Mach 3 Legend: The North American XB-70 Valkyrie

First flown more than fourty years ago, the XB-70 Valkyrie was a technological masterpiece of its time. With the help of the latest technology it could cruise at speeds of up to Mach 3 and heights of up to 77 000 ft (23 km). What would have been the highest performance bomber ever built eventually became a museum piece when it was cancelled after politicians declared the bomber obsolete in the face of surface-to-air and intercontinental ballistic missiles.

North American XB-70 taking off on 17 August 1965. (NASA) The beautiful North American XB-70 Valkyrie #2 in flight.

The aircraft
The beginning of the end
Flight history
Disaster strikes
NASA service


The XB-70 program began in 1955 when the United States Air Force (USAF) issued a request for a high-altitude, long-range bomber with a Mach 3 (three times the speed of sound) dash capability and the ability to carry a heavy nuclear and conventional warload. The new aircraft was to replace the subsonic B-52. This requirement came about when the Air Force realized bombers could have performance similar to that of fighters and therefore be much more effective. Technological breakthroughs in the 1950s and '60s made such high performance possible.

Boeing and North American Aviation submitted their proposals to the Air Force in response to its request. Both the projected aircraft had a takeoff weight of around 1 million pounds (450 tonnes), and the ability to dash at Mach 3 (by jettisoning wing sections). But they were too large to fit into existing USAF hangars and other facilities, built mainly to accommodate the B-52.

The North American design featured an airframe built in three parts: a central pod that contained the engines, crew, bombs and some fuel; and a large pod on each wing that contained most of the wing and tail area as well as fuel. The bomber would cruise at subsonic speed toward its objective; jettison the outer wings and pods and then dash at Mach 3 to the target. It would land at a little over a quarter of its takeoff weight.

In September 18 1957, the Air Force issued a new set of requirements which called for a bomber with a cruising speed of Mach 3.0 to 3.2, an over-target altitude of 70 000 to 75 000 feet, (21 336 to 22 860 m) a range of up to 10 500 miles (16 894 km), and a gross weight not to exceed 490 000 pounds (222 264 kg). Neither company's existing proposals met them and so both were rejected.

The two companies started working on new designs and soon found that they could in fact build bombers that could meet these requirements and still operate from existing facilities. On 23 December 1957, North American's design was accepted and on 24 January 1958, a development contract was awarded. The first flight was planned to take place in December 1961 and an operational wing of 30 aircraft was to be ready by August 1964. In February of 1958, the aircraft was designated the B-70 Valkyrie*.

XB-70 Valkyrie taking off in 1967. (NASA)


The aircraft

To build a bomber that could travel 7 500 miles (12 070 km) at Mach 3, engineers at North American Aviation used a principle called 'compression lift'. This had originally been researched by the National Advisory Committee for Aeronautics (NACA - NASA today). Their research formed a large part of the foundation for Mach 3 cruising flight.

Compression lift occurs when a conical body underneath the centre of a wing pushes the air to the sides, increasing pressure under the wing section and thus increasing lift dramatically. This eliminates the need for a large wing which creates a considerable amount of drag. The XB-70 was designed to use the compression lift principle by riding its own shock wave generated by supersonic flight and using the lift thus created. The conical body underneath the centre of the wing became a giant pod, slightly triangular when viewed from the bottom, which held six engines, a bomb bay and landing gear.

The outer 20 ft (6 m) sections of the wings were hydraulically pivoted downward at supersonic speeds to trap and literally ride the high-pressure shock wave that was generated. This also reduced drag caused by the engine inlet/wing shock wave. Compression lift was 30 percent more effective because the pressure under the wing was better managed.

When the XB-70 took off, the wingtips were straight. At subsonic cruising speeds, they would be lowered to 25 degrees and above Mach 1.4, to 65 degrees. The folding wingtips allowed smaller vertical stabilizers to be used and also compensated for the delta wing's backwards shift in its centre of lift as speed increased, thus reducing trim drag.

The XB-70 emerged as a giant, impressive and sleek aircraft and one of the most beautiful to have ever flown. A huge 105 ft (32.03 m) wide delta wing, which was swept at 65.5 degrees, was attached to a long, slim forward fuselage with two canards mounted on either side of the airframe behind the cockpit. A series of elevons were mounted on the trailing edge of the wing.

The 415.6 square foot (38.68 sq m) canards were all-moving and provided variable trim. They also had a trailing edge flap which could be extended downwards. Twin vertical stabilizers were mounted at either side of the engine exhausts (which were located at the rear centre of the airframe). Including the pitot tube on the nose, the aircraft was 185.83 ft (56.69 m) long, 30.75 ft (9.38 m) high and weighed a staggering 542 000 lb (246 365 kg) at maximum takeoff weight.

Flying continuously (or even for short periods) at Mach 3 generates a heavy thermal burden on an aircraft as the movement of air rushing past the airframe heats it up. The nose and other leading-edge parts of the Valkyrie rose to 330 degrees Celsius (625 degrees Fahrenheit) at Mach 3 speeds while the rest of the airframe remained at 232 degrees Celsius (450 degrees Fahrenheit). To withstand these extreme temperatures, the XB-70 was built of stainless steel honeycomb panels and titanium. The brazed honeycomb panels consisted of stainless steel just .02 inches (.508 mm) thick.

Titanium is a costly metal and was difficult to manufacture back in the '60s so it was used sparingly and only in certain heat-critical areas. The metal made up just 9 percent of the Valkyrie's structure. Using these materials helped make the XB-70 less expensive than other comparable aircraft, such as the Lockheed SR-71. The Valkyrie was still very expensive though, with each prototype costing about $750 million.

Just designing an airframe to withstand Mach 3 heat wasn't enough to cool the airframe completely. To help with this task, the aircraft's fuel was used to absorb heat. This hot fuel was then directed into the engines and burnt up, leaving the cooler fuel behind. This 'cooler' fuel was so hot that just a tiny amount (two percent) of oxygen mixed with it would have caused the fuel to burst into flames. To prevent this, inert nitrogen gas was used to fill the space vacated by the fuel.

Six General Electric J93-GE 3 turbojet engines that each delivered around 30 000 lb (13 608 kg) of thrust in afterburner, powered the Valkyrie. Because of installation design, an engine could be removed and replaced in only a couple of hours. The engines were mounted side by side at the rear of the underwing pod. Two large rectangular inlet ducts provided two-dimensional airflow. A series of variable ramps inside the intakes, called the Air Induction Control System (AICS), would expand and contract to manipulate airflow to the engines and protect them from powerful shock of supersonic air. The system detected small changes in pressure during flight and reduced supersonic air to subsonic speeds at the engine faces.

The cockpit was designed for a crew of four, consisting of a pilot, co-pilot, bombardier and defensive systems operator. The crew operated in a shirt-sleeve environment and were provided with airliner-type comfort. They sat in cocoon-like seats with clamshell doors, which, in case of pressurization loss, would provide them with individual sealed escape capsules. The capsules contained their own oxygen bottles and emergency supplies, and basic controls to close the throttles and trim for an emergency descent, whilst monitoring the instruments through a window in the capsule. The capsules could be re-opened at a safe altitude, or rocket-ejected through jettisonable roof panels at speeds of up to Mach 3 and altitudes to above 70 000 ft (21 336 m).

When the Valkyrie reached supersonic speeds, a nose-ramp forward of the cockpit was raised so the windscreen would be a smoother surface flush with the rest of the fuselage and be aerodynamically streamlined. At subsonic speeds, the windscreen sloped downward while at high speeds it was nearly horizontal. Unfortunately, this interfered with the pilot's view.

The XB-70 had a tricycle-type landing gear which weighed more than six tons. The wheels, tires and brakes alone weighed two tons. Each main gear had four wheels and the nose gear had two. Additionally, three brake parachutes were used. If these 'chutes weren't used, the landing run would be over 11 100 ft (3 380 meters).

A single weapons bay between the engine ducts and engines could carry groups of nuclear bombs. The bay had doors which slid open automatically at the last moment before weapon release.

The  Valkyrie's unique shape can be appreciated in this 1968 photo. (NASA) The Valkyrie in flight in 1965 showing the variable incidence nose in supersonic configuration. (NASA) Cockpit of the first Valkyrie. (North American/NASA)


The beginning of the end

By 1960 Intercontinental Ballistic Missiles (ICBMs) had begun entering service. They were seen as the way of the future as they were less vulnerable than aircraft and were easier to operate. In the last year of his administration, President Dwight Eisenhower came to the conclusion that the Valkyrie program made little military sense and would only enter service in 8 to 10 years time when missiles rather than bombers would be the primary threat deterrent. An economic recession put Eisenhower's administration under pressure to cut military costs.

Surface-to-Air Missile (SAM) technology had also improved greatly, thus making aircraft seem extremely vulnerable. The British government so strongly believed manned aircraft were obsolete that in April 1957 a Defense White Paper cancelled all planned fighter aircraft in favor of missiles. When American pilot Gary Powers was shot down in his 12 mile (20 km) altitude U-2 spyplane by a SA-2 SAM over the Soviet Union on 1 May 1960, this seemed to re-enforce such beliefs.

The Valkyrie had a big airframe with a large radar cross-section that would have been easy to detect and its straight and level trajectory would have made its course simple to plot and intercept. Low altitude, high-speed flight to avoid missile detection was seen as one of the best countermeasures to SAMs. However, the Valkyrie with its large, thin wings was not suited to such flight.

The cancellation of the North American F-108 program on 24 September 1959 added to the increasing doubts over the feasibility of the Valkyrie. This aircraft, under contract since 1957, was to have been a long-range interceptor, planned as a possible escort for the XB-70. It was intended to have items such as engines, escape capsules and fuel systems built in common with the Valkyrie. The XB-70 program now had to fund and develop these systems by itself, adding at least $180 million to the overall development and construction cost.

As a result of all these events, on 29 December 1959, the Air Force dramatically cut back on the entire program. The Valkyrie program was to be reduced to a single experimental prototype for research only, with most of the weapons subsystems cancelled.

But in August 1960 the Air Force announced that this downsizing decision would be reversed. The Valkyrie had become an election issue, with candidate John F Kennedy arguing that the XB-70 was needed to even out the arms race. This debate renewed interest in the program and it was announced the aircraft would again be scheduled for production and service. Twelve B-70 prototypes were called for and $265 million was appropriated. An RS-70 (RS for reconnaissance strike) variant was conceived in 1961 and offered in 1962. It was envisaged that 60 RS-70s would be available by 1969. However, its reconnaissance capability would not have been as good as the specialized aircraft designed for this role, the Lockheed SR-71, which was publicly unveiled in 1964. The RS-70 was subsequently abandoned.

Then, in March 1961, President John F. Kennedy announced that the program was to be scaled back and used for research only. High costs of over $700 million per prototype, vulnerability and diminished need for the aircraft were cited as the main reasons. The Kennedy administration felt ICBMs were more cost-effective and were less vulnerable than aircraft. Only three aircraft, consisting of two XB-70 Valkyrie flight test prototypes and one YB-70 operational prototype, were to be built. The XB-70s were to be used for research only. Most of their combat avionics, such as the bombing-navigation system, were deleted, as well as the bombardier and navigator positions. It was agreed that lessons learned during the flight testing of the first XB-70A would be applied to the construction of the second example. The YB-70 was to have complete combat systems installed so they could be fully developed and refined. If the Air Force saw the need for the B-70 to go into production, they would not have to first test all the combat equipment. However, on 3 March 1964, budget cuts resulted in the cancellation of the YB-70.

The National Aeronautics and Space Administration (NASA) was interested in an American supersonic transport aircraft (SSTs) around this time. The administration, which operated four SST research aircraft at the Flight Research Center (FRC - now the Dryden Flight Research Center at Edwards, California), were keen to gather flight data from the XB-70 since it could cruise at supersonic speeds, was the same size as a projected SST and was constructed of similar materials an SST might use. It was agreed that NASA would collaborate with the USAF on flight tests.

Underfuselage view of the Valkyrie, clearly showing the triangular engines compartment


Flight History

Problems with the airframe seriously hampered the Valkyrie program and delayed the first flight by three years. By 1962 there were severe difficulties with the hydraulics system, the secondary power generation system, there were defects with the honeycomb panels, fuel tanks and air inlets, and the wings did not fit onto the wing stubs correctly. To overcome these problems, braces were added to the generator gear boxes, hydraulic pumps were rebuilt with stronger, more heat resistant metals, nickel was plated onto defective honeycomb panels and adapters were developed so that the wings fitted correctly. All this took much time and money and even on the aircraft's first flight problems persisted - one fuel tank was still unusable.

Assembly of the first XB-70A was eventually completed in mid-1963, but solving a fuel leak problem took another eighteen months. Finally, on 11 May 1964 the first XB-70, serial 62-0001 (tail number 20001), generally referred to as AV/1 (Air Vehicle/Ship 1), was rolled out of its hangar at North American's Palmdale, California facility. On 21 September 1964, the Valkyrie took to the air on its maiden flight. Piloted by North American pilot Al White and USAF Colonel Joe Cotton, it took off from North American Aviation's Palmdale ('Plant 42') facility and flew over the Mojave Desert to Edwards Air Force Base, the Air Force's Flight Test Center (AFFTC). The aircraft was planned to break the sound barrier on this first flight. However, the landing gear didn't retract properly because of a hydraulic system malfunction, so the aircraft was forced to remain subsonic. The XB-70 landed without mishap, but suffered minor damage from a fire caused by a surge in the hydraulic system.

On its third attempt at going supersonic, on 12 October 1964, the Valkyrie burst through the sound barrier, reaching a speed of Mach 1.1. The aircraft subsequently set a number of world records, which included sustained supersonic flight for 40 continuous minutes on 24 October 1964 and sustained supersonic flight for 60 minutes on 4 March 1965, the aircraft's seventh flight. On the Valkyrie's tenth flight, the aircraft sustained 74 minutes of supersonic flight, including 50 minutes beyond Mach 2.

While traveling at Mach 2.58 on May 7, 1965, the aircraft's horizontal intake splitter was torn away and pieces landed in the engines, which were all destroyed beyond repair. Incredibly, the Valkyrie managed to land in one piece. After this flight, the splitter was replaced with a single solid piece in place of the honeycomb unit that had failed.

After 17 flights, AV/1 finally reached Mach 3 on 14 October 1965. However, two feet (.6 m) of wing leading edge was torn away during flight. Luckily, the debris wasn't drawn into the engine inlets. Because of structural problems, it was decided that AV/1 would be limited to a maximum speed of Mach 2.5.
Wind-tunnel research by NASA showed that modifications to the airframe could drastically improve the aircraft, so North American built the second XB-70 (tail number 20207, serial 62-0207 and generally referred to as AV/2) with an added 5 degrees of dihedral on the wings. It also had an improved hydraulic system, a more soundly constructed skin and an automatic AICS. These changes resulted in greatly improved handling. AV/2 made its first flight on July 17, 1965, reaching a speed of Mach 1.4. Mach 3 was attained for the first time on 3 January 1966 and on 19 May of that year AV/2 flew at Mach 3 for 33 minutes and at Mach 2.5 for a total of 62 minutes.

Technical problems nearly destroyed both aircraft. In March 1966, while AV/1 was in flight, the main and backup hydraulic systems failed and the landing gear wouldn't deploy correctly. The aircraft managed to make a controlled landing at Edwards, although the landing roll was almost 3 miles (4.8 kilometers) and the aircraft went through a slow 110-degree turn on the ground. On 30 April 1966 AV/2's nose landing gear jammed in its door and its hydraulic and secondary electrical systems failed. Two touch-and-go landings using just the main gear failed to pop the nose gear down. After a few hours of circling around burning up fuel, it was suggested that bypassing a circuit breaker in the backup electrical system would solve the problem. Co-pilot Joe Cotton improvised a jumper with a paper clip, and the nose gear went down. A 39-cent paperclip saved a $750 million aircraft!

Around this time an agreement between NASA and the Air Force was signed to use the second XB-70 for high speed research in support of the SST program. AV/2 was chosen because of its improved aerodynamics, inlet controls, hydraulics systems and superior instrumentation. It was also fitted with extra data recording equipment, which cost more than $50 million. These instruments, fitted mainly in the former weapons bay, would provide information regarding body flex, flutter, and airframe pressures, especially during supersonic flight. The aircraft would also be used to evaluate typical SST flight profiles.
NASA research flights were to begin in mid-June, once the North American Aviation Phase I tests of the vehicle's airworthiness were completed. NASA research pilot Joe Walker was selected as the project pilot.

The Valkyrie landing with three brake parachutes streaming. (USAF/NASA)


Disaster strikes

On 8 June, AV/2 took off in the early morning for a test sortie and subsequently overflew NACA recording instruments at Mach 1.4. Al white and Maj. Carl S. Cross flew the plane. The aircraft then met up with four fighter aircraft, consisting of an F-4, an F-5, T-38, NASA F-104N Starfighter and a Learjet. All aircraft were powered by General Electric engines, including the Learjet. General Electric realized that a formation such as this would create excellent publicity; hence the Learjet carried photographers who snapped away at the XB-70 and other aircraft in formation. General Electric had a sufficient number of photos by 9.30 am and ended the shoot at about 9.35, after roughly 40 minutes of formation photo work.

The photo shoot went perfectly, just as the formation was separating, disaster struck. The F-104N, flown by Joe Walker, began to move in close to the XB-70's right wingtip, too close for safety. The Valkyrie's angled-down wingtips generated strong vortices, and these caught the F-104N and flipped it over onto the Valkyrie's back. The upside-down Starfighter smashed across the Valkyrie's tail surfaces, tearing most of them away and damaging the left wing. Joe Walker was killed instantly upon impact and his F-104N fell to the floor of the Mojave Desert in a ball of flames.

For 16 seconds the Valkyrie continued in straight and level flight, but then went into two slow rolls and broke into an unrecoverable flat spin. White managed to eject, but Cross was killed when the aircraft hit the ground in an upright and level configuration. White received serious injuries during ejection, but later returned to flight status. He never flew the remaining XB-70 again.

The XB-70 Valkyrie photographed on the ramp in 1967 (NASA)


NASA service

After the crash, the remaining XB-70 was modified, with the ejection system being improved, better brakes being fitted and test instrumentation similar to that of AV/2 being added. AV/1 did not fly again until 3 November 1966. The first NASA XB-70 flight occurred on April 25, 1967 and by the end of March 1968 another 12 research flights had been completed. A top speed of Mach 2.57 was the highest speed attained during the remainder of the XB-70 program. The XB-70 suffered buffeting during high-speed, high-altitude flights and to rectify this, a turbulence reduction system was fitted in March 1968.

During NASA flights, the XB-70 flew at differing Mach numbers, altitudes and weights over an instrumented test range at Edwards as part of the National Sonic Boom Program. Housing units were specially constructed to test the shockwave generated by sonic booms. The tests showed that a large aircraft, such as the XB-70 or the projected SST, could generate a shockwave powerful enough to cause structural damage.

Because the surviving Valkyrie couldn't reach Mach 3 safely, the Air Force soon lost interest in the testing and pulled out, leaving the program to NASA. But NASA soon reached an agreement with the Air Force to fly research missions with a pair of YF-12As and an SR-71. The two XB-70s had logged just 1 hour and 48 minutes of Mach 3 flight, while an SR-71 could do this in a single flight. Consequently, NASA decided that XB-70 testing would come to an end.

The 83rd and last XB-70 flight occurred on 4 February 1969, when the Valkyrie made a subsonic structural dynamics test and ferry flight. The aircraft took off from Edwards and flew to Wright-Patterson Air Force Base in Ohio, where it was put on display at the Air Force Museum. It resides there today.

The XB-70 dramatically advanced high-speed flight and was a fantastic aeronautical achievement, especially considering the period in which it was developed. Its effects can still be seen today in the MiG-25 and in the absence of a US SST, which was cancelled in 1971. The research data the XB-70 gathered helped kill the American version of the Concorde by showing how destructive and expensive sustained supersonic flight was.

In total, the first XB-70 made 83 flights equaling 160 hours and 16 minutes, while the second XB-70 logged 46 flights in its brief life, totaling 92 hours and 22 minutes. This fantastic aircraft will always be remembered in the history of aviation and its legacy can still be felt in the aviation industry today.

Two famous North American aircraft as seen in 1967: the XB-70 Valkyrie and X-15 (NASA) This photo, taken in 1968, clearly shows the XB-70 Valkyrie's rear end. (NASA)



The B-70 program gave rise to the Russian Mikoyan-Gurevich MiG-25 'Foxbat'. The Soviets were deeply concerned about the threat the Valkyrie posed, so Soviet authorities urgently ordered the development of an interceptor that could match the Valkyrie's performance and expected in-service date of 1964. When the B-70 was cancelled, the Soviets continued work on their interceptor, which emerged as the MiG-25. The aircraft first flew in 1964 and entered Soviet service in 1971. It set many world records in 1965-67 and a variant of the aircraft, known as the E.266M, still holds the world absolute height record for airplanes at 37 650 m (123 524 ft). The MiG-25 is also still the world's fastest fighter in service.

A greatly improved all-weather, all-altitude interceptor variant of the MiG-25, the MiG-31 'Foxhound' was developed and flown in 1975 and entered service in 1982. This formidable aircraft forms a large strategic part of the Russian air force.

*A Valkyrie ('Chooser of the Slain') is a mythological horse-mounted Norse maiden, sent by the god Odin to battlefields to choose the slain that were worthy of a place in Valhalla. Once there, the dead would live blissfully until needed by Odin. Some Valkyries had the power to cause the death of the warriors they did not favor; others guarded the lives and ships of those dear to them.

XB-70 Valkyrie #1 taking off with a Convair TB-58A Hustler chaseplane (North American/NASA)

The text of this article was first published in issue 17 of Flightline magazine. Please visit their website at:


Back to articles

AddThis Social Bookmark Button

© Copyright 2005-2011 Guy Martin /


Top Aviation Sites