Maine Owl

4:00 am, thirty years ago today


Early sample of alternative media from the Maine Owl library

In March 1979 I was a student in electrical engineering at the University of Minnesota. I was working a co-op job in the electric utility industry programming big mainframe computers to display information on and control power grids. I loved wires and currents. I hated the industry.

Then Three Mile Island happened. It soured my taste for what I was doing there. Later, I would end up in Maine working on a graduate degree in physics while sniffing around the countryside for extremely low-level environmental radioactivity. (There is extremely low-level human-produced radioactivity present almost everywhere, along with much stronger natural radioactivity.)

I still object to the use of nuclear power. This is not because I think reactor meltdowns are likely or widely-distributed low-level radioactivity is super dangerous. It is not, at least not compared to a lot of other things. It's rather because of the un-democratic exercise of unaccountable government and corporate power and dominating wealth that the nuclear industry represents.

The story of what happened in Harrisburg, PA thirty years ago gives us important lessons about our over-reliance on arrogant scientific/technical decision-making processes that still harms our society. This comment is not made while failing to place risk in perspective. I'm not saying nuclear power plants cannot be operated somewhat safely. I'm just saying that we should not allow is trusting the vested corporate power structure to make all of our energy decisions. Inevitably if we do, they'll be bad ones.

Dr. Arjun Makhijani of ieer.org has the perspective that I prefer: we can have a carbon-free, nuclear-free energy system in fifty years if we start making the right decisions now. (Dr. Makhijani spoke in Maine in 2007, see HERE.) He's been posting recently that some of these bad decisions to resume building nukes are today on the table, ending the de facto thirty-year post-TMI moratorium.

Dr. Arjun Makhijani: Eight new nuclear reactors are being proposed in Texas alone. The two near Amarillo, in the panhandle, will consume 60 million gallons of water every day?more than what the entire city uses. The company proposing the plant has said there is a lake there in an unidentified location that will supply the water. In Idaho, the CEO of Alternate Energy Holdings, which wants to build a power plant there, implies that nuclear power will cost only 1 to 2 cents per kilowatt-hour, because capital cost is borne by the investors, as if Wall Street were a kind of charity for electricity consumers.

Despite government attempts to grease the skids on these things, Dr. Makhijani writes that some localities are finding it "unwise and imprudent" to pursue nukes "because of insufficient time to examine the paperwork and the risk of cost overruns and delays." That's hopeful.

The main lesson of Three Mile Island is that depending technical systems is very unwise when the risk of failure is so catastrophic. There are many parallels for our own time. Below I reproduce many illustrative details of those events of so long ago from one of the best accessible articles ever written on the subject. These authors captured my young conscience at the time. It's worth looking back at an extended excerpt...

Corporate Meltdown
The Lessons of Three Mile Island
by Bill Keisling and Ed Perrone | The Progressive, June 1979

... At four o'clock in the morning on March 28, 1979, the loud whistle of highly pressurized steam escaped the confines of Three Mile Island's Unit 2 nuclear power plant. The noise gushed into the darkness. Some nearby residents would later tell reporters that the sound had been loud enough to wake them from sleep. But others, less noted by the press, said they heard nothing, continued to sleep, unaware of the drama that was about to engulf them.

The men in the control room of Unit 2, working Three Mile Island's graveyard shift, heard the sudden escape of steam. They knew what it meant. The plant's generating turbine had for some reason shut down. This caused the shutdown of the secondary feedwater system, which carries heat away from the radioactive primary cooling system by way of two large steam generators. Within seconds, the nuclear core's primary cooling system overheated. Responding to the resulting sudden pressure increase, a relief valve on the pressurizer of the reactor?s primary cooling system opened, venting the excess steam, relieving the pressure. This was the sound some people heard in the night.

At the same time, the computer monitoring Unit 2's operations automatically directed the reactor's control rods to descend around the cylinders of uranium pellets, halting nuclear fission, shutting off the reactor.

Within thirty seconds, the computer's logic circuits turned on the auxiliary feed-water system to cool the steam generators, since the main feedwater system had shut down with the turbines. What the computer didn't know was that the auxiliary feedwater system had been shut off at valves between the auxiliary pumps and the steam generators. One emergency cooling system was useless. The computer, programmed to assume the valves were open, began to lose control of the nuclear power plant. ...

The automatic pressure relief valve in the system's pressurizer had done its job. Within fifteen seconds after the valve had opened, the pressure in the primary loop had returned to normal. At that point, the valve should have closed. It didn't. Water normally pressurized at 2,000 pounds per square inch for the purpose of quenching the nuclear core was pouring from the valve. The computer bad been programmed to assume this wouldn't happen. Worse yet, gauges in the control room indicated tremendously high water pressure in the pressurizer, when actually little water pressure was there. Within a minute, the two steam generators were drying out. And the nuclear core was being exposed to the air.

Two minutes into the accident the computer switched on the three emergency core cooling system pumps. Water began pouring over the core at 1,600 pounds per square inch.

But in the control room, Craig Faust was watching the pressurizer gauge rise off the scale. It was telling him, incorrectly, that the core's coolant was still present and terribly pressurized. When the white warning light for the emergency core cooling system flashed, Faust feared the highly pressurized water in the primary loop would back up into the emergency pumps, crippling them.

Four-and-a-half minutes into the accident, Faust shut off the first emergency core cooling system pump. Other control room operators, trained to assume a failsafe system, said they would have done the same thing. But by this time the computer was useless.

Soon the plant foreman and shift supervisor arrived on the scene. Eight minutes after the turbine had shut down, the auxiliary feedwater valves were opened.

Two explanations are offered as to how the auxiliary feedwater valves were finally opened. The first pictures Frederick as the quick-thinking, experienced control room operator: He doesn't assume fail-safe, doesn't assume the valves are open. He opens them from the control room, after discovering the problem on the flashing, ringing control panel.

The other explanation is more mundane: Either the control room operators had to be informed by management that the valves were closed, or checklist procedures on the emergency system, which the utility should have conducted before the accident, were hastily run through until the problem was isolated.

But at four o'clock in the morning, the men in the control room of Unit 2 had no time to ponder how their actions would later be viewed. Though the auxiliary feedwater flow had begun, the pressurizer gauge still was off the high end of the scale. Ten minutes into the mishap, a second emergency core cooling pump was shut off.

The temperature of the system should have been decreasing, cooled by the flow of water from the auxiliary feedwater system. But the reactor temperature had now gone over 600 degrees. The water remaining in the primary loop was superheated and vented as steam through the stuck electromagnetic valve. With little water left to cool it, the reactor was dangerously close to going completely out of control.

But as water from the auxiliary feedwater system finally entered the massive, drying steam generators, the malfunctioning pressurizer gauge returned on scale and suddenly decreased. The men in the control room realized with horror they had been led by a blind gauge.

Six-and-a-half minutes after the first emergency core cooling pump had been shut off, up to a minute after the second bad been shut off, both pumps were thrown back on. Already the metal tubing that held the uranium pellets was melting, bending, in the nuclear steam. When the water from the emergency pumps hit the melting nuclear core, the oxygen in the water combined with the zirconium, releasing hydrogen in the form of bubbles. Some of these bubbles blocked the cooling channels, while others rose to the top of the reactor's pressure vessel like the bubbles in a bottle of champagne.

Meanwhile, the water from the emergency core cooling system was following the rest of the coolant through the relief valve, which remained open. The rushing water, pouring directly over the reactor core to cool it, was highly radioactive, carrying small particles of the uranium itself. Like the rest of the primary coolant, it splashed onto the floor of the containment building.

Soon a floor sump pump came on, draining the water into a wastewater storage tank. The tank was never designed to handle the amount of water it eventually received. Fifteen minutes after the original turbine trip, the tank was filled. But with water still pouring in, a pressure disc on the tank's side blew, draining the highly radioactive water into a nearby auxiliary building. The building had no radioactive shielding; it was never meant to hold anything radioactive. From this building, a constant stream of low-level radiation would be vented into the atmosphere over the next two weeks.

During the next forty-five minutes, the systems began to stabilize. But another problem developed. The water in the primary loop doesn't usually boil because it is held under tremendous pressure. With the electromagnetic valve open, however, the coolant was boiling. The control room operators listened to the reactor coolant pumps through earphones, detecting cavitation. The same way an outboard motor stirs air into the water behind it, the delicate reactor coolant pump rotors were now running the risk of destroying themselves because they were churning air bubbles.

Both pumps were shut down. The core would now be cooled as the cooler water was heated and drawn upward, then out of the reactor at its top. But this method was less efficient than using pumps, and the core temperature began to rise again. By six in the morning the temperature of the reactor core was somewhere off the gauge scale. It would remain there until the early evening. At 6:18 am, the stuck pressure relief valve was discovered and closed, but only after 250,000 gallons of radioactive water had been spilled onto the containment room floor. By seven o'clock, someone on Three Mile Island got around to calling the state and Dauphin County emergency management agencies...

Kevin Molloy, director of Dauphin County Emergency Preparedness, received a call at his home at 7:05. Three Mile Island is within eyesight of his home. Molloy was told a site emergency had been declared at his neighborhood nuclear power plant. Racing to his office in downtown Harrisburg, MolIoy was told over his radio that Three Mile Island had declared a general emergency. He was about to begin two feverish weeks of drawing up the plans for the evacuation of Dauphin County.

By 7:30 in the morning, it was pandemonium at Unit 2. Sudden increases in pressure pointed to blockage of the cooling system, perhaps to a hydrogen explosion in the reactor vessel. Workers in the control room donned masks. TMI employes arriving for the eight o'clock shift were sent to an auditorium on the island. One technician described the scene in the control room as "twenty people doing the work of 150."

About 9:30 fifty TMI employes were evacuated from the island, pouring to the shore in a caravan of cars. Guards stopped them at the gates and ran Geiger counters over their automobiles. Some workers drove to a substation down the road to be checked for radiation contamination. Others traveled to an observation tower across the river from the island on the Middletown side. There the human chain reaction quickened. Helicopters carrying utility executives and hordes of journalists and cameramen touched down on the lawn of the observation building. Other choppers circled the island like flies around a piece of spoiling meat.

Inside the observation building a plant foreman rushed into a front room screaming, "I think there's a mixup somewhere here!"

Then he ran out....