Flawed Fermi 2 test baffles experts; defective process eluded utility, NRC for 2 decades

2/23/2007

NEWPORT, Mich. - For 20 years, Detroit Edison Co. unknowingly relied on a certain amount of luck in calculating the reliability of its main source of backup power at its Fermi 2 nuclear plant.

But even after an extensive probe of public records, the Washington-based Union of Concerned Scientists concluded in a report issued yesterday it's still unclear why a flawed testing procedure for the plant's massive emergency diesel generators went undiagnosed for so long.

Equally as baffling, according to David Lochbaum, the group's nuclear safety project director, is how fundamental oversight of such important safety devices eluded Nuclear Regulatory Commission inspectors.

"You can't blame it on any one person," Mr. Lochbaum said, explaining that both the utility and the regulator had multiple opportunities over 20 years to catch a low-voltage testing procedure that was out of kilter.

"I can't figure out how [multiple engineers over a 20-year period] missed this over and over. I think the company and the NRC need to look back and figure out what happened. Everyone was sincerely trying to do it right," he said.

Here's the situation: In 1986, Detroit Edison correctly diagnosed a design flaw in the plant's undervoltage relay setpoints for the Division 1 electrical system.

But, according to NRC findings issued in December, the utility never followed through with a commitment it made on Aug. 22, 1986, to bring its testing criteria into sync once the physical changes were made.

The utility and the NRC carried on for two decades, oblivious to the discrepancy.

Government inspectors reopened the issue July 30, 2003, then closed it July 27, 2004, with the understanding the problem was about to be fixed. A special inspection team reopened the issue again after being dispatched to Fermi 2 in November, 2005, the NRC has said.

The procedure was corrected a few months ago. Tests now performed are considered accurate, the NRC has said.

Detroit Edison had little to say, both yesterday and when the NRC issued its findings in December.

"We do recognize the minimum voltage criteria should have been changed," spokesman John Austerberry said. "We don't know why, at this point, that hadn't been caught before."

Spokesmen for both the NRC's headquarters in suburban Washington and the agency's Midwest regional office near Chicago said they were not prepared to comment on Mr. Lochbaum's follow-up report.

Michael Keegan, a Monroe activist and spokesman for Coalition for a Nuclear Free Great Lakes, said the consequences "could have been an accident on the scale of Chernobyl" if backup power ever failed during a serious accident.

He said the region cannot be lulled into a false sense of security, especially now that Detroit Edison's parent, DTE Energy, has announced plans to apply for a license to build another reactor at the Fermi site.

Mr. Austerberry noted that the diesel generators worked when called upon for backup power in two real-life incidents recently: The blackout of August, 2003, and when a main transformer failed last summer.