1、RESEARCH MEMORANDUM WIND-TUNNEL INVESTIGATION .TO DETERMINE TE3 AERODYNAMIC I I I I ! I 1; t L I 4 AT HIGH SUBSONIC SPEEDS By Richard E. Kuhn and James W. Wiggins E. i Langley Aeronautical Laboratory . Langley Field, Va. C“ RATIONAL d ADVISORY COMMITTEE B F.OR AERONAUTICS WASHINGTON January 21,1953
2、Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-1D NACA RM EX24 .- NATIONAL ADVISORY COMMITllEE - WIND-TUNNEL INVESTIGATION To DE- TRE AERODYNAMIC CKARAC-TICS IN STEADY ROLL OF A MODEL AT RIGH SUESONIC SPEEDS By Richazd E. Kuhn and James W. Wiggins A
3、erodynamic characteristics in steady roll were obtained in the Langley high-speed 7- by 10-foot tunnel on a complete model and its component parts. The wing and horizontal tail were swept back 45 at the quarter-chord line and had a taper ratio of 0.6, an aspect ratio of 4, and NACA 65AOO6 airfoil se
4、ctions parallel to the plane of symmetry. The ver- tical tail was swept back 5S0 at the qwter-chord line, had. a taper ratio of 0.5, an aspect ratio of 1.2, and an NACA 63(10)9 afrfoil section parallel to the fuselage center *e. investigation covered a Mach number-rarge from 0.40 to 0.95 and an angl
5、e-of-attack range from Oo to 6O. In general, the effects of Mach number were small and the over-all comparison of theory Hth the experimental rolling derivatives at Mach numbers below the force break was not greatly different from that which has been established at low speeds. The theoretical variat
6、ion of the damping-in-roll parameter with Mach nmiber at zero lift was in very good agreement with experiment, although the predicted variation with angle of attack and lift coefficient was only fair. The theoretical vm- iation of the slope of the curve of yawing moment due to rolling against lift c
7、oefficient Cnp/CL with Mach number was in good agreement with experiment up to the force-break Mach nlmiber, above which an abrupt reduc- tion in Cn,CL occurred. The predicted variation of the coefficient of yawing moment due to rolling Cnp wlth Uft coefficient was in excellent agreement with the ex
8、perimental data. Theoretical predictions of the coefficient of lateral force due to rolling Cyp were in poor agreement with experiment. The theoretical estimation of the effect of the rolling flow induced by the wing on the vertical-tail contribution to Cap was good, although somewhat too smallpztic
9、ulasly at the higher Mach numbers. c2P Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2 NACA RM L52KZ4 INTFiODUCTION A general research program is being carried out in the Langley high-speed 7- by ID-foot tunnel to determine the aerodynamic characte
10、r- istics in pitch, sideslip, and steady roll of various model configura- tions. This paper presents data obtained during steady-roll tests of a complete swept-wing model and its component pests. The win; and hori- zontal tail of the model were swept back bo at the quarter-chord lines and the vertic
11、al tail w-as swept back 55O at the quarter-chord line. The sting-mounted model was tested through a Mach number range from 0.40 to approximately 0.95 which gave a mean test Reynolds number range based on the mean aerodynamic chord of the wing from about 1.8 x 10 6 to approxi- mately 3.0 x lo6. Stati
12、c longitudinal stabilfty characteristics for the wing-fuselage conibinationof the present model are presented in reference 1. COEFFICIENTS AND smLs I The symbols used in the present paper we defined Fn the following list. All forces and moments are referred to the stability axes (fig. l), with the o
13、rigin at the quarter-chord point of the wing mean aerodynamic chord. CL lift coefficient, LWt/qS CD drag coefficient , Drag/qS C2 rolling-moment coefficient , Rolling moment/qSb CY lateral-force coefficient, hterd force/qS Cn yawing-moment coefficient , Yawing moment/qSb a speed of sound, ft/sec v f
14、ree-stream velocity, ft/sec M free-stream Mach number, V/a P air density, slugs/cu ft 9 dynamic pressure, pV2/2, lb/sq ft ! Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L52K24 3 L b S - C - C R . a ba P P wing area, sq ft local wing chord,
15、 ft wing mean aerodynamic chord, Reynolds nmiber based on c - angle of attack of wing, deg local angle-of-attack change due to aeroelastic Ustortion of wing, radians angle of sideslip, deg rolling angular velocity, rdians/sec -tip helix angle, radians correction factor for aeroelastic distortion asp
16、ect ratio, b2/S thickness ratio tail length; distance, measured parallel to fuselage center line, from moment reference point to center of pressure of vertical. tail, ft tail height; distance, measured normal to fuselage center line, from moment reference point to center of pressure of vertical tail
17、, ft c Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-4 GB=mRmm Subscripts and abbreviations: F fuselage v vertical tail H horizontal tail D measured dues L static loading NACA RM L52K24 L A three-view drawing of the test node1 and a tabulation of i
18、ts geometric characteristics me shown in figure 2. The wing and horizontal tail had an NACA 65006 airfoil section parallel to the plane of symmetry. The wing panels were of a composite construction, consisting of a steel insert with a bismuth-tin covering to give the section contour. The tail sectio
19、n and fuselage were constructed of slam Etlloy. A photo- graph of the model on the forced-roll sting-sqport system is shown in figure 3. Figure 4 shows a view of the complete support system used for the forced-roll tests. A schematic view of the forced-roll drive system is shown in figure 5. The mod
20、el was rotated about the x-axis of the stability axes system. The angle of attack was changed by the use of offset sting adapters as shown in figures 3 and 5. The model was driven by a constant-displacement reversible hydraulic motor, located inside the main sting body, which wm actuated by a variab
21、le-displacement hydraulic pmrp driven by a constant-speed electric motor. The rotational speed was measured by a calibrated microammeter that waa connected to a gear-driven direct-current generator mounted inside the main stfng body. Speed of rotation was varied by controlling the fluid displacement
22、 of the hydraulic pump, and the direction of rotation was changed by reversing the fluid flow through an arrangement of electrically controlled solenoid valves in the hydraulic system. The forces and moments, measured by an electrical strain-gage balance . incorporated inside the Model, were transmi
23、tted to the recording devices through an arrangement of brushes and slip rings. . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L52K24 5 TESTS AND CORRECTIONS The forced-roll tests were conducted in the Langley htgh-speed 7- by 10-foot tunn
24、el through a Mach number range from approxbmtely 0.40 to 0.95, and through an angle-of-attack range from 00 to 60. The wing- tip helix-angle (pb/2V) range, corresponding to a revolutions-per-minute range from -150 to 450, is presented in figure 6. The blocking corrections which were applied to the d
25、ynamic pressure and Mach number were determined by the method of reference 2. The size of the model cawed the tunnel to choke at a corrected Mach number of about 0.96. An investigation of the jet-boundary corrections to the rotary derivatives by the method of reference 3 indicated that these correct
26、ions are negligible. Jet-boundasy corrections applied to the lift were calcu- lated by the method of reference 4. There were no tare corrections available to apply to these data; however, the static tare tests conducted in connection with an unpublished investigation of the static lateral stability
27、characteristics of this model indicate the effect of the sting support to be very small. The support system deflected under load and these deflections, combined with any initial displacement of the mass center of gravity of the model from the roll ais, introduced centrifugal forces and moments when
28、the model was rotated. Corrections for these forces and moments were determined and have been applied to these data. The wing was known to deflect under load When the model was forced to roll, the opposing rolling moment distorted the wlng in such a mer as to reduce the angle of attack on the down-g
29、oing wing and increase the angle of attack on the up-go.ing wing. Accordingly, in 821 effort to cor- rect the measured data to correspond to the rigid case, a correction factor for the effect of this aeroelastic distortion on the rolling moment was determined with the aid of static loadin;s. The the
30、oretical spanwise load distribution due to roll of reference 5 was simulated by loading the wing at four spanwise points on the qumter-chord line. The change in angle of attack flu (fig. 7(a) was measured by dial gages at several spanwise stations in the chordwise plane parallel to the plane OP symm
31、etry. n equivalent linear variation of LU (fig. ?(a) ) was determined which corresponds to the angle-of-attack distribution produced by an increment of “tip helix angle A - . The corrected damping- in-rol1,coefficient can be written in terms of the measured values and this increment as follows e3 Pr
32、ovided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-6 where NACA Dl L52K24 where h/qCzL is the value at y = 3 (fig. 7(a) ). Aeroehstic effects On CYp and C were small and therefore neglected. nP “he angle of attack at the plane of symmetry has been correct
33、ed for the deflection of the model and support system under load. The variation of the mean test Reynolds number with Mach nuniber is presented in figure 8. h-esentation of Results The results of the investigation are presented in the following figures : Fipe Basic data . 9 and 10 CZP ll to 15 cnp 1
34、6 to 20 cyp 21 to 23 The basic data (frgs. 9 and 10) have not been corrected for aeroelastic distortion. The rotary dervLative8 in figure 9 are presented against angle bf attack at several Mach numbers; whereas, in figure 10, the derivatives are presented against Mach number at several angles of att
35、ack. A system of designating the various model configurations has been used and is defined however, the discrepancies apperent at the higher lift coefficients result in part from difficulties in establishing the experimental lift-curve slope at these lift coefficients. The high-speed free-roll data
36、of reference 8 and the low-speed data of reference 7 (wing No. 22) show similar variations. Tail contributions.- The contributiws of the vertical and horizontal tails to Czp are presented in figures 13 and 14 along with values predicted by the method of reference 9. The experimentally determined tai
37、l lift-curve slope and the locations of tail center of pressure used in the theoretical calculations were determined from unpublished static lateral-stabiuty data on the present model; however, calculations using the geometric aspect ratio and tail lengths indicated essentially the same results. The
38、 increment of Czp contributed by the tail surfaces is seen to be small and is adequately predicted. Complete model.- A comparison of the corrected experhental damping in roll Czp with predicted values for the complete model at several Mach numbers is presented in figure 15. Since the theory presente
39、d is a summation of the theoretical values from figures 12, 13, and 14, and . - Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-a - NACA RM L52K2.4 since the wing contribution to Czp is predominant, the veriation of (=ZP for the complete model with a
40、ngle of attack is quite similar to the wing-fuselage veriation presented in figure 12. Yawing Moment Due to Rolling Wing-fuselage.- A comparison of the wing-fuselage experimental and the wing-alone theoretical variation of Cnp/CL with Mach number is presented in figure 16. The experimental points of
41、 figure 16 were deter- mined from the slopes of the experimental data between zero and approxi- mately 0.1 lift coefficient. The theory of reference 10 is presented with the first term of equation (4) from reference u) corrected for the effects of compressibility by the method of reference 6. The ex
42、perimental data of reference 1 were used for evaluating the profile-drag contribution in accordance with equation (8) of reference 10. The predicted variation with Mach number is in good agreement with the experimental variation, although theory predicts somewhat more negative values. This discrep-
43、ancy may be largely due to the difficulties of determining the eqeri- mental variation of Cnp with lift coefficient because of nonlinearities even at the lowest lift coefficient (fig . 17) . An abrupt reduction in the magnitude of Cn CL occurs above the force-break Mach number (fig. 16). This reduct
44、ion probably results from the drag rise at zero lift and the decrease in the lift-curve 610pe at the higher Mach nmibers (ref. 1) and may possible be augmented by a loss of tip suction. PI Figure 17 presents a conrpmieon of the wing-fuselage experimental variation of C with lift coefficient and the
45、wing-alone theoretical variation, where theory includes the effects of both the induced and profile drag (ref. 10) . Excellent agreement is indicated at all Mach num- bers and lift coefficients. nP Tail contributions.- The contribution of the vertical tail to Cnp is presented in figure 18, along wit
46、h a comparison with theory (ref. 9) for wing-on and wing-off conditions. The tail lengths and tail lift- cur-ve slopes used In the theoretical calculations were determined from unpublished static lateral-stability data on the present model. In general, the agreement is considered good, although theo
47、ry somewhat underestimates the effect of the rolYng flow induced by the wing on the vertical tail, particularly at the higher Mach numbers. This underesti- mation is also indicated in the data presented in reference 9. Figure lg(a) shows comparison of tail center-of-pressure locations (given by the
48、length 2v and the height zv) as determined from static Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2D NACA RM L52K24 7 9 lateral-stability data and as indicated from simple geometric considera- tions. These center-of-pressure parameters are appli
49、ed in calculations of the vertical-tail contribution to Cnp in figure 19(b), and the results are compared with experiment. It is apparent that the predicted variations, using the center of pressure determined from experimental data, are in better agreement with exper-nt than the predicted varia- tions using the geometric centers of pressure. Complete m
