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

CURV resembled this old system in some ways. It consisted of a frame of aluminum tubing, which held sonar equipment, lights, cameras, and three propellers. On top rested four oblong buoyancy tanks, and off the front jutted an arm holding a metal hydraulic claw lined with rubber. The whole contraption measured about five feet high, six feet wide, and thirteen feet long.

To recover a torpedo, the Navy steered a ship to the site and lowered CURV into the water. But CURV didn't just dangle there as the barge swayed overhead. CURV was remote-controlled, attached to the surface ship with a thick umbilical cord. A surface operator, sitting at a console, could see the bottom through CURV's TV and sonar and direct its propellers with joysticks. The operator flew CURV to the lost torpedo, grasped it around the middle with CURV's claw, then carried it to the surface. If the torpedo was too heavy, the operator could jettison the claw and back away. The claw, attached to a lift line, could then be raised to the surface. Early in its career, CURV managed to recover an unheard-of four torpedoes in one day.

By the time the Air Force dropped four bombs over Palomares, CURV had been operating for about two years and had plucked up fifty-two torpedoes from as deep as 2,000 feet. The CURV team had mastered their job, but the device itself was still a work in progress, and the CURV group remained a small-scale, low-budget operation.

When the Navy requested CURV's services in Spain, the team welcomed another opportunity to show off their skills. If they were successful, it would mean more recognition and money for their project. An H-bomb, to them, was just a big torpedo, and they certainly knew how to recover those.

There was only one problem — the tether was too short.

CURV's umbilical cord, an inch-and-a-half-diameter cable that fed power and commands to the device, was about 2,000 feet long. But by March 15, when Alvin found the bomb at 2,500 feet, the Navy knew that CURV would have to go deeper. So the CURV team embarked on a crash program to lengthen its lifeline by splicing in an additional 1,000 feet of cable.

The splicing was tedious work. The cable contained fifty-five separate conductors, each of which had to be spliced individually. The team staggered the splices over about four feet of cable, then covered the wounded area with a gray, pliable goo used for sealing air ducts. (The team dubbed the material “monkey shit.”) Then they wrapped the area in black electrical tape, sealing the splices as well as they could. Larry Brady and George Stephenson, the CURV operators, spent several days on the splicing operation. When they finished, the splice was an ugly black blob. “It looked like a python had swallowed a dog,” said Brady. “Or a couple of dogs.” Dragging the device to a NOTS range, the CURV team ran a few test dives. The ugly splice held.

The cable now stretched 3,100 feet long — long enough for the job — but they couldn't go much below 2,800 feet and still maneuver. There were some other bugs as well: CURV's depth gauge didn't work well below 2,000 feet, and its altimeter, which measured CURV's distance from the bottom, didn't work at all. But on March 24, as soon as Task Force 65's first lift attempt failed and the bomb fell back into the Mediterranean, Admiral Swanson ordered CURV to Spain. On March 26, the team packed up the CURV system, loaded it onto a cargo plane, and headed to Palomares.

The CURV team set up shop on the USS Petrel, a submarine rescue ship, on March 29. The Petrel

and her commanding officer, Lieutenant Commander Max Harrell, were old hands at Navy salvage, having worked on such operations for years, and the ship proved a good home for CURV. The CURV team set up their gear but had little to do; the bomb was still missing after the failed lift attempt. So for a couple of days, they ran tests. The Navy dropped a dummy bomb into the water with the same dimensions as the missing weapon and sent CURV off to find it.

The CURV team operated the device from a small control shack on the deck of the Petrel. It took two people to run the device: one to operate the sonar and read the compass, the other to move the switches and levers that spun CURV's three propellers and sent it swimming. When it came to

“flying” CURV, Larry Brady was the undisputed master. He just imagined himself sitting on the device, sailing happily underwater, and the moves came naturally. Once he found the target, Brady grabbed it with the claw, like a kid in an arcade grabbing cheap toys from a bin.

A couple of days after his arrival on the Petrel, Brady dove CURV to 1,050 feet, found the dummy bomb, and recovered it with a special claw the team had built in California. The next day, CURV