5 ms·

Last time around it was negative learning by doing for France[1]. I wonder what magical efficiency gains they will manage to muster compared to Flamanville [2],

Last time around it was negative learning by doing for France[1]. I wonder what magical efficiency gains they will manage to muster compared to Flamanville [2], Hinkley Point C[3] and Olkiluoto 3[4] which the state owned french nuclear industry is building. Well, more than magical "simplification", "cost cutting" and "modular design" shown on a powerpoint before reality hits again. [1]: https://www.sciencedirect.com/science/article/abs/pii/S0301421510003526 https://www.sciencedirect.com/science/article/abs/pii/S03014... [2]: https://en.wikipedia.org/wiki/EPR_(nuclear_reactor)#Flamanville_3_(France) https://en.wikipedia.org/wiki/EPR_(nuclear_reactor)#Flamanvi... [3]: https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_station https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_... [4]: https://en.wikipedia.org/wiki/Olkiluoto_Nuclear_Power_Plant#Unit_3 https://en.wikipedia.org/wiki/Olkiluoto_Nuclear_Power_Plant#...

panick21_ 5y ago
Nuclear reactors have massive learning effects, the more you build with the same people and organization the cheaper it is. Nuclear own massive scale is its own worst enemy. If we could just build 500MW smaller Molten Salt cooled reactors we could literally build them like gas plants.
- Gwypaas 5y ago
  The source regarding actually seeing negative learning effects I linked in the original comment states in the abstract. > The French nuclear case illustrates the perils of the assumption of robust learning effects resulting in lowered costs over time in the scale-up of large-scale, complex new energy supply technologies. The uncertainties in anticipated learning effects of new technologies might be much larger that often assumed, including also cases of “negative learning” in which specific costs increase rather than decrease with accumulated experience. Sure, a nuclear cycle without the use of a steam turbine may have a future. Similarly to how gas plants undercut coal, and nuclear. Simply due to the cost of the steam plant. I haven't seen any proposals though which is more concrete than a pie-in-sky powerpoint design though. The other issue is that for wind you only need an axle and a generator, for sun it is solid state. It is hard to compete with the economics of solid state power generation.
  - panick21_ 5y ago
    What matter to me is that within one generation they totally de-carbonized the grid from oil and coal to nuclear and have successfully been in running it that way for decades. Germany has been doing a Green New Deal style thing with Renwables for 2 decades now, in my opinion they could have easily been heading to 100% CO2 free by now if they had just started to build nuclear plants. The French certainty made mistakes and unfortunately technology innovation and improvement and continued expatiation of the products on a global scale didn't happen. Of course I don't want to relay on learning effects from these large complex projects, but its better then just extrapolating from individual first time builds. We defiantly want to move to smaller AND more efficient nuclear plants, but for that we also need nuclear industry, nuclear education a competent nuclear regulator and a government that actually puts its resources in that place. > The other issue is that for wind you only need an axle and a generator, for sun it is solid state. It is hard to compete with the economics of solid state power generation. That is one consideration. The real question is all in, 100% making sure all citizens have as much power as they need at every moment no matter what happens. If you approach it like that, going a all nuclear route with some localized batteries for peak shaving would be the overall cheapest solution for a largish industrial economy. You just replace current coal plants with nuclear plants and use the same infrastructure. Doing a fully renewable massive industrial economy is totally unsolved problem requiring storage, smart grids and so on. And even then, when you consider the loses of solar in the case of a vulcano for example, you probably need some kind of gas backup. Had the world at Kyoto just said, lets everybody build 1 nuclear plant per 1 million people, we would have de-carbonized by now.
    - mrep 5y ago
      Nuclear looks to be about 3x as expensive as wind/solar [0] and that is assuming you use it at 100% of its capacity due to the high building cost and low marginal costs. Just looking at California, winter nights drop to about half of electric demand as summer days [1] so you'd have to pay even more for spare capacity. I think it would far cheaper to just build like 2x the wind/solar you need and use the surpluses to make methane [2] which is easily stored and already used for 40% of the US electrical grid [3]. [0]: https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#/media/File:3-Learning-curves-for-electricity-prices.png https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#... [1]: http://www.caiso.com/TodaysOutlook/Pages/default.aspx http://www.caiso.com/TodaysOutlook/Pages/default.aspx [2]: https://en.wikipedia.org/wiki/Power-to-gas https://en.wikipedia.org/wiki/Power-to-gas [3]: https://en.wikipedia.org/wiki/Electricity_sector_of_the_United_States https://en.wikipedia.org/wiki/Electricity_sector_of_the_Unit...
rsynnott 5y ago
> The plant, which has a projected lifetime of 60 years, had an estimated construction cost of between £19.6 billion and £20.3 billion > According to January 2021 estimates, the expected operational start date is June 2026 with a build cost of £22–23 billion That seems like a... fairly modest overrun for a megaproject? Flamanville and Olkiluoto were the first two of a new design. Historically, that rarely goes well. Taishan went a lot better, and Hinkley Point is basically on track.
- Gwypaas 5y ago
  It's always nice to be an authoritarian state > CNN said French energy firm EDF, which helps run the site, had warned the US government that China's nuclear regulator had raised limits on permissible levels of radiation outside the plant to avoid shutting it down. https://www.bbc.com/news/world-asia-china-58026038 https://www.bbc.com/news/world-asia-china-58026038 Hinkley Point C is looking to be an at least £50 billion cool transfer from taxes to private corporations. Renewables have also gotten way cheaper today compared to 2017. > EDF has negotiated a guaranteed fixed price – a "strike price" – for electricity from Hinkley Point C of £92.50/MWh (in 2012 prices),[20][77] which will be adjusted (linked to inflation – £106/MWh by 2021[71]) during the construction period and over the subsequent 35 years tariff period. The base strike price could fall to £89.50/MWh if a new plant at Sizewell is also approved.[20][77] High consumer prices for energy will hit the poorest consumers hardest according to the Public Accounts Committee.[81] > In July 2016, the National Audit Office estimated that due to falling energy costs, the additional cost to consumers of 'future top-up payments under the proposed HPC CfD had increased from £6.1 billion in October 2013, when the strike price was agreed, to £29.7 billion'.[82][83] In July 2017, this estimate rose to £50 billion, or 'more than eight times the 2013 estimate'.[9]

panick21_ 5y ago

Nuclear reactors have massive learning effects, the more you build with the same people and organization the cheaper it is. Nuclear own massive scale is its own worst enemy. If we could just build 500MW smaller Molten Salt cooled reactors we could literally build them like gas plants.

Gwypaas 5y ago
The source regarding actually seeing negative learning effects I linked in the original comment states in the abstract. > The French nuclear case illustrates the perils of the assumption of robust learning effects resulting in lowered costs over time in the scale-up of large-scale, complex new energy supply technologies. The uncertainties in anticipated learning effects of new technologies might be much larger that often assumed, including also cases of “negative learning” in which specific costs increase rather than decrease with accumulated experience. Sure, a nuclear cycle without the use of a steam turbine may have a future. Similarly to how gas plants undercut coal, and nuclear. Simply due to the cost of the steam plant. I haven't seen any proposals though which is more concrete than a pie-in-sky powerpoint design though. The other issue is that for wind you only need an axle and a generator, for sun it is solid state. It is hard to compete with the economics of solid state power generation.
- panick21_ 5y ago
  What matter to me is that within one generation they totally de-carbonized the grid from oil and coal to nuclear and have successfully been in running it that way for decades. Germany has been doing a Green New Deal style thing with Renwables for 2 decades now, in my opinion they could have easily been heading to 100% CO2 free by now if they had just started to build nuclear plants. The French certainty made mistakes and unfortunately technology innovation and improvement and continued expatiation of the products on a global scale didn't happen. Of course I don't want to relay on learning effects from these large complex projects, but its better then just extrapolating from individual first time builds. We defiantly want to move to smaller AND more efficient nuclear plants, but for that we also need nuclear industry, nuclear education a competent nuclear regulator and a government that actually puts its resources in that place. > The other issue is that for wind you only need an axle and a generator, for sun it is solid state. It is hard to compete with the economics of solid state power generation. That is one consideration. The real question is all in, 100% making sure all citizens have as much power as they need at every moment no matter what happens. If you approach it like that, going a all nuclear route with some localized batteries for peak shaving would be the overall cheapest solution for a largish industrial economy. You just replace current coal plants with nuclear plants and use the same infrastructure. Doing a fully renewable massive industrial economy is totally unsolved problem requiring storage, smart grids and so on. And even then, when you consider the loses of solar in the case of a vulcano for example, you probably need some kind of gas backup. Had the world at Kyoto just said, lets everybody build 1 nuclear plant per 1 million people, we would have de-carbonized by now.
  - mrep 5y ago
    Nuclear looks to be about 3x as expensive as wind/solar [0] and that is assuming you use it at 100% of its capacity due to the high building cost and low marginal costs. Just looking at California, winter nights drop to about half of electric demand as summer days [1] so you'd have to pay even more for spare capacity. I think it would far cheaper to just build like 2x the wind/solar you need and use the surpluses to make methane [2] which is easily stored and already used for 40% of the US electrical grid [3]. [0]: https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#/media/File:3-Learning-curves-for-electricity-prices.png https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#... [1]: http://www.caiso.com/TodaysOutlook/Pages/default.aspx http://www.caiso.com/TodaysOutlook/Pages/default.aspx [2]: https://en.wikipedia.org/wiki/Power-to-gas https://en.wikipedia.org/wiki/Power-to-gas [3]: https://en.wikipedia.org/wiki/Electricity_sector_of_the_United_States https://en.wikipedia.org/wiki/Electricity_sector_of_the_Unit...

rsynnott 5y ago

> The plant, which has a projected lifetime of 60 years, had an estimated construction cost of between £19.6 billion and £20.3 billion > According to January 2021 estimates, the expected operational start date is June 2026 with a build cost of £22–23 billion That seems like a... fairly modest overrun for a megaproject? Flamanville and Olkiluoto were the first two of a new design. Historically, that rarely goes well. Taishan went a lot better, and Hinkley Point is basically on track.

Gwypaas 5y ago
It's always nice to be an authoritarian state > CNN said French energy firm EDF, which helps run the site, had warned the US government that China's nuclear regulator had raised limits on permissible levels of radiation outside the plant to avoid shutting it down. https://www.bbc.com/news/world-asia-china-58026038 https://www.bbc.com/news/world-asia-china-58026038 Hinkley Point C is looking to be an at least £50 billion cool transfer from taxes to private corporations. Renewables have also gotten way cheaper today compared to 2017. > EDF has negotiated a guaranteed fixed price – a "strike price" – for electricity from Hinkley Point C of £92.50/MWh (in 2012 prices),[20][77] which will be adjusted (linked to inflation – £106/MWh by 2021[71]) during the construction period and over the subsequent 35 years tariff period. The base strike price could fall to £89.50/MWh if a new plant at Sizewell is also approved.[20][77] High consumer prices for energy will hit the poorest consumers hardest according to the Public Accounts Committee.[81] > In July 2016, the National Audit Office estimated that due to falling energy costs, the additional cost to consumers of 'future top-up payments under the proposed HPC CfD had increased from £6.1 billion in October 2013, when the strike price was agreed, to £29.7 billion'.[82][83] In July 2017, this estimate rose to £50 billion, or 'more than eight times the 2013 estimate'.[9]