5 ms·

This report provides costs for nuclear power down to about $30/MWH (LCOE) at 3% discount rate (interest rate - inflation rate): https://www.oecd-nea.org/upload

This report provides costs for nuclear power down to about $30/MWH (LCOE) at 3% discount rate (interest rate - inflation rate): https://www.oecd-nea.org/upload/docs/application/pdf/2020-12/egc-2020_2020-12-09_18-26-46_781.pdf https://www.oecd-nea.org/upload/docs/application/pdf/2020-12... Most countries should be able to reach such levels if providing guarantees for the credit, removing overly excessive regulation and allowing the most capable and experienced construction companies to lead construction and initial operations, especially if large enough numbers are built that economies of scale kick in. Renewables-only systems (apart from hydro, where available), are not able to provide constant power output at anything near such levels yet, without a significant percentage of the power being generated by natural gas. Massive capital investments in the grid, hydrogen generation, etc, MAY be able to actually provide stable power, but would be very unlikely to match nuclear if nuclear is optimized for price. This could change in a few years, obviously, especially if storage becomes super cheap. But there is no guarantee of that. As for the astronomical costs of some recent nuclear plants in western countries, those are a bit like evaluating the price of electric cars based on the Tesla Model S Plaid. We can get perfectly serviceable plants for a fraction of the price, by removing most of the bells and whistles, while still making sure that they are safer than most of the plants in operation today.

Schroedingersat 4y ago
There are a couple of problems with that analysis. Firstly I am very sceptical of their 60 year claim with an additional 20 years for LTO. Every time I look a supposed 60 year reactor in the west up it is having major problems after less than 40 -- I'd be interested in seeing a survey of reactors broken down by location (china is provably much more cost efficient at doing large projects than anyone else albeit at a high human cost) and whether they're funded by geopolitical incentives (russia subsidizing reactors to induce energy dependence). Additionally half of the reactor lifetime is irrelevant to meeting 2050 targets even if we take design lifetime as true. Even if we take the $30/MWh as an assumption. There are solar projects completing right now on the order of 80c/fixed Watt or $1.2 tilting. These produce 5-6 and 6-8 kWh/day/kW respectively. Ending the day with half of that production in a battery nets you an LCOE around $50 and that is falling by 15%/year with little indication of slowing down. Wind is a similar story (although not dropping as precipitously) and moderately anti-correlated. When 'we promise we can reduce the price by 3x and we'll follow through on that promise for 70 years even though costs have been going up and we've never followed through before' is a pretty hard sell against 'we've been consistently dropping costs for 10 years at a rate that will make even the most optimistic projections of nuclear look expensive before you can build even the first one'. Then with the shorter lived renewables you get your money back after 10-20 years, and cna spend it on a new system that is $3/MWh (or more likely 10 as land and coupling costs will dominate). And then there's all the geopolitical problems going nuclear entails for the large portion of the world who aren't allowed to make their own fuel. We did that with fossil fuels and it was awful -- and they can be refined by more than 5 countries.
- trashtester 4y ago
  > (china is provably much more cost efficient at doing large projects than anyone else albeit at a high human cost) Tesla and SpaceX is proving that with the right leadership, western countries can also do large projects. Anyway, I think we should let international corporations compete for such projects, kind of like ship construction or more remotely, the car industry. > Ending the day with half of that production in a battery nets you an LCOE around $50 and that is falling by 15%/year with little indication of slowing down. That is great, and I'm a huge believer in solar for many purposes in suitable climates. Where I live, though, we have about 4-6 hours of faint daylight in the middle of winter, when it's really cold. Also, unlike nuclear, no country has so far been able to demonstrate an electrical grid that is powered exclusively by renewables (and Europe has tried). So far, everyone who tried, ended up with a huge reliance on (mostly Russian) natural gas, while having prices quite a bit above $50/MWH LCOE (for the grid as a hole). Currently, consumer facing prices are above $500/MWH, and may reach $1000/MWH this winter. If some country wants to invest in a plan to make them independent of natural gas without nuclear, I wish them the best of luck, but in my opinion, nuclear is the proven path. Also, it's not like one cannot do both at the same time. > Wind is a similar story (although not dropping as precipitously) and moderately anti-correlated. Wind also has longer variance in supply. To some extent, making a grid the size of Australia may lower the variance, but at the cost of a grid that competes with nuclear in zero day investment costs while bringing its own set of political risks (especially in places where the grid needs to span many countries with different cultures). > spend it on a new system that is $3/MWh With all due respect, this is a fantasy number. When you're not including the storage and/or distribution needed to provide stable supply, this number means nothing. (Or rather, it's only relevant as long as only a small fraction of the power comes from such a source.) Actual costs (LCOE, including enough enough storage to provide stable supply) are much higher. Different estimates exist. Your $50/MWh is probably the lowest I've seen. Typically I see $100-$200/MWH when including moderate storage, and several times that (up to $1000/MWH) when including enough storage to last several days with very little wind. From my point of view, paying $30-$50/MWH for stable electricity production, with the lowest carbon emissions (construction emissions for concrete, etc, are about the same as the best renewables), low dependence on minerals that are in short supply (rare earths, etc) using proven technology is quite fair. Hopefully, renewables, thorium, fusion etc will continue to develop to a point where they become clearly superior to nuclear in terms of price. But if so, I see that as a potential upside. Building more nuclear provides a hedge against the risk that this will take a long time. > And then there's all the geopolitical problems going nuclear entails for the large portion of the world who aren't allowed to make their own fuel. It's not for every country, but unlike natural gas, there are enough deposits in stable democratic western countries to power all such countries. Also, there are alternatives sources for China and Russia to not worry too much about the risk either. India, being somewhat in-between these worlds would probably be able to source fuel from at least one. And unlike natural gas, transporting nuclear fuel is quite trivial, due to the energy density. Finally, it's not like we need to have 100% of one source of energy. Some should probably go mostly renewable (Australia, for example), due to climate. Some already have plenty of hydro, and some would benefit from having more (or much more) nuclear. Some might also need to go for fossil fuels, at least as backup. I'm thinking of smaller islands, for instance (like Hawaii). But it would be good if we could soon start to REDUCE the dependence on fossil fuels, instead of consuming more and more every year, globally.

Schroedingersat 4y ago

There are a couple of problems with that analysis. Firstly I am very sceptical of their 60 year claim with an additional 20 years for LTO. Every time I look a supposed 60 year reactor in the west up it is having major problems after less than 40 -- I'd be interested in seeing a survey of reactors broken down by location (china is provably much more cost efficient at doing large projects than anyone else albeit at a high human cost) and whether they're funded by geopolitical incentives (russia subsidizing reactors to induce energy dependence). Additionally half of the reactor lifetime is irrelevant to meeting 2050 targets even if we take design lifetime as true. Even if we take the $30/MWh as an assumption. There are solar projects completing right now on the order of 80c/fixed Watt or $1.2 tilting. These produce 5-6 and 6-8 kWh/day/kW respectively. Ending the day with half of that production in a battery nets you an LCOE around $50 and that is falling by 15%/year with little indication of slowing down. Wind is a similar story (although not dropping as precipitously) and moderately anti-correlated. When 'we promise we can reduce the price by 3x and we'll follow through on that promise for 70 years even though costs have been going up and we've never followed through before' is a pretty hard sell against 'we've been consistently dropping costs for 10 years at a rate that will make even the most optimistic projections of nuclear look expensive before you can build even the first one'. Then with the shorter lived renewables you get your money back after 10-20 years, and cna spend it on a new system that is $3/MWh (or more likely 10 as land and coupling costs will dominate). And then there's all the geopolitical problems going nuclear entails for the large portion of the world who aren't allowed to make their own fuel. We did that with fossil fuels and it was awful -- and they can be refined by more than 5 countries.

trashtester 4y ago
> (china is provably much more cost efficient at doing large projects than anyone else albeit at a high human cost) Tesla and SpaceX is proving that with the right leadership, western countries can also do large projects. Anyway, I think we should let international corporations compete for such projects, kind of like ship construction or more remotely, the car industry. > Ending the day with half of that production in a battery nets you an LCOE around $50 and that is falling by 15%/year with little indication of slowing down. That is great, and I'm a huge believer in solar for many purposes in suitable climates. Where I live, though, we have about 4-6 hours of faint daylight in the middle of winter, when it's really cold. Also, unlike nuclear, no country has so far been able to demonstrate an electrical grid that is powered exclusively by renewables (and Europe has tried). So far, everyone who tried, ended up with a huge reliance on (mostly Russian) natural gas, while having prices quite a bit above $50/MWH LCOE (for the grid as a hole). Currently, consumer facing prices are above $500/MWH, and may reach $1000/MWH this winter. If some country wants to invest in a plan to make them independent of natural gas without nuclear, I wish them the best of luck, but in my opinion, nuclear is the proven path. Also, it's not like one cannot do both at the same time. > Wind is a similar story (although not dropping as precipitously) and moderately anti-correlated. Wind also has longer variance in supply. To some extent, making a grid the size of Australia may lower the variance, but at the cost of a grid that competes with nuclear in zero day investment costs while bringing its own set of political risks (especially in places where the grid needs to span many countries with different cultures). > spend it on a new system that is $3/MWh With all due respect, this is a fantasy number. When you're not including the storage and/or distribution needed to provide stable supply, this number means nothing. (Or rather, it's only relevant as long as only a small fraction of the power comes from such a source.) Actual costs (LCOE, including enough enough storage to provide stable supply) are much higher. Different estimates exist. Your $50/MWh is probably the lowest I've seen. Typically I see $100-$200/MWH when including moderate storage, and several times that (up to $1000/MWH) when including enough storage to last several days with very little wind. From my point of view, paying $30-$50/MWH for stable electricity production, with the lowest carbon emissions (construction emissions for concrete, etc, are about the same as the best renewables), low dependence on minerals that are in short supply (rare earths, etc) using proven technology is quite fair. Hopefully, renewables, thorium, fusion etc will continue to develop to a point where they become clearly superior to nuclear in terms of price. But if so, I see that as a potential upside. Building more nuclear provides a hedge against the risk that this will take a long time. > And then there's all the geopolitical problems going nuclear entails for the large portion of the world who aren't allowed to make their own fuel. It's not for every country, but unlike natural gas, there are enough deposits in stable democratic western countries to power all such countries. Also, there are alternatives sources for China and Russia to not worry too much about the risk either. India, being somewhat in-between these worlds would probably be able to source fuel from at least one. And unlike natural gas, transporting nuclear fuel is quite trivial, due to the energy density. Finally, it's not like we need to have 100% of one source of energy. Some should probably go mostly renewable (Australia, for example), due to climate. Some already have plenty of hydro, and some would benefit from having more (or much more) nuclear. Some might also need to go for fossil fuels, at least as backup. I'm thinking of smaller islands, for instance (like Hawaii). But it would be good if we could soon start to REDUCE the dependence on fossil fuels, instead of consuming more and more every year, globally.