6 ms·

> this must be something with fast cold start. So black/brown coal power plan will not help you, similarly nuclear. Nuclear plants provide base load and they a

by troupo 8mo ago

> this must be something with fast cold start. So black/brown coal power plan will not help you, similarly nuclear. Nuclear plants provide base load and they are extremely fast at ramping up/lowering production. All modern nuclear plants are capable of changing power output at 3-5% of nameplate capacity per minute: https://www.oecd-nea.org/upload/docs/application/pdf/2021-12/technical_and_economic_aspects_of_load_following_with_nuclear_power_plants.pdf https://www.oecd-nea.org/upload/docs/application/pdf/2021-12... You don't shut down power plants. None of the power plants ever do a "fast cold start" > The end result now is that electricity in Europe is the most expensive on the World, so all manufacturing is moved to Asia The production moved to Asia due to extremely cheap labor, not due to electricity costs.

Scoundreller 8mo ago
For the foreseeable future, building enough nuclear for peak capacity is exceedingly expensive. > None of the power plants ever do a "fast cold start" Somewhere in each grid you will have “black start” capacity contracts, dunno if nuclear can fills this role (or if grids exclude nukes for one reason or another). Plenty of peaker plants built with the intention of running double digit hours per year and therefore the tradeoff supports being largely “off” in between those calls. Batteries might fill that gap.
crote 8mo ago
> Nuclear plants provide base load and they are extremely fast at ramping up/lowering production The obvious counterexample is Chernobyl, where a big contributor was the fact that they were unable to scale it down & back up as desired. Yes, nuclear reactors can scale down rapidly - but you have to wait several hours until it can scale back up! Besides, the linked paper only covers load-following in a traditional grid (swinging between 60% and 100% once a day) and barely touches on the economic effects. The situation is going to look drastically different for a renewables-first grid, where additional sources are needed for at most a few hours a day, for a few months per year. > You don't shut down power plants. None of the power plants ever do a "fast cold start" Gas turbines can. Hydro can. Battery storage can.
- phil21 8mo ago
  The answer is you don't scale nuclear up or down, it's a silly waste of time and effort to even think about it. The fuel costs are effectively a rounding error, so running at 100% 24x7 is the only way to ever think about how nuclear should operate. If you are going to curtail, you curtail other sources including solar and wind. Nuclear fits quite well for the baseload you need. It's more expensive, but if you are going to need X capacity 24x7 and build nuclear, you simply build enough to provide just that plus perhaps a few extra for redundancy when another one goes offline. Then use gas peakers for the "oh shit" days difference between what nuclear is providing and solar was expected to but could not. I don't understand the fascination folks have about nuclear not being able to following the grid. They don't need to, since they only ever remotely make sense when operated 24x7 at 100%. If you always have 1TW of grid usage every night during your lowest usage period - build that much nuclear as your starting point and figure out the rest from there. Nuclear's share of the total mix should be a straight line on a graph outside of plant shutdowns for maintenance.
  - dalyons 8mo ago
    That’s not the way the energy market works though. The cheapest sources (like daytime solar) will knock your expensive nuclear off the grid. Or force it to sell at significantly below operating cost, which is suicidal in the long term, since nukes need a guaranteed high price nearly 100% of the time to pencil out (pay back the capex). Your argument only works in entirely state controlled systems, not in free energy markets of independent suppliers. Which is why nukes don’t get built.
    - troupo 8mo ago
      > Which is why nukes don’t get built. Nukes don't get built because: - billions (if not trillions) of subsidies were poured into wind and solar over decades to make them viable while nuclear energy was addled with additional taxes, reactor closures, and very few new reactor licenses - decades of fear-mongering led to loss of expertise in building new nuclear power plants (and instead South-East Asia has been picking up speed in building new reactors) [1] In 2015 nuclear was significantly cheaper than most other types of energy across most markets: https://world-nuclear.org/images/articles/REPORT_Economics_Report_2017.pdf https://world-nuclear.org/images/articles/REPORT_Economics_R... (Figure 12, in some markets including the then-emerging renewables). And yet renewables were enjoying unprecedented amounts of subsidies and money poured into them while nuclear... Oh we know what was happening to nuclear, just look at Germany. [1] Here's EU's own report: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM:2025:17:FIN&pk_campaign=todays_OJ&pk_source=EUR-Lex&pk_medium=X&pk_content=Report&pk_keyword=Energy&pk_cid=EURLEX_todaysOJ https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM:2025... Renewables: 80-80 billion euro in subsidies a year. Fossil fuels: 60-140 billion euro in subsidies a year. Nuclear: good luck finding the thin orange line in the graph. (1% of subsidies) --- start quote --- As shown on Figure 4 , solar energy received by far the largest share of subsidies, both historically and in 2023 (EUR 21 bn), followed by biomass (EUR 9 bn) and wind power (EUR 7 bn). Hydropower received marginal financial support (~EUR 1 bn), while subsidies targeting multiple renewable technologies (such as tax reductions on green technology or public aid for investment projects) jumped to EUR 23 bn by 2023. Subsidies for nuclear energy dropped from EUR 7.9 bn in 2021 to 3.7 bn in 2022 and 4.1 bn in 2023. Of the 14 MS providing nuclear subsidies, France (EUR 2.9 bn) accounted for the biggest share, followed by Germany (EUR 0.8 bn) , Spain and Belgium (EUR 0.1 bn each). --- end quote ---
sehansen 8mo ago
5% per minute is not extremely fast. Simple cycle gas turbine (peaker) plants routinely go 0 to 100% in less than 10 minutes. Nuclear plants can only hit 5% per minute in the 50 to 100% interval (per your own source). And all of this is confused by the way the nuclear industry uses the term "load following". You'd think it means "changing the power output from moment to moment to match electricity demand" but for nuclear plants it means "changing from one pre-planned constant level to another pre-planned constant level, up to four times per day".[0] There are only three[1] sources of electricity that can be ramped freely enough to exactly match demand: hydro, simple-cycle gas turbines and batteries. All electrical supplies will need some of those three mixed in. Which is why France is still 10% hydro and 10% natural gas in their electricity supply. 0: Some of the most modern Russian plants can move to +-20% of their current target at 10% per minute, but "the number of such very fast power variations is limited, and they are mainly reserved for emergency situations." per your source. 1: OK, there are some obsolete ways too, like diesel generators. At least obsolete at the scale of the electricity grid.
- troupo 8mo ago
  > 5% per minute is not extremely fast. 5% of nameplate capacity. > You'd think it means "changing the power output from moment to moment to match electricity demand" but for nuclear plants it means "changing from one pre-planned constant level to another pre-planned constant level, up to four times per day" Which is clearly invalidated by the very source I provided, and which you then somehow quote back at me. > "the number of such very fast power variations is limited, and they are mainly reserved for emergency situations." per your source. Imagine if you didn't omit the full quote/context: --- start quote --- Also, AES-2006 is capable of fast power modulations with ramps of up to 5% Pr per second (in the interval of ±10% Pr), or power drops of 20% Pr per minute in the interval of 50-100% of the rated power. However, the number of such very fast power variations is limited, and they are mainly reserved for emergency situations. --- end quote --- Oh look. What's limited is an actual emergency ramp up of 5% per second or power drops of 20% per minute. Which is literally an emergency that is not needed in a power grid.
  - sehansen 8mo ago

by troupo 8mo ago

Scoundreller 8mo ago

For the foreseeable future, building enough nuclear for peak capacity is exceedingly expensive. > None of the power plants ever do a "fast cold start" Somewhere in each grid you will have “black start” capacity contracts, dunno if nuclear can fills this role (or if grids exclude nukes for one reason or another). Plenty of peaker plants built with the intention of running double digit hours per year and therefore the tradeoff supports being largely “off” in between those calls. Batteries might fill that gap.

crote 8mo ago

> Nuclear plants provide base load and they are extremely fast at ramping up/lowering production The obvious counterexample is Chernobyl, where a big contributor was the fact that they were unable to scale it down & back up as desired. Yes, nuclear reactors can scale down rapidly - but you have to wait several hours until it can scale back up! Besides, the linked paper only covers load-following in a traditional grid (swinging between 60% and 100% once a day) and barely touches on the economic effects. The situation is going to look drastically different for a renewables-first grid, where additional sources are needed for at most a few hours a day, for a few months per year. > You don't shut down power plants. None of the power plants ever do a "fast cold start" Gas turbines can. Hydro can. Battery storage can.

phil21 8mo ago
The answer is you don't scale nuclear up or down, it's a silly waste of time and effort to even think about it. The fuel costs are effectively a rounding error, so running at 100% 24x7 is the only way to ever think about how nuclear should operate. If you are going to curtail, you curtail other sources including solar and wind. Nuclear fits quite well for the baseload you need. It's more expensive, but if you are going to need X capacity 24x7 and build nuclear, you simply build enough to provide just that plus perhaps a few extra for redundancy when another one goes offline. Then use gas peakers for the "oh shit" days difference between what nuclear is providing and solar was expected to but could not. I don't understand the fascination folks have about nuclear not being able to following the grid. They don't need to, since they only ever remotely make sense when operated 24x7 at 100%. If you always have 1TW of grid usage every night during your lowest usage period - build that much nuclear as your starting point and figure out the rest from there. Nuclear's share of the total mix should be a straight line on a graph outside of plant shutdowns for maintenance.
- dalyons 8mo ago
  That’s not the way the energy market works though. The cheapest sources (like daytime solar) will knock your expensive nuclear off the grid. Or force it to sell at significantly below operating cost, which is suicidal in the long term, since nukes need a guaranteed high price nearly 100% of the time to pencil out (pay back the capex). Your argument only works in entirely state controlled systems, not in free energy markets of independent suppliers. Which is why nukes don’t get built.
  - troupo 8mo ago
    > Which is why nukes don’t get built. Nukes don't get built because: - billions (if not trillions) of subsidies were poured into wind and solar over decades to make them viable while nuclear energy was addled with additional taxes, reactor closures, and very few new reactor licenses - decades of fear-mongering led to loss of expertise in building new nuclear power plants (and instead South-East Asia has been picking up speed in building new reactors) [1] In 2015 nuclear was significantly cheaper than most other types of energy across most markets: https://world-nuclear.org/images/articles/REPORT_Economics_Report_2017.pdf https://world-nuclear.org/images/articles/REPORT_Economics_R... (Figure 12, in some markets including the then-emerging renewables). And yet renewables were enjoying unprecedented amounts of subsidies and money poured into them while nuclear... Oh we know what was happening to nuclear, just look at Germany. [1] Here's EU's own report: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM:2025:17:FIN&pk_campaign=todays_OJ&pk_source=EUR-Lex&pk_medium=X&pk_content=Report&pk_keyword=Energy&pk_cid=EURLEX_todaysOJ https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=COM:2025... Renewables: 80-80 billion euro in subsidies a year. Fossil fuels: 60-140 billion euro in subsidies a year. Nuclear: good luck finding the thin orange line in the graph. (1% of subsidies) --- start quote --- As shown on Figure 4 , solar energy received by far the largest share of subsidies, both historically and in 2023 (EUR 21 bn), followed by biomass (EUR 9 bn) and wind power (EUR 7 bn). Hydropower received marginal financial support (~EUR 1 bn), while subsidies targeting multiple renewable technologies (such as tax reductions on green technology or public aid for investment projects) jumped to EUR 23 bn by 2023. Subsidies for nuclear energy dropped from EUR 7.9 bn in 2021 to 3.7 bn in 2022 and 4.1 bn in 2023. Of the 14 MS providing nuclear subsidies, France (EUR 2.9 bn) accounted for the biggest share, followed by Germany (EUR 0.8 bn) , Spain and Belgium (EUR 0.1 bn each). --- end quote ---

sehansen 8mo ago

5% per minute is not extremely fast. Simple cycle gas turbine (peaker) plants routinely go 0 to 100% in less than 10 minutes. Nuclear plants can only hit 5% per minute in the 50 to 100% interval (per your own source). And all of this is confused by the way the nuclear industry uses the term "load following". You'd think it means "changing the power output from moment to moment to match electricity demand" but for nuclear plants it means "changing from one pre-planned constant level to another pre-planned constant level, up to four times per day".[0] There are only three[1] sources of electricity that can be ramped freely enough to exactly match demand: hydro, simple-cycle gas turbines and batteries. All electrical supplies will need some of those three mixed in. Which is why France is still 10% hydro and 10% natural gas in their electricity supply. 0: Some of the most modern Russian plants can move to +-20% of their current target at 10% per minute, but "the number of such very fast power variations is limited, and they are mainly reserved for emergency situations." per your source. 1: OK, there are some obsolete ways too, like diesel generators. At least obsolete at the scale of the electricity grid.

troupo 8mo ago
> 5% per minute is not extremely fast. 5% of nameplate capacity. > You'd think it means "changing the power output from moment to moment to match electricity demand" but for nuclear plants it means "changing from one pre-planned constant level to another pre-planned constant level, up to four times per day" Which is clearly invalidated by the very source I provided, and which you then somehow quote back at me. > "the number of such very fast power variations is limited, and they are mainly reserved for emergency situations." per your source. Imagine if you didn't omit the full quote/context: --- start quote --- Also, AES-2006 is capable of fast power modulations with ramps of up to 5% Pr per second (in the interval of ±10% Pr), or power drops of 20% Pr per minute in the interval of 50-100% of the rated power. However, the number of such very fast power variations is limited, and they are mainly reserved for emergency situations. --- end quote --- Oh look. What's limited is an actual emergency ramp up of 5% per second or power drops of 20% per minute. Which is literally an emergency that is not needed in a power grid.
- sehansen 8mo ago