4 ms·

As the article mentions towards the end, just because it's technically feasible does not mean a lot of coal plants will be converted to nuclear. Fundamentally t

As the article mentions towards the end, just because it's technically feasible does not mean a lot of coal plants will be converted to nuclear. Fundamentally this is about reducing cost. The question is if it is enough. Nuclear plants are expensive and making them slightly less expensive helps. But they might still be too expensive overall to make it worth the trouble, Also, time is a factor. A lot of coal plants have already been decommissioned and more are going offline in the next few years. Mothballed/decommissioned coal plants are plentiful but their infrastructure is not going to maintain itself and is probably actively being decommissioned instead. And even if it's not, it's probably not getting any better just sitting there unmaintained. And if anything, nuclear is a very slow moving industry. Ten years is nothing. Ten years from now, a lot of these coal plants will be history.

epistasis 4y ago
The expensive part of nuclear isn't the parts that coal plants provide, either. All those miles upon miles of precision welding made to withstand decades of harsh conditions with out a single leak? Coal doesn't need that. Large, high precision concrete pours? Coal doesn't need that. Containment structures? Coal doesn't need that. Nuclear is inherently an extremely complex technology that requires construction techniques that are advanced and expensive. The steam turbines and cooling towers are the only parts that are easy and standard construction with well-contained costs.
Schroedingersat 4y ago
If it's a functioning economically viable steam turbine, why not put a gas turbine in front of it and burn hydrogen (or no change if it is already a pulverised coal turbine)?
- Turing_Machine 4y ago
  Hydrogen is not an energy source as such. You have to use energy to make it, incurring losses, then incur still more losses when you burn it. It is, at best, an energy storage mechanism, not a source.
  - Schroedingersat 4y ago
    Yes. This is quite obviously the point. Electrolysers are dirt cheap (and prospects of technologies with battery-like efficiency commercialising are pretty good). You still only get back 50% of your energy by burning, but if the energy from the lower capacity times is enough to fully fund your panels, then electrolysing the energy from 10-3 during spring till autumn allows the use of a resource you'd otherwise dump as heat. Storage is hard, energy to power electrolysers is only cheap with solar or surplus wind, and the second stage of a ccgt is big and expensive. This would solve one of those problems. The other two are already cheaper than drilling a bunch of 20km deep holes or building a nuclear reactor.
    - Turing_Machine 4y ago
      Yes, nuclear is always going to be more expensive than systems that (as of now) exist only in the imagination of their proponents. Unless another has come online since the last time I looked, there is only one (1) grid-scale solar installation in the entire world, and it was built with massive government funding.
      - Schroedingersat 4y ago
        > Unless another has come online since the last time I looked, there is only one (1) grid-scale solar installation in the entire world, and it was built with massive government funding. What a bizarre line to draw. The technology that benefits from being distributed rather than needing massive centralisation to be viable isn't concentrated to the point it causes the grid to be over-centralised? Whatever metric you need to satisfy those mental gymnastics you can watch it come true realtime over the next year or two by adding 'unsubsidized solar park GW' to your news topic feed.
        Turing_Machine 4y ago
        There's such a thing as "economy of scale". You're not going to run an industrial civilization from a few hobby solar panels on someone's roof, any more than Mao was able to run an industrial civilization from crude cast iron smelted in people's back yards. > you can watch it come true realtime over the next year or two Yeah. Not happening. Not in a year or two, nor even in a decade or two. What powers heavy industry in your decentralized utopia? Some kind of reverse grid? Only about 34% of electricity is used for residential purposes in the United States, and I would be surprised if the proportion were much different in other countries.
- stjohnswarts 4y ago
  Where are you going to get that hydrogen?
  - Schroedingersat 4y ago
    One of many options and analyses that get you well below the cost of a still-mythical nuclear reactor from someone like nuscale: https://newatlas.com/energy/hysata-efficient-hydrogen-electrolysis/ https://newatlas.com/energy/hysata-efficient-hydrogen-electr... https://www.pv-magazine.com/2019/07/22/solar-at-17-30-mw-in-brazils-auction-is-it-a-world-record/ https://www.pv-magazine.com/2019/07/22/solar-at-17-30-mw-in-... https://www.science-climat-energie.be/2021/07/16/the-present-and-future-green-hydrogen-production-cost/ https://www.science-climat-energie.be/2021/07/16/the-present...
    - stjohnswarts 4y ago
      I'll believe those guys when I see a commercial scale version. Lots of things have been done in the lab with a large budge for small results. I would say less than 0.5% of them become a success. I hope they do, I would love for this to be true. However, the planet is dying right now and nuclear is one of those things that could help us save ourselves from ourselves so I will keep rooting for it.
      - Schroedingersat 4y ago
        As I said, it's only one of many technologies which are vying to lower costs. If 0.5% of the ones already undergoing commercialisation succeed we'll be seeing capital costs halve or better and efficiencies improve. The SMRs required to convert coal to nuclear or microwave drilling technologies are much much further from reality than "take a well understood design and put a capilliary in the middle". Here's a simple model using very pessimistic costs of last generation technology with 100% uptime for lower LCOE than new nuclear https://model.energy/?results=ee966ba4279eec9f32fcdadb8dd4dcf0#solve https://model.energy/?results=ee966ba4279eec9f32fcdadb8dd4dc... Much more importantly you're buying capacity that can come on as soon as each part is done with that money rather than capacity that comes online between 2032 and 2045 and won't see full return until 2100. As you said the planet is dying now and I'd much rather have 3W for the capital in 3 years (with a vast surplus for variable loads) than 1W in 10-20. All we have to do is stop helping the fossil fuel industry block new wind and solar projects, delay them and drive costs up. The variability will fund storage and drive fossil fuels out of business on its own. Pouring those billions of public funds into gigawatts of nationally owned electrolysers to kickstart the industry would be a vastly better use of money than spending it on nuclear projects that will never open or be a drain on public coffers for decades via tarriff guarantees 10x the projected cost of electricity.

epistasis 4y ago

The expensive part of nuclear isn't the parts that coal plants provide, either. All those miles upon miles of precision welding made to withstand decades of harsh conditions with out a single leak? Coal doesn't need that. Large, high precision concrete pours? Coal doesn't need that. Containment structures? Coal doesn't need that. Nuclear is inherently an extremely complex technology that requires construction techniques that are advanced and expensive. The steam turbines and cooling towers are the only parts that are easy and standard construction with well-contained costs.

Schroedingersat 4y ago

If it's a functioning economically viable steam turbine, why not put a gas turbine in front of it and burn hydrogen (or no change if it is already a pulverised coal turbine)?

Turing_Machine 4y ago
Hydrogen is not an energy source as such. You have to use energy to make it, incurring losses, then incur still more losses when you burn it. It is, at best, an energy storage mechanism, not a source.
- Schroedingersat 4y ago
  Yes. This is quite obviously the point. Electrolysers are dirt cheap (and prospects of technologies with battery-like efficiency commercialising are pretty good). You still only get back 50% of your energy by burning, but if the energy from the lower capacity times is enough to fully fund your panels, then electrolysing the energy from 10-3 during spring till autumn allows the use of a resource you'd otherwise dump as heat. Storage is hard, energy to power electrolysers is only cheap with solar or surplus wind, and the second stage of a ccgt is big and expensive. This would solve one of those problems. The other two are already cheaper than drilling a bunch of 20km deep holes or building a nuclear reactor.
  - Turing_Machine 4y ago
    Yes, nuclear is always going to be more expensive than systems that (as of now) exist only in the imagination of their proponents. Unless another has come online since the last time I looked, there is only one (1) grid-scale solar installation in the entire world, and it was built with massive government funding.
    - Schroedingersat 4y ago
      > Unless another has come online since the last time I looked, there is only one (1) grid-scale solar installation in the entire world, and it was built with massive government funding. What a bizarre line to draw. The technology that benefits from being distributed rather than needing massive centralisation to be viable isn't concentrated to the point it causes the grid to be over-centralised? Whatever metric you need to satisfy those mental gymnastics you can watch it come true realtime over the next year or two by adding 'unsubsidized solar park GW' to your news topic feed.
      - Turing_Machine 4y ago
        There's such a thing as "economy of scale". You're not going to run an industrial civilization from a few hobby solar panels on someone's roof, any more than Mao was able to run an industrial civilization from crude cast iron smelted in people's back yards. > you can watch it come true realtime over the next year or two Yeah. Not happening. Not in a year or two, nor even in a decade or two. What powers heavy industry in your decentralized utopia? Some kind of reverse grid? Only about 34% of electricity is used for residential purposes in the United States, and I would be surprised if the proportion were much different in other countries.
stjohnswarts 4y ago
Where are you going to get that hydrogen?
- Schroedingersat 4y ago
  One of many options and analyses that get you well below the cost of a still-mythical nuclear reactor from someone like nuscale: https://newatlas.com/energy/hysata-efficient-hydrogen-electrolysis/ https://newatlas.com/energy/hysata-efficient-hydrogen-electr... https://www.pv-magazine.com/2019/07/22/solar-at-17-30-mw-in-brazils-auction-is-it-a-world-record/ https://www.pv-magazine.com/2019/07/22/solar-at-17-30-mw-in-... https://www.science-climat-energie.be/2021/07/16/the-present-and-future-green-hydrogen-production-cost/ https://www.science-climat-energie.be/2021/07/16/the-present...
  - stjohnswarts 4y ago
    I'll believe those guys when I see a commercial scale version. Lots of things have been done in the lab with a large budge for small results. I would say less than 0.5% of them become a success. I hope they do, I would love for this to be true. However, the planet is dying right now and nuclear is one of those things that could help us save ourselves from ourselves so I will keep rooting for it.
    - Schroedingersat 4y ago
      As I said, it's only one of many technologies which are vying to lower costs. If 0.5% of the ones already undergoing commercialisation succeed we'll be seeing capital costs halve or better and efficiencies improve. The SMRs required to convert coal to nuclear or microwave drilling technologies are much much further from reality than "take a well understood design and put a capilliary in the middle". Here's a simple model using very pessimistic costs of last generation technology with 100% uptime for lower LCOE than new nuclear https://model.energy/?results=ee966ba4279eec9f32fcdadb8dd4dcf0#solve https://model.energy/?results=ee966ba4279eec9f32fcdadb8dd4dc... Much more importantly you're buying capacity that can come on as soon as each part is done with that money rather than capacity that comes online between 2032 and 2045 and won't see full return until 2100. As you said the planet is dying now and I'd much rather have 3W for the capital in 3 years (with a vast surplus for variable loads) than 1W in 10-20. All we have to do is stop helping the fossil fuel industry block new wind and solar projects, delay them and drive costs up. The variability will fund storage and drive fossil fuels out of business on its own. Pouring those billions of public funds into gigawatts of nationally owned electrolysers to kickstart the industry would be a vastly better use of money than spending it on nuclear projects that will never open or be a drain on public coffers for decades via tarriff guarantees 10x the projected cost of electricity.