The First Discoverer Missions: America’s Original (Secret) Satellite Program

As the Cold War deepened in the 1950s, Americans developed an ever worsening case of paranoia about its primary adversary, the Soviet Union. These fears were heightened not only by Soviet propaganda but also by the lack of any substantive information about Soviet forces deep behind the borders of this enormous and secretive country. Based on the sobering conclusions drawn during White House meetings between Eisenhower and his top science advisors, the USAF issued General Operational Requirement No. 80 on March 16, 1955. This document authorized the development several new systems to obtain photographs over Soviet territory in order to provide useful intelligence about their activities well behind the Iron Curtain.

 

America’s First Satellite Program

One of the systems to be developed as result of this order was called “Aquatone”. Better known as the U-2, this high-altitude aircraft was designed to photograph the Soviet Union while flying out of reach of their air defense systems. While flights starting in July of 1956 returned much valuable intelligence, the aircraft was tracked from the start by Soviet radar and MiG fighters. USAF and CIA officials knew it would only be a matter of time before an incident stopped U-2 flights over the Soviet Union. With this and the threat of Soviet protests blowing the cover on this classified (and illegal) effort, Eisenhower eventually authorized only two dozen missions over Soviet territory through 1960.

The next new reconnaissance program became known as Weapon System-117L (WS-117L). Eventually run by the USAF’s Western Development Division of the Air Research and Development Command under General Bernard A. Schriever, this program called for the development of a reconnaissance satellite capable of returning detailed images of the Soviet Union from orbit. Since the end of World War II there had been a number of studies performed on satellites and their uses including reconnaissance. The one that had the greatest impact was the classified “Project Feed Back” study of the Rand Corporation published on March 1, 1954. The culmination of a series of USAF-sponsored studies at Rand, the report outlined the development a television-equipped satellite that would orbit 480 kilometers above the Earth taking images that were 600 kilometers across with a resolution of up to 44 meters. As this detailed study circulated through the USAF, it generated much interest and convinced many that a recon satellite was actually feasible.

With the need for more advanced studies, one year contracts were awarded to Lockheed, the Glenn L. Martin Co., and RCA under the codename “Pied Piper” in 1955. By July 1956 the development plan for the covert WS-117L satellite program (also known as the Advanced Reconnaissance System) was approved. This was two months before Vanguard was publicly chosen as America’s “official” satellite program (see “Vanguard TV-3: America’s First Satellite Launch Attempt”) making WS-117L the nation’s first (albeit secret) satellite program. In October 1956 Lockheed, who had also built the U-2, received the contract to develop the new recon satellite. But with an initial allocation of only $3 million (about $28 million in today’s money), satellite reconnaissance obviously still had a comparatively low priority with military leaders.

Despite the lack of funds, the planning and system development for WS-117L proceeded. After it was shown that a television-based system would be inadequate for reconnaissance, the clearly preferred option became a film readout system launched on an Atlas ICBM. This system would expose its film and develop it in orbit for subsequent scanning and transmission to Earth. The satellite would also carry a signal intelligence package and later infrared sensors would be added to detect missile launches. A relatively easy to develop, spin-stabilized photographic system was also studied. Launched on using a Thor intermediate-range ballistic missile (IRBM), this spacecraft would return exposed film inside a small return capsule and could provide an interim reconnaissance capability.

As a weapon system, the Thor IRBM was capable of sending a nuclear warhead over a range of 2,600 kilometers. The Thor, built for the USAF by the Douglas Aircraft Company (which merged in 1967 to become McDonnell-Douglas and three decades later merged with Boeing), was about 18.6 meters long and 2.4 meters in diameter at its base. The Thor incorporated a Rocketdyne MB-3 power plant burning RP-1 (a rocket propellant grade of kerosene) and liquid oxygen (LOX) to produce 668 kilonewtons of thrust. For the proposed spin-stabilized version of the photographic reconnaissance satellite, the three-stage Thor-Able launch vehicle was considered. Mounted on top of the Thor was a modified version of the second stage of the Vanguard launch vehicle also built by Douglas. This two-stage configuration was originally developed by the USAF for high-speed testing of proposed ICBM reentry vehicle designs. The third stage was the spin-stabilized X-248 rocket motor built by Allegany Ballistics Laboratory (which today is operated by ATK under contract from the US Navy) and was also developed by the Navy for the Vanguard rocket. This three-stage version of the Thor-Able, which could place about 160 kilograms of payload into a 480-kilometer high orbit, would initially be employed in 1958 to launch the first Pioneer probes to the Moon (see “The First Race to the Moon: Getting Off the Ground”) and would later serve as the basis of NASA’s Delta launch vehicle.

But as planning for WS-117L crawled along through 1957, it was becoming clear that the program was moving too slowly. Along with the anxiety generated by the first Sputnik launches (see “Sputnik: The Launch of the Space Age”), the project’s security was compromised with press reports referring to the supposedly secret project as “Big Brother” and “Spy in the Sky”. Wanting to avoid public discussion about the sensitive subject of overhead reconnaissance while at the same time accelerating recon satellite development, the Eisenhower administration had to take drastic action to get the program back on track and under wraps.

 

The Birth of Project Corona

Over the course of the 1958, the satellite reconnaissance program was completely restructured. Along with all the other military space programs, ARPA (Advanced Research Projects Agency) took control of WS-117L when it was founded in February of 1958. In a bid to get a satellite reconnaissance capability at the earliest possible date, ARPA secretly spun-off the interim Thor-based film-return part of WS-117L on February 28 as a separate, highly classified program. Called “Project Corona”, it would operate under the joint management of the CIA represented by Richard Bissell, Jr. and the USAF represented by Major General Osmond Ritland. Lockheed would continue as the prime contractor of this smaller, more focused effort. Fairchild Camera and Instrument Corp. was chosen as the subcontractor for the camera while General Electric got the nod to develop the return capsule or SRV (Satellite Reentry Vehicle).

In the end, the spin-stabilized concept was dropped in favor of a three-axis stabilized design in part because the latter had significantly more growth potential. The reconnaissance payload would be mounted on top of a Lockheed-built propulsion system which would serve as the final stage of the launch system as well as remain attached to the payload in orbit to provide various support functions and attitude control. This stage would employ a modified Bell 8000-series rocket engine (also known by the USAF designation, XLR-81) originally designed for a RATO (Rocket Assisted Take-Off) and auxiliary power system for the Convair B-58 “Hustler” bomber as well as for rocket-powered weapon pods for the supersonic bomber.

Initial versions of this stage would use the turbopump-fed Bell 8001 (XLR-81-BA-3) engine which burned a combination of JP-4 and IRFNA to generate 67.4 kilonewtons of thrust. JP-4 is a 50/50 mix of gasoline and kerosene used as jet fuel by the USAF at the time while IRFNA (inhibited red fuming nitric acid) is a mix of 84% nitric acid and 13% nitrogen tetroxide. Hydrofluoric acid in a concentration of 0.6% in the mix produces a protective fluoride coating on the tanks and plumbing which inhibits corrosion. The stage had a diameter of 1.5 meters and a length of 4.7 meters. With its full propellant load of about 2,900 kilograms, the stage would burn for up to 120 seconds to place itself and about 250 kilograms of useful payload into a low polar orbit. With both the stage and its engine originally nicknamed “Hustler”, in June of 1959 ARPA officially named this stage “Agena” which was one of the traditional names of the bright star β Centauri (which is located at the knees, or “genua” in Latin, of the Centaur). With the initial model of this stage dubbed the “Agena A”, this adaptable propulsion system eventually would be employed in other programs as well.

Mounted on top of the Agena A was Corona’s multi-section payload. The first part was an unpressurized compartment with a mass of about 115 kilograms. Shaped like a truncated cone, this compartment housed Corona’s model “C” camera based on the HYAK camera employed on high-altitude reconnaissance balloons. This camera, built by Fairchild and incorporating a lens system developed by Itek, would expose 70° wide photographic swaths perpendicular to the satellite’s ground track with a resolution of up to 12 meters from an orbit with a perigee of about 190 kilometers. As the camera’s 70 mm-format acetate film was exposed, it would be fed into the SRV at the top of the stack for return back to Earth.

With a diameter of 83 centimeters, a height of 69 centimeters and a mass of about 135 kilograms, the SRV carried stabilizing spin rockets which would fire after separation and a solid propellant retrorocket to start its descent from orbit. The exterior of the bowl-shaped SRV was coated with ablative material to serve as a heatshield during a nose first reentry. Inside was a “bucket” plated in gold to help reflect heat away from the cargo of film. After reentry, a parachute would open and pull the bucket clear of the heatshield for the final leg of the descent. Like the manned space program, a water recovery was preferred but there was a danger that Soviet submarines might get to the payload before American naval forces. To avoid this, recovery by air was proposed. A Fairchild C-119 “Flying Boxcar” cargo plane trailing a special trapeze-like rig would snag the parachute of the descending SRV allowing it to be winched inside.

Altogether, the first Corona recon satellites would have an orbital mass of about 800 kilograms or more. Because of payload limitations, only enough film and other consumables were carried for a one-day mission. Launch would take place south over the Pacific Ocean from a modified Thor launch pad at the missile facilities at Cooke Air Force Base in California (renamed Vandenberg AFB in October 1958) into a near-polar orbit that would allow Corona to view Soviet territory in daylight during southward overpasses. This orbit also permitted the return sequence to be monitored as Corona passed over Alaska with the tricky air-recovery taking place in the Pacific south of Hawaii.

As the USAF continued with the more complex WS-117L surveillance systems, development of the now independent Corona program proceeded at a frantic pace. While the small size of the Corona team made it easier to conceal the program from the public, by early 1959 launches would begin making it impossible to easily hide the program any longer. A cover story was needed and on December 3, 1958 ARPA announced the “Discoverer” program. According to ARPA, Discoverer was a test program to develop new spaceflight technologies as well as study the space environment. Biomedical experiments would also be flown and techniques to recover payloads from orbit would be perfected.

 

The First Discoverer Launches

The first 23.8-meter tall Thor-Agena A was erected on its pad at Vandenberg AFB on January 21, 1959 in preparation for the first Discoverer launch. Since the objective for this flight was simply to test the new rocket, only a light engineering payload was carried. With no SRV or camera, the in-orbit mass of the spent stage and its light payload was only about 590 kilograms. With Agena 1019 already filled with its JP-4 propellant, the hazardous task of loading the IRFNA oxidizer began prior to loading the propellants into the Thor booster.

At about the T-60 minute mark, Agena’s Guidance and Control Auto Sequencer was tested but an alarm immediately sounded in the blockhouse. The test had inadvertently activated the stage’s timer as it would during an actual launch. Due to the quick-thinking of blockhouse personnel, power was cut to “Hustler” and the JP-4 was quickly drained from the stage and back into its holding tanks before explosive bolts detonated ejecting covers over the payload’s sensors and separating the stage from its booster. With the firing of the ullage motors that would normally separate the stages during ascent, smoke began to pour from the side of the Thor-Agena on the pad as ground crews ran for cover.

Luckily, the Thor-Agena did not explode and stood steady on the pad. After the remaining propellants had been drained from the upper stage, the rocket was carefully destacked and the cause of the failure investigated. The Thor booster’s forward fuel tank bulkhead had been damaged by the incident and the missile was sent back to Douglas for repair. Having been exposed to the highly corrosive IRFNA which spilled during the incident, Agena 1019 was relegated to use in tests and fit checks of ground equipment for fear of hidden damage which could cause a failure in flight. In the end it was found that improper procedures were responsible for the accident prompting modifications in training and engineering practices.

The next launch attempt would again focus on testing the new rocket as well as simple engineering tests of the payload. With a total in-orbit mass of 728 kilograms including 18 kilograms of scientific instruments, no SRV was carried and no recovery would be attempted. Code named “Flying Yankee”, this launch would use Thor 163 to boost Agena 1022 and its payload into orbit. The six-hour, 45-minute countdown for the first launch attempt from Pad 75-3-4 (today known as Space Launch Complex 1W or SLC-1W) started at 6:00 AM PST on February 26, 1959 but was scrubbed when problems were encountered during the final minutes while pressurizing the tanks for Thor’s liquid nitrogen supply. Along with a fuel leak issue discovered after the scrub, the launch was rescheduled for two days later.

An abbreviated countdown was started at 8:00 AM PST on February 28, 1959 and this time the rocket lifted off without incident at 1:49:16 PM PST as reporters watched from some dunes three kilometers away. The MB-3 engine shutdown 161 seconds after launch with the vernier engine shutdown following nine seconds later. After a 168-second coast, Agena’s Bell 8001 engine ignited for a planned 96-second burn which would place Discoverer 1 into orbit. About 8½ minutes after launch, all contact with the satellite was lost. Based on Lockheed engineers’ initial calculations, it was believed that Discoverer 1 was in a 159 by 974 kilometer orbit with an inclination of 90.0°. Although no further telemetry had been unambiguously received from Discoverer 1 following its ascent, the USAF officially announced on March 6 that the new satellite was in orbit.

Initially, it was speculated that the antennas on Discoverer 1 were damaged when the payload shroud failed to cleanly separate thus preventing contact with ground stations. But as time wore on with no clear signal from the satellite or an unambiguous radar detection or even any visual sightings, it became evident that Discoverer 1 had failed to reach orbit. A detailed review of the Agena’s initial telemetry showed that the pitch gyro reference had shifted possibly due to the shock of engine ignition resulting in a negative pitch angle of 2.35°and a lower than desired trajectory. In addition, the telemetry from the second half of the Agena’s burn was poor or missing entirely. The calculated orbit parameters were based on the assumption that everything had gone as planned despite the low pitch angle. The inescapable conclusion, however, was that everything did not go as planned and that Discoverer 1 had most likely failed to achieve orbit and crashed near the South Pole instead.

 

Getting Discoverer Into Orbit

For the Discoverer 2 launch, tracking ships were stationed downrange so that telemetry from the Agena could be monitored during the entire ascent to orbit. For this flight, code named “Early Time”, Thor 170 would boost 748-kilogram Agena 1018 and its payload into orbit. Agena 1018 was the first of the series to use UDMH (unsymmetrical dimethyl hydrazine) instead of JP-4 as a fuel. UDMH in combination with IRFNA is hypergolic (or self-igniting) which greatly simplifed the design of the new Bell 8048 engine increasing its performance as well as its thrust to 68.9 kilonewtons.

While no camera was included in the payload as before, this time the SRV would be carried with the first-ever recovery of a payload from orbit attempted the day after launch. Originally, a biomedical payload of mice was to be carried on this flight, but a dummy payload was carried inside the SRV instead until a successful recovery could be proven. Discoverer 2 lifted off from Pad 75-3-4 at 1:18:39 PM PST on April 13, 1959 after a three hour, 19 minute delay due to technical issues, fog at the launch site and a passing train on the nearby Southern Pacific Railroad line. Although an engine relay issue on the Agena resulted in a slightly early engine cutoff, Discoverer 2 entered a 253 by 346 kilometer orbit with an inclination of 90.4°. After the earlier failed attempts of Discoverer 1 and the secret US Navy “NOTSNIK” program in 1958 (see “NOTSNIK: The First Air-Launched Satellite Attempts”), a satellite finally had been placed into polar orbit for the first time.

After reaching orbit, Discoverer 2 assumed its intended tail-first orientation and then stabilized its attitude using its cold gas thrusters making it the first three-axis stabilized satellite in orbit. Because Discoverer 2 was in a lower orbit than planned with a period of 90.5 minutes instead of 94 minutes, the onboard timer which controlled the satellites functions had to be reprogrammed during the second orbit as the satellite passed over its ground station in Kodiak, Alaska. Unfortunately, a ground equipment issue led to the timer not being reprogrammed properly. Not only did this error turn on the communication system at the wrong time during each orbit preventing further contact with the satellite, it would also release the SRV at the incorrect time preventing it from coming down in the expected recovery zone in the Pacific.

The SRV was ejected upon command of the misset onboard timer after 17 orbits and came down near Norway’s Spitsbergen Islands in the Arctic Ocean. Reportedly some residents of the island’s Longyearbyn settlement saw the descending capsule’s parachute and starburst recovery aid. Nearby this Norwegian community were three Soviet coal mining settlements, the largest of which was located in Barentsburg, operating for decades under contract with Norway. American officials received permission from the Norwegian government to begin low-key search operations on April 16 for fear that the Soviets might recover the wayward SRV first. Unfortunately, the SRV was not located by American forces and there is no indication from available historical records that the Soviets did either. Agena 1018 remained in orbit until its orbit decayed on April 26.

With yet another set of hard-learned lessons from the problems encountered with Discoverer 2, preparations were made for the launch of the Discoverer 3 mission, code named “Gold Duke”. For this launch, Thor 174 would lift Agena 1020 and its payload, with a total in-orbit mass of 843 kilograms, into a planned 225 by 780 kilometer orbit. The SRV for this mission, with a recovery once again to be attempted after 17 orbits, would carry four black mice from the C-57 strain commonly used in various types of medical research. Their “life cell” would keep each mouse inside a small cage about twice their size and supply them with a pure oxygen atmosphere at a pressure of about 570 millibars while maintaining the humidity at 50% and the temperature in the 18° to 26° C range. The objective was to determine the effects of 26 hours of space travel on the mice for the first attempted recovery of animals after orbital flight (unlike the dog launched into orbit by the Soviets in November 1957 where no recovery was possible – see “Sputnik 2: The First Animal in Orbit“).

The first launch attempt on May 21, 1959 was scrubbed due to coastal haze. Attempts on May 23 and 24 were cancelled due to failures of a helium regulator and a LOX switch, respectively. Following one of these aborted launches, the four mice were found to have died because they had ingested paint from their life cells (which was apparently more appetizing then their food) forcing the use of a backup “crew”. During the fourth launch attempt on June 4, the countdown was stopped when the humidity inside the life cell spiked to 100%. After it was determined that in fact the mice had urinated on the sensor (thus yielding a false reading), the countdown resumed. At 1:09:00 PM PDT, Discoverer 3 finally lifted off after a three hour, ten minute delay due to low clouds over the launch site. While Thor 174 operated as intended, Agena 1020 depleted its propellants early and failed to reach orbit dooming the mission and its four would be space travellers.

Because of the bad press and attention generated by the loss of the biological payload (and because of the problems caused by the tiny passengers which distracted from the program’s secret objectives), plans for Discoverer 4 and some subsequent missions to carry mice were scrapped as were plans to carry a monkey into orbit on Discoverer 7. Future studies would need to rely on NASA which planned to launch monkeys as part of its Mercury program with any USAF studies about the effects of spaceflight with live animals relegated to other programs including, eventually, the Manned Orbiting Laboratory or MOL (see “The USAF Manned Orbiting Laboratory Test Flight”).

 

The Long Road

With its biomedical payload deleted, the objective of the Discoverer 4 mission (code named “Long Road”) was to carry the first C-model camera into orbit. Given the classified designation “Mission 9001”, Thor 179 would use Pad 75-3-5 for the first time (known as SLC-1E today) to loft Agena 1023, camera C-1 and an SRV into orbit. With an in-orbit mass of 860 kilograms, this would be the heaviest satellite of the Discoverer program to date. Discoverer 4 lifted off at 8:48:00 AM PDT on June 25, 1959 but it failed to reach orbit when Agena 1023 shutdown early due to vortexing of fuel as its tank emptied during ascent.

Bad luck continued to haunt the Corona program for the rest of 1959. After five scrubbed launch attempts between July 28 and August 12, Discoverer 5 finally lifted off at 12:00:08 PM PDT on August 13, 1959 and successfully entered a 220 by 734 kilometer orbit with an inclination of 80.0°. Unfortunately, telemetry indicated that the temperature inside the payload compartment was abnormally low preventing camera C-2 from functioning. When recovery of the SRV was attempted the following day, it was not spotted by recovery forces and was presumed lost. Five months later, the wayward SRV was instead spotted in a 209 by 1,702 kilometer orbit. The SRV’s stabilizing spin rockets had failed to ignite causing the retrorocket motor to fire in the wrong direction placing it into a higher orbit where it remained until it decayed on February 11, 1961.

During the Discoverer 6 mission launched on August 19, 1959, the camera malfunctioned after two orbits and the SRV failed to separate for recovery. After a two-month stand down to address continuing issues with the spacecraft design and procedures, Discoverer 7 was launched on November 7 but once again problems were encountered. This time the satellite’s stabilization system failed when the attitude control propellant for Agena 1051 ran out after only two orbits preventing SRV recovery. Discoverer 8 was launched on November 20 but did only slightly better. The guidance system of Agena 1050 malfunctioned and placed Discoverer 8 into a highly eccentric 193 by 1,700 kilometer orbit. Once again, the camera failed to operate but the SRV did attempt to come home after 15 orbits. Unfortunately, the parachute failed during descent and the SRV was lost.

With a string of mission failures caused by a variety of malfunctions, many improvements would have to be made to the Corona spacecraft design. With much work ahead of them, another two-month stand down was in effect until the reliability of this potential intelligence tool could be improved.

 

Follow Drew Ex Machina on Facebook.

 

Related Video

Here is a video produced by Lockheed, entitled “T+Infinity”, about the early USAF satellite program and preparations leading to the launch of Discoverer 1.

 

 

Here is a Universal Newsreel, entitled “Space Mice: Plan ‘Live’ Satellite in 1^st Recovery Test”, about the about the mice to be carried on Discoverer 3.

 

 

Related Reading

“NOTSNIK: The First Air-Launched Satellite Attempts”, Drew Ex Machina, July 25, 2018 [Post]

“The USAF Manned Orbiting Laboratory Test Flight”, Drew Ex Machina, November 3, 2016 [Post]

“Vintage Micro: The First ELINT Satellites”, Drew Ex Machina, September 30, 2014 [Post]

 

General References

Dwayne A. Day, John M. Logsdon and Brian Latel, Eye in the Sky: The Story of the Corona Spy Satellite, Smithsonian Institution Press, 1998

Sarah A. Grassley, Agena Flight History as of 31 December 1967 – Volume 1 (Declassified), Space and Missile Systems Organization – Air Force Systems Command, June 1969

J.E. Lipp and R.M. Salter (ed), Project Feed Back Summary Report Volumes 1 & 2, Rand Corporation, March 1, 1954

Curtis Peebles, The Corona Project: America’s First Spy Satellites, Naval Institute Press, 1997

Robert Perry, A History of Satellite Reconnaissance Volume 1 – Corona (Declassified), The National Reconnaissance Office, October 1973 (Revised)

Kevin C. Ruffner (editor), Corona: America’s First Satellite Program, CIA, 1995

“Missiles and Spaceflight: First Discoverer”, Flight International, p. 312, March 6, 1959

“Missiles and Spaceflight: Discoverer”, Flight International, p. 343, March 13, 1959

“Missiles and Spaceflight: ‘Dead’ Discoverer”, Flight International, p. 427, March 27, 1959

“Missiles and Spaceflight: Discoverer 3 Launch”, Flight International, p. 794, June 12, 1959