Eh. I covered the issue for ten years and went from pro-nuclear to nuclear skeptic. You hit a big part of it on the head: economically, nuclear can't compete with renewables except in powerpoint presentations. But there is also the long history of design mishaps. Even what should be no brainer things like water containment pools got messed up at Vogtle. (Ask NuScale about their passive cooling system.) 60%+ pct of the costs are steel, cement and engineering: good luck getting a Moore's Law for that! Nuclear plays on people's nostalgia for Pax Americana. Solar and storage will be cheaper and you won't have to wait 15 years to use it. Put it another way: every AI technology we know today will be gone and obsolete before an AI nuke comes up.
Look - i wrote about the reason "why now" and its pretty much that for this brief moment of $/mwh makes the most sense and is much better than solar & wind. IRA helps finance the roughest part of it too, the cost!
I understand that 60% of the costs are cement and engineering, but still think that this hits the sweet spot and the time is now.
disagree, think we will be starting brownfield before 2030 lol. I just think it's a match made in heaven in terms of the footprint, and think geothermal is brutal because it's so sight specific.
and I think the capital cycle of nuclear is not going to overlap *well* with this one, most of the dollars come down and then we will see more capacity added overtime.
And that is what makes a great debate. Who knows. It might be the time. To be honest, I'd like to be wrong. But when you look at the cost curves, the modularity, and the ability to put GWs in the ground, the renewable community has a great track record. Nuclear? Even in France and Finland, two pro nuke jurisdictions, delays and cost overruns are the order of the day.
I mean here's the thing - Us energy consumption has been pretty much 1% CAGR for the last 5 years.
Now we need it to be 5%+ a year. The alternative is legit coal, I'm hearing that coal might get restarted that the demand is so real. Nuclear will the replacement. Thinking of it as a swap is easy when things were flattish and we retired coal for a decade. Right now we need everything! Also the financing + IRA helps
But here is what you're missing. Solar, storage and wind are modular. Small to medium sized groups working independently are installing it. 32GW of solar in the US alone went up last year to raise the total to 177GW. Wind did 6GW last year and has doubled to 148GW since 2010. Batteries grew by 7.9GW last year. Nuclear? 4GW new capacity in 20 years and often it's out for unplanned outages. Why? Plants are difficult to build and harder to maintain. It's the same reason geothernal, tidal, wave and solar thermal aren't taking the world by storm either. Ask your Wall Street pals which source of energy they're more comfortable funding. And recommissioning old nuke plants will be costly too. Three Mile Island was built in 1968: three years before the microprocessor and 8 before Ethernet. And in the end it's only 835 MW! If you broke ground on a nuke today, it wouldn't be operational until 2035 at a minimum and more like 2042 (judging by completion times in the US, UK, France, Finland and the UAE.) .
Believe me. Demand is real. We need more power. But electrons are a commodity: you pick the technology that can produce them quickly, cheaply, consistently, safely etc. Whatever method clicks the boxes. In practice, nuclear, despite the best efforts and heavy doses of subsidies, typically misses the mark.
Energy density matters man - also solar great cost curves but you need FIRM power. Like we literally might not have the supply chain for lithium for the next 5 years plus that alone is very dirty.
FWIW wall street pals are looking to fund, its being taken very seriously!
I'm of the belief that it wont be the majority of power added, but think that it will move the needle all-in. Just my 2 cents, and theres a reason why its called "nuclear zeitgeist", i think the zeitgeist is back
I think that looking at LCOE can be a skewed way to look at energy deliver by wind and solar since it understated the hidden cost (extra transmission capacity, extra back-up capacity (gas turbine, energy storage) compare to nuclear in terms of total energy cost adjusting for stability factor. I think the wind and solar to be competitive the main things now is energy storage price like you said. What do you think will be the point at which energy storage is cheap enough for wind/solar to be a viable competitive both in terms of cost and reliability ?
I also have a thought that the slow progress of US nuclear powerplant could also due to lack of public support for it (resulting in lack of supporting industry, craftperson etc) compare to say China, which build nuclear power plant at fast pace and competitive cost. May be with public support this problem will be alleviate?
wind/solar and storage are already cheaper than SMRs or new nuclear by virtue of the fact that SMRs don't exist and the recent plants went way overbudget (see Finland, France, UK and US). But even if you look at historic cost curves, the argument is over. Renewables and storage go down. Nuclear rarely if ever does. And the public gets left holding the bag. Ask people in Florida. They've got a long history of having to sue utilities for nuke promises gone awry. https://www.tampabay.com/news/business/energy/duke-energy-customers-will-pay-108-million-a-year-for-cancelled-nuclear/2134867/
The regulator issue comes up a lot, but if you look, the regulators are often pro nuclear. Look at Georgia. They approved everything at Vogtle. But did regulators ask them to insert leaks in the contaimnent pool? Or was that just substandard construction work. Even MIT, probably the most pro nuke major university, says regulation is only a factor and not the primary factor. https://news.mit.edu/2020/reasons-nuclear-overruns-1118.
When it comes to China, their two advantages seem to be really low cost loans subsidized by the government and a more devil may care attitude toward public opinion. You could argue taxpayers should fund plants for AI data centers, but that could be a tough sell. Imagine the public hearing: a utility wants a $10 per month surcharge to build a facility for a single customer, who also happens to be super wealthy and wants the power to sell for-profit services back to hapless ratepayers.
Do you have any sources or reports that you are often reading that have comparison between wind/solar + storage compare to nuclear (either SMR or normal size) that accounts for both costs and other factors like reliability?
I am reading this one https://www.lazard.com/media/xemfey0k/lazards-lcoeplus-june-2024-_vf.pdf and it seems to me LCOE calculation of nuclear powerplant in the US is using mostly Vogtle data (7-7.5 bio per reactor) compare to Chinese or Russian counter-part (VVER-1200) of 3-5$ bio per reactor investment cost with good safety record (I doubt that you can hide any major nuclear accident even though I believe that Chinese government can do a terrific job of filter out what they want everyone to know).
That would reduce the capital cost in LCOE of nuclear by half (currently 7-80% of LCOE calculated by Lazard) in competitive range with solar and wind. LCOE calculation by Lazard for nuclear plant is 40 years compare to the typical 60 or even 80 years now. The assumption for storage for Wind and Solar currently is 4 hrs which I think will be increasing to 6-8-12 storage hours since the more wind/solar you have in the grid the issues of time-shifting the power become more pronounced and wind/solar are naturally natural hence not as stable as other base load sources.
You can't hide a major nuclear accident but can you hide a cost overrun? That's a little easier. Or when was the last time you saw an audited financial report from Rosatom? China was part of Hinckley Point in the UK but backed out after the problems. And look at Rosatom in Egypt. $30B for 4.8 GW. (ugh! 6B per GW before overruns. ) Deal signed in 2015. Russia providing funding and construction. Construction began this year. 9 years of planning.
Lazard is pretty good with LCOE, but you have to look at other factors too. Can you get a GW of solar in a year with storage? No problem. Nuclear? Not possible. When I started as an energy reporter, I was gung ho nuke. The endless delays made me a skeptic.
I guess, I still will bet there will be like meaningful additions (+5 GW) in the next few years cumulatively. You can say all this but the Zeitgeist is against it.
How does this sound - let's bet? 100 bucks or a nice dinner and we pay out in 2030 on new builds in the US. I don't expect power to be online, but I would do something along the lines of backlog / in process.
$100 on functionally complete. Everything done but grid connection. "in process" is where they stumble (see NuScale.) But even if reasonably close and not way past the original cost estimates I'll give it to you.
Fair point about data. I guess we will need to wait and see then. I don't think 1GW of solar can be equivalent in output (kWh) or stability compare to base load plant though.
Given that the companies looking to develop nuclear-supported AI data centers have a vested interest + war chests bigger than the market caps of most utilities (therefore economic calculations are dramatically altered), it would suggest "too expensive" isn't the hang-up. The fact that this generation will also be captive alters the regulatory framework, too (no concerns over customers' electric bills).
Additionally, if left to their own devices, don't you think companies like Google/Meta/Amazon/xAI - who specially designed their entire data infrastructure/rocket ships to optimize for their business models - will be able to experiment/iterate and come up with designs that work better than what has been done previously (or even team up to crack the problem). Obviously regulations will still have to be cleared, but just because the atrophied and fossilized nuclear construction industry can't deliver a modularized reactor on time or budget doesn't mean it can't be done.
And thanks to Wright's Law, once it is done, it will be done again, but better and faster.
I'm mixed on nuclear. It is clean (until it is not) so good for climate change. But as a human living things don't do well w nuclear. If Iran builds nuclear and Israel attacks them that is pretty scary as it could release radiation. This would make nuclear an 'own goal" in times of war, especially between far away rivals (not Ukraine vs Russia).
On a positive note, the mega cap tech companies are legit innovators and they can use AI to innovate. So the industry is moving from 65 year old nuclear engineers ready to retire to young tech nerd engineers innovating a brand new tech that is hopefully robust and safe firstly. Next gen reactors might be easy, fast and safe to build.
in general the more nuclear we get the more we will need surveillance / transparency and less war to keep it safe.
I personally believe geothermal will have bigger adoption than nuclear. I think the fear of nuclear will never go away. Meta and Google are already experimenting with it
Eh. I covered the issue for ten years and went from pro-nuclear to nuclear skeptic. You hit a big part of it on the head: economically, nuclear can't compete with renewables except in powerpoint presentations. But there is also the long history of design mishaps. Even what should be no brainer things like water containment pools got messed up at Vogtle. (Ask NuScale about their passive cooling system.) 60%+ pct of the costs are steel, cement and engineering: good luck getting a Moore's Law for that! Nuclear plays on people's nostalgia for Pax Americana. Solar and storage will be cheaper and you won't have to wait 15 years to use it. Put it another way: every AI technology we know today will be gone and obsolete before an AI nuke comes up.
Look - i wrote about the reason "why now" and its pretty much that for this brief moment of $/mwh makes the most sense and is much better than solar & wind. IRA helps finance the roughest part of it too, the cost!
I understand that 60% of the costs are cement and engineering, but still think that this hits the sweet spot and the time is now.
disagree, think we will be starting brownfield before 2030 lol. I just think it's a match made in heaven in terms of the footprint, and think geothermal is brutal because it's so sight specific.
and I think the capital cycle of nuclear is not going to overlap *well* with this one, most of the dollars come down and then we will see more capacity added overtime.
And that is what makes a great debate. Who knows. It might be the time. To be honest, I'd like to be wrong. But when you look at the cost curves, the modularity, and the ability to put GWs in the ground, the renewable community has a great track record. Nuclear? Even in France and Finland, two pro nuke jurisdictions, delays and cost overruns are the order of the day.
I mean here's the thing - Us energy consumption has been pretty much 1% CAGR for the last 5 years.
Now we need it to be 5%+ a year. The alternative is legit coal, I'm hearing that coal might get restarted that the demand is so real. Nuclear will the replacement. Thinking of it as a swap is easy when things were flattish and we retired coal for a decade. Right now we need everything! Also the financing + IRA helps
But here is what you're missing. Solar, storage and wind are modular. Small to medium sized groups working independently are installing it. 32GW of solar in the US alone went up last year to raise the total to 177GW. Wind did 6GW last year and has doubled to 148GW since 2010. Batteries grew by 7.9GW last year. Nuclear? 4GW new capacity in 20 years and often it's out for unplanned outages. Why? Plants are difficult to build and harder to maintain. It's the same reason geothernal, tidal, wave and solar thermal aren't taking the world by storm either. Ask your Wall Street pals which source of energy they're more comfortable funding. And recommissioning old nuke plants will be costly too. Three Mile Island was built in 1968: three years before the microprocessor and 8 before Ethernet. And in the end it's only 835 MW! If you broke ground on a nuke today, it wouldn't be operational until 2035 at a minimum and more like 2042 (judging by completion times in the US, UK, France, Finland and the UAE.) .
Believe me. Demand is real. We need more power. But electrons are a commodity: you pick the technology that can produce them quickly, cheaply, consistently, safely etc. Whatever method clicks the boxes. In practice, nuclear, despite the best efforts and heavy doses of subsidies, typically misses the mark.
Energy density matters man - also solar great cost curves but you need FIRM power. Like we literally might not have the supply chain for lithium for the next 5 years plus that alone is very dirty.
FWIW wall street pals are looking to fund, its being taken very seriously!
I'm of the belief that it wont be the majority of power added, but think that it will move the needle all-in. Just my 2 cents, and theres a reason why its called "nuclear zeitgeist", i think the zeitgeist is back
I think that looking at LCOE can be a skewed way to look at energy deliver by wind and solar since it understated the hidden cost (extra transmission capacity, extra back-up capacity (gas turbine, energy storage) compare to nuclear in terms of total energy cost adjusting for stability factor. I think the wind and solar to be competitive the main things now is energy storage price like you said. What do you think will be the point at which energy storage is cheap enough for wind/solar to be a viable competitive both in terms of cost and reliability ?
I also have a thought that the slow progress of US nuclear powerplant could also due to lack of public support for it (resulting in lack of supporting industry, craftperson etc) compare to say China, which build nuclear power plant at fast pace and competitive cost. May be with public support this problem will be alleviate?
wind/solar and storage are already cheaper than SMRs or new nuclear by virtue of the fact that SMRs don't exist and the recent plants went way overbudget (see Finland, France, UK and US). But even if you look at historic cost curves, the argument is over. Renewables and storage go down. Nuclear rarely if ever does. And the public gets left holding the bag. Ask people in Florida. They've got a long history of having to sue utilities for nuke promises gone awry. https://www.tampabay.com/news/business/energy/duke-energy-customers-will-pay-108-million-a-year-for-cancelled-nuclear/2134867/
The regulator issue comes up a lot, but if you look, the regulators are often pro nuclear. Look at Georgia. They approved everything at Vogtle. But did regulators ask them to insert leaks in the contaimnent pool? Or was that just substandard construction work. Even MIT, probably the most pro nuke major university, says regulation is only a factor and not the primary factor. https://news.mit.edu/2020/reasons-nuclear-overruns-1118.
When it comes to China, their two advantages seem to be really low cost loans subsidized by the government and a more devil may care attitude toward public opinion. You could argue taxpayers should fund plants for AI data centers, but that could be a tough sell. Imagine the public hearing: a utility wants a $10 per month surcharge to build a facility for a single customer, who also happens to be super wealthy and wants the power to sell for-profit services back to hapless ratepayers.
Do you have any sources or reports that you are often reading that have comparison between wind/solar + storage compare to nuclear (either SMR or normal size) that accounts for both costs and other factors like reliability?
I am reading this one https://www.lazard.com/media/xemfey0k/lazards-lcoeplus-june-2024-_vf.pdf and it seems to me LCOE calculation of nuclear powerplant in the US is using mostly Vogtle data (7-7.5 bio per reactor) compare to Chinese or Russian counter-part (VVER-1200) of 3-5$ bio per reactor investment cost with good safety record (I doubt that you can hide any major nuclear accident even though I believe that Chinese government can do a terrific job of filter out what they want everyone to know).
That would reduce the capital cost in LCOE of nuclear by half (currently 7-80% of LCOE calculated by Lazard) in competitive range with solar and wind. LCOE calculation by Lazard for nuclear plant is 40 years compare to the typical 60 or even 80 years now. The assumption for storage for Wind and Solar currently is 4 hrs which I think will be increasing to 6-8-12 storage hours since the more wind/solar you have in the grid the issues of time-shifting the power become more pronounced and wind/solar are naturally natural hence not as stable as other base load sources.
You can't hide a major nuclear accident but can you hide a cost overrun? That's a little easier. Or when was the last time you saw an audited financial report from Rosatom? China was part of Hinckley Point in the UK but backed out after the problems. And look at Rosatom in Egypt. $30B for 4.8 GW. (ugh! 6B per GW before overruns. ) Deal signed in 2015. Russia providing funding and construction. Construction began this year. 9 years of planning.
Lazard is pretty good with LCOE, but you have to look at other factors too. Can you get a GW of solar in a year with storage? No problem. Nuclear? Not possible. When I started as an energy reporter, I was gung ho nuke. The endless delays made me a skeptic.
I guess, I still will bet there will be like meaningful additions (+5 GW) in the next few years cumulatively. You can say all this but the Zeitgeist is against it.
How does this sound - let's bet? 100 bucks or a nice dinner and we pay out in 2030 on new builds in the US. I don't expect power to be online, but I would do something along the lines of backlog / in process.
$100 on functionally complete. Everything done but grid connection. "in process" is where they stumble (see NuScale.) But even if reasonably close and not way past the original cost estimates I'll give it to you.
Fair point about data. I guess we will need to wait and see then. I don't think 1GW of solar can be equivalent in output (kWh) or stability compare to base load plant though.
I am wondering when to updated
Given that the companies looking to develop nuclear-supported AI data centers have a vested interest + war chests bigger than the market caps of most utilities (therefore economic calculations are dramatically altered), it would suggest "too expensive" isn't the hang-up. The fact that this generation will also be captive alters the regulatory framework, too (no concerns over customers' electric bills).
Additionally, if left to their own devices, don't you think companies like Google/Meta/Amazon/xAI - who specially designed their entire data infrastructure/rocket ships to optimize for their business models - will be able to experiment/iterate and come up with designs that work better than what has been done previously (or even team up to crack the problem). Obviously regulations will still have to be cleared, but just because the atrophied and fossilized nuclear construction industry can't deliver a modularized reactor on time or budget doesn't mean it can't be done.
And thanks to Wright's Law, once it is done, it will be done again, but better and faster.
I'm mixed on nuclear. It is clean (until it is not) so good for climate change. But as a human living things don't do well w nuclear. If Iran builds nuclear and Israel attacks them that is pretty scary as it could release radiation. This would make nuclear an 'own goal" in times of war, especially between far away rivals (not Ukraine vs Russia).
On a positive note, the mega cap tech companies are legit innovators and they can use AI to innovate. So the industry is moving from 65 year old nuclear engineers ready to retire to young tech nerd engineers innovating a brand new tech that is hopefully robust and safe firstly. Next gen reactors might be easy, fast and safe to build.
in general the more nuclear we get the more we will need surveillance / transparency and less war to keep it safe.
I personally believe geothermal will have bigger adoption than nuclear. I think the fear of nuclear will never go away. Meta and Google are already experimenting with it
https://about.fb.com/news/2024/08/new-geothermal-energy-project-to-support-our-data-centers/
https://blog.google/outreach-initiatives/sustainability/google-fervo-geothermal-energy-partnership/