San Onofre: Design Caused Decay

CAVANAUGH: I'm Maureen Cavanaugh. It's Tuesday, June 19th. Our top story on Midday Edition, it was a standing room only crowd at last night's public meeting about the troubled San Onofre nuclear power plant. Officials from the nuclear regulatory commission told people gathered at the San Juan Capistrano community center that the investigation has focused in on possible design flaws in the steam generators. It could the cause of the tube failure that has kept the plant shut. My guests, KPBS senior metro reporter, Alison St. John. ST. JOHN: Good morning, Maureen. CAVANAUGH: Nuclear energy consultant Arnie Gunderson. GUNDERSON: Thanks for having me. CAVANAUGH: And Murray Jennex is here, associate professor at SDSU school of. He has extensive experience working at San Onofre. JENNEX: Good afternoon, morning. CAVANAUGH: Who attended and what happened? ST. JOHN: There were hundreds of people there. It was a large hall and before they began they had to ask several people to heave because it was overflowing. They were mostly from what I could see, people who lived relatively within a few miles of the plant, although there were people also there from San Diego. And the purpose of it was to share with the public the initial findings that NRC had made as to the cause of the problems at San Onofre. They had not produced a written report. So there was nothing written for the public or Edison to see. Edison was there as well to respond to those findings. And it was I think it was a sort of contained anxiety. There was no chanting or any sort of attempt to disrupt the meeting. It was very civil. But there were some very good questions asked. And I got some take of some of the questions. One was from jean Stone, and he asked a pretty pertinent question. And this also contains the response from the regional administrator from the NRC. NEW SPEAKER: How is it that thine design changes did not trigger a complete review by the NRC and complete public hearings as required by law? Has the law been broken by California Edison, Mitsubishi or the NRC? NEW SPEAKER: That's a question we're trying to answer as well. Your question is right on the money. We have to be completely clear what happened here, and how did these generators get in the plant over the NRC's review processes, what are the regulations to make sure that this went the wayy woo wanted. And we're still looking at that. ST. JOHN: And Collins to his credit was very, very open to the public, and was not exactly -- was saying this is under investigation. And I did call Barbara Boxer's office. And they are looking at both this issue of how San Onofre got into this pickle, and whether the NRC's procedures are correct. They've got hundreds and hundreds of documents, and they've assembled a team of experts to review the whole thing. CAVANAUGH: Now, everybody seemed to behave themselves very well at this meeting. Was there any statement made by the regulators that caused any kind of reaction in the audience? ST. JOHN: Well, yes, there was one that I think was a bit of a clarification for me. And it caused this gasp from these gentlemen who were sitting behind me, who turned out to be nuclear engineers from the Navy. When they talked about the problems with a simulator set up by Mitsubishi to simulate how the steam generators would operate under this new design, and that's the issue, it's a new design that's different from any other nuclear power plant in the country. So here's what the branch chief had to say. NEW SPEAKER: In our inspections we developed a simplified mathematical hydraulic simulator model, determined that the computer simulation used by Mitsubishi had underpredicted velocities by factors of 3-4 times ST. JOHN: Which is just, like, off the charts in terms of inaccuracy. So the question arises, how can you trust new designs if the simulation models they set up to test them ahead of time are so way off? CAVANAUGH: Murray, let me go to you. That last comment sort of careened over my head a little bit from what the NRC person was saying. Can you take us back a bit and give us a brief reminder of what these generators do and what danger leaky tubes can cause? JENNEX: Well, what the steam generators do is they take the heat, the hot water from the reactor, put them through the tubes, and then the heat from that water transfers across to clean water, which is turned into steam, which is then sent over to power the turbine. So when these tubes leaks, you get primary coolant around the reactor coming out into clean water, which could go into the environment, but which is usually just captured in the containment building or the enclosures right there. CAVANAUGH: And that's what happened in January when that tiny tiny radiation leak triggered this while process the JENNEX: Right. And this whole piping system is monitored for radiation. They do keep assistant monitoring on if. CAVANAUGH: And what the regulators are telling us is when Mitsubishi tested the tubes, they used the wrong amount of pressure? JENNEX: Well, when they designed the tubes. Not tested. And it wasn't necessarily the pressure. It was the velocity of flow through these tube, and velocity of flow through -- with the team and the water on the clean side. That's what causes vibration, higher than expected velocity. CAVANAUGH: One time you told us that tube failure happens in nuclear generators frequently. But last night, the regulators say they have not seen failure like this before. What makes it different? JENNEX: The number of tubes, the excessive and very quick wear. That's what really makes this different. You expect a certain number of tubes over the course of a 25-year lifetime to fail, but not so many so quickly. CAVANAUGH: Arnie, in the first report you issued for friends of the earth, you asserted that design changes in the steam generators had been made and not fully disclosed to the nuclear regulatory commission. What difference would disclosure have made? GUNDERSON: It is true. And yeah, it's interesting that they did basically confirm everything that we put in the reports. This error is a 400% error. And the reason is that they chose a vendor who had never built a stem generator like this before. They had only built a competitor's design. So the computer codes weren't validated for this design. It would have been caught by the NRC if it had been identified earlier in the process because it's not like for like. And the NRC would have identified that oh, my God, Mitsubishi has never built a generator like this, and therefore the codes would have been questioned. I think if Edison had honestly let the NRC know that this was not like forly, this 2006, this problem would have been corrected. CAVANAUGH: Do we know who initiated these design changes? Is this something Southern California Edison requested, or designer Mitsubishi did it on their own? Do we know that? GUNDERSON: Yeah, Edison did -- the ultimate responsibility, the NRC was clear, lies with Edison. Before they chose Mitsubishi, they made a decision that this was going to be a like for like to avoid the NRC's licensing process. So after that, they chose Mitsubishi, and after that, they made some series design modifications. But the strategic goal here from the very beginning was that it would be like for like to avoid any licensing hassles from the NRC. CAVANAUGH: I'm not clear on what you're saying with the like for like. Was that their intention, to build a steam generator that was like the one they had before? Or was it their intention to build something that was very different and get it to pass for the kind of generator they had before? GUNDERSON: What they told the NRC was that it was like for like, that it was essentially identical to the one that lasted 30 years. In fact, what happened once they got Mitsubishi fired up and in design and in construction was a dramatically steam generator than the one that had been installed in there for the last 30 years. CAVANAUGH: Right. Now, none of this has been asserted as actual fact. So let me go to you, Murray, what is -- how do you evaluate that assessment? JENNEX: Well, using the term like for like is probably not quite the correct term. The 50-59 rule they use to decide whether it needs to go to NRC for review is looking at if you're within the design considerations. So you can have some flexibility there. Now, I do agree with Arnie to a point that if somebody had are viewed the output of the codes, you would have looked at the flow velocities and it should have triggered a question. Now, because they're lower, somebody probably made the assumption that that was better. So they were still in design considerations. The fact they're so much lower should have triggered somebody to ask the question, is this accurate? CAVANAUGH: Why would the plant want to change the design on these steam generators to contain hundreds or tubes? JENNEX: Well, it gives you longer life and more heat transfer. Part of the upgrade in San Onofre raised the power limits. So need to have more heat. When you do that on the whole steam generators, that increased the wear. So they started degenerating a little quicker. So I'm sure what they were think think was by having more tubes and transfer it would reduce the wear on each individual tube. CAVANAUGH: Would it also increase the utpout of San Onofre? JENNEX: It potentially could. They'd have to be authorized to do that. But by having more tubes, you actually reduce the wear on all tubes. I'm pretty sure when they were going for was having less of an issue with the steam generator tubes. But because they did a bad analysis in the design phase, they made more of a problem. ST. JOHN: One of the questions last night was to really push Edison to come up with the exact number of tubes that are damaged. And I wondered, Arnie, do you have a sense of why that's so important? It seemed like it might have something to do with Edison's future plans as to how they might try to reopen the plant. GUNDERSON: Yeah, unfortunately those of us who are outside of Edison are not getting this information. But there's more damaged tubes in unit 2 than there is it in unit 3. Now, unit 2 ran longer, so that makes sense. But the most severe damage to tubes is in unit 3. So it basically confirms what we've been saying all along, that these two reactors really have a common root cause, and to treat them differently is really not the right course of action. MAUREEN CAVANAUGH: Now, Arnie, you issued another report for friends of the earth, and in that report, it says San Onofre should remain closed for another 18 months. Why is that? GUNDERSON: Well, are the modifications they're going to need to keep these tubes from rattling require some pretty significant engineering and construction efforts. They don't have to pull them out. But they have to open up a hole in the side of the generator and slide in additional braces to keep them vibrating. It's like blowing across a reed in your mouth and getting it to hum. That's what was happening in these generators. So if they can insert pieces of material between the tubes to keep them rattling, that process takes over a year. CAVANAUGH: Murray, do you foresee such an extensive repair needed? JENNEX: Well, definitely that's the repair that has to get done. You have to support the tubes. And they'll have to rerun the analysis to see how much dampening you could put it in in wear. And he's right, it could take up to 18 months to do that. The other possibility is that you run the plant at a reduced level, but I don't think they're going to. I think the report that they found some manufacturing issues with the tubes themselves would lead them to say it's probably not a good idea to even run it low power until they do do the fix. So it could be up to 18 months. CAVANAUGH: This sounds like a terribly extensive fix. Do we have any idea how much this might cost? JENNEX: Well, and to whom. ST. JOHN: Right. JENNEX: I would suspect this would have to be a cost borne by Mitsubishi, and we are probably looking at a $1 million or so. CAVANAUGH: Is it possible that this whole process with possible sanctions against Southern California Edison, if they find out that the proper paperwork was not issued to the NRC, is it that it could be more cost effective to shut down than to get back up and running? JENNEX: $100 million is not cost effective to shut down. This is a plant that's about $2 billion, it's just -- and it makes so much money, and it provides so much of the power base load here in Southern California, even though that's a large amount of cost, it's still cost effective to run these plants. >> It's interesting that you say the cost would fall on Mitsubishi. One of the questions last night was if the rate payers have to pay, is there a point that it becomes to expensive? And Edison said it will come before the utilities commission to decide how much of the burden of this the rate payer would have to pick up. JENNEX: And probably the rate payer would pay, and then get reimbursed, once they decided how much fault was that Mitsubishi and Edison. CAVANAUGH: And Arnie, what are your thoughts on this? GUNDERSON: Well, I think that there's some limitations to ability in the Mitsubishi contract that limit the liability to about $130 million. They may have already ponied up $130 million in the process we've done so far. I think at that point, they may have to sue. And a lot of the parties -- a lot of lawyers are going to get rich before this problem ever gets solved. CAVANAUGH: Alison, what is the NRC going to do next? ST. JOHN: Yes, they're going to produce a report in writing. It'll be useful to have something in writing that everyone can refer to within the next 30 days, then they're going to have a series at these public meetings. I think they're very committed to at least appearing to be Franz parent by allowing the public to ask questions. CAVANAUGH: Okay then. I'm sure we're going to be revisiting this particular issue. Thank you all. JENNEX: Thank you. GUNDERSON: Thank you.