Eye in the Sky: The Story of the CORONA Spy Satellites
Page 22
Unlike their American counterparts, Soviet engineers traditionally preferred to shield the most sensitive systems of their satellites from the space vacuum. Creating an artificial atmosphere inside pressurized containers was easier than designing new vacuum-rated equipment. While ensuring higher reliability, this approach had a price, since the resulting satellites were larger and heavier. The Zenit’s cameras worked in an artificial environment inside the return module. Temperature change substantially influenced image quality; the environmental control system was required to keep air temperature fluctuations to a maximum of no more than 0.1 degrees Celsius per hour.17 Since the optical transparency of the thick porthole glass was affected by internal pressure and external contrasting temperatures, the cameras had special adjustable lenses. Movement of the satellite also created a problem of image displacement—especially for lenses with long focal lengths. Image motion compensation was ensured through a slow motion of the film itself, which was adjustable to the rate of terrain motion in the camera’s view.18 The necessity for constant control and adjustment of the satellite’s equipment required a new multichannel telemetry and radio-command system with a capability to encrypt information. Radio exchange between Zenit and a ground control center was expected to be 10 times more intensive than that of the manned missions.19
The Zenit-2 final layout was very different from CORONA. The Soviet satellite was heavier (up to 4,740 kg) and was separate from the third stage of the launch vehicle. Its two modules were about 5 meters long and 2.43 meters in diameter. The nonpressurized instrument module housed electrical batteries, equipment for orientation, environmental control, and radio communications, and a liquid-propellant braking engine. That engine, called TDU-1, was not restartable, and the satellite was incapable of changing orbits. The instrument module was similar to that of Vostok, but with the addition of a cylindrical adapter between two conical shapes. Thus, its increased length was about 3 meters (2.43 meters in diameter) and the module weighed about 2,300 kg. A large spherical return (reentry) module (2.3 meters in diameter and about 2,400 kg in weight) was pressurized and contained cameras, film, parachutes, and an emergency destructive device to prevent the module from being captured by a foreign country.
The Zenit-2 return module accommodated several different camera arrangements. Initially it was equipped with one camera, called SA-20, with a 1,000mm focal length lens and a second camera, SA-10, with a 200mm lens. It also had a special photo-television system, called Baikal (after a Siberian lake). The latter was a film readout device that scanned photo images and transmitted them to the control center electronically. A similar device had been used by the Luna-3 probe in 1959 to produce pictures of the moon’s dark side. The film readout system was installed on six early test models of Zenit-2 in 1961–63. Four of those satellites (Kosmos-4, 7, 9, 12) tested that system in space (two were launch failures). The Baikal system was apparently an early attempt to obtain “near-real-time” image return, but its performance as a reconnaissance tool was not satisfactory. Hence, the later models of Zenit-2 relied on film cameras only.20
The operational camera arrangement for Zenit-2 was called Ftor-2R (“Fluorite”). It consisted of four fixed cameras: three 1,000mm SA-20 units and one SA-10 with a 200mm lens. The latter camera was used for low-resolution pictures—to provide a location reference for the high-resolution images taken at the same time. Removal of the readout system allowed the designers to increase the film load of each camera to 1,500 frames. Unlike CORONA’s panoramic images on narrow film strips, the SA-20 made pictures on square frames (apparently 300 × 300 mm). That made possible stereoscopic flat field photographs. The image format of the smaller camera probably was 180 × 180 mm.21 With an average flight altitude of 200 km, the main cameras scanned a terrain strip 60 × 180 km. Hence, each camera covered an area of 3,600 square kilometers (60 × 60 km) per frame. Dr. Frumkin, one of the Zenit-2 principal designers, stated that an area of about 10 million square kilometers (more than the entire territory of the United States) could be photographed in each mission.22 Various Russian publications provide conflicting data on the camera’s ground resolution. Officially it was said to be about 10–15 meters, but some sources have suggested much better resolution.23
The Soviet Zenit photoreconnaissance spacecraft. Unlike its American counterpart, Zenit returned both cameras and film to Earth. (Illustration courtesy of Peter Gorin)
The Zenit cameras were developed by the Krasnogorsk Optical-Mechanical Plant near Moscow.24 For decades that enterprise—one of the largest of that type in the USSR—was known for commercial production of the single-lens reflex cameras called Zenit. A freely available photo-camera was named after a super-secret military program!
In orbit, Zenit kept a horizontal orientation (its longitudinal axis was pointed along the flight path). Since the satellite flew in a low orbit, this orientation was useful to minimize drag from the atmosphere.25 The cameras looked through the side portholes of the return module. Pictures were taken in three different modes: continuous scanning (with all three cameras constantly taking photographs), a single picture (three-frame strip) along a ground track, or a single picture on either side of the flight path. For the latter mode, the satellite performed a controlled roll, using its vernier thrusters.
Zenit satellites were normally launched to a near-circular orbit with an average altitude of 200 × 350 km. Orbital inclination was 51.8–65 degrees (from Tyura-tam) and 72–81 degrees (from Plesetsk), depending on the target location. Due to the natural drift of the orbit, the altitude of each mission was preselected so that the satellite would cover the same area twice in seven days.26 The Zenit’s layout made it impossible to have several return capsules, as was done on later versions of CORONA. Thus, a standard mission lasted for eight days, although it could be prolonged to up to twelve days. At the same time, Zenit’s layout had a substantial economic advantage: the flown cameras were reused several times.27
Apart from overhead photography, Zenit-2 was designed to perform signals intelligence—interception of U.S. and NATO air defense radar frequencies. For that purpose, the satellite was equipped with a special system, Kust-12M (“Bush”). Intercepted information was recorded and stored aboard until the next communication session with the ground control center. Zenit-2 had a high-gain parabolic antenna, which was used for downloading electronic reconnaissance data.28
The USSR achieved its first successful space recovery on August 20, 1960—just nine days after the first American recovery of Discoverer XIII. That recovery of the Vostok prototype (1K-2), carrying two dogs, signaled a green light to both the manned Vostok and automated Zenit programs. Apparently, the Soviet leadership gave the manned program a higher priority, since it promised greater political advantages. However, in some of the launches of unmanned Vostok prototypes, experiments were conducted on photography from space.29
Zenit-2 had a rocky start: the launch vehicle failed on November 11, 1961, during the first Zenit-2 launch attempt. Contrary to Western reports, the spacecraft was not lost in Siberia but was destroyed by an emergency destructive device.30 The Zenit-2 test program resumed on April 26, 1962, when the second satellite was put into orbit under the false designation of Kosmos-4. The orientation system malfunctioned and the cameras did not produce satisfactory pictures, but reentry was successful after a three-day flight. The first reconnaissance pictures from space were obtained by the USSR from the third Zenit-2 (Kosmos-7, July 28–August 8, 1962). This was almost two years after CORONA returned its first images.
In the first two launches of Zenit-2, the launch vehicle used was the same as for the manned Vostok. That was the three-stage version of the R-7 ICBM (U.S. code name: SL-3). Later, Zenit-2 satellites were launched by another booster, the three-stage version of the improved R-7A ICBM.31 Twice (on June 1, 1962, and July 10, 1963) the rockets, along with their Zenit satellites, exploded on a launch pad at Tyura-tam, causing substantial damage to ground facilities and delays in test flights.
Laun
ch vehicles for Soviet and American photoreconnaissance spacecraft. (Illustration courtesy of Peter Gorin)
Western analysts usually consider 1962 the starting point of Soviet reconnaissance activities in space. This is not quite correct. It took nearly two years (November 1961–October 1963) and thirteen satellites (three of them failed at launch) to complete the test program.32 The last Zenit-2 test vehicle (Kosmos-20) was launched on October 18, 1963.33
SOVIET NATIONAL TECHNICAL MEANS OF VERIFICATION
That the United States was developing photoreconnaissance satellites was not a mystery to the Russians. In the early 1960s, Soviet propaganda viciously criticized the United States for “espionage from space.” Prime Minister Nikita Khrushchev promised to shoot down American spy satellites. This was not just a bluff—the USSR was developing antisatellite weapons. The Soviet Union even sponsored a UN General Assembly resolution on banning espionage from space. As the Zenit-2 system became operational, however, Soviet antisatellite propaganda was gradually put to an end. For almost twenty years the governments of the Soviet Union and the United States pretended that space reconnaissance systems did not exist. This so-called secrecy was intended to fool not the potential enemy but the general public. In major arms reduction agreements both superpowers not only acknowledged but stressed the ultimate importance of what they called “national technical means” for treaty verification.
In the USSR, the system of national technical means of verification—space reconnaissance—was declared operational on March 10, 1964, when the Ministry of Defense commissioned the Zenit-2 complex by a secret decree.34 Apart from the satellite itself, that complex included several other major elements.
Starting from Sputnik, all space missions in the USSR were prepared and launched by special military units—the “space troops”—of the Ministry of Defense. In December of 1959, those units were incorporated into the newly organized Strategic Rocket Forces (RVSN). The military also controlled all three rocket/space test ranges: Kapustin Yar (from 1947), Tyura-tam (from 1955), and Plesetsk (from 1957). Initially, Zenit missions were launched from Tyura-tam (better known as “Cosmodrome Baikonur”), where two launch pads for the R-7 rocket had been constructed. Starting from 1966, the spy missions more heavily relied on Plesetsk in the northern part of Russia. Plesetsk had four R-7 launch pads that had been used as operational R-7 ICBM sites until about 1965.
The reception and processing of data from all space objects was controlled by the military as well. From 1956, the NII-4 supervised the development of the ground-control station network. The most prominent role in that development was played by Colonel Yuri Mozzhorin, then-deputy commander of the NII-4. (Later, by then Lieutenant General Mozzhorin for thirty years headed TsNIIMASH [former NII-88]—the leading research center of the Soviet rocket/space industry.)35 The Command and Data Processing Center (KIK) was established in the city of Krasnoznamensk (code named “Golitsino-2”), Moscow Region, while the reception and relay stations were placed along the whole territory of the USSR—from the Black Sea to the Pacific. Initially that system was used to track ballistic missile tests, but it was later adapted for space missions. The launch control center at Tyura-tam and KIK were among the first in the USSR to be equipped with first-generation Soviet-made computers. The KIK became fully operational in May 1958, when it processed data from Sputnik 3. The reception and data processing network was substantially modified and enlarged in 1959–60 to accommodate the Vostok and Zenit programs. In addition to new ground stations, satellite tracking capability was enhanced by introduction of four floating stations aboard modified commercial ships.36
To ensure the timely recovery of cosmonauts and evacuation from the Vostok landing sites, a special Search and Rescue Service was organized by the Soviet Air Force in 1960. The service included seven groups of paratroopers (3–4 persons each) and was equipped with twenty large passenger planes and ten helicopters.37 The same Search and Rescue Service probably was also involved in recovery operations for Zenit which, unlike the CORONA reentry vehicles, came down on land.
Zenit-2 was the first military space system employed by the USSR Ministry of Defense on a regular basis. Growing military activity in space required greater coordination. Hence, the Central Directorate of Space Systems (TsUKOS) was established under the Strategic Rocket Forces in October 1964.38 It was a precursor of the modern Military Space Forces of Russia. The TsUKOS incorporated the launch facilities, NII-4, KIK with its network of control stations, and the RVSN Directorate of Satellites and Space Objects. It also became the major procurer of new military space systems.
According to Frumkin, the results of the first Zenit-2 photographs exceeded even the most optimistic expectations.39 The consumer of space reconnaissance data apparently was Soviet military intelligence—the GRU (Chief Intelligence Directorate) under the Soviet Armed Forces General Staff. Strategic reconnaissance in the interests of all services of the Soviet Armed Forces was the GRU’s prime objective. Some sources indicate the existence of a GRU Satellite Intelligence Directorate, which interpreted and analyzed space photos, providing results to the high political and military leadership.40 Other presumed photoreconnaissance consumers included the Topographical Directorate of the USSR Armed Forces General Staff and the Intelligence Department of the RVSN Commanding Staff. The first was responsible for military mapping; the latter needed satellite information for precision ICBM targeting.
To disguise military space missions, the Russians actually used an approach similar to the American Discoverer coverup, but they did so in a more elaborate way. They concealed not just one military project under a false cover, but all of them! They mixed various military and civilian satellites, failed probes, and other orbital objects, and called them all part of the Kosmos program. That program represented about 95 percent of Soviet space missions. At least one-third of all Kosmos satellites in reality were photoreconnaissance spacecraft.41 Nearly all launches were announced, but it required a painstaking analysis of orbital and other parameters to distinguish one Kosmos satellite from another. Very few actual scientific or applications satellites received different designations.
A comparison of the American CORONA and Soviet Zenit photoreconnaissance spacecraft. Some specific details of the Zenit remain unknown, but it was roughly comparable to CORONA and served the same role. (Illustration courtesy of Peter Gorin)
In 1968, a modified version—Zenit-2M—was introduced. There is no data available on the improvements to the new satellite. It is only known that the satellite incorporated a new launch vehicle and some improved systems from its sister spacecraft—Zenit-4 (discussed below). The signals intelligence equipment was apparently removed.
The Zenit-2 program remained operational for almost seven years and was retired after the launch of Kosmos-344 (May 12, 1970). Satellites of this type (including test vehicles and modified versions) were launched eighty-one times with fifty-eight completely successful missions. Eleven missions were only partially successful. Twelve launches failed, seven from launch vehicle failures and five from malfunctions of satellite systems.42
ZENIT-4
Even before Zenit-2 became operational, OKB-1 began development of a modified version, Zenit-4 (4K). Detailed technical data of Zenit-4 has not yet been revealed, except for a single cutaway drawing. Instead of three cameras with 1,000mm lenses, Zenit-4 apparently had one camera with a 3,000mm lens. In order to install such a large telescope into the reentry vehicle (which had a diameter of only 2,300 mm), the designers folded the camera’s light path, bending it backwards.43 It is also possible that this high-resolution variant was only one of several models of Zenit-4.
Although it had the same dimensions as its predecessor, Zenit-4 was much heavier—at least 5,500 kg (compared to 4,740 kg for Zenit-2). It required a new launch vehicle, which was developed at OKB-1 by 1963. That rocket was based on the R-7A ICBM with a new, more powerful, third stage (U.S. code name: SL-4). It was also used to launch later missions of the Zenit-2 and Zenit-2M.<
br />
Apparently the concept of a universal spy satellite with a combination of cameras and electronic sensors was rejected due to the short mission duration. In the mid-1960s, that role was transferred to smaller specialized satellites. Hence, Zenit-4 did not carry signals intelligence equipment.
Zenit-4 was the last photoreconnaissance satellite project developed by the OKB-1. In 1964, the Zenit and R-7 programs were transferred to the Branch #3 of OKB-1 under Dr. Dmitry Kozlov. Branch #3 in the city of Kuibyshev (currently Samara) had its own large factory and was responsible for serial production of OKB-1-designed missiles and spacecraft. That enterprise in 1974 became independent from the OKB-1 and is known now as TsSKB Progress (Central Specialized Design Bureau).44 Later models of Soviet photoreconnaissance satellites and most of their launch vehicles have been developed by the TsSKB.
The first Zenit-4 launch presumably occurred on November 16, 1963 (Kosmos-22). According to Western observers, the next satellite of that type was not launched until 1965 (Kosmos-59).45 At the same time there were probably three Zenit-4 missions in 1964 (Kosmos-30, 34, 45). Those satellites are identified in the West as photoreconnaissance vehicles but they are not in the Zenit-2 launch log and are therefore probably Zenit-4s.46 A modified version, Zenit-4M, began regular service in 1968. Another model—Zenit-4MK—was introduced at about the same time. The latter was probably used for mapping.