1、NASA TECHNICAL NOTE COMPARISON OF SEVERAL METHODS STABILITY DERIVATIVES FOR TWO AIRPLANE CONFIGURATIONS FOR ESTIMATING LOW-SPEED by Herman S, Fletcher Langley Research Center Hampton, Va. 23365 NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, 0. C. NOVEMBER 1971 Provided by IHSNot for Resal
2、eNo reproduction or networking permitted without license from IHS-,-,-ERRATA NASA Technical Note D-6531 COMPARISON OF SEVERAL METHODS FOR ESTIMATING LOW-SPEED STABILITY DERIVATIVES FOR TWO AIRPLANE CONFIGURATIONS By Herman S. Fletcher November 197 1 Page 10: Under the section entitled ltCZr (fig. 17
3、),“ delete the last sentence: “The equation given in reference 4 for the tail contribution to Clr appears to have a sign error, which causes the Clr Then, reword the entire section as follows: value to be too negative.“ Czr were in fair agreement with each other for the unswept- and swept-wing confi
4、gurations. Page 33: Replace figure 17 with the attached corrected figure. (fig. 17).- Figure 17 shows that the estimated values of Clr Issued October 1972 NASA-Langley, 1972 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TECH LIBRARY KAFB, NM 2. Gov
5、ernment Accession NO. T - 1. Repon No. NASA TN D-6531 4. Title and Subtitle COMPARZSON OF SEVERAL METHODS FOR ESTIMATING I 1111 I I rrr 3. Recipients utaiog NO. 5. Report Date November 1971 7. Author(s) Herman S. Fletcher 9. Performing Organization Name and Address 8. Performing Organization Report
6、No. L-7914 10. Work Unit No, 136-62-02-04 NASA Langley Research Center 2. Sponsoring Agency Name and Address 11. Contract or Grant No. 13. Type of Report and Period Covered Technical Note - . 7. Key -Words (Suggested by Author(s) ) Calculated and experimental stability Swept-wing and unswept-wing co
7、nfigurations Low speeds derivatives - - Washington, D.C. 20546 - 5. Supplementary Notes 18. Distribution Statement Unclassified - Unlimited .- - . . . - 6. Abstract Methods presented in five different publications have been used to estimate the low- speed stability derivatives of two unpowered airpl
8、ane configurations. had unswept lifting surfaces; the other configuration was the D-558-11 swept-wing research airplane. wind-tunnel data, and with flight-test data for the D-558-11 configuration to assess the rela- tive merits of the methods for estimating derivatives. that, in general, for low sub
9、sonic speeds, no one text appeared consistently better for esti- mating all derivatives. One configuration The results of the computations were compared with each other, with existing The results of the study indicated Provided by IHSNot for ResaleNo reproduction or networking permitted without lice
10、nse from IHS-,-,-COMPARISON OF SEVERAL METHODS FOR ESTIMATING LOW-SPEED STABILITY DERIVATIVES FOR TWO AIRPLANE CONFIGURATIONS By Herman S. Fletcher Langley Research Center SUMMARY Methods presented in five different publications have been used to estimate the low- speed stability derivatives of two
11、unpowered airplane configurations. had unswept lifting surfaces; the other configuration was the D-558-11 swept-wing research airplane. with existing wind-tunnel data, and with flight-test data for the D-558-11 configuration to assess the relative merits of the methods for estimating derivatives. On
12、e configuration The results of the computations were compared with each other, In general, it was found that all the methods gave reasonably accurate predictions Even in these instances, however, there was some for those derivatives which are attributed primarily to the wing and horizontal tail - ma
13、inly, the longitudinal derivatives. variation in the estimated horizontal tail and fuselage contribution to the pitching moments. There were large differences between some of the lateral derivatives computed by using the various estimation methods. Most of the differences can be traced to the esti-
14、mated vertical-tail effectiveness. A detailed comparison of tail-effectiveness estimates is not feasible because of differences in definitions of effective areas, span, interference effects, and so on. The results of this study indicate that, in general, for low subsonic speeds, no one text appeared
15、 consistently better for estimating all derivatives. INTRODUCTION Aerodynamic derivatives of airplanes are required for several types of analyses, such as stability calculations, motion response, and man-machine simulation. In order for the results of such analyses to be valid, it is necessary that
16、the derivatives be accu- rate. plete airplane, and these are There are three general methods of obtaining aerodynamic derivatives of a com- Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-(1) Analytical methods based on theory and on empirical relati
17、ons derived from accumulated wind-tunnel data (2) Wind-tunnel tests of the airplane or a model of the airplane (3) Analysis of flight data Each of these basic methods is subject to some limitations and interpretations. There are several documents available in which techniques are presented for estim
18、ating deriva- tives (e.g., refs. 1 to 5). derivatives from flight data. There are also several techniques available for extracting A recent publication (ref. 6) compared the stability derivatives of a Navion aircraft as determined by several textbook methods, from wind-tunnel tests, and from flight
19、data. There were large differences in some of the more important derivatives. for the differences in some instances were not identified. The reasons The present study was initiated to determine whether there are basic differences in the various published methods, to point out the differences found,
20、and to assess the rela- tive merits of the methods for estimating the low-speed stability derivatives. The study is based on computation by various methods of the derivatives for two specific airplane configurations for which much wind-tunnel data were available. Considerable flight data also were a
21、vailable for one of the configurations. In addition, some other comparisons are available for flight, wind-tunnel, and theoretical derivatives (for example, refs. 7 to 19). However, these references are different in scope and for other airplane configura- tions than those considered herein. SYMBOLS
22、The calculated, experimental, and flight -extracted derivatives are presented in the form of standard NASA coefficients and moments about the stability axes. Values are given in both SI and U.S. Customary Units. The measurements and calculations were made in U.S. Customary Units. The coefficients an
23、d symbols used herein are defined as follows: b span, meters (feet) - C mean aerodynamic chord, meters (feet) qV dynamic pressure at vertical tail, newtons per meter2 (pounds per foot2) q, free-stream dynamic pressure, newtons per meter2 (pounds per foot2) S wing area, meters2 (feet2) 2 Provided by
24、IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-t M time, seconds angle of attack of body reference line, radians sideslip angle, sin-l v, radians change in downwash angle with angle of attack change in sidewash angle at tail with change in sideslip angle Mach n
25、umber rolling velocity, radians per second pitching velocity, radians per second yawing velocity, radians per second free-stream velocity, meters per second (feet per second) velocity along Y-axis, meters per second (feet per second) wing-tip helix angle, radians yawing-angular-velocity parameter ,
26、radians lift, newtons (pounds) side force, newtons (pounds) rolling moment, meter-newtons (foot-pounds) pitching moment about center of gravity, meter -newtons (foot-pounds) yawing moment, meter -newtons (foot-pounds) lift coefficient, FL/q,S vco 3 Provided by IHSNot for ResaleNo reproduction or net
27、working permitted without license from IHS-,-,-rolling- moment coefficient, MX/q,Sb pitching-moment coefficient, My/q,Sc yawing-moment coefficient, MZ/q,Sb side-force coefficient, Fy/q,S pitching-moment-curve slope, aCm/aa: per radian angle-of-attack damping parameter, Wm/e, per radian rolling momen
28、t due to sideslip or effective-dihedral parameter, aCi/ap, per radian yawing moment due to sideslip or directional-stability parameter, aCn/ap, per radian side force due to sideslip, aCy/ap, per radian damping-in-pitch parameter, aCm/B,. 8 10 c . 11 % Cn p. 13 Clp . 14 cy 15 Cn 16 Czr . 17 6 Provide
29、d by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-RESULTS AND DISCUSSION The results are discussed in two separate sections: one related to longitudinal derivatives and the other to lateral derivatives. The wind-tunnel data are used as a basis for comparison
30、of the results obtained by the various methods of estimating deriva- tives. agreement between estimates and wind-tunnel results is poor, additional comparisons are made for various components of the configurations to try to identify the factors responsi- ble for the differences. The comparisons are
31、first made for the complete airplane configuration. If the Longitudinal Derivatives CL, (fig. 3).- The estimated values of CL, were within 10 percent of the wind- tunnel results (refs. 21, 22, and 23). (See fig. 3.) The wing is the primary contributor to this parameter, and the differences in estima
32、tes obtained from the various references can be traced to differences in the wing contribution. These can, in turn, be associated with small differences in suggested values of section lift-curve slope, neglect of taper- ratio effects, or the form of the equation for CL,. It appears that all the meth
33、ods used were about equally good for the two configurations of this study and that reference 5 was the best for the swept-wing configuration as is evident from a comparison of wind-tunnel data (refs. 22 and 23) and flight test data (ref. 24). It also appears from unpublished cal- culations that the
34、smaller total CL, values obtained for the swept-wing configuration by using references 1 to 4 are due partly to neglect of or incorrect wing-fuselage effects. - Cma (fig. 4).- All except reference 4 of the analytical methods predicted a small stable (negative) value of Cma for the unswept-wing confi
35、guration. (See fig. 4(a).) The wind-tunnel value was a small positive value. However, the data source (ref. 21) indi- cated that the experimental value probably was in error because of geometric asymme- tries and should have been neutrally stable Cm, = Cm, was from about 0.07 to -0.13 (fig. 4(a), wh
36、ich corresponds to a static-margin range of from about -0.02 to 0.03. computational methods were equally good for the unswept-wing configuration. . The range of computed values of ( O) This range is quite reasonable, and it appears that all the The estimated values of Cm, for the complete swept-wing
37、 configuration varied from -0.42 to -0.87. The wind-tunnel (ref. 23) and flight-extracted values (ref. 24) were about -0.65. Since there was such a large spread in the calculated values, additional curves are shown in figure 4(b) to isolate the causes of the differences. It can be seen that the prim
38、ary causes of the differences are in the estimated values of the fuselage and horizontal-tail contributions to Cm,. Additional factors which are not readily apparent in figure 4(b) but which also have some effect are differences in downwash (fig. 5) and 7 Provided by IHSNot for ResaleNo reproduction
39、 or networking permitted without license from IHS-,-,-Y interference factors. wind-tunnel and flight-test values (fig. 4(b). The value estimated by use of reference 3 came closest to the Cmq (fig. 6). - The primary contribution to Cmq comes from the horizontal tail, and a small increment is produced
40、 by the wing. The analytical methods of references 1 to 5 are all approximately the same when nondimensionalized in the same manner and, therefore, yield comparable results. suggestec! vaiues ef section lift-curve slope or neglecting taper-ratio effects in estimating the horizontal-taii lifi-curve s
41、lope. Interference effects, however, are responsible for the poor agreement for tile swept wing by using reference 5. There were no experimental values availa,ble for comparison with the estimated values. Differences which do occur are associated with (See fig. 6.) Cmb (fig. 7).- The primary contrib
42、ution to Cm . comes from the horizontal tail a! - also. they all yielded approximately the same result. (See fig. 7.) There were no experimental values available for comparison with estimated values. Since all the analytical methods used herein were based on the same reference, Cmq + Cm however, the
43、re was a source of values for the swept-wing configuration. These data were for free- flight tests with a model at M = 0.6 (ref. 25). The results from the model tests are in good agreement with the computations. For this study the computed values of the component coef- The combination parameter can
44、be (See fig. 8.) Lateral Derivatives (fig. 9).- The calculated data show large differences in the values of Cy as cyP B estimated by the methods of the various references. account for the major differences in the values obtained by the various methods. (See fig. 9.) The following factors Reference 1
45、: (a) No procedure is given to account for the end-plate effect of the fuselage on the vertical-tail effectiveness. fuselage contribution to horizontal tail on the vertical tail (end-plate effect) if the horizontal tail is located some- where other than at the base of the vertical tail. (b) No proce
46、dure is given to account for the . (c) No procedure is given to account for the effect of the cyP P Reference 3: No procedure is given to account for any fuselage contribution to Cy Reference 4: (a) No method is given to estimate the end-plate effect of the blunt-tail fuselage on the aspect ratio of
47、 the vertical tail. (b) No procedure is given to estimate the 8 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-end-plate effect of the horizontal tail on the vertical tail for a position of the horizontal tail other than at the extremities of the ve
48、rtical tail. both airplane conl“igu!ratims. Most or* the differences are associated with the estimated contribution of the vertical tail io C The low value obtained from the use of reference 4 for the swept-wifig coxfiguration is caused primarily by not properly accounting for the effects of the fus
49、elage and horizontal tail on the lift-curve slope of the vertical tail. (See fig. 10.) Cip (fig. 11).- The values of Ci estimated by use of ref3rences 3, 4, and 5 are much higher (more negative) than the wind-tunnel value (ref. 21) or the values estimated by use of references 1 and 2 for the unswept-wing configuration. Th
