06/23/03

The "near-death experience." It's what Nuclear Regulatory Commission officials still call the bruising week in early June 1998, when the future of the agency that oversees the na tion's nuclear plants suddenly seemed in doubt.

In the months leading up to its brush with fiscal mortality, the NRC had become as invigo rated a regulator as anyone could remember. Stung by news of a string of long-stand ing, uncorrected design lapses at the Millstone reactor in Con necticut, the normally temper ate agency went on a tear.

After a decade of talking tough about rooting out unad dressed design flaws and un certainties systemwide - prob lems that have been at the heart of the nuclear industry's biggest debacles, from the par tial meltdown at Three Mile Is land in 1979 to risky fuel-stor age practices at Millstone in the 1990s - the NRC was doing something about it.

Thirteen unlucky facilities landed on the NRC's "watch list" for extra scrutiny in 1997. Special inspection teams formed. The harder the agency looked for so-called design-ba sis issues, the more it found. Regulators shut down nuclear reactors; their operating com panies' profits took a nose dive.

The NRC insisted the extra scrutiny was necessary to pro tect the public. Plant owners criticized the initiative as costly and unnecessary.

That's when the powerful in dustry began what insiders call "pushback."

The NRC found itself in the gunsights of the Senate subcom mittee that controls the agency's budget. Armed with an efficiency and cost-cutting study done by a nuclear industry consultant, the subcommittee's chairman, pro- nuclear Republican Sen. Pete Do menici of New Mexico, declared that the NRC could get by just fine with a $90 million budget cut, 700 fewer employees, and a greatly reduced inspection effort.

The agency got the message. The beefed-up inspections aimed at resolving design concerns ended soon after the budget showdown.

The NRC's new view? Most cases in which reactors were found to be operating outside their design limits posed little or no public risk. There were plenty of backup systems and lots of safety margin. With many plants in their second or third decade of operation, the majority of design issues already had been identi fied. The NRC's regular inspec tions would find the rest.

But it didn't happen that way at Ohio's Davis-Besse, where last year plant workers discovered significant design deficiencies or questions involving five vital emergency systems. These design issues were not found in the course of regular or even special NRC inspections, but because Davis-Besse engineers were in vestigating an unrelated problem of a rust hole in the reactor's lid.

Plant engineers said some of the defects had existed unde tected since the Toledo-area-re actor's construction in the mid- 1970s. The equipment in ques tion still may have worked dur ing an accident, but there was no way for the utility or federal reg ulators to be sure.

NRC and nuclear industry offi cials contend that the discoveries at Davis-Besse are an aberration and not the resurgence of an old problem, but critics aren't buying it. They say the agency isn't find ing design issues at other plants because the NRC has stopped looking aggressively.

"The NRC has tried to ignore design-basis problems at Davis- Besse because they [the agency] supposedly had programs in place" to catch them, said Green peace's Jim Riccio, author of a study by the watchdog group Public Citizen that documented more than 500 instances be tween 1996 and 1999 in which U.S. reactors were operating out side design specifications. The agency "is taking credit for safety systems that may not work. Pull this string, and the emperor has no clothes."

With a team of four or five in spectors, "I could find enough design-basis issues in the first day at any plant in the country to shut them down," asserts nuclear safety engineer David Lochbaum, a former reactor worker and con sultant now with the Union of Concerned Scientists. Faced with pressure from the nuclear indus try and the Senate committee's ultimatum, "the NRC caved and stayed caved," he said.

Long delays

Agency officials insist that they have dealt firmly and responsibly with the design-basis issue. As evidence, they point to the NRC's stern 1996 dictate that all nu clear utility chief executives sub mit, under oath, information that would show they had pro cesses in place to assure that their reactors were operating as designed.

The reports were supposed to be made within four months. Nearly seven years later, though, some plants still haven't wrap ped up their final responses to the NRC, and the agency doesn't seem especially concerned.

New York's Indian Point asked for and got three multiyear ex tensions to finish its design veri fication work, the last of which isn't due until the end of this year. An investigation by the NRC's inspector general found that agency staffers considered the delays "reasonable" and felt that the company was making "steady but slow progress."

As of late February at FirstEn ergy Corp.'s Davis-Besse, an NRC inspection found that the plant still had not resolved 200 of the 1,000 design deficiencies its re views had turned up in response to the 1996 letter. The lapse earned Davis-Besse executives an NRC scolding about lack of prior ities, but no regulatory penalty.

"The work wasn't done, and it should have been," FirstEnergy nuclear division Executive Vice President Gary Leidich recently acknowledged to the agency.

Senior agency officials weren't immediately able to say how many other plants might also still be working on their re sponses to the 7-year-old letter. But they say the number of unre solved design issues at Davis- Besse can be misleading.

"I've been on inspections where inspectors come back and say there are 300 discrepancies more than a year old and it sounds bad," said Stu Richards, chief of the NRC's inspection program branch. "But when you look at what those discrepancies are, they're minor items that, even if they're not corrected in a short period of time, it's really not going to contribute to risk. And it's not an issue we want to spend any kind of time on, be cause the payback is not that great."

From the nuclear industry's perspective, reactor operators al ready have ferreted out the worst design issues; those that remain are mostly low-significance items that don't merit the million-dol lar cost and monthslong aggrava tion of a full-blown design-basis review at every plant.

Plant managers "are often in volved in fairly esoteric issues with the NRC when their time would be better spent focusing on the day-to-day safe and reli able operations of their plant and performance of their people," warned industry leader Zack Pate in 1997, the last time the NRC geared up for increased de sign-basis scrutiny.

Pate, who at the time headed the Institute of Nuclear Power Operations, the industry's re search group, made the remarks to the trade publication Inside NRC.

Existing government and plant inspection routines are adequate to pinpoint the isolated design- basis issues that remain, indus try officials say. Such discrepan cies, usually minor, will inevita bly be found as the nuclear industry matures and engi neering reviews grow more so phisticated, they contend.

"When Davis-Besse was de signed, we all had slide rules, you know, and we have come a ways since then," Lew Myers, chief op erating officer of FirstEnergy's nuclear division, told NRC offi cials who were quizzing him about why design issues are still emerging there.

Most other plants shouldn't have the degree of design ques tions unearthed at the Ohio plant, according to the industry's trade group, the Nuclear Energy Institute.

"I think the situation at Davis- Besse is kind of unique," said NEI engineering director Alex Marion. "The rest of the fleet doesn't have the lack of imple mentation programs they had."

Digging deeper

To an extent, the NRC agrees with the industry's position.

Although recent inspections at Davis-Besse have raised some significant design questions that previous agency and plant checks didn't disclose, the design issues there "are not pervasive," said Jack Grobe, chairman of the NRC panel overseeing the plant's re habilitation.

And compared with the rest of the nation's reactors, "I would like to tell you it [Davis-Besse] is an anomaly," said Gene Imbro, chief of the agency's mechanical and civil engineering branch. "The expectation would be that by this time, after the efforts that went on in the early 1990s, there would not be any major design issues still outstanding."

Improved training for reactor operators, multiple layers of backup equipment and the safety margins built into the plants all combine to minimize the chances that a design flaw will cause a major problem, the agency offi cials said.

At first glance, the NRC's most recent examination of design is sues, published in November 2000, seems to support that view.

The percentage of design-basis issues deemed capable of causing a serious reactor accident stead ily fell from 1991 to 1997, the re view found. In 1997, the last year studied, 22 percent of the design issues that investigators screened had potentially significant risk. The majority posed minimal or no danger.

But other findings were more worrisome.

The NRC study turned up far more instances of nuclear plants operating outside their design specifications than even the study by the anti-nuclear Public Citizen. The watchdog group identified more than 500 cases between 1996 and 1999; the NRC documented 569 in only one year, 1997. Overall, there were more than 3,100 from 1985 to 1997, or an average of about 240 per year among the nation's 100-plus reactors.

"I suspect that [design defi ciencies] permeate most of the plants, though not necessarily to the degree of what has been found at Davis-Besse," said Hal Ornstein, a retired NRC senior safety analyst whose department studied design-basis issues be fore the agency abolished it in 1999.

The NRC review found that de sign flaws tended to be concen trated in the safety systems that in an accident would be relied on to keep the plant supplied with electricity, the reactor core cool, and the reactor isolated from the environment.

As with the latest crop at Da vis-Besse, more design issues re vealed themselves through equipment problems or hap penstance than because of NRC or plant efforts to find them.

And finally, as the number of hours that NRC inspectors spent at a plant increased, so did the number of design issues that they or the utility's engineers dis covered. As Riccio points out, the NRC's own study found that "the more design-basis issues you look for, the more you find."

Influential industry

So why hasn't the NRC looked more?

The unavoidable answer is be cause of the powerful resistance and lobbying efforts of the nu clear industry.

When the NRC has tried in the wake of past design fiascoes to institute stricter and more wide- ranging checks, the industry has marshaled its cost-versus-benefit argument, correctly but selec tively noting that most - but not all - design-basis issues pose little risk.

Most of the NRC's budget comes from government fees on nuclear utilities rather than tax dollars. As a result, the industry's concerns about how the agency stewards the programs that plant owners pay for get a receptive hearing in Congress.

The pushback that eventually threatened the NRC's well-being began in the 1980s, in response to the agency's rising concern about design-basis issues.

A 1985 close call at Davis- Besse helped focus the NRC's at tention. Design flaws and opera tor errors combined to temporar ily disable the plant's main and backup water supplies to the steam generators, driving the re actor's temperature and pressure abnormally high and risking damage to the radioactive core.

The NRC decided that more-thor ough inspections of plants' safety systems were needed. The late- 1980s inspections, in turn, found that plants weren't keeping de sign information properly and were modifying equipment with out an understanding of its effect on the safe operation of the plant, according to an agency summary.

The NRC in 1990 asked the nu clear industry's trade group to consider mounting an organized effort to get plant operators to collect, re-examine and revali date their design information. The trade group declined, saying that most of its members were al ready doing so voluntarily.

The agency's response was to issue a 1992 policy statement stressing the importance of nu clear plants operating as they were designed. As an incentive, the statement pledged that the NRC would not fine plant opera tors for lesser design deficiencies as long as it was the utilities and not the agency that found them.

Over the next few years, the NRC dropped its costly and in trusive design-oriented inspec tions. It relied on the industry's insistence that it would police it self. But in 1996, Millstone changed everything. A blistering Time magazine cover story re vealed that the Connecticut plant had, with the NRC's knowledge, operated for 20 years outside its design basis by routinely unload ing its entire reactor core into a storage pool without waiting the required time for the radioactive fuel bundles to cool. If the pool accidentally drained or its cool ing system failed, the conse quences would have been disas trous.

Follow-up inspections there and at several other plants turned up more design flaws and questions. Embarrassed by the bad publicity, the agency's new chairwoman, Shirley Jackson, made it clear she intended to launch a broad, no-holds-barred hunt for design problems.

NEI, the industry group, con tinued to insist that the issues were minor and only at a few plants, not systemwide. "Look at the bigger picture," the organiza tion's president urged Jackson.

NEI's analysis, sent to the NRC chairwoman, warned of "the ma jor upset that will result if the NRC staff continues on its appar ent path of modifying . . . a long- standing regulatory scheme" without evidence the changes would increase public protection.

The agency pressed on with its plans, penning the 1996 letters that required plant CEOs to swear they had reasonable de sign-review processes, and form ing special inspection teams to scour reactors' safety systems for design flaws.

Davis-Besse was one of the plants to undergo the checks in 1997 and '98. While the inspec tors missed the problems at the Toledo-area reactor that have come to light in the past year, they identified significant issues at other plants, including enough design-related uncertainties to force a three-year, $500 million shutdown of Michigan's D.C. Cook facility.

By 1997, the number of react ors on the agency's problem plant list had doubled, fines were increasing and the amount of electricity - and dollars - generated by nuclear plants had dipped sharply. The NRC planned to continue its design inspections.

But the nuclear industry and certain members of Congress were fed up. Domenici, using an analysis done at no cost to the Senate subcommittee by a nu clear industry consulting firm, made it clear he believed that the agency had gone too far in its regulatory efforts.

The industry consultant's study estimated that the 103 U.S. nuclear plants spent at least $952 million per year responding to NRC initiatives that didn't in crease safety. The agency should drastically reduce its activities, the study recommended, elimi nating 704 jobs and cutting its $489 million budget request by $90 million. The subcommittee agreed. The analysis especially singled out the NRC's design-ba sis reviews, saying they should be abolished because they imposed high costs on reactor operators but uncovered few significant flaws.

Domenici, a powerful Republi can who has championed the ex pansion of nuclear power pro duction, made his point. The NRC followed many of the study's recommendations and preserved most of its budget. Most notably, the agency ended the design team inspections sev eral months later, well short of its goal of checking 35 plants, though it continues a smaller- scale "sampling" effort.

NRC Chairwoman Jackson, who maintained that the severity of the design-basis errors found at a minority of plants justified the reviews, left the agency in 1999 for the presidency of New York's Rensselaer Polytechnic In stitute. She repeatedly has de clined to discuss the confronta tion with Domenici; the senator's office did not respond to an in terview request.

"Millstone exposed huge defi ciencies in conforming to design basis," said former NRC commis sioner Victor Gilinsky. "It had a perverse effect, however, in that it mobilized the industry to come down on the NRC by means of an attack on its budget."

Lochbaum, the nuclear safety engineer, is more blunt. "The NRC is as good a regulator as Congress permits it to be," he said. "Right now, Congress doesn't want a good regulator."

Three months ago, the NRC made public its action plan for applying the lessons the agency has learned from Davis-Besse to the way it oversees the nation's nuclear reactors. The 33-page re port contained 54 activities in tended to make for better, safer plants.

The phrase "design basis" was nowhere to be found.

To reach this Plain Dealer reporter:

jmangels@plaind.com, 216-999-4842