30GW, even your 16 hours is around 1.5x the current battery manufacturing capabilities. You'd be competing against every car, laptop, tablet and phone manufacturer for that capacity.
        ben_w 6y ago
        Given battery factories are being built as fast as possible, and given the batteries in the cars are often suggested as overnight storage (and that transport is such a major part of overall power use), I disagree that this a problem. I would instead describe it as an economic opportunity. Also, the cost of grid storage batteries is dominated by the short lifespan rather than by the number of days or hours it can replace supply. If the battery is enough for a week of zero power from all sources in your entire grid, the wear from normal use takes longer to become a significant cost.
sir_bearington 6y ago
> Therefore the only cost estimate we can rely on is what we have actually managed thus far. Correct. And we've managed to bring the cost of nuclear down to 1-2 billion dollars per GW when building plants at scale [1]. This was done during the late 60s and early 70s when we built lots of nuclear plants at once [2]. 1. https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151... 2. https://ars.els-cdn.com/content/image/1-s2.0-S0301421516300106-gr2.jpg https://ars.els-cdn.com/content/image/1-s2.0-S03014215163001...
- ben_w 6y ago
  > These results show that there is no single or intrinsic learning rate that we should expect for nuclear power technology, nor an expected cost trend. Your first link. Second link is invalid for me.
freeflight 6y ago
> In the absence of evidence, it’s entirely conceivable there isn’t enough economically recoverable uranium in the world for mass production of reactors to be cost effective. That was actually the official outlook of the US Atomic Energy Commission back in the early 1960: They feared the advent of the nuclear age would see world uranium supplies rapidly exhausted. This potential outlook is what triggered research into early attempts at thorium fuel cycles [0] [0] https://thebulletin.org/2014/05/thorium-the-wonder-fuel-that-wasnt/ https://thebulletin.org/2014/05/thorium-the-wonder-fuel-that...

maxerickson 6y ago

The cost is a measure of efficiency.

fastball 6y ago
In an ideal world, yes. In this case, not really.

Retric 6y ago

Nuclear’s construction costs aren’t the actual issue. Nuclear power plants need 24/7/365 security and highly trained staff, that’s not going to change if you start mass producing them. Raw fuel cost is low, but enrichment is expensive. Further decommissioning is again extremely expensive as is insurance etc. Dig down into the numbers and a 50% drop in construction costs wouldn’t make unsubsidized nuclear even close to cost competitive with unsubsidized wind or solar.

fsflover 6y ago
Care to show some sources?
- Retric 6y ago
  If you want to do a deep dive into subsidized historical numbers this is a good starting place. https://www.nrc.gov/docs/ML0400/ML040070512.pdf https://www.nrc.gov/docs/ML0400/ML040070512.pdf You can get actual construction costs for comparison from various sources like Wikipedia. More recent US numbers are inflated due to the age of these reactors, but looking at costs over time shows they where never cheap to operate.
  - fsflover 6y ago
    2003. Are you sure it's up to date?
    - Retric 6y ago
      Yea, the US hasn’t been building nuclear reactors for the last 20 years. Here is the list: https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_States#Nuclear_power_plants https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_St... Excluding Watts Bar which is a very odd situation: The plant, construction of which began in 1973, has two Westinghouse pressurized water reactor units: Unit 1, completed in 1996, and Unit 2, completed in 2015. Unit 1 has a winter net dependable generating capacity of 1,167 megawatts. Unit 2 has a capacity of 1,165 megawatts. Both units are the newest operating civilian reactors to come online in the United States, and Unit 2 is the first and only new reactor to enter service in the 21st century in the US as of 2021. https://en.wikipedia.org/wiki/Watts_Bar_Nuclear_Plant https://en.wikipedia.org/wiki/Watts_Bar_Nuclear_Plant Their lifetime capacity factor is also 73.45% which makes generalizing based on them a poor choice.

fennecfoxen 6y ago

> Make a nuclear power plant factory and you can get those numbers way down. That's exactly what modern nuclear proposals look like. You make a nuclear power plant factory, mass-produce smallish mostly-sealed reactor units of a few hundred megawatts each, truck a few of the units to a power plant for installation, and truck them back to the factory occasionally for service. Look at what the DoE has to say about things like "Advanced Small Modular Reactors". Or https://en.wikipedia.org/wiki/Small_modular_reactor https://en.wikipedia.org/wiki/Small_modular_reactor

acidburnNSA 6y ago
You can also do large nuclear reactors in a factory if you use shipyard construction. In the 1970s Westinghouse and Newport News had a joint venture called Offshore Power Systems to do this. They employed 1000 people at their peak and did serious design and construction work of their nuclear plant manufacturing facility in Jacksonville Florida. In fact they actually got a manufacturing license to build eight gigawatt-class floating nuclear power plants in it from the NRC in 1982. https://www.sciencefriday.com/segments/floating-nuclear-power-plants/ https://www.sciencefriday.com/segments/floating-nuclear-powe... https://www.youtube.com/watch?v=0awiL0BeZ-k https://www.youtube.com/watch?v=0awiL0BeZ-k https://whatisnuclear.com/blog/2020-01-26-offshore-power-systems.html https://whatisnuclear.com/blog/2020-01-26-offshore-power-sys...

sir_bearington 6y ago

Nuclear's economy of scale is not at all "basically zero" even with traditional PWR reactors. Here's a graph of nuclear plant cost relative to construction year: https://ars.els-cdn.com/content/image/1-s2.0-S0301421516300106-gr2.jpg https://ars.els-cdn.com/content/image/1-s2.0-S03014215163001... [1] There's a big cluster of plants built for 1-2 billion per GW during the late 60s and early 70s. That's nuclear's economy of scale. 1. https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151...

fastball 6y ago
That is definitely not what I would call "scale".
- sir_bearington 6y ago
  What do you mean? Over the course of a decade and a half we brought the share of nuclear generation to 20%. That's way more impressive than what intermittent sources have achieved.
  - ben_w 6y ago
    Persistently near-exponential growth over multiple decades isn’t impressive to you? If this carries on, peak PV output alone will exceed 100% of current total power demand by 2030. (Naturally this means I think the growth seen since 1992 won’t last until 2030, but the point remains)
    - sir_bearington 6y ago
      Assuming that exponential growth will be steady is a very risky assumption. Where's my single-core 100 GHz CPU? Moore's law says we should have that by now.
      - ben_w 6y ago
        Hence second paragraph