Excerpt from Tickling the Dragon’s Tail: How I Once Blew Up a Nuclear Reactor and Went on to Save the World’s Wild Places

Boyd Norton

One of the legendary stories of the Manhattan Project at Los Alamos was about famed physicist Richard Feynman. A risky experiment had been proposed by physicist Otto Frisch, which entailed dropping a piece of fissionable uranium 235 through a subcritical mass of the same material, making it supercritical for an instant. The burst of fissions would help in refining calculations for the final critical mass needed for the atomic bomb. At the meeting where the presentation was made for the experiment, Feynman began chuckling. When asked why he thought it humorous he said, “That’s like tickling the tail of a sleeping dragon.” Thereafter, it was named the Dragon Experiment. Later, modified versions of the Dragon Experiment killed two physicists.

In 1960 I went to work in a project in Idaho that was a descendant of the Dragon Experiment. What followed was a lifetime of saving wilderness worldwide.

A-long the trail you’ll find me lopin’
Where the spaces are wide open
In the land of the old A-E-C, (Yahoo)
Where the scenery’s attractive
And the air is radioactive
Oh the Wild West is where I want to be-e-e-e . . .
     —From a song by Tom Lehrer

I love all deserts, but I have a particular fondness for that bleak and barren piece of wasteland that sprawls for hundreds of miles west and south of Idaho Falls, Idaho. It is blistering hot in August, bitter cold in January, barely tolerable the rest of the time.

It’s called the Snake River Plain or the northern Great Basin Desert, depending upon which map you look at. Nothing there grows higher than your knees, and it’s mostly scrubby sagebrush that claws at your legs when you walk through it. In spring, after the snow melts and the rains fall, this desert has a green blush that masks the outcroppings of black lava bedrock. When it rains, the air is heavy with the camphorous odor of sage. In September the sage and grasses turn pale and, after those searing hot days of August, the desert smells like burnt grass and everything is crackling dry. In winter, just before dawn, the snow is tinged icy blue, like a scene out of Dr. Zhivago. The cold, dry air stings your nostrils when you inhale.

Perhaps the real reason I like this desert so much is because it’s a land of extremes, and I worked in it doing extreme things. It was important, cutting-edge science. In recent times one reporter has called us “The Right Stuff” of nuclear science because we were pushing the outside of the envelope of reactor safety. And on November 5, 1962, I punched through that envelope by blowing up a reactor. Deliberately.

My God, but it had been fun. And exciting.


Headed west in 1960, fresh out of Michigan Tech with a degree in physics and an eye toward California, land of sunshine and the aerospace industry, I took a wrong turn in the middle of Wyoming, or thought I did at the time. I wanted to see Yellowstone, make one of those typical in-and-out-of-the-park trips, mostly to confirm the wild stories I’d read about the place. My road map had been lost for days, buried somewhere in a backseat that was full of clothes and empty beer cans, but I knew where I was going and I knew that somewhere, just before Yellowstone, I would pass a mountain range called the Tetons. That name conjured up dim memories, from an old calendar photograph, perhaps, or a travel article, of towering peaks covered with a frosting of dazzling snow.

West of Riverton and across hot, shimmering desert, the blue forms of distant mountains took shape. Past Dubois they were confirmed; rough, rugged mountains jutting out of a deep-green sea of forest—the Tetons, obviously. (Actually, as I was to learn later, what I was seeing was part of the southern Absaroka Range.) The highway continued on, climbing upward and apparently through the range. Funny. I didn’t recall the map showing any highway cutting across the Tetons. Near the top of Togwotee Pass I stopped to breathe in that crisp air, take a picture of the great cliffs towering above, and listen to a Californian brag to another tourist about why he lived in the land of smog instead of here: “Ya can’t eat scenery.” (“Damned good thing,” I muttered to myself.)

I was totally unprepared for what happened next. Having seen and crossed the “Tetons,” I continued my journey. In a few miles I rounded a bend on the west side of Togwotee Pass and very nearly drove off the road. Before me was the most stunning panorama I had ever seen. Laid out like a lush green carpet below me was the valley of Jackson Hole. And at the far edge of that carpet, poking upward like carnivorous teeth biting into a deep-blue Wyoming sky, were the most incredible mountains. No. They couldn’t be real. They must be the leftover backdrop from a ridiculous Hollywood extravaganza. Even in my wildest dreams I could not have conceived of a range of peaks so awful, so bristling, so wild and rugged and magnificent. But there they were. As I drove closer, the first impressions were amplified. They were awful, looming jaggedly over Jackson Hole. Frightening. Wild. And fascinating.

I never made it to the aerospace industry, choosing instead to settle somewhere near these magnificent Tetons—my Tetons. (I secretly laid claim to them that day on Togwotee Pass.) And thus began what was to become—and remain—a love affair with a great mountain range and wild country everywhere.

Fortunately, before leaving the East Coast for my cross-country jaunt, I had submitted a resume in response to an ad in a technical magazine. A prime contractor for the United States Atomic Energy Commission was seeking physicists for reactor-safety studies at the National Reactor Testing Station in Idaho.

Idaho! I had never been west of Wisconsin, so I hadn’t the foggiest notion of what Idaho was like. I suppose in my mind there was some vague vision of a Sun Valley ski resort rising like Mount Fuji, abruptly and snow-capped, out of a vast sea of potatoes. It didn’t matter; if it was close to the Tetons and Yellowstone, Idaho it would be. Especially if I could be in the nuclear field as well.

I had some deep roots in nuclear science. When I was ten years old, I could write complex chemical equations, and I knew the symbols and atomic weights for most of the known elements. By then I had graduated from the simple chemistry sets and had a fully stocked chemical laboratory in our basement. News of the atomic bomb drew me into physics like a magnet. By the time I was twelve I could explain in great detail the concept of critical masses and the fission process. My career path was becoming clear.

I received my first AEC/FBI security clearance at age eighteen, right out of high school. This gave me access to information classified up to Top Secret. In order to earn money for college I had taken a job in Attleboro, Massachusetts, as a lab technician in a metallurgical laboratory doing research on fuel elements for nuclear submarines. I guess I impressed them enough, for they hired me back each summer during college. It was there that I handled uranium 235 (the highly enriched fissionable isotope of an atomic bomb) for the first time, and I got to play with exotic elements from a big chunk of the periodic table. Then, at nineteen, I was accepted with a scholarship to Michigan College of Mining and Technology. It was the poor man’s MIT, located in the wonderful boondocks of Michigan’s Upper Peninsula. I majored in physics with a minor in beer drinking.

So, when I hauled into Idaho Falls, only two hours’ drive from the Tetons, I was pretty excited. I put on my only suit and tie and went to see a man about a job. I was ecstatic when he offered me the job.

From the start I knew I’d fit in. I was pissed that I had missed all the fun and excitement of the Manhattan Project. But this program had a definite Manhattan Project atmosphere and excitement to it: expansive desert country; rusting, drafty old metal buildings (your basic early AEC architecture); battered pickup trucks tooling back and forth between the control center and the reactor buildings. Above all, the total environment was informal. The only suits and ties seen around here were worn by the occasional Washington dignitary or job interviewee. Everyone else, including the project manager, ran around in Levis or chinos and wore shit-kickin’ boots. I needn’t have bothered with the suit and tie for the interview.

It was an important program—in fact, the only study of nuclear reactor safety that the AEC had at the time. Being at the cutting edge of this research was exciting. Or, as one of my colleagues termed it, “Pushing back the foreskin of science.”

The project was called SPERT, one of those great AEC acronyms, which translated to Special Power Excursion Reactor Test. There were four reactors at the SPERT Project, each of them operated remotely from a control center a half-mile distant. These reactors were stripped-down nuclear hot rods; they had no radiation shielding and no elaborate safety systems. Sitting as they were, in the middle of more than nine hundred square miles of desert, there wasn’t much concern over such things. Not back then.

The manager of the SPERT Project had nuclear roots in the reactor field extending back to day one. As an undergraduate student at the University of Chicago, he had assisted Enrico Fermi in building the world’s first nuclear reactor under the football stands at Stagg Field in 1942. He was one of three agile young physicists who were known as the “suicide squad.” Had the nuclear chain reaction proven uncontrollable—and no one, not even Fermi, seemed to know for certain about the safety—it was the task of this trio to scramble over the scaffolding atop the reactor and drench the beast with buckets of neutron-absorbing solution. As the reactor had no shielding, the task could have been suicidal in the event of a nuclear runaway.

My arrival at the SPERT Project in 1960 coincided with the start of the first real push by the AEC to utilize nuclear power commercially on a large scale. During the 1950s, little had been accomplished in realizing those great postwar promises of abundant, virtually meterless atomic power. First, there had been the problem of separating the peaceful atom of industry from the top-secret military atom, a task that proved almost as difficult as the isotopic separation of uranium 235 from uranium 238. Then there were technical problems, lots of them. Harnessing nuclear fission in safety simply wasn’t as easy as had been thought. The urgency of the wartime Manhattan Project allowed a lot of risk-taking. For peacetime uses, however, public safety had to be assured.

Finally, no single, unified approach had been taken toward the commercial development of nuclear power. Hundreds of millions of dollars were spent in pursuing various reactor concepts. Some of them, in retrospect, seem rather weird. There were research programs investigating gas-cooled reactors, heavy water-moderated reactors, boiling water reactors, high-enrichment reactors, low-enrichment reactors, pressurized water reactors, organic-moderated-and-cooled reactors, fast reactors, breeder reactors—reactors that were cooled and fueled by just about everything imaginable. And, of course, each concept had its strong proponents.

I arrived at the National Reactor Testing Station at about the peak of this confusing technical diversity. Among its various programs, NRTS had more than thirty different research reactors of every size, shape, color, and flavor. Their names were a nuclear alphabet soup: MTR and ETR, the Materials Testing Reactor and Engineering Test Reactor, huge neutron factories used for important irradiation studies on materials; NRF, or Naval Reactors Facility, with its landlocked nuclear submarine, an aircraft carrier sans ocean and a tyrannical admiral in charge; ML-1, Mobile Low-Power Reactor number one; SL-1, Stationary Low-Power Reactor Number One, scene of a grisly nuclear tragedy in 1961 (SL-1 was only five miles from my project and the deaths of the three men had great impact on us); OMRE and EOCR, Organic-Moderated Reactor Experiment and Experimental Organic-Cooled Reactor; EBR I and II, Experimental Breeder Reactors numbers one and two; GCRE, Gas-Cooled Reactor Experiment.

Finally there was the most ludicrous of all AEC projects, the ANP or Aircraft Nuclear Propulsion program. Picture an airplane larger than a Boeing 747 carrying a nuclear reactor containing highly radioactive fission products of such quantity that, should it crash, it could kill the population of most cities and perhaps a few states. Fortunately, reason finally prevailed and ANP was canceled even though the hangar and the reactor had been built for it. I can’t even begin to imagine writing an environmental impact statement today for that baby.

SPERT Project, National Reactor Testing Station

In 1960, SPERT represented the Atomic Energy Commission’s foremost research program in studying reactor safety. At that time, before anyone appreciated the danger of the Three Mile Island’s loss-of-coolant accident, the greatest concern was the supercritical power excursion or nuclear transient (which, by the way, initiated the Chernobyl accident). Each of SPERT’s four reactors served a different and highly important function in the field of safety studies.

SPERT I, built in 1954, was the simplest of the four, with a large, open tank containing the core and water moderator. In basic design it was similar to many research reactors of the day, and it had been the workhorse of the project. Before it was shut down and dismantled in 1967, seven different cores had been tested in it and more than two thousand power excursions run.

SPERT I Reactor

SPERT I Reactor

SPERT II was altogether different. Essentially a small-scale prototype of a modern nuclear power plant, it consisted of a large pressure vessel and complex plumbing for circulating the coolant/moderator. The major difference, however, between SPERT II and modern nuclear plants was that it used heavy water as a moderator. SPERT II was a strange machine to run. Because of certain properties of heavy water, the core continued to generate large numbers of neutrons after the reactor had been scrammed and shut down. It was sometimes hours after a test before we could safely reenter the reactor building.

Then there was SPERT III. I hated that machine, even though I eventually ended up being in charge of its operation. Maybe hated is not the proper term. Perhaps feared would be better. Either way, this was the most complex facility we had, essentially a prototype nuclear power plant of the pressurized water reactor type, but only about one-fifth the size of modern plants. It was a plumber’s nightmare, with its primary and secondary cooling systems, canned motor pumps, rod drive seals, check valves, dump valves, pressurizer, and so on ad complexium. The core was contained in a large pressure vessel. The total system was designed to be operated at 650 degrees Fahrenheit and 2,500 pounds per square inch, with a coolant flow of twenty thousand gallons per minute—parameters similar to those in one of today’s nuclear power plants.

SPERT III Reactor and Associated Plumbing

SPERT III Reactor and Associated Plumbing

SPERT III had problems—something always seemed to go wrong. It appeared to be hexed: pressure fittings blowing out (and short-circuiting crucial control-system wiring), electronic and mechanical malfunctions, fuel plates bowing and warping, neutron flux suppressors breaking loose in control rods, mysterious shifts in the criticality position of control rods, relay failures, wiring failures, and more. But it held together, and we ran a lot of power excursion tests on it until, in 1968, we had an accidental nuclear transient during startup for a test. I was in charge of the operation of the reactor at the time, and the accident was largely my fault. SPERT III was shut down permanently after that.

SPERT IV was a return to simpler design. Another open-tank system—though much larger than SPERT I—this facility originally was intended to study in detail the phenomenon, discovered in certain early SPERT I tests, known as instability. It had been observed that operating a reactor at a high-power level—a few hundred megawatts—in the boiling-water mode could allow the onset of wildly divergent and dangerous power oscillations because of certain reactivity feedback mechanisms. I wasn’t at the project when they ran those instability tests, but I was told that it got pretty hairy in the control room when the power began oscillating out of control and threatened to blow the thing apart. Being at the reactor operating console was described to me as a total exercise in sphincter control.

The focus of reactor safety research in the early sixties was on the so-called reactivity accident, one where a reactor might suddenly be made greatly supercritical and undergo a severe power excursion or transient. In just hundredths of a second the power, or fission rate, could leap from zero to billions of watts, with the potential for severe core damage and spread of radioactivity. At the SPERT Project we ran many supercritical power excursions on our reactors, though the tests were normally conducted in such a way as to prevent damage to the cores and the release of any radioactivity.

In 1962 it was decided to run the ultimate super prompt-critical power excursion on the SPERT 1 reactor. Blow it up, in other words, to see how a reactor would disassemble itself and how severe it would be. I was chosen to run the test.


November 5, 1962
“Console power on.”

I say it to no one in particular as I insert the key into a special lock on the gray, sloping panels, and turn it. It is a habit from many months of reactor operation to call out certain tasks, usually double-checked or at least observed by another operator. This time, however, no one seems to pay attention. Turning the key brings the console to life. Lights of different colors glow against the drab metal panels. This is the simplest of control panels, unlike the complex systems used for operating nuclear power plants or any of a number of special research reactors. There are no safety systems on this or our other reactors. By grabbing that pistol-butt handle protruding from the console and turning it, I could pull all the control rods out of that core a half mile away. And nothing would halt the massive power excursion, which could blow the reactor apart. Which is what I will do, but under a more precise and controlled condition, which will allow recording of the reactor’s temperature, pressure, and fission rate as it comes apart.

The control room is crowded, and there is tension in the air as everyone goes about assigned and carefully rehearsed jobs in preparation for the test. Snatches of other conversations float into my consciousness.

“ . . . and there’s a drift on channel twenty-three, in-core thermocouple . . . ”

“Check, power to camera six is on . . . ”

“Peak power versus alpha curve predicts now about two point three gigawatts . . . ”

It is T minus thirty minutes and counting. We’ve been here almost five hours, having driven out from Idaho Falls in the grim, cold overcast of dawn, wondering about the skill of the meteorologists who predicted good weather and winds for the test. Now the weather has improved, the countdown has progressed, and, after months of preparation, it seems the test finally will take place.

Preparations are nearly complete half a mile away in the SPERT I reactor building. Electronic data-gathering systems ready, high-speed cameras loaded and on standby to be activated in the last split-second. Radiation monitors, air samplers, closed-circuit television, emergency equipment—all are prepared. As the last of these items is checked off on the master list, the order is given to evacuate the reactor area. The health physicist makes the announcement, and a loud warning buzzer sounds. As the technicians return to the control center, the road to the reactor is sealed off. A black-and-white-striped semaphore gate with flashing red lights blocks the way. A nose count is made to be sure no one has been left behind.

Once more all checklists are scanned. Radio contact with meteorological teams confirms the right winds and weather. Security and health physics teams are prepared. A final consultation in the control room among the physicists, engineers, and electronics technicians. All systems ready. Then I’m given the word: “Start nuclear operation.”

In starting up from a shutdown condition, all reactors are made slightly supercritical by pulling up the control rods slowly. The rate of fission is allowed to increase in a carefully controlled way until the desired power level is reached. Then the control rods are inserted slightly and adjusted until the fission process is at steady state, neither increasing nor decreasing in time. When that steady state condition was achieved, the reactor was termed to be “critical.” If something unforeseen should occur, there is a large red button on the console labeled SCRAM. Punching it cuts off the current to the control rod magnets and allows the neutron absorbing rods to drop back into the core, stopping the fission process.

A few more checks are made. Then I grasp a handle that protrudes from the console. Turned to the right, it begins withdrawing the control rods. Immediately, a warning horn sounds in the now-abandoned reactor building, a loud bellow which lasts for only ten seconds. It’s a last minute warning to anyone left behind to get the hell out of there.

SPERT I Operating Console

SPERT I Operating Console

As I watch the digital indicator of rod position, I tune in mentally to the audible neutron detectors, a static pop-pop-pop like a Geiger counter, coming over speakers above the console. As the rod-position indicator clicks in a blur of numerals past nine, ten, eleven inches, the pop-pop-pop rapidly increases in frequency. Twelve, thirteen, fourteen inches, and now the speakers become a frantic shriek of static as neutrons give birth to multitudes of progeny from the infinite wombs of splitting uranium 235 atoms. I release the control-rod handle, and it snaps back to the stop position. As the loudness of the neutron detectors approaches the threshold of pain, I flip a switch which silences them and focus my attention on a panel immediately in front and at eye level, where a pen on a chart of slowly moving paper begins to trace a lovely exponential curve of increasing fissions across the lines of the graph. Supercritical.

I allow the pen to rise through three decades of power, pushing a range-change button each time the red-inked line threatens to leave the right side of paper. Finally, I turn the pistol-grip to the left for a moment, to insert the rods slightly, repeat, and continue to jog the handle to adjust the distant rods to achieve that delicate equilibrium of neutron absorber to fissions. Then the crimson line moves unwaveringly straight in the middle of the chart. Critical.

I make a cryptic notation in the logbook: crit @ 13.65 inches. Pulling the TR. After conferring for a moment with the others, I grasp another pistol-grip to the right of the first one and below a white-etched label which reads TRANSIENT ROD. As I turn it to the right to begin raising this upside-down control rod up into the distant reactor core, I become aware of increasing activity in the control room as last-minute checks are made on all systems:

“CEC’s on standby . . . OK, mag tape backup loop is on . . . check, and switches A through E are on . . . slight drift on channel thirty-two, but it’s in limits . . . timer settings: transient rod fire at thirty point oh, oh, cameras A on at twenty-point oh, oh, transient rod latch open at . . . ”

As the transient rod continues to move upward into the core, it perturbs that delicate balance of fissions by absorbing neutrons and causes the red line on the chart to move slowly to the left as the fissions die off. Subcritical.

We confer for a few moments on final calculations for the test. Then I raise the control rods well above their normal critical position to a carefully predetermined level. The reactor remains subcritical because the transient rod is now fully into the core and is quenching any fissions which might take place. At the proper time the transient rod will be ejected, leaving the reactor greatly supercritical.

We are only moments away now. A brief hold in the countdown is called while we wait for an airplane with radiological measuring equipment to circle into position. We expect little radioactivity to be released because, unlike that of power reactors, this is a clean, cold core containing only a tiny amount of radioactive fission products. Nonetheless, it still is necessary to monitor even the minute amounts that may be released.

Finally, the moment is at hand. Everything is ready. With sweaty and somewhat shaky hands, I insert another key into the console, this one to start the sequence timer. One last check of the console. Then I turn the key. My mouth is dry as I call out the last seconds, shouting now over the whine of the recording oscillographs: “three, two, one, FIRE!”

In those carefully dissected parts of a second—too small, really, to comprehend—the neutron-absorbing transient rod is fired rapidly from the core, and what follows, unseen and unheard by any person, is the silent scream of neutrons, which multiply a billion- trillion-fold in the classic chain reaction. What we do hear in the control room is an enormous explosion over the reactor’s intercom system, the frightening sounds of unleashed energy, followed by the crashing of things being hurled and broken. In less than one-hundredth of a second, the reactor’s power zooms from zero to more than two-and-a-half-billion watts. Released in a core only two feet by three feet in dimension, this tremendous surge in power causes melting and vaporization of the fuel and violent expulsion of the water moderator to a height of nearly a hundred feet. But I’m too busy at the console to glance out the window and see the spectacle. I have to view it later on film.

In the moments that follow there is a great deal of excitement in the control room. Procedures must be followed to assure there will be no further danger. In order to avert a possible secondary supercritical excursion from pieces of the fuel falling together in a critical mass, we begin the process of draining water moderator from the reactor tank. Radiological teams in protective clothing and special breathing apparatus prepare to enter the reactor area to retrieve film from the motion picture cameras.

Eventually, all the important procedures are carried out, and the system is made secure. With the tension lessened, there is elation, almost giddiness, among my colleagues in the control room, with much backslapping and handshaking. When I have a chance to relax and reflect on the events of the day, I realize: Holy shit! I have just blown up a nuclear reactor. And it was fun. Moments later I called my wife, Barbara. “We did it,” I said.

“Well, I didn’t see any mushroom cloud,” she quipped.

As we were to discover later, by way of Three Mile Island, Chernobyl, and Fukushima, there is a much greater danger than the power excursion in reactor technology: the core meltdown, or China Syndrome—so-named because some wag once suggested that a molten radioactive core might bore its way entirely through the earth to China. It wouldn’t happen, of course, but it could contaminate the hell out of a large region as it did at Fukushima and Chernobyl.

The author looking at the core of SPERT I prior to blowing it up.

The author looking at the core of SPERT I prior to blowing it up.

After the test.

After the test.

The dragon roared.

Back to top ↑

Sign up for Our Email Newsletter