Category AERONAUTICS

The most dramatic interaction of airplane structure with aerodynam­ics is "flutter”: a dynamic, high-frequency oscillation of some part of the structure. Aeroelastic flutter is a rapid, self-excited motion, potentially destructive to aircraft structures and control surfaces. It has been a par­ticularly persistent problem since invention of the cantilever monoplane at the end of the First World War. The monoplane lacked the "bridge truss” rigidity found in the redundant structure of the externally braced biplane and, as it consisted of a single surface unsupported except at the wing root, was prone to aerodynamic induced flutter. The simplest example of flutter is a free-floating, hinged control surface at the trail­ing edge of a wing, such as an aileron. The control surface will begin to oscillate (flap, like the trailing edge of a flag) as the speed increases. Eventually the motion will feed back through the hinge, into the struc­ture, and the entire wing will vibrate and eventually self-destruct. A similar situation can develop on a single fixed aerodynamic surface, like a wing or tail surface. When aerodynamic forces and moments are applied to the surface, the structure will respond by twisting or bending

about its elastic axis. Depending on the relationship between the elas­tic axis of the structure and the axis of the applied forces and moments, the motion can become self-energizing and a divergent vibration—one increasing in both frequency and amplitude—can follow. The high fre­quency and very rapid divergence of flutter causes it to be one of the most feared, and potentially catastrophic, events that can occur on an aircraft. Accordingly, extensive detailed flutter analyses are performed during the design of most modern aircraft using mathematical mod­els of the structure and the aerodynamics. Flight tests are usually per­formed by temporarily fitting the aircraft with a flutter generator. This consists of an oscillating mass, or small vane, which can be controlled and driven at different frequencies and amplitudes to force an aerody­namic surface to vibrate. Instrumentation monitors and measures the natural damping characteristics of the structure when the flutter gener­ator is suddenly turned off. In this way, the flutter mathematical model (frequency and damping) can be validated at flight conditions below the point of critical divergence.

Traditionally, if flight tests show that flutter margins are insuffi­cient, operational limits are imposed, or structural beef-ups might be accomplished for extreme cases. But as electronic flight control tech­nology advances, the prospect exists for so-called "active” suppression of flutter by using rapid, computer-directed control surface deflections. In the 1970s, NASA Langley undertook the first tests of such a system, on a one-seventeenth scale model of a proposed Boeing Supersonic Transport (SST) design, in the Langley Transonic Dynamics Tunnel (TDT). Encouraged, Center researchers followed this with TDT tests of a stores flutter suppression system on the model of the Northrop YF-17, in concert with the Air Force Flight Dynamics Laboratory (AFFDL, now the Air Force Research Laboratory’s Air Vehicles Directorate), later implementing a similar program on the General Dynamics YF-16. Then, NASA DFRC researchers modified a Ryan Firebee drone with such a system. This program, Drones for Aerodynamic and Structural Testing (DAST), used a Ryan BQM-34 Firebee II, an uncrewed aerial vehicle, rather than an inhabited system, because of the obvious risk to the pilot for such an experiment.

The modified Firebee made two successful flights but then, in June 1980, crashed on its third flight. Postflight analysis showed that one of the software gains had been inadvertently set three times higher than planned, causing the airplane wing to flutter explosively right after launch

A Drones for Aerodynamic and Structural Testing (DAST) unpiloted structural test vehicle, derived from the Ryan Firebee, during a 1980 flight test. NASA.

from the B-52 mother ship. In spite of the accident, progress was made in the definition of various control laws that could be used in the future for control and suppression of flutter.[714] Overall, NASA research on active flutter suppression has been generally so encouraging that the fruits of it were applied to new aircraft designs, most notably in the "growth” ver­sion of the YF-17, the McDonnell-Douglas (now Boeing) F/A-18 Hornet strike fighter. It used an Active Oscillation Suppression (AOS) system to suppress flutter tendencies induced by its wing-mounted stores and wingtip Sidewinder missiles, inspired to a significant degree by earlier YF-17 and YF-16 Transonic Dynamics Tunnel testing.[715]

Ceramic tiles, of the kind used in a blast furnace or fireplace to insulate the surrounding structure from the extreme temperatures, were far too heavy to be considered for use on a flight vehicle. The concept of a light­weight ceramic tile for thermal protection was conceived by Lockheed and developed into operational use by NASA Ames Research Center, NASA Johnson Space Center, and Rockwell International for use on the Space Shuttle orbiter, first flown into orbit in April 1981. The result­ing tiles and ceramic blankets provided exceptionally light and efficient thermal protection for the orbiter without altering the external shape. Although highly efficient for thermal protection, the tiles were—and are— quite fragile and time-consuming to repair and maintain. The Shuttle program experienced considerable delays prior to its first flight because of bonding, breaking, and other installation issues. (Unlike the X-15 grad­ual envelope expansion program, the Shuttle orbiter was exposed to its full operational flight envelope on its very first orbital flight and entry, thus introducing a great deal of analysis and caution during flight prep­aration.) Subsequent Shuttle history confirmed the high-maintenance nature of the tiles, and their vulnerability to external damage such as ice or insulation shedding from the super-cold external propellant tank. Even with these limitations, however, they do constitute the most prom­ising technology for future lifting entry vehicles.[757]

The first computers were analog computers. Direct analog computers are networks of physical components (most commonly, electrical components: resistors, capacitors, inductances, and transformers) whose behavior is gov­erned by the same equations as some system of interest that is being mod­eled. Direct analog computers were used in the 1950s and 1960s to solve problems in structural analysis, heat transfer, fluid flow, and other fields.

The method of analysis and the needs that were driving the move from classical idealizations such as slender-beam theory toward computational

methods are well stated in the following passage, from an NACA-sponsored paper by Stanley Benscoter and Richard MacNeal (subsequently a cofounder of the MacNeal Schwendler Corporation [MSC] and member of the NASTRAN development team):

The theory is expressed entirely in terms of first-order differ­ence equations in order that analogous electrical circuits can be readily designed and solutions obtained on the Caltech ana­log computer. . . . In the process of designing thin supersonic wings for minimum weight it is found that a convenient con­struction with aluminum alloy consists of a rather thick skin with closely spaced spars and no stringers. Such a wing deflects in the manner of a plate rather than as a beam. Internal stress distributions may be considerably different from those given by beam theory.[794]

Their implementation of analog circuitry for bending loads is illus­trated here and serves as an example of the direct analog modeling of structures.[795]

Direct analog computing had its advocates well into the 1960s. "For complex problems [direct analog] computers are inherently faster than digital machines since they solve the equations for the several nodes simultaneously, while the digital machines solve them sequen­tially. Direct analogs have, moreover, the advantage of visualization;

computer setups as well as programming are more closely related to the actual problem and are based primarily on physical insight rather than on numerical skills.”[796]

The advantages came at a price, however. It could take weeks, in some cases, to set up an analog computer to solve a particular type of problem. And there was no way to store a problem to be revisited at a later date. These drawbacks may not have seemed so important when there was no other recourse available, but they became more and more apparent as the programmable digital computer began to mature.

Hybrid direct-analog/digital computers were hypothesized in the 1960s: essentially a direct analog computer controlled by a digital computer capable of storing and executing program instructions. This would have overcome some of the drawbacks of direct analog com­puters.[797] However, this possibility was most likely overtaken by the rapid progress of digital computers. At the same time these hybrid ana – log/digital computers were just being thought about, NASTRAN was already in development.

A different type of analog computer—the active-element, or indi­rect, analog—consisted of operational amplifiers that performed arith­metic operations. These solved programmed mathematical equations, rather than mimicking a physical system. Several NACA locations— including Langley, Ames, and the Flight Research Center (now Dryden Flight Research Center)—used analog computers of this type for flight simulation. Ames installed its first analog computer in 1947.[798] The Flight Research Center flight simulators used analog computers exclusively from 1955 to 1964 and in combination with digital computers until 1975.[799] This type of analog computer can be thought of as simply a less precise, less reliable, and less versatile predecessor to the digital computer.

NASA operated two Lockheed YF-12As and one "YF-12C” (actually an early nonstandard SR-71A, although the Air Force at that time could not acknowledge that it was allowing NASA to operate an SR-71) between 1969 and 1979.[936] These aircraft were used for a variety of research proj­ects. In some projects, the YF-12s were the test articles, exploring their performance, handling qualities, and propulsion system characteristics in various baseline or modified configurations and modes of operation. In other projects, the YF-12s were used as "flying wind tunnels” to carry test models and other experiments into the Mach 3+ flight environment. Testing directly related to structural analysis methods and/or loads pre­diction included a series of thermal-structural load tests from 1969 to 1972 and smaller projects concerning ventral fin loads and structural

(a) Surface temperature distribution (deg K) at cruise

(b) Time history of typical wing spar temperatures

Temperature time histories from YF12 flight project. NASA.

dynamics.[937] The flight-testing was conducted at Dryden, which was also responsible for project management. Ames, Langley, and Lewis Research Centers were all involved in technical planning, analysis, and supporting research activities, coordinated through NASA Headquarters. The U. S. Air Force and Lockheed also provided support in various areas.[938] Gene Matranga of Dryden was the manager of the program before Berwin Kock later assumed that role.[939]

The thermal-structural loads project involved modeling and test­ing in Dryden’s unique thermal load facility. The purpose was to corre­late in-flight and ground-test measurements and analytical predictions of temperatures, mechanical loads, strains, and deflections. "In all the X-15 work, flight conditions were always transient. The vehicle went to high speed in a matter of two to three minutes. It slowed down in a matter of three to five minutes. . . . The YF-12, on the other hand, could stay at Mach 3 for 15 minutes. We could get steady-state temperature
data, which would augment the X-15 data immeasurably.”[940] The YF-12 testing showed that it could take up to 15 minutes for absolute tem­peratures in the internal structure to approach steady state, and, even then, the gradients—which have a strong effect on stresses because of differential expansion—did not approach steady state until close to 30 minutes into the cruise.[941]

NASTRAN and FLEXSTAB (a code developed by Boeing on contract to NASA Ames to predict aeroelastic effects on stability) were used to model the YF-12A’s aeroelastic and aerothermoelastic characteristics. Alan Carter and Perry Polentz of NASA oversaw the modeling effort, which was contracted to Lockheed and accomplished by Al Curtis. This effort produced what was claimed to be the most extensive full-vehicle NASTRAN model developed up to that time. The computational models were used to predict loads and deflections, and also to identify appro­priate locations for the strain gauges that would take measurements in ground – and flight-testing. The instrumentation included strain gauges, thermocouples, and a camera mounted on the fuselage to record air­frame deflection in flight. Most of the flights, from Flight 11 in April 1970 through Flight 53 in February 1972, included data collection for this project, often mixed with other test objectives.[942] Subsequently, the air­craft ceased flying for more than a year to undergo ground tests in the high-temperature loads laboratory. The temperatures measured in flight were matched on the ground, using heated "blankets” placed over dif­ferent parts of the airframe. Ground-testing with no aerodynamic load allowed the thermal effects to be isolated from the aerodynamic effects.[943]

There were also projects involving the measurement of aerodynamic loads on the ventral fin and the excitation of structural dynamic modes. The ventral fin project was conducted to provide improved understand­ing of the aerodynamics of low aspect ratio surfaces. FLEXSTAB was used in this effort but only for linear aerodynamic predictions. Ground tests had shown the fin to be stiff enough to be treated as a rigid surface. Measured load data were compared to the linear theory predictions and to wind tunnel data.[944] For the structural dynamics tests, which occurred near the end of NASA’s YF-12A program, "shaker vanes”—essentially oscillating canards—were installed to excite structural modes in flight. Six flights with shaker vanes between November 1978 and March 1979 "provided flight data on aeroelastic response, allowed comparison with calculated response data, and thereby validated analytical techniques.”[945] Experiences from the program were communicated to industry and other interested organizations in a YF-12 Experiments Symposium that was held at Dryden in 1978, near the end of the 10-year effort.[946] There were also briefings to Boeing, specifically intended to provide informa­tion that would be useful on the Supersonic Transport (SST) program, which was canceled in 1971.[947] There have been other civil supersonic projects since then—the High-Speed Civil Transport (HSCT)/High-Speed Research (HSR) efforts in the 1990s and some efforts related to super­sonic business jets since 2000—but none have yet led to an operational civil supersonic aircraft.

One particular facility of many, a spin test rig built at Lewis in 1983 is mentioned here because its stated purpose was not primarily the test­ing of engine parts to verify the parts but the testing of engine parts to verify analysis methods: "The Lewis Research Center spin rig was con­structed to provide experimental evaluation of analysis methods devel­oped under the NASA Engine Structural Dynamics Program. Rotors up to 51 cm (20 in.) in diameter can be spun to 16,000 rpm in vacuum by an air motor. Vibration forcing functions are provided by shakers that apply oscillatory axial forces or transverse moments to the shaft, by a natural whirling of the shaft, and by an air jet. Blade vibration is detected by strain gages and optical tip blade-motion sensors.”[1012]

During the 1960s, two major events would unfold in the United States that had very strong influence on the development and eventual
introduction into operational service of advanced computer-controlled fly-by-wire flight control systems. Early in his administration, President John F. Kennedy had initiated the NASA Apollo program with the goal of placing a man on the Moon and safely bringing him back to Earth by the end of the decade. The space program, and Apollo in particular, would lead to major strides in the application of the digital computer to manage and control sensors, systems, and advanced fly-by-wire vehicles (eventually including piloted aircraft). During the same period, America became increasingly involved in the expanding conflict in South Vietnam, an involvement that rapidly escalated as the war expanded into a con­ventional conflict with dimensions far beyond what was originally fore­seen. As combat operations intensified in Southeast Asia, large-scale

U. S. strike missions began to be flown against North Vietnam. In response, the Soviet Union equipped North Vietnamese forces with improved air defense weapons, including advanced fighters, air-to-air and surface-to-air missiles, and massive quantities of conventional antiaircraft weapons, ranging in caliber from 12.7 to 100 millimeters (mm). U. S. aircraft losses rose dramatically, and American warplane designs came under increasing scrutiny as the war escalated.[1132] Analyses of combat data revealed that many aircraft losses resulted from battle damage to hydromechanical flight control system components. Traditionally, pri­mary and secondary hydraulic system lines had been routed in paral­lel through the aircraft structure to the flight control system actuators. In the Vietnam combat, experience revealed that loss of hydraulic fluid because of battle damage often led to catastrophic fires or total loss of aircraft control. Aircraft modification programs were developed to reroute and separate primary and secondary hydraulic lines to reduce the possibility of a total loss of fluid given a hit. Other changes to exist­ing aircraft flight control systems improved survivability, such as a mod­ification to the F-4 that froze the horizontal tail in the neutral position to prevent the aircraft from going out of control when hydraulic fluid was lost.[1133] However, there was an increasing body of opinion that felt a
new approach to flight control system design was necessary and tech­nically feasible.

From mid-1983 through mid-1984, components for the Automated Maneuvering Attack System and related avionics systems were installed into the AFTI/F-16 at GD in Fort Worth in preparation for the Phase II effort. Precision electrical-optical tracking pods were installed in the wing root area on both sides of the aircraft. First flight of the AFTI/F-16 in the AMAS configuration was on July 31, 1984, with Phase II flight-testing at Edwards beginning shortly after the aircraft returned to Dryden on August 6, 1984. Beginning in September 1984 and continuing through April 1987, improved sensors, integrated fire and flight control, and enhancements
in pilot-vehicle interface were evaluated. During Phase II testing, the system demonstrated automatic gun tracking of airborne targets and accurate delivery of unguided bombs during 5-g curvilinear toss bomb maneuvers from altitudes as low as 200 feet. An all-attitude automatic ground collision avoidance capability was demonstrated,[1185] as was the Voice Command System (for interfacing with the avionics system), a helmet-mounted sight (used for high off bore sight target cueing), and a digital terrain system with color moving map.[1186] The sortie rate dur­ing Phase II was very high. From the start of the AMAS tests in August 1984 to the completion of Phase II in early 1987, 226 flights were accom­plished, with 160 sorties being flown during 1986. To manage this high sortie rate, the ground maintenance crews worked a two-shift operation.

Follow-On AFTI/F-16 Testing

Following Phase II in 1987, the forward fuselage-mounted ventral fins were removed and the AFTI/F-16 was flown in support of other test efforts and new aircraft programs, such as evaluating strike technologies pro­posed for use in the next generation ground attack aircraft, which even­tually evolved into the Joint Strike Fighter (JSF) program.

The Adaptive Engine Control System (ADECS) improved engine perfor­mance by exploiting the excess stall margin originally designed into the engines using capabilities made possible with the integrated comput­erized flight and engine control systems.[1263] ADECS used airframe and engine data to allow the engine to operate at higher performance levels at times when inlet distortion was low and the full engine stall margin is not needed. Initial engineering work on ADECS began in 1983, with research flights beginning in 1986. Test results showed thrust improve­ments of between 8 and 10 percent depending on altitude. Fuel flow reductions of between 7 and 17 percent at maximum afterburning thrust at an altitude of 30,000 feet were recorded. Rate of climb increased 14 percent at 40,000 feet. Time required to climb from 10,000 feet to 40,000 feet dropped 13 percent. Acceleration improved between 5 and 24 per­cent at intermediate and maximum power settings, depending on altitude. No unintentional engine stalls were encountered in the test program.

ADECS technology has been incorporated into the Pratt & Whitney F119 engine used on the Air Force F-22 Raptor.[1264]

As one set of NASA and contractor engineers worked on improving the design of the various types of jet engines, another set of researchers rep­resenting another science discipline were increasingly interested in mar­rying the computer’s capabilities to the operation of a jet engine, much in the same way that fly-by-wire systems already were in use with air­craft flight controls.

Beginning with that first Wright Flyer in 1903, flying an airplane meant moving levers and other mechanical contrivances that were directly connected by wires and cables to control the operation of the rudder, elevator, wing surfaces, instruments, and engine. When Chuck Yeager broke the sound barrier in 1947 in the X-1, if he wanted to go up, he pulled back on the yoke and cables directly connecting the stick to the elevator, which made that aerosurface move to effect a change in the aircraft’s attitude. The rockets propelling the X-1 were activated with a switch throw that closed an electrical circuit whose wiring led directly from the cockpit to the engines. As planes grew bigger, so did their control surfaces. Aircraft such as the B-52 bomber had aerosur – faces as big as the entire wings of smaller airplanes—too bulky and heavy for a single pilot to move using a simple cable/pulley system. A hydrau­lic system was required and "inserted” between the pilot’s input on the yoke and the control surface needing to be moved. Meanwhile, engine
operation remained more or less "old fashioned,” with all parameters such as fuel flow and engine temperatures reported to the cockpit on dials the pilot could read, react to, and then make changes by adjust­ing the throttle or other engine controls.

With the introduction of digital computers and the miniaturiza­tion of their circuits—a necessity inspired, in part, by the reduced mass requirements of space flight—engineers began to consider how the quick-thinking electronic marvels might ease the workload for pilots flying increasingly more complex aircraft designs. In fact, as the 1960s transitioned to the 1970s, engineers were already considering aircraft designs that could do remarkable maneuvers in the sky but were inher­ently unstable, requiring constant, subtle adjustments to the flight con­trols to keep the vehicle in the air. The solution—already demonstrated for spacecraft applications during Project Apollo—was to insert the power of the computer between the cockpit controls and the flight con­trol surfaces—a concept known as fly-by-wire. A pilot using this system and wanting to turn left would move the control stick to the left, apply a little back pressure, and depress the left rudder pedal. Instead of a wire/cable system directly moving the related aerosurfaces, the move­ment of the controls would be sensed by a computer, which would send electronic impulses to the appropriate actuators, which in turn would deflect the ailerons, elevator, and rudder.[1328]

Managed by NASA’s Dryden Flight Research Facility, the fly-by-wire system was first tested without a backup mechanical system in 1972, when a modified F-8C fighter took off from Edwards Air Force Base in California. Testing on this aircraft, whose aerodynamics were known and considered stable, proved that fly-by-wire could work and be reliable. In the years to follow, the system was used to allow pilots to safely fly unstable aircraft, including the B-2 bomber, the forward-swept winged X-29, the Space Shuttle orbiter, and commercial airliners such as the Airbus A320 and Boeing 777.[1329]

As experienced was gained with the digital flight control system and computers shrunk in size and grew in power, it didn’t take long for pro­pulsion experts to start thinking about how computers could monitor
engine performance and, by making many adjustments in every vari­able that affects the efficiency of a jet engine, improve the powerplant’s overall capabilities.

The first step toward enabling computer control of engine operations was taken by Dryden engineers in managing the Integrated Propulsion Control System (IPCS) program during the mid-1970s. A joint effort with the U. S. Air Force, the IPCS was installed on an F-111E long – range tactical fighter-bomber aircraft. The jet was powered by twin TF30 afterburning turbofan engines with variable-geometry external com­pression inlets. The IPCS effort installed a digital computer to control the variable inlet and realized significant performance improvements in stallfree operations, faster throttle response, increased thrust, and improved range flying at supersonic speeds. During this same period, results from the IPCS tests were applied to NASA’s YF-12C Blackbird, a civilian research version of the famous SR-71 Blackbird spy plane. A digital control system installed on the YF-12C successfully tested, mon­itored, and adjusted the engine inlet control, autothrottle, air data, and navigation functions for the Pratt & Whitney-built engines. The results gave the aircraft a 7-percent increase in range, improved handling char­acteristics, and lowered the frequency of inlet unstarts, which happen when an engine shock wave moves forward of the inlet and disrupts the flow of air into the engine, causing it to shutdown. Seeing how well this computer-controlled engine worked, Pratt & Whitney and the U. S. Air Force in 1983 chose to incorporate the system into their SR-71 fleet.[1330]

The promising future for more efficient jet engines from develop­ing digitally controlled integrated systems prompted Pratt & Whitney, the Air Force, and NASA (involving both Dryden and Lewis) to pur­sue a more robust system, which became the Digital Electronic Engine Control (DEEC) program.

Pratt & Whitney actually started what would become the DEEC pro­gram, using its own research and development funds to pay for configura­tion studies beginning during 1973. Then, in 1978, Lewis engineers tested a breadboard version of a computer-controlled system on an engine in an altitude chamber. By 1979, the Air Force had approached NASA and asked if Dryden could demonstrate and evaluate a DEEC system using an F100 engine installed in a NASA F-15, with flight tests beginning in

1981. At every step in the test program, researchers took advantage of les­sons learned not only from the IPCS exercise but also from a U. S. Navy – funded effort called the Full Authority Digital Engine Control program, which ran concurrently to the IPCS program during the mid-1970s.[1331]

A NASA Dryden fact sheet about the control system does a good job of explaining in a concise manner the hardware involved, what it moni­tored, and the resulting actions it was capable of performing:

The DEEC system tested on the NASA F-15 was an engine mounted, fuel-cooled, single-channel digital controller that received inputs from the airframe and engine to control a wide range of engine functions, such as inlet guide vanes, compres­sor stators, bleeds, main burner fuel flow, afterburner fuel flow and exhaust nozzle vanes.

Engine input measurements that led to these computer – controlled functions included static pressure at the compres­sor face, fan and core RPM, compressor face temperature, burner pressure, turbine inlet temperature, turbine discharge pressure, throttle position, afterburner fuel flow, fan and com­pressor speeds and an ultra violet detector in the afterburner to check for flame presence.

Functions carried out after input data were processed by the DEEC computer included setting the variable vanes, position­ing compressor start bleeds, controlling gas-generator and augmentation of fuel flows, adjusting the augmenter segment – sequence valve, and controlling the exhaust nozzle position.

These actions, and others, gave the engine—and the pilot— rapid and stable throttle response, protection from fan and compressor stalls, improved thrust, better performance at high altitudes, and they kept the engine operating within its limits over the full flight envelope.[1332]

When incorporated into the F100 engine, the DEEC provided improve­ments such as faster throttle responses, more reliable capability to restart an engine in flight, an increase of more than 10,000 feet in altitude when firing the afterburners, and the capability of providing stallfree operations. And with the engine running more efficiently thanks to the DEEC, overall engine and aircraft reliability and maintainability were improved as well.[1333]

So successful and promising was this program that even before test­ing was complete the Air Force approved widespread production of the F100 control units for its F-15 and F-16 fighter fleet. Almost at the same time, Pratt & Whitney added the digital control technology in its PW2037 turbofan engines for the then-new Boeing 757 airliner.[1334]

With the DEEC program fully opening the door to computer control of key engine functions, and with the continuing understanding of fly­by-wire systems for aircraft control—along with steady improvements in making computers faster, more capable, and smaller—the next logi-

cal step was to combine together computer control of engines and flight controls. This was done initially with the Adaptive Engine Control System (ADECS) program accomplished between 1985 and 1989, followed by the Performance Seeking Control (PSC) program that performed 72 flight tests between 1990 and 1993. The PSC system was designed to handle multiple variables in performance, compared with the single-variable control allowed in ADECS. The PSC effort was designed to optimize the engine and flight con­trols in four modes: minimum fuel flow at constant thrust, minimum turbine temperature at constant thrust, maximum thrust, and minimum thrust.[1335]

The next evolution in the combining of computer-controlled flight and engine controls— a legacy of the original DEEC program—was inspired in large part by the 1989 crash in Sioux City, IA, of a DC-10 that had lost all three of its hydraulic systems when there was an uncon­tained failure of the aircraft’s No. 2 engine. With three pilots in the cockpit, no working flight controls, and only the thrust levels available for the two remaining working engines, the crew was able to steer the jet to the airport by using variable thrust. During the landing, the airliner broke apart, killing 111 of the 296 people on board.[1336]

Soon thereafter, Dryden managers established a program to thor­oughly investigate the idea of a Propulsion Controlled Aircraft (PCA) using variable thrust between engines to maintain safe flight control. Once again, the NASA F-15 was pressed into service to demonstrate the concept. Beginning in 1991 with a general ability to steer, refine­ments in the procedures were made and tested, allowing for more precise maneuvering. Finally, on April 21, 1993, the flight tests of PCA concluded with a successful landing using only engine power to climb, descend, and maneuver. Research continued using an MD-11 airliner, which success­fully demonstrated the technology in 1995.[1337]

The energy crisis of the 1970s brought about renewed interest in the development of alternative energy sources, including harnessing wind power for the generation of electricity. This renewed interest led to the establishment of the Federal Wind Energy Program in 1974 as part of the Nation’s solar energy program. The initial program overview, technical analysis, and objectives were formalized by the Project Independence Interagency Solar Task Force that was formed in April 1974 and chaired by the National Science Foundation (NSF). Approximately 100
individuals—representing various Government agencies, universities, research laboratories, private industries, and consulting firms—par­ticipated in the task force project. Thirteen of the participants were from NASA, including six from NASA Lewis (now NASA John H. Glenn Research Center). The task force’s final findings were outlined in the November 1974 "Project Independence Blueprint” report. The task force identified the six following "most promising” technologies for converting solar energy to a variety of useful energy forms: (1) solar heating and cooling of buildings, (2) solar thermal energy conversion, (3) wind energy conversion, (4) bioconversion to fuels, (5) ocean ther­mal energy conversion, and (6) photovoltaic electric power systems. The task force noted that the objective of the wind energy conversion part of the program was to improve the efficiency of wind turbine sys­tems in a variety of applications and to reduce their costs. In regard to site selection, the task force concluded that the first attainment of eco­nomic viability in the United States would occur in areas such as the Great Plains, Alaska, the Great Lakes, the Atlantic and Pacific coasts, New England, and Hawaii. It concluded that the key to large-scale application of wind energy conversion systems was the reduction of costs through advanced technology, new materials, mass production, and the use of field fabrication techniques. Finally, the task force noted that a closely moni­tored program of proof-of-concept experiments was expected to reduce cost and constraint uncertainties.[1492]

As a prelude to the formation of the wind energy program, NASA Lewis made significant contributions to a wind energy workshop that reviewed both the current status of wind energy and assessed the poten­tial of wind power. This workshop was held as part of the Research Applied to National Needs (RANN) project that led to the National Science Foundation’s role in the initial planning of a 5-year sustained wind energy program. In January 1975, the wind energy program was transferred to the newly formed Energy Research and Development Administration (ERDA), which was incorporated into the newly formed

U. S. Department of Energy (DOE) in 1977.

Pursuant to the initial agreement between NASA and the NSF, which had no research centers of its own, NASA’s Lewis Research Center at

Lewis Field in Cleveland, OH, was given overall project management for the portion of the Wind Energy Program that involved the development and fabrication of large experimental horizontal-axis wind turbines. NASA Lewis’s responsibilities also included the conduct of supporting research and technology for the wind turbine conversion systems. This sponsorship continued under the Department of Energy once DOE took over the Federal Wind Energy Program. Louis Divone, who initially selected NASA Lewis to participate in the program, was the wind energy program manager for the NSF and later for ERDA and DOE. The pro­gram goal was the development of the technology for safe, reliable, and environmentally acceptable large wind turbine systems that could gen­erate significant amounts of electricity at costs competitive with con­ventional electricity-generating systems.

NASA Lewis engineers were very interested in getting involved in the Wind Energy Program and realized early on that they could make significant contributions because of the Research Center’s long expe­rience and expertise in propeller and power systems, aerodynamics, materials, and structures testing. The selection of NASA Lewis also rep­resented an interesting historical context. Over 85 years earlier, in 1887— 1888, in Cleveland, OH, an engineer by the name of Charles F. Brush constructed a 60-foot, 80,000-pound wind-electric dynamo that is gen­erally credited as being the first automatically operating wind turbine for electricity generation. Brush’s wind turbine, which supplied power for his home for up to 10 years, could produce a maximum 12,000 watts of direct current that charged 12 batteries that in turn ran 350 incandescent lights, 2 arc lights, and a number of electric motors. His dynamo made 50 revolutions to 1 revolution of the wind wheel, which consisted of 144 wooden blades and was 56 feet in diameter, accounting for 1,800 square feet of total blade surface swept area. The wind dynamo had an automatic regulator that prevented the power from running above 90 volts at any speed. Brush later dismantled his wind dynamo, apparently without attempting to develop a unit that could feed into a central power network.[1493]

The use of wind power to generate electricity achieved a degree of success in rural and remote areas of the United States in the 1920s and 1930s. These generators, however, were small, stand-alone wind-
electric systems such as those designed and marketed by Marcellus and Joseph Jacobs, who built three-bladed systems, and the Windcharger Corporation, which built two-bladed generators. Most of these efforts were abandoned in the 1940s and 1950s because of the expansion of electrical utility networks, especially in response to passage of the Rural Electrification Act of 1937 and the availability of low-cost fossil fuels.

The first American effort to build a large wind turbine to feed into a power network was undertaken by Palmer Cosslett Putnam. This effort was funded by the S. Morgan Smith Company, which constructed and installed a 1.25-megawatt wind turbine at Grandpa’s Knob, VT. Prior to fabrication of his turbine, Putnam considered a number of questions that were still being debated years later, including whether to build a vertical – or horizontal-axis wind turbine; if horizontal, how many blades should there be; whether the generator should be aloft or on the ground; whether the drive should be mechanical or hydraulic; whether the tower should rotate or be stationary; and what size generator should be used. He noted that examples of all of these configurations existed in writings on wind power. Putnam, with the concurrence of both Beauchamp and Burwell Smith, decided on using the horizontal-axis, two-bladed stain­less steel configuration, with a mechanically driven synchronous gener­ator mounted aloft. He then concluded that the optimum size of a wind turbine generator (WTG) was close to 2 megawatts and noted that stud­ies indicated that increased efficiency appeared to be flat between 2 and 3 megawatts.[1494] The Smith-Putnam wind turbine, which supplied power to the Central Vermont Public Service Corporation’s power network, started operations on October 19, 1941, and operated intermittently for a total electric generation period of approximately 16 months. A bearing fail­ure caused a blade separation accident, and the project was terminated in March 1945. While the turbine was not rebuilt, the system’s operation demonstrated that wind could be harnessed on a large scale to produce electricity. The power company, as well as others, envisioned that wind turbines would operate in conjunction with hydroelectric power systems.[1495]

In the late 1950s, a German engineer by the name of Ulrich Hutter also built a smaller, 100-kilowatt wind turbine generator (the Hutter-

Allgaier wind turbine) that was tied into a power utility grid. Hutter’s machine used a 112-foot-diameter, two-bladed downwind rotor with full span pitch control. The blades were mounted on a teetered hub. In prep­aration to commence work on its own wind turbines, NASA purchased the plans from Hutter and considered or incorporated a number of design criteria and features of both the Smith-Putnam and Hutter-Allgaier wind turbine generators.[1496] NASA also participated in a joint NASA-Danish financ­ing of the restoration of the wind turbine, which was completed in 1977. NASA Lewis later did aerodynamics testing and modeling of the Gedser wind turbine information using the Mod-0 testbed turbine.

NASA Lewis’s involvement in wind energy leading up to its selection to oversee the wind turbine development portion of the Federal Wind Energy Program included designing, at the request of Puerto Rico, a wind tur­bine to generate electricity for the Island of Culebra. This project grew out of an unrelated NASA Lewis 1972 project to take wind measurements in Puerto Rico. Later on, under the Wind Energy Program, NASA returned to Puerto Rico to build one of the Agency’s first-generation (Mod-0A) wind turbine machines. NASA Lewis’s involvement in the Wind Energy Program also was enhanced by its research of past wind energy projects and its projection of the future feasibility of using wind power to gener­ate electricity for U. S. power utility networks. NASA’s overview and find­ings were presented as a paper at a symposium held in Washington, DC, that brought together past developers of wind turbines, including Palmer Putnam, Beauchamp Smith, Marcellus Jacobs, and Ulrich Hutter, as well as a new group of interested wind energy advocates.