1、WUJ.RM A55K09-.RESEARCH MEMORANDUM-2$ -l:tkA,ktiliiiA i)/jy) T)b.-DA7t!“”Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TECHLIBRARYKAFB.NMivNACA R4 A55K09NATIONAL ADVISORY COMMITTEE FORRESEARCH MEMORANDUMIlllllllllillllllllllllllllljlllltlli335iAERO
2、NAUTICSFORCE, MOMENT, AND PKESSURE-DISTRIBUTIONCHARACTERISTICSOF RECTANGULAR WINGS AT HIGH ANGLES OF ATTACKAND SUPERSONICSPEEDSBy Williem C. PittsSJMMARYExperimental force and mcment data are presented for rectangularwings of aspect ratios 1, 2, and 3. The angle-of-attack range is aboutThe No. 2 tun
3、nel is an intermittent-operation,nonreturn, variable-pressurewind tunnel that has a Mach number rangefrom 1.2 to 4.0. In both tunnels the Mach number is changed by varyingthe contour of flexible plates which comprise the top and bottom wallsof the tunnels. The No. 1 tunnel was used to obtain.the e m
4、easurements were made on the pressure-distributionwing for :) (3.20).43 0.33 0.40 0.44(.35) ( .45) (*47)2.12 .45 .451.* :)( .47) (:)1.96 ;:j;) (:$ (2*14)2“43 (%) (i%) (;:% 2.43 (:;)( .49) ( “49)3.36 ;:;, (;:;) (%)(Qdn1 2 31.45 0.016 0.018 0.0201.96 .012 .014 a71 0152.43 .010 .011 .0123.36 .007 .007
5、.008I (L/D)M I2 31.45 4.9 6.2 6.1L*96 597 5.7 5.82.43 5.8 5.8 5.83.36 6.4 I 6.7 5.7The numbers in the parenthesis are linear-theory values. The trends in.c%) with and A are we redicted by linear theory, but theProvided by IHSNot for ResaleNo reproduction or networking permitted without license from
6、IHS-,-,-wNACA RM A55K09 9.predicted magnitudes of the lift-curve slope are somewhat low. Thecenter-of-pressureposition predtcted by linear theory is about 3 per-cent of the wing chord too far aft for all Mach numbers and aspectratios. This is primarily due to second-order effects of thickness.The ce
7、nter-of-pressuretravel with Mach number is primarily due to thewing-tip effect rather than section effects. This is apparent from thefact that the center-of-pressureposition for the aspect-ratio-3 wing,which approaches a two-dimensional airfoil, is nearly constant. Regard-ing (L/ll)mx, it 3s not sur
8、prisingthat no general trends occur sincethe drag due to the lift and Chin have opposite effects upon (L/D)Was Mach number and aspect ratio vary.CORRELATION AND DISCUSSIONBasic Physical PhenomenaBefore discussing the method used to correlate the rectangular-wing data, it is well to describe first so
9、me of the basic physicalphenomena of the flow over a-three-dimensional,rectangular wing. Asketch of an aspect-ratio-2 semispan wing is shown in figure 6. Theestimated Mach waves from the wing tip for and, two, the positionof the Mach wave is predicted incorrectly, as shown by the insert. An .obvious
10、 modification is to stretch the Busemann theory as shown by thedashed curve so that it agrees with two-dimensional,shock-expansiontheory at the correct Mach wave position. (This is essentially themethod used in ref. 12.) However, the experimentaldata are still notwell predicted. A linearized, conica
11、l-flow theory that considers theeffect of the wing vortices is presented in reference 13. However, thistheory is not in good agreement with the experimental results of thisinvestigation as shown by figure 9. In this figure the theoretical andexperimentalvalues of the local loading (both surfaces) ar
12、e normalizedby the two-dimensional section loading and plotted against the usualconical parameter J3q/x. For = O there are no vortices present and .thethetheory reduces tothat of Busemann. It is apparent that-the flow intip region is not conical from the fact that when plotted against the .Provided
13、by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM A55K09 Uconical Parameter, j3q/x,the experimental loading pressures taken along. the q/s = 0.J.25station differ from those taken along the /s = 0.437station. Thus the poor agreement with the conical-flow
14、 theories ofreferences 11 and 13 is not surprising and a nonconical theory is needed.The-nonconical nature of the flow is probably due primarily to the factthat the surface of the wing is not conical at the tip. Better agree-ment with these theories shouldbe obtained for surfaces that areconical fro
15、m the leading edge of the wing tip.b the absence of an adequate theory, a semiempiricalmethod Wasformulated for predicting the span loading in the tip region. FigureIO shows the basis of this method. The experimental section load dis-tribution in the tip region is presented in normalized form for bo
16、th theupper and lower surfaces of the wing. The abscissa, q/q*, is the frac-tion of the distance from the wing tip, q/q* = O, to the total width ofthe tip region. The dashed curves are approximate fairtngs of theexperimental data for the lower surface of the wing. The shape of thisfamily of curves w
17、as based on data at all test Mach numbers and anglesof attack below shock detachment. It is apparent that the loadingincreases more rapidly with a in the tip region than in the two-dimensionalregion,and that it approaches rectangular loading as. pointed out in references 5 and 6. The solid curves sh
18、ow the variationdue to angle of attack ofcnJcns* at a.fixed geometric position onthe wing, y/s = 0.875. The upper and lower surface curyes cross neardthe Busemann theory curve. This is to be expected since, by symetry,the two experimental curves must cross at u = 0, and Busemann*stheory becomes exac
19、t as a approaches zero. The similarity of the spanloading curves in the tip regi,onsuggests the follofing seemPtricalmethod: (1) Use Busemannrs theory to give the basic shape of the loading “for a= OO. (2) Use shock-eansion theory to give the absolute magni-tudes at point (1,1) in figure 10. (3) Use
20、 an empirical correction toaccount for the effect of rx. This empirical correction will in generalbe a small percentage of the loading, so that great accuracy in thecorrection is not necessary.The form of this empirical correction can be seen from figure 11where ens/ens* is plotted against a for sev
21、eral,valuesof q/q*and for several lOSVCS% 12 92.4%= 1,t/c.2 .025-.O5 (x/c-.5) ; c= 2 (8.580)2-(Vc-.586)2-(8.560j; OSx/cS IPressure distribution wings Force test wingsFigure 1.- Wing dimensions.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. .24 16
22、8 O 0 c1 oaFigure 2.- The effects of Mach number and aspect ratio on the lift, , r8 16c2c3effIcien+,24 32 40 46from force teste.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-.tiCA RM A55K09 21cLO -.8 .6 -.4 -.2 0 .2 .4 .6CLFigure 3.- The effects of
23、 Mach number and aspectcoefficient from force tests.8 Lo 1.2ratio on the dragProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Ill-.12 .08 .04 0 0 0 0 -.04 -.00 -.12I cmFigure 4.- The effects of Mach number and aspect ratio on the pitching-moment coeff
24、icientfrom force teelm., ,I1. 14 1, , , ! il ,! , 1 IiProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-4 *,” -1.0 -.8.6 -.4 -.2 0Figure ?. - The effects of hkch numbero 0cand aqect ratio on thefrom force tests.o .2root -bending-moment,4 .6 ,acoefficie
25、ntProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-.-nl1=Boundary layer plate F-1 ,I ,I IIIII/15 II /I . _d.- Wing imagelower-surface/Mach wavesa. 5Mach waves 1:,Hmre 6.- Hp Mach curves on the pressure-distribution wings as computed by shock-eansiontheory; A= 2, = 1.97, lower surfacej m = 10,X/C = 0.617.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-
