^ RM L58E09 i?ffi£V, 'Miff:: <8£fS8Si-M < .'< ((8Hito«*M«-v»V.'"**4' i 3WMBflf r NACA RESEARCH MEMORANDUM WIND-TUNNEL INVESTIGATION OF THE AERODYNAMIC AND STRUCTURAL DEFLECTION CHARACTERISTICS OF THE GOODYEAR INFLATOPLANE ByBennieW. Cocke, Jr. Langley Aeronautical Laboratory Langley Field, Va. UNIVERSITY OF FLORIDA DOCUMENTS DEPARTMENT 1 20 MARSTON SCIENCE LIBRARY P.O. BOX 11 7011 GAINESVILLE, FL 32611-7011 USA NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WASHINGTON September 10, 1958 ( and 10, wing stall was reached at each speed and the value of C T obtained was reduced J_ij IHELX NACA RM L58E09 with increased speed. The C T mQV value of 1.0 at a r = 10 Ib/sq ft J_i^ IBcUX civ was reached at an a of approximately -2° and gave a load factor slightly- less than 2 with moderate wing deflections but no signs of wing failure. With the tunnel speed increased to approximately 71 mph /a = 12\, a run was made with angle of attack increased "by 1° increments, and when an angle of attack of approximately -5° was reached wing buckling occurred suddenly after approximately 30 seconds time had elapsed at this condition. The wing recovered quickly when load was reduced after buckling without any apparent damage, however observation of the wing behavior indicated that if a propeller had been installed and operating the wing would have been destroyed. As it was important to obtain the loadings for this condition, the run was repeated with angle of attack increased by increments of 0.25° up to the -5° attitude of which three load readings were taken prior to collapse. This information showed a steady increase in load with time with the fuselage attitude held constant thus indicating that stretch in the nylon fabric at high loadings was allowing the wing to increase effective attitude with respect to the fuselage. The last recorded load prior to wing buckle for this condition was 1,15^ pounds for a load factor slightly in excess of 2. No apparent damage to the airplane resulted from these first two buckling experiences; therefore, an additional buckling test was pro- gramed with more complete motion-picture and still-photographic cover- age for study of the rapid motions of the wing during the 2 or 3 cycles of buckling and recovery which the wing went through despite prompt shutdown of the wind tunnel. For this additional photographic run the airplane attitude was slowly increased from -10° to -5 and continuous movie coverage was taken as the wing loaded and buckled. In this sequence the rear wing-guy- cable patch on the lower surface of the left wing tore on the second buckling cycle and the wing contacted the engine and was punctured by the spark plugs and propeller shaft. Photographs showing the wing at the onset of buckling and just after puncture are shown in figure 5- Motion pictures and still photographs of the wing buckling showed a column-type failure inboard approximately midway between the fuselage and wing-guy-cable-attachment points with the wing folding inboard and moving up and in so as to bring the inboard wing sections well into the propeller disk area. Close study of photographs of the buckling runs made prior to the failure of the rear guy- cable patch and wing puncture showed further that the rear guy cable had fouled on the model support system during these runs and actually snub- bed the wing in its upward travel, thus probably preventing wing punc- ture due to contact with engine during the first buckling tests. Initial buckling in all cases occurred on the left wing panel which developed a slightly higher loading than the right panel. This load asymmetry, also indicated by the higher magnitude of the left-wing-guy-cable loads NACA RM L58E09 and by positive airplane rolling moments, was attributed at least in part to negative camber evident in the airfoil sections in the region of the right wing tip. Airplane with additional wing guy cables . - On the basis of the ■ basic tests it was desirable to modify the wing attachment system to improve the aerodynamic characteristics of the airplane at higher speeds and to improve its load-carrying ability and arrest its motions after wing failure. From studies of the data and photographs it was felt that the aeroelastic effects shown in the data were largely associated with the deflection of the inboard wing sections which resulted in increasing wing incidence with load. This increase was believed to produce an unfavorable downwash at the tail resulting in the loss in stability at higher loadings. As the wing failure was si mi lar to that of an eccen- trically loaded column it also was obvious that some additional restraint inboard should give higher load capability, while, at the same time, offering some possibility of improving stability. During the time used for patching the wing punctures, provisions were made for addition of two new guy cables on the lower surface of each wing panel. Attachment points for these cables were located on the wing at chordwise locations approximately 0.23c" and 0.60c~ at a span- wise point approximately k feet from the fuselage beneath the point where buckling was first observed. The front cable was rigged taut to take load and the rear cable was left slack to serve only to reduce wing motion should buckling occur. This approach was taken as it was felt that adding the cable near the center of pressure should provide the necessary restraint and offer the greatest chance of reducing unfavorable wing twist on the inboard wing sections. The results of the tests made at normal inflation pressure (7 lb/sc[ in.) with the additional guy cables installed (fig. 6) showed an appreciable improvement in the aerodynamic characteristics along with higher load capability. For this configuration the degree of instability resulting at the higher loadings is greatly reduced and the increase in lift-curve slope with speed was noticeably less. Also for the higher speed condition, wing buckling finally occurred only after a condition of intermittent stall of the left wing developed which produced a series of wing oscillations which were followed by buckling along a chordwise line approximately 2 feet out from the fuselage center line. For this configuration a lift load of approximately 1,300 pounds was recorded for an airplane attitude approximately 1° below stall (a = -3.I ). Stall and buckling occurred as the airplane attitude was further increased to -2.1°. Loads for this condition can only be estimated but it is reasonable to believe that a value of C T „ of at least 1.0 was L,max reached with the maximum load approaching 1,^0 pounds for a load factor of approximately 2.5- 8 NACA RM L58E09 Although buckling occurred at this condition, it is conceivable that the longer time lapse and the intermittent stall preceding buckling would be an effective warning of the approaching buckling boundary. It is also felt that by further modification to the wing-guy-cable and wing- root attachments, additional improvements in the stability and load limit could probably be obtained) however, such development was beyond the intended scope of this program. As an additional point of interest in connection with load char- acteristics of the pneumatic structure, a short series of tests was made with wing and fuselage pressures reduced from normal pressure to ascer- tain the possibility of maintaining flight near minimum speed in case of an emergency caused by loss of air pressure. The results of the reduced pressure tests (fig. 7) d-id- not show any drastic changes in aerodynamic characteristics which should rule out flight down to the lowest test pressure of 3 lb/sq in. At this pressure the wing did not buckle for the test speed with the maximum load factor reaching approxi- mately 1, and brief aileron and elevator control checks indicated that control could be maintained. Wing-Guy-Cable Loads and Deflection Studies Loads measured in the wing guy cables for the two cable configura- tions tested are summarized in figures 8 and 9 where the individual cable loads are plotted as a function of total configuration lift. Wing- deflection photographs for some of the more pertinent conditions are presented in figures 10 to l4. These photographs have airplane angle of attack and total configuration lift noted on each to permit correla- tion with the proper cable loads and aerodynamic data plots . Horizontal stripes on the vertical deflection target bar shown at the wing tips in the photographs are spaced 2 inches apart. The long stripe on the horizontal bar provided general horizontal reference. The cable-load data for the original configuration (fig. 8) indi- cate that the wing bending due to lift is primarily restrained by the front guy cables and the bending due to chordwise forces by the rear guy cables. At zero lift the rear-guy-cable load is therefore larger than the front-cable load at all speeds; but as lift is increased, the front-cable load increases rapidly and for the high loading condition (fig. 8(d)) the front-cable loads reach values over three times as large as the rear-cable loads. With this cable configuration the front-cable load was approximately twice the rear-cable load for the 1 g condition (550-pound lift) condition at all test speeds. Cable-load data for the modified cable system (fig. 9) show that the additional cable attached forward and inboard on the wing NACA RM L58E09 appreciably reduced the load in. the original front guy cable at the higher load conditions but had only a small effect on the rear- cable load. The maximum load reached on the inboard cable was approximately 200 pounds for the maximum loading condition (fig. 9( c )) which imposed loads of over 600 pounds on the outboard front wing cable. The inboard cable could probably have been made to carry more load by preloading; however, the additional cable as installed raised the allowable wing load to a value slightly above stall onset at maximum design speed (71 mph). If a higher allowable load is required, it is felt that some further improvement could best be obtained by adding another light guy cable at the new initial failure point which for the modified cable system was approximately 2 feet out from fuselage center line. The wing-deflection photographs for the two cable systems at all speeds tested (figs. 10 to ik) show only small differences in deflec- tion for the two cable systems at load conditions approximating steady level flight ( L = 550 pounds ) . At the higher loading conditions obtained at the higher speeds, however, the wing deflections are noticeably dif- ferent. For the original cable installation the deflection inboard is seen to build up with load (fig. 12) until failure is reached at a load just over 1,100 pounds. For the modified cable system at the same speed (fig. lh) the deflection inboard is less and at the higher loadings reached with this cable system the tip sections show a more pronounced deflection. Control Characteristics The static longitudinal characteristics of the airplane with the elevators deflected are presented in figure 15 for test speeds averaging ^-1 and 6k miles per hour. The data for these speeds chosen to represent minimum- speed and cruising- speed flight conditions, respectfully, show a marked change in static stability with speed but little change in the effectiveness of the elevator to produce trim. Comparison of the eleva- tor angles obtained for a given deflection of the control stick indicated somewhat lower response of control motion to a given stick motion at the higher speeds. Adequate control would appear to be available, however, and the loss of response is apparently the result of stretch within the semirigid control system at higher loads. The amount of twist occurring in the elevator control surface was small as may be seen in figure 15 • Longitudinal and lateral aerodynamic data obtained with the ailerons deflected for the same test velocities previously discussed are shown in figures l6 and 17. A reduction in rolling-moment coefficient for a given stick deflection is evident at the higher speed condition. This reduction apparently comes from aeroelastic effects and from a reduction in control motion with stick deflection at the higher loading condition. 10 NACA RM L58E09 The reduced control motion is most apparent for the up-aileron in all cases because the up-aileron is acutated by a simple bungee cord attached to the upper surface so that preset tension in this cord pulls the con- trol up when tension is relaxed in the lower actuating cable by deflec- tion of the stick. Rolling-moment coefficients higher than those shown for the condition at a =10 could have been obtained by utilizing full control travel; however, with the model rigidly mounted through the fuselage for these tests, a nondesign condition existed with the wing rolling moment applied with fuselage restrained. Maximum rolling- moment tests, therefore, were not made at the higher speed. SUMMARY OF RESULTS The results of tests on the Goodyear Inflatoplane may be summarized as follows: 1. The airplane, with original guy cables, was stable through stall for low speed conditions but for the higher speed conditions exhibited instability in the lift- coefficient range representing load factors greater than 1. 2. With the original wing-guy-cable configuration, wing stall occur- red without any wing buckling for test speeds to 6h mph ( load factor just under 2) but at approximately 70 mph wing buckling occurred with a load factor slightly higher than 2. 3. The longitudinal instability noted for the higher load factor conditions was apparently the result of increased wing incidence inboard due to growth and stretch in the nylon fabric. k. Wing buckling which occurred as a column type failure inboard approximately midway between the fuselage and wing -guy -cable attachment points resulted in the inboard wing sections folding upward in a manner to bring them within the propeller disk area. When wing puncture did not occur due to contact with the engine (without propeller), wing recovery from a buckled condition was instantaneous with load reduction and without apparent damage. 5. The addition of one wing guy cable attached on the lower wing surface at the point of initial buckling appreciably reduced the static longitudinal instability at higher speeds and allowed the airplane to reach stall at 70 mph before buckling occurred. Buckling followed the wing oscillations produced by stall. Maximum lift loads reached approximately 1,^00 pounds (load factor approximately 2.5) before stall. NACA RM L58E09 1:L 6. For all configurations studied, the wing behavior following buckling was such that an operating propeller would have struck and destroyed the wing. 7. Tests made with airplane inflation pressure reduced and air speeds considered minimum for maintaining level flight indicate that flight should be possible in an emergency for inflation pressures less than one-half the normal inflation pressure. 8. Elevator and aileron control characteristics were modified some- what by changes in speed due to flexibility in the structure and con- trol system; however, adequate control should be maintained throughout the design speed range. Langley Aeronautical Laboratory, National Advisory Committee for Aeronautics, Langley Field, Va., April 17, 1958. 12 NACA RM L58E09 a 3 o -p & O o o o o o C\J •H Ik NACA RM L58E09 H -4" I 1 •H > a) U V o -p 3 •H -P a a; 3 S H ft O •P oJ H •H P-4 NACA RM L58E09 15 (d) Main support yoke. L-57-3^95 (c) Tail support fitting. L-57-5^96 Figure 3.- Concluded. 16 NACA RM L58E09 to -p ft •H O o 3 H O 0) > H -el cd d fH •H -P CD h d cd CD rH > ? CD U en P d -p o cd o CD ••> d—* cd • h a ft -H o -p a 1 cd en H\ Ch ,Q &~ CD CD s% O en O ttl CD ft xi -p d o O -P td en H ■P -H M •H H h cd oj a -p a cd o d d o o -p •h cd d so fn O CD O • <; Ti H 0) I td r-\ d h cd M -P •H en M d O -P -d" -H 0) 1 cd o 3 IXA h 18 NACA RM L58E09 CM K\ 1 t- m 1 • tJ 0) H 2 0) a 01 H o o o 2 p' a) h > m -P aj I & Id M a •H t- O O , O a p Hh to o •iH -P CO •H ^ CD -P o cd § o o a •H crj 9 O ■H -P •H T3 < co CO CD h ft a o •H 3 I o u I cd > O ^ *&? 600 Total lift, lb (a) V = h\ mph; q = h.06 lb/sq, ft. Figure 8.- Variation of wing- guy- cable loads with airplane total lift for several wind velocities. Original configuration; normal inflation pressure (7 lb./sq. in.); controls neutral. 22 NACA RM L58E09 400 1 ± uu tut tut ttt tut rat tut rat rat a: rat tt 5: rat rat ^^m:ittdtt+ta::flTtttfatt -<3#7 ^ <## 600 ^^ /00 ° Total lift, lb (b) V = 5^ mph; q ay = 7-07 lb/sq ft. Figure 8.- Continued. NACA RM L58E09 25 SCO -Zoo 200 400 600 Total lift, lb (c) V = 6h mph; q ay = 10.15 lb/sq ft. Figure 8.- Continued. /ooo 2k NACA RM L58E09 600 500 L = 639 lb. Figure 11.- Continued. L-58-1638 *>k NACA RM L58E09 Upper camera; a = -h ; L = 7&5 lb. Lower camera; a = -k ; L = 765 lb. Figure 11.- Continued. L- 58- 1639 MCA RM L58E09 35 Upper camera; a = -3.1 ; L = 90^ lb. Lower camera; a = -3-1 ; L = 90^ lb, Figure 11.- Concluded. L- 58-1640 36 NACA RM L58E09 Upper camera; a = -6.8 ; L = 616 lb. Lower camera; a = -6.8 ; L = 616 lb. L- 58-1641 Figure 12.- Deflection study photographs for original configuration. V = 71 mph; q = 12.2 lb/sq. ft. av NACA RM L58E09 57 Upper camera; a = -6 ; L = 850 lb. Lower camera; a = -6 ; L = 85O lb. L-58-l6^2 Figure 12.- Continued. 58 NACA RM L58E09 Upper camera; a = -5 .75°} L = 890 lb. Upper camera; a = -5 -50°; L = 958 lb. Figure 12.- Continued. L-58-I6U3 NACA RM L58E09 39 Upper camera ; a = -5-35 } L = 1,011 lb. Lower camera; a = -5-1°; L = 1,100 lb. L-58-1644 Figure 12.- Continued. ko NACA RM L58E09 Upper camera; a = -4.75 . Figure 12.- Concluded. L- 58-1645 NACA RM L58E09 1+1 Upper camera; a = -4.9 ; L = 65O lb. Lower camera; a = -4.9°j 1 = 65O lb. L-58-1646 Figure 13.- Deflection study photographs with additional wing guy cables installed. V = 6k mph; q^ = 10.15 lb/sq. ft. 42 NACA RM L58E09 Upper camera; a = -3.1 ; L = 8^9 lb- Lover camera; a = -3.1 ; L = 8^9 lb. L-58-l6^7 Figure 13.- Continued. NACA RM L58E09 ^3 Upper camera; a = -2.2 ; L = 971 lb. Lower camera; a = -2.2 ; L = 971 lb. Figure 1J.- Continued. L-58-I6W 44 NACA RM L?8E09 Upper camera j a = -1.2°; L = 1,086 lb. Lower camera; a = -1.2 j L = 1,086 lb. L-58-164 - Figure 13.- Continued. NACA RM L58E09 k$ Upper camera; a = -0-3° J L = 1,126 lb. Lower camera; a = -0.3 ; L = 1,126 lb. Figure 13.- Concluded. L- 58- 1650 kG NACA RM L58E09 Upper camera; a = -6.8 ; L = 55^ lb. Lower camera; a = -6.8 ; L = 55^ lb. L-58-1651 Figure lh.~ Deflection study photographs with additional wing guy cables installed. V = 71 mph; q ay =12.4 lb/sq ft. NACA RM L58E09 kj Upper camera; a = -4.9°; L = 889 lb. Lower camera; a = -4.9 ; L = 889 lb. Figure lk.- Continued. L- 58-1652 48 NACA RM L58E09 Upper camera; a = -k.l°- L = 1,089 lb. Lower camera; a = -k.l ; L = 1,089 lb. Figure 1^.- Continued. L-58-1653 NACA RM L58E09 h9 Upper camera; a = -3.2 ; L = 1,300 lb. Lower camera; a = -3-2 ; L = 1,300 lb, Figure 1^.- Continued. L- 58-165^ NACA - Langley Field, Vj 50 NACA RM L58E09 Lower camera; a = -5 Upper camera; a = -3 . Figure lU.- Concluded. L-58-1655 NACA RM L58E09 51 -p to II J ,d -p ■H O O H d & O o a- 0) ,d -p O tQ O •H -P- tQ •H CD -p o o3 Xi o d 03 O a 1 to tQ tQ u Ph a o ■H -P a3 ^ t3 O u a; H (U =H O •P o cd o 1 • ■tf «•«. a> Xi xt ft a H rH O -4- a VO t_> *\ >> 1 ■p • •H i*\ O rH O H 0) %-4 !> g. >d •H d (X) NACA RM L58E09 55 : ::!• 1 ■ ::::-- " . — 1 ; : ~[~> ri : J :!.. -_ 1 _J_ K- >J ■ ... : :i ■Jff- .... .... ...4..,. TV\ ^fi ■r— "S. 1 :> -;;;;;. 3 '"■'. . .... ■■ ■ ■ ■ ; !.. ■ : ■ - — t— - — 1— ■ : 1 — ■■:: -•j-:: ^i : : ' j ! . "Vl M ^3 .... mi; 1 ■ : ■ .':':'-' : : : :: :ii 1 : ;:! : .■:. :::: : :::t:::: ::::J:::: i .-..,j- !:::■ . j ^ U S/ vl s| I smi&UUi iii^«ijLii«i Mtfmi iKBti i.nii.ii initiiiiVfiilirii ^ w Q Q y<^ ^ ^ b 5V a ^5 o ft •H S -P ^^ O ti o a d •H WD •H u o u ■p to ■H u (U -p o o & >^ P •H O o H 'd d •H 5 o P CO d •H >, Pa O O u a d o ■ d d O -H •H p a 1 O CQ o o^ d — -