Skunk Works

12 THE CHINA SYNDROME

THE MOST SENSITIVE project during my years at the Skunk Works was begun in 1962 and code-named Tagboard. Inside the Skunk Works and out, fewer than one hundred people were involved or knew about it, even though the program began under President Kennedy and became operational under Richard Nixon. To this day, Tagboard remains mist-shrouded, in part because it was basically a failure. But still, the operations were truly spectacular: flying spy drones over Communist China’s most remote and secret nuclear test facilities.

I was fascinated by drones. To me, a remote-controlled or preprogrammed pilotless vehicle was the pragmatic solution to spying over extreme hostile territory without worrying about loss of life or political embarrassments of the Francis Gary Powers variety. If the drone traveled high enough and fast enough, the enemy could not stop it—indeed might not even spot it. Several of us in the analytical section were drone boosters and from time to time tried to lobby Kelly Johnson into joining our fan club. But Kelly resisted. Drones, he argued, were too big and complex to be economically feasible or operationally successful. But over the years he gradually changed his mind. He was aware of the ominous crash development of nuclear and rocket weapons on the Chinese mainland and the loss of four Taiwanese U-2s, shot down while trying to film those sensitive sites. The most significant of these development facilities was Lop Nor, two thousand miles inland, practically to the Chinese border with Mongolia. Lop Nor was situated inside a two-thousand-foot depression twenty miles wide on an otherwise four-thousand-foot plateau. This rugged, isolated test facility was the primary target of U.S. intelligence, especially in the face of an aggressive and increasingly hostile Chinese foreign policy. The Chinese had split with Moscow, triggered border clashes with India and Tibet, and were causing major concerns in nearly every foreign capital around the world.

Lop Nor was two thousand miles away and a very tough round-trip for even the most experienced U-2 pilot. Those of us promoting the drone idea inside the Skunk Works argued with Kelly that it was the best way to overfly forbidding places like Lop Nor. We envisioned a delivery system where a drone would be piggybacked on top of the Blackbird and launched off the China coast, rocket up to 100,000 feet, and zip over Lop Nor at speeds faster than Mach 3, take its pictures, then turn around and fly back to the launch point, where on electronic command it would drop its film package by parachute to a waiting naval frigate in the sea below. The drone would then self-destruct. The technology involved was not only feasible but within our grasp.

Kelly got a negative reception to the drone idea from John Parangosky, who had replaced Dick Bissell at the CIA. The Air Force was only slightly more receptive. Nevertheless, Kelly found an ally in Brig. Gen. Leo Geary, director of special projects in the Air Force, who coordinated programs between the CIA and the blue-suiters. Geary obtained half a million dollars in seed money from “black project” contingency funds, and we put together a small team to plot and plan a design. I was the propulsion man. “This is a most peculiar situation,” Kelly told us. “The agency has turned its back, so Lockheed might wind up launching this damned thing ourselves. I have no instructions from anyone in Washington, but I think I know what they want: we will try to get six-inch ground resolution photographically, a range of at least three thousand nautical miles, a camera payload of 425 pounds, and a guidance system of about 400 pounds. We should detach the payload bay holding the camera, film, and guidance system and float it down by parachute. Make it reusable and save a bundle of money that way. That’s it. Go to it.”

The drone we designed had the flat triangular shape of a manta ray, was forty feet long, weighed about seventeen thousand pounds, would be built from titanium, powered by the same kind of Marquardt ramjet we once used for an experimental ground-to-air missile developed in the 1950s, called Bomarc. The drone had the lowest radar cross section of anything we had ever designed and could cruise faster than three times the speed of sound. It was equipped with a star-tracker inertial guidance system that could be constantly updated via computer feeds from the system aboard the mothership until the moment of launch. The system was fully automated, and the drone’s steering was directed by stored signals to its hydraulic servo actuators. It was capable of a sophisticated flight plan, making numerous turns and twists to get where it was going, then repeating them in reverse to return to where it came from. The payload was detached on radio command after the mission and parachuted to a waiting cargo plane equipped with a Y-shaped catching device. After the nose detached, the drone self-exploded.