geoharvey

Running to Failure

(This article first appeared in the NEC Newsletter, published by the New England Coalition on Nuclear Pollution.)

The most important question is not whether Vermont Yankee will close; it will certainly do that.  The most important question is not even when it will close.  The single, overwhelmingly important question is, “How will it close?”

There are two ways a power plant can shut down.  One is planned retirement.  The other is unplanned failure. Which way depends to some extent on how it is run.  If its operating mode is run-to-retirement, it has a chance of retiring as gracefully as it can.  When a plant is retired, the employees, community, and other stakeholders can be prepared for the change.

When a plant is run to failure, however, the specifics of shutting down are almost entirely unpredictable.  Employees will unexpectedly get pink slips as the plant will go offline.  But since we cannot know in advance why a nuclear power plant will fail, we have no way of predicting the economic, environmental, or human impact.

Looking at the history of nuclear power in the United States, we see one striking fact:  Nuclear power plants are almost never run to retirement.  Those that have gone offline have nearly all done so for operational or economic reasons – they ran until they failed.  Utilities that ask to keep plants running beyond their designed service life have never had a license renewal application denied.  The Nuclear Regulatory Commission and the utilities seem to agree on one essential fact: as long as it is economically feasible for a plant to operate, it will.  In other words, almost all of our 104 commercial nuclear power plants are operating in run-to-failure mode.

In the old days, the Atomic Energy Commission promised there was only a 1 in 10,000 chance of an event that would damage the core of a given reactor in any given year.  There was an assumption made in this calculation, which was that the reactor was run as it was designed to be run. But the design basis assumes a reactor would retire, and not be run until it failed.  The chance of core-damaging failure has never been recalculated based on a run-to-failure operating mode.  So the industry is not currently gambling at 1 in 10,000, but on uncalculated odds, with the potential greatest losers being the public.

Vermont Yankee is already operating outside its design basis, producing more power than its designers ever intended.  Now, continued operation compounds the problem, as the overloaded reactor continues past the end of its intended life.

We can be sure of one thing: if we persist in operating in a run-to-failure mode, the probability of failure is 100%.