Vogtle Part 2: Murphy's Law

00:00:01:00 - 00:00:23:06

Chris Keefer

Welcome back to Decouple for a much anticipated part two of a recent episode examining Vogel and the nuclear Renaissance. That wasn't James. It's been fabulous. We've had lots of feedback to that first episode. People are really clamoring for the follow up that we did promised them. It's been a couple of weeks. It's always good to make people wait a little while, right?

00:00:23:08 - 00:00:26:08

Chris Keefer

But thank you for taking the time to come back on.

00:00:26:10 - 00:00:48:12

James Krellenstein

Yeah, no, I'm glad to, you know, sort of report here back from the Death Star that I'm currently sort of recording this in. But, you know, I think that, you know, there has been a lot of feedback to the episode. And one of the things that I think is most interesting about the Vogel and summer saga is, you know, where the community has responded and hasn't responded.

00:00:48:12 - 00:01:11:00

James Krellenstein

And I want to take maybe a second to think about how we are all thinking about where we are in the nuclear space right now and how Vogel and Sommer fits into that narrative. You know, I think as you know, from the 30,000 foot level, if we wanted to talk about what went wrong at Vogel, we would basically say, you know, Westinghouse hadn't completed the design yet.

00:01:11:02 - 00:01:45:05

James Krellenstein

The about the EPC hadn't completed their part of the design. That's the first major problem. We had a second challenge in that we did not have a supply chain that was ready to supply a modular plant like Vogel at the quality that was needed. And at the time and schedule accuracy that was needed. And the third issue was that we hadn't built a nuclear power plant in decades in the United States, and we had a general lack of familiarity of the workforce with, you know, building things at the level that is needed for a nuclear power plant build.

00:01:45:07 - 00:02:21:04

James Krellenstein

But I want to put this now zoom out from that and just talk about where we are today in 2024, right where we see that there is no new nuclear power plant under construction in the United States. None. There's, you know, some test reactors that are being built for the advanced for the advanced or the yet to advance to use your recently coined term or nodes or something to coin term and we have some, you know, light water reactor projects going on in Europe, particularly Hinkley Point C that are not exactly obviously not AP1000, but are are not exactly a stunning success so far.

00:02:21:06 - 00:02:48:03

James Krellenstein

And I think this should all cause us a moment to pause, pull back and ask the question, have we really learned the lessons that are available to us from Vogel? Some are old three Farmingville, three And are we as a community and as an industry, are we are we actually taking those lessons and rapidly trying to learn from them and ensure that we don't repeat the same mistakes?

00:02:48:05 - 00:03:30:23

James Krellenstein

And I think that advocates in particular have some responsibility, not responsibility, but in some ways this is one of the most important tasks that advocates can do, which is ensure that we are accurately seeing what happened and ensuring that it won't happen again. And that, I'm not so sure is happening. And I think some of the response to this episode on the more critical side maybe exemplifies that as well as as we'll go into some of the sort of mythmaking or that has surrounded this project in light of the fact in lieu of the fact that we haven't had a major sort of lesson learned report that has been issued by the industry about what

00:03:30:23 - 00:03:43:23

James Krellenstein

went wrong at Vogel three in that summer, two and three that we could utilize to publicly insure both customers that this won't happen again, but also make sure that we don't make the same mistakes.

00:03:44:01 - 00:04:05:05

Chris Keefer

So and the topic of of that lessons learned side of things and the industry being transparent about it, I'm not sure if I heard it from you or somewhere else, but I do understand such a document exists. But apparently just kind of access it. You need to get into Fort Knox, go through a whole number of security clearances, leave your phone at the door, you know, sign NDAs, etc..

00:04:05:07 - 00:04:07:11

Chris Keefer

That's the rumor I heard. Or is that is that.

00:04:07:12 - 00:04:19:00

James Krellenstein

You know, I think it's a rumor. I've heard that rumor as well. You know what? What I'll say is I have never seen such a report that exists. And you've tried everything you might.

00:04:19:02 - 00:04:22:16

Chris Keefer

Not many people that try harder than you. I.

00:04:22:18 - 00:04:44:00

James Krellenstein

I, I think maybe the nuclear energy Institute or the American Nuclear Society has such a report. I don't think it's publicly available. One of the things that is really, really clear is that if you go and talk to American utilities and American independent power producers, they don't really understand with much detail what happened at Vogel and what happened at summer, why it will or will not happen again.

00:04:44:05 - 00:05:06:01

James Krellenstein

And so this is a major, major barrier, I think, to new nuclear in the United States right now. And, you know, one of the things that I talked about in the last episode was I mentioned a sort of piece that Rod Adams made about blaming the aircraft impact assessment role. And I wasn't saying that the aircraft impact assessment rule wasn't a major driver of this.

00:05:06:01 - 00:05:31:06

James Krellenstein

Of course, what I'm saying is, is we can't say that the aircraft impact assessment rule, which was not applied in China, that the NRC, which does not exist, you know, doesn't have authority in China. When they took China at 9.4 years of nuclear construction to get an AP 1000 online, which is, you know, three standard deviations longer than what the 40 to recent large light water reactors they built, built was built.

00:05:31:12 - 00:05:53:23

James Krellenstein

We can't blame the NRC for for the root cause of this, nor can we blame the aircraft impact assessment rule for the root cause of this because we took away the root cause, we've got nearly the same outcomes and those outcomes also, I mean, what's amazing that the Chinese example is not only the Chinese on a median build a reactor in 5.5 years, the sun Min and hyang builds didn't happen during COVID, right?

00:05:53:23 - 00:06:27:10

James Krellenstein

Whereas Vogel three and Vogel four had to go through COVID, which is a major disruption of productivity. It didn't go through sort of they have a labor force that is highly, highly able to build nuclear power plants and churn them out in a five and a half year sort of timeline. This really was to a large extent, not only not saying there wasn't challenges and major challenges induced by the regulator, but this was primarily, I would say, a result of the industry not really being able to deliver a project.

00:06:27:12 - 00:06:51:21

James Krellenstein

And we shouldn't shy away from that fact because the only way that we can ensure that the industry will be ready for the next project build will be to really, really ensure that the lessons that happen at Vogel and Summer, which are a treasure trove of of lessons learned, are understood, you know, socialized and capitalized upon in the broader community and the broader industry.

00:06:52:01 - 00:07:17:11

James Krellenstein

And this applies not only to AP 1000 projects, but AP 300 problems projects. BW 300 just general nuclear project builds. This is really, really important and we need to have a much better story to be able to tell other people who are considering building the nuclear as customers about what happened and can't just say, it was the Nuclear Regulatory Commission is fault because they are, you know, they're about to make a multi-billion dollar investment.

00:07:17:13 - 00:07:57:17

James Krellenstein

They can see the same information that I can see that China somehow was able wasn't able to build it either. They didn't have the NRC nor an aircraft impact assessment role on time and on budget. So let's really actually be serious about this. And, you know, there's some other examples that we could dive into the sort of mythmaking that has happened in the community around Vogel And Sommer, if if we're interested to dive into that, I think sort of just show this compulsion of, I think a compulsion, very understandable impulse of a lot of nuclear advocates to sort of blame other folks, external sort of forces for the challenges that we face in this build.

00:07:57:19 - 00:08:19:09

Chris Keefer

Yeah, no, absolutely. I think, you know, as as we outlined in the previous episode, having this case control experience over in China, hugely, hugely helpful to start breaking down what are the relevant and most important variables. Last last week or I guess was a couple of weeks ago, we were really putting ourselves in the perspective of, you know, had we been there in 2005, what what would this all look like?

00:08:19:11 - 00:08:43:18

Chris Keefer

Could we have seen it coming essentially? And this episode we talked about sort of moving into the noughties. I think we're still going to be drawing comparisons between China and the US, but getting a little bit more into the specifics of the projects at Vogel and Summer. So, you know, unfortunately, I spent all morning arguing over this silly post on advanced and yet to address reactors rather than preparing for this much more important episode.

00:08:43:18 - 00:08:50:13

Chris Keefer

James So I'm going to let you sort of steered in terms of the architecture and hopefully pipe in and somewhat intelligent ways throughout right?

00:08:50:13 - 00:09:19:03

James Krellenstein

Yeah. So I mean we can restart back to I think we sort of ended pretty early in the sort of experience, I think before even nuclear construction started where we just saw the sort of supply chain starting to break down at the Shoreham modular factory over in Lake Charles. And we really started to see the project just become, you know, behind schedule because we weren't able to supply the modules that were necessary to build the plant on time and on schedule.

00:09:19:05 - 00:09:53:02

James Krellenstein

And one of the things that I think is important, as we talked a little bit about, is that each of these large modules, right, are broken down into dozens and dozens of sub modules. And this is a necessity, not in all power plants, right? But in the place, for example, Vogel doesn't have navigable barge access, direct Lee at the site, which means that we would go to maybe the port savannah or another port and sort of drop off the sort of sub modules and then ship them by land right to the site.

00:09:53:04 - 00:10:16:16

James Krellenstein

And this would be turned out to be one of the major differences that if you do a historical comparison, as we start building out the comparisons that we had compared to other sites and other previous project builds, in particular, I think the advanced boiling water reactor builds that we saw in Japan that everyone sort of keeps as this example of, hey, we're able to build large light water reactors in 38 months, 36 months, maybe even.

00:10:16:20 - 00:10:32:11

James Krellenstein

Why couldn't we do that the same thing? Well, one of the things and this is not you know, one of the things that they did there is they completed the design impeccably before the project build was done. But one of the things that also is a direct comparison, a dark difference, excuse me, is the fact that in that build.

00:10:32:13 - 00:10:52:05

James Krellenstein

Right, what we saw was the modules, right. Being much, much larger and literally being barge up to the site and then the the large lift crane literally taking those modules directly off the barge and shipping them, you know, sort of moving them maybe a couple thousand feet over where to where they need to be placed in the site.

00:10:52:07 - 00:11:13:00

James Krellenstein

And what what was the difference here is that when we have even those modules, when they arrive from shore down in Lake Charles, they see a to zero, right? They had to be then assembled and built in a module assembly building. Right. And, you know, welded together and then ship, you know, sort of moved over to where they would be placed in the plant site.

00:11:13:02 - 00:11:36:00

James Krellenstein

And this is a contrast to what we have seen in other sort of pre module or generation three builds that may have been more successful. And this doesn't explain all of it, of course, but the Chinese builds did have direct sea access directly to the site and that didn't make that much of a difference. But one of the things that we talk about when we're talking about modular construction is the idea that we can literally be building these things in factories.

00:11:36:00 - 00:11:58:19

James Krellenstein

Well, not only do the factory fail in the Vogel and summer builds, at least initially, and they were unable to delay to deploy the sub modules on time and on budget. We started to see the fact that we had to build a whole new factory called the module assembly building at the site, right in order to assemble those sub modules into the big modules that they wouldn't be placed at the site.

00:11:58:19 - 00:12:27:03

James Krellenstein

So one of the things that I think that we should all be looking for when we're looking at, you know, where do we want to deploy the next large light water reactor, things that seem quite you know, I don't want to say very practical, much less exotic than, say, how, you know, a molten salt reactor would be built or, you know, core neutron x, But the very basic logistics of how we're going to ship things and what the size that we're going to be able to ship things to at the site really should be a critical factor.

00:12:27:03 - 00:12:51:01

James Krellenstein

One of the critical factors that we look at when we're elucidating the or sort of distinguishing what the attractive elements of a site are and not out of sight because this turned out to be a sort of, you know, interesting issue that was was faced that. VOGEL Now, in the defense of the Vogel workers, there's a lot of the sub modules that would arrive if the module assembly building and they would be at such a low quality that they would have to literally be re welded.

00:12:51:07 - 00:13:20:07

James Krellenstein

Even the work that was done at the module at at shore, let's say you really have cases of workers having to grout, you know, grind out the wells and resell them to basically meet the Q HQ said you see so many faces having this module assembly building at the Vogel site was a blessing and having that sort of experienced of a workforce of a craft labor workforce, I was able to do that module assembly and integration was a really great thing, but it shouldn't have been necessary.

00:13:20:11 - 00:13:35:14

James Krellenstein

We should have had a supply chain that really was ready to do it and a logistics sort of sort of delivery chain that's able to do it in a way that minimized onsite sort of, you know, work that needed to actually get those modules ready to be placed and unhook.

00:13:35:16 - 00:13:56:00

Chris Keefer

So modularity is definitely a major theme here, and I don't want to skirt it at all whatsoever. Maybe we'll come back to it, but maybe before we do, just for those of us with a little less experience with nuclear construction, I mean, I watched a video of sort of a time lapse video of an a BWR being built in Japan, which is pretty miraculous.

00:13:56:00 - 00:14:27:18

Chris Keefer

But it just gave me that little bit of perspective on what actually goes into building a nuclear power plant. Obviously, there's a whole bunch of excavation. There's a base pore, there's then triple as balance of plant containment. I don't know if you can if it's too much for a podcast, but if you can take a minute or two and just describe what the various steps are just, you know, from really 30,000 foot view and maybe, you know, from Vogel or in general what seem to be the challenges I've heard, again, based on that that might work that the plant is actually not a major cost driver, but is it a time driver, etc..

00:14:27:18 - 00:14:35:10

Chris Keefer

I just want to get a sense of, you know, what the five or six steps are, and that's probably far more than that. But grant our listeners a little bit before we.

00:14:35:10 - 00:15:07:11

James Krellenstein

Dive in, breaking it down to five or six steps. I think it's a little bit, but there are many more than that. But if you want to just think about what happened, right? You know, when we talk about this ten years to build. Vogel Right. So from March of 2020 to 2013, when the license was issued to March of 2023, when it actually sort of went critical and commercial operation, you know, there was about a couple of years before that, they were actually doing a lot of work on the site.

00:15:07:11 - 00:15:30:01

James Krellenstein

They got something called an early site permit, 2009, 2008. And that whole idea was they couldn't pour any sort of safety critical concrete, but they could prepare the site, clear the site off. Right. Prepare stuff for the balance of plant for the non-nuclear side, and also begin, you know, preparing the fill, as we call or the backfill that the plant will actually sit down.

00:15:30:01 - 00:15:53:19

James Krellenstein

So at Vogel in particular, what we did is we dug down 100 feet right below the plant grade down to a layer of clay, and then we rebuilt up to about 50 feet, right. With with a sort of bare CD of safety critical fill. Right. That actually was regulated by the Nuclear Regulatory Commission because it impacted the seismic response of the nuclear island.

00:15:53:21 - 00:16:15:22

James Krellenstein

And actually, even before we started the early, you know, the actual licensed nuclear activity, we were already close to a year behind the scheduled construction. And just because that process of actually filling the safety critical fill wasn't actually in compliance with the license or the early requirements of the early site permit. So this really gives you an idea of that.

00:16:15:22 - 00:16:37:23

James Krellenstein

Even before we started the nuclear grade, sort of concrete and construction pouring, you know, sorry, concrete pouring and construction, we were already behind the mark on this project. And it also gives you an idea that that should have been actually a time that Westinghouse had the ability to really be catching up. But it was just really falling way, way behind in every single way.

00:16:37:23 - 00:17:00:01

James Krellenstein

So so here we are. So we've we've done what we call, you know, the sort of site preparation. We've obviously cleared it of, say, you know, trees of boulders we've dug down to wherever the sort of base mat. You know, we generally go way below the base mat maybe for we're at if we're pouring directly on or we're building directly on bedrock, we do rock inspection, right?

00:17:00:01 - 00:17:34:11

James Krellenstein

But we generally do some sort of inspection to make sure that where where we've, you know, dug down to is actually responding to the requirements that was assumed when the plant was going to be built. So if you look at the early pictures, for example, of the Volga site, they really have sensors that are measuring, say, movement of the sort of of the safety critical fill and the sort of the sort of dense layer of clay that's beneath there to make sure it's in sort of requirements that is necessary for the seismic sort of qualifications that we need the structure to be qualified to.

00:17:34:13 - 00:18:00:23

James Krellenstein

So I guess you could call that sort of, you know, rock inspection or base mat preparation just to make sure that we're actually building on a literally a solid foundation on the earth to make sure. So let's try that the first step, if you will. And that's, you know, it looks like a big hole, but a very, very neat hill hole that needs to meet a bunch of regulatory requirements as part of the, you know, sort of safety basis of this plant.

00:18:01:01 - 00:18:04:22

James Krellenstein

So, okay, so we have that. So I'm going Yeah, go ahead. Do you have a question, Chris?

00:18:05:00 - 00:18:26:23

Chris Keefer

Well, no, I mean, I'm just you know, we were both over in Dubai and I was watching some videos afterwards about, you know, the construction of the Burj Khalifa, the 380 piles or whatever that were driven into, you know, through the sandstone. They actually run an electric current through the the rebar inside of those concrete steel piles because there's saltwater that's kind of seeping in and would erode the foundation.

00:18:26:23 - 00:18:32:06

Chris Keefer

And if that electric current wasn't kept there with the cathode sort of concentrating all the corrosion once.

00:18:32:06 - 00:18:33:06

James Krellenstein

Clonic.

00:18:33:08 - 00:18:47:14

Chris Keefer

Wild stuff that goes into these, you know, the world's tallest building or in Fogell, I mean, we're saying maybe is one of the world's most expensive buildings. Whatever moniker that it earns. But it is fast because we do hear a lot and I do want to get sidetracked on this. But, you know, people saying, hey, you know, how will these things perform in earthquakes?

00:18:47:14 - 00:19:03:20

Chris Keefer

And so getting a sense just of how, you know, 100 feet down is a long way. And the ways the ways that these things are seismically isolated, probably beyond the scope of this podcast. But certainly I love that description so far. So if you want to nerd out more on that or move on to stage two of our more than five things, go for it.

00:19:03:20 - 00:19:27:05

James Krellenstein

To your exact point, right? I'm not going through everything else, but there's generally, depending on the soil and the sort of bedrock and geology of the site. Right. Generally we do have to do a de-watering. It's called, as anyone who has done civil construction knows this, right. If you're going down below, say, the water table. Right. There's a possibility that there's going to be a lot of, you know, an aquifer or groundwater that's going to want to seep in.

00:19:27:10 - 00:19:47:18

James Krellenstein

So often you do have, you know, variously needle pumps or other sorts of pumps that allows this site to be watered and to be dry. Right. And then we're basically filling it back up to about 50 feet. And then we begin on the, you know, the way that the AP1000 sort of and this is very varying on from site to site.

00:19:47:20 - 00:20:10:18

James Krellenstein

But one of the interesting things actually about the AP1000 is at least four places that are not in the high seismic intensity. So the reference AP1000 design is designed for a reference peak ground acceleration or PGA of 0.3 G, Right. And G, of course, is, you know, the gravitational accelerate little G gravitational acceleration, which is 9.8 meters per second squared.

00:20:10:19 - 00:20:32:12

James Krellenstein

So you can do that math, right? It's around three meters per second squared. And more importantly, in a lot of ways, right, there's what we call a spectra that that talks about the the the way that actually an earthquake will move the plant. Right. It's not just the peak acceleration, but the distribution of velocities in different directions at the plant is seismically designed to handle.

00:20:32:13 - 00:20:58:21

James Krellenstein

Right. So given this standardized plant design that Westinghouse came up with, they came up with a reference base map design. Right. That sits over what we call mud. Not right. So what happens is we're at that 50 foot say, roughly speaking, and I might be slightly off on the levels. I think that the actual base grade of base that's like 47 or 46 feet below grade of the AP1000.

00:20:58:23 - 00:21:28:08

James Krellenstein

So the idea is we put a mud mat on top of that. What about safety critical, you know, foundation that the plant is going to be built upon and then a mud match. So everyone knows it's an unreinforced bit of concrete, right? Like it's like a you pour concrete like and basically with no reinforcement there. Right. And it's just to give a flat surface upon which to start building the the plant up, as its name implies.

00:21:28:08 - 00:21:59:13

James Krellenstein

It's like, you know, it sits on the mud, writes a mat on top of the mud that we then sort of have a level concrete surface that we can begin the base mat construction of the nuclear island. Right. So maybe we want to start transitioning from there. And that's where we get into nuclear island licensed activities. Right. At this point, this will be a highly regulated activity that we will have to basically be, as we're pouring the base mat, pouring the foundation of the plant, This is going to be really, really important to the safety case, particularly the seismic case of the plant.

00:21:59:18 - 00:22:29:04

James Krellenstein

So it can be highly regulated. And so we have to have a site license at this point or a construction permit in the United States. In the United States, sort of regulatory infrastructure to proceed to this point. So so we've poured this mud mat. And then in the what's really interesting with the AP1000 design is of the six plants that were reactors that were licensed at multiple different sites in the US, one of the interesting things that we saw, none of those were particularly high seismic areas.

00:22:29:04 - 00:22:57:20

James Krellenstein

They're all on the East coast of the United States, which is not particularly seismically active, although there is obviously some seismic activity. One of the cool things is that there might have been some difference in the mud mat design, but the actual basement was standardized across all six or six different sea issued cables. And the NRC was able to basically say, you can actually use the reference base mat design for the AP1000.

00:22:57:20 - 00:23:29:18

James Krellenstein

That includes, of course, Vogel and Summer, but it also includes licensed plants that were never get built like Turkey point down in Florida. Right includes William State's Lee as well as Levi County, which are all different geographic geographies, different states. But we're able to actually standardize even the base of the plant. Right. Which shows, I think, not to be an 1000 fanboy too much, but it really shows the robustness of the ability to standardize the plant even in different soil structure interaction and different site geologies.

00:23:29:20 - 00:24:00:12

James Krellenstein

So that that's sort of where we go. And that base mat is sort of six feet thick, If I remember right, for the AP1000, it's a highly reinforced concrete structure where even the details of the rebar are highly, highly regulated and the tolerances that are necessary on that rebar are very, very regulated. One thing I will say that's interesting, if you think about it, what's really going on here is that mud that that base mat is sitting basically on top of the mud now.

00:24:00:14 - 00:24:22:01

James Krellenstein

And so if you actually look at the safety case that's going on, they really are doing analysis to make sure that the plant won't float up right from that sort of from that from that basement and also doing frictional calculations to see how in a seismic event, let's say that the the the basement of the plant is sort of shifting alongside the the mud mat.

00:24:22:01 - 00:24:43:09

James Krellenstein

And to the other point there's general is a waterproofing membrane of some sort that is put in between the the different mud mat layers as well as the, the sort of basement. Am I going into too much detail? I have no idea like this. Like I've tried probably even something as simple to right, even something as illustrative. Pouring a basement here is complex.

00:24:43:09 - 00:25:20:22

James Krellenstein

I'm trying to do it. I won't go into. But we got into not only do we get into problems with the regulator and with them sort of deviating from what the quality of the safety related feel is. And literally, actually we get the first litigation even by 2012, 2013. We have Southern suing CB and I and Westinghouse because the Westinghouse and CB and I are claiming that there's an issue here with the with the way that they're doing the safety critical fill and southern says it should actually be your responsibility based on the contract that we signed here for you to actually be able to meet this.

00:25:20:22 - 00:25:42:19

James Krellenstein

And they're suing for $30 million, $38 million. They were unable to sort of reach a non litigation based settlement. So they literally go into court to begin actually fighting this out based on this issue that we have just in the safety critical sill that that needs to be before we even begin nuclear safety, you know, sort of regulated construction.

00:25:42:21 - 00:26:19:18

James Krellenstein

Right. We already are in an acrimonious relationship between the the three sort of parties. So I'm not trying to I know probably a lot of people are like, you know, Tony, you know, tuning out here, what I'm trying to say is these things are not only important from like being pedantic and being really nerdy about it, but they're actually commercially really, really important for everyone to understand what is necessary for building a nuclear project, because literally these parties are going to litigation before we think of this project being in any trouble in 2012, 2013 over issues before they even begin nuclear safety construction.

00:26:19:18 - 00:26:44:09

James Krellenstein

And this is maybe why I, I have expressed not doubt or hesitation, but just a note of concern or of something that people should be aware of when we're talking about even smaller projects that require such huge civil infrastructure to be built, it's really important that we drill down and understand these can be drivers of both cost and schedule overrun even before we get building it.

00:26:44:09 - 00:27:15:16

James Krellenstein

So when we're looking at the BW 300 and we're talking about a big shaft that goes way, way down of all nuclear related sort of safety concrete that maybe has to be down 140 feet or 130 feet below grade. That's something that we should be, you know, sort of flagging. Same thing with, of course, new scale as we previously talked, which had a much larger and almost equally deep sort of civil structure below below grade in order to deal with the aircraft impact assessment role and actually as part of that design.

00:27:15:22 - 00:27:34:12

James Krellenstein

So these I know sound the devil is in the detail and nuclear projects and I'm trying to bore your audience almost to sort of emphasize how sort of detail oriented this process actually is. So apologies to everyone out there who is no doubt saying why is he talking so much about the base?

00:27:34:12 - 00:27:58:14

Chris Keefer

I think decouple and the relative wonkiness and nerdiness of the podcast has already filtered out a lot of folks that that might have been turned off by that. So that was good. I think it's just just a matter of the interest of time. But again, I think as you're saying, very illustrative. I don't if it's possible and if it's possible, but if you could again, just try and walk through a few more other stages with with a little less excruciating detail, unless you want to make some pit stops for, again, other things that came up.

00:27:58:14 - 00:28:02:16

Chris Keefer

I'm not sure if this is the best way to organize the episode, but again, I just would love to have that sense of.

00:28:02:18 - 00:28:24:00

James Krellenstein

Well, I kind of like this because I think we're kind of doing an abbreviated fast sort of staging. So we begin sort of so if you think about a nuclear project from a 30,000 foot level, there's the improvements, sort of the work to get the site prepped, prepped and ready for actually, you know, putting, digging down, getting down, doing rock inspection or fill inspection, doing some safety critical fill.

00:28:24:00 - 00:28:45:01

James Krellenstein

So you have a nice base to basically build the plant on. Then we begin the actual process. And once again, I'm focusing on the nuclear island. I'm not focusing on the turbine island and what needs to be done in the circulating waterworks. So if you think about it for just from a labor perspective, obviously we have manual labor that's helping with all of this products, but, you know, processes.

00:28:45:01 - 00:29:03:15

James Krellenstein

But we have basically, you know, the base, the thing that we're going to build as we're building the base mat, which is the foundation upon which we're building the plant, we're using, you know, experts in civil works, construction, concrete, rebar, that's sort of where we are at the stage right now, as you'll know. Right. We're going have to finish that up.

00:29:03:15 - 00:29:22:10

James Krellenstein

And then we're going to have to not only continue building the plant up, but at a certain point we're going to have to start actually doing the h vac, the sort of welding, the piping. Right. And the electrical work that will be necessary to actually build the components and assemble the components that make up the nuclear steam supply system.

00:29:22:13 - 00:29:55:14

James Krellenstein

We're not even there yet. Right. We're still at the point upon which we are building the foundation of the plant so that we can actually build that nuclear steam supply system and assemble it onto the sort of foundations of this plant that we're building right now. But at Vogel and at summer, we begin to have an initial problem right away, which is that as part of, as I said, the the basement is highly, highly regulated because it's so critical to the seismic response of the plant in a design basis earthquake.

00:29:55:16 - 00:30:28:06

James Krellenstein

Right. So as part of the license, Westinghouse and the licensees had to basically submit tables and design information that contained the details, for example, of how the rebar was going to be designed and configured. Right. And what we saw in that plant in this build was the rebar, you know, specifications that the applicant themselves submitted to the NRC that was part of this one stop licensing process, which basically said, okay, you're going to tell us how that rebar is going to be built.

00:30:28:06 - 00:31:07:02

James Krellenstein

We'll do our safety and you'll do a safety analysis. We'll evaluate that safety analysis. And all you need to do is build in conformance with how you said you were going to build the plant. And we be in the first series of mistakes that occur here that require license, amendments. What turns out being the case here is that it turns out for minor, very minor detail in the tolerances and in some cases the actual design of the rebar that was actually fabricated and placed at the plant actually deviated from the way that Westinghouse Southern end.

00:31:07:02 - 00:31:39:03

James Krellenstein

KB And I told the Nuclear Regulatory Commission they were going to build it. And the NRC inspection force flags this issue and says you're out of conformance with your own license. How do you want to deal with this? In some cases, I'm not going to go into excruciating detail because it would bore the audience in some cases. We literally had to jack out a lot of that of a lot of that, found that rebar, that foundation and basically redo it in conformance to the license in other other parts of the of the basement.

00:31:39:05 - 00:32:22:04

James Krellenstein

We had to work with the Nuclear Regulatory Commission to basically change some parameters, particularly the compressive strength of the concrete, to be able to deal with the fact that we had non-compliant rebar placed by the licensee themselves that violated what they themselves had told the NRC they'd be building in design in spec with. So we get to this exact idea of one of the downsides, like with the modular construction of this one step licensing process, is it really assumes that you have a very, very rigorous quality assurance and quality control sort of, you know, in place for both your designs and your procurement and your placing to make sure everything's in conformance with the way

00:32:22:04 - 00:32:40:08

James Krellenstein

you say you're going to build the plant. But if you don't build the plant in the conformance with the license that you yourself submitted, you're going to have a problem and you're going to need, you know, license amendments or some sort of design change in order or replacement of the actual component to be able to sort of build the plant.

00:32:40:09 - 00:32:40:14

James Krellenstein

And I mean.

00:32:40:14 - 00:32:51:13

Chris Keefer

I've seen some video, I think it was Hinkley Point of this rebar. I mean, it almost seems that there's more rebar than concrete. It's just an incredible amount of steel. And so the issue was just the angle at which, like it was just the angle which it was late because.

00:32:51:15 - 00:32:52:17

James Krellenstein

Because I hear people saying.

00:32:52:17 - 00:33:00:14

Chris Keefer

Hey, like industry should just kind of go easy on these guys. Like, does it actually make any difference in the end? Like, just let it go? Like, what was your perception there?

00:33:00:20 - 00:33:26:15

James Krellenstein

So there were a half a dozen. I mean, I think I mean, don't I can go through the I don't want to walk through there were at least I think a half a dozen license amendment requests that were required to basically get the rebar and concrete into spec. Here's the issue. And some of it had to do with, for example, the way that the NRC versus the licensee were interpreting American Concrete Institute codes on actually how the rebar was coming in compliance with that code.

00:33:26:18 - 00:33:51:13

James Krellenstein

Just to give you an example of the sort of minutia that's involved here, here's I think the takeaway for this podcast that you would want to say is that most of the issues, I would argue maybe even all of the issues here, the NRC is basically saying we're holding you to the standards. You yourself set your set up in the license, write the license criteria that were put here, were put not by the NRC but by the licensee.

00:33:51:15 - 00:34:07:04

James Krellenstein

The licensee said NRC. This is what we're going to build into. Is this yay or nay? Do you like this or not? And the NSC said yes. And then it turned out, for example, on some of these issues, on leveling the sort of basement, it turned out that they said they were going to be within a certain tolerance.

00:34:07:04 - 00:34:27:00

James Krellenstein

And it turns out that the actual build was a lot more was a lot less tolerant when we actually tried to do this in the real world. And they had to go back to the NRC and say we're out of the sort of tolerances that we were ourselves said we were going to build, too. And that's where I think we get here's the the sort of pro-nuclear that sort of bull nuclear case now.

00:34:27:05 - 00:34:46:20

James Krellenstein

It's because we actually had a lot of these issues. Yes. Were caused by the lack of design finality. So it turned out how actually perfect how often we were going to able to meet the licensing requirements in one case was a lot of times was just driven by the fact that we couldn't actually build it as the license sort of specified because the design wasn't complete.

00:34:46:20 - 00:35:08:05

James Krellenstein

We weren't able to accurately simulate what that that design would look like in reality, because it wasn't done when we submitted the license. In other cases, it was just practical sort of construct ability issues. Westinghouse hadn't constructed a plant in a long time right? S.B. and I and Stone Webster had really not constructed a plant and not done the civil works on a nuclear plant for a while.

00:35:08:07 - 00:35:26:23

James Krellenstein

So their expectations of the tolerances and quality that they were actually going to be able to build was much higher than it was in reality. The good thing here is we actually have that experience. We have all that licensing material. So if I was building an AP1000 today, I would basically say, okay, I learned all of these license amendments from Vogel that you approved.

00:35:27:03 - 00:35:44:07

James Krellenstein

I'm going to make sure that I'm able to build to the actual real world expectations today before I start building. And the one step licensing process allows us to do this. But it really does point to you got to know your design well and now you have to know your design in, you know, in silico or on blueprints or on paper.

00:35:44:09 - 00:36:08:06

James Krellenstein

You got to think about the practical ways that you're going to construct the plant in the real world, which is going be a lot less perfect and design that into your licensing basis before you sort of make yourself held to overly stringent quality control, quality assurance sort of targets or, you know, tolerances in the licensing basis. I'll say one more point about this.

00:36:08:08 - 00:36:32:05

James Krellenstein

One of the interesting things I think that we have learned that a huge credit to G Itachi is there's a reason why G Itachi is not building the first BWR x 300 under this 50 to 1 step process. They're doing a construction permit, two step licensing approach for the first plant. They're going to learn all the practical lessons that we get from building a plant in reality, constructing it.

00:36:32:05 - 00:36:55:18

James Krellenstein

And then maybe their regulatory engagement plan is to be believed, turn it around and then do a design certification and a sort of licensing basis under one step so that they actually can have that real world construct ability, a final design and real world design basis together so that they can have a one step license that actually is buildable.

00:36:55:19 - 00:36:58:18

Chris Keefer

Okay. So I'm getting no, no, I'm getting but.

00:36:58:19 - 00:37:07:22

James Krellenstein

These are really important lessons, man. I got to say, I hate to tell you that, that the real world of nuclear construction is is not a pretty one. And it's slightly.

00:37:08:02 - 00:37:33:22

Chris Keefer

Just, you know, if you're just so used to these like, CAD illustrations, these animations on PowerPoints that I see all the time. But, you know, I think we've spent probably a time that's appropriate to how long the civil works elements take. And I'm curious to like how much, you know, we're going to go through. Some of the other main sort of chunks of process is the civil works, the interplays, etc. about the plant whatever content, how we're going to break down those other papers, other five or six sections, 1050 sections.

00:37:34:00 - 00:37:49:14

Chris Keefer

But it is going to be curious to see. Okay, Well, on the civil work side, was Vogel a huge anomaly compared to the average nuclear build? Like what sections of the build went over over time and over budget? Or is it just Murphy's Law and just literally everything that could go wrong went wrong at every stage?

00:37:49:16 - 00:38:18:04

James Krellenstein

So I want to say two things here. The first is the Vogel three and four was not particularly long build. Well, it was a longer build in the United States sort of history, but Vogel One in two actually took longer to build than. VOGEL Three and four less expensive, but longer. And a lot of that in fairness on one and two, that a lot of that was the NRC is fault.

00:38:18:06 - 00:38:38:04

James Krellenstein

VOGEL One and two was being built the post-mi it was one of the longer builds in US history, but we have a lot of long builds in the United States is nuclear construction history. So, you know, I want to be careful here in saying that everything that went wrong at Vogel did go wrong. That's actually kind of true.

00:38:38:06 - 00:38:58:23

James Krellenstein

We saw challenges as we'll talk about in almost every aspect of the build that we had, which in some ways may have been expected because this was the first nuclear power plant that we'd started. Construction had been ordered since literally 1978. That was and you know, slightly before that, that actually beginning construction. So maybe that's not so surprising.

00:38:58:23 - 00:39:25:01

James Krellenstein

What the note of caution I will just tell everyone here I will tell the audience is let's go back to the Chinese control example. Right. The nuclear power plants that China is building. You know, they're all with the exception of the VR product line. Right. Which is Russian Rosatom. Obviously, they're all derivatives of America or American or French nuclear power plant.

00:39:25:01 - 00:39:52:20

James Krellenstein

So even a long one, the so-called indigenous sort of Chinese nuclear power plant, nuclear steam supply system. It's history, of course, if you look at it and it's not a secret, it's based on the French three loop pressurized water reactor, the sort of spy series, you know, with changes. And that, of course, is based on the American Westinghouse three loop product line at Beaver Valley and and at I'm forgetting the name the Virginia plant that's North China.

00:39:53:02 - 00:40:36:02

James Krellenstein

And so those plants are coming together in China in five and a half years. And they have as much civil works, actually slightly more civil works then saying if you want thousand project did. Here's the point. Despite the fact that China is able to on a median so 50% are building faster than this 5.5 years of nuclear construction with all that civil works with a more classical sort of basement and containment design, the Chinese build took 9.4 years for the AP1000 and maybe a little bit slower, you know, 8.5 years on Hyang to right so these were much, much longer, you know, standard deviations, multiple standard deviations longer even in a workforce and a country

00:40:36:04 - 00:41:11:06

James Krellenstein

that is very, very experienced in nuclear civil construction. So there was really issues and challenges that were AP1000 specific. Now on civil civil works in civil construction, the Chinese plants were actually on time and on schedule. So so they actually did get that far. But then they started hitting other challenges that weren't associated with the base construction. So that's just a1i want us to to to to be clear here is that we have had building problems of nuclear power plants going back decades and in the United States and this build in particular was problematic even for very experienced builders.

00:41:11:08 - 00:41:38:11

James Krellenstein

So let's move on sort of from the basement sort of construction right into where we're starting to begin building the actual, you know, containment, though the safety critical welding that's going to be needed, the piping modules and so on. So one thing I want to just go before to your earlier point, before we move off of concrete. I'm sorry to to belabor the concrete issue.

00:41:38:11 - 00:42:04:00

James Krellenstein

You were correct at saying that the rebar is very, very dense nuclear power plants. One of the general issues that we see as problems called consolidation in concrete, the ability to make sure that there aren't voids or pocketing within that very heavily dense rebar. And so this is not trivial and this is something that is really, really important in nuclear power plant, civil construction and concrete work in concrete engineering.

00:42:04:01 - 00:42:27:11

James Krellenstein

And this has been a challenge that we said one thing that should work in the AP 1000 favor and disfavor designs like Nuscale, for example, or the EPR, is that the actual safety critical footprint of that nuclear island is much tinier than we historically see large, mainly because of passive safety systems. So it should be better, but obviously in this case we're already in trouble here.

00:42:27:13 - 00:42:52:12

James Krellenstein

We were already behind schedule before this even happens. Okay. So we're starting to build basically the structural modules that consist of the of AP1000 as we start building up and we begin starting to see major, major license amendment requests that are having to basically fill in to basically be able to build on top of the space mat that's happening.

00:42:52:14 - 00:43:14:01

James Krellenstein

But before we go into the regulatory challenges, remember how we started this podcast about all of the problems that we had with these modules that were coming all of a sudden this is starting this is a point to the construction that we're starting to get in trouble because of this. So C 8 to 0, it's basically the entire auxiliary building, including the spent fuel pool.

00:43:14:03 - 00:43:52:02

James Krellenstein

Right. Is in this one mega thousand metric ton module that obviously is going to need to be put place on the base mat placed into concrete sort of integrated in here's the problem when that when those sub modules haven't gotten to the module assembly building or have gotten to the module assembly building and need to be reworked on site to basically be able to get up to the Q A that's needed to be built where all the strings trying to run into sort of scheduling issues because maybe we've built the base mat up, but 8 to 0, maybe two years or a year late in actually being placed on to that, that onto the actual

00:43:52:02 - 00:44:13:17

James Krellenstein

base map. So all the mod, the advantages of modular construction where we're supposed to be snapping things on this is where we start really getting into trouble and then we begin to actually do the sort of forms major areas of craft nuclear construction that we're doing. So in nuclear projects, we generally distinguish between what we call craft labor and manual labor.

00:44:13:17 - 00:44:39:18

James Krellenstein

So craft labor are going to do things like electricians, like HVAC technicians, welders, etc., that have a very specific skill set and a pretty amazing artisans that are going to be able to actually do this really. And, you know, technically challenging issues in actually assembling the plant. Whereas shit man, we're already at 1250.

00:44:39:18 - 00:44:44:02

Chris Keefer

Five, it's all good. So that's going to be a part. It's part three, it's no problem.

00:44:44:04 - 00:44:46:07

James Krellenstein

Go fuck, man.

00:44:46:09 - 00:44:50:01

Chris Keefer

Carry on. It's good.

00:44:50:03 - 00:44:52:18

James Krellenstein

I just look at the time I start the day.

00:44:52:20 - 00:44:54:03

Chris Keefer

It's all good.

00:44:54:05 - 00:45:33:10

James Krellenstein

I'm sorry. I'm sorry. You're out of the. no, no, no, no. It's. It's okay. It's okay. So. So we're starting to see those challenges really start piling up on on the the project. And we're also beginning to see that the detailed work packages that are being distributed out to the the the work face to the work teams of craft laborers and manual laborers right there have to go in and actually literally do, you know, build a structure, pour some concrete, do a weld or assemble a cable tray.

00:45:33:15 - 00:46:01:21

James Krellenstein

Right. They're starting to realize the work teams are coming back and literally realizing that the work packages that they're being handed over are unbelievable. Right? That there's literally physically they're physically unbelievable because the design hasn't been finished at Westinghouse headquarters, let's say, or at CBI headquarters and the construction maybe three or six months ahead of where the engineering bases are.

00:46:01:23 - 00:46:31:03

James Krellenstein

So we literally have the cases where work teams are literally attempting to build this plant and they're months like they're really waiting around for weeks and weeks and weeks, right. To be able to actually build the plant out because they have to wait for it to be redesigned again. An engineering design and change review that has to go all the way back up to the headquarter, you know, sort of engineering company and be re-engineered and reissued to the spot.

00:46:31:03 - 00:46:54:12

James Krellenstein

So you have all these high priced laborers, right? The labor. One of the things we love about nuclear construction is these craft laborers are very expensive, right? And that's really good for them because they get high wages as a skilled artisan job, Right. What, you want them to be properly compensated becomes even more expensive when all of a sudden you don't have the actual buildable work package that actually can be built.

00:46:54:14 - 00:47:10:07

James Krellenstein

Now, this is one of the challenges that we have with first of a kind builds, but it's one of the things that why it's so important that we build an AP1000 today because we know we have work packages that are buildable because we built them. And so this sort of major consideration.

00:47:10:09 - 00:47:29:01

Chris Keefer

I mean, just as a slight tangent or offsite here, I'm thinking about this project like almost like a war and, you know, the incredible complexity of, of, you know, that project delivery, you know, it's like you have a general, you have marshals, you have like a whole communications structure. Things need to move down and you have the morale of your soldiers.

00:47:29:03 - 00:47:44:13

Chris Keefer

And you know, this craftsmen sitting on their hands for days, weeks at a time. I mean, I can imagine the lunchroom conversations and just you wouldn't feel particularly as motivated as you might if you if you lose faith in your commander in chief kind of thing. I don't know. Like I don't know if that's an appropriate analogy or if that was a major factor.

00:47:44:13 - 00:47:46:23

Chris Keefer

But it's just on the human side. I can just imagine.

00:47:47:00 - 00:48:19:18

James Krellenstein

I mean, one of the issues that we had here at this project, let's just be frank, was was retention of the skilled labor force over this massive time. And part of that was compensation. Part of that was shortages of workers that were highly skilled, that were needed for other projects in the region. But another problem, as you just said, is, is you have that problem is going to be compounded if you're not able to use the labor force most effectively in terms of actually getting project completion, you know, actually getting a work package actually built.

00:48:19:23 - 00:48:43:14

James Krellenstein

If people are just literally sitting around because they're waiting for the work package was supposed to be buildable, redesigned so it can be built, you start seeing how these projects start, these problems start piling on, even as we start building up from the foundation and start doing more craft labor, more welding and more electrical work. Now let's go back to what's going on at the project, at the project level.

00:48:43:19 - 00:49:06:16

James Krellenstein

So everyone's realizing this is 2013, 2014, we're way behind schedule. We're not getting the sort of project completion rate that we're needed as part of the sort of master project schedule. And there's a problem here that Chicago Bridge and Iron has bought, you know, and the old joke is, of course, that Chicago Bridge and Iron doesn't is not based in Chicago and doesn't build bridges.

00:49:06:16 - 00:49:28:17

James Krellenstein

But, you know, Chicago Bridge and Iron buys the old epic which that Shaw which is building it and all of a sudden looks at this project figures out that it's a fixed price contract for the plant and realizes it's a massive massive liability that the company has on its books to finish this project that is not going very well.

00:49:28:19 - 00:49:57:04

James Krellenstein

So this is both Seabee and I. This is both Summer and Vogel at this time are both being built and they're both way behind schedule with a really, really steep road ahead of them. So the commercial issues begin piling up, which is that there is tension between Westinghouse and CB and I, tension building between Vogel, I, between Southern and SCANA as the customers and the consortium.

00:49:57:04 - 00:50:33:16

James Krellenstein

And basically everyone is threatened, beginning to threaten to sue each other. And the first person who sort of really starts going out is Chicago Bridge and Iron basically says, I want out, I want to quit. I get rid of these liabilities. How are we going to do this? And they propose at this point a deal where Westinghouse itself buys the old stone in Webster Division of Shaw slash, CB and I to basically buy it up and sort of assume the entire project themselves.

00:50:33:18 - 00:50:58:16

James Krellenstein

But that means not only is Westinghouse going to be responsible for the nuclear sort of scope of this project and the nuclear engineering scope, they're going to be responsible for the entire delivery and assume all of the roles of the EP. And remember, at this time, Westinghouse is backed up by Toshiba. It's owned by Toshiba. So there's this idea that Toshiba, well, we are very experienced in building stuff.

00:50:58:18 - 00:51:24:18

James Krellenstein

Toshiba is going to take on this sort of all this risk and all the completion responsibility. And that's where we start beginning and get into a little bit of challenges here. How's the Internet connectivity going? But so I don't know. I want to just maybe end at that point. I want to end at this point and basically say, you know, maybe we need a part three because we're about to watch Westinghouse go into bankruptcy.

00:51:24:20 - 00:51:51:23

James Krellenstein

Right? Toshiba almost collapse as a result of this and a rapid succession of two new pieces that are going to be assigned to finish this project up, with the last one being Bechtel, who ultimately finishes this project. And with the project management being ultimately assumed by Southern and Scan abandoning the project middle place. So it gets exciting. But we're already at or 50% complete maybe at these projects and things are going pretty bad right on.

00:51:51:23 - 00:52:12:05

Chris Keefer

James Yeah, I mean, I have a feeling there's going to be more than just an episode number three, but I'm personally loving it and we'll we'll let the audience decide. So if you like this show, get some audience participation, let us know about it on Twitter. Demand Part three James Crown Sign on on the Murphy's Law. Vogel Okay, James, thanks again.

00:52:12:06 - 00:52:19:18

Chris Keefer

I got to run off to an appointment, but it's been a pleasure and we'll get the Internet issues fixed for next then.

00:52:19:19 - 00:52:20:15

James Krellenstein

Cool. All right.