1、ANDNASA TECHNICAL NOTE NASA TN D-7721I-N74-31506(NASA-TN-D-7721) WIND-TUNNELINVESTIGATION TO DETERMINE THE LOW SPEEDYAWING STABILITY DERIVATIVES OF A TWINIJET FIGHTER MODEL AT HIGH ANGLES OF Unclas,ATTACK (NASA)41 _HC$325 CSCL 01C H102 47881WIND-TUNNEL INVESTIGATIONTO DETERMINE THE LOW-SPEEDYAWING S
2、TABILITY DERIVATIVESOF A TWIN-JET FIGHTER MODELAT HIGH ANGLES OF ATTACKby Paul L. Coe, Jr., and William A. Newsom, Jr.Langley Research CenterHampton, Va. 23665N 276 .191NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. AIIGUST 1974Provided by IHSNot for ResaleNo reproduction or network
3、ing permitted without license from IHS-,-,-1. Report No. 2. Government Accession No. 3. Recipients Catalog No.NASA TN D-77214. Title and Subtitle 5. Report DateWIND-TUNNEL INVESTIGATION TO DETERMINE THE LOW- August 1974SPEED YAWING STABILITY DERIVATIVES OF A TWIN-JET 6. Performing Organization CodeF
4、IGHTER MODEL AT HIGH ANGLES OF ATTACK7. Author(s) 8. Performing Organization Report No.Paul L. Coe, Jr., and William A. Newsom, Jr. L-966410. Work Unit No.9. Performing Organization Name and Address 501-26 -04 -02NASA Langley Research Center 11. Contract or Grant No.Hampton, Va. 2366513. Type of Rep
5、ort and Period Covered12. Sponsoring Agency Name and Address Technical NoteNational Aeronautics and Space Administration 14. Sponsoring Agency CodeWashington, D.C. 2054615. Supplementary Notes16. AbstractAn investigation was conducted to determine the low-speed yawing stability derivativesof a twin-
6、jet fighter airplane model at high angles of attack. Tests were performed in a low-speed tunnel utilizing variable-curvature walls to simulate pure yawing motion.The results of the study showed that at .angles of attack below the stall the yawing deriv-atives were essentially independent of the yawi
7、ng velocity and sideslip angle. However, atangles of attack above the stall some nonlinear variations were present and the derivativeswere strongly dependent upon sideslip angle. The results also showed that the rolling momentdue to yawing Clr was primarily due to the wing-fuselage combination, and
8、that at anglesof attack below the stall both the vertical and horizontal tails produced significant contribu-tions to the damping in yaw Cnr. Additionally, the tests showed that the use of the forced-oscillation data to represent the yawing stability derivatives is questionable, at high anglesof att
9、ack, due to large effects arising from the acceleration in sideslip derivatives.17. Key Words (Suggested by Author(s) 18. Distribution StatementYawing derivatives Unclassified - UnlimitedAcceleration in sideslip derivativesHigh angles of attackF-4 Phantom jet model STAR Category 0219. Security Class
10、if. (of this report) 20. Security Classif. (of this page) 21. No. of Pages 22. Price*Unclassified Unclassified 39 $3.25For sale by the National Technical Information Service, Springfield, Virginia 22151Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-
11、WIND-TUNNEL INVESTIGATION TO DETERMINE THE LOW-SPEEDYAWING STABILITY DERIVATIVES OF A TWIN-JET FIGHTERMODEL AT HIGH ANGLES OF ATTACKBy Paul L. Coe, Jr., and William A. Newsom, Jr.Langley Research CenterSUMMARYAn investigation was conducted to determine the low-speed yawing stability deriv-atives of
12、a twin-jet fighter airplane model at high angles of attack. Tests were performedin a low-speed tunnel utilizing variable-curvature walls to simulate pure yawing motion.The results of the study showed that at angles of attack below the stall the yawingderivatives were essentially independent of the y
13、awing velocity and sideslip angle. How-ever, at angles of attack above the stall some nonlinear variations were present and thederivatives were strongly dependent upon sideslip angle. The results also showed thatthe rolling moment due to yawing Clr was primarily due to the wing-fuselage combina-tion
14、, and that at angles of attack below the stall both the vertical and horizontal tailsproduced significant contributions to the damping in yaw Cnr. Additionally, the testsshowed that the use of the forced-oscillation data to represent the yawing stability deriv-atives is questionable, at high angles
15、of attack, due to large effects arising from theacceleration in sideslip derivatives.INTRODUCTIONThe National Aeronautics and Space Administration is currently engaged in a broadresearch program designed to supply fundamental information in the areas of automaticspin prevention, inherent spin resist
16、ance, and development of theoretical techniques forstall/spin studies. A major requirement for such a research program is an understand-ing of aerodynamic phenomena at high angles of attack, including techniques for the meas-urement of these characteristics.Previous wind-tunnel studies of swept wing
17、s (refs. 1 to 4) have shown that the clas-sical dynamic stability derivatives of swept wings at high angles of attack require spe-cialized test techniques in order to identify derivatives due to pure angular rates (suchas rolling and yawing velocities) and derivatives due to linear accelerations (su
18、ch as rateof change of sideslip). The present investigation was conducted in order to determineProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-the dynamic yawing stability derivatives of a contemporary fighter-airplane configurationat high angles of
19、attack. The tests were conducted in a curved-flow wind tunnel whichpermitted the simulation of pure yawing motion rather than the combined yawing andsideslipping motion normally produced by other dynamic test techniques, such as theforced-6scillation test technique described in reference 5. The resu
20、lts of the presenttests are compared with the results of forced-oscillation tests previously conducted atthe Langley Research Center (see ref. 6) in which the same model was used.SYMBOLSAll aerodynamic data are presented with respect to the stability system of axes asshown in figure 1. Moment data a
21、re presented with respect to a center-of-gravity posi-tion of 33 percent of the wing mean aerodynamic chord. Measurements and calculationswere made in U.S. Customary Units and are presented herein in the International Systemof Units (SI) with equivalent values given parenthetically in the U.S. Custo
22、mary Units.b wing span, m (ft)c wing mean aerodynamic chord, m (ft)Ct horizontal-tail mean aerodynamic chord, m (ft)CD drag coefficient, FD/qSCL lift coefficient, FL/qSC1 rolling-moment coefficient, MX/qSbCm pitching-moment coefficient, My/qSECn yawing-moment coefficient, MZ/qSbC side-force coeffici
23、ent, Fy/qSFD drag force, N (lb)FL lift force, N (lb)FY side. force, N (lb)2Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-MX rolling moment, m-N (ft-lb)My pitching moment, m-N (ft-lb)Mz yawing moment, m-N (ft-lb)q free-stream dynamic pressure, N/m2
24、(lb/ft2)r yawing velocity, rad/secrbnondimensional yawing-velocity parameter2VS wing area, m2 (ft2)V free-stream velocity, m/sec (ft/sec)X,Y,Z stability axes (fig. 1)a angle of attack, degSangle of sideslip, degrate of change of sideslip angle, rad/secaC aC acCyC ap a YP apaC/ aCn aCyCl= Cn - CY/ a
25、b n a Ob Y0 aOb2V 2V 2VaC aC aCC C C _ anC _ CYr rb nr rb Yr rb2V 2V 2VModel component designations:F fuselage3Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-H horizontal tailV vertical tailW wingMODEL AND APPARATUSModelThe investigation was conduct
26、ed by using a 0.0915-scale model of a twin-jet fighterairplane. The model was primarily of fiber-glass construction with blocked inlets. Athree-view sketch of the model is presented in figure 2, and pertinent dimensional char-acteristics of the full-scale airplane are given in table I.Wind TunnelThe
27、 data presented herein were obtained in a low-speed tunnel (previously known asthe Langley stability tunnel) which has a 1.83- by 1.83-m (6- by 6-ft) curved-flow testsection. The tunnel was acquired by the Virginia Polytechnic Institute in 1958 and iscurrently operated at that institute. The tunnel
28、is used with a straight test section toobtain conventional static test data. The tunnel also has a unique capability in that thevertical sidewalls of the test section are designed with sufficient flexibility so that theymay be deflected into a curve, thus creating a curved airflow past the model. Ja
29、ckscrewsare positioned at regular intervals along each wall to allow the curvature to be set atprescribed values.In order to simulate flight in a curved path it is also necessary to redistribute thevelocity profile in the radial direction. This is accomplished by installing vertical wirescreens in t
30、he flow upstream of the test section. These screens vary in mesh across thewind tunnel, with the densest portion of the screens located nearer the center of curvature.A sketch showing a typical curved-flow test arrangement is shown in figure 3. A completedescription of the tunnel and its operation i
31、s given in reference 7.TESTS AND CORRECTIONSThe force tests were conducted at a Reynolds number of approximately 0.73 x 106,based on the mean aerodynamic chord of the wing. Tests in both curved and straight flowwere conducted for the fuselage-wing combination, the fuselage-wing-vertical-tail com-bin
32、ation, and the complete model. The model was sting mounted, and measurements weremade of the six force and moment components by using an internal strain-gage balance.4Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-The tests were conducted over an an
33、gle-of-attack range from -100 to 450 for an angle -of-sideslip range from -100 to 100 in straight flow. For the curved-flow tests the angle-of-attack range was 00 to 450 for an angle-of -sideslip range from -50 to 50. Three cur-vatures representing yawing flight to the left were selected for the cur
34、ved-flow tests andresulted in values of the nondimensional yawing velocity rb/2V of -0.0327, -0.0483, and-0.0637.Experimentally determined blockage corrections have been applied to the databecause of the relatively large size of the model in relation to the test-section area.Additionally, the side-f
35、orce coefficients have been corrected to account for the radialpressure gradient in the tunnel in accordance with the method of reference 7.RESULTS AND DISCUSSIONResults of Straight-Flow Static TestsLongitudinal characteristics.- The variations of the static longitudinal aerodynamiccharacteristics o
36、f the model with angle of attack are shown in figure 4. These data showthat the onset of wing stall occurred at approximately a = 150, with a gradual well-definedstall at higher angles of attack.Comparison of the data with and without the horizontal tail (fig. 4(a) indicates thatthe horizontal tail
37、remained effective in providing static longitudinal stability throughoutthe angle-of-attack range tested. Presented in figure 4(b) are data from reference 8which were obtained with the same model at approximately the same value of Reynoldsnumber, but in a 3.66- by 3.66-m (12- by 12-ft) octagonal tes
38、t section of a low-speedwind tunnel at the Langley Research Center. The data are in relatively good agreement,but variations in Cm were noted at higher angles of attack. The close proximity of thehorizontal tail to the tunnel floor in the present investigation probably contributed to thedifferences
39、shown.Lateral-directional characteristics. - The variation of the static lateral-directionalcharacteristics of the model with angle of sideslip is presented in figure 5 for the variousairframe component combinations tested. The data show that up to an angle of attack of350 the variation of Cy, Cn, a
40、nd C1 with 0 is generally linear over a 0 range from-50 to 50. When the 3 range was extended to -100 to 100, the variation of Cy, Cn,and C1 with 0 generally became nonlinear with the greatest nonlinearity at a = 200The variation of static lateral-directional stability derivatives with angle of attac
41、kobtained over a range of sideslip angles from -50 to 50 is summarized for the variousmodel components in figure 6. The data for the wing-fuselage combination, the wing-fuselage -vertical-tail combination, and the complete model show a loss of effective dihe-dral and a rapid decrease in directional
42、stability at the stall.5Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-As pointed out in reference 8, the loss of effective dihedral at the stall is associ-ated with flow separation which causes reduced lift on the leading wing in a sideslippedcondi
43、tion. The analysis of reference 8 also shows that the factors producing the loss ofdirectional stability at angles of attack near the stall are a loss of dynamic pressure atthe vertical tail and an adverse sidewash field produced by flow separation on the wing-fuselage combination. As the angle of a
44、ttack is increased above the stall, the verticaltail enters this adverse sidewash field which results in a complete loss of vertical-taileffectiveness as shown by the data of figure 6(a).Shown in figure 6(b) is a comparison of the static lateral-directional derivatives asmeasured during the present
45、study with those determined from the data of reference 6.Although the trends shown by the data are in fairly good agreement, some differences inthe magnitudes of the derivatives occurred at high angles of attack.Results of Curved-Flow TestsLongitudinal characteristics.- The longitudinal aerodynamic
46、characteristics of themodel, obtained in the curved-flow tests, did not differ to any significant extent from thedata in the straight-flow tests, and as a result the data are not presented herein.Lateral-directional characteristics.- The variations of the static lateral-directionalaerodynamic coeffi
47、cients Cy, Cn, and C1 with nondimensional yawing velocity rb/2Vat 0 = 00 and 3 = 50 are presented for the fuselage-wing combination, the fuselage-wing-vertical-tail combination, and the complete configuration in figures 7, 8, and 9,respectively. These data are faired by using a least-squares linear
48、curve fit. The vari-ations of Cy, Cn, and Cl with rb/2V are shown to be relatively linear for angles ofattack up to the onset of stall; however, as the angle of attack was increased above thestall some nonlinearities are noted.The stability derivatives CYr, Cnr, and Clr obtained by using the least-s
49、quareslinear curve fit over the range of nondimensional yawing velocities from 0 to -0.0637, forthe data of figures 7 to 9, are summarized in figures 10 and 11.Analysis of the data of figure 10 indicates that the magnitude of the rolling momentdue to yawing derivative C/r was primarily due to the wing-body combination, since thehorizontal and vertical tails provided only small change
