Some Notes on the early Discoverer flights

Sven Grahn

In early 1958, a few months after the Soviet Union launched Sputnik-1, President Eisenhower authorized the development of a top-priority reconnaissance satellite project operated by the CIA Directorate of Science & Technology with assistance from the US Air Force. It was used for photographic surveillance of the Soviet Union, China and other areas from 1960 until May 1972. The project's code name is CORONA, but to hide its true purpose, it was given the cover name Discoverer and described as a scientific research and technology development program.

There were 144 Corona satellites launched, 102 of which returned useful pictures. Detailed accounts of this program can be found here, here, and here.

The Discoverer program was very much in the media in my youth. This may seem strange for a highly classified program, but the cover story was successfully used and lots of technical details were released at the time. Here I will just touch upon a few items about this program that has caught my interest.

The flight of Discoverer-1

A USAF photo of Discoverer-1.

The first Discoverer launch attempt that took place at Vandenberg Air Force base in California on 21 January 1959 failed because of the launch vehicle. The second launch took place on 28 February 1959. Initially the satellite was reported to be in orbit, but I have always wondered what really happened. It seems that contact was lost with the spacecraft very early in the mission. In Swedish media the satellite was called "the whispering satellite". Even the Swedish radio telescope at R on the Onsala peninsula outside Gothenburg was reported to have picked up weak signals.

I asked Dwayne Day, the foremost expert on the U.S. reconnaissance satellite effort, about this mystery and here is his reply:

"The person I talked to was Frank Buzard ("buz ard") who was the test officer for Discoverer.  He was assigned to write up the reports on the launches after they were over.  Of course, Discoverer failed repeatedly for the first dozen missions or so (although a few reached orbit).  Buzard has a very good memory and remembers what specifically went wrong with most of the early missions. In the case of Discoverer 1, I think that the Air Force launched it and they had no bad telemetry during the launch, but the telemetry did not follow the entire Agena burn.  Based upon something like 50% of the Agena data being good, the Air Force assumed that the other 50% must be good and it reached orbit. So they issued a press release saying it had reached orbit.

'Well that's a bad assumption,' Buzard told me, and said that he did not like to assume anything as far as rockets were concerned.  So the Air Force said it was in orbit and then everybody listened to hear it fly overhead.  At the time, the idea of getting a radar skin paint on a satellite was still pretty much nonexistent.  They relied upon picking up the radio beacon.  But nobody got a good beep beep beep. They had one or two stations that claimed that they got a couple of beeps, but nothing more substantial. 

I think that Buzard looked at some data printouts and tried hard to interpret some spikes out of the background noise as a signal from the satellite.  He then turned this into a report saying what the Air Force wanted him to say, but he was never happy with it. And nobody else ever got good beeps from the satellite. Later, after Discoverer was clearly successful (more than any of the public realized at the time), those involved were willing to concede that they had probably bent the data to serve public relations purposes .."

The flight of Discoverer-2

Discoverer-2 was launched from Vandenberg Air Force base at at 2122 UT on 28 February 1959.  It entered an orbit between 239 and 346 km at an inclination of 89.9 degrees, a truly polar orbit - very rare in the history of space flight (2). It was widely reported as having come down on Svalbard and this event has even stimulated the Alistair MacLean's novel "Ice Station Zebra" (1963) and the movie with the same name (1968). The capsule was never found and the fate of the mission remains somewhat of a mystery. An internal CIA publication describes this flight as follows(3):

"... The [U.S.] Air Force reported on 15 April that plans to recover the capsule near Hawaii had been abandoned and that the capsule might descend somewhere in the Arctic. The announcement slightly understated the known facts. The capsule had ejected in the 17th orbit as planned, but a programming malfunction (actually a human programming error)  had caused the ejection sequence to be initiated too early. The capsule was down, probably somewhere in the near vicinity of the Spitsbergen islands north of Norway. In fact, there were later reports that the falling capsule had actually been seen by Spitsbergen residents. The Air Force announced on the 16th that the Norwegian government had authorized a search for the capsule which would begin the following day. Planes scoured the area, and helicopters joined the search on the 20th. Nothing was found, however, and the search was abandoned on the 23rd. [two lines deleted by censors] ..."

There were many reports and rumors that the capsule had been recovered by Soviet authorities using staff based at Soviet mining sites at Spitsbergen. However as(4) recounts

"... Not everyone was convinced that the capsule had been recovered. Major Buzard initially felt that the capsule had come down on Spitsbergen, but later changed his mind and suggested that the vehicle probably came down in the water and sank, or never came down at all, suffering the same fate as later satellites that had problems with their retro-rocket de-spin system. There was no evidence on the ground that  Discoverer-2 had been recovered. 'Witnesses' had been told exactly what to look for and may have merely repeated what they were told ..."

Dwayne Day has written an excellent account of the flight of Discoverer-2 and the attempts to recover it.

As one can see from the plot below, which shows the groundtrack for the climb out of Vandenberg on 13 April and the planned recovery orbit on 14 April (landing would have occurred on the 15th, Universal Time), the ground track of Discoverer-2 indeed passed over Spitsbergen, and also straight over my home town - Stockholm, Sweden!

Discoverer-13, the first recovery from orbit 

Flight profile

Discoverer-13 was launched in a fog bank from Vandenberg AFB at 2038 UT on 10 August 1960 and entered an orbit between 253.5 and 690 km at 82.85 degrees inclination. At 2115 UT (5) on 11 August the recovery sequence was initiated when the spacecraft was over the "Indian Ocean Station" in the Seychelles. The station at Kodiak island transmitted the capsule ejection command at 2311 UT on orders from the USAF Satellite Test Control Center at Sunnyvale, California - "The Blue Cube". The parachute opened at 2326 UT and the capsule descended about 480 km northwest of Hawaii. The capsule was recovered at 0130 UT on 12 August 1960 200 km north-west of Honolulu. A helicopter from Haiti Victory picked up the capsule.

The map below shows the ground trac for the launch out of Vandenberg AFB on 10 August 1960 and the two last passes over the Pacific the following day. The plots have been made with orbital element sets from the old NASA OIG web site. 

Dan Hill's eyewitness account of the recovery of Discoverer-13


USAF Generals T.D. White (left) and Bernard
Schriever, Col. Charles G. "Moose" Mathison,
and Discoverer-13 capsule. Note beacon antenna
- see below. USAF photo.
In late 2009 I received a very interesting message from Dan Hill (9) who participated in this historic event:

"... I was part of the 6593rd Test Squadron "Special", which we had formed in the fall of 1958 at Edwards AFB, CA., I occasionally search for any new data or input.   The purpose of our new squadron was to develop, test and practice aerial recoveries of future space capsules.  Our new Squadron relocated to Hickam AFB, Hawaii, in early December 1958, and we continued practice missions. 

Our aircrew was very close to Discoverer 13 space capsule when it re-entered Earth right on target on August 11, 1960.  We were prime recovery position, but unfortunately we were directed away from the capsule by one of the control aircraft and by the time our navigator realized they had sent us in the opposite direction, it was too late. 

He altered our course, we flew back to where we should have been all along, and found the capsule and parachute floating in the ocean.  By that time another one of our recovery aircrews were flying over the area as well.  We dropped dye markers and smoke bombs and flew surveillance for several hours till the U.S. Navy got there and retrieved the historical capsule and chute into a helicopter and flew them back to their ship ..." 

Discoverer-14, the first aerial recovery

A mere eight days after the launch of Discoverer-13 the next satellite is launched. Thus, Discoverer-14 was launched from Vandenberg AFB at 1957 UT on 18 August 1960. The film capsule from Discoverer-14 was snagged by a n airplane over the Pacific at 2314 UT on 19 August 1960. In 1999 Bill Obenauf wrote to me (6)  about Disoverer-14:

"I worked on Corona at Vandenberg 1959 - 1967. Your pictures are very familar. In my desk drawer, under this keyboard, is a bolt from Discoverer 14, the first capsule that brought back film. Thanks for bringing back many memories."  

Snagging a capsule with a C-119 plane. This picture
shows a practice run. USAF photo

Dan Hill wrote to me in 2009  (9) and described the recovery of Discoverer-14 and his part in it:

"... The following week, our aircrew was way, way down range in the outfield of the recovery area, the least desirable recovery position for Discoverer-14.  As we all waited and waited, our navigator finally received a signal, which became stronger.  He directed our aircrew to the signal, we scanned the skies overhead and finally our aircraft commander saw the orange/silver parachute and capsule high above us.  As you know, history was in the making again, when on our third and last possible aerial recovery attempt, we snagged the parachute with capsule attached [at 2600 meters altitude (11) ], had it in tow and our winch operator brought them to the rear of our aircraft.  At that point, we Loadmasters lifted and pulled the first historical aerial recovered chute and space capsule on board our now famous aircraft.

I just turned 21 years old (the youngest of our aircrew), the day that Discoverer-14 lifted off and when we returned to Hickam AFB the next afternoon on August 19, 1960 with the capsule, I was advised by our wing commander in front of the spectators and press people as I was presented a birthday cake: 'Danny, yesterday they sent your birthday present into orbit and today it was delivered to you'.   We were naturally, very proud of our historical catch, but the fact that now, Discoverer-14 space capsule had been officially designated my birthday present and my picture with me and cake went out to various parts of the world.  That made me that much prouder ..."

Newspaper clipping showing Dan Hill receiving his
birthday cake.

The primary recovery area encompassed a 320 x 96 km rectangle, and six C-119's and a C-130 flew within this "ballpark" area. Three other C-119's, including the actual recovery aircraft, patrolled an "outfield" area embracing an additional 640 km. the capsule was hauled aboard the aircraft 30 minutes from the time the navigator had first reported a steady signal from the capsule's radio beacon (11).

Air crew of C-119 (tail nr 18037) unload the canister with the Discoverer-14
capsule inside. Dan Hill is the righthand person standing on the ground.

Discoverer-14 recovery crew and its commander Captain Harold E Mitchell
being praised by Pacific Air Force Commander General Emmett O'Donnell. 
The stand in front of their C-119 with tail number 18037. This plane is pre-
. Dan Hill is at the far right of the line of airman. USAF photo.

Discoverer telemetry, command, and tracking systems

Telemetry and command system

Telemetry antennas on an Agena-A. USAF photo.

The Agena stage/spacecraft used an FM/FM telemetry system initially operating on P-band (215-260 MHz) in the VHF range, called "Type V" (1). Orbital determination and  command reception was provided by an S-band beacon transponder. The telemetry system relayed 85 data points(1) .

Bill Obenauf, who worked for Lockheed at Vandenberg Air Force Base, gave more details about the early telemetry system of Corona (10):

"The Agena telemetry was FM/FM during the first few years. It used 17 channels of sub-carrier oscillators modulating an FM transmitter. A few of the oscillators were commutated using 60 point motor driven 'switches'. Some oscillators monitored film rollers in the camera. One of the commutator points monitored film quantity in the capsule."

Bill Obenauf also described the command link (10) :

"Commands were sent up from tracking stations to the Agena via the S-band link to the tracking beacon. Two pulses, ten microseconds spacing, were decoded by the agena beacon to generate a reply to the ground tracking radar station. A third pulse, between the other two, was modulated position-wise at audio rates to generate commands to the Agena. S-band was 10 cm [SG: 3 GHz]."

In (1) the Philco radar transponder is said to have responded to a combination of two audio tones of the middle pulse. Fifteen commands from the decoder were possible by using six audio frequencies in all possible combinations of two.

There was a mechanical sequencer for executing commands out of sight of grund stations. It had two magazines of punched Mylar tape. The programmer could execute twenty-six commands repeatedly for 256 orbits. There was also a programmer for 52 preprogrammed commands and a flight loadable storage register for 32 commands(1) . In addition to the main command system there was also a UHF back-up command system.

Telemetry and recovery beacon frequency

We can learn something about the frequencies used for the VHF telemetry system by referring to observations made by radio trackers at the time, and "leaks" in the literature. The telemetry frequencies used were in P-band and the most common telemetry frequency during early flights was 237.8 MHz. It is not easy to locate antennas for this frequency band on pictures of Agena vehicles. Only in a picture of what purportedly is Discoverer-1 can a ring of antennas be seen around the engine compartment of the Agena rocket stage which formed part of the spacecraft. These antennas seen in the picture (the stripes on the antennas coloured red in the picture) below are probably so-called quadra-loop antennas and the size is compatible with frequencies in P-band.

Bill Obenauf also wrote:

"... I asked a friend, who also worked on Corona.  After thinking about the Agena telemetry radio carrier frequency 238 MHz  comes to mind Anyhow, the Corona capsule beacon used a carrier of 235 Megacycles (we didn't use mega hertz in those days). It was amplitude  modulated by a tone that sounded like weep - weep - weep. The period of the "weep" varied depending on events that happened during the recovery, like  parachute deploy ..." 

We can also get a glimpse of what the Agena telemetry sounded like by listening to Agena signals on P-band picked up by satellite trackers at the University of Florida Student Tracking Station in the 60's while they monitored telemetry from Agena target vehicles for the Gemini program.  

Radar transponder






Waveguide antenna seen on an Agena stage on display at
the National Air and Space Museum in Washington D.C.

Gemini-Agena had two radar transponders, one with a downlink in S-band at 2910 MHz and another in C-band with a downlink at 5765 MHz. A few years ago, there was an Agena on display at the National Air and Space Museum in Washington D.C. It carried device that looked like the sketch on the right. The size and general look of this device suggest that it is a waveguide antenna for C-band, possibly the radar transponder frequency 5765 MHz.

Ground station operations 

On a listserver, I cannot recall which, I found the following personal memory from someone working at USAF tracking stations in the early 60's (7) :

"... I worked at several of the Air Force tracking stations for these things including Vandenberg AFB in the early 60's. The satellite command officer, or civilian was told what commands to send after we turned the bird on. Any indications they got from the consoles was minimal and certainly didn't let you know about film. It would be a reading, like: meter number 5 reads 50 percent. 

For each 'pass' we set up the decommutator outputs to drive specific meters. We had no idea what these meters were reading, but in real time we gave out those numbers. So the sequence would be:'KODI Transmit command 1020'. Satellite turns on-telemetry, decommutators lock onto the downlink, signal-meters start to act. During the short pass we give the satellite command console operator the read-outs. The console operator passes them to the 'Blue Cube'.  He then checks what is required and then sends back a response; 'Indi, send command 2050, command 1010 and command 1230.' ...

... The console operator then dials up those commands, they are picked from a library of commands back in the computer room running a SperryRand computer, and then he presses transmit. He then tells Palo Alto the commands he sent. The only site that sent deorbit commands was the KODI site at Kodiak, Alaska.  The satellite would be heading south toward Hawaii the pickup area ..."

Antennas on the re-entry vehicle

The eleventh International Astronautical Congress (IAC) was held in Stockholm in August 1960 as I have described in another article at this web site. As you can read there a very detailed mock-up (actually a "mechanical analog") of the Discoverer re-entry vehicle (including retro-rocket) and film capsule was on display at the Congress and I (at the age of fourteen) was given the task of describing it to the general public during "open day" on the Congress' last day (see picture below), actually some fourty hours after that the capsule from Discoverer-14 was snagged by a a C-119 plane.

I hold my hand on the Discoverer film capsule on display at the eleventh International Astronautical
Congress. The venue is the Royal Institute of Technology in Stockholm. The date is 21 August 1960. I have
struggled to bring myself out of the horribly bad contrast of this picture.

I also took some close-up pictures of the display. The picture below shows the heat shield with the retro-rocket ona sheet metal cone on the lower left and part of the film capsule on the upper right.

Discoverer heat shield with retro-rocket on top and film capsule (upper
right) on display at IAC 1960 in Stockholm.

By blowing up the little part of this picture that shows the film capsule and turning it 90 degrees we can examine some deails on top of the capsule.

Details on top of the Discoverer film capsule on display at the IAC 1960 in Stockholm.

I can state with certainty that device on the left is a strobe light and the device on the right looks like a typical loop antenna half a wavelength long (see this article) with one end grounded in the capsule chassis and the other end insulated from chassis and fed by a transmitter. The size matches relatively well the beacon frequency 235 MHz. But what is the device marked "?"? It looks like an antenna too with one end grounded and the other fed by a radio device. But what is the frequency and purpose? Could it for a frequency in the range 1-2 GHz and is it a radar transponder antenna? Interestingly this device can not be found on pictures of the recovered Discoverer-13 capsule. On one side of this capsule one can see the strobe light an a hole where "?" is seen above. On the other side of the capsule there is no strobe light either, just a hole. The "?" can be seen in a picture by James Plummer.

Re-entry and recovery sequence

In (11) Some basic numbers are given:

Orbital parameters of early Discoverer missions

The obital elements of early Discoverer satellites were similar between flights, but not identical. To able to find a pattern I plotted the mean motion (revolutions/day) against the inclination for the first 39 launches that made it to orbit. (See the figure below) The inclination varied less than the mean motion. It seems that the orbital period (together with the inclination) was used to achieve coverage of differenet areas. To prove this and extenisve study of the coverage over the Soviet Union of each of these missions. I am not sure the photo coverage of all these missions have been made public.


Back to Space History Notes