 |
| New Steam Generator Support Plates picture courtesy of Areva |
When I was project manager at the Electric Power Research Institute (EPRI), my speciality was corrosion prevention in steam generators. Due to this background, I am very interested in the problems at the San Onofre Nuclear Generating Station (SONGS) new steam generators. I care, even though SONGS is far away from Vermont.
SONGS problems are due to vibration, and my expertise was materials and water chemistry. But still. The Steam Generator Project Office was a pretty small group at EPRI: there were about ten of us. We all learned about everything concerned with steam generators. We held formal seminars for our utility sponsors, and we had informal talks and seminars within the group. EPRI wouldn't have asked me to perform a vibrational analysis (I'm a chemist), but we all discussed the steam generator research results.
Steam Generators at SONGS
SONGS recently replaced its steam generators, but the new ones didn't work well. They showed a great deal tube-to-tube wear. Due to vibration, the tubes bump into each other and fret each other.
The steam generators at Unit 3 suffered a lot of tube-to-tube wear, and one tube even had a small leak. Due to the leak, the operators shut the plant down.
Unit 2, on the other hand, had only two tubes with indications of significant tube-to-tube wear. These tubes have been plugged. Five hundred more tubes were plugged as a preventative measure. Steam generators have huge numbers of tubes (more then 9700 tubes per generator at SONGS). When a generator is built, it is assumed that a certain percentage of tubes will be plugged over the lifetime of the plant. The tube plugging at Unit 2 is a lot for a young plant, but hardly near any kind of limit. Five hundred tubes is about 2% of the tubes.
 |
| Aerial view of SONGS from Wikipedia |
I think the most complete description of the tube situation is in the letter from Southern California Edison to the NRC, explaining how the NRC criteria for restarting Unit 2 will be addressed. More information can be found at the SONGS site: the Information Package is especially readable.
Vibration at SONGS Or "I Told You So"
Luckily, vibration problems are usually comparatively easy to fix. Change the speed of the liquid, and the vibrational mode will change. It can be as simple as that.
In April, I wrote a guest post at Atomic Power Review: San Onofre and Steam Generator Design. In this post, I wrote:
The root cause analysis will come up with something vibration connected.
To cure it, some kind of derating of water flow (primary side and/or secondary side) with ameliorate the problem well enough to keep the generators in operation while the owners and the manufacturers sue each other. The generators will be replaced early.
Well, the last part hasn't happened, but the first part is on its way.
In July, a post at my own blog called San Onofre, Gundersen and Vermont Yankee received quite a few comments. In the comment section, I predicted that San Onofre would be back on-line in a few months---except that it might take a long time to convince the NRC that the new vibrational models were correct. This also seems to be exactly what is happening.
Also, starting Unit 3 will be harder much harder than starting Unit 2, because the damage at Unit 3 was greater when discovered. My crystal ball was good, but not perfect.
Am I Being Boastful?
Well, yes I am. The vibrational problems at Unit 2 are solvable, just as I predicted.
I am tired of those who predict doom claiming to always be right, while people who know something about nuclear power, and people who predict good things on occasion--we don't bother to take any credit.
In other words, I'm not boastful. I am just standing up for the nuclear industry!
Oh. Okay. I'm boastful. In other words: "I told you so."
-----
More Resources
- Atomic Power Review on how steam generators are designed
- Atomic Power Review on SONGS, with good links
- Atomic Power Review, my guest post on Steam Generator development at EPRI
- Rod Adams on whether nuclear power has been captured by overly-conservative group-think.
- SONGS website on issues, with many links to documents.
- Find out about NRC meeting yesterday by searching on twitter for #SanOnofre
I have to hand it to you, Meredith; you called it! Bravo. I was happy to see you tweeting along with the meeting last night, and I look forward to your future posts on this subject.
ReplyDeleteWell done Meredith. Have a great week! Mary
ReplyDeleteModels are powerful tools, but are not perfect, as demonstrated by the unexpectedly high wear on the steam generators. Failures such as this drive engineers bring their models closer to the perfect point where the model and reality do the same thing. The best model of reality is reality itself. It is time to restart at least one of the San Onofre reactors at 70% power to see if the improved model is correct. The safety risk is tiny, even if a steam generator should leak again. The risk is much smaller than the danger caused by the normal operation of the fossil fuel power plants providing the replacement power.
ReplyDeleteI think the following bit is particularly prescient, "while the owners and the manufacturers sue each other."
ReplyDeleteFrom everything I've seen, it appears this problem was the fault of the manufacturer (I believe it was Mitsubishi?)
If I were SoCalEd, I'd be prepping a lawsuit against Mitsubishi to A) get a new Steam Generator free, so I could run at 100% power instead of 70%, B) Installation related costs for putting in the new generator, C) all costs associated with the plant outage (regulatory costs, replacement power costs, loss of revenue, etc).
It just seems to me that if the steam generator was faulty, then SoCalEd should be able to hold the manufacturer liable for all those costs.
So, lawsuits it will be, I suspect - and this is no 'frivolous' lawsuit - this is the sort of real problem that courts are designed to address.
Well, SCE has a share of the blame. The original S/G u tubes were alloy 600. They wanted the new ones made out of alloy 690. The problem? 690 does not have NEAR as good heat transfer as 600. Since space in containment is limited, the size and footprint could not change. Tube HAD to be added, and thinner. These were 4 first of a kind S/G's. You just don't know what you don't know. MHI has made lots of S/G's, and is good at it. Hindsight is 20/20, their computer codes did not even anylize for the vibrations that did occur....
DeleteThere is a sadly amusing twist to this story. Scuttlebutt in the industry is that Mitsubishi "bought" the contract to supply the steam generators by offering a significantly reduced price compared to its competitors. It then made some design decisions aimed at reducing the production cost - shifting to a "simpler" set of tube support structures that did not provide exactly the same kind of movement restriction.
ReplyDeleteSupposedly, the reason Mitsubishi was so interested in doing this work for SCE was that they wanted to establish a larger US presence in support of their effort to license and sell their 1700 MWe PWRs.
That effort must be judged as a failure on many levels.
This should not, by any stretch of the imagination, be a negative reflection on nuclear technology. It is one more sad story of a failed business strategy that was made worse by an overemphasis on price over value.
I wonder how the PUC oversight played into this saga? Did they help push SCE to go with the low bidder?
Well..... now they are shut down for good.... such a shame!! I have spent 26 years in nuclear, all in operations. I am deeply saddened by the whole US Nuclear situation!
ReplyDelete