Reactor Safety Margins
Speech to the American Nuclear Society, June 8, 1998
by David Lochbaum
Imagine that you are a parent picking up your child from elementary school. You stand on the corner watching your son or daughter in the cross-walk heading towards you. Suddenly, there's the sound of squealing tires and a racing engine. A car darts out of a side street. Before you can react, it speeds into the intersection. For a few seconds, your child is hidden from sight as the car hurtles past. After an endless moment, the car disappears from view and you see your child—unharmed, but visibly shaken by the near-miss.
In that split second of horror, you noticed that I was driving the car. It's safe to assume that you voiced your concerns about my driving the next time you saw me. Think about your reaction if I had dismissed your concerns with a discussion about the dry pavement conditions, the quick response from rack and pinion steering, the unobstructed vision through my windshield, and my excellent driving record. Chances are pretty good that you might have mentioned that your child could have reacted to my recklessness by running towards the safety of your arms, into the path of my car. Refusing to be drawn into such pointless 'what if' scenarios, I responded, "look, you just don't want to discuss this matter rationally. I missed your kid by a good foot, foot and a half. Get over it." Half of this audience probably would have punched me in the nose. I think that all of you would have wanted to slug me, but I'm guessing that I could outrun half of you.
The moral of this story is that safety margin is not just something that you measure or quantify and just dismiss so long as the number is positive. There's another component to safety margin which I call comfort level. Comfort level is very subjective. It is also a function of perspective. As the driver, I had a high comfort level because I saw your child and I knew I had plenty of room to evade any sudden movements. Standing on the corner, you had a low comfort level, and high distress level, because you only saw a speeding car too close to a loved one.
But, enough about my driving skills. Let's turn to the subject of today's session -- reactor safety margin. Part 50 to Title 10 of the Code of Federal Regulations defines minimum acceptable safety standards for nuclear plant operation. These minimum standards do not result in zero risk, but reduce the risk from nuclear plant operation to an acceptably low level.
The minimum acceptable safety standards essentially draw the upper line for what is called "reactor safety margin." The lower line is crossed when one or more members of the public is exposed to excessive amounts of radiation. A nuclear plant operating at or above the minimum standards can experience a transient or accident with reasonable assurance that public health and safety will not be affected. In other words, the transient or accident will, in all likelihood, cause the plant to drop below the minimum standards and use up some of its reactor safety margin, but there is reasonable assurance that the important lower line will remain intact.
UCS believes that the upper and lower lines for reactor safety margin are acceptably drawn. We do not necessarily advocate raising the minimum acceptable safety standards to provide additional reactor safety margin, although quite honestly we would not strenuously oppose any such efforts. We are truly concerned, however, that too many nuclear plants are venturing too often, and too far, below the upper line into their reactor safety margins.
For example, there have been numerous findings recently that safety-related equipment would or may not have functioned as required during an accident (see the figure). Maine Yankee discovered in 1996 that 16 feet of control cable had been inadvertently cut and removed three to four years earlier. The missing cable would have prevented the high pressure injection (HPI) pump from automatically starting in event of an accident. In 1995, boiling water reactors began correcting a problem with the suction strainers for their emergency core cooling system pumps. Debris such as insulation blown off piping could have plugged the strainers and disabled all of the pumps during an accident. And Millstone Unit 3 recently completed over 70 physical changes to the plant and made several hundred changes to its Final Safety Analysis Report to get the facility back above the minimum acceptable safety standards.
|
Recent Examples of Safety-Related Equipment Problems
|
The NRC and the industry invariably tell the public that these violations are little cause for alarm. These positions talk about defense-in-depth and probabilities. Even if the public understood, yet alone believed, these complex arguments, they would still be unsatisfied by them. These arguments fail to adequately address the comfort level component of reactor safety margins. All of my reasons for thinking that I did not unduly jeopardize your child may be sincere and valid, but they are tossed aside in a heartbeat when someone, or perhaps that inner voice we all have, says, "sure he missed your kid this time, but what about next time?"
Nuclear plant problems are almost always evaluated by the NRC and the industry on the basis of the as-built reactor safety margin. The as-built reactor safety margin is that margin which exists when the plant meets the minimum acceptable safety standards -- the upper line. When a problem leaves the plant at or above the upper line, then that approach is valid. We would agree that the public was protected this time and would probably be protected next time.
But some problems drop plants below the upper line. In these cases, the as-found reactor safety margin must be considered. We define the as-found reactor safety margin as the margin existing with the plant in its degraded condition. For some of the events that have been recently reported, the nuclear plants did not satisfy the minimum acceptable safety standards. Had these plants experienced an accident starting from those degraded conditions, it seems possible that the lower line might have been crossed.
We really should not debate whether the lower line would have been crossed. We all have both a legal and moral obligation to the public to know that answer. Not to guess, not to presume, but to know the answer. When a problem is discovered that indicates a plant operated below the upper line, we must know the as-found reactor safety margin for that degraded condition. We must also know whether the lower line would have been crossed had the plant experienced an accident from that degraded condition. [In other words, would an accident initiating from the degraded plant condition ended up at Point A' or B' on Figure 1?]
There probably will be times when the answer indicates that the lower line would have been crossed. That will clearly be bad news which will not please the public. But by self-identifying and reporting these near-misses, the nuclear industry will regain the public's confidence. The public will then be more receptive to the industry's explanations about corrective actions and recurrence control measures.
If the NRC and the industry continue to discuss problems only in the context of as-built reactor safety margins, then UCS's job becomes very simple. All we have to do is point out, "sure the plant didn't kill you and your family this time, but what about next time?" We may not be quite so blunt, but that will be our basic message.
I challenge the nuclear industry and the NRC to make my job as difficult as possible. Please take the "next time" message away from me. When reporting problems, talk about the as-found reactor safety margin. Not the as-built safety margin, but the actual safety margin for the plant. Explain what would have happened had an accident occurred with the plant in its degraded condition. If public health might have been affected, say so.
The harder you make me work in that way, the easier you'll find it to regain some public trust. Perhaps, they would even come to accept license renewal, stranded cost recovery, and giving nuclear plants some credit for avoiding air emissions.
I began my speech asking you to picture yourself on the sidewalk while I raced past your child in my car. It turns out that you are in the driver's seat, not me. If you start delivering responsible, credible messages in response to nuclear plant events, the public may even stop punching you in the nose.
