The Day We Lost the H-Bomb: Cold War, Hot Nukes, and the Worst Nuclear Weapons Disaster in History

In 1953, Randy Maydew's boss asked him to work on the project. The Air Force, at that time, wanted to drop nuclear bombs from 2,000 feet, at speeds greater than Mach 1. (Later, they requested drops under 200 feet.) The only way to do this and give the pilot time to escape, Maydew figured, was to slow the bomb down with parachutes. Knowing nothing about parachutes, he surveyed the literature and found that no parachute in the world could withstand the stress of being blown open at the speed of sound. So the Sandia engineers set out to design one that could. Along the way, they came up with other ways a weapon could survive a low-altitude drop: a bomb with a spike on its nose that stuck in the ground like a dart; a bomb with a metal honeycomb tip that could endure a bruising dent in the nose. In a couple of years, Sandia gave the Air Force plans for a laydown weapon. Maydew, meanwhile, became an expert on high-performance parachutes.

The Mark 28 bomb missing in Palomares packed an elaborate four-parachute system into its hind end. In a low-altitude drop, when the system worked correctly, three nylon parachutes would open in sequence — an elegant bit of fancy footwork in the sky. Soon after a bomb fell from the belly of a plane, a ring of explosive bolts fired on the back of the bomb, knocking the tail plate off. The inside of the tail plate had an eyebolt in the center, tied to a lanyard. The lanyard attached to a four-foot-diameter guide parachute. When the tail plate fell, it pulled the small chute out behind it. The pilot chute, in turn, heaved out a sixteen-foot-diameter ribbon parachute. This sixteen-foot chute slowed the bomb for two to three seconds, then cut itself loose. As it drifted away, it yanked another pack out of the bomb, pulling the cover off a sixty-four-foot chute. The sixteen-foot chute floated away, carrying the empty bag, as the larger chute finished the job. This monstrous canopy opened and slowed the bomb to about twenty-eight feet per second by the time it hit the ground, giving the pilots time to get away. To complicate matters for the bomb search, the Mark 28 was also designed to free-fall from a high altitude. In that case it would release only a small, thirty-inch chute, which would stabilize the bomb as it sailed down to its target.

Because bomb number four had been torn from its rack in an explosion, any — or none — of the parachutes might have opened. The three bombs found on land only emphasized the range of possibilities. The first bomb, which had hit the ground at about 140 feet per second, had deployed the sixteen-footer but nothing else, but that had been enough to keep the bomb intact. The second bomb had landed without deploying any chutes, smashing into the ground at about 325 feet per second. This high-speed impact had detonated the high explosive, scattering plutonium dust, case fragments, and remnants of parachute over the Spanish countryside. The third bomb had deployed the sixteen-foot chute, but because the chute was damaged, it hadn't supplied enough drag. The bomb had hit the ground at about 225 feet per second, igniting its high explosive and scattering radio active debris.

There were hundreds, maybe thousands, of possible scenarios for bomb number four. But despite the dearth of data, the Sandia engineers made some quick calculations over the weekend. Since bomb number four's tail plate had come off, it seemed likely that at least one of its parachutes had popped out behind it. If the big, sixty-four-foot chute had deployed right after the explosion, and if the wind had been blowing out to sea, and if the gusts had been strong enough, the engineers thought, the bomb could have splashed into the Mediterranean. Only one of many possibilities, it was an early hint that the bomb might not be on land. On Monday, one week after the accident, Sandia told the Pentagon that the bomb might have gone into the sea. A few days later, Sandia's computer spit out an estimated location: 37° 13.9' N, 01° 42.3' W. This was not good news; according to this calculation, the bomb had plunked far out in the Mediterranean, closer to Africa than to Spain.

On January 27, General Wilson requested that Randy Maydew or his boss, Alan Pope, fly to Spain.

Wilson was building an advisory team to help define the search area, and he wanted someone from Sandia on board. Having experts on site, he had decided, would be better than “furnishing data to unseeing computers.” As the engineers in Albuquerque continued to hog the IBM 7090 for their calculations (ultimately generating a three-foot stack of paper printouts), Maydew prepared to jet off to Spain. As he was packing, a fellow engineer dropped by his office with a gift: a forked stick, like those used by diviners to search for hidden water.