APR: your source for nuclear news and analysis since April 16, 2010

Friday, April 1, 2011

The battle over blame begins: Fukushima Daiichi.

The results of the nuclear accident at the Fukushima Daiichi nuclear generating station in Japan, which was caused by what is now known as the Great East Japan Earthquake, are now beginning to be thrown around in the media in terms of blame, of excuse, of implication, of supposition. We knew this would happen - it happened with Three Mile Island, and to some extent with Chernobyl although really since everything was state owned in the former USSR it was impossible to blame private companies. So let's get ahead of the crowd here with some background information to figure out just what the primary players in this whole thing are responsible for and what they're not responsible for.

REACTOR VENDOR. This is the company responsible for the overall design and construction features of the primary reactor plant. This includes the reactor core. This company will also be responsible for all nuclear calculations regarding operation and safety, operating and safety procedures, and direct support to the eventual owner/operator entity. United States based reactor vendors for large commercial nuclear power plants have included the "big four," namely Westinghouse, General Electric, Babcock & Wilcox, and Combustion Engineering. Any company acting as a reactor vendor must have a great deal of internal infrastructure to support such an operation, and in fact each of the above operated its own nuclear energy research center at one time or another.

ARCHITECT / ENGINEER. This is the company responsible for the overall layout of the entire reactor plant including the primary containment and the turbine building. Sometimes a company will handle everything in house, and sometimes will subcontract actual design work for portions of the plant to another design firm.

CONTRACTEES / SUPPLIERS. These are very numerous, and include many subcontracts for all sorts of reactor parts. For example, while the original Shippingport Atomic Power Station was the responsibility of Westinghouse, it contracted the actual manufacturing of the reactor pressure vessel to Combustion Engineering. A number of General Electric plants were built with pressure vessels made near site by Chicago Bridge & Iron. While the reactor plant for the N.S. Savannah was the responsibility of Babcock & Wilcox, three of its main coolant pumps were made by Westinghouse while the fourth was made by Allis-Chalmers (also a reactor vendor itself at the time.) Of course this list extends to the turbine building as well, and it's not unheard of to find a Westinghouse PWR plant with a GE turbine-generator. And there are myriad other systems and parts as well.

CONSTRUCTORS. The firms subcontracted by the reactor vendor and / or the architect-engineers to actually erect the plant, the containment, the turbine building and install equipment. Some smaller equipment is installed by third parties or by its OEM company.

OWNER-OPERATOR. The utility company that owns the plant, for whom it will produce electricity, and who employs the personnel that will operate and maintain the plant.

Now, as you can see, there are a lot of parties involved. And when the blame game starts up, well, there are going to be a lot of potential targets. Let's pick out the ones we can identify for Fukushima Daiichi and just give a brief sample of what has been said about each already from both sides of the argument.

REACTOR VENDOR: General Electric, and partners Toshiba and Hitachi. While what GE says about its BWR plants with Mk I containments is true -- that is, that they have been in service for about 40 years and still get licensed and extended and extensively examined and approved -- it is also true that one of the reasons that the Mk II containment was developed was to eliminate a chance that gaseous fission products could bypass the suppression pool and get directly to the atmosphere inside the reactor building. Also on GE's side is the fact that these plants were subjected, it is now known (just today) to earthquake forces that exceeded (in at least two plants) the original contractually stipulated earthquake forces they were designed to withstand. You can't blame GE for meeting the original requirements that many years ago vis a vis earthquake forces for what happened this year.

OWNER-OPERATOR: Tokyo Electric Power Company, or TEPCO, who apparently in the face of this disaster did prevent total core meltdown and release but could not stop severe core damage. TEPCO's people have been under absolutely unprecedented stress dealing with three simultaneous reactor accidents on one site in close proximity and four spent fuel site incidents at the same site at the same time. Probably no company could have been ready for this; but then again if any should have been, it should have been TEPCO, which operates one of the largest fleets of reactors of any one utility companies in the world. TEPCO's operations at the site have been at times aggressive, at times cautious, at times right by the book and at times clearly ignoring the book. Certainly if failure to act properly by operating staff is proven, this will blacken the eyes of TEPCO but if TEPCO can prove design flaws led to the core melt then the battle is on again and likely to spread to other plants elsewhere of the same general design.

We must remember in all this that the magnitude of the disaster here is beyond the DBA or Design Basis Accident. Yes, the plants did survive an earthquake above or at their designed limit, at least from what we know of the short time between quake and full scram, and the tsunami wave arrival. The total loss of cooling is part of the assumed SBO DBA or Station Blackout, but I can be pretty safe in guessing that if the control rooms had five feet of water in them nobody calculated against that. The plants were designed for about the most serious earthquake that had been encountered in recent time (and that's WAY oversimplifying the process for seismic consideration) and the most severe tsunami it would generate.

Nuclear plants only make the news when the worst-case scenario is exceeded by a situation or disaster previously inconceivable. No one would have suspected the vast range of equipment deficiencies (some not the fault of the vendor, but broken items unfixed) and training errors and personnel errors that led to the Three Mile Island accident, but it happened. Vast acres of paperwork were generated after this, which led to mountains of legislation and many years of action so that the possibility of recurrence is absolutely tiny. In Japan, the worst known earthquake and tsunami scenario, against which the plants were theoretically designed, was exceeded by a once in a lifetime, perhaps once every 250 years type of event that is still producing figures and estimates astonishing to geologists, seismologists and oceanographers. These events in the nuclear power field, like many before them, set new levels of expectation, of safety, of prudence in design not considered necessary before. They expanded the envelope. What is most impressive is that, in the field of Western nations' nuclear energy, it took an act of God to cause the first nuclear reactor accident since 1979.. a period of three decades of vast employment of nuclear reactors numbering many hundreds, throughout many nations and oceans all over the world. When the blame begins, we must keep this fact squarely in focus.

4:12 PM Eastern Friday 4/1


  1. As usual very informative; but unusually, I take issue with a couple of statements.
    "You can't blame GE for meeting the original requirements..."
    Is it clear that the quake itself caused the reactors' containments to fail, and not the pressure from the later explosions? If it was due to the quake, fine, nobody designed for that. But if it was the explosions, that was predicted back in the 1970s about the Mk I design, wasn't it? And that would be GE.

    "The plants were designed for about the most ... and the most severe tsunami it would generate."
    The NY Times reported that there has been very good science out for several years now predicting tsunami's on >30' from quakes of 7.5 magnitude. 7.5 is reported to be within the size the reactors were designed for. Yet TEPCO did nothing to improve their 14' seawall. That seems more egregious than not predicting a mag 9 quake.

    I hope you will comment on these points as they are based on assertions that have at least appeared in reputable media. And thanks for your diligence in reporting all this.

  2. We won't know if it was due to the quake for a while, and may not ever considering the damage. However, my point was that GE met the contractual obligation as built originally .. meaning that what they were required to do, they did. Go back and read my statement again- I am on your side about the Mk I containment. Note my mention of FP's bypassing the suppression pool -- that would hold for radiolytic and metal-water hydrogen too, and as I said that's part of why the Mk II was designed. HOWEVER, if anyone is going to say that GE should have replaced the containment (and we all know that's a practical impossibility) then that's just absurd. That would be TEPCO's call as owner-operator.

    I don't know what TEPCO has been reviewing in the last few years vis a vis improvements to the protection from tsunami. What we do know from its constant reporting in the news is the set of original standards required for the plant. If that had been the United States, those requirements would have been made public and opened for public debate and discussion. I do not know if such was done there, and even if it were it wouldn't likely be in English. We'll have to wait for better information to see. My point there was that TEPCO and GE did in fact pick a point that seemed rational to defend against given the best data when the plants were built.

    No one is going to get a free pass from me in this situation, but I'm also not going to start throwing corporations, governments, or utility companies under the wheels either. The reason for this article in the first place is to STOP the blame game before it starts, so that reasonable interpretation of how we got to this point can be made calmly and logically. We must strive to avoid hysteria and name-calling which does no one any good.. and we know how helpful the media was after Three Mile Island vis a vis that subject (and that's not at all.)

  3. Would you care to comment on the risk involved in continuing to operate the remaining 48 reactors located on the Japanese coastline?

  4. To be clear, I don't blame GE re the original requirements either. The issue with both the Mk I containment and Fukushima's level of tsunami preparedness seems to be, what happens when new science raises issues with an old plant? I guess it's a regulatory matter first and foremost. But suppliers and utilities may be greatly remiss if they obstruct or obfuscate in these cases. (And it would sure be interesting to look at the power outage scenarios for US Mk I reactors and see if they are realistic in light of Fukushima. Have you got experience with those?)

  5. @david... The risk involved in operating the rest of the reactors in Japan is probably roughly as great as the percentage chance that another event, greater than the design basis accident against which any given affected plant is designed, will occur before plants are modified to meet a new standard. Chubu Electric's Hamaoka plant is leading the way in upgrades already. This might sound like an evasive answer but we can't quantify risk, really, of events as violent as the recent tsunami because one this large has not happened in Japan in modern times.

  6. So the chances of another accident in Japan, or anywhere, are the same or greater than the one that has occurred.

    This is unacceptable and nuclear power should be abandoned as a bad idea.

    No other form of generation has these possible consequences - exclusion zones, food contamination, health effects for generations - ITS NOT WORTH IT.

  7. @david... No, the chances are very high that a similar natural disaster will NOT occur, so that the chances of another such accident are very remote.