REG NACA-TR-669-1939 Airfoil section data obtained in the N A C A variable-density tunnel as affected by support interference and other corrections.pdf

1、l=NATIONAL ADVISORY COMMI_EEFOR AERONAUTICSREPORT No. 669AIRFOIL SECTION DATA OBTAINED IN THE N. A. C. A.VARIABLE-DENSITY TUNNEL AS AFFECTED BYSUPPORT INTERFERENCE AND OTHER CORRECTIONSBy EASTMAN N. JACOBS and IRA H. ABBOTTCASE FI LECOPY1939For role by the Supedntenden! of Dorumengs, Washington, D.C

2、. “ l_lce 10 centsSubscription price, S3 per yearProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Length Time Force AERONAUTIC SYMBOLS1. FUNDAMENTAL AND DERIVED UNITSPower .Speed .SymbolP1“MetricAbbrevia-Unit tionmeter . msecond sweight of 1 kilogram

3、kghorsepower (metric)_ _ .)kilometers per hour k.p.h._meters per second . m.p.s.I2. GENERAL SYMBOLSEnglishUnitfoot (or mile) .second (or hour) .weight of 1 pound .horsepower .miles per hour feet per second Abbrevia-tionft. (or mi.)sec. (or hr.)lb.l_, Weight=msg, Standard acceleration of gravity-9.80

4、665m/s 2 or 32.1740 ft./see3Wrn, 51 ass _- gI, Moment of inertia=m_ “2. (Indicate axis of_, Kinematic viscosityp, Density (mass per unit volume)Standard density of dry air, 0.12497 kg-m4-s I15 C. and 760 ram; or 0.002378 lb.-ft. -4 see2Specific weight of “standard“ air, 1.2255 kg/m 30.07651 lb./cu,

5、ft.radius ofgyration k by proper subscript.)Coefficient of viscosity3. AERODYNAMIC SYMBOLSatorS, Area iw,S_, Area of wing “.G, Gap _“b, Spanc, Chord Q,b2 fl,_,_ Aspect ratio _lV, True air speed P _1q, Dynamic pressure= _oV“-L, Lift, absolute coefficient C_=_S SD, Drag, absolute coefficient Co=_ST_Do

6、, Profile drag, absolute coefficient CDO=_ a,D_, Induced drag, absolute coefficient C._=_-_ ao,D OttjDp, Parasite drag, absolute coefficient CDp=_ a_,C, Cross-wind force, absolute coefficient Ce=_ 7,R,Cp,Angle of setth-g of wings (relative to thrustline)Angle of stabilizer setting (relative to thrus

7、tline)Resultant momentResultant angular velocityReynolds Number, where l is a linear dimension(e.g., for a model airfoil 3 in. chord, 100m.p.h, normal pressure at 15 C., the cor-responding number is 234,000; or for a modelof 10 cm chord, 40 m.p.s., the correspondingnumber is 274,000)Center-of-pressu

8、re coefficient (ratio of distanceof c.p. from leading edge to chord length)Angle of attackAngle of downwashAngle of attack, inflnife aspect ratioAnglo of attack, inducedAngle of attack, absolute (measured from zero-lift position)Flight-path angleResultant forceJj_JfJProvided by IHSNot for ResaleNo r

9、eproduction or networking permitted without license from IHS-,-,-REPORT No. 669AIRFOIL SECTION DATA OBTAINED IN THE N. A. C. A.VARiABLE-DENSITY TUNNEL AS AFFECTED BYSUPPORT INTERFERENCE AND OTHER CORRECTIONSBy EASTMAN N. JACOBS and IRA H. ABBOTTLangley Memorial Aeronautical LaboratoryProvided by IHS

10、Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-NATIONAL ADVISORY COMMITTEE FOR AERONAUTICSHEADQUARTERS, NAVY BUILDING, WASHINGTON. D. C.LABORATO tIES. LANGLEY FIELD. VA.Created by act of Congress approved March 3, 1915, for the supervision and direction of the sci

11、entific study of the problems offlight (U. 8. Code, Title 50, 8ec. 151). Its membership was increased to 15 by act approved March 2, 1929. The members areappointed by the President, and serve as such without compensation.JOSEPH S. AMES, Ph.D., Chairman,Baltimore, Md.VANNEVAR BUSH, 8C_ D., Vice Chair

12、man,Washington, D. C.CHARLES G. ASBOT, Se. D.,Secretary, Smithsonian Institution.:HENRY H. ARNOLD, Major General, United States Army,Chief of Air Corps, War Department.GEORGE iT. BRETT, Brigadier General, United States Army,Chief Mat6riel Division, Air Corps, Wright Field, Dayton,Ohio.LY2_IA_ J. BRI

13、GGS, Ph.D.,Director, National Bureau ofStandards.Ct,INTON _I. I_ESTER, A. _., LL.B.,Administrator, Civil Aeronautics Authority,ROBERT tI. HINCKLEY, A. B.,Chairman, Civil Aeronautics Authority.JEROME C. HUNSAEER, Sc. D.,Cambridge, Mass.SYDNEY _I. KRAVS, Captain, United States Navy,Bureau of Aeronauti

14、cs, Navy Department.CHARLES A. LINDBERGH, LL.D.,:New York City.FRANCIS W. REICHELDERFER, A. B.,Chief, United States Weather Bureau.JoHr H. TOWERS, Rear Admiral, United States Navy,Chief, Bureau of Aeronautics, Navy Department.EDWARD WARNER, So. D.,Greenwich, Conn.ORVILLE WRIGHT, Be. D.,Dayton, Ohio.

15、p.?GEORGE W, LEWIS, Director of Aeronautical ResearchJOHN F. VICTORY, SecretaryHENRY J. E. REID, Engineer-in-Charge, Langley Memorial Aeronautical Laboratory, Langley Field, Va.JOHN J. IDE, Technical Assistant in Europe, Paris, FranceJAERODYNAMICSPOWER PLANTS FOR AIRCRAFTAIRCRAFT MATERIALSLANGLEY ME

16、MORIAL AERONAUTICAL LABORATORYLANGLEY FIELD. VA.Unified conduct, for all agencies, of scientific research on thefundamental problems of flight.TECHNICAL COMMITTEESAIRCRAFT STRUCTURESAIRCRAFT ACCIDENTsINVENTIONS AND DESIGNSCoordination of Research Needs of Military and Civil AeiationPreparation of Re

17、search ProgramsAllocation of ProblemsPrevention of DuplicationConsideration of InventionsOFFICE OF AERONAUTICAL INTELLIGENCEWASHINGTON, D. 7.Collection, elass!fication, compilation, and disscminatio6 ofscientific and technical information on aeronautics.Provided by IHSNot for ResaleNo reproduction o

18、r networking permitted without license from IHS-,-,-REPORT No. 669AIRFOIL SECTION DATA OBTAINED IN THE N. A. C. A. VARIABLE-DENSITY TUNNELAS AFFECTED BY SUPPORT INTERFERENCE AND OTHER CORRECTIONSBy EASTMAN N. JACOBS and I1tA H. ABBOTTSUMMARYThe results of an investigation oJ the effect of supportint

19、erference on airfoil drag data obtained in the variable-density tunnel are presented. As a result of the supportinte?ferenee, previously published airfoil data from theeariable-density tunnel have ._hown too large drag coeffi-cients and too large a rate of increase of drag coefficientwith airfoil th

20、ickness. Thdpraetieal effect of the correc-tions on the choice q the optimum section is brieflyconsidered and corrected data for a selected list of airfoilsare presented as a convenience to the designer. 1Iethodsof correcting published data for other airfoils are presented.INTROD UCTIONAirfoil data

21、obtained in the variable-density tunnel(reference 1) have been published (references 2 to 6) informs that were considered at tlle time of publicationto be most useful to the airplane designer. In theearlier publications (references 1 and 2) no correctionsother than flmse for tunnel-wall effects and

22、to infiniteaspect ratio were applied to the data, and emphasis wasplaced on the pressing problem of obtaining good com-parative data for judging the relative merits of airfoilsrather than on ol)taining absolute accuracy.It was recognized that certain consistent errors werepresent in the data, but it

23、 was tlmught that the effectof these errors on the comparative value of the datawas not of primar 3, importance. Support-strut inter-terence, for example, was considered to be a possiblesource of systematic error, but it was thought that thisinterference would not affect the order of merit of theair

24、foils tested except possibly in the case of very sensi-tive airfoils, which might also be similarly affected bythe wing-strut intersections of biplanes common at thetime. The turbulence of the air stream was thoughtnot seriously to impair the comparative value of thedata aml, perhaps, even to be des

25、irat)le, because theextensive turbulent t oundary layers occurring on timnmdels in the tunnel as a result of the turbulence wotfldalso be found in practice at high values of the ReynoldsNumt)er on conventional airfoils with the. usual mod-erately rough surfaces. It was also considered thaterrors ari

26、sing from failure of the conventional airfoiltheory to predict sect ion characteristics accurately fromthe model tests would largely disappear when tile dataso derived were used to predict the characteristics ofwings approximating the same plan form and aspectratio as the models.The absolute accurac

27、y of the data was, however,improved from time to time by the investigation ofconsistent errors. An attempt to ewduate the effectof support in t erferen c e on the m ea sured drag co efficien tswas inconclusive (reference 4) and no corrections wereapplied. The data were further improved by the ap-pli

28、cation of corrections for turbulence and for improve-ment of the approximations to section characteristics.The corrected coefficients were designated by lower-casesymbols, such as ca0, as contrasted to the older CD o.One of the chief effects of these corrections was to re-duce the profile-drag coeff

29、icients, particularly for thethicker airfoils.As airfoil data at large values of the Reynolds Num-ber became available from the N. A. C. A. full-scaletunnel (reference 7) and from foreign sources (references8 to 13), even the corrected profile-drag coefficients ob-tained in the variable-density tunn

30、el appeared to be toolarge. The discrepancy increased with airfoil thickness.The important practical effect is that the data from thevariab.e-density tunnel apparently showed too largea variation of drag coefficient with airfoil thickness.Correct information regarding this variation may be ofprimary

31、 importance to the airphme designer in choosingthe optimum airfoil sections for actual wings.Further investigations of this subject were under-taken, one of the most important being an investiga-tion of three symmetrical sections, N. A. C. A. 0009,0012, and 0018, under conditions of low turbulence i

32、nthe N. A. C. A. full-scale tunnel. Results from thisinvestigation (references 14 and 15) indicate a smallerincrease in drag with airfoil thickness than is indicatedby the results from the N. A. C. A. variable-densitytunnel. Furthermore, comparative tests were madein the t_-o tunnels by applying str

33、ings to the surfaceof the N. A. C. A. 0012 airfoil to nmve the transitionpoint to a predetermined position. These tests indi-cated that, for this airfoil, the discrepancies were toolarge to be ascribed to failure of the effective ReynoldsNumber concept to correct approximately for the dragas affecte

34、d by transition.1Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2 REPORT NO. 669_NATIONAL ADVISORY COMMITTEE FOR AERONAUTICSAnother correction, however, was suggested by tileinvestigation in the full-scale tunnel. Differencesbetween the results from

35、 force and momentum methodsof measurement suggested the presence of incrementsof support-interference drag that increased with sectionthickness. Further tests, made with additional dummysupports, verified the presence of this type of supportinterference in the full-scale tunnel. Tests weretherefore

36、started in tile variable-density tunnel toinvestigate any variation of support interference withairfoil thickness, in spite of the fact that previousinvestigations (see appendix of reference 4) had showqqno definite corrections for two airfoils, the N. A. C. A.0012 and 4412. Improvements of the bala

37、nce of thevariable-density tunnel were expected to enable greateraccuracy than was obtainable from tlle previous bal-ance arrangement. Tile results of this investigationindicate that marked increments of support-interfer-ence drag, easily measurable, are “present in the dragresults from the variable

38、-density tmmel, the incrementincreasing with airfoil thickness.The purpose of this report is to present the cor-rections for application to published results from timvariable-density tunnel to give more reliable valuesof section profile-drag coefficient for airfoils of variousthicknesses. The practi

39、cal effect of the correctionson tim choice of tile optimmn section is briefly con-sidered. Comparison is made between some correcteddrag data fronl the variable-density tunnel and fromother serecos to show tile extent of the existing agree-meat. Corrected data for a selected list of airfoils arealso

40、 present,d as a convenience to tile designer.METHODThe standard nletllod of testing in the variable-density tunml, the model supports, and the method ofdeterI_ning the tare forces are described in reference 1.The usual tare tests de_ermine the tare forces on thesupl)orts including, the interference

41、of the model onthe supports. The c,mventinnal method of dcternfin-ing the balance-alinenwnt correction by testing asymmetrical airfoil through positive and negativeangles of attack determines the eltects of balance andair-stream misalinement and any interference of thesupports on the model tlmt is e

42、quivalent to a changeill ,fir-flow direction.The method selected for investigating the additionalinterference of the supports on the model was the sameas that described in the appendix of reference 4. Testswere made of each airfifil supported by three lifferentnwthods. Besides the method of using ti

43、le usual sup-port struts, tests were made with tile models mmmtedon the usual supports with the addition of special sup-ports and with the lnodels mounted only on the specialsupports. The special supports consisted of three wiresattached to the quarter-chord point of the model ateach wing tip an,I o

44、f a sting and nn angle-of-attackstrut so located as to be as free as possible from aero-dynanfic interference with the regular supports. Thesting was symmetrical with respect to the airfoil andwas attached near tile trailing edge instead of to thelower surface, as usual.The tares due to the special

45、supports were deter-mined from data obtained from the tests with the modelson the regular supports with and without the specialsupports. These tares were then applied to the dataobtained with the model on the special supports alone;the results were then compared with the data obtainedin the customar

46、y manner to detemnine the unevahmtedinterference caused by the usual supports. This m_thoddoes not eliminate bahmce deflections arising fromsources other than aerodynamic forces on the modeland the supports. A test was accordingly made withno motlel nor supports in the tunnel; the result showedthat

47、no such balance deflections were present.Tlle scope of the present investigation was linfitedto tim study of tile profile drag at low and moderatelift coefficients at the highest value of the test ReynoldsNumber ordinarily obtained (about 3,000,000). Testswere made of the N. A. C. A. 0012, 0018, 002

48、5, 0030,and 0040 symmetrical airfi)ils to study tlle variation ofsupport interference with airfifil thickness. The N. A.C. A. 43012, 43018, and 8318 .firfoils were also testedto obtain an indication of the variation of support inter-ference witll camber.RESULTS AND DISCUSSIONMINIMUM PROFILE-DRAG COE

49、FFICIENTSThe effect of the support interference on the measuredsection minimum profile-drag coefficients is shown infigure 1. The increment of the mininmm profile-drag coefficients caused by the support interference isplotted against airfoil thickness for the five symmetricaland the three cambered airfoils tested. The pointsfor the five symmetrical airfoils lie on a fair curvepassting t_rough zero at zero airfoil thickness,