The final competition

In early July, and much to Johnson’s surprise, Lockheed was instructed to redesign the A‑11 incorporating RCS reduction techniques, even to the detriment of cruise altitude. In contrast, Convair was instructed to undertake a colossal amount of work completely redesigning FISH as a single‑stage platform, not reliant upon a mother ship and utilizing two J58 turbojets instead of ramjets. They called their new design KINGFISH.

The Convair team now pulled out all the stops, working through three or four design variations. Remarkably, in just over two weeks they had not only decided upon the basic design, but had also considerably modified the FISH radar model being used for RCS testing at Indian Springs AFB in Nevada, to the larger KINGFISH configuration. The KINGFISH model then undertook a series of 70 MHz radar tests from August 15 – the results of which would turn out to be similar to those achieved by the A‑12.

^{The addition of fuselage chines to the “Blackbird” family’s circular forebody transformed its cross‑section into a “stealthy” two‑dimensional “flying saucer,” a discovery made independently by both Frank Rodgers and Ed Lovick. (Paul F. Crickmore)}

^{This rare color shot shows engineers within Skunk Works Building 309/310 at Plant B‑6, getting ready to position the inboard wing leading‑edge serrations onto an A‑12. (Lockheed Martin)}

“LD” MacDonald and Ed Lovick advised Dick Fuller how best to modify the A‑11 design into the reduced RCS A‑12. This included using chines, suggested by Frank Rodgers during the A‑4 series of designs. Once these were applied a cross‑section of the fuselage forebody resembled a two‑dimensional flying saucer! The wings were blended into both the fuselage and engine nacelles, whilst the twin tails were made of composite material and also canted inboard at 15 degrees. Other more subtle RCS reduction innovations included incorporating Ed Lovick’s serrations that were applied to the wing leading and trailing edges as well as to the chines, which were then filled with dielectric material, as explained earlier. The A‑11’s rectangular inlets were replaced with round, axisymmetric units; and the two translating spikes, used to regulate mass airflow into each inlet, not only helped to shield the front of the compressor face from incident radar energy, but also were covered in RAM.

On July 14, Bissell and the Land Committee together with James Killian met and discussed the pros and cons of FISH, KINGFISH, the A‑11, and A‑12, in addition to reports speculating about the possible capabilities of the new Soviet Bar Lock radar. Like the June 9 meeting, a final decision regarding which design to recommend wasn’t made; however, those present did begin to formulate a briefing to be taken by Dulles and others to President Eisenhower six days later.

Final proposals from both Convair and Lockheed were submitted on August 20 and nine days later Lockheed received official notification that their design had been selected by the Committee, and that consequently the company would receive $4.5 million by way of an advanced feasibility contract, covering the period September 1, 1959 to January 1, 1960. The name of the new program was to be Oxcart and Project Gusto was to be terminated. On August 31, 1959, Lockheed began building a full‑scale RCS model with which to validate their RCS claims. However it was decided that the site at Indian Springs AFB couldn’t provide the required level of security to test a full‑size A‑12 model, so Johnson instructed that these tests would instead be conducted at the radar target scatter (or Rat Scat) range, located at Area 51.

Area 51, or “Watertown Strip” as it had been known to the U‑2 community, offered exceptional remoteness. Located about 100 miles northwest of Las Vegas, the site would also be home for Oxcart. However, a vast amount of work was necessary both at the airfield and the Rat Scat facility to bring them up to the standards required for this new advanced program.

In December 1959, John Parangosky was appointed as the CIA’s program manager of Oxcart. Following validation of the A‑12’s RCS tests, a contract worth $96.6 million was signed between the Agency and Lockheed on February 11, 1960, for the manufacture and testing of 12 A‑12s, including one two‑seat conversion trainer. The CIA had included a clause within the contract providing for periodic re‑evaluation of costs – a clause that had to be invoked a number of times in the coming five years due to spiraling costs attributed to technical difficulties.

Mayday

Despite misgivings, Eisenhower continued to sanction a limited number of U‑2 Soviet overflights on the basis of the continuing need to collect such intelligence and the ongoing failure of the Corona satellite reconnaissance program to plug the gap. But on May 1, 1960, Soviet radar and SAM technology finally caught up with Project Chalice, when the U‑2 being flown by Gary Powers was shot down near Sverdlovsk, deep within the Soviet Union. Two days later, assuming that Powers had been killed, approval was given by Eisenhower to put out a prearranged cover story in which NASA claimed that one of their U‑2 research aircraft “apparently went down” after the pilot “reported that he was experiencing oxygen difficulties.” This played perfectly into Premier Khrushchev’s hands and on Saturday, May 7, having returned to the Supreme Soviet, he revealed to the world the truth: “We have the remnants of the plane – and we also have the pilot, who is quite alive and kicking!” The wreckage, maps and even film were recovered proving, beyond dispute, the precise reason behind the flight. A subsequent mid‑May summit meeting in Paris was a disaster, during which Khrushchev walked out after Eisenhower refused to apologise publicly for sanctioning the U‑2 overflights (despite having assured the Soviet Premier that there would be no more such flights during the remaining eight months of his administration).

A serious question thus arose regarding the future status of Oxcart. On the one hand, it vindicated the decision to develop a U‑2 replacement; but it also opened up a debate as to whether the United States should continue violating Soviet airspace with its manned strategic reconnaissance program.

In late May, Eisenhower told his military aide Brig Gen Andrew Goodpaster that he believed the project should go forward on low priority, for Air Force use in time of war. In a subsequent memo for the record, Goodpaster noted that Eisenhower “Did not think the project should be pushed at top priority. In fact, they might come to the conclusion that it would be best to get out of it if we could.” But with Corona still struggling, there was no viable option to Oxcart available in the short‑to‑mid‑term, so the program survived.

^{A 1/8th scale A‑12 model undertakes RCS testing. The extensions aft of the engine nacelles were added to simulate the aircraft’s exhaust plume. (Lockheed Martin)}

Whilst production work on the A‑12 got underway at the Skunk Works plant in Burbank, California, so too did construction on a 12,000ft runway extension, workshops, hangars, and a raft of other support facilities out at Area 51. The operational unit for Oxcart was designated the 1129th Special Activities Squadron (SAS), and the selection process for its pilots was evolved in 1961 by Brig Gen Flickinger, Col Houser Wilson at the Pentagon, and the Director of the Agency’s Office of Special Activities (OSA), Brig Gen Jack Ledford (later succeeded by Brig Gen Paul Baclais), who together defined the required physical and experience criteria. Pilots were required to be qualified in the latest high‑performance aircraft, highly proficient in air refueling, emotionally stable, married, and especially well motivated. In addition, candidates had to be between 25 and 40 years of age, weigh less than 175lb, and be under 6ft tall. There followed a week of psychological assessments in various hotel rooms in Washington, DC and then an extensive medical examination at the Lovelace Clinic at Albuquerque, New Mexico. Once the process was completed, the 11 successful candidates began taking trips to the David Clark Company in Worcester, Massachusetts, to be outfitted with their personal S‑901 full pressure suits.

Following further production delays in Burbank, the May 1961 planned delivery of aircraft number one to the Area 51 test site proved overly optimistic and slipped to late summer. Problems in procuring and working with titanium, combined with difficulties experienced by Pratt & Whitney, prompted Johnson to write to CIA officials explaining that schedules could slip from between three to four months. The response from Bissell was predictably curt, his memo finishing with, “I trust that this is the last of such disappointments short of a severe earthquake in Burbank.”

Despite the production problems, Oxcart preparations continued in various parts of the world, with high‑capacity fuel tank farms constructed at Air Force bases in California, Alaska, Greenland, Okinawa, and Turkey, to provide worldwide storage of the A‑12’s special fuel in readiness for the jet’s operational sorties. Additional fuel storage facilities were also prepared at bases in Arkansas and Florida to support transcontinental training flights. The 903rd Air Refueling Squadron based at Beale AFB, California, was designated to support Oxcart air refueling operations and was equipped with special KC‑135Q tankers, which possessed separate “clean” tankage and plumbing to isolate the A‑12’s fuel from the tanker’s own JP‑4. These tankers were also equipped with special ARC‑50 distance‑ranging radios to facilitate discrete, precision, long‑distance, highspeed join‑ups.

^{The first A‑12, Article 121 (60‑6924), being removed from its crate at Area 51, having arrived by road. (Roadrunners Internationale)}

During January and February 1962, final checks were at last successfully conducted on aircraft number one (Article Number 121; USAF serial number 60‑6924). The airframe was then dismantled, loaded onto a custom‑built trailer and on February 26, 1962, transported to Area 51. Following reassembly at its new desert home and whilst being prepared for the first flight, yet another delay occurred when the aircraft began leaking fuel profusely from its “wet” tanks. An investigation revealed that the specially designed sealant had failed to adhere to surfaces between the fuel tanks and the metal skin of the aircraft. The defective sealant was removed and the tanks relined – but tank sealant issues on all variants of the aircraft remained an issue throughout its life.

Having successfully completed a series of low‑ and medium‑speed taxi tests, Lockheed Chief Project pilot, Lou Schalk, was scheduled to take Article 121 on a high‑speed test on April 24, 1962. The test‑card called for the evaluation to culminate in a momentary lift‑off and landing roll‑out onto the dry salty lakebed. For this first “hop” the A‑12’s stability augmentation system (SAS) was left uncoupled – it was planned to test this properly in flight. However, as Lou Schalk recalls, immediately after lift‑off everything went badly wrong:

I really didn’t think I was going to be able to put the aircraft back on the ground safely because of lateral, directional and longitudinal oscillations. The aircraft was very difficult to handle but I finally caught up with everything that was happening, got control back enough to set it back down, and chop engine power. Touchdown was on the lakebed instead of the runway, creating a tremendous cloud of dust into which I disappeared entirely. The tower controllers were calling me to find out what was happening and I was answering, but the UHF antenna was located on the underside of the aircraft [for best transmission in flight] and no one could hear me. Finally, when I slowed down and started my turn on the lake‑bed and re‑emerged from the dust cloud, everyone breathed a sigh of relief.

^{Photographed during a test flight on December 22, 1962, Article 123 (60‑6926) was the third A‑12 to be built. The innovative “blended body” design merging both the fuselage and engine nacelles into the wings to reduce RCS is clearly depicted. (Lockheed Martin)}

^{A‑12 chief test pilot Lou Schalk (center, in flight suit) is congratulated by “Agency” and Lockheed officials following successful completion of the A‑12’s first “official” flight on April 30, 1962. Note the F‑104 Starfighter chase plane in the background. (Lockheed Martin)}

Two days later, Schalk successfully completed a trouble‑free first real test flight lasting 35 minutes – for which the SAS dampers remained switched on!

Over the next few months Article 121 was joined by more of its stablemates. Article 122 (60‑6925) arrived on June 26, but was destined to spend three months conducting ground radar tests before taking to the air. Aircraft number three (Article 123; 60‑6926) arrived in August and flew in October. In November the two‑seat pilot trainer (Article 124; 60‑6927) was delivered, which was planned to help smooth transition training. The aircraft was to have been powered by J58 engines, but as engine production problems persisted, it was decided to equip the two‑seater with J75 engines and let the checked‑out pilots go on to high‑Mach flight on their own. Therefore the AT‑12 trainer aircraft, nicknamed the “Titanium Goose,” undertook its maiden flight in January 1963 fitted with the less‑powerful engines. Aircraft number five (Article 125; 60‑6928) was delivered to the site on December 19, 1962.

The Cuban Missile Crisis once again demonstrated the U‑2’s vulnerability to SA‑2 attack in spectacular fashion when Air Force Maj Rudy Anderson was shot down and tragically killed during a reconnaissance mission over the Caribbean island on October 27, 1962. But still there was no sign of Oxcart entering service.

On January 15, 1963, the first flight of an A‑12 powered by two J58s finally occurred and by the end of the month ten engines were available and the test program began to gain momentum. The biggest hurdle facing both test pilots and engineers was perfecting the air induction system, designed to vastly augment engine thrust. To achieve Johnson’s design goal of sustained Mach 3.2 flight, the air inlet spike’s aft‑movement, together with the precise position of various bypass doors, had to initially be manually programed extremely accurately by the test team to ensure that the terminal shock wave was positioned in exactly the correct position in order to stabilize airflow in the inlet duct for future flights. When these parameters were finally achieved, the A‑12’s thirst for fuel – particularly during the transonic phase of acceleration – was notably reduced. In all, it took 66 flights to extend Oxcart’s speed envelope out from Mach 2.0 to Mach 3.2.

^{1. A‑12 Serial 60‑6933, this aircraft was the tenth A‑12 built.}

^{2. AT‑12, The two‑seat A‑12 pilot trainer.}

^{3. YF‑12A, Serial 60‑6935, as it appeared when operating with NASA, configured with instruments for a series of “cold‑wall” experiments.}

^{4. SR‑71A The A‑12’s replacement as the United States’ Mach 3 reconnaisance aircraft. The two‑seat SR‑71 had a radar antenna in the nose, equipment bays in the underside of its chine, a longer “boat‑tail,” and a circular “window” in the upper fuselage for its astroinertial navigation system.}

^{Article 123 was the first of five A‑12s lost in accidents. The aircraft crashed on May 24, 1963, whilst being flown by Ken Collins during a subsonic flight test – the airframe had accumulated just 135.3 flight hours. (Lockheed Martin via Tony Landis)}

But success came at a price. On May 24, 1963, Ken Collins was forced to eject from Article 123 during a subsonic engine test sortie, following an aircraft pitch‑up and subsequent loss of control. The cause was found to be ice encrustation in the pitot static system, leading to the display of erroneous flight data. Article 133 (60‑6939) was lost on July 9, 1964, just as Lockheed Test Pilot Bill Park was turning onto final approach into Area 51 having just completed a tri‑sonic test flight. The aircraft experienced a complete flight control lock‑up and Park was forced to eject at about 200kts as the aircraft continued to increase bank‑angle at just 200ft above the desert floor. The cause was loss of hydraulic fluid to the flight control system.

^{The four hangars and workshops in the foreground were just part of the major redevelopment necessary to support Oxcart flight operations up at Area 51. To enable the A‑12 to be tested both at speed and in a secure environment when airborne, the Yuletide Special Rules Area was established above the base; it extended up from 24,000ft to 60,000ft and was approximately the size of England! (Roadrunners Internationale)}

On December 28, 1964, Agency pilot Mele Vojvodich taxied out in Article 126 (60‑6929) for a Functional Check Flight (FCF), after the aircraft had undergone deep maintenance. With both burners lit and immediately upon rotating the aircraft, it yawed viciously to one side; corrective rudder application caused 126 to pitch‑up. It became apparent that all pilot control inputs were having a reverse effect to those intended – in the midst of these uncontrollable divergent effects, Vojvodich was forced to eject from the aircraft not even 100ft above the ground. With just one swing on the open parachute, Vojvodich narrowly missed the flaming pyre of 126, which signified the end of yet another aircraft. The sortie had lasted just six seconds – the shortest of any “Blackbird” flight. A subsequent inquiry established that the SAS had been wired back into the aircraft incorrectly.

On November 20, 1965, the final stage of the validation process was completed when a maximum‑endurance flight of six hours and 20 minutes was achieved, during which time an Oxcart demonstrated sustained speeds above Mach 3.2 at altitudes approaching 90,000ft. But the question remained – where to deploy the bird?