Choristers have recently referred to the nuclear N-stamp. The choir is easily misled. To prevent this, here is a slightly expanded repeat of an early 2023 sermon.
Michael's link is interesting as it relates to this post about the "value" of an N-Stamp, but it also provides the basis for an interesting discussion about your "learning curves" position as described in your last post.
Studying the debacle at Shaw Modular and implementing some straightforward, affordable corrections could provide a step change in cost and schedule performance that would look a lot like "learning" at a rapid rate.
Shipyards with engineering, quality, inspection and manufacturing processes that have been refined through decades of practice might not display much learning with a series of ships, but that is because they have achieved a very high level of performance already.
It's a lot easier to be "most improved" when you're not very good and difficult to achieve that award if you are already at a world leading performance level.
True, but since the technical skills required to build a nuclear plant are pretty much the same as the skills required to build a coal plant, in the past we have seen people who are good at building coal plants switch to nuke without much trouble, and we don't see much first time effect. See Figure 2. In the US, the first full scale plants were the cheapest.
This time around it could be different. Nobody has built coal plants in the US in along time. Nobody has built much of anything big in the US in a long time. The nuclear malaise has metastasized. It is hard to see any improvement unless we bring in the Koreans or the Japanese as we did for autos. But I doubt they will come under the current regulatory system.
Jack - for better or for worse, the engineering and quality skills needed for successful nuclear plant construction in the U.S. are different from those used for coal plants.
Though the U.S. hasn’t built many coal plant in the last few decades, we have built a large number of combined cycle plants that include high temp, high pressure steam plant bottoming cycles.
Those steam plant aren’t as large as coal steam plants, so maybe that’s another argument for smaller nuclear plants.
If you are talking about the "skills" to navigate an NRC-style regulatory process, I can assure you it is for the worse, much worse., much , much worse. But I'm talking about the skills to actually pour the concrete, bend the metal, and pull the wire. The skills that really count.
There is no such thing as nuclear quality as Shaw demonstrated. There is just good quality and lousy quality.
Good point about Combined Cycle, but it is not an argument for small scale. It is a reason for optimism if we regulate nuclear like we regulate gas.
There is ice speed rail, train being built between Dallas and Houston by the Japanese. They have very strict manufacturing standards.
From your reporting it sounds like a Korean shipyard could do the same thing with a nuclear reactor? The only thing that would be damaged is American pride.
Technically, of course they could. And under the present yard system, the quality would improve a lot. For example, in order to fit together 3000 ton super-blocks, the dimensional tolerances that the yards must achieve are far tighter that those for a stick built nuclear plant.
But the yards are petrified of nuclear. They know if they allow current nuclear regualtion into the yard, their whole carefully tuned system will grind to a disastrous halt. The last thing a yard needs is an NRC inspector telling them how to build things.
Shaw bought the inside track to selection as the module supplier by purchasing 20% of Westinghouse as part of the group that purchased the company from BNFL in Oct 2006. The total purchase price was $5.4 B; Shaw Group invested $1.08 billion.
From a New York Times article discussing the deal:
"Under the deal, Shaw said it would get certain exclusive agreements for engineering, procurement and construction services on future Westinghouse AP 1000 nuclear power plants, the chosen design for 10 proposed new power units in the United States and four in China."
This is a good example of the potential risks associated with taking substantial investments from "strategic" investors whose business is to provide important parts of the supply chain for the final product.
It seems like Terrapower, Nuscale, and other "advanced" nuclear developers are repeating this mistake by taking money from Doosan, Hyundai and the like. The problem here is an exorbitantly expensive, often ineffective and counterproductive, and massively overhyped quality control system, not where the money comes from.
https://www.enr.com/articles/43325-witness-to-the-origins-of-a-huge-nuclear-construction-flop
Michael,
Thanks. I should have included this link. Everybody should check it out.
The depth of detail at the link will be interesting to some, but it's non-essential to your main point (which is very compelling as presented).
Glad to add a little to the conversation. Cheers.
Jack:
Michael's link is interesting as it relates to this post about the "value" of an N-Stamp, but it also provides the basis for an interesting discussion about your "learning curves" position as described in your last post.
Studying the debacle at Shaw Modular and implementing some straightforward, affordable corrections could provide a step change in cost and schedule performance that would look a lot like "learning" at a rapid rate.
Shipyards with engineering, quality, inspection and manufacturing processes that have been refined through decades of practice might not display much learning with a series of ships, but that is because they have achieved a very high level of performance already.
It's a lot easier to be "most improved" when you're not very good and difficult to achieve that award if you are already at a world leading performance level.
Rod,
True, but since the technical skills required to build a nuclear plant are pretty much the same as the skills required to build a coal plant, in the past we have seen people who are good at building coal plants switch to nuke without much trouble, and we don't see much first time effect. See Figure 2. In the US, the first full scale plants were the cheapest.
This time around it could be different. Nobody has built coal plants in the US in along time. Nobody has built much of anything big in the US in a long time. The nuclear malaise has metastasized. It is hard to see any improvement unless we bring in the Koreans or the Japanese as we did for autos. But I doubt they will come under the current regulatory system.
Jack - for better or for worse, the engineering and quality skills needed for successful nuclear plant construction in the U.S. are different from those used for coal plants.
Though the U.S. hasn’t built many coal plant in the last few decades, we have built a large number of combined cycle plants that include high temp, high pressure steam plant bottoming cycles.
Those steam plant aren’t as large as coal steam plants, so maybe that’s another argument for smaller nuclear plants.
If you are talking about the "skills" to navigate an NRC-style regulatory process, I can assure you it is for the worse, much worse., much , much worse. But I'm talking about the skills to actually pour the concrete, bend the metal, and pull the wire. The skills that really count.
There is no such thing as nuclear quality as Shaw demonstrated. There is just good quality and lousy quality.
Good point about Combined Cycle, but it is not an argument for small scale. It is a reason for optimism if we regulate nuclear like we regulate gas.
Wow, this was enlightening. I assumed the N stamp was the nuclear equivalent of an ASME pressure vessel stamp wrt inspections, etc.
David,
Shaw's QA manual undoubtedly listed all sorts of inspections and tests, but obviously it did not require any outside inspection.
Trusting but not verifying vendors is bad practice. Agreed. Thanks for the insight.
When I was running a tanker fleet we had a mantra:
Bid everybody; trust nobody.
US nuclear does just the opposite in part because the regulatory system prevents them from bidding everybody and focuses verification on paperwork.
There is ice speed rail, train being built between Dallas and Houston by the Japanese. They have very strict manufacturing standards.
From your reporting it sounds like a Korean shipyard could do the same thing with a nuclear reactor? The only thing that would be damaged is American pride.
Late,
Technically, of course they could. And under the present yard system, the quality would improve a lot. For example, in order to fit together 3000 ton super-blocks, the dimensional tolerances that the yards must achieve are far tighter that those for a stick built nuclear plant.
But the yards are petrified of nuclear. They know if they allow current nuclear regualtion into the yard, their whole carefully tuned system will grind to a disastrous halt. The last thing a yard needs is an NRC inspector telling them how to build things.
Shaw bought the inside track to selection as the module supplier by purchasing 20% of Westinghouse as part of the group that purchased the company from BNFL in Oct 2006. The total purchase price was $5.4 B; Shaw Group invested $1.08 billion.
From a New York Times article discussing the deal:
"Under the deal, Shaw said it would get certain exclusive agreements for engineering, procurement and construction services on future Westinghouse AP 1000 nuclear power plants, the chosen design for 10 proposed new power units in the United States and four in China."
This is a good example of the potential risks associated with taking substantial investments from "strategic" investors whose business is to provide important parts of the supply chain for the final product.
Rod,
It seems like Terrapower, Nuscale, and other "advanced" nuclear developers are repeating this mistake by taking money from Doosan, Hyundai and the like. The problem here is an exorbitantly expensive, often ineffective and counterproductive, and massively overhyped quality control system, not where the money comes from.