NASA-TP-3133-1991 Full-scale semispan tests of a business-jet wing with a natural laminar flow airfoil《带有自然层流机翼商务喷气机翼的全面半翼展试验》.pdf

1、_*/,NASA_Technical:_iPaper_:3133 September 1991 I“- Full-Scale SemispanTests of a Business-JetWing With a NaturalLaminar Flow AirfoilDavid E. Hahne andFrank L. Jordan, Jr.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo

2、 reproduction or networking permitted without license from IHS-,-,-NASATechnicalPaper31331991National Aeronautics andSpace AdministrationOffice of ManagementScientific and TechnicalInformation ProgramFull-Scale SemispanTests of a Business-JetWing With a NaturalLaminar Flow AirfoilDavid E. Hahne andF

3、rank L. Jordan, Jr.Langley Research CenterHampton, VirginiaProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-The use of trademarks or names of manufacturers in thisreport is for accurate reporting and does not constitute anofficial endorsement, either

4、expressed or implied, of suchproducts or manufacturers by the National Aeronautics andSpace Administration.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-AbstractAn investigation was conducted in the Langley30- by 60-Foot Tunnel on a full-scale semi

5、span modelto evaluate and document the low-speed, high-liftcharacteristics of a business-jet class wing that uti-lized the HSNLF(1)-0213 airfoil section and a single-slotted flap system. In addition to the high-liftstudies, boundary-layer transition effects were exam-ined, a segmented leading-edge d

6、roop for improvedstall/spin resistance was studied, and two roll-controldevices were evaluated.The wind-tunnel investigation showed that de-ployment of a single-slotted, trailing-edge flap waseffective in providing substantial increments in liftrequired for takeoff and landing performance. Fixed-tra

7、nsition studies to investigate premature trippingof the boundary layer indicated no adverse effectson lift and pitching-moment characteristics h)r eitherthe cruise or landing configuration. The full-scale re-sults also suggested the need to further optimize theleading-edge droop design that was deve

8、loped in thesubscale tests.IntroductionWhile much research on natural laminar flow(NLF) airfoils has recently focused on drag reduc-tion for improved cruise performance, few studieshave addressed the use of high-lift systems for takeoffand landing with this wing class. Although large im-provements i

9、n cruise performance have been shown,these NLF airfoils will only be used if they carl beequipped with a viable flap system that is capableof generating enough lift to meet takeoff and landingrequirements.Prior to this investigation, some two-dimensionalwind-tunnel tests had been conducted to evalua

10、tehigh-lift characteristics of NLF airfoils and to sup-port associated theoretical studies of flap effective-ness (refs. 1 and 2). These tests were focused onthe use of simple split flaps. Other studies wereconducted that used theoretical methods to designmore complex flap systems for NLF airfoils (

11、ref. 3).One of the airfoil sections used for the study inreference 3 was the high-speed HSNLF(1)-0213 air-foil. This airfoil was developed to extend the nat-ural laminar flow concepts that were developed forlow-speed airfoils to airfoils intended for higher speedand Reynolds number applications (ref

12、s. 4 to 6). Asstated in references 6 and 7, the HSNLF(1)-0213 air-foil was designed for a cruise section lift coefficientof 0.26 at a Mach number of 0.7 and a Reynoldsnumber of 9 x 106. Theoretical data on the airfoilpredicted that large increments in lift could be ob-tained with a slotted flap desi

13、gn (ref. 3). As ex-pected, the amount of additional lift and the angle ofattack for maximum lift depended on the flap gcom-etry. Two-dimensional theoretical studies indicatedthat a single-slotted flap design would offer a goodtrade-off between CL,max and flap complexity for useon lightweight busines

14、s jets (ref. 3). However, be-cause theoretical techniques cannot reliably predictmaximum lift for three-dimensional wings with flaps,experimental tests are necessary to accurately evalu-ate any flap system.In the present investigation, tests were conductedin the Langley 30- by 60-Foot Tunnel on a fi

15、ll-scalesemispan model that incorporated the HSNLF(1)-.0213 airfoil section. The main objective of these testswas to evaluate and document the low-speed, high-lift characteristics of a business-jet class wing thatused the HSNLF(1)-0213 airfoil section and a single-slotted flap system that was design

16、ed with the aidof the computer code described in refercnce 8. Thisflap system was the same as the one discussed in ref-erence 4. Photographs of the model mounted for testsare shown in figure 1. Figure l(a) shows the modelwith the flap retracted and the flow going from rightto left. Figure l(b) shows

17、 a close-up of the undersideof the model with the flap deflected 40 . In additionto the high-lift studies, boundary-layer transition ef-fects were examihed, a segmented lcading-cdge droopfor improved stall/spin resistance wins studied, andtwo roll-control devices were evaluated.SymbolsLongitudinal f

18、orces and moments are presentedin the stability-axis system, and lateral forces andmoments are presented in the body-axis system. Amoment reference center of 0.25_ was used for alltests.b wing span, ftCD drag coefficient, _L_q_ :_CL lift coefficient, LiftCL,ma x maximum lift coefficientC l rolling-m

19、oment coefficient,Rolling momentq_Sb/2pitching-moment coefficient,Pitching momentq_,S_pressure coefficient, q,x_incremental rolling-momentcoefficientlocal wing chord with droop off, ftCmProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Ppocq_RSxYzo_6at

20、_fAbbreviations:FSMCARFNLFVGWSmean aerodynamic chord, ftlocal static pressure, lb/ft 2free-stream static pressure, lb/ft 2free-stream dynamic pressure, lb/ft 2Reynolds number based onsemispan reference area, ft 2chordwise distance from wingleading edge, positive aft, ftspanwise distance from wing ro

21、ot, ftnormal distance from wing leadingedge, positive up, ftangle of attack, degaileron deflection, positive trailingedge down, degflap deflection, positive trailing edgedown, degspoiler deflection, positive trailingedge up, degfuselage station, in.Multi-Component Airfoil AnalysisProgramnatural lami

22、nar flowvortex generatorwing station, measured plane ofwing, in.Model Description and ApparatusThe geometry of the semispan model tested isshown in figure 2, and a summary of the geo-metric characteristics is contained in table I. TheHSNLF(1)-0213 airfoil section used in these tests isshown in figur

23、e 3, and section coordinates for this air-foil are given in table II. The wing incorporated 3of twist between wing station 0.0 and the 50-percentsemispan station. An additional 1 of twist was in-corporated between the 50-percent semispan stationand the wingtip for a total of 4 washout. The in-board

24、portion of the wing was twisted about the 30-percent chord line, and the outboard portion of thewing was twisted about the 78-percent chord line. Asmall winglet was located at the wingtip. The modelalso incorporated an aileron and spoiler for roll con-trol. (See fig. 2.) A half body of revolution wa

25、sincorporated to simulate the presence of a fuselagenear the wing. This fuselage was representative of abusiness jet in both size and shape. A vortex gener-ator was mounted on the fuselage (fig. 4) just abovethe wing-body juncture to delay flow separation onthe inboard panel of the wing. A multiposi

26、tion flapsystem was incorporated in the model for evaluation.(See figs. l(b) and 5.) The flap was a 28-percentchord flap that extended from the wing root to asemispan location of 2y/b = 0.79. Flap deflections(0 , 20 , and 40), flap gap, and flap overlap wereset by changing three brackets that were l

27、ocated onthe lower surface of the wing. Flap overlap was de-fined as the distance from the trailing edge of thewing upper surface (0.92c) to the leading edge of theflap (negative when the wing overlaps the flap). Flapgap was defined as the shortest vertical distance be-tween the wing upper surface (

28、0.92c) and the flapleading edge. The nominal flap overlap and gap were0 and 2 percent of the local wing chord for the flapdeflected 40 and -3 and 4 percent of the local wingchord for the flap deflected 20 .In an attempt to improve stall/spin resistance, asegmented leading-edge droop was developed fo

29、r thewing prior to the full-scale senlispan tests. These ex-ploratory tests used two subseale models. The mod-els incorporated the same airfoil section and wingptanform as the full-scale semispan model. However,the subscale model wings that were used for droopdevelopment were not twisted, and they d

30、id not in-corporate a winglet. Static force tests, which wereconducted in the Langley 12-Foot Low-Speed _ln-nel at a Reynolds number of 3.1 x 105, were usedto develop several candidate droop geometries. Theroll-damping characteristics of these droop designswere then evaluated in dynamic force tests

31、in theLangley 30- by 60-Foot Tunnel. This evaluation wasused to select the final droop configuration for thefull-scale tests. These roll-damping tests were con-ducted at a Reynolds number of 9.7 x 105. Refer-ence 9 contains a description of the techniques usedto develop this droop design. Figure 6 s

32、hows theleading-edge droop location and droop section thatresulted from the subscale tests. The droop designconsisted of two segments: an outboard segment thatextended from the tip inboard to approximately the70-percent semispan location and a smaller segmentthat was mounted farther inboard between

33、the 40-and 50-percent semispan locations. The droop sec-tion was derived from another NLF airfoil (NLF(1)-0215F). This approach was adopted in an attempt toachieve natural laminar flow on the drooped as wellas the undrooped portions of the wing. Coordinatesfor the drooped airfoil section are given i

34、n table III.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Unlessotherwisenoted,thedatapresentedhereinarefor configurationsthat donot includethedroop.Static force and momentmeasurementsweremadein the Langley30-by 60-FootTunnelwiththeexternalscalesys

35、temthat is describedin refer-ence10. Theloadsonboth thewingandthe fuse-lageareincludedin all forceandmomentdata. Inadditionto forceandmomentmeasurements,static-pressuredata,flowvisualization,andhot-fihndatawereobtained.A total of 322pressuretapswerespacedalongthe spanof tile wing at eight sta-tions:

36、20-,35-,45-,55-,62-,75-,85-,and95-percentsemispanlocations.Chordwiselocationsof boththeupperandlowersurfaceportsarelistedin tableIV.A total of 45hot-filmsensorsweremountedonthewing (both upperand lowersurfaces)to measureboundary-layerbehaviorat threespanwiselocations.(Seefig. 7.) Surfacetufts wereus

37、edto visualizethesurfaceflowconditions,especiallystall progression.Thesetuftswereremovedwhilepressureandhot-filmdatawereobtained.Test Conditions and CorrectionsTestswere conductedover an angle-of-attackrangefrom-10 to 40. Aerodynamicforceandmo-mentdatawereobtainedat free-streamvelocitiesof55,66,and7

38、7mph,whichcorrespondto Reynoldsnumt)ersbasedon_of about3.05x 106,3.67x 106,and4.26x 106,respectively.Mostof the testswereconductedat a free-streamvelocityof 66mph;un-lessotherwisenoted,thedatapresentedarefor thiscondition. Althoughsomeroll dataweretakentoevahmteaileronand spoilereffectiveness,the te

39、stsfocusedon longitudiimlcharacteristicsof the semi-spantoo(tel.A wind-tunnelcalibrationwas madeprior tomodelinstallationto determinebuoyancyandflowangularitycorrections.Flow-fieldsurveysweremadeto determineflowblockagecorrectionsin themannerdescribedin reference11.Correctionsforjet bound-ary interf

40、erenceweremadein accordancewith themethodof reference12and aredescribedin refer-ences13to 15. TheLangley30-by 60-FootTunnelhas a measured turbulence factor of 1.1, which cor-responds to an average turbulence level of approxi-mately 0.1 percent of the mean flow velocity (ref. 16).Results and Discussi

41、onPressure DistributionsChordwise pressure distributions for (5f = 0 and5f = 40 are presented in figures 8 and 9. Data wereobtained for angles of attack between -2.2 and 16.8 at the eight semispan stations that were previouslydiscussed. Problems with the pressure measurementsystem resulted in a limi

42、ted amount of reliable pres-sure data. The pressure data presented in figure 8(b)for cz = 1.5 with the flaps retracted (CL _ 0.3)show that pressure gradients are conducive to lanfi-nar flow over much of the upper and lower surfacesof the wing. This conduciveness is indicated by thedecreasing values

43、of the pressure coefficient with thechord station up to about the 60-percent chord sta-tion. These results indicate that, even at low speeds,the amount of laminar flow possible over the wing atcruise angles of attack should be significant.A comparison of the pressure data in figures 8and 9 (_i = 0 a

44、nd fI = 40) indicates that largeincreases in lift result from flap deflection. Eventhough the vortex generator wa_s on for the bf = 40 data, the generator was believed to have no effect onthe pressure data because flow visualization studiesindicated that the vortex generator primarily affectedthe fl

45、ow inboard of the 2y/b = 0.2 pressure portstation. Integration of the pressure distribution datato calculate the lift on the wing and flap indicatesthat this increase in total lift results not only from thelift generated on the flap but also from the enhancedlift characteristics on the main wing.Eff

46、ect of Reynolds NumberThe effect of Reynolds number on the longitudi-nal characteristics for the model without the vortexgenerator and with the flaps retracted is shown infigure 10. Changes in Reynolds number had no ef-fect on lift characteristics except in the maxinmmlift and post-stall angle-of-at

47、tack regions. The in-crease in both the maximmn lift coefficient and stallangle of attack at the higher Reynolds numbers re-sulted from the increased resistance of the boundarylayer to separation. With the flaps deflected 40 andthe vortex generator installed, Reynolds number ef-fects on lift and pit

48、ching moment were limited to asmall angle-of-attack range just past the stall. (Seefig. 11.) The maximum lift coefficient and stall an-gle of attack, however, were not affected by Reynoldsnumber variations.Two-dimensional data from reference 7 for thebasic airfoil with _f = 0 showed very little chan

49、gein minimum drag for Reynolds numbers between4.0 x 106 and 9.0 x 106 at low subsonic Mach numbers.However, these two-dimensional tests did indicatethat there were some Reynolds number effects onmaximum lift between R = 4.0 x 106 and R =6.0 x 106. Since the full-scale data indicated virtuallyno Reynolds number effects between R = 3.67 x 106and R = 4.26 x 106 (the maximum Reynolds numberob