NASA Technical Memorandum 110311

Research and Applications in Structures at the NASA Langley Research Center

Irving AbelLangley Research Center, Hampton, Virginia

January 1997

National Aeronautics andSpace AdministrationLangley Research CenterHampton, Virginia 23681-0001

1

RESEARCH AND APPLICATIONS IN STRUCTURES AT THE NASA LANGLEYRESEARCH CENTER

Irving AbelNASA Langley Research Center

Hampton, Virginia USA

ABSTRACT

An overview of recently completed programs instructures research at the NASA Langley ResearchCenter is presented. Also included is a descriptionof the unique facilities used to support thestructures program. Methods used to performflutter clearance studies in the wind-tunnel on ahigh performance fighter are discussed. Recentadvances in the use of smart structures and controlsto solve the aeroelastic problems of fixed- androtary-wing vehicles, including flutter, loads,vibrations, and structural response are presented.The use of photogrammetric methods in space tomeasure spacecraft dynamic response isdiscussed. The use of advanced analyticalmethods to speed up detailed structural analysis ispresented. Finally, the application of cost-effectivecomposite materials to wing and fuselage primarystructures is illustrated.

INTRODUCTION

The Langley Research Center (LaRC) has beendesignated the NASA "Center of Excellence" forStructures and Materials research. The StructuresDivision at LaRC conducts analytical andexperimental research (figure 1) in Aeroelasticity,Structural Mechanics, Computational Structures,Structural Dynamics, and Thermal Structures tomeet the technology requirements for advancedaerospace vehicles. The charter for each researcharea is given in figure 2.

The Structures Division supports the developmentof more efficient structures for airplanes,helicopters, spacecraft, and space transportationvehicles. Analytical methods for improvingstructural analysis and design are developed andvalidated by experimental methods. New structuralconcepts for both metal and composite structuresare developed and evaluated through laboratorytesting. Research is conducted to integrateadvanced structural concepts with active-controlconcepts and smart materials to enhance structuralperformance. Studies of impact dynamics focus on

increased survivability in the case of crash impact.Research in thermal structures is aimed at efficientstructural concepts for future high-speed aircraftand space transportation systems that exploit thebenefits of advanced composite and metallicmaterials. Research in aeroelasticity ranges fromflutter clearance studies of new vehicles usingaeroelastic models tested in the wind tunnel, to thedevelopment of new concepts to control aeroelasticresponse, and to the acquisition of unsteadypressures on wind-tunnel models for providingexperimental data to validate unsteady theories.Analytical methods are developed and validated tosolve the aeroelastic problems of fixed- and rotary-wing vehicles, including the control of instabilities,loads, vibration, and adverse structural response.

This paper presents a brief overview of the testfacilities operated by the Structures Division and theresults of some selected studies in structuralmechanics and dynamics during the last 2 years.This paper begins with an overview of how flutterclearance studies are performed in the wind tunnel.The paper then addresses research aimed at usingsmart materials to suppress aeroelastic response, atacquiring an experimental data base to validatecomputational fluid dynamics codes, at the use ofsmart materials and control surfaces to reduce thebuffeting response of a modern twin-tail fighter, atthe use of flaperons to control tiltrotor vibratoryloads, at using advanced photogrammetric methodsto measure the dynamic response of spacecraft inorbit, at developing new computational methodsthat significantly reduces the time and cost toperform structural analysis, at the use of advancedconcepts to improve the survivability of occupantsin a composite aircraft during a crash, and at aprogram to exploit the use of composites in primarystructure of advanced transport aircraft.

EXPERIMENTAL FACILITIES

The structures research program at LaRC requiresthe support of a unique set of experimentalfacilities. These facilities include the TransonicDynamics Tunnel, the Structural Mechanics

2

Laboratory, the Aircraft Landing Dynamics Facility,the Impact Dynamics Research Facility, the ThermalStructures Laboratory, the Structural DynamicsLaboratory, and the Combined Loads TestingSystem (figure 3). Replacement value of thesefacilities is estimated to be in excess of $350 millionU.S. dollars.

A short description of each of these experimentalfacilities follows:

Transonic Dynamics Tunnel

The Transonic Dynamics Tunnel (TDT) is a unique"national" facility dedicated to identifying,understanding, and solving aeroelastic problems.The TDT is a closed-circuit, continuous-flow,variable-pressure, wind tunnel with a 16-foot squaretest section. The tunnel uses either air or a heavygas as the test medium and can operate atstagnation pressures from near vacuum toatmospheric, has a Mach number range from nearzero to 1.2, and is capable of maximum Reynoldsnumbers of about 3 million per foot in air to 10million per foot in heavy gas. The TDT is speciallyconfigured for flutter testing, with excellent modelvisibility from the control room and a rapid tunnelshutdown capability for model safety. Model mountsystems include two sidewall turntables forsemispan models, a variety of stings for full-spanmodels, a cable-mount system for “flying” models, arotorcraft testbed for rotor blade loads research, anda floor turntable for launch vehicle ground-windloads studies. The TDT also offers an airstreamoscillation system for gust studies and supportingsystems for active controls testing. Testing in heavygas has important advantages over testing in airincluding improved model to full-scale similitude,higher Reynolds numbers, and reduced tunnelpower requirements. The TDT is the only windtunnel in the world capable of flutter testing large,full-span, aeroelastically-scaled models at transonicspeeds.

Structural Mechanics Laboratory

Built in 1939 to contribute to the development andvalidation of aircraft structural designs during WorldWar II, this laboratory currently supports a broadrange of structural and materials developmentactivities for advanced aerospace structures. Statictesting, environmental testing, and materialfabrication and analysis are performed. Emphasis is

on the development of structural mechanicstechnology and advanced structural conceptsenabling the verified design of efficient, cost-effective, damage-tolerant, advanced-compositestructural components subjected to complexloading and demanding environmental conditions.This facility contains unique specially designed120-, 300-, and 1,200-kip test machines withspecial platens for precision compression testingand a strong-back load-reaction structure capable oftesting large test specimens subjected to loadsfrom multiple hydraulic actuators.

Aircraft Landing Dynamics Facility

The Aircraft Landing Dynamics Facility (ALDF) is atest track used primarily for landing gear, tire, andrunway surface research studies. The ALDF uses ahigh-pressure water-jet system to propel a testcarriage down a 2800-ft track. The propulsionsystem consists of a vessel that holds 28000gallons of water pressurized up to 3150 psi. Aquick-opening shutter valve releases a high energywater jet, which catapults the carriage to the desiredspeed. The propulsion system produces a thrust inexcess of 2 million lbs, which is capable ofaccelerating the 54-ton test carriage to speeds of220 knots within 400 ft. This thrust creates a peakacceleration of approximately 20g’s on the carriage.The carriage coasts through an 1800-ft test sectionand decelerates to a velocity of 175 knots or lessbefore it intercepts five arresting cables that spanthe track at the end of the test section. Thearresting system brings the test carriage to a stop in600 ft or less. Essentially, any aircraft landing gearand tire can be mounted on the test carriage andvirtually any runway surface and weather conditioncan be duplicated on the track.

Impact Dynamics Research Facility

The Langley Impact Dynamics Research Facility isused to conduct crash testing of small full-scaleaircraft under controlled conditions. Using apendulum method, aircraft (maximum weight30,000 lbs) are suspended by cables from the 240feet high gantry and swung into impact surfaces--either soil or concrete. Free flight conditions areestablished when the swing cables arepyrotechnically separated from the vehicle just priorto impact. Flight path angles at impact may be variedfrom 0° to -60°. The maximum velocity obtainable,without rocket assistance, is approximately 88 ft/sec(60 mph). Instrumentation for these tests include

3

on-board cameras, strain gages, load cells,displacement transducers, and accelerometers.Transducers in the aircraft are hard-wired through along umbilical cable to the data acquisition room.

Thermal-Structures Laboratory

The Thermal-Structures Laboratory is used toconduct a broad range of research to characterizethe behavior of advanced thermal-structuressubjected to combined thermal and mechanicalloading conditions. The structures can be passivelyand/or actively cooled and range from innovativelightweight, durable thermal protection systems andcryogenic propellant tanks for reusable launchvehicles to actively cooled engine and stagnationregion structures for hypersonic air breathingvehicles. This facility contains one 22 kip, one 110kip, three 220 kip, and two 500 kip servo-hydraulicload machines. Thermal/mechanical load tests canbe conducted under the application of high thermalloads with a temperature range of -420°F to 2500°Fon specimens that are up to 4 ft by 8 ft in size.

Structural Dynamics Laboratory

The Structural Dynamics Laboratory consists of twofacilities designed for structural dynamics andpointing control research on aerospace structuresand components: 1.-The Dynamics Test andResearch Laboratory (DTRL) is a 5,200 square ft,80-ft-high building equipped with advancedsuspension devices. The photograph in figure 3shows a model of a spacecraft under dynamicstesting in the DTRL. This facility has a large dataacquisition system for acquiring dynamic data.2.-The Structural Dynamics Research Laboratory(SDRL) has a 38-ft-high vertical backstop and a12- by 12- by 95-ft tower. These facilities aresupported by dynamic test and signal processingequipment including 10-in.-stroke shakers, near-zero spring-rate suspension systems, an arc-second attitude measurement system, dataacquisition, a real-time control computer, videomonitoring, and environmental controls.

C o mbined L oads T esting S ystem - COLTS

The Combined Loads Test System (COLTS) isbeing developed at the Langley Research Centerand will be a unique structures research test facility.This facility will enable complex, combined loadstesting of large aerospace structures under

representative operating conditions. Typicalaerospace structures will include panels fromtransport aircraft fuselages, full fuselage barrels, andpanels from launch vehicles. The COLTS willconsist of two Pressure Box Test Machines and aCombined Loads Test Machine. One Pressure BoxTest Machine is currently operational (Figure 4a)and is used to apply pneumatic pressure (up to 20psig) to curved panel specimens to achieve a biaxialtension stress field at ambient conditions. Loadactuators apply additional longitudinal loads of up to450 kips to the specimen. Typical test specimensare 72 in. long and 63 in. wide, and have a radius of125 in. A second Pressure Box Test Machine willbe used to conduct biaxial tension testing atelevated and cryogenic conditions.

The Combined Loads Test Machine is underconstruction (anticipated startup date November1997). This test machine will be able to applycombined mechanical, pneumatic pressure, andthermal loadings to broad classes of aerospacestructures including panel and barrel specimens(Figure 4b). The Combined Loads Test Machine willhave a 2700-kip axial load, a 600-kip vertical shearload, a 6000-ft-kip torsion load, and a 20-psigpressure load capacity. Specimens may be testedat temperatures up to 400°F, and at cyclic, spectrumfatigue loading conditions. Typical test specimenswould include curved panels that are 120 in. longand 96 in. wide, and have a radius of 125 in. as wellas full shells that are as much as 45 ft long and 15 ftin diameter.

This paper will now present selected results ofsome recent studies in the areas of structuraldynamics and mechanics.

F/A-18E/F FLUTTER CLEARANCE

The Transonic Dynamics Tunnel plays a significantrole in providing flutter clearance data for newaircraft configurations. Tunnel tests performed onan aeroelastic model tested in a heavy gas can beused to predict the aeroelastic characteristics of thefull-scale vehicle flying in the atmosphere. Thisinformation can then be used to minimize the flutterrisk of new configurations, to provide data so thatfull-scale calculations can be performed with greaterconfidence, and to minimize the time required toperform airplane flutter clearance flights. Whenmilitary fighters are tested in the tunnel manydifferent store configurations can be cleared withrelative speed and safety.

4

Such an example is the F-18E/F flutter studiesrecently completed in the TDT. The flutterclearance study utilized a full-span 18-percent scalemodel. As shown in figure 5, the model can besting- or cable-mounted in the test section. Beforeflutter testing commenced, a rigid model was testedon the cable-mount system to assure flying stabilityin the tunnel. During tests in the TDT the followingaccomplishments were achieved; the flexiblevehicle components were flutter cleared throughM=1.2 on the sting mount, the flexible wing andfuselage configuration were cleared for flutter onthe cable-mount system, numerous storeconfigurations were flutter cleared on the cable-mount system, the stability of all-moveablestabilators with mil-spec freeplay was verified, andthe stability of the model with several failure modeswas determined. Limited parametric studies werealso performed to determine the effect of stabilatorfree play, wing and fuselage fuel, wing-tip-and wing-pylon-mounted stores/tanks, and control surfacerestraint springs on flutter.

PIEZOELECTRIC AEROELASTICRESPONSE TAILORING INVESTIGATION

The Piezoelectric Aeroelastic Response TailoringInvestigation (PARTI) was the first study in whichpiezoelectric materials were chosen to control theaeroelastic response of a relatively large, aeroelasticmodel. Piezoelectric materials possess the ability todevelop a mechanical strain when subjected to anelectrical charge. Therefore, piezoelectric materialscan be used as actuators to control aeroelasticmotion. The relationship between an appliedelectric field and the corresponding behavior of apiezoelectric actuator is well documented in [1, 2,3]. The conventional configuration for an in-planedisplacement piezoelectric actuator consists of asingle piezoelectric wafer sandwiched between twoelectrodes. Increased in-plane actuation can beobtained by grouping multiple wafers into multiplelayers.

The model, shown in figure 6a, is a five-foot long,high-aspect-ratio semi-span wing designed to flutterin the TDT. The model is comprised of an exteriorfiberglass shell to provide the proper aerodynamiccontour and an interior composite plate as the mainload carrying structure. A sketch of the majorcomponents of the PARTI wing are shown in figure6b. Piezoelectric actuator patches were attached tothe upper and lower surfaces of the compositeplate. Fifteen groups of piezoelectric actuator

patches covered the inboard 60% of the span. Dueto the ply orientation of the material used in thecomposite plate and the wing sweep, the actuatorswere able to affect both the bending and torsionalresponse of the model. Ten strain gauges and fouraccelerometers were used as sensors to providefeedback signals to the piezoelectric actuators. Themodel is also equipped with wing-tip flutter-stopperand a trailing-edge control surface. The flutter-stopper was used as a safety device during wind-tunnel testing.

The purpose of the study was to demonstrate anincrease in flutter dynamic pressure and a reductionin subcritical response by using piezoelectricactuators. Experimental open-loop fluttercharacteristics and response time histories belowthe flutter boundary as a function of eachpiezoelectric actuator group were first determined inthe TDT. These results were then used to designcontrol laws to suppress flutter and reduce theaeroelastic response. Twenty-eight control lawswere designed and tested. Control laws weredesigned using both single-input/single-output(SISO) and multi-input/multi-output (MIMO)methods that utilized up to five inputs and nineoutputs. Each control law varied in designtechnique, actuator and sensor choices, andcomplexity of the controller. The most successfulcontrol law demonstrated a 12% increase in flutterdynamic pressure and reduced the power spectraldensity of peak response due to tunnel turbulenceat subcritical speeds by 75%. These experimentalresults are shown in figure 7.

The PARTI program successfully demonstrated thecontrol of aeroelastic response using piezoelectricactuators on a large aeroelastic model tested in theTDT. Results of this investigation are fullydocumented in [4, 5, 6].

BENCHMARK ACTIVE CONTROLSTECHNOLOGY

The Benchmark Active Controls Technology(BACT) model is one of a series of five wind-tunnelmodels developed for the Benchmark ModelsProgram (BMP). The original goal of the BMP wasto obtain experimental data for validating unsteadyCFD codes. An example of the type of dataacquired in this program is presented in [7]. TheBMP uses highly instrumented rigid models thatare tested in the TDT on a flexible sidewall mountknown as the Pitch and Plunge Apparatus or

5

“PAPA”. It provides the two degrees of freedomthat are required for classical flutter [8]. Unsteadypressure distributions can then be obtained duringsustained model oscillations at flutter onset andcan compared with analytical predictions.

The objectives of the current BACT model testsare to: obtain high quality data to validatecomputational-fluid-dynamics and computational-aeroelasticity codes; to verify the accuracy ofcurrent aeroservoelastic design and analysis tools;and to provide an active controls testbed forevaluating new and innovative controlmethodologies. Some early results for this BACTmodel are presented in [9]. The model has arectangular planform with an NACA 0012 airfoilsection and is equipped with a trailing-edge controlsurface and a pair of independently actuatedupper and lower-surface spoilers. All surfaces aremoved with independent miniature hydraulicactuators. A photograph of the model on the“PAPA” mount system is shown in figure 8a and aview of the model mounted in the wind tunnel isshown in figure 8b. Instrumentation includespressure transducers and accelerometers on themodel, and strain gages on the mount-system.

During a recent BACT wind-tunnel test entry theprimary objective was to investigate a variety ofcontrol algorithms, designed using variousmethods, to suppress flutter and alleviate gustloads. The plot in figure 9a presents theperformance of three semi-adaptive flutter-suppression control laws. The solid line is theopen-loop flutter boundary of the BACT model.The circle symbols correspond to the points wherecontrol laws were tested. The GPC control lawused a Generalized Predictive Control algorithmand employed an analytical representation of theplant to predict future model responses and selectcontrol surface commands to minimize thatresponse. The Inverse Control used linear neuralnetworks to learn the plant inverse and employedexperimental data. The NPC system used a NeuralPredictive Control (NPC) algorithm. All control lawsused only the trailing-edge control surface. Asindicated in the figure, all three semi-adaptivesystems were very successful in suppressingflutter. Figure 9b presents open- and closed-loopmodel responses due to flow oscillationsproduced by the TDT flow oscillator system.Oscillation frequencies ranged from zero to 5.5 Hz.The NPC control law was a gust load alleviation

design and reduced acceleration responses by upto 80%.

ACTIVELY CONTROLLED RESPONSE OFBUFFET AFFECTED TAILS

Buffeting is an aeroelastic phenomenon whichplagues high performance aircraft, especially thosewith twin-vertical-tails. For aircraft of this type at highangles of attack, vortices emanating fromwing/fuselage leading-edge extensions burst,immersing the vertical tails in their wake. As shownin figure 10, for an F/A-18 undergoing high angle-of-attack tests at the NASA Dryden Flight ResearchCenter, vortices emanating from the wing/fuselageleading-edge extensions burst and immerse thevertical tail in their wake. The resulting buffetingloads on the vertical tails are a concern from afatigue standpoint. For example, for the F/A-18aircraft, special and costly 200-flight-hourinspections are required to check for structuraldamage due to buffet loads. Buffeting loadalleviation through the use of active controls is apromising solution to this problem. The researchobjective of the current work is to apply activecontrols technology, using a variety of forceproducers, to perform buffeting load alleviation on atwin-vertical-tail wind-tunnel model.

A 1/6-size, rigid, full-span model of the F/A-18 A/Baircraft was tested in the TDT. The model, shown infigure 11 mounted on a sting in the tunnel, wastested with flexible and rigid tail surfaces. Threeflexible tails were built to test different controlconcepts. The flexible tails were instrumented witha root strain gage aligned to measure bendingmoment and with two tip accelerometers. Each tailwas equipped with a different concept for buffetalleviation: the first was equipped with an activerudder; the second could be equipped with eitheran active tip vane or an active embedded slottedcylinder; and the third was equipped with activepiezoelectric actuation devices. Of the differentconcepts, early open-loop tests in the tunnelindicated that the rudder and the piezoelectricactuators appeared to be the most promisingcandidates. A photograph of the piezoelectricactuator on the flexible tail is shown in figure 12.

Two single-input/single-output control laws wereimplemented on the model. One control law usedthe rudder and the other used the piezoelectricactuator. Both control laws used a vertical tailleading-edge tip accelerometer as the sensor. The

6

control laws added damping to the system byproviding a 90° system phase lag between theadded force and the accelerometer response. Theresults are presented in figure 13 in terms of rootbending moment versus angle-of-attack.Reductions in root bending moment as much as60% at certain angles-of-attack are evident. Resultsof this investigation are to be published in [10]. Theresults of this wind-tunnel test illustrate thatbuffeting alleviation of the vertical tails can beaccomplished by using the rudder or piezoelectricactuators as active controls.

TILTROTOR WING VIBRATORY LOADSREDUCTION USING ACTIVE

SWASHPLATE AND FLAPERON

The fundamental vibration problem in tiltrotor aircraftis caused by blade passage in front of the wingwhile in high-speed airplane-mode flight. Wingcirculation creates an azimuthally unsymmetric flowthrough the rotor system which is the primarycontributor to fixed system (pylon, wing, andfuselage) vibrations. The Wing and RotorAeroelastic Testing System (WRATS) tiltrotor modelis a semispan testbed developed from a V-22 1/5-scale aeroelastic tiltrotor model. It was designedand fabricated by Bell Helicopter Textron, Inc.(BHTI). In an effort to control vibrations in the fixedsystem, BHTI developed a system called theMultipoint Adaptive Vibration Suppression System(MAVSS). The objective of the current research isto evaluate the ability of the MAVSS system tocontrol multiple modes of fixed-system vibrations onthe WRATS model during tests in the TDT.

A photograph of the WRATS model mounted in theTDT is shown in the figure 14. An active controlsystem using three high-frequency hydraulicactuators to tilt the swashplate and an activeflaperon was fabricated and installed on the WRATSmodel. The actuators were driven by a signalproduced by the MAVSS system at frequencies upto 50 Hz. The basic test procedure was to identifyflight conditions of high vibration, activate MAVSS,and compare the resulting loads. The MAVSSsystem operates in the following manner: It obtainsfeedback signals from response sensors (beam,chord, and torsion strain gage bridges); quantifiesmodel vibration levels in an objective function;identifies the system using a series of test signals;computes and then applies commands to the activeswashplate/flaperon to lower the objective function.If the optimized vibration level rises above a given

threshold, the controller will automatically reactivateitself.

The bar chart shown in figure 15 contains resultsfrom the wind-tunnel test and illustrates the successof the MAVSS system in controlling vibratory loadsin three wing modes simultaneously. Each set ofthree vertical bars grouped together indicates thethree-per-rev (3P) wing beam, chord, and torsionloads at one instant of time. For each of the fourairspeeds there are a set of bars shown with theMAVSS system both off and on. The plot shows atrend of increasing baseline 3P vibration level in allthree wing modes with airspeed, but, moreimportantly, also shows significant reductions (89%to 99%) in all 3P vibratory wing loads at eachairspeed. Although not shown on the figure, theswashplate and flaperon motions required toaccomplish these reductions are within acceptablelimits. Results of this investigation will be fullydocumented in [11]. This test has confirmed that anactive control system is a viable candidate foralleviating multiple modes of tiltrotor vibration.

PHOTOGRAMMETIC APPENDAGESTRUCTURAL DYNAMICS EXPERIMENT

The Photogrammetric Appendage StructuralDynamics Experiment (PASDE) was developed todemonstrate the use of photogrammetrictechniques for structural dynamic responsemeasurements of spacecraft solar arrays and similarstructures. Development and demonstration ofpassive, on-orbit structural response measurementmethods will increase the amount of spacecraftengineering data available. The availability of low-cost, on-orbit engineering data for the InternationalSpace Station is essential for mathematical modeland design load verification and subsequentdetermination of proper operational procedures andconstraints.

A photogrammetric structural dynamic responsemeasurement instrument was designed, fabricated,assembled, and tested to meet a flight experimentopportunity on a NASA Shuttle/Mir mission (STS-74) in November 1995. The instrument consistedof six video cameras with 50mm lens and motorizedirises, six video tape recorders, video time inserters,and interface electronics. The instrument waspackaged in three standard Shuttle canisters. Amission plan was developed to obtain video imagedata during STS-74 mission events consideredlikely to result in structural response motion of the

7

Mir Kvant-II module lower solar array as shown infigure 16.

The PASDE mission was implemented from aControl Center at the Goddard Space Flight Centerduring the mission. Video data was collected on thePASDE video recorders during the followingevents: docking of the Shuttle with Mir, three setsof specific Shuttle jet firing sequences designed toexcite structural motion of the combined Shuttle/Mirspacecraft, transitions of the combined spacecraftfrom night-to-day and day-to-night, and sun trackingmovements of the solar array. The video data wasretrieved from the instrument following the mission,and digitized into standard computer image format.From the digitized data, displacement time historiesat points in the video images were computed. Thetime history data was then triangulated using theknown geometry of the instruments in the Shuttlepayload bay and the coordinate systems of theShuttle and Mir to obtain high resolution three-axismotions of multiple points on the solar array. A timehistory of the normal displacement at the pointindicated by the white circle in figure 16 is shown inthe graph. A description of the experiment ispresented in [12].

The PASDE experiment demonstrated the use ofpassive photogrammetric techniques to make highresolution structural response measurements ofsolar arrays and other spacecraft appendages on-orbit. The International Space Station has adoptedthis technique for on-orbit measurement of solararray response.

INTERFACE TECHNOLOGY

Detailed analysis of complex aircraft structures canseverely tax today’s computing environment.Therefore, it is highly desirable to use detailedmodeling only when necessary. Embedding localrefinement in a single model of the entire structuremay lead to highly complex modeling due to the useof transition modeling between highly refinedregions and regions with less refinement.Additionally, transition modeling typically introducesdistorted elements into the finite element modelwhich may adversely affect the accuracy of thesolution.

Interface technology was developed [13] that allowsthe independent modeling of differentsubstructures or components without the concernfor one-to-one nodal coincidence between the

finite element models. The interface element actsas a “glue” between independent finite elementmodels with different mesh densities and nodallayouts. Interface technology provides a local/global model which is fully coupled, hasdisplacement compatibility, and captures changesin load and load path. Interface technologyprovides the analyst with increased modelingflexibility. Since the grid points along the commonsubstructure boundaries need not coincide, theneed for potentially complex transition modeling iseliminated. Different levels of approximations maybe used in each of the substructures allowing theuse of substructures only where needed.Examples of how this technology can be used arepresented in figure 17. An example of interfacetechnology applied to a built up fuselage panel ispresented in figure 18. This technology provides ameans of rapidly assembling diverse structuralmodels subject to mechanical, thermal or dynamicloads. These models can come from differentsources or from previous designs where similarcomponents were used. The engineer working onpreliminary vehicle design could create structuralmodels of unparalleled accuracy by combiningcomponents from a “library” of models. Structuralmodeling assembly and design could become a“plug and play” operation.

CRASHWORTHINESS

The objectives of crashworthiness research are todevelop a fundamental understanding of theresponse of composite structures to impact loads,and to apply this information for improved crash-worthiness designs [14]. The crashworthinessprogram has four main elements: full-scale crashtesting, crash analysis, scale model testing, and thedevelopment of innovative concepts for improvedcrashworthiness. A recent accomplishment in thisprogram was the full-scale crash test of an all-composite Lear Fan 2100 aircraft. NASA acquiredtwo prototype aircraft after the company endedproduction. The aircraft was never put intoproduction.

The Lear Fan aircraft is fabricated of graphite epoxycomposite material with construction that consists ofsemi-circular frames which are bonded and rivetedto the skin. The subfloor of the aircraft contains fourlongitudinal aluminum beams which were used forgrounding and lightning protection. The presenceof the aluminum subfloor beams can cause theimpact response of this composite aircraft to be

8

similar to those of previously tested metallic generalaviation aircraft. Therefore, the first Lear Fan aircraftwas tested with the aluminum beams in place. Forthe second Lear Fan aircraft, the aluminumsubfloors will be removed and an energy absorbingcomposite subfloor will be installed. Both aircraft areto be tested under identical impact conditions of 82ft/sec horizontal velocity, 31 ft/sec vertical velocity,and a flat impact attitude. On board the aircraft wereside-by-side seat tests with one anthropomorphicdummy seated in a standard GA aircraft seat (pilot),and a second dummy seated in an energyabsorbing seat (co-pilot).

A photograph of the aircraft undergoing flighttesting is shown in figure 19. A post-testphotograph of the aircraft after crash testing is alsoshown in the figure. Post-test damage assessmentindicated a large fracture in the top of the fuselage,several smaller fractures emanating from the doorsand windows, and all the composite frames werefractured close to the impact point. No damage wasobserved in any of the aluminum floor beams.Measured acceleration levels at the floor locationwere in the range of 160-200 g’s. As shown infigure 19, the energy absorbing seats offered moreprotection to the occupant than did the standardseat. For example, the co-pilot dummy only slightlyexceeded the FAR 23 requirement of no more than1500 lb load in the lumbar region, whereas the pilotdummy seated in a standard seat exceeded thelevel by a factor of two. A comprehensivedevelopment program is being conducted todesign a composite energy absorbing floor beamfor retrofit on the second Lear Fan aircraft. Light-weight, cost-effective composite floor beamconcepts are being developed to improve theenergy absorbing capability of the airframe. Staticand dynamic tests are being conducted on thesestructural concepts to determine their responsecharacteristics when subjected to crash loads and toevaluate their energy absorbing capability.

Several composite fuselage subfloor-beamconfigurations have been evaluated for their staticand dynamic response characteristics, their energyabsorption capability, their ability to controltransmitted loads, and their structural integrity aftercrushing. A foam-filled composite floor-beamconcept has been identified that has enoughenergy-absorption capability to limit to anacceptable level the high-acceleration crash loadsthat are transmitted to a passenger seat during acrash event. Test results shown in figure 20

indicate that a relatively simple flat-sided box-beamconfiguration with a foam core that is integrated withan aircraft seat rail can satisfy the energy absorption,load transmission level, and structural integrity goalsfor crashworthiness. A desired, relatively constant,collapse load, designated as the Sustained CrushLoad (SCL), of 240 lbf/in. was achieved with theconcept as shown in the figure. A SCL value of 200 - 300 lbf/in. is generally required for crash-worthiness. The load attenuation characteristics ofthe test specimen are excellent as shown in figure20. Approximately 230 g’s of acceleration force wasimposed on the test specimen and a response ofapproximately 25 g’s was recorded.

The results of the tests with the foam-filledcomposite floor-beam concept indicate that it ispossible to design energy-absorbing crashworthycomposite structures. The results of these testswill help designers of future aircraft fuselagestructures develop designs with improved energy-absorbing subfloors.

COMPOSITE STRUCTURES

For three decades, NASA has worked in closepartnership with the U.S. aircraft industry to developthe materials, the structures, and the essentialscience that provides the means to fully exploit theuse of composites in aerospace vehicles.Throughout this period, NASA has supported thedevelopment of materials synthesis, structuralanalysis methods, fracture mechanics, and testprocedures. Building upon this essential sciencefoundation, the aircraft industry has systematicallyexplored the application of composite aircraftstructures. NASA and industry began in the 70’s todevelop lightly loaded aircraft components such asspoilers for the Boeing 737 and upper aft rudder forthe Douglas DC-10. A significant number of thesestructures were built, ground tested, and subjectedto long-term flight tests. This work establishedindustry confidence in the performance andenvironmental durability of composites for aircraftuse. In the decade of the 80’s, NASA and industryfocused their development efforts on medium-loaded primary structures that included thehorizontal stabilizer for a Boeing 737 and the verticalstabilizer for the Douglas DC-10. These primarycomponents received FAA certification and are stillin flight service.

NASA studies in the mid 80’s established that thefull potential of composites in aircraft could only be

9

achieved with the development of composite wingand fuselage structure. The major barriers wererecognized to be damage tolerance and cost. In1989 NASA launched its Advanced CompositesTechnology (ACT) Program aimed specifically atdeveloping cost-effective composite primarystructures for commercial transports. While thiswork has not reached the technology readinesslevel required for industry to produce a new aircraftwith a composite wing or fuselage, significantprogress has been made. This progress includes:(1) a basic understanding of compression strengthafter impact damage, (2) new materials andapproaches for increasing compression strengthafter impact damage, (3) materials data base, (4)tests methods and analysis, (5) repair approaches,and (6) exploring new fabrication methods toreduce costs.

One of the primary goals of the ACT program is todevelop the enabling technology that will allowcomposite materials to be used in the primary wingand fuselage structures of the next generation ofadvanced subsonic transport aircraft. Compositestructures offers the greatest potential for reducingthe direct operating cost (>10%), reducing weight(up to 40%), and the elimination of corrosion andfatigue issues. The following sections will describeresearch activities that are being pursued in bothwing and fuselage technology.

Wing Stub Box

To evaluate the potential of a stitched graphite-epoxy material for use on commercial transportaircraft wings, a section of a wing box was designedand fabricated by the McDonnell DouglasAerospace Company under the NASA ACTprogram. The wing stub box represents the inboardportion of a high-aspect-ratio wing box for a civiltransport aircraft. The wing box was fabricated usingan innovative manufacturing process that haspotential for reducing manufacturing cost andproducing damage tolerant composite primaryaircraft structure. The objectives of the tests wereto evaluate the behavior of a wing box structure andto verify analysis methods for predicting thestructural response of the wing box [15 and 16].The wing box was designed to simulate a section ofa commercial transport wing and was subjected tobending loads.

As shown in figure 21 the wing stub box testspecimen consists of a metallic load transition

structure at the wing-root, a composite wing stubbox, and a metallic extension structure at the wing-tip. The load transition structure and the wing-tipextension structure are metallic end fixturesrequired for appropriate load introduction into thecomposite wing stub box during the test.

Layout of the composite stub box is shown in figure22. The stub box consists of ribs, spars, and upperand lower cover panels (each of which has stringersand intercostals stitched to the skin). The skin ofthe upper and lower cover panels range inthickness from about 0.29 to 0.90 inches. Theupper cover panel has 10 stringers along the lengthof the wing box and the lower cover panel has 11stringers. The upper cover panel has an accessdoor cutout. At the stringer runout locations, thestringer is terminated and the tapered stringer webprovides a mechanism for smoothly transferring theload from the stringer to the skin. At the runout,fasteners were installed to prevent skin stiffenerdebonding at these locations. The ribs and sparswere stiffened with blade stiffeners to preventbuckling. The ribs were connected to the coverpanels at the intercostals.

A series of structural tests were conducted byloading the wing in bending with no damage, withdetectable damage, with nondetectable damage,and with a repair. Damage was inflicted to the uppercover panel by using a dropped-weight impactor.Strains at 254 locations and displacements at 15locations on the structure were recorded duringeach test. A photograph of the wing stub box priorto testing is shown in figure 23.

No failures and no damage growth occurred inpreliminary tests of the undamaged structure withthe applied load up to Design Limit Load (DLL). Thestructure satisfied the requirement of supportingDLL with detectable damage caused by dropped-weight impact condition imposed on the wing box.The damaged region was repaired with a simplealuminum bolt-on patch by American Airlinesmaintenance personnel.

In the final test the structure supported 140% ofDLL prior to failure through a stiffener runout regionwith nondetectable impact damage. Twoindependent failures occurred in the stub box. Thefirst failure occurred in an unsupported region at therunout of a stringer where it terminates at a rib (in thesame bay as the access panel cutout). The stringerhad extensive damage including delaminations

10

between the skin and flange, broken stitches, boltspulled through the flange, and cracking of theblade. This failure involved large out-of-planedeformations in the upper cover skin bay outboardof the access door. The second failure wascatastrophic and initiated at a nondetectable impactdamage sight. The failure occurred across theentire width of the upper cover panel and into bothspars. The maximum load carried by the structurewas 154 kips (93% of design ultimate load). A viewof the failure is shown in figure 24.

The other objective of the test was to verifyanalytical methods for predicting the structuralresponse of the wing box. An initial finite elementmodel [17] accurately predicted the global behaviorof the stub box but did not accurately predict thebehavior at the center portion of the upper coverpanel outboard of the access door or at the splicejoint between the composite stub box and themetallic extension box. A refined model wascreated to improve the accuracy of the analyticalpredictions. A comparison of strain responsebetween the original and refined analyses andexperiment is shown in figure 25.

The tests verified the ability of the wing boxstructure to satisfy most of the designrequirements. The global behavior of the structureis in good agreement with the analytical predictionsand the displacements and strains predicted by ageometrically nonlinear analysis using a refinedfinite element model compare very favorably withexperiment.

Fuselage Structures

As part of the ACT program, the Boeing CommercialAircraft Group has been working to develop costeffective and structurally efficient compositefuselage structure. The focus of this work has beenon the fuselage section just aft of the main landinggear wheel well of a modern wide body transport asshown in figure 26. This fuselage section is 33 feetlong and 20 feet in diameter and contains crown,side, and keel quadrant sections as shown in thefigure. Sandwich structure is being considered inthe design of side and keel quadrant sectionsbecause it has the potential for high structuralefficiency and low cost manufacturing.

The application of sandwich structure has beenrestricted in the past due to undesirable moistureabsorption and retention, and due to an insufficient

understanding of low-speed impact damagemechanisms and the effect of such damage, as wellas penetration damage, on the structuralperformance of sandwich structures.Understanding these issues are important ifcomposite sandwich concepts are to be acceptedfor primary structures. A joint NASA/Boeing studyof the technology issues associated with the use ofcomposite face sheet sandwich construction in theside and keel panels is being conducted as part ofthe ACT program [18 and 19].

Keel Panel

A composite sandwich fuselage keel test panel wasfabricated by the Boeing Commercial AirplaneGroup, and is representative of a highly loadedfuselage keel structure (figure 27). The test panelwas machined from a larger demonstration panelthat was fabricated to gain experience with towplaced composite structures with dropped plies.The dropped plies result from the reduction incompression loads as the structure moves aft in thekeel section. The purpose of these studies is tounderstand the load distribution in thick-face-sheetcomposite honeycomb-sandwich structures withand without impact damage, and to understand theload distribution and panel failure mechanisms inthe presence of both impact damage and discrete-source damage. Compression tests of the panelwith three different conditions were conducted:undamaged; barely visible impact damage (BVID) intwo locations; and with BVID in two locations and anotch through both face sheets. The impact-energy level necessary to inflict BVID on the panelwas determined from an impact-damage screeningstudy conducted on another panel of the samedesign. BVID was assumed to have occurred whenthe residual dent depth on the facesheet at theimpact location was equal to or greater than .05 in.or the impact energy was greater than 100 ft-lbs.Finite element analyses of the undamaged paneland the notched panel were also performed.

The tests of the impact damaged and notchedpanel identified the notch as being the most criticalof the three damage sites. Analytical resultscompared well with the experimental results. Thenotched panel failed at 202 kips, which is thedesign ultimate load for the panel. The failure modewas compression failure of the face sheets at thenotch location. A photograph of the failed panel isshown in figure 28. Since the panel supported

11

design ultimate load with BVID, the design isdamage tolerant for nonvisible impact damage.

An important finding from the impact-damagescreening study [18] was that significant internaldamage occurred at relatively low impact-energylevels and that the corresponding surface damageat the impact sites as measured by the residual dentdepth were very small, making it nonvisible to aground-crew inspection. This internal damage atthe impact locations can significantly reduce theresidual strength of the panel. These resultssuggest that the present approach of using residualdent depth as a means of assessing the effect ofBVID on the strength degradation of a compositehoneycomb-sandwich structure needs to be re-evaluated.

To identify a more suitable criterion for assessingthe effects of BVID on the strength degradation ofthick-face-sheet composite honeycomb-sandwichstructure, compression-after-impact residual-strength studies were performed on specimens thathave been impacted with a wide range of damage-inducing impact-energy levels [19].

For this study, 5 inch wide by 10 inch longspecimens were machined from the original keeldemonstration panel. These specimens weretested in compression in both the undamaged anddamaged conditions. A typical test set up in a 300kip testing machine is shown in figure 29. Impactdamage was generated using a dropped weightimpact apparatus. Impact energies ranged from 40ft-lbs to 100 ft-lbs. The damage area was measuredfrom ultrasonic C-scan images of the damage site.Impact screening tests that were performed earlierindicated that significant internal damage can occurfor impact energies significantly lower than 100 ft-lbs even when the residual depth dent is muchsmaller than .05 in.

Results for seven specimens are presented infigure 30. As shown in the figure global stiffness(except for specimen 6) of the specimens is notaffected by the impact damage. However, as shownin figure 31, failure loads as high as 40% lower thanthe undamaged specimen were experienced bypanels with impact damage as low as 40 ft-lbs andwith residual dent depth less than .01 in.Appreciable reductions in compression strengthoccurred for all specimens even for conditionswhere the impact damage would be considerednon-visible. A typical failure is shown in figure 32.

As shown in the figure, a compression failureoccurred in the facesheet that was impact damaged.Following this failure, the sandwich specimenexperienced significant bending and the remainingfacesheet failed in bending. Results from ultrasonicC-scan inspections of the impacted specimensindicated that large areas of internal damage werecaused by the impact. It appears as a result of thesestudies that further investigation is required toestablish a criteria for the affect of impact damage oncomposite sandwich structures.

Crown Panel

Another study being performed in support of theACT program is shown in figure 33. Shown in figure33 are the results of an all composite crown paneltested in the pressure box test machine. Theobjective of this test was to evaluate theperformance of a stiffened composite fuselagecrown panel fabricated using cost-effectivemanufacturing techniques and subjected to internalpressure and axial loads.

A fuselage crown test panel shown in the figure wasfabricated by the Boeing Commercial AirplaneGroup, and is representative of a fuselage crownstructure designed for internal pressure and highaxial tension loads. The skin of the panel was tow-placed from a graphite-epoxy material system. Theframes were fabricated using a braided fiber preformcured by the resin transfer molding process andsecondarily bonded to the skin. The stringers werefabricated from graphite-epoxy tape and cocurredwith the skin. There are no shear clips that connectthe stringers to the frames at the frame-stringerintersections or "mouse-hole" regions. Thefuselage crown panel was subjected to internalpressure and axial loads in the pressure-box testmachine. A finite element analysis was conductedand the results of the analysis were correlated withtest results.

The panel was loaded to 4,000 lb/in. of axial loadand 11 psi of internal pressure prior to failure. Theeffect of combined internal pressure and axial loadon the local bending gradients was studied. Theeffect of load eccentricities on panel response wasstudied analytically. The effect of combined internalpressure and axial load on local bending gradientswas determined. The failed panel is shown in thefigure.

12

The composite fuselage crown panel structuralresponse was not influenced by axial loadingeccentricities. Combinations of axial load andpressure influenced the bending stresses at theframe-stringer intersection or mouse hole region.The test results verified the structural integrity ofthis advanced design concept.

SUMMARY

This paper has presented the results of somerecently completed research programs in structuresat the NASA Langley Research Center. The resultsthat have been presented indicate the wide rangeof research being conducted. NASA will continueto conduct research in this area in an effort to fullyunderstand and predict structural response ofaerospace vehicles so that future designs can fullyexploit these technology advances.

ACKNOWLEDGMENT

The author wishes to acknowledge the work of themembers of the Structures Division at NASALangley Research Center whose research is beingoverviewed in this paper.

REFERENCES

1. Weisshaar, T. A.: Aeroservoelastic ControlConcepts with Active Materials. ASMEInternational Mechanical EngineeringCongress Exposition, Special Symposium onAeroelasticity and Fluid/Structure InteractionProblems. Proceedings of the 1994 ASMEWinter Meeting, November 1994.

2. Heeg, J.: Analytical and ExperimentalInvestigation of Flutter Suppression byPiezoelectric Actuation. NASA TP-3241,March 1993.

3. Crawley, Edward F.; and Anderson, Eric H.:Detailed Models of Piezoceramic Actuation ofBeams. Journal of Intelligent MaterialSystems and Structures, Vol. 1, pp. 4-25,January 1990.

4. Heeg, J.; McGowan, A-M. R.; Crawley, E. F.;and Lin, C. Y.: The Piezoelectric AeroelasticResponse Tailoring Investigation: Analysisand Open-Loop Testing. Proceedings of theCEAS International Forum on Aeroelasticity

and Structural Dynamics, Manchester, UK,June 1995.

5. Lin, C. Y.; Crawley, E. F.; and Heeg, J.:Open-Loop and Preliminary Closed-LoopResults of a Strain Actuated ActiveAeroelastic Wing. 36th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics,and Materials Conference, New Orleans, LA.AIAA Paper No. 95-1386, April 10-13, 1995.

6. McGowan, Anna-Maria Rivas; Heeg, Jennifer;and Lake, Renee C.: Results From Wind-Tunnel Testing from the PiezoelectricAeroelastic Response Tailoring Investigation.37th AIAA/ASME/ASCE/AHS/ASCStructures, Structural Dynamics, and MaterialsConference, Salt Lake City, Utah. AIAAPaper No. 96-1511, April 15-17, 1996.

7. Rivera, Jose A., Jr.; Dansberry, Bryan E.;Farmer, Moses G.; Eckstrom, Clinton V.;Seidel, David A.; and Bennett, Robert M.:Experimental Flutter Results With Steady andUnsteady Pressure Measurements of a RigidWing on a Flexible Mount System. AIAA 91-1010, April 1991.

8. Farmer, M. G.: A Two-Degrees-of-FreedomFlutter Mount System with Low Damping forTesting Rigid Wings at Different Angles ofAttack. NASA TM-83302, 1982.

9. Scott, R. C.; Wieseman, C. D.; Hoadley, S. T.;and Durham, M. H.: Pressure and LoadsMeasurements on the Benchmark ActiveControls Technology Model. AIAA 35thAerospace Sciences Meeting and Exhibit,January 1997.

10. Moses, R. W.: Active Vertical Tail BuffetingAlleviation on a Twin-Tail FighterConfiguration in a Wind Tunnel. Proceedingsof the CEAS International Forum onAeroelasticity and Structural Dynamics, June1997.

11. Nixon, M. W.; Kvaternik, R. G.; and Settle, T.B.: Tiltrotor Vibration Reduction ThroughHigher Harmonic Control. AmericanHelicopter Society Forum 53, Virginia Beach,Virginia, April 29-May 1,1997.

13

12. Gilbert, M. G.; and Welch, S. S.: STS-74/MIRPhotogrammetric Appendage StructuralDynamics Experiment. 37th AIAA/ASME/ASCE/AHS/ASC Structures, StructuralDynamics, and Materials Conference, SaltLake City, Utah. AIAA Paper No. 96-1493,April 1996.

13. Ransom, J. B.; McCleary, S. L.; andAminpour, M. A.: A New Interface Element forConnecting Independently ModeledSubstructures. AIAA Paper 93-1503, 1993.

14. Jones, L. E.; and Carden, H. D.: CompositeAircraft Crash Testing. AerospaceEngineering, December 1995.

15. Jegley, D. C.; and Bush, H. G.: TestDocumentation and Results of the StructuralTests on the All-Composite McDonnellDouglas Wing Stub Box. NASA TM-110204,1996.

16. Wang, J. T.; Jegley, D. C.; Bush, H. G.; andHinrichs, S. C.: Correlation of StructuralAnalysis and Test Results for the McDonnellDouglas Stitched/RFI All-Composite WingStub Box. NASA TM-110267, July 1996.

17. Wang, J. T.: Global and Local StressAnalyses of McDonnell Douglas Stitched /RFIComposite Wing Stub Box. NASA TM-110171, March 1996.

18. McGowan, D. M.; and Ambur, D. R.:Compression Response of a SandwichFuselage Keel Panel With and WithoutDamage. NASA TM-110302, February1997.

19. McGowan, D. M.; and Ambur, D. R.: Damage-Tolerance Characteristics of CompositeFuselage Structures With Thick Facesheets.NASA TM-110303, February 1997.

AeroelasticityComputational Structures

Structural Dynamcis Thermal Structures

Structural Mechanics

Figure 1. Structures Division research areas.

Aeroelasticity Branch

• Experimental Aeroelasticity• Aeroservoelastic Analysis• CFD Methods Application

Thermal Structures Branch

• Thermal Protection Systems• Hot Structures• Cryogenic Tanks

Structural DynamicsBranch

• Landing Dynamics• Spacecraft Dynamics• Control of Dynamic Response

Structural Mechanics Branch

• Composite Structures• Structural Concepts• Crashworthiness

Computational StructuresBranch

• Advanced Analysis• Multidisciplinary Optimization• Methods for Next Generation

Computers

STRUCTURESDIVISION

Figure 2. Structures Division research charters.

Transonic Dynamics Tunnel

Structural Mechanics Laboratory

Impact Dynamics Research Facility

Structural Dynamics Laboratory

Thermal Structures Laboratory

Aircraft Landing Dynamics Facility

Figure 3. Structures Division research facilities.

D-Box Test Fixture

End platen

Hinge fittingI-beams

Support beam assembly

Reacting platen support

assembly

Reacting platen assembly

Loading platen assembly 15 ft diameter

45 ft long

Test panel

32 ft

47 ft 72 ft

a) Pressure box test machine. b) Combined loads test machine.

Figure 4. Combined loads testing system.

Stores flutter clearance

• Flexible wings & stores • Cable mounted

• Flexible wings & fuselage • Cable mounted

Vehicle flutter clearance

• Rigid model • Cable mounted

Flying-stability verification Components flutter clearance

• Flexible components • Sting mounted

Figure 5. F/A-18 E/F flutter clearance program.

��������������������

4

7

1013

151

7

3

4

82

4

5

3

10

9

1

2

6

15

Aerodynamic control surface

15

14

14

14

13

1312

12

1111

10

109

9

88

7

7

66

5

55

4

3

3

2

2

11

1/4 chord swept 30°Flutter stopper tip mass ensemble

Aerodynamic shell section

Piezoelectric actuator (with group number)

Strain gage

Accelerometer

a) Model mounted in the TDT. b) Sketch of major components.

Figure 6. Piezoelectric aeroelastic response tailoring investigation.

Open Loop

Closed Loop

Mode 1 Mode 2

0

5

10

15

Dynamic Pressure (psf)50 60 70 80 90

x x

Nat

ural

freq

uenc

y (

Hz)

Frequency (Hz)0 5 10 15 20 2510

10

10

10

10

10

-1

0

1

2

3

4 Open Loop

Closed Loop

PS

D

((µε

) /H

z)2

Strain response at 60 psf

a) Flutter results. b) Subcritical response results.

Figure 7. Experimental results.

a) Model on flexible mount.

b) Model installed in the TDT.

Figure 8. Benchmark active controls model.

200Open-loop unstable

Open-loop stable

NPC Inverse control GPC

180

160

Dyn

amic

pre

ssur

e,

psf

140

1200.6 0.7 0.8

Mach number0.9 1.0

a) Flutter suppression results.

Open-loop Closed-loop

Mach number = 0.77 Dyn. pressure = 125 psf

Acc

el. r

espo

nse

RM

S, g

's

Flow oscillator frequency (Hz)

1.0

0.8

0.6

0.4

0.2

0 1 2 3 4 5 6

b) Gust load alleviation results.

Figure 9. Experimental results.

Figure 10. Leading-edge extension vortex burston a F/A-18 at 30° angle-of-attack.

Figure 11. 1/6-scale F/A-18 model mounted in theTDT.

Figure 12. Piezoelectric actuator on flexible wing.

1.0

Angle of Attack, deg26 30 34 37

0

Open-loop Closed-looprudder

1.0

028 30 32 34 37

Open-loop Closed-loopPiezoelectricactuators

Angle of Attack, deg

PS

D o

f roo

t ben

ding

mom

ent n

orm

aliz

edto

max

imum

ope

n-lo

op v

alue

PS

D o

f roo

t ben

ding

mom

ent n

orm

aliz

edto

max

imum

ope

n-lo

op v

alue

a) Rudder results.

b) Piezoelectric actuator results.

Figure 13. Normalized power spectral density of root bending moment at first bending frequency for open-loop and closed-loop systems.

a) Model installed in the TDT.

b) Active control system installed on model.

Figure 14. Wing and rotor aeroelastic testing system.

75 100 125 150Airspeed, knots

160

120

80

40

0

3P wing

loads, in-lbs

Wing BeamWing ChordWing Torsion

MAVSS- Off On Off OnOff OnOff On

Figure 15. Effect of Multipoint Active Vibration Suppression System on loads as a function of airspeeed.

0 20 40 60 80 100

Time, sec

b) Data image.

a) Area of Mir imaged.

120 140 160 180854

855

856

c) Displacement at o due to Shuttle jet firing, in.

Figure 16. Photogrammetric Appendage Structural Dynamics Experiment.

Conventional Technology

New InterfaceTechnology

Detail Modeling

Assembly Modeling

Repair Modeling

Repair

Model

Figure 17. Interface technology applications.

Fuselage Panel Component Models

Skin

Frames

Predicted Deformation Due to Pressurization

Radial displ.(inches)

frame locations

frames

.18

.12

.06

.00

p=10 psi.

Figure 18. Demonstration of interface technology for component synthesis.

Lear Fan 2100 Aircraft Pre-test Lear Fan Aircraft Post-test

Seat Data from Lear Fan Crash Test

-4000

-3000

-2000

-1000

0

1000

0 20 40 60 80 100Time, msec

Lumbarload, lb

Pilot

Co-pilot

Part 23 Criteria

Pilot - Standard Non-EA SeatCo-Pilot - JAARS EA Seat

Figure 19. Crashworthiness research.

Acc

el.,

g's

250500

400

300

200

100

0 40 80

200

150

100

50

-50

0

0 10 20Time, ms

Mid-point stroke, %

Cru

sh lo

ad,

lbf/i

n.

Foam filled beam with seat rail

Figure 20. Composite floor beam concept.

3 ft

8 ft

12 ft

12 ft

Load transition structure

Wing stub box

Wing-tip extension structure

Figure 21. Wing stub box test article.

Front spar

Spar access holes

Rib cutouts

Rear spar

12 ft

8 ft

Intercostal

Impact point

Upper cover layout

Interior of the composite stub box

Figure 22. Layout of composite stub box.

Wing root load transition structure

Wing-tip extension structure

Hydraulic jack

Hydraulic jack

Composite wing stub box

Composite wing stub box

Vertical reaction structure

Figure 23. Wing stub box test setup.

Failure line

Runout

RunoutImpact site

Figure 24. Upper cover of wing box after failure.

0

40

80

120

160

200

20000

Gages 63 and 64

-2000-4000-6000-8000Strain, micro-inches/inch

Load

, kip

s

0

40

80

120

160

200

20000-2000-4000-6000-8000Strain, micro-inches/inch

Load

, kip

s

Strain gage 64Strain gage 63Analysis

TestTest

Strain gage 64Strain gage 63Analysis Test

Test

a) Initial analysis.

b) Refined analysis.

Figure 25. Comparison of strain response between analysis and experiment.

Exploded view

Side quadrant - sandwich

Crown quadrant - skin stringer

Aft fuselage section

Keel quadrant - sandwich

Figure 26. Generic wide-body transport aircraft fuselage structure.

122-in. outer radius1.53-in. thick (constant)

100 ft-lb energy impact

100 ft-lb energy impact

2-in. x 0.19-in. notch through both face sheets

Figure 27. Forward fuselage keel panel testspecimen.

Figure 28. Failed keel panel specimen.

Unsupported sides Loading platen

Notched test specimenSupport fixture

Figure 29. Typical axial compression test setup.

250,000

200,000

150,000

100,000

50,000

.14.12.10.08.06.04.02

Specimen 7 (undamaged) Specimen 4 (100 ft-lbs) Specimen 5 (60 ft-lbs) Specimen 6 (40 ft-lbs)

300,000

250,000

200,000

150,000

100,000

50,000

.14.12.10.08.06.04.02

Specimen 2 (undamaged) Specimen 1 (100 ft-lbs)

Specimen 3 (60 ft-lbs)

Load

, lbs

Load

, lbs

End shortening, δ, in.

(b) Specimens 4 - 7

End shortening, δ, in.

(a) Specimens 1 - 3

0

0

δ

δ

Specimen 2 failure

Specimen 3 failure

Specimen 1 failure

Specimen 7 failure

Specimen 6 failure

Specimen 4 failure

Specimen 5 failure

Load

Load

Impactdamagelocation

Impactdamagelocation

Figure 30. Experimental load-shortening resultsfrom compression after impact tests.

1.0

0.8

0.6

0.4

0.2

10080604020

Specimens 1-3 Specimens 4-7

Nor

mal

ized

failu

re lo

ad

Impact energy, ft-lbs0

Figure 31. Normalized failure loads as a function of impact energy.

Figure 32. Typical failure mode.

Test panel

Failed panel

Tow- placed skin

Braided frames

Bolted secondary flange

Mouse hole

Bonded aluminum doublers

Figure 33. Composite fuselage crown panel testresults.