CASE
5
Dynamically Scaled Free-Flight Models
The earliest flying machines were small models and concept demonstrators, and they dramatically influenced the invention of flight. Since the invention of the airplane, free-flight atmospheric model testing—and tests of “flying” models in wind tunnel and ground research facilities—has been a means of undertaking flight research critical to ensuring that designs meet mission objectives. Much of this testing has helped identify problems and solutions while reducing risk.
Case-5 Cover Image: Hovering flight test of a free-flight model of the Hawker P.1127 V/STOL fighter underway in the return passage of the Full-Scale Tunnel. Flying-model demonstrations of the ease of transition to and from forward flight were key in obtaining the British government’s support. NASA.
On a hot, muggy day in summer 1959, Joe Walker, the crusty old head of the wind tunnel technicians at the legendary NASA Langley Full-Scale Tunnel, couldn’t believe what he saw in the test section of his beloved wind tunnel. Just a few decades earlier, Walker had led his technician staff during wind tunnel test operations of some of the most famous U.S. aircraft of World War II in its gigantic 30- by 60-foot test section. With names like Buffalo, Airacobra, Warhawk, Lightning, Mustang, Wildcat, Hellcat, Avenger, Thunderbolt, Helldiver, and Corsair, the test subjects were big, powerful fighters that carried the day for the United States and its allies during the war. Early versions of these aircraft had been flown to Langley Field and installed in the tunnel for exhaustive studies of how to improve their aerodynamic performance, engine cooling, and stability and control characteristics.
On this day, however, Walker was witnessing a type of test that would markedly change the research agenda at the Full-Scale Tunnel for many years to come. With the creation of the new National Aeronautics and Space Administration (NASA) in 1958 and its focus on human space flight, massive transfers of the old tunnel’s National Advisory Committee for Aeronautics (NACA) personnel to new space flight priorities such as Project Mercury at other facilities had resulted in significant reductions in the tunnel’s staff, test schedule, and workload. The situation had not, however, gone unnoticed by a group of brilliant engineers that had pioneered the use of remotely controlled free-flying model airplanes for predictions of the flying behavior of full-scale aircraft using a unique testing technique that had been developed and applied in a much smaller tunnel known as the Langley 12-Foot Free Flight Tunnel. The engineers’ activities would benefit tremendously by use of the gigantic test section of the Full-Scale Tunnel, which would provide a tremendous increase in flying space and allow for a significant increase in the size of models used in their experiments. In view of the operational changes occurring at the tunnel, they began a strong advocacy to move their free-flight studies to the larger facility. The decision to transfer the free-flight model testing to the Full-Scale Tunnel was made in 1959 by Langley’s management, and the model flight-testing was underway.
Joe Walker was observing a critical NASA free-flight model test that had been requested under joint sponsorship between NASA, industry, and the Department of Defense (DOD) to determine the flying characteristics of a 7-foot-long model of the North American X-15 research aircraft. As Walker watched the model maneuvering across the test section, he lamented the radical change of test subjects in the tunnel with several profanities and a proclamation that the testing had “gone from big-iron hardware to a bunch of damn butterflies.”[1] What Walker didn’t appreciate was that the revolutionary efforts of the NACA and NASA to develop tools, facilities, and testing techniques based on the use of subscale flying models were rapidly maturing and being sought by military and civil aircraft designers—not only in the Full-Scale Tunnel, but in several other unique NASA testing facilities.
For over 80 years, thousands of flight tests of “butterflies” in NACA and NASA wind tunnel facilities and outdoor test ranges have contributed valuable predictions, data, and risk reduction for the Nation’s high-priority aircraft programs, space flight vehicles, and instrumented planetary probes. Free-flight models have been used in a myriad of studies as far ranging as aerodynamic drag reduction, loads caused by atmospheric gusts and landing impacts, ditching, aeroelasticity and flutter, and dynamic stability and control. The models used in the studies have been flown at conditions ranging from hovering flight to hypersonic speeds. Even a brief description of the wide variety of free-flight model applications is far beyond the intent of this essay; therefore, the following discussion is limited to activities in flight dynamics, which includes dynamic stability and control, flight at high angles of attack, spin entry, and spinning.
Birthing the Testing Techniques
The development and use of free-flying model techniques within the NACA originated in the 1920s at the Langley Memorial Aeronautical Laboratory at Hampton, VA. The early efforts had been stimulated by concerns over a critical lack of understanding and design criteria for methods to improve aircraft spin behavior.[2] Although early aviation pioneers had been frequently using flying models to demonstrate concepts for flying machines, many of the applications had not adhered to the proper scaling procedures required for realistic simulation of full-scale aircraft motions. The NACA researchers were very aware that certain model features other than geometrical shape required application of scaling factors to ensure that the flight motions of the model would replicate those of the aircraft during flight. In particular, the requirements to scale the mass and the distribution of mass within the model were very specific.[3] The fundamental theories and derivation of scaling factors for free-flight models are based on the science known as dimensional analysis. Briefly, dynamic free-flight models are constructed so that the linear and angular motions and rates of the model can be readily scaled to full-scale values. For example, a dynamically scaled 1/9-scale model will have a wingspan 1/9 that of the airplane and it will have a weight of 1/729 that of the airplane. Of more importance is the fact that the scaled model will exhibit angular velocities that are three times faster than those of the airplane, creating a potential challenge for a remotely located human pilot to control its rapid motions.
Initial NACA testing of dynamically scaled models consisted of spin tests of biplane models that were hand-launched by a researcher or catapulted from a platform about 100 feet above the ground in an airship hangar at Langley Field.[4] As the unpowered model spun toward the ground, its path was tracked and followed by a pair of researchers holding a retrieval net similar to those used in fire rescues. To an observer, the testing technique contained all the elements of an old silent movie, including the dash for the falling object. The information provided by this free-spin test technique was valuable and provided confidence (or lack thereof) in the ability of the model to predict full-scale behavior, but the briefness of the test and the inevitable delays caused by damage to the model left much to be desired.
The free-flight model testing at Langley was accompanied by other forms of analysis, including a 5-foot vertical wind tunnel in which the aerodynamic characteristics of the models could be measured during simulated spinning motions while attached to a motor-driven spinning apparatus. The aerodynamic data gathered in the Langley 5-Foot Vertical Tunnel were used for analyses of spin modes, the effects of various airplane components in spins, and the impact of configuration changes. The airstream in the tunnel was directed downward, therefore free-spinning tests could not be conducted.[5]
Meanwhile, in England, the Royal Aircraft Establishment (RAE) was aware of the NACA’s airship hangar free-spinning technique and had been inspired to explore the use of similar catapulted model spin tests in a large building. The RAE experience led to the same unsatisfactory conclusions and redirected its interest to experiments with a novel 2-foot-diameter vertical free-spinning tunnel. The positive results of tests of very small models (wingspans of a few inches) in the apparatus led the British to construct a 12-foot vertical spin tunnel that became operational in 1932.[6] Tests in the facility were conducted with the model launched into a vertically rising airstream, with the model’s weight being supported by its aerodynamic drag in the rising airstream. The model’s vertical position in the test section could be reasonably maintained within the view of an observer by precise and rapid control of the tunnel speed, and the resulting test time could be much longer than that obtained with catapulted models. The advantages of this technique were very apparent to the international research community, and the facility features of the RAE tunnel have influenced the design of all other vertical spin tunnels to this day.
This cross-sectional view of the Langley 20-Foot Vertical Spin Tunnel shows the closed-return tunnel configuration, the location of the drive fan at the top of the facility, and the locations of safety nets above and below the test section to restrain and retrieve models. NASA.
When the NACA learned of the new British tunnel, Charles H. Zimmerman of the Langley staff led the design of a similar tunnel known as the Langley 15-Foot Free-Spinning Wind Tunnel, which became operational in 1935.[7] The use of clockwork delayed-action mechanisms to move the control surfaces of the model during the spin enabled the researchers to evaluate the effectiveness of various combinations of spin recovery techniques. The tunnel was immediately used to accumulate design data for satisfactory spin characteristics, and its workload increased dramatically.
Langley replaced its 15-Foot Free-Spinning Wind Tunnel in 1941 with a 20-foot spin tunnel that produced higher test speeds to support scaled models of the heavier aircraft emerging at the time. Control inputs for spin recovery were actuated at the command of a researcher rather than the preset clockwork mechanisms of the previous tunnel. Copper coils placed around the periphery of the tunnel set up a magnetic field in the tunnel when energized, and the magnetic field actuated a magnetic device in the model to operate the model’s aerodynamic control surfaces.[8]
The Langley 20-Foot Vertical Spin Tunnel has since continued to serve the Nation as the most active facility for spinning experiments and other studies requiring a vertical airstream. Data acquisition is based on a model space positioning system that uses retro-reflective targets attached on the model for determining model position, and results include spin rate, model attitudes, and control positions.[9] The Spin Tunnel has supported the development of nearly all U.S. military fighter and attack aircraft, trainers, and bombers during its 68-year history, with nearly 600 projects conducted for different aerospace configurations to date.
Wind Tunnel Free-Flight Techniques
Charles Zimmerman energetically continued his interest in free-flight models after the successful introduction of his 15-foot free-spinning tunnel. His next ambition was to provide a capability of investigating the dynamic stability and control of aircraft in conventional flight. His approach to this goal was to simulate the unpowered gliding flight of a model airplane in still air but to accomplish this goal in a wind tunnel with the model within view of the tunnel operators. Without power, the model would be in equilibrium in descending flight, so the tunnel airstream had to be at an inclined angle relative to the horizon. Zimmerman designed a 5-foot-diameter wind tunnel that was mounted in a yoke-like support structure such that the tunnel could be pivoted and its airstream could simulate various descent angles. Known as the Langley 5-Foot Free-Flight Tunnel, this exploratory apparatus was operated by two researchers—a tunnel operator, who controlled the airspeed and tilt angle of the tunnel, and a pilot, who controlled the model and assessed its behavior via a control box with a fine wire connection to the model’s control actuators.[10]
The Langley 5-Foot Free-Flight Tunnel was mounted in a yoke assembly that permitted the test section to be tilted down for simulation of gliding flight. Its inventor, Charles Zimmerman, is on the left controlling the model, while the tunnel operator is behind the test section. NASA.
Very positive results obtained in this proof-of-concept apparatus led to the design and construction of a larger 12-Foot Free-Flight Tunnel in 1939. Housed in a 60-foot-diameter sphere that permitted the tunnel to tilt upward and downward, the Langley 12-Foot Free-Flight Tunnel was designed for free-flight testing of powered as well as unpowered models. A three-person crew was used in the testing, including a tunnel airspeed controller, a tunnel tilt-angle operator, and an evaluation pilot.
The tunnel operated as the premier NACA low-speed free-flight facility for over 20 years, supporting advances in fundamental dynamic stability and control theory as well as specific airplane development programs. After the 1959 decision to transfer the free-flight activities to the Full-Scale Tunnel, the tunnel pivot was fixed in a horizontal position, and the facility has continued to operate as a NASA low-cost laboratory-type tunnel for exploratory testing of advanced concepts.
Test setup for free-flight studies at Langley. The pitch pilot is in a balcony at the side of the test section. The pilot who controls the rolling and yawing motions is at the rear of the tunnel. NASA.
Relocation of the free-flight testing to the Full-Scale Tunnel made that tunnel the focal point of free-flight applications at Langley for the next 50 years.[11] The move required updates to the test technique and the free-flight models. The test crew increased to four or more individuals responsible for piloting duties, thrust control, tunnel operations, and model retrieval and was located at two sites within the wind tunnel building. One group of researchers was in a balcony at one side of the open-throat test section, while a pilot who controlled the rolling and yawing motions of the model was in an enclosure at the rear of the test section within the structure of the tunnel exit-flow collector. Models of jet aircraft were typically powered by compressed air, and the level of thrust was controlled by a thrust pilot in the balcony. Next to the thrust pilot was a pitch pilot who controlled the longitudinal motions of the model and conducted assessments of dynamic longitudinal stability and control during flight tests. Other key members of the test crew in the balcony included the test conductor and the tunnel airspeed operator.
A light, flexible cable attached to the model supplied the model with the compressed air, electric power for control actuators, and transmission of signals for the controls and sensors carried within the model. A portion of the cable was made up of steel cable that passed through a pulley above the test section and was used to retrieve the model when the test was terminated or when an uncontrollable motion occurred. The flight cable was kept slack during the flight tests by a safety-cable operator in the balcony who accomplished the job with a high-speed winch.[12]
Free-flight models in the Full-Scale Tunnel typically had model wingspans of about 6 feet and weighed about 100 pounds. Propulsion was provided by compressed air ejectors, miniature turbofans, and high thrust/weight propeller motors. The materials used to fabricate models changed from the simple balsa free-flight construction used in the 12-Foot Free-Flight Tunnel to high-strength, lightweight composite materials. The control systems used by the free-flight models simulated the complex feedback and stabilization logic used in flight control systems for contemporary aircraft. The control signals from the pilot stations were transmitted to a digital computer in the balcony, and a special software program computed the control surface deflections required in response to pilot inputs, sensor feedbacks, and other control system inputs. Typical sensor packages included control-position indicators, linear accelerometers, and angular-rate gyros. Many models used nose-boom–mounted vanes for feedback of angle of attack and angle of sideslip, similar to systems used on full-scale aircraft. Data obtained from the flights included optical and digital recordings of model motions and pilot comments as well as analysis of the model’s response characteristics.
The NACA and NASA also developed wind tunnel free-flight testing techniques to determine high-speed aerodynamic characteristics, dynamic stability of aircraft, Earth atmosphere entry configurations, planetary probes, and aerobraking concepts. The NASA Ames Research Center led the development of such facilities starting in the 1940s with the Ames Supersonic Free-Flight Tunnel (SFFT).[13] The SFFT, which was similar in many respects to ballistic range facilities used for testing munitions, was designed for aerodynamic and dynamic stability research at high supersonic Mach numbers (Mach numbers in excess of 10). In the SFFT, the model was fired at high speeds upstream into a supersonic airstream (typically Mach 2.0). Windows for shadowgraph photography were along the top and sides of the test section.
Data obtained from motion time histories and measurements of the model’s attitudes during the brief flights were used to obtain aerodynamic and dynamic stability characteristics. The small research models had to be extremely strong to withstand high accelerations during the launch (up to 100,000 g’s), yet light enough to meet requirements for dynamic mass scaling (moments of inertia). Launching the models without angular disturbances or damage was challenging and required extensive development and experience. The SFFT was completed in late 1949 and became operational in the early 1950s.
Ames later brought online its most advanced aeroballistic testing capability, the Ames Hypervelocity Free-Flight Aerodynamic Facility (HFFAF), in 1964. This facility was initially developed in support of the Apollo program and utilized both light-gas gun and shock tube technology to produce lunar return and atmospheric entry. At one end of the test section, a family of light-gas gun was used to launch specimens into the test section, while at the opposite end, a large shock tube could be simultaneously used to produce a counterflowing airstream (the result being Mach numbers of about 30). This counterflow mode of operation proved to be very challenging and was used for only a brief time from 1968 to 1971. Throughout much of the 1970s and 1980s, this versatile facility was operated as a traditional aeroballistic range, using the guns to launch models into quiescent air (or some other test gas), or as a hypervelocity impact test facility. From 1989 through 1995, the facility was operated as a shock tube–driven wind tunnel for scramjet propulsion testing. In 1997, the HFFAF underwent a major refurbishment and was returned to an aeroballistic mode of operation. It continues to operate in this mode and is NASA’s only remaining aeroballistic test facility.[14]
Outdoor Free-Flight Facilities and Test Ranges
Wind tunnel free-flight testing facilities provide unique and very valuable information regarding the flying characteristics of advanced aerospace vehicles. However, they are inherently limited or unsuitable for certain types of investigations in flight dynamics. For example, vehicle motions involving large maneuvers at elevated g’s, out-of-control conditions, and poststall gyrations result in significant changes in flight trajectories and altitude, which can only be studied in the expanded spaces provided by outdoor facilities. In addition, critical studies associated with high-speed flight could not be conducted in Langley’s low-speed wind tunnels. Outdoor testing of dynamically scaled powered and unpowered free-flight models was therefore developed and applied in many research activities. Although outdoor test techniques are more expensive than wind tunnel free-flight tests, are subject to limitations because of weather conditions, and have inherently slower turnaround time than tunnel tests, the results obtained are unique and especially valuable for certain types of flight dynamics studies.
One of the most important outdoor free-flight test techniques developed by NASA is used in the study of aircraft spin entry motions, which includes investigations of spin resistance, poststall gyrations, and recovery controls. A significant void of information exists between the prestall and stall-departure results produced by the wind tunnel free-flight test technique in the Full-Scale Tunnel discussed earlier and the results of fully developed spin evaluations obtained in the Spin Tunnel. The lack of information in this area can be critically misleading for some aircraft designs. For example, some free-flight models exhibit severe instabilities in pitch, yaw, or roll at stall during wind tunnel free-flight tests, and they may also exhibit potentially dangerous spins from which recovery is impossible during spin tunnel tests. However, a combination of aerodynamic, control, and inertial properties can result in this same configuration exhibiting a high degree of resistance to enter the dangerous spin following a departure, despite forced spin entry attempts by a pilot. On the other hand, some configurations easily enter developed spins despite recovery controls applied by the pilot.
To evaluate the resistance of aircraft to spins, in 1950 Langley revisited the catapult techniques of the 1930s and experimented with an indoor catapult-launching technique.[15] Once again, however, the catapult technique proved to be unsatisfactory, and other approaches to study spin entry were pursued.[16] Disappointed by the inherent limitations of the catapult-launched technique, the Langley researchers began to explore the feasibility of an outdoor drop-model technique in which unpowered models would be launched from a helicopter at higher altitudes, permitting more time to study the spin entry and the effects of recovery controls. The technique would use much larger models than those used in the Spin Tunnel, resulting in a desirable increase in the test Reynolds number. After encouraging feasibility experiments were conducted at Langley Air Force Base, a search was conducted to locate a test site for research operations. A suitable low-traffic airport was identified near West Point, VA, about 40 miles from Langley, and research operations began in 1958.[17]
As testing progressed at West Point, the technique evolved into an operation consisting of launching the unpowered model at an altitude of about 2,000 feet and evaluating its spin resistance with separately located, ground-based pilots who attempted to promote spins by various combinations of control inputs and maneuvers. At the end of the test, an onboard recovery parachute was deployed and used to recover the model and lower it to a ground landing. This approach proved to be the prototype of the extremely successful drop-model testing technique that was continually updated and applied by NASA for over 50 years.
Initially, two separate tracking units consisting of modified power-driven antiaircraft gun trailer mounts were used by two pilots and two tracking operators to track and control the model. One pilot and tracker were to the side of the model’s flight path, where they could control the longitudinal motions following launch, while the other pilot and tracker were about 1,000 feet away, behind the model, to control lateral-directional motions. However, as the technique was refined in later years, both pilots used a single dual gun mount arrangement with a single tracker operator.
F/A-18A drop model mounted on its launch rig on a NASA helicopter in preparation for spin entry investigations at the Langley Plum Tree test site. NASA.
Researchers continued their search for a test site nearer to Langley, and in 1959, Langley requested and was granted approval by the Air Force to conduct drop tests at the abandoned Plum Tree bombing range near Poquoson, VA, about 5 miles from Langley. The marshy area under consideration had been cleared by the Air Force of depleted bombs and munitions left from the First and Second World War eras. A temporary building and concrete landing pad for the launch helicopter were added for operations at Plum Tree, and a surge of request jobs for U.S. high-performance military aircraft in the mid- to-late 1960s (F-14, F-15, B-1, F/A-18, etc.) brought a flurry of test activities that continued until the early 1990s.[18]
During operations at Plum Tree, the sophistication of the drop-model technique dramatically increased.[19] High-resolution video cameras were used for tracking the model, and graphic displays were presented to a remote pilot control station, including images of the model in flight and the model’s location within the range. A high-resolution video image of the model was centrally located in front of a pilot station within a building. In addition, digital displays of parameters such as angle of attack, angle of sideslip, altitude, yaw rate, and normal acceleration were also in the pilot’s view. The centerpiece of operational capability was a digital flight control computer programmed with variable research flight control laws and a flight operations computer with telemetry downlinks and uplinks within the temporary building. NASA operations at Plum Tree lasted about 30 years and included a broad scope of free-flight model investigations of military aircraft, general aviation aircraft, parawings, gliding parachutes, and reentry vehicles. In the early 1990s, however, several issues regarding environmental protection forced NASA to close its research activities at Plum Tree and remove all its facilities. After considerable searching and consideration of several candidate sites, the NASA Wallops Flight Facility was chosen for Langley’s drop-model activities.
The last NASA drop-model tests of a military fighter for poststall studies began in 1996 and ended in 2000.[20] This project, which evaluated the spin resistance of a 22-percent-scale model of the U.S. Navy F/A-18E Super Hornet, was the final evolution of drop-model technology for Langley. Launched from a helicopter at an altitude of about 15,000 feet in the vicinity of Wallops, the Super Hornet model weighed about 1,000 pounds. Recovery of the model at the end of the flight test was again initiated with the deployment of onboard parachutes. The model used a flotation bag after water impact and was retrieved from the Atlantic Ocean by a recovery boat.
Outdoor free-flight model testing has also flourished at NASA Dryden Flight Research Center. Dryden’s primary advocate and highly successful user of free-flight models for low-speed research