X-3 Stiletto

Of all the experimental research planes ever built, Douglas Aircraft Company's rakish X-3 was the one which, of them all, looked like it was built for speed. Which it undeniably was. During the heady era of the early 1950s, X-series aircraft were appearing one after another to explore numerous aspects of controlled human flight. Aircraft were pushing simple speed and altitude limititations on a daily basis, but also evaluating controllability, and studying various wing and tail configurations. The speed of sound was routinely being surpassed, but data was lacking on the aerodynamic phenomena of sustained high-speed flight. The "thermal barrier," in particular--the heating effects of high-speed air friction upon airframe components--remained little known.

The Douglas X-3 Stiletto was something of a disappointment. Because its planned J46 engines never materialized and its fuselage was too narrow for more powerful ones, the X-3 never achieved useful speeds or altitude. Almost by accident, however, its extremely long, narrow planform and short-span wings revealed the cause and the remedy for the deadly phenomenon of roll-coupling. All high speed aircraft since have benefited from research conducted with this craft.

A contract was issued for the design and construction of two transonic research aircraft was approved by the Secretary of War as early as 30 June, 1945, more than two years before supersonic speeds had been reached. Conceptual research and design work occupied the next six years.

The resulting X-3 was designed to reach Mach 2, and to sustain that speed for not less than 30 minutes. Unlike the contemporary X-1 bis and X-2 which were air-launched from carrier aircraft, it was designed to take off and land independently, under its own power. The X-3's final design appeared exotic, even by Edwards standards, yet it was driven entirely by the logic of its requirements. Even as early as the late 1940s, it was apparent that a straight wing with a low aspect ratio was going to be the most efficient design for sustained flight beyond Mach 1. Very short and highly-loaded, the X-3's wings mandated a proportionately longer fuselage to contain the landing gear and fuel cells. The very long, sharply-pointed nose section, designed to carry a variety of test instrumentation, heightened the effect. To cut down on drag, the fuselage was also made as narrow as possible, leaving barely enough frontal area for the pilot and a pair of turbojet engines. Very short exhaust ducts were used in order to get the maximum thrust, which in turn required that the tail surfaces be mounted on a slim boom located behind and above the engine nozzles. A cramped cockpit, beautifully faired in, contributed little drag and added to the fineness of the overall design. For the first time, titanium was used extensively in major airframe components to save weight and increase strength. By the time the final assembly of the X-3 was completed 30 September, 1951, it seemed that the nation had another superlative research aircraft.

No matter how carefully an airframe is designed and crafted, however, it will come to nothing if the engines cannot measure up. The X-3's narrow fuselage was literally designed around the early design specifications for a pair of Westinghouse J46-WE-1 engines, each anticipated to deliver some 4,200 lbs of thrust. Development of the J46 proved to be troublesome, however, and while the X-3 was taking shape, the new engine was not only falling behind schedule, but was growing in size and weight. Douglas was forced to install a pair of smaller J34-WE-17s, yielding only 3,000 lbs of thrust each. It was hoped by all that these engines would prove to be interim only, but the J46 fell ever farther behind schedule and no other engines could fit into the constricted engine bays of the new research plane.

Douglas test pilot Bill Bridgeman took the X-3 into the air for its first test flight on 20 October, 1952. Even before the jet made its first landing 20 minutes later, it was apparent that it was sadly underpowered and would never reach the performance levels necessary for its designed mission. Worse, the low power levels allowed a host of awkward handling problems to leap into prominence. Suddenly the tiny wings were difficult to control, and made the long fuselage extremely sensitive to pitch effects. The heavy wing loading, harmless enough in high speed flight, also meant a high takeoff speed--no less than a sizzling 260 mph. Naturally enough, there was no comparable effect at the other end of the speed range. With its laboring J34s, the X-3 was not only incapable of reaching Mach 2, it was firmly subsonic. The aircraft could only be nudged past Mach 1 in a power dive. The highest speed the Stiletto ever reached was a comparatively sedate Mach 1.21 after a thirty-degree dive. By the time the contractor and the Air Force had completed their initial evaluations, it was obvious that NACA's hot new transonic jet was no more than a lumbering and treacherous dog.

Once in a great while, however, real life mimics a Hollywood epic, and a disgraced hero is allowed to redeem himself at the last moment. So it proved with the X-3. When NACA test pilot Joe Walker, systematically following a somewhat dispirited test program, performed a routine left aileron roll and suddenly found himself wildly out of control, the way had unknowingly opened for the X-3 to redeem itself. That eventuality was the very last thing on Walker's mind during the long seconds it took him to regain control over the slipping and rolling aircraft. What had happened was a phenomenon which was coming to be known as roll coupling, when an aircraft suddenly becomes unstable along all three axes. In this instance, the X-3's nose had abruptly pitched up at the onset of a left roll; corrective action only made the problem more extreme until the airplane was riding nose high in a heavy sideslip. Ever the professional, Walker cautiously tried the maneuver once more, and once again the X-3 slammed out of control. Recovering a second time, Walker landed and the post-flight analysis began at once.

Roll coupling, also known as inertia coupling or roll divergence, had been predicted several years earlier in design studies. Air Force test pilot Chuck Yeager had experienced it himself during the preceding December, when his X-1A suddenly flipped out of control after a high-speed test run and, as the flight analysts later reported, went into divergent angular rotations about all three flight axes. Worse was to come. Nearly two years afterward, the X-2 went into roll coupling and crashed on 27 September 1956, killing Air Force Captain Milburn Apt.

Even more troubling, the deadly phenomenon was not confined to the rarified world of flight research. Even as Joe Walker made his eventful flight, roll coupling was beginning to appear in the operational Air Force. North American's revolutionary new F-100 Super Sabre, the world's first fighter plane capable of supersonic speeds in level flight, was being plagued by an unexplained yawing motion which appeared at high speeds, occasionally with disastrous results. The high-tailed, long-coupled X-3, with its unexpected tendency to replicate the same aerodynamic fault, proved invaluable in studying both its cause and its remedy. Based on the new data, North American increased the F-100's fin area and wingspan and the jet went on to be one of the nation's hardest-working and most successful fighters of the 1960s and 1970s.

As for the X-3: only the single aircraft was ever completed. After NACA completed its test program on May 23, 1956, the X-3 was returned to the Air Force, refurbished, and placed on display at the U.S. Air Force Museum at Wright-Patterson AFB in Ohio, where it remains to this day.