North American XP-86 Sabre

Last revised June 3, 2015

The North American F-86 Sabre was without question one of the greatest fighter aircraft of all time, ranking right up there with such aircraft as the Fokker D.VII, the Sopwith Camel, the Supermarine Spitfire, the Messerschmitt Bf 109, the Focke Wulf Fw 190, the Mitsubishi Zero, and the North American P-51 Mustang. It first entered service with the USAAF in 1949, and was instrumental in denying air superiority to the Communist forces during the Korean War. After the Korean War ended, many Sabres entered service with dozens of foreign air arms, becoming the primary fighter equipment of many Allied nations. It was built under license in Canada, Japan, Italy, and Australia. Its service was so long-lived that the last operational F-86 was not withdrawn from service until 1993, which must be some sort of record for a combat aircraft.

The F-86 Sabre began its life as North American Aviation's company project NA-134, which was originally intended for the US Navy. As the war in the Pacific edged toward its climax, the US Navy was making plans to acquire jet-powered carrier-based aircraft which, it was hoped, could be pressed into service in time for Operation Olympic-Coronet, the invasion of Japan planned for May 1946. The Navy had planned to acquire four jet fighters, the Vought XF6U-1 Pirate, the McDonnell XFD-1 Phantom, the McDonnell XF2D-1 Banshee, and the North American XFJ-1 Fury.

Work on the NA-134 project began in the late autumn of 1944. The NA-134 had a straight, thin-section wing set low on a rather tubby fuselage. It featured a straight-through flow of air from the nose intake to the jet exhaust that exited the aircraft under a straight tailplane. The wing was borrowed directly from the P-51D, and had a laminar-flow airfoil. It was to be powered by a single General Electric TG-180 gas turbine which was a license-built version of the de Havilland Goblin. The TG-180 was designated J35 by the military and was an 11-stage axial-flow turbojet which offered 4000 lb.s.t. at sea level. The Navy ordered three prototypes of the NA-134 under the designation XFJ-1 on January 1, 1945. On May 28, 1945, the Navy approved a contract for 100 production FJ-1s (NA-141).

At the same time that North American was beginning to design the Navy's XFJ-1, the USAAF issued a requirement for a medium-range day fighter which could also be used as an escort fighter and a dive bomber. Specifications called for a speed of at least 600 mph, since the Republic XP-84 Thunderjet already under construction promised 587 mph. On Nov 22, 1944, the company's RD-1265 design study proposed a version of the XFJ-1 for the Air Force to meet this requirement. This design was known in company records as NA-140. The USAAF was sufficiently impressed that they issued a Letter Contract on May 18, 1945 which authorized the acquisition of three NA-140 aircraft under the designation XP-86.

The Navy's XFJ-1 design had to incorporate some performance compromises in order to support low-speed carrier operations, but the land-based USAAF XP-86 version was not so constrained and had a somewhat thinner wing and a slimmer fuselage with a high fineness ratio. However, the XP-86 retained the tail surfaces of the XFJ-1.

The XP-86 incorporated several features not previously used on fighter aircraft, including a fully-pressurized cockpit and hydraulically-boosted ailerons and elevators. Armament was the standard USAAF equipment of the era--six 0.50-inch Browning M3 machine guns that fired at 1100 rounds per minute, with 267 rounds per gun. The aircraft was to use the Sperry type A-1B gun/bomb/rocket sight, working in conjunction with an AN/APG-5 ranging radar. Rocket launchers could be added underneath the wings to carry up to 8 5-inch HVARs. Self-sealing fuel tanks were to be fitted, and the pilot was to be provided with some armor plating around the cockpit area.

In the XP-86, a ten percent ratio of wing thickness to chord was used to extend the critical Mach number to 0.9. Wingspan was to be 38 feet 2 1/2 inches, length was 35 feet 6 inches, and height was 13 feet 2 1/2 inches. Four speed brakes were to be attached above and below the wings. At a gross weight of 11,500 pounds, the XP-86 was estimated to be capable of achieving a top speed of 574 mph at sea level and 582 mph at 10,000 feet, still below the USAAF requirement. Initial climb rate was to be 5850 feet per minute and service ceiling was to be 46,000 feet. Combat radius was 297 miles with 410 gallons of internal fuel, but could be increased to 750 miles by adding a 170 gallon drop tank to each wingtip. As it would turn out, these performance figures were greatly exaggerated.

A mock-up of the XP-86 was built and approved on June 20, 1945. However, early wind tunnel tests indicated that the airframe of the XP-86 would not be able to reach the desired speed of 600 mph. It is highly likely that the XP-86 project would have been cancelled at this time were it not for some unusual developments.

After the surrender of Germany in May of 1945, the USAAF (along with a lot of other air forces) was keenly interested in obtaining information about the latest German jet fighters and in learning as much as they could about secret German wartime research on jet propulsion, rocket power, and ballistic missiles. American teams were selected from industry and research institutions and sent into occupied Germany to investigate captured weapons research data, microfilm it, and ship it back to the USA.

By the summer of 1945, a lot of German data was pouring in, much of it as yet untranslated into English. As it turned out, German aeronautical engineers had wind-tunnel tested just about every aerodynamic shape that the human mind could conceive of, even some ideas even only remotely promising. A particular German paper dated 1940 reported that wind tunnel tests showed that there were some significant advantages offered by swept wings at speeds of about Mach 0.9. A straight-winged aircraft was severely affected by compressibility effects as sonic speed was approached, but the use of a swept wing delayed the effects of shock waves and permitted better control at these higher speeds. Unfortunately, German research also indicated that the use of wing sweep introduced some undesirable wing tip stall and low-speed stability effects. American researchers had also encountered similar problem with the swept-wing Curtiss XP-55 Ascender, which was so unstable that it flipped over on its back and fell out of the sky on one of its test flights.

In 1940, these German studies were of only theoretical interest, since no powerplants were available even remotely capable of reaching such speeds. However, such studies caught the attention of North American engineers trying to figure out ways to improve the performance of their XP-86.

It would do no good to build an aircraft capable of high speeds that would be so unstable that it would fall out of the sky at low speeds. The cure for the low-speed stability problem that was worked out by North American engineers was to attach automatic slats to the wing leading edges. The wing slats were entirely automatic, and opened and closed in response to aerodynamic forces. When the slats opened, the changed airflow over the upper wing surface increased the lift and produced lower stalling speeds. At high speeds, the slats automatically closed to minimize drag.

In August of 1945, project aerodynamicist L. P. Greene proposed to Raymond Rice that a swept-wing configuration for the P-86 be adopted. Wind tunnel tests carried out in September of 1945 confirmed the reduction in drag at high subsonic speeds as well as the beneficial effect of the slats on low speed stability. The limiting Mach number was raised to 0.875.

Based on these wind-tunnel studies, a new design for a swept-wing P-86 was submitted to the USAAF in the fall of 1945. The USAAF was impressed, and on November 1, 1945 it readily approved the proposal. This was one of the most important decisions ever made by the USAAF--had they not agreed to this change, the history of the next forty years would undoubtedly have been quite different.

North American's next step was to choose the aspect ratio of the swept wing. A larger aspect ratio would give better range, a narrower one better stability, and the correct choice would obviously have to be a tradeoff between the two. Further tests carried out between late October and mid November indicated that a wing aspect ratio of 6 would be satisfactory, and such an aspect ratio had been planned for in the proposal accepted on November 1. However, early in 1946 additional wind tunnel tests indicated that stability with such a narrow wing would be too great a problem, and in March the design reverted to a shorter wingform. An aspect ratio of 4.79, a sweep-back of 35 degrees, and a thickness/chord ratio of 11% at the root and 10% at the tip was finally chosen.

All of these changes lengthened the time scale of the P-86 development in comparison to that of the Navy's XFJ-1. The XFJ-1 took to the air for the first time on November 27, 1946, but the XP-86 still had almost a year more of work ahead of it before it was ready for its first flight.

On February 28, 1946, the mockup of the swept-winged XP-86 was inspected and approved. In August of 1946, the basic engineering drawings were made available to the manufacturing shop of North American, and the first metal was cut. So excited was the USAAF over the performance of the XP-86, on December 20, 1946, a Letter Contract for 33 production P-86As was approved by the USAAF. No service test aircraft were ordered. Although the 4000 lb.s.t. J35 would power the three XP-86 prototypes, production P-86As would be powered by the General Electric TG-190 (J47) turbojet offering 5000 lb.s.t.

The wing of the P-86 was to be constructed of a double-skin structure with hat sections between layers extending from the center section to the outboard edges of the outer panel fuel tanks. This structure replaced the conventional rib and stringer construction in that region. This new construction provided additional strength and allowed enough space in the wing for fuel tanks.

The wing-mounted speed brakes originally contemplated for the XP-86 were considered unsuitable for this type of wing, so they were replaced by a hydraulic door-type brake mounted on each side of the rear fuselage and one brake mounted on the bottom of the fuselage in a dorsal position. The speed brakes opened frontwards. These speed brakes had the advantage in that they could be opened at any attitude and speed, including speeds above Mach One.

The first of three prototypes, 45-59507, was rolled out of the Inglewood factory on August 8, 1947. It was powered by a Chevrolet-built J35-C-3 turbojet rated at 4000 pounds of static thrust. The aircraft was unarmed. After a few ground taxiing and braking tests, it was disassembled and trucked out to Muroc Dry Lake Army Air Base, where it was reassembled.

Test pilot George "Wheaties" Welch took the XP-86 up into the air for the first time on October 1, 1947. The flight went well until it came time to lower the landing gear and come in for a landing. Welch found to his shock that the nosewheel wouldn't come down all the way. After spending forty minutes in fruitless attempts to shake the nosewheel down into place, Welch finally brought the plane in for a nose-high landing. Fortunately, the impact of the main wheels jolted the nosewheel into place, and the aircraft rolled safely to a stop. The swept-wing XP-86 had made its first flight.

The maximum speed of the XP-86 was over 650 mph, 75 mph faster than anything else in service at the time. With the bubble canopy, the pilot's field of vision was excellent. The noise and vibration levels were considerably lower than those of other jet-powered aircraft. However, the J35 engine did not produce enough thrust, and the XP-86 could only climb at 4000 feet per minute. However, since production P-86As were to be powered by the 5000 lb.s.t. General Electric J47, no one was too worried.

On October 16, 1947, the USAF gave final approval to the Fixed Price contract for 33 P-86As, plus they authorized 190 P-86Bs. The P-86B was to be a strengthened P-86A for rough-field operations.

There is actually a possibility that the XP-86 rather than the Bell XS-1 might have been the first aircraft to achieve supersonic flight. During some of his early flight tests, George Welch reported that he had encountered some rather unusual fluctuations in his airspeed and altitude indicators during high speed dives, which might mean that he had exceeded the speed of sound. However, at that time, North American had no way of calibrating airspeed indicators into the transonic range, so they were not sure just how fast Welch had gone. On October 14, 1947, Chuck Yeager exceeded Mach 1 in the XS-1. Although the event was kept secret from the general public, North American test crews heard about this feat via the grapevine and persuaded NACA to use its equipment to track the XP-86 in a high-speed dive to see if there was a possibility that the XP-86 could also go supersonic. This test was done on October 19, five days after Yeager's flight, in which George Welch was tracked at Mach 1.02. The tests were flown again on October 21 with the same results. Since Welch had been performing the very same flight patterns in tests before October 14, there is the possibility that he, not Chuck Yeager, might have been first to exceed the speed of sound.

In any case, the fact that the XP-86 had exceeded the speed of sound was immediately classified, and remained so for several months afterward. In May of 1948, the world was informed that George Welch had exceeded Mach 1.0 in the XP-86, becoming the first "aircraft" to do so (an aircraft being defined as a vehicle that takes off and lands under its own power). The date was set as April 26, 1948. This flight did actually take place, but George Welch was not the pilot. In fact, it was a British pilot who was checking out the XP-86 who inadvertently broadcasted that he had exceeded Mach 1 over an open radio channel. However, the facts soon became common knowledge throughout the aviation community--the June 14, 1948 issue of *Aviation Week* published an article revealing that the XP-86 had gone supersonic.

The XP-86 could go supersonic in a dive with only a moderate and manageable tendency to nose-up, although below 25,000 feet there was a tendency to roll which made it unwise to stay supersonic for very long. Production Sabres were limited to Mach 0.95 below 25,000 feet for safety reasons because of this roll tendency.

XP-86 number 45-59597 was officially delivered to the USAF on November 30, 1948. By that time, its designation had been changed to XF-86.

Phase II flight tests (those flown by USAF pilots) began in early December of 1947. . An Allison-built J35-A-5 rated at 4000 lbs of static thrust was installed for USAF tests. The second and third XP-86 prototypes (45-59598 and 45-59599 joined the test program in early 1948. There were different from the first prototype as well as being different from each other in several respects. Nos 1 and 2 had different fuel gauges, a stall warning system built into the control stick, a bypass for emergency operation of the hydraulic boost system, and hydraulically-actuated leading-edge slat locks. The number 3 prototype was the only one of the three to have fully-automatic leading-edge slats that opened at 135 mph. Nos. 2 and 3 had SCR-695-B IFF beacons and carried the AN/ARN-6 radio compass set.

For the second and third prototypes, the ventral brake was eliminated, and the two rear-opening side fuselage brakes were replaced by brakes which had hinges at the front and opened out and down. These air brakes were adopted for production aircraft.

Prototype number 3 was the only one to be fitted with armament. The armament of six 0.50-inch M3 machine guns were mounted in blocks of three on either side of the cockpit. Ammunition bays were installed in the bottom of the fuselage underneath the gun bay, with as many as 300 rounds per gun. The guns were aimed by a Mk 18 gyroscopic gunsight with manual ranging.

In June of 1948, the new US Air Force redesignated all Pursuit aircraft as Fighter aircraft, changing the prefix from P to F. Thus the XP-86 became the XF-86. XP-86 number one was officially delivered to the USAF on November 30, 1948. The three prototypes remained in various test and evaluation roles well into the 1950s, and were unofficially referred to as YP-86s. All three prototypes were sold for scrap after being used in nuclear tests at Frenchman Flats in Nevada

Specifications of the XP-86:

One Chevrolet-built J35-C-3 turbojet rated at 4000 pounds of static thrust. Dimensions: wingspan 37 feet 1 7/16 inches, length 37 feet 6/1/2 inches height 14 feet 9 inches. Weights: 9730 pounds empty, 13,395 pounds gross, 16,438 pounds maximum takeoff. Performance: Maximum speed of 599 mph at sea level, 618 mph at 14,000 feet, and 575 mph at 35,000 feet. Initial climb rate was 4000 feet per minute. An altitude of 20,000 feet could be attained in 6.4 minutes, and 30,000 feet in 12.1 minutes. service ceiling was 41,300 feet. Takeoff run was 3030 feet, and the aircraft could clear a 50-foot obstacle in 4410 feet.


  1. F-86 Sabre in Action, Larry Davis, Squadron/Signal Publications, 1992.

  2. The North American Sabre, Ray Wagner, MacDonald, 1963.

  3. The American Fighter, Enzo Angelucci and Peter Bowers, Orion, 1987.

  4. The World Guide to Combat Planes, William Green, MacDonald, 1966.

  5. Flash of the Sabre, Jack Dean, Wings Vol 22, No 5, 1992.

  6. Fury-The Navy's Sabre, Robert F. Dorr, Air International vol 44 No. 1 1993.

  7. North American F-86 Sabre, Larry Davis, Wings of Fame, Vol 10., 1998.