There is another aspect, a political aspect, that applies to the UK and Europe:
In the UK and Europe, all nuclear plants will have to be shut down within a decade or two.
And all fuel and spent fuel removed to other regions.
Why?
Because the UK and Europe are well on course to becoming Islamic. Such materials represent a significant security risk - radioactive dirty bombs could be constructed.
The UK has 140 tonnes of Plutonium. Crude but effective atomic bombs only require a few kilograms of Plutonium. The yield of the explosion is not all that significant if it is delivered by boat or by truck. The yield is only significant if delivered by plane or rocket.
Additionally, in the case of the UK and France, with nuclear weapons, these weapons must be removed to safer areas, such as to the USA.
I have written to UK Members of Parliament detailing the urgency of the task:
Your panegyric for St. Lucie 2 is a bit more than misleading. Florida Power and Light(FPL) was indeed pretty good at building nuclear plants. According to the Lovering database, Turkey Point 3/4, a Westinghouse 3 loop 800 MW 3 loop design, came in at an overnight cost of $795/kW in2024 dollars. As far as I know, this is the lowest NNP real ovnite cost ever. The construction time for TP 3/4 was mediocre 6 years. I don't why. Other US plants at this time were going up in around 4 years. Still an FPL executive told me that Turkey Point could produce power at around 1 cent per kWh at the gate. This was around 2010. I lived in the Upper Keys between 1980 and 2018, and enjoyed some of the cheapest retail power in the country for 40 years. Even I will admit Turkey Point was pretty close to should-cost nuclear power.
FPL's St. Lucie 1 is an 900 MW. two loop, CE design. Construction started in mid 1970 and finished in mid 1976 at an overnight cost of $2885 2024 USD, 3.6 times Turkey Point's real ovnite cost. Things got even worse for St. Lucie 2. St. Lucie 2, an identical CE plant, took 6 years to build and came in at an overnight cost of $4957 2024 USD, 6.2 times Turkey Point' real cost. No where near should-cost.
The CE design should have had a real unit cost less than the Westinghouse TP's. The FPL guys probably did get better at building NPP's, but those improvements were vastly over-whelmed by an autocratic regulatory ratchet. The fact that St Lucie 2 was one of the cheaper US plants completed in the early 1980's, only makes the case against misguided regulation stronger.
Interesting comment from a listener, thanks for forwarding.
First, citing Turkey Point 3 and 4 as a “should cost” benchmark is really germane to the "basics that people get wrong". Famously, Turkey Point was the last “firm fixed price” plant Bechtel ever built. The reason they never signed another contract like that is simple: the actual cost to build was 88% higher than the price they got paid by FPL. Bechtel lost billions of dollars, effectively subsidizing FPL and its ratepayers.
If the argument is that a “should cost” plant is one where the engineering and construction firm bleeds billions to build it, that might be a great deal for ratepayers, but it is hardly a sustainable business model. The listener correctly notes this was before the AEC/NRC regulations tightened, but they conveniently ignore the financial disaster that made those prices impossible to replicate. In defense of Bechtel, GE and Westinghouse lost far more money (tens of billions) in building the “turnkey plants” (Oyster Creek, Point Beach, etc.), which is why they both stopped building turnkey plants in 1966 – again, way before the AEC/NRC regulations tightened.
It is further odd that your listener points to the cost escalation between St. Lucie 1 (SL1) and St. Lucie 2 (SL2) as evidence for their conclusory theory that the increase was due to the “regulatory ratchet.” As is well established, overnight costs for both nuclear and fossil power plants were skyrocketing in the 1970s and 1980s, driven by factors like labor costs inflating by 25% per year by 1976. It is hard to overestimate this impact: on an inflation-adjusted basis, the overnight cost per kWe of a coal or oil plant coming online in 1983 (the year SL2 came online) was 2.5 times higher than the cost of the same plant coming online in 1976 (SL1's year), a finding detailed by many, see, e.g. Komanoff (1984)
Critically, the overnight cost increase for SL2 was only 1.8 times higher than SL1 (on an inflation-adjusted per kWe basis)—a significantly lower escalation than that of the average coal or oil plant built during the same period. One must ask why the listener concludes, with zero analysis, that the NRC's “regulatory ratchet” was the primary culprit when the NRC-regulated nuclear construction inflated significantly slower (at St. Lucie) than the average non-NRC regulated construction of a fossil fuel plant over the same period.
Luckily, we need not speculate on the actual cost drivers here; the data is established.
Contrary to the conclusions of your listener, St. Lucie 2 actually faced significantly lower regulatory-driven cost increases under the NRC than Unit 1 did under the AEC. Cohen (1987), analyzed the actual EBASCO and FPL change order databases (rather than just hand waving) for both SL1 and SL2, revealed a stark contrast: while 87% of cost increases at St. Lucie 1 (AEC) were driven by regulatory issues, that figure dropped to just 48% for St. Lucie 2 (NRC). While this is, of course, still too high, it certainly is in direct contradiction to the listeners conclusion.
As Cohen demonstrated, the shift from ad-hoc, unpredictable licensing approach under the AEC toward more standardized NRC approach actually generated substantial savings.
I do not write this to discourage the listener; their passion is clear, and the industry needs advocates. However, advocacy detached from engineering reality is of limited value. I encourage them to stop making conclusions without analysis or familiarity with the previous literature, and instead acquaint themselves with actual engineering history of the plants. Understanding the basics, after all, is the precise point where passion can become useful, rather than internet pablum.
Of course, I always disparage the low road, but that's a lie. I'm happy to get down in the mud with James. So here goes.
I do not listen to podcasts. They take way too long, and the content density is minuscule. My name is not listener. If you like me, I'm Jack. If you don't, I'm Devanney. I answer to either.
Krellenstein claims I am "detached from engineering reality". After working at Newport News and Electric Boat, I was recruited to the MIT faculty and earned tenure in the minimum time possible. Number of reasons, but perhaps the most important was I designed and tested the hull form of the DDX, which later became the Spruance Class destroyer. The Navy was so happy with the hull's efficiency and sea keeping performance, that they used a slightly elongated version for the Ticonderoga guided missile cruiser.
Shortly after getting tenure, I left MIT to seek my future in the tanker market. I put together an 8 ship fleet of laid up ULCC's for the Loews Corporation (the Tisch family). We bought these ships for 47 million dollars and a few years later sold 50\% in a deal that valued the fleet at \$315 million dollars. In the meantime, I ran the fleet.
I later convinced Jim Tisch to invest in 8 new buildings in Korea, 4 VLCC's and 4 ULCC's. The latter were the largest double hall tankers every built by about 30%. Total project cost about $700,000,000. I wrote the specs and oversaw the building and later operation. These ships were sold at an extremely high rate of return. As far as I know, Krellenstein has never built anything other than a podcast. I'm sure he will correct me if that is not the case.
Most importantly, I have worked in both the Navy style environment, on which the US nuclear regulation is modeled and in an effectively competitive market. They are on different planets in terms of both cost and performance. My goal is to somehow articulate that difference.
OK time to climb out of the gutter, and get to the substance of the matter. In the mid late 1960's, according to Lovering, there were 14 US nuclear plant built under turnkey contracts. The vendors did take losses on some of those contracts in order to get a dominate position in a highly moated market and lock in the profits from providing the fuel. But the losses were in the tens of million of dollars, not billions.
The vendors quickly learned that the AEC rules could change on them after the contract had been signed. AEA Section 187 explicitly says this is the case. The only thing you could be sure of is that the rules would keep getting more expensive. So everybody switched to cost plus contracts, which in regulated monopoly allowed the costs of regulatory uncertainty and volatility to be transferred to the rate payer, It also screwed up everybody's incentives. But that took a while to sink in. Meanwhile coal plants continued to be built on turnkey contracts.
According to Lovering, Turkey Point 3/4 were not turnkey contracts. Bechtel may have been the EPC but my discussions with FPL executives tells me FPL was very much in charge of the project. They managed it. Bechtel did the working drawings. I'm quite sure Bechtel did not lose billions on a two unit plant whose overnight cost was about 100 million 1970 dollars. Perhaps James will point us to his sources for this claim.
I also know this. Turkey Point has been a cash cow for FPL; and, as a 40 year customer of FPL in the Upper Keys. I enjoyed some of the cheapest electricity in the US. Win for the ratepayer; win for the investors. Despite Krellenstein's claims, these plants truly were dirt cheap.
I don't know that much about St Lucie, but I do know by then FPL had developed a strong nuclear plant management team. Effectively, they were their own EPC. The two plants did come in at well below the average cost for US plants delivered at the same time, including plants where Bechtel was the engineer. This isn't saying much. Despite FPL leadership, the REAL cost over Turkey Point was a factor of 3 for St Lucie 1 and 6 for St Lucie 2. Real cost means we have factored in monetary inflation. Does this mean FPL became more incompetent the more plants they built?
The problem was not monetary inflation, it was a massive jump in materials: steel, concrete, cabling, and most of all paperwork. Please see Crowley and Phung. There's a brief summary at
In about 1981, the FPL team was brought into resurrect Seabrook. FPL needed more investor money so they offered the contrarian Tisch family a share in the project. Despite the fact I had never been in an nuclear plant, I was as close to a nuclear inspector as the Tisch's had. So they sent me to Seabrook to check the plant out. We were escorted around by FPL guys who clearly were in charge of the project. They were even counting the number of pipe hangers installed each day. If there were any Bechtel/Fluor etc people around I never saw them.
I was astounded by the over-building. All welds ground so smooth you could barely tell where the welds were. The emergency diesel building had concrete walls that had to have been 2 feet thick, You felt like like you were on the Atlantikwall. Of course, a fire in this space would take out all the backup generators.
I was most blown away by the cable spreading room. This area had hundreds of cables coming in each with scores of conductors. The conductors were separated in this space and fed to the control room above, I asked my guides where are the multiplexers? The FPL guys said the NRC won't let us use them. I asked why not, They said "safety". I said that's ridiculous, With multiplexers you could afford to have two or more separate paths to the control room. As it stands if you have a fire in this space, you are going to leave the whole plant blind. They sort of chuckled and said "It has already happened." They were referring to the Browns Ferry fire in 1975, which being a dumb naval architect I was unaware of.
I went back to the Tisch's and said this is one nutty business. Don't touch it with a barge pole. They decided to pass.
One final point, Chris, I was responding to your post, not James'. You can't all of a sudden turn into some kind of referee, counting Krellenstein points and Devanney points. James has to snuggle up to the NRC. That's his job and that's what he has been indoctrinated to do. You like me do not have that excuse. You have to decide whether you are going to try and make a real dent in energy poverty and global warming and help nuclear fission realize its promethean promise; or go with the nuclear establishment and support a misdirected, autocratic, rotten regulatory system
excellent exposition. the integration process knowledge is lacking and the decisive task is to build it, to try out in 'prototype' projects and refine it before the big time. this is the sort of analysis of the needs of state capacity prompting corporate endeavor that the advanced countries need to emerge from their arrested development, trapped in a resource curse of software engineering global sales alongside atrophied local plant too costly to automate and reconstruct
What is needed is an update to what is the roadblock in the Carney/Ford governments on building CANDUs. Why the nonsensical cost claim on refurbishing Pickering. Why is Carney focusing so much effort on pipelines and ignoring the CANDU. Maybe Chris Adlam can fill us in?
Thanks! This was the best run down on what it will take to get nuclear up and running again in North America that I have seen.
Please like, subscribe, pledge and share!
That is a highly informative article, thank you.
There is another aspect, a political aspect, that applies to the UK and Europe:
In the UK and Europe, all nuclear plants will have to be shut down within a decade or two.
And all fuel and spent fuel removed to other regions.
Why?
Because the UK and Europe are well on course to becoming Islamic. Such materials represent a significant security risk - radioactive dirty bombs could be constructed.
The UK has 140 tonnes of Plutonium. Crude but effective atomic bombs only require a few kilograms of Plutonium. The yield of the explosion is not all that significant if it is delivered by boat or by truck. The yield is only significant if delivered by plane or rocket.
Additionally, in the case of the UK and France, with nuclear weapons, these weapons must be removed to safer areas, such as to the USA.
I have written to UK Members of Parliament detailing the urgency of the task:
https://hellish2050.substack.com/p/spreadsheet-list-of-uk-mp-email-addresses
You may wish to read the book: "Islamic Britain With Nukes", available here:
https://hellish2050.substack.com/p/book-catalogue
Chris,
Your panegyric for St. Lucie 2 is a bit more than misleading. Florida Power and Light(FPL) was indeed pretty good at building nuclear plants. According to the Lovering database, Turkey Point 3/4, a Westinghouse 3 loop 800 MW 3 loop design, came in at an overnight cost of $795/kW in2024 dollars. As far as I know, this is the lowest NNP real ovnite cost ever. The construction time for TP 3/4 was mediocre 6 years. I don't why. Other US plants at this time were going up in around 4 years. Still an FPL executive told me that Turkey Point could produce power at around 1 cent per kWh at the gate. This was around 2010. I lived in the Upper Keys between 1980 and 2018, and enjoyed some of the cheapest retail power in the country for 40 years. Even I will admit Turkey Point was pretty close to should-cost nuclear power.
FPL's St. Lucie 1 is an 900 MW. two loop, CE design. Construction started in mid 1970 and finished in mid 1976 at an overnight cost of $2885 2024 USD, 3.6 times Turkey Point's real ovnite cost. Things got even worse for St. Lucie 2. St. Lucie 2, an identical CE plant, took 6 years to build and came in at an overnight cost of $4957 2024 USD, 6.2 times Turkey Point' real cost. No where near should-cost.
The CE design should have had a real unit cost less than the Westinghouse TP's. The FPL guys probably did get better at building NPP's, but those improvements were vastly over-whelmed by an autocratic regulatory ratchet. The fact that St Lucie 2 was one of the cheaper US plants completed in the early 1980's, only makes the case against misguided regulation stronger.
Hi Jack.. A reply from James.
Interesting comment from a listener, thanks for forwarding.
First, citing Turkey Point 3 and 4 as a “should cost” benchmark is really germane to the "basics that people get wrong". Famously, Turkey Point was the last “firm fixed price” plant Bechtel ever built. The reason they never signed another contract like that is simple: the actual cost to build was 88% higher than the price they got paid by FPL. Bechtel lost billions of dollars, effectively subsidizing FPL and its ratepayers.
If the argument is that a “should cost” plant is one where the engineering and construction firm bleeds billions to build it, that might be a great deal for ratepayers, but it is hardly a sustainable business model. The listener correctly notes this was before the AEC/NRC regulations tightened, but they conveniently ignore the financial disaster that made those prices impossible to replicate. In defense of Bechtel, GE and Westinghouse lost far more money (tens of billions) in building the “turnkey plants” (Oyster Creek, Point Beach, etc.), which is why they both stopped building turnkey plants in 1966 – again, way before the AEC/NRC regulations tightened.
It is further odd that your listener points to the cost escalation between St. Lucie 1 (SL1) and St. Lucie 2 (SL2) as evidence for their conclusory theory that the increase was due to the “regulatory ratchet.” As is well established, overnight costs for both nuclear and fossil power plants were skyrocketing in the 1970s and 1980s, driven by factors like labor costs inflating by 25% per year by 1976. It is hard to overestimate this impact: on an inflation-adjusted basis, the overnight cost per kWe of a coal or oil plant coming online in 1983 (the year SL2 came online) was 2.5 times higher than the cost of the same plant coming online in 1976 (SL1's year), a finding detailed by many, see, e.g. Komanoff (1984)
Critically, the overnight cost increase for SL2 was only 1.8 times higher than SL1 (on an inflation-adjusted per kWe basis)—a significantly lower escalation than that of the average coal or oil plant built during the same period. One must ask why the listener concludes, with zero analysis, that the NRC's “regulatory ratchet” was the primary culprit when the NRC-regulated nuclear construction inflated significantly slower (at St. Lucie) than the average non-NRC regulated construction of a fossil fuel plant over the same period.
Luckily, we need not speculate on the actual cost drivers here; the data is established.
Contrary to the conclusions of your listener, St. Lucie 2 actually faced significantly lower regulatory-driven cost increases under the NRC than Unit 1 did under the AEC. Cohen (1987), analyzed the actual EBASCO and FPL change order databases (rather than just hand waving) for both SL1 and SL2, revealed a stark contrast: while 87% of cost increases at St. Lucie 1 (AEC) were driven by regulatory issues, that figure dropped to just 48% for St. Lucie 2 (NRC). While this is, of course, still too high, it certainly is in direct contradiction to the listeners conclusion.
As Cohen demonstrated, the shift from ad-hoc, unpredictable licensing approach under the AEC toward more standardized NRC approach actually generated substantial savings.
I do not write this to discourage the listener; their passion is clear, and the industry needs advocates. However, advocacy detached from engineering reality is of limited value. I encourage them to stop making conclusions without analysis or familiarity with the previous literature, and instead acquaint themselves with actual engineering history of the plants. Understanding the basics, after all, is the precise point where passion can become useful, rather than internet pablum.
Chris,
James' response forces me to argue on two levels
1) The Low Road: ad hominum arguments
2) High Road: the substance of the issue,
Of course, I always disparage the low road, but that's a lie. I'm happy to get down in the mud with James. So here goes.
I do not listen to podcasts. They take way too long, and the content density is minuscule. My name is not listener. If you like me, I'm Jack. If you don't, I'm Devanney. I answer to either.
Krellenstein claims I am "detached from engineering reality". After working at Newport News and Electric Boat, I was recruited to the MIT faculty and earned tenure in the minimum time possible. Number of reasons, but perhaps the most important was I designed and tested the hull form of the DDX, which later became the Spruance Class destroyer. The Navy was so happy with the hull's efficiency and sea keeping performance, that they used a slightly elongated version for the Ticonderoga guided missile cruiser.
Shortly after getting tenure, I left MIT to seek my future in the tanker market. I put together an 8 ship fleet of laid up ULCC's for the Loews Corporation (the Tisch family). We bought these ships for 47 million dollars and a few years later sold 50\% in a deal that valued the fleet at \$315 million dollars. In the meantime, I ran the fleet.
I later convinced Jim Tisch to invest in 8 new buildings in Korea, 4 VLCC's and 4 ULCC's. The latter were the largest double hall tankers every built by about 30%. Total project cost about $700,000,000. I wrote the specs and oversaw the building and later operation. These ships were sold at an extremely high rate of return. As far as I know, Krellenstein has never built anything other than a podcast. I'm sure he will correct me if that is not the case.
Most importantly, I have worked in both the Navy style environment, on which the US nuclear regulation is modeled and in an effectively competitive market. They are on different planets in terms of both cost and performance. My goal is to somehow articulate that difference.
OK time to climb out of the gutter, and get to the substance of the matter. In the mid late 1960's, according to Lovering, there were 14 US nuclear plant built under turnkey contracts. The vendors did take losses on some of those contracts in order to get a dominate position in a highly moated market and lock in the profits from providing the fuel. But the losses were in the tens of million of dollars, not billions.
The vendors quickly learned that the AEC rules could change on them after the contract had been signed. AEA Section 187 explicitly says this is the case. The only thing you could be sure of is that the rules would keep getting more expensive. So everybody switched to cost plus contracts, which in regulated monopoly allowed the costs of regulatory uncertainty and volatility to be transferred to the rate payer, It also screwed up everybody's incentives. But that took a while to sink in. Meanwhile coal plants continued to be built on turnkey contracts.
According to Lovering, Turkey Point 3/4 were not turnkey contracts. Bechtel may have been the EPC but my discussions with FPL executives tells me FPL was very much in charge of the project. They managed it. Bechtel did the working drawings. I'm quite sure Bechtel did not lose billions on a two unit plant whose overnight cost was about 100 million 1970 dollars. Perhaps James will point us to his sources for this claim.
I also know this. Turkey Point has been a cash cow for FPL; and, as a 40 year customer of FPL in the Upper Keys. I enjoyed some of the cheapest electricity in the US. Win for the ratepayer; win for the investors. Despite Krellenstein's claims, these plants truly were dirt cheap.
I don't know that much about St Lucie, but I do know by then FPL had developed a strong nuclear plant management team. Effectively, they were their own EPC. The two plants did come in at well below the average cost for US plants delivered at the same time, including plants where Bechtel was the engineer. This isn't saying much. Despite FPL leadership, the REAL cost over Turkey Point was a factor of 3 for St Lucie 1 and 6 for St Lucie 2. Real cost means we have factored in monetary inflation. Does this mean FPL became more incompetent the more plants they built?
The problem was not monetary inflation, it was a massive jump in materials: steel, concrete, cabling, and most of all paperwork. Please see Crowley and Phung. There's a brief summary at
https://jackdevanney.substack.com/p/nuclear-power-not-only-should-be
In about 1981, the FPL team was brought into resurrect Seabrook. FPL needed more investor money so they offered the contrarian Tisch family a share in the project. Despite the fact I had never been in an nuclear plant, I was as close to a nuclear inspector as the Tisch's had. So they sent me to Seabrook to check the plant out. We were escorted around by FPL guys who clearly were in charge of the project. They were even counting the number of pipe hangers installed each day. If there were any Bechtel/Fluor etc people around I never saw them.
I was astounded by the over-building. All welds ground so smooth you could barely tell where the welds were. The emergency diesel building had concrete walls that had to have been 2 feet thick, You felt like like you were on the Atlantikwall. Of course, a fire in this space would take out all the backup generators.
I was most blown away by the cable spreading room. This area had hundreds of cables coming in each with scores of conductors. The conductors were separated in this space and fed to the control room above, I asked my guides where are the multiplexers? The FPL guys said the NRC won't let us use them. I asked why not, They said "safety". I said that's ridiculous, With multiplexers you could afford to have two or more separate paths to the control room. As it stands if you have a fire in this space, you are going to leave the whole plant blind. They sort of chuckled and said "It has already happened." They were referring to the Browns Ferry fire in 1975, which being a dumb naval architect I was unaware of.
I went back to the Tisch's and said this is one nutty business. Don't touch it with a barge pole. They decided to pass.
One final point, Chris, I was responding to your post, not James'. You can't all of a sudden turn into some kind of referee, counting Krellenstein points and Devanney points. James has to snuggle up to the NRC. That's his job and that's what he has been indoctrinated to do. You like me do not have that excuse. You have to decide whether you are going to try and make a real dent in energy poverty and global warming and help nuclear fission realize its promethean promise; or go with the nuclear establishment and support a misdirected, autocratic, rotten regulatory system
excellent exposition. the integration process knowledge is lacking and the decisive task is to build it, to try out in 'prototype' projects and refine it before the big time. this is the sort of analysis of the needs of state capacity prompting corporate endeavor that the advanced countries need to emerge from their arrested development, trapped in a resource curse of software engineering global sales alongside atrophied local plant too costly to automate and reconstruct
Nuclear journalism par excellence
What is needed is an update to what is the roadblock in the Carney/Ford governments on building CANDUs. Why the nonsensical cost claim on refurbishing Pickering. Why is Carney focusing so much effort on pipelines and ignoring the CANDU. Maybe Chris Adlam can fill us in?