Walking a Nuclear Tightrope: Unlearned Lessons of Year-plus Reactor Outages

[kristopher]

From The Union of Concerned Scientists:

Walking a Nuclear Tightrope:
Unlearned Lessons of Year-plus Reactor Outages

No U.S. nuclear power reactor has experienced significant core damage since the partial meltdown at Pennsylvania’s Three Mile Island (TMI) in 1979. This passage of time suggests nuclear power is safer today, but is it safe enough? A car speeding through a school zone at 90 miles per hour is safer if it slows to “only” 75 mph, but it isn’t safe enough. Likewise, nuclear power may be safer today but it is not safe enough.

In the 27 years since the TMI meltdown, 38 U.S. nuclear power reactors had to be shut down for at least a year while widespread problems within each plant were fixed and safety margins were restored to minimally acceptable levels. Including those prior to TMI, 51 reactor outages of a year or longer have occurred. While these reactors shut down before they experienced a major accident, we cannot assume our luck will continue.

Federal regulations require nuclear plant owners to have quality assurance (QA) programs that find and fix problems in a timely manner. But the recurring theme of the year-plus outages has been owners failing to find serious problems and/or failing to properly fix them, leading to an erosion of safety margins. The accompanying theme has been the Nuclear Regulatory Commission (NRC) either being unaware of the QA program deficiencies or knowing but being unable or unwilling to enforce the applicable federal safety regulations.

Nuclear power is clearly not safe enough when so many reactors have to shut down for so long to restore safety to the minimum level considered acceptable. The recurring year-plus outages constitute prima facie evidence of how far safety levels fell below minimally acceptable levels, making nuclear power far more dangerous and costly than necessary. The chronic violations of federal safety regulations must stop before the increased safety risks yield an even higher cost – human tragedy.

Our review of year-plus reactor outages found a silver lining. The extended outages fell into three cause categories: (1) damage recovery outages resulting from an event, such as an accident, causing extensive damage that took a long time to repair, (2) component replacement or repair outages resulting from degradation to a major component, such as the steam generator, that forced its replacement or repair, and (3) safety restoration outages resulting from cumulative, systemic degradation of many reactor components, which flawed QA programs failed to prevent, detect or fix. The figure strongly suggests that damage recovery outages and component replacement/repair outages have been eliminated – no such outage has occurred in over a decade – due to NRC attention to these problems. The lack of effective NRC attention to the third problem allowed safety restoration outages to proliferate. The silver lining is that proper NRC attention to the causes of safety restoration outages will likely be as successful in curbing their recurrence as proper attention to the other causes has been.

UCS recommends the following steps be taken to improve safety levels at nuclear power plants:

1. The NRC must significantly improve its assessment of QA programs at nuclear power plants.

The most crucial reform is in the way the NRC evaluates QA at nuclear power plants. Federal regulations
require power plants to have QA programs that find and fix problems at the plant. If NRC inspectors find
problems when they visit, it means that the plant’s QA program has failed. Yet today when NRC inspectors find
problems, that information is merely entered into the very same QA program just demonstrated to be flawed.

2. The NRC must alert plant owners about non-hardware problems that have reduced safety levels.

The NRC currently alerts plant owners about hardware problems identified at a specific plant. During our
review of the year-plus outages, we did not find a single instance in which the NRC had alerted plant owners
about programmatic breakdowns—problems with procedures or training—that led to broken equipment and,
ultimately, extended outages. The lessons of non-hardware problems must also be communicated so that they
can be incorporated into procedures and training at all nuclear power plants.

3. The NRC must expand the scope of its oversight efforts when programmatic breakdowns at a nuclear
plant are identified to ensure that other plants operated by the same company do not experience
declining performance while the company focuses on restarting the troubled plant.

Companies operating multiple nuclear power plants seldom shut down all of their reactors when one plant
experiences a programmatic breakdown of the kind that has caused most of the year-plus outages. Yet
programmatic breakdowns are unlikely to be confined to only one facility owned by a poorly managed
company. When programmatic breakdowns are identified at a plant, the NRC must (1) determine whether other
plants operated by the same company have the same problems, and (2) ensure that performance does not
deteriorate at those plants while the company focuses on restarting the troubled facility.

4. When longstanding problems are identified at a plant, the NRC must require the owner to determine
why its testing and inspections failed to find the problems earlier and remedy those failures.

Our review of year-plus reactor outages revealed the fact that numerous safety problems “invisible” to plant
workers and NRC inspectors during years of tests and inspections magically materialized after an extended
outage caused a sea change in how that plant was perceived. The NRC must break the longstanding pattern of
NRC inspectors and plant workers repeatedly overlooking safety problems until operations grind to a halt.

5. The NRC must develop a central repository for all current information about plant safety levels,
potential safety problems, and generic safety issues so that all agency employees have access to the
same data when making regulatory decisions.

The post-mortems on year-plus outages conducted by the NRC, the NRC’s inspector general, and the General
Accountability Office show that the NRC had known about many of the problems but had not “connected the
dots” to see the picture of a plant headed for trouble. Why? Because the dots resided in numerous places within
the agency: some with regional staff, some with headquarters staff, and some with a different program office.
There is no excuse today for not making all of the dots readily available to all NRC staff.

6. The Congress must require the NRC to provide periodic status reports on progress made towards
implementing these five steps.

The NRC has much on its plate (e.g., security issues, license renewals, new reactor design certifications, etc.). It
is critical that the Congress provide active oversight of the important work being conducted by the NRC,
especially progress on reforms intended to prevent future safety problems at U.S. nuclear plants.

[AlecBGreen]

1. BS

"Nuclear power is clearly not safe enough when so many reactors have to shut down for so long to restore safety to the minimum level considered acceptable. The recurring year-plus outages constitute prima facie evidence of how far safety levels fell below minimally acceptable levels, making nuclear power far more dangerous and costly than necessary."

First, a question: who says these shutdowns are "to restore safety to the minimum level considered acceptable." This sounds more like a diatrabe than a coherent arguement.

A few points:

1) Just because its takes a while to fix something doesnt mean its unsafe.

2) Routine maintenance is performed in all power plants. Why should nuclear be any different?

3) Compare a 1GW nuclear plant to 500 2MW wind turbines. Do you think the one plant will take more time to maintain than hundreds of 100-meter towers?

:shrug:

I think there are better alternatives to nuclear power. This article does not make a strong case in support of that.

[kristopher]

3. The article isn't arguing for or against nuclear power

The document in the OP was produced by the Union of Concerned Scientists, a non-profit public policy education group that is not given to "diatribes" and is well respected for its objective science based approach to problem analysis.

The article is arguing that the current regulatory regime is lacking in performing (surprise) its oversight function.

A separate piece of evidence is the utilization rate of Japanese nuclear plants, which is considerably lower than that of US plants. While there could be other interpretations of this data, the one most probable is that the Japanese are less liable to allow the profit motive to dictate matters of operational safety. Comparing their excellent record of engineering excellence and long range planning to the US record of slipshod operations and cutting "safety" corners to enhance profitability makes the differing utilization rates troubling.

[NNadir]

2. Um, the capacity utilization of nuclear reactors is public record on the EIA site.

We don't need the circle jerk of anti-nuke websites, including the misnamed "Union of Concerned 'Scientists'."

Here's the data, and the capacity utilization of nuclear power can be directly calculated - if one can calculate, something the anti-nuke diciples can't do - from the figures.

Generation:

http://www.eia.doe.gov/pub/international/iealf/table27.xls

Installed capacity:


http://www.eia.doe.gov/pub/international/iealf/table27.xls

The latter figure, 99,209 MWe of installed capacity - capacity which the anti-nukes are trying to vandalize because they are paranoids who can't think straight - represents, at a theoretical capacity utilization of 100%, 86400 seconds a day, 365.25 days a year, 3.13 exajoules of pure electricity. (The primary energy production is correspondingly larger reflecting the 2nd law of thermodynamics.) The previous figure, 787 billion kwh, (2006) represents 2.83 of exajoules of pure electricity.

Thus the capacity utilization of nuclear facilities in the United States was as of 2006, 89.1%

This compares with 20% capacity utilization with yuppie toy solar PV power, and 25% with yuppie hyped wind power - figures that can also be calculated from EIA, IPCC, or IEA data.

I note with all due contempt that we are having this bullshit lazily posted on this site with zero critical thinking by a member of the "renewables will save us" set who are in my opinion, status quo enablers.

As a result of such uncritical thinking the dependence of the United States upon dangerous fossil fuels - about which the anti-nukes couldn't care less - is rising, not falling.

http://www.eia.doe.gov/cneaf/alternate/page/renew_energy_consump/table1.html