In the 1980s, I was a manufacturer’s representative for electronic test set manufacturers. About 1986, I was asked to demonstrate a cable locating test set for TVA’s (Tennessee Valley Authority) Browns Ferry nuclear power plant. This plant, located in Alabama, had been shut down in 1975 because of a fire that had nearly severed critical control cables. It, like the Japanese Fukushima nuclear power plant, is a General Electric (GE) Mark 1 reactor design from the 1960s.
What caused this emergency shutdown and near nuclear runaway? A candle! Yes, a candle was used by technicians to locate a leak in the seal around cables between two rooms. Here’s a complete report. Here is a simplified report, with actual photographs. The photo of the cables on fire is frightening, when you realize that those cables are all that controls the fission reaction. As with most catastrophic system failures, several subsystems failed, and more than one person goofed.
These paragraphs from Candle In The Wind are chilling:
As the fire burned insulation off electrical cables, they shorted out. Equipment that had been running stopped running. Some equipment that had not been running started running. When low-voltage instrument cables shorted, the control room lost indication of equipment status. Because the cable spreading room was common to reactor Units 1 and 2, both units were affected.
The fire disabled all the emergency core cooling systems for Unit I and most of these systems for Unit 2. The water level in the reactor vessel was nearly 17 feet above the nuclear fuel during operation. During the fire, the Unit 1 water level dropped to just 4 feet above the reactor core. The fire was, in the words of David Lilienthal, a former director of TVA and the first chairman of the Atomic Energy Commission (AEC, the NRC’s forerunner), “An incredible piece of negligence.”
Investigation of the fire had revealed that at least one supposedly redundant cable resided in the same cable tray as its twin. Since the two cables lay next to each other in the same cable tray, if one cable were damaged, its redundant cable would probably also be damaged by the same physical assault. The plant designers specified that redundant cable pairs would run alongside opposite walls — not in the same cable tray. The NRC (Nuclear Regulatory Commission) subsequently required that the paths of all control cables be traced — which is why I was called.
Who would have guessed that cables would not have been installed as planned? Who would have guessed that electricians in a nuclear power plant would use candles right next to flammable insulation on critical control cables? Who would have guessed that proper fire extinguishing procedures wouldn’t be followed? Who would have guessed that the units at the Fukushima nuclear power plant would be unable to cool their cores because they had no electricity, because the standby generators resided in a flooded basement?!
Maybe we need to plan for not only likely mistakes, but unlikely mistakes, as well, because humans do very dumb things.