NASA NACA-RM-A53I17-1953 A study of the effects of body shape on the vortex wakes of inclined bodies at a Mach number of 2《在马赫数为2时 飞机形状对倾斜车身涡粒的影响研究》.pdf

《NASA NACA-RM-A53I17-1953 A study of the effects of body shape on the vortex wakes of inclined bodies at a Mach number of 2《在马赫数为2时 飞机形状对倾斜车身涡粒的影响研究》.pdf》由会员分享，可在线阅读，更多相关《NASA NACA-RM-A53I17-1953 A study of the effects of body shape on the vortex wakes of inclined bodies at a Mach number of 2《在马赫数为2时 飞机形状对倾斜车身涡粒的影响研究》.pdf（26页珍藏版）》请在麦多课文档分享上搜索。

1、SECURITY INFORMATIONCOPY 235RMA531171,. -=ii:EA - .-. . . ,-_nsmely, a steady symmetricpair, a steadyasymmetric configuration of two or more vortices, and an unsteady con-figuration of two or more vortices. For most models the steady symmetricpair of vortices, such as illustrated in figure 3(a), is

2、observed in thelow ane-of-attack range (angles of attack,.lessthan about 150)= Asteady asymmetric configuration, figure 3(b)= is found in the inter- mediate ane-of-attack rsnge (angles of attack between about 15 and 280),and the unsteady configuration is found at large angles of attack.The angle-of-

3、attackrange for which any particular vortex pattern existsdepends upon a number of factors such as model geometry and Reynoldsnumber. The present investigation is concernedprimarily with theeffects of these two factors on the angle of attack at which the unsteadyvortex wake occurs. The results of th

4、is study are presented in thefollowing discussion.Effects of Body Expansion .The analogy between the developmentwith time of the flow about atwo-dimensional circular cylinder impulsively set in motion from restand the developmentwith distance along the body of the crossflow aboutan inclined body has

5、 been pointed out in reference 1. This analogysuggests that if the body is designed with a large rate of increase of.body cross-sectionalarea with distance along the body, both thesynmetry smd stability of the vortex pattern might be retained tolarge angles of attack. In order to investigate the eff

6、ects on thewake vortex configurationof various rates in increase in bcdy cross-secti.onalarea, a series of cones, cone-cylinders,and other simplemodels were tested. The results of these tests will be considered intwo parts: the expansion of the body nose, and expsmsion of theafterbody.-Expansicm of

7、the body nose.- The results of the tests with conesand cone-cylindersare presented in figure b(a) in which the angle ofattack at which unsteadiness in the crossflow first appeared is plottedas a function of nose apex angle. In thts figure the plotted positionof each symbol designates the lowest angl

8、e of attack at which wakeunsteadinesswas observed for the cone. In order to denote thatunsteadiness in the wake was not observed within the angle-of-attack *range of the tests for certain of the models, an arrow has been attachedto the symbol and the symbol has been plotted at the maximum angle ofat

9、tack of the tests. The curve drawn through the symbols thus indicates b11.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM A53117an approximateand the regiondivision line between the region for steady wake flowfor unsteady wake flow.1 This cur

10、ve shows that anincrease in the cone apex angle resulted in am increase in the angle ofattack at which wake unsteadiness first occurred. Although the fairingof the experimental curve for angles of attack greater than 30 mayappear somewhat srbitrary, it is based upon a number of observations ofsimila

11、r wakes and, in addition, on the foldmwing evidence. The datafor the 19 smd 30 apex angle cones (models 2-B and l-A, respectively)indicate that at the maximum angles of attack for these two models(35 d 40, respectively) no unsteadiness in the wake flow occurredat a Reynolds number of 0.35 million. H

12、owever, when the Reynoldsnumber for the 19 cone was increased to 0.85 million, the wake becameunsteady at an angle of attack of about 33. Comparison of the corre-sponding vapor-screen photographs indicated that the wake flow for thetest at the lower Reynolds number would probably have been unsteady

13、atam angle of attack about 2 to 4 above the maximum available for thismodel. Therefore, the experimental curve has been faired accordingly.The second curve included in figure l+(a)was obtained from reference 5and shows the theoretical variation with apex angle of the smallestangle of attack at which

14、 the laminar crossflow boundary layer separateswith the presumed formation of wake vortices. It is apparent from thetrends of the two curves in figure h(a) that increasing the apex anglesof cones and conical-nosed bodies (i.e., increasing the longitudinalrate of growth of the body cross section) inc

15、reases both the angle ofattack for the initial formation of the crossfluw vortices and theangle of attack at which the vortex flow becomes unsteady. For aparticular cone at the test Reynolds number of figure l(a), the lowestangle of attack at which the vortex flow became unsteady was about fivetimes

16、 the angle of attack at which the boundary layer theoreticallyseparates.The results for conical-nosed bodies of revolution have indicatedthat the apex e is a dominant factor in determining the nature ofthe wake vortex pattern. One might also,reason that for planar bodiessuch as triangula?-plan-formw

17、ings the apex singlemight also have animportant effect, and comparison of the wake flow for a body of revolu-tion, and a wing of similar plan-form apex angle would be of interest.Hence, a triangular-plan-formwing-body combination with a 12 apexangle (model 15) and an il.-l/2apex angle cone-cylinder

18、(model 3-C)were tested with similar mbient test conditions throughout the angle-of-attack range. For both models in the low angle-of-attack range, thevortex patterns were observed to be both steady and symmetric. Thevortex configuration for the cone-cylinder became asymmetric at anangle of attack of

19、 22; whereas the vortex wake for the triangular wingremained symmetric to an angle of attack about 2 or 3 greater. ForlIt should be noted that the datum point for model 5-II(8 apex angle .“cone-cylinder) does n“otfall on the correlation curve. The reason forthis deviation will.be discussed in a late

20、r section.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-16the singleof att=ck at which the vortex flowbecme Unstetiy ws greater than that for the equivalent cone. Thissecond result was evidenced by the data obtained from the tests withmodels 7-A an

21、d 19, circular-arc and parabolic-arc noses, respectively,where both the asymmetry and the unsteadiness in the wake occurred atlarger angles of attack than for the equivalent cones. The resultsobtained for models 8-B, 9-B, sad 1O-B (3/4-power, l/-power, and bluntHaack noses, respectively) showed that

22、 the wake flow was both symmetricand steady throughout the available angle-of-attack and Reynolds numberranges. On the other hand, the wake flow for the equivalent cone (mcdel2-B) became asymmetric at an angle of attack of about 300 and becameunsteady for the highest Reynolds number at an angle of a

23、ttack of about33.The foregoing results have shown that eansion of the forebody inthe vicinity of the body apex is an important geometric parameter, butnot necessarily the only factor which determines the nature of the wakevortex configuration.Expansion of the afterbcdy.- The effect of aa expanding a

24、fterbodyon the angle of attack at which the crossflow wake first becomes unsteadyis of interest for application to configurationswhere a rearward center-of-pressure position is achieved by the use of either a conical afterbody(e.g., ref. 7) or triangular-plan-formwings of low aspect ratio. Ithas bee

25、n shown that, for conical noses, the lsrger the axial rate ofincrease in cross-sectional area, the greater the angle of attack atwhich unsteady crossflow in the wake first occurs. Therefore, it wasexpected that a similar, although smaller, effect might be obtained if,for a given nose shape, various

26、rates of afterbody expamsion were used.Since expsnsion of the afterbody with distance from the body nose may besimply accomplished by the use of conical afterbcdiesj or by the additionof low-aspect-ratio, triangular-plan-formwings to a model, tests weremade using both methods of expansion. The model

27、s with conical after-bodies consisted of a 33-1/3-caliber ogival nose in combination withthree afterbodies (models 7-A, 7-E, and 7-F) in which the conicalexpsnsion angle of the afterbody varied from 0 to 8 as shown in figure2(a). The results of the tests with these bodies of revolution indicatedthat

28、 expanding the afterbciiyhad little effect on the wake vortex con-figuration. In fact, for alJ three of the tests the angles of attackat which unsteady wake flow first occurred were within about *l” ofeach other. (This is approximately the uncertainty in the test results.)The effect of afterbody exp

29、ansion utilizing low-aspect-ratio,triangular-plan-formwings in combination with a cylindrical afterbodywas investigated with models 13 and 14, the former having a conical noseand the latter am ogival nose. Vapor-screen photographs obtained forthese two models tith semiapex angle of the wings, A, equ

30、al to 4 areProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-8 NACA RMA53117.shown in figur”e5.3 These photographs show the wake vortex patterns atthree longitudinal stations: one near the juncture of the wing leadingedge and the body, the second about

31、 midway along the wing root chord,snd the third at the trailing edge of the wing. The pictures for themost forward stations show that the symmetricpair for the ogival-nosedmodel and the asymmetric configuration for the conical-nosedmodel wereestablished ahead of the wings on the noses of the models.

32、 The sequenceof photographs for the conical-nosedmodel indicates that the asymmetricstructure of the vortex pattern established on the nose remainedasymmetric at the downstream stations in spite of the presence of thewings. From a consideration of these results and the results with theconical.afrbod

33、ies, it appears that once a given vortex pattern isestablished on the nose of a body, expansion of the afterbody has littleinfluence on the symmetry of the wake v“ortexpattern.,The Effect of Body Cross-SectionalAqmmetryConsideration of two-dimensional-flowphenomena indicates that thenature of the fl

34、ow within the crossflowwake depends upon the manner inwhich the crossflow separates from the body surface. Also, sinceasymmetry of the body cross section (relative to the oncoming crossflow)can cause asymmetric sepm?atlon in the crossfluw plane, it was reasonedthat control bf both the symmetry and s

35、teadiness of the flow might beachieved by artificiallyfixing the crossflow separation lines alongeach side of the body. In order to establish separation and to investi-gate the effects of both symmetric and asymmetric separation on thewake vortex pattern, sharp-edged separation strips were installed

36、 alongthe entire length of an 8 cone-cylinder (model 16) as shown in figure2(c). Both the anglk of attack and the angle of rold.could be remotelycontrolled so about 36. -CONCLUDINGREMARKSThe results of the present investigationhave”brought out certainsalient facts which may be important to the desig

37、ner of aircraftrequired to fly at large angles of attack at supersonic speeds. Sincean unsteady vortex wake in the lee of a body may lead to undesirablecontrol characteristicsand to tail buffeting, the factors which affectthe behavior of the vortices are important. The results of the presentinvestig

38、ationindicate that both body shape and Reynolds number affectthe vortex configuration. Of the various body-shape variables, itappears that the shape of the nose has the greatest influence on thebehavior of the vortices. If the designer has to select one ofseveral nose shapes,which may be equally acc

39、eptable from other con-siderations, then, to minimize adverse effects of the body vortex wake,the nose shapes of lower fineness ratio and more blunt contour appesxdesirable. The addition of external longitudinal fairings (tunnels)or other protuberances,which result in a noncircular cross sectionyaff

40、ects the confuration of body vortices and may lead to undesirablerolling moments.Ames Aeronautical LaboratoryNational Advisory Committee for AeronauticsMoffett Field, Calif., Sept. 17, 1953REFERENCES1. Allen,Over2. Gowen,Body3. Seiff,H. Julian, and Perkins, Edward W.”: Characteristics of Flow - -Inc

41、lined Bodies of Revolution. NACARMA50L07, 1951.Forrest E.: Buffeting of a Vertical Tail on an Inclinedat SupersonicMach Numbers. NACARMA53A09, 19-53.Alvin, Sandabl, Carl A., Chapman, Dean R., Perkins, Edward W.,and Gowen, Forrest E.: Aerodynamic Characteristics of Bodies atSupersonic Speeds - A Coll

42、ection of Three Papers. NACA RM A5LJ25,1951.4. Mead, Merrill H.: Observations of Unsteady Flow Phenomena for anInclined Body-Fitted With Stabilizing Fins. NACA RM A51K05, 1952.,.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACARM A53117 135. Moore

43、, Franklin K.: Laminar Boundary Layer on Cone in SupersonicFlow at Large Angle of Attack. NACA TN 2844, 1952.6. Haack, W.: Projectile Forms for Minimum Wave Resistance. (Trans-lation) Douglas Aircraft Co., MC., Rep. 288, 1946.7. Ross, F. W., and Dorr=ce, W. H.: An Introduction to a SupersonicBody De

44、velopmental Study, University of Michigan. W-m,Dec. 1949.8. Roshko, Anatol: On the Development of Turbulent Wakes from VortexStreets. NACATN 2913, 1953.9. Cooper, Morton, Gapcyuski, JohnP., Hasel, Lowell E.: A Pressure-Distribution Investigation of a Fineness-Ratio-lZ.2 ParabolicBody of Revolution (

45、NACARM-1O) at M = 1.Z and Angles of AttackUp to 36. NACARML52Glka, 1952.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM A53117cProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-a71 a15 * J-.-.+.

46、A-18338Figure 1.- Schematic diagram of vapor-screenapparatusshowing vortices from a lifting bcdy ofrevolution.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-,.i7r Hoack nosesTde .f.87 .50.0300 19,05.00 11.5675Z15 n7.I5 8.011.455;5 f%180.01/i Otheth6

47、 Oogel11.455.75 f9;75.75 19.7%Cyf.Ion/colGy/.IbonicalI(a) Models.with conicsl and cylindrical afterbcd.ies.Figure 2.- Sketches of models., ., 1, IProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACARM A53117 254V.a=25Q= 29”a=33(a) Re=O.lhxlOeA-18344(b) Re = 0.69 x 106Figure 9.- Vapor-screen photographs showing the effect of Reynoldsnumber on the vortex pattern for an ogival nose with an expandngafterbody (model 7-F).vNAcA-LRJqley -12-l-53 - 925Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-