|Bell Model D2127 / X-22|
The U.S. Navy's X-22 was developed as a V/STOL research aircraft. As such, it proved to be one of the most versatile and longest-lived of the many V/STOL aircraft that were developed. It also had the distinction of being the only aircraft to have a variable stability system incorporated into the basic design from the beginning. This feature contributed to its versatility and long service life by allowing it to perform V/STOL research that applied to a wide range of aircraft, not just to the peculiarities of the X-22 configuration itself.
The X-22's history goes back to the Tri-Service V/STOL Transport Program that addressed needs of the Army, Navy, and Air Force. One goal of this program was to develop a small number of prototype V/STOL transport aircraft that used different concepts and to perform operational evaluations of their usefulness. While the Air Force was interested in the tilt-wing concept, such as the XC-142, the Navy faced tougher problems caused by shipboard compatibility requirements. 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 in the cramped environment of 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 was no newcomer to the V/STOL business. They already were building the world's first commercially licensed helicopter. In addition, Bell built and flew the Air Test Vehicle and X-14 VTOL research aircraft, and was at the mock-up stage for the XF-109, a supersonic V/STOL fighter that never was built. While the X-22 was to be a research aircraft, it was representative of a possible small V/STOL transport. Thus, it could carry a 540kg payload and represent a commercial aircraft that could carry up to six passengers. Its length and wingspan were each a little over 11.9m, and maximum gross weight was 7530kg.
The basic configuration of the X-22 was four ducted fans that could rotate together between vertical and horizontal positions for the various flight modes. Four General Electric YT58-GE-8B/D turboshaft engines rated at 1250hp each were mounted in pairs at each wing root. They powered a common drive shaft that turned all four props. The "B" and "D" designations on the engines referred to two engine configurations that differed only by their fuel controllers. 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.
Hamilton Standard built the four propellers. The 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. The X-22 performed a vertical take-off with the ducts in the vertical position and then transitioned to wing-borne flight by rotating the ducts forward. The ducts acted as wings when rotated to the horizontal position for forward flight. In this mode, the X-22 basically had the efficiency of a canard design, rather new and radical for the 1960s (actually not, the Wright Brothers' first aircraft all were canard designs!).
All four props operated from a common drive shaft and thus always turned at the same speed. With each prop turning at high speed, very quick and precise 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.
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. The thrust forces now being off vertical caused the aircraft to move in the appropriate direction. Yaw was controlled by moving the rudder pedals, but now the computer caused 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. In today's terminology, it would be called an in-flight simulator. Every airplane is unique, having its own set of flight characteristics. 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 Calspan Corporation of Buffalo, NY (then known as the Cornell Aeronautical Laboratory), designed the variable stability system. Calspan had a long history of developing and operating in-flight simulators for the Air Force. Some of the requirements for incorporating variable stability placed difficult design requirements on Bell. For example, there had to be virtually no hysteresis in any of the flight controls. To explain hysteresis, think of the pilot moving the control stick and then letting the stick go. The stick and surface may not return to the exact starting point after the pilot takes his hand off the stick. This is the result of friction or surface irregularities in the mechanical connections and hinges. Hysteresis is a measurement of how far a moving component ends up from its original starting point. For normal aircraft, an error of a few percent is tolerable and easily compensated for by the pilot. However, in order for the variable stability system to function properly, the hysteresis has to be virtually zero so that the computer knew the precise location of the controls. Bell engineers accounted for this and other Calspan requirements in their design, assuring the proper operation of the X-22 for variable stability programs.
When operating in the variable stability mode, the pilot in the left seat would experience different flight characteristics. The pilot in the right seat served as the safety pilot. The X-22 always had the same X-22 characteristics when flown from the right seat. Thus, the safety pilot always knew exactly what flight characteristics to expect when taking over control from the left-seat pilot.
The first X-22, tail number 1520, rolled out on May 25, 1965, and was followed by fifty hours of propulsion tests in a test stand. 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. Unfortunately, the first X-22 was damaged beyond repair on its fifteenth flight on August 8, 1966. Although it had flown only 3.2 hours, it 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 high 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. This event gives an interesting example of the principle of dual redundancy, because the two systems were identical and suffered the same failure within minutes of each other. 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 second X-22, tail number 1521, flew on January 26, 1967. 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's stability augmentation system helped the pilot immensely during transition and hover. 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.
Numerous problems were discovered and fixed during these flights. These should not be taken as design faults, since the purpose of building a research aircraft is to test new concepts and see how well they work! 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:
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 as summarized below:
By the time the last research program was completed, the military's interest in V/STOL research was all but gone, and 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.
S.Markman & B.Holder "Straight Up: A History of Vertical Flight", 2000
In the early 1950s, Bell was exploring VTOL flight and aircraft in various directions. A new and unconventional method of achieving this involved the study of the ducted-fan configuration. In 1953, the US Navy sponsored a one-year study of a VTOL assault transport equipped with tilting ducted propellers. This programme led to several very interesting designs among which was the promising Bell Design D-190 sea-air-rescue (SAR) utility aircraft, which eventually reached the full-scale mock-up stage. Several configurations of the D-190 were studied, one of them being a variant capable of being carried under the fuselage of a specially modified Lockheed C-130 Hercules. Another related design was the D-2005 which appeared in 1959 and which evolved into the D-2022. Design D-2022 was intended to be used as an assault transport to meet the US Marines' VTOL Assault Transport System Requirement and the Army's ASR 3-60 study. The D-2022 was to be powered by four Lycoming T55-L-5 turboshaft engines, have a gross weight of 12850kg and accommodate up to 30 armed troops and a crew of two. But like its relatives, this remained a project. Early in 1961, a competition was held for a tilt-wing VTOL aircraft answering the Tri-Service Transport Aircraft Specification TS-152. To meet this Bell teamed up with Lockheed to propose the Design D-2064 with ducted-fan configuration. The D-2064 was not chosen and the competition was won by the Vought-Hiller-Ryan team with the tilt-wing Vought XC-142A which never went beyond the prototype stage. But, fortunately for Bell, this Tri-Service programme was subsequently extended to include the testing of a research aircraft of the ducted-fan type. Consequently, on 30 November, 1962, a 42-month contract was awarded to Bell for the building of two examples of the smaller Design D-2127, this proposal having been chosen in preference to the Douglas design in competition.
These two experimental aircraft, which were to become the last aircraft built at Niagara Falls, were designated X-22A. During the design phase, Bell researches focused on three aspects: a wide range of centre of gravity movement during hover and transition, sufficient control forces for precise control in hover and transition, and a low empty weight to attain maximum operational payload. These three requirements led to a configuration that consisted of two pairs of interconnected ducted fans. With such an arrangement, large pitch trim and control forces could be attained by varying blade pitch differentially.
From February 1963, Bell conducted numerous tests in wind tunnels with eight separate scale models both at NASA Ames and NASA Langley: a 1/20th scale model was used for spinning tests; a 1/6th scale model was used to collect basic aerodynamic data; a 1/5th scale model was used for performance and stability dynamic data; a 1/3rd scale powered duct model and a full-scale powered duct model were prepared for propeller performance investigation; a 0.0032-scale ground effect model was used to examine hover characteristics; a 0.018 scale flight model was tested by NASA for transition characteristics. The eighth model was an elevon effectiveness model. In addition several test-beds for the major dynamic components were used and a full-scale mock-up of the cockpit was built and formally approved in September 1963. Meanwhile, in Bell's Niagara Falls facility, the first of the two prototypes (which had received US Navy Bureau Number 151520 in spite of its tri-Service purpose) was nearing completion. The official roll-out took place on 25 May, 1965, and preparation began for flight trials. The Bell X-22A was powered by four 1,250shp General Electric YT-58-GE-8D shaft-drive turbines rated at 19500rpm and mounted in pairs in streamlined nacelles fitted to each side of the rear fuselage. Each pair of engines was connected to a gearbox driving a transmission shaft itself connected to the two rear ducted propellers. This shaft also drove, through a 'T' gearbox, another transmission shaft linked to the two forward ducted propellers. This arrangement was fail-safe; the failure of one engine was not critical because the three remaining operative engines could still drive the four propellers but, in this case, performance was degraded. The propellers were four three-bladed 2.1m diameter Hamilton Standards. Take-off rpm was 2590. 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. It should be noted that the X-22A had no conventional ailerons, flaps or rudder, but nevertheless the pilots' controls remained conventional. In addition, a collective pitch lever, duct rotation switches and a variable stability system (VSS) developed by the Cornell Aeronautical Laboratory were fitted. As part of the pre-flight tests, pilot evaluation of the handling characteristics of the aircraft were made on a six-degree-of-freedom analogue simulator.
Roll-out of the second aircraft (BuNo 151521) took place on 30 October, 1965, seven months after the first prototype. After half a year of static tests, a first 10-minute hover flight was made with the first aircraft on 17 March, 1966. For this flight, Bell test pilots Stanley Kakol and Paul Miller were at the controls. Unfortunately, the first X-22A had a relatively short career. On 8 August, 1966, it was damaged beyond repair when a dual hydraulic system failure caused the aircraft to make a hard landing. The crew escaped safely and the salvaged components were eventually used by Calspan Corporation to set up a VTOL flight simulator and other parts were used as spares for aircraft BuNo 151521. The test programme was thus continued with the remaining prototype which accomplished its maiden flight on 26 January, 1967, with Stanley Kakol and Richard Carlin in the cockpit, the first transition being successfully made on 3 March.
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. Even this programme did not lead to a production contract, but it contributed to a great extent in better knowledge of VTOL technologies, especially the ducted fan-configuration. Projects
Some derivatives were studied such as an armed variant known as the X-22A-1 with redesigned forward fuselage; a 1400shp T58-GE-5 powered general purpose X-22B and the X-22C, an enlarged cargo variant with rear ramp and 2650shp T55-L-7 turboshaft engines.
A.J.Pelletier "Bell Aircraft since 1935", 1992
Unlike its predecessors, the Hiller X-18 and Curtiss-Wright X-19, the Bell X-22 was designed to study the possibility of a V/STOL tactical transport aircraft and had annular wing surfaces containing ducted propellers. It was built to a US Navy contract, and was derived from an earlier project for which only a mock-up had been produced. Two X-22A prototypes were built and made hundreds of flights with conventional, short and vertical take-offs. Although they proved to be far the most efficient aircraft of the kind yet developed, they were not considered suitable for operational service, as the maximum speed was only 370km/h, as compared with the 525km/h envisaged.
G.Apostolo "The Illustrated Encyclopedia of Helicopters", 1984
Technical data for Bell X-22
Engine: 4 x General Electric YT58-GE-8D turboshaft, rated at 932kW, wingspan: 11.96m, length: 12.06m, height: 6.3m, take-off weight: 8172kg, empty weight: 4302kg, max speed: 509km/h, cruising speed: 343km/h, range: 716km