Although at the end of World War II the United States had captured large quantities of German photos and documents on the Soviet Union, this material was rapidly becoming outdated. The main source of current intelligence on the Soviet Union’s military installations was interrogation of prisoners of war returning from Soviet captivity. To obtain information about Soviet scientific progress, the intelligence community established several programs to debrief German scientists who had been taken to the Soviet Union after the end of the war but were now being allowed to leave.
Interrogation of returning Germans offered only fragmentary information, and this source could not be expected to last much longer. As a result, in the late 1940s, the US Air Force and Navy began trying to obtain aerial photography of the Soviet Union. The main Air Force effort involved Boeing RB-47 aircraft (the reconnaissance version of the B-47 jet-propelled medium bomber) equipped with cameras and electronic “ferret” equipment that enabled aircrews to detect tracking by Soviet radars. At that time the Soviet Union had not yet completely ringed its borders with radars, and much of the interior also lacked radar coverage. Thus, when the RB-47s found a gap in the air-warning network, they would dart inland to take photographs of any accessible targets. These “penetration photography” flights (called SENSINT—sensitive intelligence—missions) occurred along the northern and Pacific coasts of Russia. One RB-47 aircraft even managed to fly 450 miles inland and photograph the city of Igarka in Siberia. Such intrusions brought protests from Moscow but no Soviet military response.
In 1950 there was a major change in Soviet policy. Air defense units became very aggressive in defending their airspace, attacking all aircraft that came near the borders of the Soviet Union. On 8 April 1950, Soviet fighters shot down a US Navy Privateer patrol aircraft over the Baltic Sea. Following the outbreak of the Korean war in June 1950, the Soviet Union extended its “severe air defense policy” to the Far East. In the autumn of 1951, Soviet aircraft downed a twin-engine US Navy Neptune bomber near Vladivostok. An RB-29 lost in the Sea of Japan on 13 June 1952 was probably also a victim of Soviet fighters. The United States was not the only country affected by the new aggressive Soviet air defense policy; Britain and Turkey also reported attacks on their planes.
The Soviet Union’s air defense policy became even more aggressive in August 1952, when its reconnaissance aircraft began violating Japanese airspace over Hokkaido, the northernmost Japanese home island. Two months later, on 7 October 1952, Soviet fighter aircraft stalked and shot down a US RB-29 flying over Hokkaido. Aerial reconnaissance of the Soviet Union and surrounding areas had become a very dangerous business.
Despite the growing risks associated with aerial reconnaissance of the Soviet Bloc, senior US officials strongly believed that such missions were necessary. The lack of information about the Soviet Union, coupled with the perception that it was an aggressive nation determined to expand its borders—a perception that had been greatly strengthened by the Soviet-backed North Korean invasion of South Korea in June 1950—increased US determination to obtain information about Soviet intentions and capabilities and thus reduce the danger of being surprised by a Soviet attack.
New Approaches to Photoreconnaissance
While existing Navy and Air Force aircraft were flying their risky reconnaissance missions over the Soviet Union, the United States began planning for a more systematic and less dangerous approach using new technology. One of the leading advocates of the need for new, high-altitude reconnaissance aircraft was Richard S. Leghorn, a Massachusetts Institute of Technology graduate and employee of Eastman Kodak who had commanded the Army Air Forces’ 67th Reconnaissance Group in Europe during World War II. After the war he returned to Kodak but maintained his interest in photoreconnaissance. Leghorn strongly believed in the need for what he called pre-D-day reconnaissance, that is, reconnaissance of a potential enemy before the outbreak of actual hostilities, in contrast to combat reconnaissance in wartime. In papers presented in 1946 and 1948, Leghorn argued that the United States needed to develop such a capability, which would require high-altitude aircraft and high-resolution cameras. The outbreak of the Korean war gave Leghorn an opportunity to put his ideas into effect. Recalled to active duty by the Air Force, Lieutenant Colonel Leghorn became the head of the Reconnaissance Systems Branch of the Wright Air Development Command at Dayton, Ohio, in April 1951.
In Leghorn’s view, altitude was the key to success for overhead reconnaissance. Since the best Soviet interceptor at that time, the MIG-17, had to struggle to reach 45,000 feet,6 Leghorn reasoned that an aircraft that could exceed 60,000 feet would be safe from Soviet fighters. Recognizing that the fastest way to produce a high-altitude reconnaissance aircraft was to modify an existing aircraft, he began looking for the highest flying aircraft available in the Free World. This search soon led him to a British twin-engine medium bomber—the Canberra—built by the English Electric Company. The Canberra had made its first flight in May 1949. Its speed of 469 knots (870 kilometers per hour) and its service ceiling of 48,000 feet made the Canberra a natural choice for high-altitude reconnaissance work. The Royal Air Force quickly developed a reconnaissance version of the Canberra, the PR3 (the PR stood for photoreconnaissance), which began flying in March 1950.
At Leghorn’s insistence, the Wright Air Development Command invited English Electric representatives to Dayton in the summer of 1951 to help find ways to make the Canberra fly even higher. By this time the Air Force had already adopted the bomber version of the Canberra, which the Glenn L. Martin Aircraft Company was to produce under license as the B-57 medium bomber. Leghorn and his English Electric colleagues designed a new Canberra configuration with very long high-lift wings, new Rolls-Royce Avon-109 engines, a solitary pilot, and an airframe that was stressed to less than the standard military specifications. Leghorn calculated that a Canberra so equipped might reach 63,000 feet early in a long mission and as high as 67,000 feet as the declining fuel supply lightened the aircraft. He believed that such a modified Canberra could penetrate the Soviet Union and China for a radius of 800 miles from bases around their periphery and photograph up to 85 percent of the intelligence targets in those countries.
Leghorn persuaded his superiors to submit his suggestion to the Pentagon for funding. He had not, however, cleared his idea with the Air Research and Development Command, whose reconnaissance division in Baltimore, headed by Lt. Col. Joseph J. Pellegrini, had to approve all new reconnaissance aircraft designs. Pellegrini’s unit reviewed Leghorn’s design and ordered extensive modifications. According to Leghorn, Pellegrini was not interested in a special-purpose aircraft that was only suitable for covert peacetime reconnaissance missions, for he believed that all Air Force reconnaissance aircraft should be capable of operating under wartime conditions. Pellegrini therefore insisted that Leghorn’s design meet the specifications for combat aircraft, which required heavily stressed airframes, armor plate, and other apparatus that made an aircraft too heavy to reach the higher altitudes necessary for safe overflights of the Soviet Bloc. The final result of Leghorn’s concept after its alteration by Pellegrini’s staff was the RB-57D in 1955, whose maximum altitude was only 64,000 feet. Meanwhile Leghorn, frustrated by the rejection of his original concept, had transferred to the Pentagon in early 1952 to work for Col. Bernard A. Schriever, Assistant for Development Planning to the Air Force’s Deputy Chief of Staff for Development.
In his new position Leghorn became responsible for planning the Air Force’s reconnaissance needs for the next decade. He worked closely with Charles F. (Bud) Wienberg—a colleague who had followed him from Wright Field—and Eugene P. Kiefer, a Notre Dame-educated aeronautical engineer who had designed reconnaissance aircraft at the Wright Air Development Center during World War II. All three of these reconnaissance experts believed that the Air Force should emphasize high-altitude photoreconnaissance.
Underlying their advocacy of high-altitude photoreconnaissance was the belief that Soviet radars would not be able to track aircraft flying above 65,000 feet. This assumption was based on the fact that the Soviet Union used American-built radar sets that had been supplied under Lend-Lease during World War II. Although the SCR-584 (Signal Corps Radio) target-tracking radar could track targets up to 90,000 feet, its high power consumption burned out a key component quickly, so this radar was normally not turned on until an early warning radar had detected a target. The SCR-270 early warning radar could be left on for much longer periods and had a greater horizontal range (approximately 120 miles) but was limited by the curvature of the earth to a maximum altitude of 40,000 feet. As a result, Leghorn, Kiefer, and Wienberg believed that an aircraft that could ascend to 65,000 feet before entering an area being swept by the early warning radar would go undetected, because the target-tracking radars would not be activated.9
The problem with this assumption was that the Soviet Union, unlike Britain and the United States, had continued to improve radar technology after the end of World War II. Even after evidence of improved Soviet radar capabilities became available, however, many advocates of high-altitude overflight continued to believe that aircraft flying above 65,000 feet were safe from detection by Soviet radars.
The Air Force Search for a New Reconnaissance Aircraft
With interest in high-altitude reconnaissance growing, several Air Force agencies began to develop an aircraft to conduct such missions. In September 1952, the Air Research and Development Command gave the Martin Aircraft Company a contract to examine the high-altitude potential of the B-57 by modifying a single aircraft to give it long, high-lift wings and the American version of the new Rolls-Royce Avon-109 engine. These were the modifications that Richard Leghorn had suggested during the previous year.
At about the same time, another Air Force office, the Wright Air Development Command (WADC) in Dayton, Ohio, was also examining ways to achieve sustained flight at high altitudes. Working with two German aeronautical experts—Woldemar Voigt and Richard Vogt—who had come to the United States after World War II, Air Force Maj. John Seaberg advocated the development of a new aircraft that would combine the high-altitude performance of the latest turbojet engines with high-efficiency wings in order to reach ultrahigh altitudes. Seaberg, an aeronautical engineer for the Chance Vought Corporation until his recall to active duty during the Korean war, was serving as assistant chief of the New Developments Office of WADC’s Bombardment Branch.
By March 1953, Seaberg had expanded his ideas for a high-altitude aircraft into a complete request for proposal for “an aircraft weapon system having an operational radius of 1,500 nm [nautical miles] and capable of conducting pre- and post-strike reconnaissance missions during daylight, good visibility conditions.” The requirement stated that such an aircraft must have an optimum subsonic cruise speed at altitudes of 70,000 feet or higher over the target, carry a payload of 100 to 700 pounds of reconnaissance equipment, and have a crew of one.
The Wright Air Development Command decided not to seek proposals from major airframe manufacturers on the grounds that a smaller company would give the new project a higher priority and produce a better aircraft more quickly. In July 1953, the Bell Aircraft Corporation of Buffalo, New York, and the Fairchild Engine and Airplane Corporation of Hagerstown, Maryland, received study contracts to develop an entirely new high-altitude reconnaissance aircraft. In addition, the Glenn L. Martin Company of Baltimore was asked to examine the possibility of improving the already exceptional high-altitude performance of the B-57 Canberra. By January 1954 all three firms had submitted their proposals. Fairchild’s entry was a single-engine plane known as M-195, which had a maximum altitude potential of 67,200 feet; Bell’s was a twin-engine craft called the Model 67 (later the X-16), which had a maximum altitude of 69,500 feet; and Martin’s design was a big-wing version of the B-57 called the Model 294, which was to cruise at 64,000 feet. In March 1954, Seaberg and other engineers at Wright Field, having evaluated the three contending designs, recommended the adoption of both the Martin and Bell proposals. They considered Martin’s version of the B-57 an interim project that could be completed and deployed rapidly while the more advanced concept from Bell was still being developed.
Air Force headquarters soon approved Martin’s proposal to modify the B-57 and was very much interested in the Bell design. But word of the competition for a new reconnaissance airplane had reached another aircraft manufacturer, the Lockheed Aircraft Corporation, which submitted an unsolicited design.
Lockheed had first become aware of the reconnaissance aircraft competition in the fall of 1953. John H. (Jack) Carter, who had recently retired from the Air Force to become the assistant director of Lockheed’s Advanced Development Program, was in the Pentagon on business and dropped in to see Eugene P. Kiefer, an old friend and colleague from the Air Force’s Office of Development Planning (more commonly known as AFDAP from its Air Force office symbol). Kiefer told Carter about the competition for a high-flying aircraft and expressed the opinion that the Air Force was going about the search in the wrong way by requiring the new aircraft to be suitable for both strategic and tactical reconnaissance.
Immediately after returning to California, Carter proposed to Lockheed Vice President L. Eugene Root (previously the top civilian official in the Air Force’s Office of Development Planning) that Lockheed also submit a design. Carter noted that the proposed aircraft would have to reach altitudes of between 65,000 and 70,000 feet and correctly forecast, “If extreme altitude performance can be realized in a practical aircraft at speeds in the vicinity of Mach 0.8, it should be capable of avoiding virtually all Russian defenses until about 1960.” Carter added, “To achieve these characteristics in an aircraft which will have a reasonably useful operational life during the period before 1960 will, of course, require very strenuous efforts and extraordinary procedures, as well as nonstandard design philosophy.” Some of the “nonstandard” design characteristics suggested by Carter were the elimination of landing gear, the disregard of military specifications, and the use of very low load factors. Carter’s memorandum closed with a warning that time was of the essence: “In order that this special aircraft can have a reasonably long and useful life, it is obvious that its development must be greatly accelerated beyond that considered normal.”
Lockheed’s senior officials approved Carter’s proposal, and early in 1954 the corporation’s best aircraft designer—Clarence L. (Kelly) Johnson—began working on the project, then known as the CL-282 but later to become famous under its Air Force designator—the U-2. Already one of the world’s leading aeronautical engineers, Kelly Johnson had many successful military and civilian designs to his credit, including the P-38, P-80, F-104, and Constellation. Johnson quickly came up with a radical design based upon the fuselage of the F-104 jet fighter but incorporating a high-aspect-ratio sailplane wing. To save weight and thereby increase the aircraft’s altitude, Johnson decided to stress the airframe to only 2.5 units of gravity (g’s) instead of the military specification strength of 5.33 g’s. For the power plant he selected the General Electric J73/GE-3 nonafterburning turbojet engine with 9,300 pounds of thrust (this was the same engine he had chosen for the F-104, which had been the basis for the U-2 design). Many of the CL-282’s design features were adapted from gliders. Thus, the wings and tail were detachable. Instead of a conventional landing gear, Johnson proposed using two skis and a reinforced belly rib for landing—a common sailplane technique—and a jettisonable wheeled dolly for takeoff. Other features included an unpressurized cockpit and a 15-cubic-foot payload area that could accommodate 600 pounds of sensors. The CL-282’s maximum altitude would be just over 70,000 feet with a 2,000-mile range. Essentially, Kelly Johnson had designed a jet-propelled glider.
Early in March 1954, Kelly Johnson submitted the CL-282 design to Brig. Gen. Bernard A. Schriever’s Office of Development Planning. Eugene Kiefer and Bud Wienberg studied the design and recommended it to General Schriever, who then asked Lockheed to submit a specific proposal. In early April, Kelly Johnson presented a full description of the CL-282 and a proposal for the construction and maintenance of 30 aircraft to a group of senior Pentagon officials that included Schriever’s superior, Lt. Gen. Donald L. Putt, Deputy Chief of Staff for Development, and Trevor N. Gardner, Special Assistant for Research and Development to the Secretary of the Air Force. Afterward Kelly Johnson noted that the civilian officials were very much interested in his design but the generals were not.
The CL-282 design was also presented to the commander of the Strategic Air Command (SAC), Gen. Curtis E. LeMay, in early April by Eugene Kiefer, Bud Wienberg, and Burton Klein from the Office of Development Planning. According to Wienberg, General LeMay stood up halfway through the briefing, took his cigar out of his mouth, and told the briefers that, if he wanted high-altitude photographs, he would put cameras in his B-36 bombers and added that he was not interested in a plane that had no wheels or guns. The general then left the room, remarking that the whole business was a waste of his time.
Meanwhile, the CL-282 design proceeded through the Air Force development channels and reached Major Seaberg at the Wright Air Development Command in mid-May. Seaberg and his colleagues carefully evaluated the Lockheed submission and finally rejected it in early June. One of their main reasons for doing so was Kelly Johnson’s choice of the unproven General Electric J73 engine. The engineers at Wright Field considered the Pratt and Whitney J57 to be the most powerful engine available, and the designs from Fairchild, Martin, and Bell all incorporated this engine. The absence of conventional landing gear was also a perceived shortcoming of the Lockheed design.
Another factor in the rejection of Kelly Johnson’s submission was the Air Force preference for multiengine aircraft. Air Force reconnaissance experts had gained their practical experience during World War II in multiengine bombers. In addition, aerial photography experts in the late 1940s and early 1950s emphasized focal length as the primary factor in reconnaissance photography and, therefore, preferred large aircraft capable of accommodating long focal-length cameras. This preference reached an extreme in the early 1950s with the development of the cumbersome 240-inch Boston camera, a device so large that the YC-97 Boeing Stratocruiser that carried it had to be partially disassembled before the camera could be installed. Finally, there was the feeling shared by many Air Force officers that two engines are always better than one because, if one fails, there is a spare to get the aircraft back to base. In reality, however, aviation records show that single-engine aircraft have always been more reliable than multiengine planes. Furthermore, a high-altitude reconnaissance aircraft deep in enemy territory would have little chance of returning if one of the engines failed, forcing the aircraft to descend.
On 7 June 1954, Kelly Johnson received a letter from the Air Force rejecting the CL-282 proposal because it had only one engine and was too unusual and because the Air Force was already committed to the modification of the Martin B-57. By this time, the Air Force had also selected the Bell X-16; the formal contract calling for aircraft was signed in September. Despite the Air Force’s selection of the X-16, Lockheed continued to work on the CL-282 and began seeking new sources of support for the aircraft.