The birth of the spy satellite

When the first American spy satellite failed in January of 1959, no one knew exactly what happened. There hadn't been a pilot on board, so engineers had nothing but telemetry to use to try and fix the problem. And the pressure was on. The missile gap between the United States and Soviet Union was thought to be increasing, and the American government was keen to see for itself just what was going on on Russian soil. This had been the goal of the failed satellite: it was meant to be the first orbital photoreconnaissance mission in the Corona program.

Orbital Reconnaissance

America’s first aerial spy program was the U-2 airplane. Conceived by Lockheed Martin in the early 1950s and introduced in 1956, the U-2 could cruise at 70,000 feet and take pictures of enemy nations safely (it was thought) out of range of any anti-aircraft missile systems. The first U-2 flight over the Soviet Union was on July 4, 1956, and the aircraft was spotted right away. The Americans had lost the element of surprise in aerial reconnaissance. So while the U-2 returned a wealth of valuable information, the Americans were keen to develop a successor to the high flying airplane that would be harder to track and truly impervious to ground or air-based missiles. The space age presented the perfect solution.

Right around the time the U-2 made its first flight, the US Air Force began researching the possibility of a satellite-based reconnaissance system. It already had a rocket: the Atlas intercontinental ballistic missile was at that point under development as a high priority, and if it could launch a mid-sized warhead across an ocean, it could launch a small satellite into orbit. Investigating the concept fell to the Ballistic Mission Division of the Air Force’s Air Research and Development Command.

The spy satellite program gradually took shape, and was formally approved in July as project WS-117L. It was to be a three-phase program: phase 1 would be a series of test launches on Thor missiles; phase 2 would focus on Atlas-boosted test launches; and phase 3 would be the operation phase. Slated to begin sometime in 1960, the operational phase was similarly divided into stages marked by increasingly sophisticated reconnaissance systems.

In the interest of speed, and spurred in part by the U-2 situation, the military sought to fast-track a photoreconnaissance program. The photographic aspect of the WS-117L system was separated and developed on its own as an interim system. This new program was called Corona, and it was run jointly by the Air Force, the Central Intelligence Agency, and the Advanced Research Projects Agency (ARPA) of the Department of Defense. It was the same managerial arrangement that had yielded the U-2.

A reconstructed camera of the kind launched on Corona missions in the late 1960s.

Developing the Mission

The Corona mission was intended to be simple. A spacecraft would launch on a two-stage rocket (a Thor for the first stage and an Agena for the second) on a polar trajectory from Vandenberg Air Force Base in California. Once in orbit, the Agena would flip 180 degrees to put the spacecraft in a rear-facing orientation. The orbiting, unpressurized satellite would stabilized around its three axes with respect to the Earth, keeping the camera pointed towards the same spot on the ground as it took pictures. The resolution of returned images was expected to be poor, but it would at least show the American government a previously unseen area of the Soviet Union.

Recovering the pictures was a much more difficult matter. Sending the information to Earth via radio signals wasn't sufficiently secure; if the Soviets interrupted a transmission, it would negate the reconnaissance gains. It was safest to recover the film manually, in a special recovery canister, as it fell from orbit.

The recovery canister, with the film sealed up inside, would be jettisoned from the Agena over the ocean. A heat shield would keep it safe during the first stages of reentry, and then a main parachute would deploy, slowing the canister for the most difficult part of the mission: an airplane flying over the canister would snag the parachute shrouds with a trapeze-like hook. From there, the canister could be pulled on board and flown to a secure location where the film would be developed and read.

If the mid-air recovery failed, the canister would float long enough for an attempted helicopter recovery. Then it would sink into the ocean to avoid falling into enemy hands.

Testing showed that this recovery method was, at best, difficult. Just 49 of 74 test canisters were recovered when they were slowed with a larger, personnel-type parachute, and only four of 15 canisters were recovered when an operational parachute was used. Modifications finally delivered a parachute that could slow a recovery canister’s fall enough for an airplane to have five chances to snag it from mid-air, making multiple passes if necessary.

A Discoverer satellite, snagged in mid-air.

The Cover Story

The Corona program was never formally approved in writing, save for a note scrawled by President Eisenhower to General Cabell, Deputy Directory of Central Intelligence, on the back on an envelope. The contract and work order for Corona’s construction were delivered to Lockheed on April 25, 1958. The contractor began designing the satellite’s systems on the 28th, and had a complete mockup finished on May 14. To keep things moving, the design was frozen on July 26.

Progress was swift throughout 1958, but Corona wasn't without its setbacks. Costs steadily increased, forcing program managers to appeal to the President directly for more money. There was also the issue of devising and planting a believable cover story, since no one involved in Corona was under any illusions that the launches would go unnoticed by the rest of the world.

The program’s fake details were presented in a press release issued on December 3, 1958. It explained that the early launches would be simple tests of a launch vehicle, while later launches would carry small probes into space to return data on the unknown environment. It also mentioned biological launches, like mice and primates, that would return important data for manned missions. ARPA’s involvement was explained by the agency’s need to develop a radiometric payload package to study navigation of space vehicles and test early warning systems. Interestingly, ARPA did develop such a system, though it was never part of Corona’s real plan.

Trying to Take Flight

The first Corona payloads were the Discoverer series of satellites, and the first launch attempt came on January 21, 1959. It was a test flight of the Thor-Agena launch vehicle, and it failed disastrously. An hour before launch, controllers applied power to the Agena stage. It inexplicably sprang to life and started going through its orbital activities while still on the launch pad. Explosive bolts detonated and small maneuvering rockets fired. The Agena fell straight down and settled into the faring on top of the Thor stage. The vehicle didn't fall over, but it was badly damaged.

Controllers were able to find and fix the problem, but not before finding more deficiencies. This was the way of things for Corona’s first year. Between April 1959 and June 1960, the program failed over and over. Three satellites failed to reach orbit while two entered into highly eccentric orbits. One recovery canister ejected from its orbital carrier far too early. In one instance, retrorockets malfunctioned, and one spacecraft was lost when it experienced damagingly low temperatures. There were also issues with the spin-stability system: designed to control the canister during reentry, the gas had a tendency to explode. Twice the onboard camera sprang to life but only worked briefly before failing, and one camera failed entirely. Those cameras that did work were plagued by film problems, as the acetate-based film had a tendency to tear in the vacuum of its capsule, jamming the camera’s mechanisms.

The recovery profile for a KH (Key Hole) spy satellite mission, a class that flew after Discoverer.


Engineers found fixes for all the problems that dogged the Corona Discoverer mission; they introducing a cold gas into the stability system and changing the acetate film for a polyester-based type. Pushing them inexorably forward was the overwhelming need for intelligence at the time. By 1960, the United States knew that the Soviets had missiles that could reach targets 5,000 miles away. The "missile gap" was fast becoming a very real concern.

In August of 1960, Corona found success with the Discoverer XII mission. It was a diagnostic flight without a camera on board, and though it landed some 250 miles away from its planned target, it was generally a success. All the fixes developed over the previous year had worked. The next mission, Discoverer XIII, launched on August 10 and was another success. Recovered from the ocean rather than grappled in mid-air, it was the first time a payload was recovered from orbit.

The first true success came a week later with Discoverer XIV on August 18, 1960. The camera sprung to life and exposed its full 20 pound load of film before transferring it into the recovery canister. The canister was ejected perfectly during the 17th orbit, and it fell to Earth right over its planned recovery zone. Captain Harold E. Mitchell of the 6593rd Test Squadron flying a C-119 snagged the canister’s parachute with his hook as it fell. The film was flown directly to a secure location where it was developed and studied.

The resolution of these pictures was lower than what the CIA could get from U-2 flights. The images were also damaged. Fittingly, an electrostatic discharge known as a corona had left density bars running across all the images. But they were valuable nonetheless, showing American leaders previously unseen details of the Soviet Union.

The Soviet launch site at Baikonur as seen by the Discoverer 14 satellite in orbit.

In the End

Corona was never expected to be a long-term program; it was always intended as a short term, high risk developmental program to meet the immediate need of photo reconnaissance. There was also the obvious drawback that orbital photography is much more limited in range than reconnaissance gathered with a piloted vehicle. The last Discoverer satellite, Discoverer XXXVIII, launched on April 18, 1962. But the Corona program continued with a variety of different reconnaissance payloads.

The Corona program was a milestone in US history. Those who developed the film-return photoreconnaissance satellite met and overcame a host technological challenges early in the space age and set the standards for all similar missions that followed: both physical recovery of film and spin stabilized reentry have been used on photoreconnaissance missions since.

The 145th and final Corona mission launched on May 25, 1972 and the final recovery came six days later on May 31, 1972. Over the course of its lifetime, Corona provided photographic coverage of about 750 million square miles of the Earth’s surface. These missions returned a significant amount of intelligence, including the truth about the missile gap, which was that there wasn't a gap at all. Without these insights, it’s possible the United States might have unwittingly entered into a Third World War.

Further Reading: "Corona: America's First Satellite Program" (Kevin Ruffner Ed.) 

For the latest tech stories, follow DVICE on Twitter
at @dvice or find us on Facebook