tmi2This past weekend marked the 30th anniversary of the nuclear accident at Three Mile Island.  If you are old enough you may remember where you were at that time and what it was like.  I remember VERY well because I was on my way to the crippled plant near Harrisburg, Pennsylvania.  Our President at the time, Jimmy Carter, was also a micro-manager and a former nuclear engineer: he wanted his own eyes and ears on the scene.  Our little group eventually coalesced into the Presidental Commission on the Accident at Three Mile Island, led by Dartmouth College president John Kemeny, who was also the co-author of BASIC.

The lessons of Three Mile Island have been, for the most part, forgotten.  The nuclear industry changed and improved somewhat, but the deep understanding of what went wrong was lost on the public in general and the real lessons that we could have learned as a society were, too.  The financial mess we are experiencing right now isn’t all that different from Three Mile Island.  If we’d taken better to heart the true lessons of TMI we might not be in our present jam.

I spent a year of my life coming to understand TMI and even wrote a book about it, now long out of print.  I was there.

There was a partial nuclear core meltdown at TMI.  We all knew what that meant because, ironically, The China Syndrome, had just swept through American movie theaters six weeks before.  Years later there was a much more severe accident at the Chernobyl nuclear power plant in the Soviet Union.  Some people argue that TMI was actually worse than Chernobyl in terms of the actual core damage.  I don’t know.  There’s no doubt that Chernobyl killed a lot of people and TMI didn’t.  The difference was that TMI had a concrete containment vessel and Chernobyl had none.  Building nuclear power plants without containment vessels was insane and Chernobyl proved that.

Looking back at the accident with the benefit of knowing what it took to clean it up and what the workers found when they were finally able to send robots inside the containment, the TMI accident was very bad indeed.  There were pressure spikes during the accident that would have cracked an average containment vessel, releasing radioactive gases into the atmosphere.  Fortunately the Unit 2 containment wasn’t average.  TMI-2 was built on the final approach path to Harrisburg International Airport, a former U.S. Air Force base, and was therefore beefed-up specifically to withstand the impact of a B-52 hitting the structure at 200 knots.  A normal containment would have been breached.

TMI wasn’t caused by a computer failure but the accident was made vastly worse by an error of computer design.  Specifically, TMI-2 had a terrible user interface.

We had a confluence of bad design decisions at TMI, some of them made by the U.S. Congress.  U.S. law specifically prohibited using computers to directly control nuclear power plants.  Men would do that and nearly all of those men would be former nuclear reactor operators from the U.S. Navy.  Computers could be used to monitor the reactor and in fact it would probably have been close to impossible to monitor it without the help of computers.  There were just too darned many valves and sensors for any team of humans to keep track of reliably, 24/7.

So the computer (there was one) monitored the plant and raised an alarm if specific parameters changed.  Then a guy would flip a switch to open or close some valve, solving the problem.

Here’s how it was supposed to work.  Something went wrong.  The computer noticed what went wrong, set off audible and visual alarms, then sent a description of the problem to a line printer in the control room.  The operator would read the print-out, check the trouble code in one of many manuals, then make the adjustment specified in the manual.  Simple, eh?

Too simple, it turned out.

What happened at Unit 2 was a little more complex.  A cascading series of events caused the computer to notice SEVEN HUNDRED things wrong in the first few minutes of the accident.  The ONE audible alarm started ringing and stayed ringing continuously until someone turned it off as useless.  The ONE visual alarm was activated and blinked for days, indicating nothing useful at all.  The line printer queue quickly contained 700 error reports followed by several thousand error report updates and corrections.  The printer queue was almost instantly hours behind, so the operators knew they had a problem (700 problems actually, though they couldn’t know that) but had no idea what the problem was.

So they guessed.

Not good.

U.S. Navy reactor operators, the sort who served under Jimmy Carter in the 1950s, were selected primarily for their temperament.  This was a nuclear device, remember, so having trustworthy operators was most important. Besides, their Navy job – as at TMI – was to follow the manual.  All knowledge was inside the book.  So knowing the book was everything.  Unfortunately knowing the book isn’t the same as knowing the reactor.  This approach was extended to most civilian U.S. reactors, where knowing the book meant passing the test on the book NOT really understanding the guts of the machine.  Civilian reactor operator training in those days was nearly all about how to pass the test, not how to operate the reactor.

So when a real accident happened the operators weren’t prepared to handle it.  Their superiors at General Public Utilities weren’t prepared to handle it, either.  Nor were the experts at the Nuclear Regulatory Commission.  And don’t even get me started about FEMA.  The outcome of Hurricane Katrina was no surprise to me.

Every level of command waited too long to ask for help at TMI.  Sometimes this was because they thought they were on top of the situation, but more often it was to avoid embarrassment or – in the case of General Public Utilities – to avoid hurting the stock price.  The FEMA guys were just plain stupid.

Nobody died, eventually the reactor was brought under something like control, and a lot of lessons were learned in the process.  Reactor operators learned better how their reactors worked, for one.  The NRC generally gave up the job of promoting atomic power that had been its primary calling as the old Atomic Energy Commission, for another.  Reactor control rooms everywhere were dramatically improved and line printers banished as interface devices.  And for the next 29 years we didn’t build another nuclear power plant, leaving that mainly to the French and the Japanese.

Now nuclear energy can be mighty dangerous and is not something to be messed with lightly, but another irony in this story is that nuclear power is actually pretty simple compared to many other industrial processes.  The average chemical plant or oil refinery is vastly more complex than a nuclear power plant.  The nuke plant heats water to run a steam turbine while a chemical plant can make thousands of complex products out of dozens of feedstocks.  Their process control was totally automated 30 years ago and had an amazing level safety and interlock systems.  A lot of effort was put into the management of chemical plant startup, shutdown, and maintenance.  The chemical plant control system was designed to force the highest safety. So when manufacturing engineers from chemical plants looked at TMI, they were shocked to see the low-tech manner in which the reactors were controlled and monitored.  To the chemical engineers it looked like an accident waiting to happen.

The folks at TMI did not really know how to manage the technology of a nuclear power plant, and that led to a huge mess.  The same thing has now happened to our economy.  Congress changed the banking and mortgage lending rules without regard to their purpose.  Many firms bought derivative securities without the slightest thought to the math behind them or the risk they were incurring.  Nuclear power plants run on a chain reaction process of atomic decay.  Our government and investment community created a chain reaction of economic decay.

Chemical plants were better designed than nuclear power plants in part because Congress did not legislate how the chemical industry designed their plants.  But more importantly most chemical firms of that era had CEO’s with engineering degrees.  They had respect for the technology and the risk of misusing it.  But that doesn’t make the chemical industry blameless.  With the off-shoring of manufacturing a lot of chemical production is now being done in places where there is little respect for the dangers of technology.  The chemical industry’s TMI was Bhopal.  There will be more Bhopal’s coming because those companies are now being managed by bean counters, not engineers.

There is a place for nuclear power in our energy future.  I say this not as a particular nuclear advocate but as a realist.  The end of the Cold War has left us with a legacy of weapons grade nuclear materials that must be dealt with.  Thanks to the 1950s we’re stuck with all the issues of storing this stuff no matter what Obama or any other U.S. President does.  It just makes sense to me to take this stuff that used to be bombs and degrade it into something that can no longer make bombs but, oh by the way, can power millions of homes with no CO2 emissions.  It seems like making lemonade to me.  Yes, there are other renewable power sources that are even better than nuclear, but I seriously doubt whether they will add up to enough total watts in the time available.  We’ll need all of them.

Just as we neglected the economy for the last decade or more, we have also neglected nuclear energy.  We don’t have a national storage system for spent fuel.  We don’t have a spent fuel recycling process.  We don’t have a standard national reactor design.  We add incredible costs to power plants for an amazing list of things, many of which contribute nothing.

Life doesn’t get simpler, it gets more complex.  TMI led us to repudiate nuclear power as a nation – something in the long run we probably can’t afford to do.  We just have to find ways to manage technology – all technologies – more responsibly.  Sadly, we tend these days to put the wrong people in charge.