A few hours into the near-disaster at Three Mile Island the plant's operators had all the information they needed to prevent a serious nuclear accident but failed to recognize it, a Nuclear Regulatory Commission official said yesterday.
The operators may have failed to spot what was really happening, the official said, because of a phenomenon known as the "hassle factor"-the roomful of flashing lights, sounding alarms and sometimes contradictory instruments, all demanding attention, which the operators inevitably face during an emergency.
The hassle factor may have kept anyone from noticing for two days that there had been a small explosion inside the reactor containment building, according to Roger Mattson, director of the NRC's division of systems safety.
In fact, Mattson revealed to the NRC's Advisory Committee on Reactor Safeguards, the explosion, if that is what is was, lasted six minutes. Knowing about it at the time, 1:50 p.m. the day of the accident, would have indicated that something, probably hydrogen, was burning inside the reactor containment and there had therefore been severe damage to the core of uranium fuel, vital facts that were not known for two days.
"Somebody probably just rolled up the recorder sheet in a hurry and it wasn't until Friday [March 30] that it was unrolled and somebody said, 'Hey, look here,'" Mattson said.
That recorder sheet showed a "spike" in containment pressure levels, a rise to 28 pounds per square inch. But no one knows whether that means there was a six-minute fire or just that the recording instruments took six minutes from a blast to reach the 28-pound level and subside. Such ambiguity is part of the problem in the control room, Mattson noted.
Stephen Hanauer of the NRC's plant systems division noted that procedure guides tell operators to beware of a dangerous loss of pressure inside the reactor when two gauges agree. But at Three Mile Island, one gauge said yes and another, which turned out to be misleading said no, and the guidelines were silent on how to deal with that problem, Hanauer said.
"An operator can't always make the correct inferences with all that's going on in the room," he added.
Layouts of control rooms differ among utilities, he said, and some are "horrendous" from the point of view of the busy operator - "People standing on each other's feet, red lights that are green except when they're blue, indicators 14 feet off the ground and behind you," Hanauer added.
The utilities have "whims of iron" on the ways they want the control rooms laid out and officials cannot assume that what is true in one plant is true in another, he said.
Discussing the same subject Monday, Mattson told the committee that detailed electrical and structural drawings of the Three Mile Island plant "as built" would have been extremely useful to the NRC during the crisis and should be obtained for all operating plants.
Congressional amendments that would have halted construction and/or licensing of all nuclear power plants for six months were defeated during heated debate yesterday in the House energy subcommittee.
Democratic Reps. James Weaver of Oregon and Edward Markey of Massachusetts argued that the amendments would "send a message to the public" that Congress is concerned about nuclear safety, but the committee voted down two proposed changes, 11 to 9 and 15 to 6, while a third amendment was withdrawn.
Weaver said he would try again when the full Interior Committee votes on the NRC's fiscal 1980 authorization of $373.3 million.
The Three Mile Island computer was thinking about money at one point during the near-disaster, Mattson revealed. Temperature sensors registering more heat than they were programmed to record caused the computer to print out not question marks, as had previously been reported, but dollars signs, Mattson said.
The difficulty was that when the computer resumed printing temperatures as the reactor cooled, operators did not trust the readings, even though all 52 sensors agreed, Mattson continued.
Incomplete information also led to erroneous worry about the possibility of a hydrogen gas explosion that would break open the containment building, he said. "There was never any potential for a hydrogen explosion" of that much power, Mattson stressed. "We just asked the staff the wrong questions, and that is a very bitter pill."