The X-22A was designed for evaluation of the tilting-duct concept in an airframe that might form the basis of a light V/STOL transport. It was also designed to provide a versatile platform capable of general research on V/STOL handling qualities using a variable stability control system.
The Tri-Service V/STOL Transport Program addressed needs of the Army, Navy, and Air Force, to develop a small number of prototype V/STOL transport aircraft that used different concepts and to perform operational evaluations of their usefulness.
The Navy studies showed that a duel tandem ducted fan configuration permitted a shorter wing span for a given weight, allowing a stubbier design that could fit on existing carrier elevators and would eliminate the need for complex wing folding mechanisms. The duct around each of the four props also would improve propeller efficiency and provide a safety benefit to personnel working on a ship’s flight deck.
The Navy awarded a $27.5 million contract for the design and development of two identical X-22s to Bell Helicopter of Niagara Falls, NY, in November 1962. Bell’s internal designation was the Model D2127. Bell had already Bell built and flew the Air Test Vehicle and X-14 VTOL research aircraft. Representative of a possible small V/STOL transport, the X-22 could carry a 540kg payload and could carry up to six passengers. Its length and wingspan were each a little over 11.9m, and maximum gross weight was 7530kg.

The rectangular fuselage accommodated at its rear a wide-chord wing fitted on its leading edges with two groups of two 1250-shp (932-kW) General Electric YT58-GE-8B/D turboshafts to drive four propellers (or fans) located inside annular ducts. These last were located at the tips of the wings and the short span foreplane and could be turned between the vertical (for vertical take-off and landing) and the horizontal (for forward flight). Change in the propeller pitch produced the thrust modulation for control, and was supplemented by movement of aileron in the slipstream of each duct. The “B” and “D” designations on the engines referred to two engine configurations that differed only by their fuel controllers. They powered a common drive shaft that turned all four props. These engines had both controllers and switched between them automatically based on whether the X-22 was operating in hover mode or cruise mode. 465 gallons of useable fuel was carried in fuselage tanks.
The power transmission system consisted of a total of ten gearboxes. It reduced the engine’s nominal 19,500 revolutions per minute speed down to the propellers’ nominal 2,600 revolutions per minute. This arrangement also allowed all four props to continue operating with any number of engines failed or intentionally shut down.
The four Hamilton Standard propellers, 2.1m diameter, 3-bladed props were fabricated of fiberglass bonded to a steel core, making them 25 percent lighter than metal props yet giving them three times the fatigue strength. A nickel sheath was mounted over the leading edge. Very high prop efficiency was achieved by placing the props inside of the ducts, so much so that the X-22 could still take off on three engines, fly on two, and make a conventional landing with only one. The two forward ducts were mounted to small pylons on the forward fuselage, and the two rear ducts were mounted to stubby, dihedralless wings on the aft fuselage. Hydraulic actuators rotated each of the four ducts, but mechanical and electrical interconnections insured that all rotated together.
Thrust control could be obtained by varying the blade angle. Four elevens, one placed at the rear of each ducted fan assembly, were the only control surfaces. Placing the elevons in the prop slipstream made them very effective, even at low airspeeds. Despite the significant looking vertical stabilizer, there was no rudder. Movement of the elevons and changes to the prop pitch achieved all flight control.
Flight control in horizontal flight was achieved using a conventional looking control stick for pitch and roll. Moving the stick caused the elevons to move, either differentially between the front and rear for pitch, or differentially on left and right sides for roll. Yaw was achieved by moving the rudder pedals, which changed the propeller blade angles to produce differential thrust. There were also throttles for each engine and a lever to control the angle of the ducts. In forward flight, the front ducts were rotated to 3 degrees up from horizontal and the rear ducts rotated to 2 degrees below horizontal. This gave an optimum incidence of 5 degrees between the two pairs.
The cockpit arrangement included two zero-zero ejection seats set side-by-side, full conventional instrument displays (plus a master tachometer for propeller revolutions and a duct angle indicator) duplicated for each pilot. Engine controls were gathered on the central console.
While hovering, the pilot used the same control stick to control pitch and bank motions. Stick inputs caused the flight control computer to command minute changes to the prop blade angles to vary the thrust, causing the X-22 to tip forward, aft, or sideways. Yaw was controlled by moving the rudder pedals, by differential movements of the elevons between the left and right sides. The pilot also could rotate the ducts to assist the fore/aft motion during hover.
During transition, with the ducts at some intermediate angle, the pilot’s control inputs produced mixed propeller pitch and elevon deflections. The ratio of mixing between the props and elevons was a function of the duct angle. The ducts rotated at 5 degrees per second.
The X-22’s flight controls also included a variable stability system. This was another flight control computer that modified the basic airplane responses so that the characteristics of other aircraft, either real or imagined, could be produced. The variable stability system followed algorithms that were developed specially for each test and programmed into the computer. They produced extra control surface motions that caused the X-22’s flight characteristics to be varied, thus producing motions that are not characteristic to the X-22 airframe, but rather to the aircraft being simulated. This gave the X-22 the capability to perform research that would be applicable to a broad range of other aircraft, not just the unique characteristics of the X-22 itself. The Cornell Aeronautical Laboratory designed the variable stability system.
The first X-22, US Navy Bureau Number 151520, rolled out on May 25, 1965, and was followed by fifty hours of propulsion tests in a test stand. The flight test program was undertaken by Calspan Corporation, in Buffalo, New York, under the auspices of the U.S. Navy. The first flight in hovering mode was not made until March 17, 1966. On this 10-minute flight, four vertical take-offs and landings and a 180 degree turn were made. It then performed a series of STOL take-off and landing tests with the ducts tilted at 30 degrees. The first X-22 was damaged beyond repair on its fifteenth flight on August 8, 1966. It had flown only 3.2 hours, but suffered a dual hydraulic failure about four miles from its base at Niagara Falls Airport. The first transition from wing-borne flight to vertical flight was made under the stress of an emergency landing. The fuselage broke in half, with the rear section coming to rest inverted. While the aircraft was lost, neither pilot was injured. Swivel fittings were used in the ducts to provide hydraulic fluid to the elevon actuators. Both failed due to excess vibration. The fix included replacing the swivel fittings with loops of flexible tubing, replacing the aluminum hydraulic lines with ones made of stainless steel, and placing additional clamps on the hydraulic lines to minimize vibration.
The first was damaged beyond economical repair and it was cannibalized to keep the second aircraft flying, although the fuselage was retained for use as a ground simulator at Calspan.
The second X-22, BuNo 151521, rolled out took place on 30 October, 1965 and flew on 26 January, 1967, with Stanley Kakol and Richard Carlin in the cockpit, for a first 10-minute hover flight. Bell test pilots Stanley Kakol and Paul Miller were at the controls. The first transition being successfully made on 3 March.
With pilots from Bell, the Army, Navy, and Air Force, the X-22 flew frequently over the next several years. At the completion of the Tri-Service testing in January of 1971, the X-22s completed 228 flights, 125 flight hours, performed over 400 vertical take-offs and landings, over 200 short take-offs and landings, and made over 250 transitions. It also hovered at 2440m of altitude and achieved forward speeds of 507km/h. These flights demonstrated that the X-22 had good basic stability and that vertical take-offs and landings could be performed easily. Operation in ground effect was a little less stable, but still positive. Hovering was easier than in most helicopters. In horizontal flight, all responses to pilot control inputs were excellent. Transitions were accomplished with minimum pilot workload. Landing position could be controlled precisely. The aircraft was still controllable without augmentation, but required a significant increase in pilot workload. This system provided rate damping in pitch, roll, and yaw only during hover and low speed flight.
With the Navy satisfied with the basic operation, they awarded a contract to Cornell in July 1970 to operate and perform flight research using the X-22, with particular emphasis on operating in the variable stability mode.

Over the next ten years, Calspan flew five test programs:

• August 1971 – February 1972: evaluation of steep STOL approach paths of 6-10 degree glide slopes, at airspeeds of 120-150km/h, with a variety of oscillatory characteristics in the pitch mode.
  June 1972-February 1973: continuation of previous effort, but with variation of roll and yaw oscillatory characteristics.
• October 1973 – April 1975: evaluation of control, display, and guidance requirements for STOL instrument approaches. Determined desirable control system requirements and information displays for pilot use to permit transition from forward flight to a decelerating steep approach and then to a hover at 30m, followed by a touchdown, all under instrument conditions.
• February 1977 – March 1978: expanded the previous experiment by evaluating the usefulness of a head-up display for STOL instrument approaches. Used precision radar distance measuring equipment to establish the aircraft’s position within 7.62mm. Evaluated a variety of display formats that presented data to the pilot, such as attitude, airspeed, altitude, horizontal location, and range to touchdown. Also, simulated the AV-8B, a specific aircraft, rather than a generic set of aircraft characteristics.
• November 1978 – May 1980: generated flying quality and flight control design requirement data for V/STOL aircraft performing shipboard landings. In performing this task, typical shipboard pitching and rocking motions had to be added to the guidance beam, then various compensation schemes were programmed into the flight computer on the X-22. Hover, and then simulated touchdowns at 30m were performed.

Numerous problems were discovered and fixed during testing. While all were fixed, they were not necessarily fixed in an optimum manner, as would be done for a production aircraft. Further development may have provided an optimal solution. But, being a research aircraft, a fix that worked for the intended mission was good enough. Some of them included the following:

• Failure of the linkage synchronizing the front and rear duct angles, resulting in the front ducts rotating to 30 degrees while the rear ducts remained vertical. Fortunately, this happened on the ground. The aluminum shaft that transmitted the proper duct angle was replaced with one made of stainless steel.
• In forward flight with the ducts near the stall angle of attack, the airflow from the lower lip separated from the duct surface as it entered the duct, causing a very loud buzzing as the turbulent flow hit the prop. Installing a number of vortex generators on the bottom inner-lip of each duct reduced this problem significantly.
• A number of fatigue cracks developed on the inside of the duct skin and ribs. Apparently this was caused by a wake of very low pressure being pulled behind each prop blade. Thus, for each revolution of the prop, the surface was hit with three pulsations of high, then low pressure. This was corrected by replacing the ribs with slimmer ones, then building up an eighteen-inch wide ring of fiberglass inside each duct, maintaining the 0.95cm clearance between the prop tip and the wall.
  The program office within the Navy that oversaw the X-22’s testing was disbanded. Calspan sought added research programs, but the most they could accomplish was to get the Naval Test Pilot School to use the aircraft for some V/STOL demonstration flights for their students during 1981 and 1982. The aircraft made its last flight in October 1984. Ownership was transferred to the Naval Aviation Museum at Pensacola, Florida, but the museum never had any desire to display unique aircraft that were not typical of Naval aviation use. The X-22 remained in storage at Calspan’s facility at Buffalo Airport in hopes that further projects would return this unique test vehicle to service, or at least be acquired by an aviation museum.
  No further work ever arose, and many efforts to transfer the X-22 to an appropriate museum in the western New York area fell through. In 1995, Calspan moved it outdoors because they needed the hangar space. To protect it from the elements, the Buffalo & Erie County Historical Society paid to cover the X-22 in a plastic wrapping. In 1998, the newly formed Niagara Aerospace Museum at Niagara Falls, NY, acquired the X-22 and placed it on display.

For two full years, the X-22A was involved in a flight-test programme with Bell and the NASA. During this period some 220 flights and 110 flying hours were logged. This phase was followed, in January 1968, by a first military preliminary evaluation during which the X-22A was examined by pilots and engineers of the three Services and accomplished fourteen flights. A second military evaluation took place at the beginning of the following April. During this period, the X-22A demonstrated. good performance such as a sustained hover at an altitude of 2400m. On 19 May, 1968, the X-22A was officially taken on charge by the US Navy which turned it over almost immediately to Calspan Corp responsible for the test programme on behalf of the Navy. The prototype had been equipped with an automatic flight control system known as LORAS (Linear Omnidirectional Resolving Airspeed System). This programme, which was broken down into several tasks, totalled 273 flights, 279.9 flying hours, 130 VTOL take-offs and 236 VTOL landings. The aircraft was flown until the autumn of 1984 when flight testing was considered terminated. The X-22A made its last flight in 1988.

Engines: 4 x General Electric YT58-GE-8B/D turboshaft, 932kW / 1250hp
Wingspan: 11.96m / 39 ft 3 in
Length: 12.06m / 39 ft 7 in
Height: 6.3m
Take-off weight: 8172kg
Empty weight: 4302kg
Max speed: 370 km/h
Cruising speed: 343km/h
Range: 716km
Ceiling: 15,000 ft.
Fastest Flight: 255 mph
Total Flights: 500+
Highest Flight: 27,800 feet (approx)