top of page

Is Regulation Strangling Nuclear Energy?

James Krellenstein

Monday, June 24, 2024

00:00:00:07 - 00:00:20:13

James Krellenstein

And I don't want it to sound like I'm saying that there's nothing that needs to change at the NRC. Nothing could be further from the truth. What I'm saying is, if we're looking for the root cause of what is going on in the US nuclear industry, particularly on the Newbuild side, I think we have way overstated the case for it's the nuclear regulations, stupid.


00:00:20:18 - 00:00:30:18

James Krellenstein

And we're going to pass law after law after law and not actually deal with some of the fundamental issues that we have in this industry.


00:00:30:20 - 00:00:58:19

Chris Keefer

Welcome back to Decouple. Today I'm joined once again by James Brownstein. a guest at this point who needs no introduction. but to me really typifies an ideal decouple. Guest. I'm going to fluff your feathers here, James, as I'm prone to do, but, the right the right mixture of, you know, technical expertise, encyclopedic knowledge, and, a very rare, characteristic, which is what seems to be like pretty total freedom of speech.


00:00:58:19 - 00:01:15:22

Chris Keefer

So we don't do heterodoxy for heterodoxy sake here, on decouple. But, it certainly is refreshing, to be able to have, these kind of conversations, a lot of people who reach out to me and, you know, would like to be able to, to share opinions but just are not able to, so hopefully we can keep you that way.


00:01:15:22 - 00:01:20:19

Chris Keefer

James. I noticed the decouple t shirt. it's nice to see. Welcome back.


00:01:20:21 - 00:01:25:23

James Krellenstein

Thanks. So I guess we're talking about nuclear regulation. we are, we are.


00:01:25:23 - 00:01:49:19

Chris Keefer

I think, you know, I don't know if you've been formally branded as an NK apologist. but you reached out to me a little while ago wanting to, talk regulation and nk, and, just last week, I believe, Ted Nordhaus came out with a piece. it's a regulation, stupid. And, you know, there's this question that dogs, as nuclear advocates, which is that, particularly Western nuclear, it's just not not affordable.


00:01:49:19 - 00:02:19:04

Chris Keefer

Not cost effective, certainly, in the construction phase, and everyone wants to find out the reasons why. And I think a really convenient reason that's been identified and is kind of a widespread perspective, is that this is just an area that's regulation, that's kind of taxing it to death. So, really excited to engage with that. you know, we have the example of the big cost plume of, of, you know, overnight construction costs in the US after Three Mile Island, which a lot of people tribute to overregulation.


00:02:19:04 - 00:02:36:08

Chris Keefer

We have, you know, new scale having to pour 500 million into a, design certification process. Oklo being turned down. There's lots of reasons why people are angry at the NRC and angry at, at, you know, overregulation or the perception of regulation in general. And I really hope to, to break that down with you.


00:02:36:10 - 00:03:08:15

James Krellenstein

So I think this is a great topic. And, you know, I think sort of, you know, the piece, although Nordhaus, I think where Ted was trying to make it seem like there are some apparent hoards of progressive, of nuclear advocates who are fighting against, you know, the sort of, dogmatic, position that he sort of, exemplified by his, position, you know, paraphrasing, James Carville, the 1992, Clinton, George H.W. Bush election.


00:03:08:15 - 00:03:29:06

James Krellenstein

You know, it's the economy, stupid. It's, it's the nuclear regulation, stupid. And, you know, I was actually talking about this, this morning with, a nuclear engineering colleague of mine, and, you know, all of us were sort of at this, this sort of after reading it was sort of like, does Ted secretly live in 1983?


00:03:29:09 - 00:03:57:16

James Krellenstein

Because I think the the real question that we have is not was nuclear regulation, particularly the nuclear regulatory system that we had in the 80s post Three Mile Island. Was it a huge driver during the post Three Mile Island and even to some extent during the 70s, pre TMI in in in spring of 79. Absolutely. Was a very, very large driver.


00:03:57:18 - 00:04:18:16

James Krellenstein

and arguably there's still obviously major improvements that can be done here. But the question that we need to ask is there's been a huge seismic shift in the way that the United States regulates nuclear power plants, and we're talking about the US here. This was really focused on the Nuclear Regulatory Commission. And bizarrely, none of those shifts were sort of dealt with in the piece.


00:04:18:16 - 00:04:40:12

James Krellenstein

And I wonder how many people actually appreciate how different the sort of regulatory approach that a nuclear power plant can take, to get regulated by the NRC today versus is what it was like in the 70s and 80s and even some of the shifts that were already happening in the 70s and 80s, we don't seem to talk about.


00:04:40:17 - 00:05:06:17

James Krellenstein

So but before we go into that and before also, I think it's useful to sort of separate this from the nuclear regulation component of this to, from the sort of National Environmental Policy Act, Clean Water Act, and other sort of non-nuclear regulation that has to get done. I want to ask three basic framing questions that should be really easy to answer if it is, in fact, is the nuclear regulation stupid?


00:05:06:19 - 00:05:34:05

James Krellenstein

Right? And the first question is, imagine a situation in which we're trying to build a newly designed reactor by the United States, and we haven't one country. We have a system that is really, really good, go through whatever means it's churning out light water reactors, gigawatt scale light water reactors, in median five and a half years.


00:05:34:05 - 00:06:11:07

James Krellenstein

And it's literally done dozens of this just over two decades. Okay. And then we build another reactor in the United States, the exact same design. And we're gonna argue, oh, it's the the nuclear regulator, right? You know, that's the cause, the main cause of construction and schedule overruns that we've seen plaguing the nuclear industry. And obviously, we have a country where the nuclear regulator, whatever they are, is able to construct plants in a way or allows construction in a way that they're happening in five and a half years, literally, and 50% being faster than five and a half years, by the way.


00:06:11:09 - 00:06:36:17

James Krellenstein

Right. And then we build it in the United States, almost surely if the nuclear regulator, is the, cause of this in the United States, the US sort of build is going to be much, much slower, much, much more problematic. Right? Because not the industry that's fundamentally the problem. It's ultimately the regulators. Stupid. Right? Great. So let's run this experiment.


00:06:36:18 - 00:07:02:08

James Krellenstein

Oh wait, we actually did do this experiment. And we built two AP one thousands in Georgia. Vogel unit three in units four we got for 81,000 in China, which is sun Moon unit no. Number one and number two and high ING unit one and unit number two. And lo and behold, they hit nearly the exact same time. Sun moon unit number one takes nine years, four months.


00:07:02:10 - 00:07:34:02

James Krellenstein

And then, Vogel three, which gets caught up in the Covid epidemic. Right. As we've discussed earlier, remote work for nuclear power plant construction doesn't really work. Zoom calls for, you know, placing rebar doesn't work. But that was only about ten years, three months for nuclear construction. And so literally, they're about the same. construction duration. So obviously the nuclear regulator in either one, it's not the rate limiting factor here.


00:07:34:04 - 00:08:08:09

James Krellenstein

But wait, you say, oh, of course, that was a Western design. It was designed to the nk's unrealistic safety requirements. And therefore the design became, buildable. Right. And so that's why even the Chinese couldn't build it. Right? Because they were too, you know, catering to the Nuclear Regulatory Commission for that design certification. Great point. Now imagine that we had, let's say, the first nuclear reactor design that ever went through the modern NRC process.


00:08:08:11 - 00:08:50:19

James Krellenstein

Right? Like the first one that went through a part 52 design certification process and got it. By the way, the NRC granted that design certification. Let's build that reactor and let's compare it to the generation two reactors that came before that we built in the United States and around the world. Right. If, in fact, the NRC is requiring this to be these new safety requirements, to be so onerous that no one can build this plant, surely that reactor design, which is the first one that goes through this new regulatory process, is is going to be slower, more expensive and just impossible for anyone to build.


00:08:50:21 - 00:09:33:12

James Krellenstein

But wait, and that reactor design, this is not a hypothetical anymore. We built the general Electric, Hitachi, Toshiba Advanced Boiling Water Reactor, the first reactor to get a design certification from the US Nuclear Regulatory Commission. Not only did we build the first two in Japan the exact same design basis that is certified in the United States, right? We built it in less than 40 months of nuclear construction for the first of a kind unit, and that first of a kind unit was 40% cheaper than reactors that were being built at the exact same time in Japan, that were the generation two technology, that is, the boiling water reactor fired in the Mach two containment, 40%


00:09:33:12 - 00:09:54:02

James Krellenstein

cheaper for the first of a kind versus the seventh BWR five in the Mach two containment. So my question is if it's so if I'm am I just so stupid that I don't see the empirical support for it? Let me ask you the question here. We have an example where we build we have a US builder has many, many problems.


00:09:54:05 - 00:10:18:10

James Krellenstein

And they have that exact same problem in a world with the NRC. Without the NRC, we have a design certification process that works, that we go through, that we not just give it for the for the system. 80 plus from Combustion Engineering for the, yes, BWR for the Apr 1400 that everyone holds up as this ideal, you know, that the South Koreans built at Baraka and at Corey.


00:10:18:15 - 00:10:44:14

James Krellenstein

Right. Which is generally held as a very buildable reactor, although we could have some arguments about that. That also has an NRC design certification, the Ap1000, the new scale, Voyager, the US 600, the ESB, all all of these reactors have design certifications and we've in some cases built them faster and cheaper than the previous second generation, legacy nuclear power plants.


00:10:44:16 - 00:11:00:10

James Krellenstein

And we know from a just a sheer design basis that these are dramatically simpler. So my question back to you is if this is such an obvious question, explain to me the very basic empirical reality that has happened in nuclear construction after the last 30 years.


00:11:00:12 - 00:11:18:10

Chris Keefer

That's a pretty compelling counterargument. I think we can just finish the episode right now. James. no. But, you know, that's that's a fascinating story. And I think a lot of people are aware of that Abwehr build. well, I should say a lot of people, people within the nuclear advocacy community. And it's certainly celebrated. But, I think you brought some interesting context there.


00:11:18:10 - 00:11:33:07

Chris Keefer

And just what a kind of compare contrast, again, a first of a kind coming in, so affordably, so quickly, you know, making that comparison with the AP 1000 experience in the US and China. And I know the Abwehr is being considered, in the US to be built, I think in the South Texas project. You're telling me it was.


00:11:33:07 - 00:11:37:14

James Krellenstein

Licensed, full combat combined operating license was issued for units, three units.


00:11:37:19 - 00:11:44:17

Chris Keefer

And there was a there was a Finnish reference plant available. I mean, why didn't they just build yours? I know this isn't the focus of today's discussion, but.


00:11:44:17 - 00:12:10:23

James Krellenstein

I some other discussion. But I think it we'll we'll touch back on this. But it's not just that I think what's weird about this is let's go back historically. So we've talked a little bit about good nuclear design, nuclear construction, which is rare but happened, and bad nuclear construction, which is unfortunately in the United States, was especially in the 70s and 80s, was the dominant case, the majority.


00:12:11:01 - 00:12:47:16

James Krellenstein

So let's talk about one of those successful ones, Palo verde. Right. The reason why I bring up palo verde is units one through three is because that represents not only a very advanced generation two reactor design, but the first standardized, arguably reactor licensing process that we had in the United States. You know, there's a common myth that is repeated ad nauseum at the US Nuclear Regulatory Commission has never before the Ap1000 were built and finished at Vogel, had never actually issued like a license or design certificate to design that was actually built.


00:12:47:18 - 00:13:09:22

James Krellenstein

That's not true, right? The Combustion Engineering System 80 design that was first built at Palo Verde and was also being built at Washington Nuclear Project three and five, as well as at Yellowknife by Texas. Tennessee Valley Authority actually arguably was the first time that the NRC fully standardized. It was under what was called ten part 50, appendix. Oh, right.


00:13:09:22 - 00:13:34:06

James Krellenstein

The what's it now with an equivalent of a final design approval that was actually done fully by the NRC. Right. The first it was a two step licensing process. And we we should go back and talk a little bit about how the NRC licensing processes have worked. but because it's really important to understand. But that was actually given we called preliminary and final design approval all by the NRC.


00:13:34:08 - 00:14:01:20

James Krellenstein

The AEC did not exist for either one of these. And the construction permit for Palo Verde referenced a standardized design. It was something called the CSR or Caesar, depending on how you pronounce it. E sa combustion engineered Standardized Safety Analysis report, and what that provided was a reference under the ten part 50 appendix. Oh, if you know your regulations, in 2007 this was moved over to ten part 52, subpart E.


00:14:01:22 - 00:14:25:14

James Krellenstein

this is a final design approval process that allowed us to standardize the nuclear island and the nuclear steam supply system such that we would have to repeat the design and engineering each time. And the regulatory process. So that was issued by the NRC and Palo Verde went through it. And then one of those successful builds, arguably in the US, not only because of a first because it took didn't take actually that short.


00:14:25:14 - 00:14:49:20

James Krellenstein

It was ten years still, but we saw a very dramatic over 40% price drop in overnight costs between unit number one and unit number two, right. And we we had a very challenging, you know, sort of environment to do this in. We had to obviously post Three Mile Island, the plant was in the middle of the build. and it was a very and a very challenging no waterways, no logistics, no cooling resources.


00:14:50:02 - 00:15:22:01

James Krellenstein

And this was a massive plant was then and just until Vogel three and Vogel four came on line was the most large nuclear power plant in the United States. And and the idea that we did not see, even at the beginnings of the appendix O processes and the appendix N processes, which then paved the way for part 52, that we were starting to see some real systematic ways that we're going to start addressing the major ways that we that nuclear regulation was contributing to cost overruns.


00:15:22:03 - 00:16:06:18

James Krellenstein

it's just not true. We saw that process happen over 40 years. And the problem is, is that after Sharon Harris, unit number one was construction permit was issued in 1978, we did not start construction on a new nuclear power plant until Vogel and Summer happened, in, you know, 2012 for nuclear construction. These and so the sort of regulatory changes that have occurred that arguably addressed a lot of the systematic challenges that were faced in nuclear regulation, we just really haven't had a very large sample size to be able to, to really elucidate how those new processes are going to work and, and improve potentially the, the, the ways that regulations get into,


00:16:06:20 - 00:16:08:21

James Krellenstein

into had got us into trouble.


00:16:08:21 - 00:16:27:04

Chris Keefer

So, so at the outset, you mentioned that, you know, the article felt like Ted's article felt like it was written in 1983. let's talk a little bit about that era. you know, framing I've heard as well from recognizes that the new coal industry stopped building new plants and just sort of became a self licking ice cream cone, installing endless upgrades and safety systems.


00:16:27:04 - 00:16:44:14

Chris Keefer

So there's this era right after Three Mile Island, in which I guess it's under part 50, where, you know, the operating license, is not given upfront or, you know, you don't get full approval right from the get go was do with part 52. And so with some of the lessons learned through Mile Island, there's just, just a plethora of new regulations that come along.


00:16:44:14 - 00:16:46:23

Chris Keefer

And these six bloom.


00:16:47:00 - 00:17:11:14

James Krellenstein

let's before we go into how part 50 versus 52 works, let's let's pause here. And just also note one other thing we have in the nuclear regulatory Commission's is that, like when there's no reactor construction, it just goes home. You know, it's main job historically, especially before we were building plants, was at least in the nuclear energy industry, was regulating and maintaining the existing fleet.


00:17:11:19 - 00:17:38:18

James Krellenstein

Right. And as you know, each nuclear power plant has two resident inspectors who are employees of the NRC, who actually be stationed at the power plant, who are constantly inspecting, and each plant gets hundreds, if not more of hours of inspection, each year and has to continuously maintain, you know, safety standards coincident with, what the NRC regulations require.


00:17:38:20 - 00:18:06:12

James Krellenstein

And remember, yes, it's true that we stopped building plants during this time, but whatever happened between the industry and the NRC, the NRC did not get in the way of the US nuclear fleet becoming the highest capacity factor large fleet in the US. And we did not see, with some notable exceptions. And we'll talk about them, that many shutdowns of plants being ordered rather than retrofits being required.


00:18:06:12 - 00:18:44:02

James Krellenstein

And I think that there are some really good examples, industry led, government enabled regulation, particularly post Fukushima Daiichi, right, where we saw that the US nuclear industry really put together a pretty comprehensive response called flex to what happened. at, you know, Fukushima Daiichi and the NRC sort of went along with that framework and allowed the retrofitting to happen in a way that did not overly interrupt, the, operations of existing power plants and I think materially dealt with the safety concerns of that accident happen.


00:18:44:02 - 00:19:09:21

James Krellenstein

So I, I just want to remember, the NRC is also running and maintaining the licenses for operating plants and also doing obviously all of the license renewals for 20 years after that 40 year requirement. And then the supplemental, the sellers, the supplemental license renewals as we push from 60 to 80 years, of license activities. Now, these are all places where there's a lot of improvement.


00:19:09:23 - 00:19:30:22

James Krellenstein

And I don't want it to sound like I'm saying that there's nothing that needs to change at the NRC. Nothing could be further from the truth. What I'm saying is, if we're looking for the root cause of what is going on in the US nuclear industry, particularly on the Newbuild side, I think we have way overstated the case for it's the nuclear regulation, stupid.


00:19:31:04 - 00:20:01:21

James Krellenstein

And we're going to pass law after law after law and not actually deal with some of the fundamental issues that we have in this industry. So let's go back to what you brought up earlier. this part 50 versus part 52 challenge. So historically, after the Atomic Energy Act of 1954 passed, and we started first getting the first civilian nuclear power plants, as you mentioned, we utilized a two step licensing process called ten part 50.


00:20:01:23 - 00:20:24:18

James Krellenstein

And with that licensing process means is, is that what you do is you go to the US Nuclear Regulatory Commission and initially you file what's called a preliminary safety analysis report among many other things, that sort of analyzes preliminarily what the safety consequences of this nuclear power plant that you're going to build and how it's going to comply with the NRC regulations.


00:20:24:20 - 00:20:50:22

James Krellenstein

And if and then the NRC not only, convenes its own hearing, convenes public hearings, chaired by the Atomic Safety and Licensing Board in the communities that the plant is going to be built in. Also, there's a second committee that is, convened called the Advisory Committee on Reactor Safeguards that is sort of separate to the NRC, that sort of independently or the AEC at this time independently reviewing that construction permit application.


00:20:51:00 - 00:21:17:07

James Krellenstein

and if all of them sort of go through all of those, those processes, they issue a construction permit. Now, there's a lot of opportunity here, by the way, for interveners like anti-nuclear activists to come in and either sue the AEC or NRC, on these processes before they can issue the construction permit application. as well as actually intervene in some of these proceedings themselves, like the SLB.


00:21:17:09 - 00:21:49:05

James Krellenstein

and then that construction permit authorizes the plant licensee, as well as its contracting partners and its reactor vendor partners, to actually construct the nuclear island, including safety critical components of the nuclear power plant. as we're sort of building that plant generally. And there's actually there's an old probabilistic method that you use to actually know when you would sort of, apply, you sort of apply for a second step under the construction permit.


00:21:49:07 - 00:22:12:01

James Krellenstein

And this is a very top level, sort of I'm not going into all the details, only because it's very hard and, a lot of people, I don't know why, surprisingly, find it boring. but, I'm trying to get a tad entertaining for your audience. So that construction permit process is only a construction. It's only a permit to construct.


00:22:12:03 - 00:22:45:08

James Krellenstein

It is not a license. It does not allow the reactor. licensee, which is not a licensee at this point, to actually load fuel or bring the reactor critical or commence power operations. It is rather just a permission to build the, nuclear island, including all the components necessary. But in order for you to get that permission to, actually load fuel, to let the reactor go critical and so on.


00:22:45:10 - 00:23:20:16

James Krellenstein

you have to, as you noted, apply for a second sort of process called an operating license. And this involves you rather than having your preliminary safety analysis report you combining everything that you've learned from the construction of the plant, your final safety analysis report, which reflects how the plant actually was constructed, and then the NRC again and the Akers again and the SLB again, have to go through a whole nother process to grant you an operating license for the reactor to turn off, for you to have permission to operate the reactor.


00:23:20:18 - 00:23:47:01

James Krellenstein

Now, there's three major challenges that this immediately poses. The first is you may not get the operating license. And that has happened, for example, I was just, as you know, on Long Island enjoying a very nice beach vacation. And, I on Long Island there is Shoreham Nuclear Power Plant, which got a construction permit, finished all of their nuclear construction.


00:23:47:03 - 00:24:10:22

James Krellenstein

But the the utility at the time line lighting coal Company filed a final safety analysis report, went through all of the processes, actually got the NRC to agree with everything. But part of the issue was historically, in order to get an operating license, you needed to have the local communities agree that you could actually evacuate with the evacuation plans.


00:24:11:00 - 00:24:41:06

James Krellenstein

And because the state of New York and Suffolk County, the county that it's in, never agreed to the licensing plans, mainly because of steadfast anti-nuclear opposition, the NRC at that point was unable to basically issue an operating license to that plant and all that, though the NRC tried to force that along with a little Co by giving them permission to do low power physics testing, i.e. bringing the reactor to about 5% power so that you'd have to, you know, decommission the plant because you actually made it radioactive.


00:24:41:11 - 00:25:15:14

James Krellenstein

They never actually were able to get the operating license issue because there they're such steadfast interveners from not only the anti-nuclear community, but also from the state of New York that was trying to really hard prevent the issuance of this. Well, so I never got permission to operate. And $6.5 billion now more than $14 billion in inflation adjusted term, that every ratepayer in Long Island, you know, much of my, my childhood paying that that bill for a power plant that was just never issued that operating license to turn off.


00:25:15:16 - 00:25:36:00

James Krellenstein

So that's a big problem, I would say. And this happened not only at Shoreham, of course, but in Seabrook in New Hampshire, where we had communities in Massachusetts and in New Hampshire say that they couldn't evacuate. And they finally had the NRC basically overrun the communities and basically just force an operating license to be issued because they didn't want to have a basically a repeat.


00:25:36:02 - 00:25:57:10

James Krellenstein

What occurred, at Shoreham. So that's the first thing that can happen. But those are kind of the edge cases and really bad. We don't want that to happen. Obviously, that's a really bad situation that doesn't really make sense. If you're going to build the plant and you build the plant up to regulations and demonstrate that you can build up to regulations, you should be allowed to operate.


00:25:57:12 - 00:26:22:16

James Krellenstein

The second challenge that we had in this two step process is the reactor design is not necessarily done when you're starting that construction. Now, on the one hand, that gives you a lot of flexibility when you're building the plant, right? It allows you to, you know, if you have some challenges that you've identified as you're constructing the plants to not necessarily be locked in to a single design.


00:26:22:18 - 00:26:51:01

James Krellenstein

But on the other hand, if the Nuclear Regulatory Commission begins or the Atomic Energy Commission, this proceeds way before the NRC was created in February of 1975. If you have the, NRC changing the safety regulations, there is no guarantee that the NRC is going to utilize the regulatory basis and standards and regulatory codes that were in place when the construction permit position.


00:26:51:03 - 00:27:18:18

James Krellenstein

So what you get happening is as the NRC changes their regulatory basis, right. Let's ask this basic question. you're in the middle of construction. You may even finished the construction of this plant. what happens all of a sudden if a new regulation comes in and that already built system is not compliant, no longer compliant with this new regulation issued by the NRC.


00:27:18:20 - 00:27:37:14

James Krellenstein

Well, you don't have an operating license. So there's a very good chance when you go ahead and submit your FSR to the NRC. The NRC is going to say this system is not compliant with the regulations and so, well, it's compliant with the regulations. When I filed a construction permit, they're going to say, great, but you're not. You're filing for an operating license.


00:27:37:14 - 00:28:07:04

James Krellenstein

Now and it's no longer compliant with code. And so you have to rebuild. And this back fitting phenomenon was a really, really large, driver of cost overruns, a massive cost driver. And as Ted correctly points out, especially after Three Mile Island, where so much of the regulatory code started to change, this really became a systematic problem in a really, really serious way, where we had, in some cases, hundreds of millions of dollars.


00:28:07:04 - 00:28:45:00

James Krellenstein

If not billions of dollars of rework being required. And then the third thing, or back fitting to to basically meet regulations that did not exist when the construction plans were initially dropped. And then the third thing that you had initially was a very big problem is you had this general perspective of how are these regular ations that are basically in ten part, as you know, ten part, 50, 51 for the environmental codes, you have obviously other sections of title ten that you're sort of dealing with, but that doesn't really tell you how you're going to build a power plant.


00:28:45:02 - 00:29:10:01

James Krellenstein

Right. and because not necessarily there wasn't necessarily a fully like license design, there's a lot of guesswork, especially initially when the amount of regulatory guidance was initially done to actually see if the design that you were going to build was compliant with what the NRC would interpret the regulations. And so you had this sort of three major issues.


00:29:10:01 - 00:29:48:06

James Krellenstein

The first is that there could be intervener and other intervention in the operating license process that prevents you, even if you built the plant, demonstrated compliance from getting an operating license. So you're screwed. That plant can't, you know, make any power if it can't turn on and can't do commercial operations. The second thing is, is that you have no surety under part 50 that the NRC is not going to change the regulations need construction, and that you're not going to literally have to go back and destroy systems that had already been built to comply with regulations that literally did not exist with construction started.


00:29:48:08 - 00:30:13:23

James Krellenstein

And then the final third thing is, it's not necessarily clear because there was not really a design approval or certification process for much of this time that they that the designs that you were doing actually are compliant. The NRC may have a different or the AEC may have a different interpretation of the regulatory code than you do, and you don't necessarily find that out until you're submitting your FSR for the operating license.


00:30:14:00 - 00:30:22:18

Chris Keefer

Right? Right. So as you're saying, in the 1980s, perhaps, it was the regulation stupid. so what changes? And you're in your view.


00:30:22:20 - 00:31:08:08

James Krellenstein

So even before we start reorganizing completely, the, this two step process, we begin starting in 1972 at the Atomic Energy Commission. They recognize that this process of going plant by plant by plant, having each power plant sort of a custom, safety analysis report, and a custom regulatory process is not a great idea. at the end of the day, if you're standardizing some aspects of the plant systems, we should have a way from a regulatory process to have that reflected and not repeat the same system over and over again, and have it evaluated and maybe come up, by the way, with different results depending on which reviewer and regulator was reviewing


00:31:08:10 - 00:31:36:12

James Krellenstein

that application at the time. So we begin seeing opening up processes for standardization that happen. And there's three major avenues that happen before we totally rewrite the code. One is called appendix M, where we allow us to basically apply seven, you know, multiple different reactor construction purpose and operating licenses can be submitted together at multiple sites because they're going to basically use a single design basis.


00:31:36:12 - 00:32:02:17

James Krellenstein

It's identical from plant to plant to plant. And this happened of course, at Wolf Creek and Callaway using SNPs or standardized nuclear unit power plant systems, which is a Westinghouse for loop nuclear steam supply system, a General Electric turbine, and using that tool as the ultimate architect and engineering constructor. And that was done through a unified common licensee process where all the plants are initially six.


00:32:02:19 - 00:32:35:11

James Krellenstein

We only had a building two that actually started on this construction permit operating license process. That's the first thing. And the second thing that happens is the NRC recognizes that at some point, the nuclear island vendors are going to want to license and standardize the design. So they introduced ten part 50 appendix up, which allows us to do design approvals, which basically says if you're combustion engineering, you can basically make a standardized safety analysis report that can be for that, you can reference in your construction permit application your PCR.


00:32:35:13 - 00:32:53:16

James Krellenstein

That's what we call a preliminary design approval or PDA. And then you also can do a final design approval and a standardized safety analysis report for your FSR for that design. And not only are we going to do it for the nuclear island, we're also going to allow the balance of plant vendors to be able to, actually do that.


00:32:53:17 - 00:33:14:15

James Krellenstein

So the architects, an engineer, a of Stone and Webster, they all began actually standardizing their balance of plants and the interfaces it had with a standardized nuclear island to be able to do, to be able to actually use the same process over and over again, such that when you applied your PCR and your FSR, it would just reference the standardized design approvals.


00:33:14:15 - 00:33:41:06

James Krellenstein

Are preliminary design approvals for that under appendix Out, which you can see in the Palo Verde PCR and FSR. For example, it's referencing a Combustion Engineering standardized safety analysis report. So that's the first thing that happened. But that didn't really solve the ratcheting problem. It is still possible if you still have this two step licensing process, you can absolutely still in the middle of the reactor design, in the middle of reactor construction, change the plan.


00:33:41:08 - 00:34:10:19

James Krellenstein

I changed the regulations that could require back fitting. That has already happened. Okay. So what they began doing is realizing we need to actually fix this completely. And it's just a one step process. What does that mean? It means that before you start nuclear construction, you get an operating license as well. And as part of that operating license or coal combined construction and operating license, you have a set of tests that are part of your license.


00:34:11:01 - 00:34:36:03

James Krellenstein

They're actually called a tax inspection test, analyzes and acceptance criterion or criteria that literally say, this is how you are going to demonstrate that you built the power plant in conformance with what your license had. And if you have demonstrated and closed out that itac process, we will literally say you now have there's no more hearing that has to happen.


00:34:36:03 - 00:35:10:11

James Krellenstein

There's a mandatory hearing. There is one possibility for an intervenor to be able to intervene there, if only they have on its face or prima facia, obvious evidence that in fact I tax have not been met and close. But if that hasn't happened literally, the NRC just issues a quick letter. And this happens in the case of local three and less than a week from when they notified that the eye tax were close, they say, okay, you can go ahead and begin loading fuel, pre-clinical pre-clinical testing and bring the plant operation.


00:35:10:13 - 00:35:35:03

James Krellenstein

So we basically do this. We take that entire two step process and say, no, give me the operating license. Before you start nuclear construction, you need to demonstrate that your design is done, that you're building the plant in conformance with the design. And there we go. So we really eliminate the back fitting problems, and we also eliminate the opportunities for motivated anti-nuclear activists to get to stop operation of the plant.


00:35:35:03 - 00:35:57:03

James Krellenstein

Once the plant is constructed. That doesn't necessarily mean there's not a huge opportunity for intervention before the coalition issue. There is. There's still work that needs to be done there, but once at least, we started nuclear construction, right? And we can demonstrate that the plant is built in conformance with the design basis and licensing basis. We can, operate the plant.


00:35:57:05 - 00:36:16:06

Chris Keefer

So, you know, I think another major concern, voiced in the, in the advocacy community, again, is, you know, there's just way too much regulation. and I think, again, people really refer back to that cost plume. I wish I had it here and I could put it up on the screen. But again, that you had, you know, a pretty flat, overnight construction costs throughout the 60s.


00:36:16:06 - 00:36:20:09

Chris Keefer

And then it just goes nuts with plants that were again under construction or started construction.


00:36:20:10 - 00:36:25:02

James Krellenstein

Well, the grid starts going way up in the 1970s, way before this. Right.


00:36:25:04 - 00:36:43:01

Chris Keefer

But but I think, you know, through looking at this graph, a lot of people say, hey, like, why aren't we just building the reactors that were affordable to build back in the 60s before? And to add on all this extra, you know, safety bells and whistles, nuclear safe and off. Look, when a meltdown happens, it doesn't even, you know, as as proven with Fukushima, it's not really resulting in any significant public health impacts.


00:36:43:03 - 00:37:00:09

Chris Keefer

maybe we'll just kind of step back from from the detail here and just talk theoretically about, you know, I like I like making aviation comparisons. when I talk about nuclear safety, personally, when I talk about waste and things like that, maybe this is a point to just talk more, kind of, you know, first principles of, of, you know, maybe.


00:37:00:11 - 00:37:34:11

James Krellenstein

This is one of the things I don't really understand and maybe I'm missing something here. I, you know, I, I've been to a lot of generation two nuclear power plants, and I know the design of generation three nuclear power plants pretty well. I have never really seen that those designs were simpler or more constructible. Right. If we look at what a generation two pressurized water reactor is, let's talk about like, a big 1200 megawatt one that was built like like Indian Point, unit number two, unit number three, that was built.


00:37:34:11 - 00:37:54:23

James Krellenstein

And, you know, came on line in the early 70s. Right. And very affordable. Right. And then I look at that as an AP 1000. If you were to believe what the many of the nuclear advocates say, right. At least in the US, you would expect the AP 1000 is a much bigger plant. It has, you know, double or triple containments.


00:37:55:01 - 00:38:17:14

James Krellenstein

It has so many more pumps, so many more safety systems. Look the exact opposite, right. If we compare that to like Indian point unit number two where we're going to find 50% safety, fewer safety related valves, we're going to find 35% fewer pumps, 80% 80% less safety related piping. That means that we have so many less welds to inspect.


00:38:17:20 - 00:38:39:01

James Krellenstein

As an example, we have so much less of the new one process and ten part 50 AP that's a chock every single pipe as we go through the process, with 85% less control cable and just the total amount of building volume that is built to withstand an earthquake, which, as we've talked about many times before, is actually one of the major drivers.


00:38:39:03 - 00:39:03:19

James Krellenstein

That's 45% less than that Indian point, unit number two. So my question is great. We could go and try to build another large pressurized water reactor of generation two style. In fact, we could try to build another Palo Verde style reactor. We had the SSR P and F were on the books for years, right. That could allowed. In fact, many people got construction permits.


00:39:03:21 - 00:39:31:14

James Krellenstein

Woops. Got construction purpose. Yellowknife got construction permits. We could build the sort of slightly improved version of that system 80 plus. They got a design certification right after the A did in 1997. I don't understand how making a bigger, more complex reactor, and with a lot more safety systems really is going to be cheaper or easier to build than a possibly safe, dramatically simpler reactor.


00:39:31:19 - 00:40:02:15

James Krellenstein

What is good? I mean, obviously it's the nuclear regulation, stupid. So I'm missing a really obvious thing here, but the whole point that I think people forget is that the industry has moved to passive safety in the 70s and 80s and 90s, was motivated primarily by cost concerns. Right. If you think about what, say, an emergency car cooling system requires in an active safety plan, it requires, at a minimum for active safety system, for a pressurized water reactor or boiling water reactor.


00:40:02:17 - 00:40:26:18

James Krellenstein

Right. It requires a high pressure safety injection system or coolant injection or core spray system. Right. That's because we have certain scenarios where we have or hold small brake loss of coolant accidents. Those are accidents where a valve gets stuck open, like what happened that three mile, which means that the reactor coolant system is still at a at very, very high pressure.


00:40:26:23 - 00:40:48:18

James Krellenstein

But we're losing reactor coolant inventory, and we have to be able to pump new inventory into the system right at a very, very high pressure. But on the other hand, if you have a very large brake and you depressurized the reactor coolant system, all of a sudden you need to be able to put a huge amount of volume in at a much lower pressure as a whole.


00:40:48:18 - 00:41:11:08

James Krellenstein

Separate system, your low pressure systems, low pressure coolant injection, low pressure safety injection. Lipsey. or low pressure core sprayer, low pressure core flutter. And for each one of those right, you need to have your own independent sets of instrumentation and control. And it allows the operator to see how those systems are running right. You need to have your own emergency diesel generators.


00:41:11:08 - 00:41:46:12

James Krellenstein

Class one E right. They're able to withstand an accident and can spin up and load, the, you know, serve the loads of those hips. Your lipsey systems in a time before you start damaging fuel. and you need to have that then redundant a couple of times because ultimately pumps sometimes break and don't work if you have to have one or 2 or 3, or in the case of the EPR, for different backups or three different backups in one primary system, or to for independent safety trains or two independent safety trains at a minimum.


00:41:46:16 - 00:42:09:09

James Krellenstein

To do this, compare that to what we do in, in on, in Ap1000 which is utilize natural convection natural forces to be able to withstand large the largest credible break loss of coolant accident without any emergency diesel generators with systems that are reliable enough that we don't need as much redundancy as we do with these active safety systems.


00:42:09:12 - 00:42:43:15

James Krellenstein

We don't need these independent systems for engineering safety feature actuation. although we do obviously have, you know, quadruple redundancy and in the protection and monitoring system and then also a diversity system as well. this dramatically reduces the costs that it takes to try to build a nuclear power plant. On the other hand, remember that the actual components that are part of a nuclear power plant, only about 20% of the overnight cost of the nuclear steam supply system, that is the reactor pressure vessel, the pumps, the pipes, and so on.


00:42:43:17 - 00:43:17:18

James Krellenstein

As we've talked about before, so much of the cost of a nuclear power plant are driven by the building volumes, by the balance of plant systems. So being able to shrink the volume that we're being able to build by making it passively safe also saves a huge amount of money and not having to have also, by the way, safeguard buildings for each one of those safety systems is a huge amount of safety savings, not just in the actual component costs, but literally allowing us to shrink down the size, the safety of a safety, related nuclear island.


00:43:17:23 - 00:43:27:09

Chris Keefer

So, you know, it was interesting. Ted, did, reference to EPR? you just mentioned it there. is it the regulation stupid over in Europe?


00:43:27:11 - 00:43:28:11

James Krellenstein

where I think one of.


00:43:28:11 - 00:43:30:02

Chris Keefer

The supervision design philosophy.


00:43:30:06 - 00:43:53:17

James Krellenstein

That I thought was a kind of a tell is he's going in, that Ted is going on about the NRC all the time, and he chooses the one, light water reactor design that's being built in Western countries that does not have an NRC design certification. we're arguing the NRC had the least amount of involvement in that, although there was an application for NRC design certification that was abandoned, by parameter.


00:43:53:19 - 00:44:27:00

James Krellenstein

So one of the things that I think people sometimes get confused is, there was actually a huge regulatory split between the United States and Europe. that's still a process. And a lot of Western European countries simply don't credit passive safety systems in the same way that the NRC does. They have far more prescriptive regulations that really preclude the, the, sort of simplistic the simplifying systems that you can demonstrate or not only cheaper but dramatically safer.


00:44:27:02 - 00:44:47:17

James Krellenstein

So you don't need these redundancies. So, for example, an AP 1000 does not have a secondary containment. It just has a primary containment that you can demonstrate pretty analytically has a very low probability of failure. So you you basically are trading rather than having this backup system. You're saying, no, I'm going to actually make it just much more reliable and simplify the system.


00:44:47:19 - 00:45:11:14

James Krellenstein

There's a lot of, European regulators who have historically not credited such an idea, which is why, of course, the EPR has two containments, right? It has a primary and secondary containment system, of which are are designed completely redundantly right, to be able to withstand a total failure of the primary containment system. But they also did not credit these, these, passive safety systems at all.


00:45:11:16 - 00:45:30:15

James Krellenstein

So what the EPR has, really catering at this point to German regulations, actually, remember the EPR was initially a Siemens Kraftwerk Union Crematorium joint project. and given that it was going to be having to be built in Germany, the Germans had not done the same amount of research that the NRC had done through the 80s and 90s.


00:45:30:17 - 00:45:59:07

James Krellenstein

to, to understand how to analyze passive safety systems. They instead basically made for a 100% capacity fully redundant safety systems. Each one has a high pressure system, a low pressure system, its own instrumentation and control system that may even have its own redundancies and actuation logic for each safety train. It has its own diesel generator for each one of those safety trains.


00:45:59:12 - 00:46:29:09

James Krellenstein

For three of them, at least, there's one that has a common chair, one set that has a common chair. But there's three independent what are called safeguard buildings that are each independent buildings that house each one of these high pressure and low pressure safety injection systems and have their own independent water sources that are also protected. So, for example, for RCS, for reactor coolant system inventory make up, they have their own protected RCA inventory makeup sources at each one.


00:46:29:11 - 00:46:52:02

James Krellenstein

So here we have a two example. This is what what real regulatory inflexibility and prescriptive regulation looks like. If you're forcing a design that just says, no, we can't innovate at all, the only way you're going to do it is make more backups. After backups after backups, you get something like the EPR, which really is a very, very difficult plant to build.


00:46:52:04 - 00:46:59:06

James Krellenstein

and has 7 to 10 times more concrete than a, an Ap1000.


00:46:59:12 - 00:47:17:16

Chris Keefer

So in terms of just the sheer volume of regulations that piled up, you know, many analysts have looked at that and sort of correlated that to, construction delays. I think you have a different, different take on that. Again, in terms of this process of arriving, I think your thesis is lots of problems in the 80s, but we're not stuck in the 80s anymore.


00:47:17:16 - 00:47:27:00

Chris Keefer

Things have changed a lot. We've actually got better reactors now because of the regulations that that piled up. This is you know, I haven't heard of this anywhere else. GM so, if you can.


00:47:27:02 - 00:47:57:17

James Krellenstein

So, clarify. Well, so let's take another you know, we've been responding to Ted's piece and, you know, overall, you know, it's actually a pretty compelling piece. I recommend everyone read it. It's not it's certainly a well-articulated and and, and provocative, albeit predictable position. take, take, another famous guy who I always hear brought up, nuclear advocate Jack Devaney, who gives this example of something called the leak, called the double Duty and break.


00:47:57:19 - 00:48:21:09

James Krellenstein

Okay. So the double team break rule was this rule that was put in the 90s, in the early 70s when the Nuclear Regulatory Commission actually atomic Energy Commission at this point required that nuclear power plants have these emergency core cooling systems, these abilities to be able to demonstrate that they're able to withstand a design basis, loss of coolant accident.


00:48:21:11 - 00:48:39:02

James Krellenstein

And the question is, what is the design basis, loss of cool and accident? And obviously you can think about, well, maybe a pipe breaks for example. And that you have, you know, a pressurized water reactors at thousands of pounds per square inch. The whole system is pressurized. So if you break integrity of that system, you have a little break.


00:48:39:04 - 00:49:07:12

James Krellenstein

A lot of coolant is going to get out and you can uncover the core damage cladding and maybe release fission products as a result of that. So the question, though was, well, how big of an X do I have to build? What is the sort of capacity that it needs to build that? So the initial rule that they couldn't place was something called the Dublin guillotine break, saying, we actually don't know the answer to this question of how pipes are going to leak before they break.


00:49:07:14 - 00:49:28:02

James Krellenstein

So instead we're going to create a bounding analysis. Right. What is a bounding analysis mean? It means literally we're going to take the maximum break size that can occur and basically design the systems so that they can respond to that. That's what what does and how do you let's think about this. How would you say what's the maximum break that occurs.


00:49:28:02 - 00:49:55:11

James Krellenstein

So imagine we took the largest diameter pipe in the reactor coolant system. And we like take a guillotine and we split it open. Right. We literally just slice it open and we slice it open on both sides. Right? So that, like, even some of the water could jet through, literally, it's all broken. You just take out a slice of that pipe and now it's vented to the containment atmosphere.


00:49:55:13 - 00:50:26:19

James Krellenstein

That is a huge leak of the reactor coolant system. And not only that, as Jack correctly points out, the NRC and Arc first, but then the NRC said, well, you have to actually now not only calculate, you know, building pumps that can be able to recover volume to this, this maximum size break, but also that water that's shedding out at thousands of pounds per square inch, that's going to cause dynamic effects, you know, pursuant to general just general design criterion four.


00:50:26:20 - 00:50:49:05

James Krellenstein

Right. We have to also calculate the dynamic effects of that pipe whipping around. Right. That pipe is not going to just stay there as, literally wall of water at thousands of psi flashes into steam as it comes out. And we have to say that, you know, this jet of water, what if it hits a, a, you know, a motor control center for a safety critical load?


00:50:49:09 - 00:51:18:04

James Krellenstein

How is that going to impact things? so how is it going to impact the internals of the reactor pressure vessel, like the core, like the sort of core barrel and core internals, as if you have suddenly a massive asset metric load on the different internals. Maybe that will cause literally the fuel assemblies to move in such a way that you can't, you know, over can get your control rods into the fuel assemblies.


00:51:18:06 - 00:51:50:08

James Krellenstein

So all of a sudden this simplified bounding analysis becomes a very, very complex set of engineering and regulatory requirements that you have to design your plant and certain plants. For instance, Indian Point Unit number one could not comply with these new x x rules, not units two and units three. It just recently got shut down. But in 1973 I believe the Atomic Energy Commission ordered that India Point, unit number one, using the back fit rule, shut down because it could not comply with the new X rules.


00:51:50:13 - 00:52:17:05

James Krellenstein

And for many other power plants, this required they even were operating required very, very expensive back fits and retrofits to be able to comply with. But people correctly pointed out once again that that's how pipes break, right? Wouldn't it be great if there's a leak before break rule? And I think we all can agree on that. And so I'm reading Why Nuclear Power is a flop in 2022.


00:52:17:07 - 00:52:50:08

James Krellenstein

And, in my, my bedroom trying not to, well, unsuccessfully not to throw it against the wall. And literally you had him talk as if the leak before break rule. If if we're still just doing double and getting breaks as if it's 1972, like, you know, as if people are still wearing bellbottoms and, you know, flower power and, you know, therefore, pintos driving everywhere except we've been we've we've studied this problem and there's literally an entire section of the standard review plan.


00:52:50:08 - 00:53:20:12

James Krellenstein

It's in 3.6 that is literally like week before break analysis. Right. And Ap1000, EPR, the EPR before it got canceled, the design certification, multiple different design licenses have leaked before break rather than double into getting break based analysis. Now, why did it take from 72 to 83? Actually, when, this was issued? Well, it turns out it's really, really complicated to figure out how pipes are going to break and what the credible maximum break size is.


00:53:20:18 - 00:53:54:16

James Krellenstein

Right. So you couldn't just do the initial regular regulation, which is a couple sentences, which is literally like, you know, basically the power plant needs to be able to withstand the largest pipe in the reactor coolant system having a double in to guillotine break. No. Now we need thousands of pages of regulatory guidance to figure out how we are going to do an analysis that the NRC will accept that allows us to do a leak before break analysis, rather than go to the maximum bounding sort of condition of the double guillotine break.


00:53:54:18 - 00:54:16:03

James Krellenstein

But here's the problem. If you have this perspective that the way that you should be viewing regulations, it's the number of pages that you do. A performance based regulation is always going to be far more complex and, lengthy then a, a prescriptive, simple regulation. That's a couple cents.


00:54:16:05 - 00:54:24:02

Chris Keefer

And so if you 1000 then gets around this by not having an active emergency core cooling system, not having to worry about guillotine breaks.


00:54:24:04 - 00:54:54:10

James Krellenstein

Well, no it does. It has to it has to it has to be it has an emergency core cooling system. It is all based on depressurized the reactor coolant system utilizing and eventually convection, condensation and gravity based cooling to basically maintain RCS inventory. What it doesn't have to do is withstand a double ender guillotine break. It's a little complicated here, and I being for certain piping, it does not have to qualify for double entry guillotine breaking.


00:54:54:12 - 00:55:16:10

James Krellenstein

and it's been able to demonstrate using a methodology that was first elucidated in 1061 of saying, well, we can actually use a leak before break analysis to demonstrate that the maximum credible break size is going to be smaller than a double into guillotine breaks is for certain systems. Now, I should be clear. Well, there's hundreds of existing.


00:55:16:12 - 00:55:37:23

James Krellenstein

there's dozens and dozens of pipes and even existing orders that have been able to use leak before break breaks. We have not been able to use it. boiling water reactors because of, a failure phenomenon enters into granular stress corrosion cracking, which basically made the NRC think that actually it would be possible that you could actually break a weld completely open and have a guillotine break.


00:55:38:05 - 00:55:52:16

James Krellenstein

But for parts like the AP 1000, you have been able to utilize, a leak before break analysis. Same thing with the EPR. By the way, the EPR uses a leak for break analysis for for crediting its dynamic effects and also sizing the emergency cocoon system.


00:55:52:21 - 00:55:54:00

Chris Keefer

Well, I mean.


00:55:54:02 - 00:55:54:12

James Krellenstein

I think some.


00:55:54:12 - 00:56:16:14

Chris Keefer

People, you know, really outside of the nuclear industry, obviously, which has its own, you know, engineering culture, which is, you know, hyper vigilant, very safety conscious. And, you know, maybe we'll be accused of overengineering we'll say, you know, we've got such emergencies facing us in terms of, you know, climate, breakdown. we should regulate nuclear plants to the same level of, of safety as, say, a coal plant.


00:56:16:14 - 00:56:37:20

Chris Keefer

I think that was not a not Ted's word is exactly. But that's we have to look at the consequences of, you know, being so overregulated that we, you know, build a fossil infrastructure rather than nuclear. So why like why why do we need to, to regulate such a high degree again, given the relatively low public health implications of, you know, even, I would say the total worst case scenarios, but things like Fukushima.


00:56:37:22 - 00:57:09:15

James Krellenstein

Okay. So I have two questions for you. The first is a political question, which is that human beings, I think, unfortunately, do not evaluate risks, mathematically or quantitatively. We we require a dramatically higher safety standard for commercial airplanes than we do for passenger vehicles. 40, you know, 35,000 people in the United States alone, somewhere that in that range die in car crashes each year.


00:57:09:17 - 00:57:38:21

James Krellenstein

We have not had a car a a, you know, a, a passenger, civilian passenger death in the US aerospace industry for years at this point, not a single death. And yet right now, there are committee. You know, the CEO of Boeing was just a couple of days ago hold before the United States Congress and basically grilled because of issues with the 737 Max.


00:57:38:21 - 00:58:02:00

James Krellenstein

Immediately that did lead to a fatal accident, but also just to end an example of something where a door, door plug literally exploded out of a Boeing 737 Max, injuring and killing nobody. But it was a national news story, and a lot of people said they'll never fly a 737 max after this story. My question to you is, is that logical anyway?


00:58:02:06 - 00:58:21:11

James Krellenstein

Is that a rational response? Here we have the car drive to the airport, as you know, 10 to 100 times more on a per kilometer basis, or even a thousand fold higher probability of you dying. Yet more people, so many more people are concerned. Make it a political priority to talk about the safety of Boeing and Airbus. They do about the airplanes.


00:58:21:13 - 00:58:44:01

James Krellenstein

That's a logical that's not fair. That's not right. And we probably would be a lot safer if we could transition a lot more of travel to commercial airplanes rather than, rather than leaving on, on automobiles. it's the same argument. And maybe we have to lower the safety standards. Maybe we don't need two pilots. Like, how often ultimately do we need a pilot and copilot?


00:58:44:03 - 00:59:13:10

James Krellenstein

The question I have for for everyone here is, is that the reaction? Sure. Fukushima Daiichi did not end up having a meaningful public health impact at all. And we know even if you believe in the most conservative, linear, no threshold hypothesis modeling, we know that the evacuation, Fukushima Daiichi killed more people than even the worst case scenario L.A. modeling says would be dying from the radiation is that doesn't mean we didn't evacuate people.


00:59:13:12 - 00:59:43:00

James Krellenstein

And that doesn't mean that people weren't freaked out at the entire global nuclear industry did not have a massive change of that of of, you know, the prospects of it and public acceptance at the Fukushima Daiichi until we figure out as an industry and maybe there is a way to do this, I don't know. This is not my my area of expertise until we have a way for the public to appreciate that nuclear safety does not need to be particularly high.


00:59:43:00 - 01:00:09:06

James Krellenstein

I think maintaining a high level, safety is prudent for the industry's future, because ultimately, Fukushima didn't probably killed far more people by leading to the shut downs of the German and Japanese nuclear power plants and the excess coal power emissions. Did the we know that they did the actual radiation released, but even in the most conservative modeling kill.


01:00:09:07 - 01:00:31:19

James Krellenstein

But it still happened. And my question is, are we so confident in our apparently glorious nuclear industry PR skills to be able to communicate after we have a higher frequency of core melt events, the core damage events after we've say, weakened safety standards to be able to to do this. The second question I have is what are we talking about?


01:00:31:19 - 01:00:59:19

James Krellenstein

Which safety systems do we want to eliminate? Which regulatory require ments need to be eliminated? Because, you know, I have this huge amount of obsession that I see nuclear advocates engaging in linear no threshold hypothesis versus, you know, a sigmoidal model of versus a hermetic model for LMT. But most of the cost drivers in terms of safety systems are providing large source term releases.


01:00:59:21 - 01:01:25:17

James Krellenstein

They're not to prevent, like the small minuscule releases. And we never this wasn't always the case. So take the example of three mile Island unit number two, the accident there. We correctly say this was an accident that had almost certainly no public health impact besides maybe psychological impact. that it happened, it released less radiation than the NRC allows a nuclear power plant in a year to routinely release.


01:01:25:19 - 01:01:55:19

James Krellenstein

Right. It released about 80mg right. The NRC rig, of course, is 100mg a year. Right? No nuclear power ever comes close to that, right? a dirty plant would be 5/10 of one milligram in a single year. But the safety system. But inside the containment building of Three Mile Island during this incident, the radiation dose reading inside the containment dome was 800 rem an hour, eight sieverts now.


01:01:55:21 - 01:02:29:02

James Krellenstein

So forget about LMT, forget about cancer risk or anything. If you went into that containment dome in 20 minutes, you would have received a dose that you would have a 50% chance of being dead in two weeks. Now, no one did. No one died because we had a massively overengineered containment building. But that wasn't always the case. Take a power plant that was built like San Onofre unit number one, not songs two and three, the ones that just were shut down in 2011.


01:02:29:04 - 01:02:55:03

James Krellenstein

But songs unit number, unit number one had so little shielding initially around its, vapor shield. Right. What was before we had containment that in 1976, Southern California Edison had to build the sphere enclosure building, right, to be able to comply with new AEC regulations to prevent literally just if there was an accident, even without containment failure. Right.


01:02:55:03 - 01:03:25:00

James Krellenstein

Just the shine. The gamma shine would be so significant in a serious accident to actually harm people in the vicinity made plant workers only because it would shine. Not actually just so fission product release. But we forget that a huge amount of the safety that we do have in the US nuclear world comes from actual regulatory requirements, or even the high pressure safety injection system that ultimately was able to terminate the loss of coolant accident at Three Mile Island.


01:03:25:03 - 01:03:42:14

James Krellenstein

That was an emergency call cooling system that was mandated in 72, 73 that plants necessarily before that regulation had did not need to have. So I think we had the classic regulators dilemma, which is that you make an industry really safe by regulations. We have obviously some work that we need to do in streamlining some of these effects.


01:03:42:16 - 01:03:47:21

James Krellenstein

But what are the safety systems, I ask you that you want to eliminate that will drive the cost down so much.


01:03:47:23 - 01:03:50:20

Chris Keefer

I can't say I have a great answer for you there, James.


01:03:50:22 - 01:03:52:17

James Krellenstein

what what he thinks motivating.


01:03:52:19 - 01:04:10:17

Chris Keefer

you know, we've said, you know, we're searching for what the root causes of of the cost spiraling and obviously trying to control that. if we misidentify what those root causes are, we're going to spend a lot of efforts, focus on the wrong thing. so, you know, what do you think's leading to this? Focus on regulation right now?


01:04:10:17 - 01:04:25:20

Chris Keefer

I mean, again, I started the intro by talking about, you know, this is something I hear commonly is, it's just so onerous to get a, a license. in terms of the, experience of new skill, half $1 billion or. Yeah. Yeah.


01:04:25:22 - 01:04:48:02

James Krellenstein

Personally, I think that it is too onerous to get a combined operating license, for example, or design certification. you know, I think I was at the NRC in January, you know, working with with actual people who are interested in building nuclear reactors. not just advocating about it and trying to work through simplifications and new strategies to deal with it.


01:04:48:02 - 01:05:18:05

James Krellenstein

That's absolutely true. My question is a more simple one. Not that this doesn't need attention, not that there isn't actually dramatic safety, you know, changes that need to happen at the NRC. the better question is, is this really the rate limiting factor? Because as onerous as it is to get a coal in the United States, right. this part 52 and below, we ultimately issued 70GW of nuclear fully licensed capacity.


01:05:18:07 - 01:05:46:13

James Krellenstein

No one wanted. You know, we haven't we only built two point, you know, 2 to 2.4 gigawatts of that 17GW so far. There's a huge number of actively, fully licensed shovel ready for a regulatory perspective with full environmental impact statements. The discharge permits that literally could build their entire nuclear power plant. So they built in conformance with the design and turn on and they're not building it.


01:05:46:15 - 01:06:14:00

James Krellenstein

So what exactly is preventing us from doing that? I think there's a lot of factors. I don't think right now regulations are the ones that are doing it. And what's good. We had this example, of China where China is building. It has built dozens and dozens of gigawatts of new nuclear, in the, in the last just two decades, and is building it in 5.5 years.


01:06:14:00 - 01:06:34:12

James Krellenstein

And we try to build our power plants there. The problems that we've talked about previously are the Ap1000, for example, dominate even with their regulatory system that has a proven track record of turning around plants at five and a half years. So I think it's a really convenient story that the regulatory system has once again failed US nuclear industry.


01:06:34:12 - 01:06:41:09

James Krellenstein

But that ignores the dramatic progress that we've made in nuclear regulation over the last 40.


01:06:41:11 - 01:07:00:10

Chris Keefer

This is maybe a little bit off topic, but, I think there's an interesting controlled experiment as well that we didn't touch on, which is that the EPA and the Ap1000 took a pretty similar amount of time, to get online in China. And you mentioned sort of the two different regulatory philosophies. Yeah. Underlying them. So, you know, if you 1000 is so much simpler.


01:07:00:12 - 01:07:04:20

Chris Keefer

as fewer, you know, active systems, like why did it take a similar amount of time?


01:07:04:22 - 01:07:33:00

James Krellenstein

So, the EPR is primarily a stick built power plant, right. and by that we mean it's not a Modularize nuclear power plant, as I think we've talked a lot on this, podcast, the AP 1000 utilized 342 modules, of which, you know, some of these know modules have 72 submodules in them. as an example, the the design issues that we had, at at the Ap1000 level, right.


01:07:33:02 - 01:07:51:09

James Krellenstein

really stopped the entire critical path of the project. Right. If the modules weren't available on site, it was, for many of them, difficult, if not impossible to fabricate them on site. Or if it did, it would take a huge amount of time to build the capacity to build those modules. The EPR is a traditional stick built nuclear power plant.


01:07:51:09 - 01:08:11:04

James Krellenstein

Very little modularization. So even though the design completeness was a lot lower, you could build it like a regular nuclear power plant is. And if you know, you could literally improvise a huge amount of that civil engineering, for example, and you weren't dependent on a module supply chain being able to spit out modules and ship you a module at the to the plant.


01:08:11:06 - 01:08:39:02

James Krellenstein

But what we have and as we've talked about before, unfortunately the the audience has, has really probably suffered a lot for the last four episodes of the ad nauseum, of me, of of my voice, when we didn't have a complete design, that even the Chinese project doesn't matter how great your constructors are or how great your welders are, how good you are at pouring cement, if literally the module isn't there, you can't proceed.


01:08:39:04 - 01:09:23:00

James Krellenstein

So the better question to ask is, what are we seeing now in these seven Ap1000 that are being built right now? We've done this work with some colleagues. and what we are fighting right now is just for the big six module placements on the nuclear island in China. We are going 75% faster in nuclear island construction than we were on the first generation of Ap1000 that we are building in China, which really gives you the, the, the, the massive sort of, speed advantages that you can get from a much smaller, much more, complex, much less complex and modularized power plant.


01:09:23:00 - 01:09:45:03

James Krellenstein

But once again, if the design isn't done, if the modules aren't there and if the supply chain isn't running, you can have the most brilliant design. On paper, it can be as simple as as peach pie, but it doesn't matter. You might as well choose a more complex design, because what ultimately matters here, just like nuclear regulation, often isn't the rate limiting factor.


01:09:45:03 - 01:10:12:10

James Krellenstein

It can be. It's not going to be that that the nuclear islands, you know, that you're 45, 45% less seismic, building volume. It's going to be that the modules aren't there and you're waiting a year for them to finish, get fabricated, produced. But now that we have the design complete and we have a design certification process, by the way, and a regulatory process that allows us to replicate for the first time, that's something we didn't talk about the standardization of it.


01:10:12:15 - 01:10:17:14

James Krellenstein

We have a real a real, you know, strong argument that this is the right way for us to be.


01:10:17:16 - 01:10:40:12

Chris Keefer

And that was, the other side of Ted's argument is that, you know, it's the regulation stupid. whereas these securities referring to sort of progressive nuclear advocates say it's, you know, failure to standardize, Ted was saying, you know, basically things were fairly standard. It was, you know, two big companies delivering quires and BW hours. He also says that, you know, standardization may have great hope with this kind of coming next generation of reactors.


01:10:40:12 - 01:10:42:10

Chris Keefer

That didn't make a ton of sense to me, but.


01:10:42:10 - 01:10:47:09

James Krellenstein

Well, I think that that does make sense because these basically argue, well, if you're man, if that's manufacturing the reactor, you.


01:10:47:13 - 01:10:49:03

Chris Keefer

Know, it makes sense.


01:10:49:05 - 01:10:50:03

James Krellenstein

Right? Listen, I'm a.


01:10:50:03 - 01:10:58:15

Chris Keefer

I'm a big fan of standardization across the board. But it was just kind of it was sort of like it didn't matter in the previous build that it will matter in the future. It was just it was just a strange superpower.


01:10:58:19 - 01:11:30:11

James Krellenstein

Let's go back and ask this very basic question, right. What percentage like when we talk about a Westinghouse or a GE boiling water reactor? So first of all, within that class of of things there is huge variety, right? So GE boiling water reactors, I mean there have been six different models, each different of which each one can vary in power by a factor of three.


01:11:30:13 - 01:11:56:21

James Krellenstein

Often you have some BWR fours that are at 1200 megawatts. You have others that are 500MW. But they're also built at three different massively different containment systems. Right. Which valve that vary in volume? Your Mach one containment is five times smaller than a mach three containment is. And what you're talking about building something, you know, that is really different.


01:11:56:21 - 01:12:25:14

James Krellenstein

Dramatically different, totally different designs. Totally different safety system designs, totally different civil engineering. But let's go back to this basic argument. What we're really talking about here is we're talking about who's provided the nuclear steam supply system. The nuclear steam supply system, equipment represents about 20% of the cost of a nuclear power plant. And saying that, you know, we basically we're doing the same things like, I nuclear engineering colleague, you know, after reading Ted's article, I was like, what the fuck is this?


01:12:25:16 - 01:12:54:00

James Krellenstein

Like basically said, that's like saying that a, a Boeing 747, which has General Electric engines, turbofan engines is sort of the same thing ultimately as a Boeing 78 787 that has General Electric turbofan engines. Yeah, the edges are 20% of the cost of the plant, of the plane. The engines represent 20%, of the new supply system components represent a 20% of the cost.


01:12:54:05 - 01:13:17:14

James Krellenstein

What actually matters is the massive building, the massive airplane that we have to build to house that in. And there's such huge design variation between A and the in the GE side, between a mach one and a mach three containment between a BWR two and a, A six, and the pressurized water reactor side, an ice condenser or a sub atmospheric versus a dry, large atmosphere.


01:13:17:16 - 01:13:53:04

James Krellenstein

These are containment building. These are completely different. If it's a western US tube. Western has three. Loop West has four loop A, Babcock's and Wilcox two loop, a combustion engineer 2 or 3 loop pressurized water reactor that we're sort of missing the forest for the trees. The vast majority of the cost that goes into a nuclear power plant is not for the nuclear steam supply system components, but for the, civil engineering and excavation that needs to happen on the plant to build the buildings that are the triple S's housing, as well as the balance of plant and the engineering services.


01:13:53:06 - 01:14:15:15

James Krellenstein

In the case of almost all of those components, none of them were standardized and had dramatic variation from plant to plant. Because even though you might have a combustion engineering pressurized water reactor, let's say you had totally different architects and engineers designing totally different containment buildings, totally different safety system designs, like auxiliary feed water systems, for example. Right.


01:14:15:15 - 01:14:38:11

James Krellenstein

You had totally different control room designs that were being designed for each one, even though they had the same end as that's all in the remit of the E of the architecture and engineers. Just like a McDonnell Douglas plane that uses GE turbo, fans might have a totally different cockpit design than a Boeing. you know, airplane that uses GE turbo fans, right?


01:14:38:11 - 01:15:05:12

James Krellenstein

They have totally different cockpits. They totally different balance of plants. They've totally fuselages. These the amount of variation in the nuclear industry is not just measured by the 56 different N triple S systems that were deployed in the fleet of only 112. It also goes down to the massively different architects and engineers that we're building. The vast majority of the cost of the powerplant.


01:15:05:13 - 01:15:17:01

Chris Keefer

So going forward, do you think the, are you a fan of, it's the lack of standardization, stupid. And we're in a new era. We're, you know, through the regulatory environment. so.


01:15:17:03 - 01:15:55:06

James Krellenstein

So I think it's all of these things. It's not there's not any stupid single multifactorial, you know, this is a multifactorial problem, but I think it's sort of naive to look at the massive progress and changes that have been made in the nuclear regulatory infrastructure and just say we that's still the regulation stupid, right? If you look at an AP 1000, which was not licensable, by the way, in the 1970s and 80s, heads of design, there's a huge amount of research that needs to go into how we actually evaluate and credit passive safety systems in both design basis and beyond design basis, accident scenarios, right.


01:15:55:08 - 01:16:16:20

James Krellenstein

we also have the advantage of not only having that those passively safe designs, as well as actively safe designs that grow plenty your your free to build the advanced boiling water reactor of course is mainly with the exception. This is even arguable of whether the reactor core isolation cooling system is passive. It doesn't require AC power necessarily, but it's certainly a turbine and it's moving water.


01:16:16:22 - 01:16:38:07

James Krellenstein

But almost everything else is, of course, active emergency cooling system and so on. Let's let's say that you can build your active safety plant if you want to, but the passive safety plants really have a huge advantage, I would argue, in terms of economics. But and that's your choice if you want to go down that line in terms of simplification.


01:16:38:09 - 01:17:04:06

James Krellenstein

but we also kind of equally importantly, have a standard, sizable licensing process that allows us to replicate the exact same plant over and over and over again and have surety when you commence nuclear construction. Right. And let me explain why that that matters, because it's not one of the interesting things. And we I think actually I this you worked on this, you know, we were talking to Yakupov, a journo at MIT, about this.


01:17:04:06 - 01:17:24:21

James Krellenstein

What are the really interesting things that we've been able to do with a new regulatory paradigm? If, for example, on the Ap1000, if you look at the licenses that have actually been issued at multiple sites across the country, one of the really interesting things is that the design for the nuclear island has been standardized from the basement and above.


01:17:24:23 - 01:17:51:11

James Krellenstein

There may be some geotechnical sort of fill, and the mud in that configuration is going to have to change a lot based on, you know, the site geology and site soil characteristics. But we have really standardized above at least where the site where we don't have to go beyond what the peak ground acceleration and the sort of, the spectra that we have for the earthquake, we have been able to standardize entirely the nuclear island.


01:17:51:12 - 01:18:28:20

James Krellenstein

That's the first that's really a first and allows us to have regulatory certainty about doing that and really actually imagine us doing a more fleet like deployment, that it was never the case before in the US regulatory environment. And allows us to actually incorporate the lessons that we learned at one plant, incorporate that into the design control document or into the FSR, and get surety that the NRC will be okay if we build that in in the same way for the next plant that did not exist under the part 50 framework did not really exist even under the appendix oh framework of of of preliminary and final design.


01:18:28:22 - 01:18:43:13

Chris Keefer

All right, James, we could go on for hours. I got to pick up my little one from, aftercare. but it's been great chatting. you know, I've learned so much in this. I think it's going to be a great contribution. there are some cameras on Twitter from Angelica and others, about this very question.


01:18:43:13 - 01:18:47:06

Chris Keefer

So, glad to, glad to get a chance to get your perspectives.


01:18:47:08 - 01:18:48:12

James Krellenstein

Thank you for having me.


01:18:48:14 - 01:18:49:08

Chris Keefer

Until next time.



bottom of page