Abbot and Doenhoff Naca Airfoil Data

CONTENTS Susser... Twrmopverion. Srunoca. Hisronicat Devevorues. Descapriox o7 Atarouta, Method of Combining Distributions... NACA Four-Digit Series Airfo ‘Numbering system. ‘Thickness distributions. Mean lines NAGA Five-Digit Series Airfolls__ ‘Numbering system. ‘Thickness distributions. Mean lines. NAGA I-Series Aifols Nombering systems. ‘Thickness distributions. Mean ines. NAGA 6-Series Airfola. ‘Numbering system_ ‘Thickness distributions Mean Hines. NAGA T-Series Aipfoll.- ‘Numbering system. ‘Thickness distributions... ‘Terorsricat, ConsiDenarioxs. Pressure Distributions. Rapid estimation of pressure distributions. Numerical examples. Effet of camber on pressure distribution Critical Mach Number. Namerical examples... Angle of Zero Lit. ‘Methods of ealeulation. ‘Numerical examples. Description of Flow around Alefoils, Exprariextan CHAmacremerie, Sourees of Data... Drag Characteristics of Smooth Aifols Drag characteristics ia low-drag range Drag characteristics outside low-~, fhe, and 250-series airfoils. The general methods ‘used to derive the basic thickness forms for NACA 6- and Teseries airfoile and their corresponding pressure distributions ‘are presented. Data and methods are giten for rapidly obtain ing the approximate pressure distributions for NACA four digit, fee-digit, 8-, and T-series airfoils. The report includes an analysis of the lift, drag, pitching- moment, and critical-speed characteristics of the airfoils, together with a discussion of the effects of surface conditions. Data on hightift devices are presented. Problems associated ‘with lateral-control decices, leading-edge air intakes, and interference are briefly discussed. The data indicate that the effects ‘of surface condition on the lift and drag characteristics are at least as large as the effects of the airfoil shape and must be considered in airfoil eelection and the prediction of wing charae- teristics. Airfoile permitting extensive laminar flow, such as the NACA 6-series airfoils, have much lower drag coeficiente at high speed and eruieing lift coeficients than earlier types of airfoils if, and only if, the wing surfaces are sufficiently smooth and fair. The NAGA 6-series airfoils also have facorable ceritical-speed characteristics and do not appear to present ‘unusual protlema associated with the application of Kigh-lift and lateral-control devices. INTRODUCTION A considerable amount of airfoil data hes been accumulated | from tests in the Langley two-dimensionel low-turbulence ‘tunnels. Date have also been obtained from tests both in other wind tunnels and in flight and include the effects of high4ift devices, surface irregularities, and interference. ‘Some data are also available on the effects of airfoil section on aileron characteristics. Although a large amount of these data has been published, the scattered nature of the data and the limited objectives of the reports have prevented adequate analysis and interpretation of the results. The purpose of this report is to summarize these data and to correlate and interpret them insofar as possible. Recent information on the aerodynamic characteristics of NACA airfoils is presented. The historical development of NACA eirfoils is briefly reviewed. New data are presented that permit the rapid calculation of the epproximate pressure distributions for the older NACA four-digit and five-digit airfoils by the same methods used for the NACA 6-series airfoils. ‘The general methods used to derive the basic thickness forms for NACA 6- and 7-series airfoils together with their corresponding pressure distributions are presented. Detail data necessary for the application of the airfoils to wing design are presented in supplementary figures placed at the end of the paper. The report includes an analysis of the lift, drag, pitching-moment, and critical-speed charac teristics of the airfoils, together with e discussion of the effects of surface conditions. Available data on high-lift devices are presented. Problems associated with lateral- control devices, leading-edge air intakes, and interference are briefly discussed, together with aerodynamic problems of application. Numbered figures are used to illustrate the text and to present miscellaneous date, Supplementary figures and tables are not numbered but are conveniently arranged at the end of the report according to the numerical designation of the airfoil section within the following headings: I—Basic Thickness Forms T1—Date for Mean Lines TH —Airfoil Ordinates IV—Predicted Critical Mech Numbers ‘V—Acrodynamic Characteristics of Various Airfoil Sections ‘These supplementary figures and tables present the basio date for the airfoils. SYMBOLS A aspect ratio ‘Ay, Ba Fourier series coefficients a mean-line designetion, fraction of chord from leed- ing edge over which design load is uniform; in derivation of thickness distributions, basie length usually considered unity 6 ‘ing span by Hapspan, inboard b, flap span, outboard Co drag coefficient C,,_, drag coefficient at zero lift, 7," lift coefficient 4C;, increment of meximum lift eaused by flap deflection 250Aen Acad a ey Oma ema Ov, 0 ®, @, % REPORT NO. 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS chord aileron chord section drag coefficient minimum section drag coefficient, fap chord, inboard flap chord, outboard flap-chord ratio tection aileron hinge-moment coeficient (<4) increment of aileron hinge-moment coefficient at constant lift hhinge-moment perameter section lit coeficient design section It coefficient moment coeficient about aerodynamic center moment coeficient about quarter-cord point section normal-foree coeficient drag Joss of total pressire fee-stream total pressure section aileron hinge moment exit height constant re Mach number critial Mach number typical points on upper and lower surfaces of airfoil presoure coefficient (&B) critical pressure cocficient| resultant pressure coeficient; difference between local upper- and lower-surface pressure coefficients local static preesure; also, angular velocity in rollin pb)2V freestream static pressure halix angle of wing tip freostream dynamic pressure Reynolds number czitieal Reynolds number est (Ho? pressure coeficient ( 5 ) frst aitfol thickness ratio second eirfoil thickness ratio freo-atream velocity inlet velocity Tocal velocity increment of local velacity inerement, of local velocity enused by additional type of load distribution velocity ratio corresponding to thickness f, velocity ratio corresponding to thickness f distance long chord meancline abscissa abscissa of lower surface abscissa of upper surface chordwise position of transition distance perpendicular to chord ©, Ye meancline ordinate ti ordinate of lower surface vi: ordinate of symmetrical thickness distribution Yo ordinate of upper surface 2 complex variable in circle plane % complex varinble in noarirele plane @ angle of attuck 420 section aileron effectiveness parameter, ratio of change in seetion angle of attack to inerement of aileron deflection at a constant value of lift coefficient ay angle of zero lift section angle of attack ‘Bae increment of seetion angle of attack ai section angle of attack corresponding to design lift coefficient 3 flop or aileron deflection; down deflection is posi 3, flap defection, inboard %, Bap deflection, outboard ‘ airfoil parameter (¢—€) fer value of eat trailing edge r complex variable in airfoil plane &@ angular coordinate of 2’; also, angleof whieh tangent is slope of mean line taper ratio (gar ion Effective Reynolds nuinber' + turhulene factor (MST Reysolia number) 6 angular coordinate of ¢ ¥ airfoil parameter determining radi! coortinate of 2 Ye avorage value of ¥ (jt: ("" va) HISTORICAL DEVELOPMENT ‘The development of types of NACA airfoils now in com- ‘mon uso was started in 1029 with a systematic investigation of a family of airfoils in the Langley varinbledensity tumel. this family hed the same basic thickness distribution (reference 1), and the amount and type of camber was cally varied to produce the family of related airfoil investigation of the NACA airfoils of the four-tigit. scrics produced airfoil sections having higher maximum Tift coefficients and lower minimum drag cocfficients than those of sections developed before that, time. The also provided information on the ehunges in acrodynamie characteristics resulting from variations of gcometry of the ‘mean line and thickness ratio (reference 1).SUMMARY OF AIRFOT DATA ‘The investigation was extended in references 2 and 3 to include airfoils with the same thickness distribution but with positions of the maximum eamber fer forward on the airfoil. These airfoils were designated by numbers having five digits, such as the NACA 23012 airfoil. Some airfoils of this family showed favorable aerodynamic characteristics exept for a lange sudden loss in lift at the stall. Although these investigations were extended to include a limited number of airfoils with varied thickness distributions (references 1 and 3 to 6), no extensive investigations of thickness distribution were made. Comparison of experimental drag deta at low lift coefficients with the skin friction coefficients for flat plates indicated that nearly all of the profile drag under such conditions was attributable to skin friction. It was therefore apparent that any pronounced reduction of the profile drag must be obtained by a reduction of the skin friction through increasing the relative extent of the laminar boundary laye Decreasing pressures in the direction of flow and low airstream turbulence were known to be favorable for laminar flow. An attempt was accordingly made to increase the relative extent of laminar flow by the development of air foils having favorable pressure gradients over @ greater proportion of the chord than the airfoils developed in references 1, 2, 3, and 6. The actual attainment of extensive laminer boundary layers et lenge Reynolds numbers was @ previously unsolved experimental problem requiring the development of new test equipment with very low air stream turbulence. This work was greatly encouraged by the experiments of Jones (reference 7), who demonstrated the possibility of obtaining extensive laminar layers in flight at relatively large Reynolds numbers. Uncertainty with regard to factors affecting seperation of the turbulent boundary layer required experiments to determine the possibility of making the rather sharp pressure recoveries Tequired over the rear portion of the new type of airfoil. New wind tunnels were designed specifically for testing jrfoils under conditions closely approaching flight conditions of airstream turbulence end Reynolds number. The resulting wind tunnels, the Langley two-dimensional low- turbulence tunnel (LTT) and the Langley two-dimensional low-turbulence pressure tunnel (TDT), and the methods ‘used for obtaining end correcting data are briefly deseribed in the appendix. In these tunnels the models completely span the comparatively narrow test sections; two- dimensional flow is thus provided, which obviates difficulties previously encountered in obtaining section data from tests of finitespan wings and in correcting adequately for support interference (reference 8). Difficulty was encountered in attempting to design air foils having desired pressure distributions because of the lack of edequate theory. ‘The Theodorsen method (reference 9), as ordinarily used for calculating the pressure distributions about airfoils, was not sufficiently accurate near the leading edge for prediction of the local pressure gradients. In the absence of e suitable theoretical method, the 9-percent- thick symmetrical airfoil of the NACA. 16-series (reference 10) s4a:07—90—18 261 was obtained by empirical modiffeation of the previously used thickness distributions (reference 4). These NACA 16-series sections represented the first family of the low- > yunse0o bosp wasces @ 16 “eo a Section angle of ottook, ey og ‘roves 17-Comparto othe seryeal ance fa ACA aris fon wt ls the Langley tweens ow-tarbalane pra ane, “16 “26 Seas > ON 19 quaeayjoc 44 LOND2S ~ 8 F 8 2 F 4 "> wstayj902 suewON 36 22 28 24 20]SUMMARY OF AIRFOTL DATA Bective aspect retio.—The combination of high dregs at, high lift coefficients, low drags at moderate lift coefficients, and the nonreguler variation of drag with lift coefficient shown by the NACA 6-series airfoils may lead to pare- doxical results when the span-efficiency concept (reference 29) is used for the calculation of sirplane performence. In the usual application of this concept, the airplane drag characteristics are approximated by a curve of the type Co=Copay ERO? an ‘This curve is usuelly matched to the actual drag characteristics at a rather low and at a moderately high value of the lift coefficient (reference 30). The application of this concept to two hypothetical airplanes with NACA 230- end 65-series sections, respectively, is illustrated in figure 18 (a). ‘The wing drags of the airplanes have been calculated by adding the induced drags ‘corresponding to an aspect ratio of 10 with elliptical loeding to the profiledrag coefficients of the NACA 23018 and 65-418 airfoils. These sections are considered. representative of average wing sections for e large airplane of this aspect ratio, Ordinate scales are given in figure 18 (a) for the wing drag end for the total airplane drag cocflicients obtained by adding © representative constant value of Di Spares aaelatal Ses Sie nonce at 2 ie Beane a ae — wets oe 7 Sie et gmk = 3S Las a s Bo 048 308 axaomna pane pal Sars a @ er ae cena Gre ene LM coercion (a) NACA ips and 08 ngs of pect te 1. 279 0.0150 to the wing drag coeficients. ‘The resulting dreg coefficients have been approximated by two curves corre sponding to equation (17) and metched to the drag curves at lift coefficients of 0.2 and 1.0. These two curves correspond to effective aspect ratios of 9.29 for the airplane with NACA 23018 sections and of 8.30 for the airplane with NACA 65,418 sections and illustrate the typical large reduction in the effective aspect ratio obtained with such sections. Tt should be noted, however, that although equation (17) provides a reasonably satisfactory approximation to the drag of the airplane with NACA 23018 sections, such is not the case for the airplane with the NACA 65,-418 section. ‘The most important reason for using high aspect ratios on lange airplanes is to reduce the drag at cruising lift coefficients and to obtain high meximum values of the lift-drag ratio. For the two wings considered, the maximum value of this _ ratio is appreciably higher for the eixplane with NACA 65,-418 sections (19.8 as compared with 18.5) despite the fact that this airplene shows the lower effective aspect ratio. Figure 18 (b) shows « similer comparison with similer results for two airplanes of aspect ratio 8 and NACA 2416 and 65,415 airfoils. It is accordingly concluded that the effective aspect ratio is not a satisfactory criterion for use in airfoil selection. ° JF Bae abate “ 2 NACA GMS wing ospect rota, 8° oo Le Mel Bares eho eth eet ro — rota, a7 || effective aspect roti, 146. fe a, got ay as 8 Fi Pe. ae 8 GL gr toass ane a (5) = k oes hose af mew Sas rae 01] I eS ber e o » 0 2 4 6 8 10 Lint coohicient,C. (0) NACA thet and 288 wing apt tos, iocer Comper of foltespecato drag starsat fr tro ixpes of als obtained by ang tbe Induced dng eureazondlog te 6 ots! span nding toe setn dag een[BEFECT OF SURFACE IRREGULARITIES ON DRAG Permissible roughness—Previous work has shown large drag increments resulting from surface roughness (reference 31). Although e large pert of these drag increments was shown to result from forward movement of transition, substantial drag increments resulted from surface roughness in the region of turbulent ow. It is accordingly important to maintain smooth surfaces even when extensive laminar flow cannot be expected, but the gains that may be expected from maintaining smooth surfaces are greater for NACA 6- or 7-series airfoils when extensive laminar flows are possible. No accurate method of specifying the surface condition necessary for extensive latninar flow at high Reynolds numbers has been developed, although some general conclusions have been reached. It mey be presumed that for a given Reynolds numbet and chordwise position, the size of the permissible roughness will vary directly with the chord of the airfoil. It is known, at one extreme, that the surfaces do not have to be polished or optically smooth. Such polishing or waxing has shown no improvement, in teste in the Langley two-dimensional low-turbulence tunnels when applied to satisfactorily sanded surfaces. Polishing or waxing 2 surface that is not aerodynamically emooth will, of course, result in improvement and such finishes may be of considerable practical value because deterioration of the finish may be easily seen and possibly postponed. Large models having chord lengths of 5 to 8 feet tested in the Langley two- dimensional low-turbulence tunnels are usually finished by sanding in the chordwise direction with No, 320 earborindum paper when an aerodynamically smooth surface is desired. Experience has shown the resulting finish to be satisfactory fat fiight values of the Reynolds number. Any rougher surface texture should be considered as a possible source of transition, although slightly rougher surfaces have appeared to produce satisfactory results in some cases. Windl-tunnel experience in testing NACA 6-series sections and data of reference 32 show that. smell protuberances extending above the general surface level of an otherwise satisfactory surface are more likely to cause transition than small depressions. Dust particles, for example, are more effective than small scratches in producing transition if the material at the edges of the scratches is not forced above the general surface level. Dust particles adhering to the oil left on airfoil surfaces by fingerprinls may be expected to cause transition at high Reynolds numbers. Transition spreads from an individuel disturbance with an included angle of about 15° (references 31 and 33). A fow scattered specks, especially near the leading edge, will cause the flow to be largely turbulent. ‘This fact makes necessary fan extremely thorough inspection if low drags are to be realized. Specks suificiently large to cause premature transition on full-size wings can be felt by hend. ‘The is spection procedure used in the Lengley two-dimensional Jow-turbulence tunnels is to feel the entire surface by hand after which the surface is thoroughly wiped with @ dry cloth. Tt has been noticed that transition resulting from individual KEPORT NO. 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS small sharp protuberances, in contrast to waves, tends to occur at the protuberance. ‘Transition caused by surface waviness appears to approach the wave gradually as the Reynolds number or wave size is increased. The height of a small cylindrical protuberance necessary to canse trans tion when located at 5 pereent of the chord with its axis normal to the surface is shown in figure 19. ‘These data were 050 t- 5800 Bag a ee Cree aers eremasersaelan ean ‘Wing Reycolds mambar, Ft Ficene 18--Vasatioa wit vlog Reynlas utero the mlaoium bh oe xine rotate neasary to eutve promatre canon Pretberae aw wba ‘Sue wit ass marmal to winged load a percent urd oles Syntace lection of I-perta thchess and wit ini Pe a 70: show a large decrease in allowable height Reynolds number. This effect of Reynolds number on permissible surface roughness is also evident in figure 20, jn which a sharp inerease in drag at a Reynolds number of approximately 20%10* occurs for the model painted with camouflage lacquer. ‘The magnitude of the favorable gradient appears to have a small effect on the permissible surfaee roughness for laminar flow. Figure 21 shows thut the roughness becomes more important at the extremities of the lowaleg range where the favorable pressure gradient is reduced on one surface. ‘The effect of increasing the Reynolds number for a surface of marginal smoothness, which has an effect similar to increasing the surface roughness for a given Reynolds number, is to reduce rapidly the extent of the lowairag range and then to inerease the minimum drag coefficient (fig. 21). The date of figure 21 were specially chosen to show Uhis effect. Tn most cases, the effect of Reynolds number pre- dominates over the effect of deercasing tie magnitude of the favorable pressure gradient to such an extent that the only effect is the elimination of the low-dreg range (reference 34). Permissible waviness.—Moro difficulty is generally encountered in reducing the waviness to permissible values for the maintenance of laminar flow than in obtaining the quired surface smoothness. In addition, the specification of the required freedom from surface waviness is more difficult. than that of the required surface smoothness. ‘The problem is not limited morely to finding the minimum wave size that will cause transition under given conditions because the number of waves and the shape of the waves require consideration.SUMMARY OF AIRFOIL. DATA Sector wy soe § a a es Reyrotds number, A 281 a Remnotds raster, R a we Baa (a moot onton: TOT east 28 (8) Coogee camoigeunsgrove alr paatng, TD tt rocax m\—Vasation of dra etl wt Reynolds number for 8 Odnchshard model ofthe NACA Wan abl to rae conten. oT TTT Tt rt SF bela sto" sno epncttion e 0/49 ee Ravel 24¢ | Syntnetic enamel conoutioge nih LOTR SE | ct Sects cur off mh ie 4 i 2 oy We ¥ & as B16 20 o# Stction'tit coefficient, @ 2.—Drag charter of NACA tSi-43) all for to ste eons, ‘Tor tees am ma wo. If the wave is sufficiently lange to affect the pressure distribution in such e manner thet laminar separation is encountered. there is little doubt that such a wave will cause premature transition at all useful Reynolds numbers, A re- lation between the dimensions of a wave and the pressure distribution may be found by the method of reference 35. ‘The size of the wave required to reverse the favorable pressure gradient increases with the pressure gradient. Large negative pressure gradients would therefore appear to be favorable for wavy surfaces. Experimental results have shown this conclusion to be qualitatively correct. Little information is available on waves too small to cause laminar seperstion or even reversal of the pressure gradient. Data for an airfoil section having a relatively long wave on the upper surface are given in figure 22. Marked increases in the drag corresponding to a rapid forward movement of the transition point were not noticeable below a Reynolds number of 44X10". On the other hand, transition has been caused at comparatively low Reynolds numbers by a series of smell waves with a wave height of the order of a few ten- thousandths of an inch and a wave length of the order of 2 inches on the same 60-inch-chord model. For the types of wave usually encountered on practical- construction wings, the test of rocking a straightedge over the surface in a chordwise direction is a fairly satisfactory criterion. ‘The straightedge should rock smoothly without jarring or clicking. ‘The straightedge test will not show the existence of waves that leave the surface conves, such as the wave of figure 22 and the series of small waves previously mentioned. Tests of a large number of precticel-eonstruction models, however, have shown thet those models which passed the straightedge test were sufficiently free of smell waves to permit low drags to be obtained at flight values of the Reynolds number. It is not feesible to specify construction tolerances on air foil ordinates with sufficient accuracy to ensure adequete freedom from waviness. If care is taken to obtain fair surfaces, normel tolerances may be used without causing serious alteration of the drag characteristics.282 Center of wove eee 8 Section drag coethinent, Cea omen ere: REPORT NO, 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS 2336 GB Wine Reynotas number. A ‘Prowns 2.—Rrpertmntel eave sboriog vara f ding eonflatwithRepacde mami fer te NACA @Sqqy- all ton with Drog with fixed tronsition—If the airfoil surface is sufi- ntly rough to cause transition near the leading edge, large drag increases are to bo expected. Figure 23 shows that, although the degree of roughness has some effect, the increment in minimum drag coefficient caused by the smallest roughness capable of producing transition is nearly as great fs thet caused by much larger grain roughness when the Toughness is confined to tho leading edge. ‘The degree of roughness has a much larger effect on the drag at high lift coefficients. If the roughness is sufficiently large to cause transition at all Reynolds numbers considered, the drag of the airfoil with roughness only at the leading edge decreases with inereesing Reynolds number (fg. 10 and reference 36). ‘The effect of fixing transition by means of a roughness strip of carborundum of 0.011-inch grain is shown in figure 24. ‘The minimum drag increases progressively with forward movement of the roughness strip. ‘The effect on the drag at high lift cocffcients is not progressive; the drag increases rapidly when the roughness is at the leading edge. Figure 25 shows that the drag coefficients for the NACA 65(223)~422 and 63(420)-422 airfoils were neatly the same throughout most of the lift range when the extent of laminar flow waa limited to 0.30¢. All recent airfoil date obtained in the Langley two-dimensional low-turbulence pressure tunnel include results with roughened leading edge, and these data are included in the supplementary figures. ‘Tests with roughened leading edge were formerly made only for e limited number of airfoil sections, especially those having large thickness ratios (reference 37). ‘The standard roughness selected for 24-inch chord models consists of 0.011-inch carborundum grains applied to the airfoil surface at the leading edge over a surface length of 0.08¢ measured from the leading edge on both surfaces. The grains are thinly spread to cover 6 to 10 pereent of this area, This standard roughness is considerably more severe than that caused by the usual manufacturing irregularities or deterioration in service but: is considerably less result of accumulation of ice or mud or damage in military combat. ‘The variation of minimum drag coefficient with thickness ratio for a number of NACA airfoils with standard roughness is shown in figure 12. These data show that tho magnitudes of the minimum drag coofiicients for the NACA G-series tirfoils are less than the values for the NACA four- and five-digit-series airfoils. ‘The rate of increase of drag with thickness is greater for the airfoils in the rough condition than in the smooth condition. Drag with practical-construction methods. drag coefficients of several airplane wings have been measured in flight by the wake-survey method (reference 38), and a number of practicel-construction wing sections have been tested in the Langley two-dimensional low-turbulence pressure tunnel at fight values of the Reynolds numbe Flight data obtained hy the NACA (reference 38) are sum- marized in figure 26 and some data obtained by the Consol dated Vultee Aireraft Corporation are presented in figure 27 Data obtained in the Eangley two-dimensional low- turbulenee pressure tunnel for typical practical-construction sections are presented in figures 28 to 32. Figure 83 presents comparison of the drag coefficients obtained in this wind tunnel for a model of the NACA 0012 section and in flight for the sume model mounted on an airplane. For this case, the wind-tunnel and fight data agree to within the expet mental error. All wings for which flight date are presented in figure 26 were carefully finished to produee smooth surfaces. Great care was taken to reduce surface waviness to a minimum for all the sections exeept the NACA 2414.5, the N-22, the Republic $-3,18, and the NACA 27-212, Curvature-gage measurements of surface waviness for some of these airfoils are presented in reference 38, Surface conditions correspond= ing to the date of igure 27 are described in the figure. ‘These data show that the sections permitting extensive laminar flow had substantially lower drag coefficients whet ef The section severe than that likely to be encountered in service as a | smooth than the other sections.26 a2] 24 2o| = ‘Section lift coetticiont & 18 226, 8 & 5 8 Section dhog coefiicient & ‘Airfoil, NACA i082 422 Be 26510" ‘Chard: 38 Fests Tor. 76 ie 7 @ ee z Section engle of artook, dog ° Sethian it Soot tetach, & ‘room Lit and drag ehuasarie ofan NAOA tt) orl with various Agra oughnas tha ad ee. Viva TORY 40 REVIEIEAS 88%35 22| 24 24 20| 5 5 = section litt coefficient, 6 2} 5 Ar ai Nace BROT az et z 26x00 11 Tl Gere 3: tTegh 107, Se ee Mee el 9 ee Section angie of a*0ch, ey eg oad 0K 8 i P Spa : Rigen re 2 oe eis area ee Section drag coefficient, &¢ 8 o16| oral BO Ee ee Section Itt Sect icieth 6 ‘acs Me Lita dng earner ofa NACA 600-42 all wth tacts ronhoa 4 rr chatr ston. ‘82 3 3 } 2 2 7 Z i 3 a z 3SUMMARY OF ATRFOTL DATA ih r & Section cho coef 8 Inca aaz29-dee Onoattiod| a ‘Section bit costhciant, 6 Flora 2—Drag chactcte of fro NAGA toe lefls wth Lach rae Tagiacer at 0302 aH ise, ‘The wind-tunnel tests of practical-construction wing sections as delivered by the manufacturer showed minimum drag coefficients of the order of 0.0070 to 0.0080 in nearly all cases regardless of the airfoil section used (figs. 28 to 32). Such values may be regarded as typical for good current, construction practice. Finishing the sections to produce smooth surfaces always produced substantial drag reductions although considerable waviness usually remained. None of the sections tested had fair surfaces at the frontspar. Unless special care is taken to produce fair surfaces at the front spar, the resulting wave may be expected to ceuse transition either at the spar location or a short distance behind it, One practical-construction specimen tested with amooth surfaces maintained relatively low drags up to Reynolds numbers of approximately 30X10* (NACA 66(2x15)-116 airfoil of fig. 10). This specimen had no sper forward of about 35 peteent chord from the leading edge end no spanwise stiffeners forward of the spars. This type of construction resulted in unusually fair surfaces and is being used on some modem high-performance airplanes. ‘A comparison of the effect of airfoil section on the mini mum drag with practical-construction surfaces is very diffi cult because the quality of the surface has more effect on the drag than the type of section. Probably the best comparison can be obtained from pairs of models constructed at ‘sen107—80—20 285 xq jgeeese! g NACA 642-(.403.5)) & mais SS aca 52-204" eC (ch 27raep | Racubhc $3.5] wee q 208 [alone ssa. Lashes S37 Zutcn 5-25 3 Wack 652-2147) g “NACA 27-212 04 f ikon Bae a + jp Toeeest a ‘Faves —Comparion of maton dag cetisentsobtlod n ght on uso ies "Tai of CAGA Pe and S215 mets ate os Bove Hck 000} nce 24145 BO 8s ] « & Section cog coefficient, o 2 5 a it coefitcient Go ‘ocas —Ceaslated-Vule ft mesures of Ue ect fg ste colton (on deg ofan NACA BOISE) wag set the same time by the same manufacturers. Data for such pairs of models are presented in figures 30 to 32. ‘The results indicate that as long es current construction practices are used the type of section has relatively little effect at flight values of the Reynolds number for military airplanes.org ; Ey gh Pee tes ie dd sa 8 T B cos i eo I 5 oF a 15 20a Reynctaé nurber, R lowe 28—Dng wale eee on 1olnchsbord practeemstoctin mode of Ue [NACA 68Gi0-4080 (prox) aol ston. c=: (prox). 8 a 2, Soaaeaaneaeeaaes 2 He Seton 8 REPORT NO. 824—-NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS Simuiated doar joint ‘iovxx Variation of rg entice wth Reyulde nur a the NACA 20th ‘Soci logan wth loa and arbuent eaten eeicns e&a ite 8 > Section crag costticiont, a er) Séétion iti cootticiont, 6 Figonx 31—Drg carats ofthe NACA SAD)-417 (approx) and NACA 46 (Gpptor) al sections ult by prec oasrustn mtb by te oust an ‘Bete Res25KIm, = a 016; 2 Bont ing erga! cotton IT. I 8 5S 8 NAC Bo cerae g 8 Sectondrog coethent, Ce ae eo a NS a pee on ee Reynotis nunber: R Pros ®2—Drareoe ofc fra ode othe NACA Saco mtn, 827 pce ‘ler na the Davis el wotln, 1827 pert thes, bul by beste! enction ‘metods bythe soe menuitem. (peo) BT g Section hog coetticient, € 20 ae amber, R loons Stouts eet on dg ofthe NACA ONG)-116 and NACA BOI al wetlons Dl by pacteateuastrction method theme masacturrand ted are 78 216 Reynotes Cea ae re eer oe ee Reynolds nunzer, A er ‘ovns S—Comparion of dag exelent aren gt and wind tuna for the "NACA On aol eto aoSUMMARY OF AIRFOIL DATA 287 s TTT Try e Taos L | ach 23ors tapi okdmlh decor 08627 Er Pie ate Et be on loner surface qn $a. 8 hee BAI Galan Gorse coco | claah 3 8 fe SETA ah ar ; ic Bega end th Gece roncved Ie 3 | > Loe == = $ == Lit 5m as OEE 62028 a8 éynotcs number, R Fiorae Eset fdr the dag eto practi consrction a sections wih lately month sats, Important savings in drag mey be obtained at high Reynolds numbers by keeping the surfaces smooth even if extensivelaminar flowisnotrealized. Drag increments resulting from surface roughness in turbulent flow heve been shown tobeimportant (reference31). Theeffects ofsurfaceroughness fon the variation of drag with Reynolds number are shown, in figure 29, in which the favorable scale effectusually expected at high Reynolds numbers wes not realized. ‘This type of scale effect may be compared with that shown for the NACA. 63(420)-422 airfoil with rough leading edge but otherwise smooth surfaces (fig. 10). Drag increments obtained in flight resulting from roughness in the turbulent boundary layer with fixed transition are presented in reference 39. ‘The effect of the application of de-icers to the leading edge of two smooth eirfoils is shown in figure 34. The de-icer “boots” were installed in both cases by the manufacturer to Airfout sechons ‘Root MOK BO L2xI-O18 Tig MACH OF, POSH Agpect ratio, 592 s Propel tip rodius: Sov ' i o eter maha 008} 8 Prebeliee Penoved : Bead § 3a ad caesar? Distance from moda center tine, ff igoan 25 ~The eet of propel operation ox ston rag cola of hte alt= ikse Oem testa ode in tbe Laney ox preare uate. Cem Rt XI represent good typical installations. The minimum drag coefficients for both sections with de-icers installed were of the order of 0.0070 at high Reynolds numbers. Effects of propeller slipstream end airplane vibration — Very few data are evailable on the effect of propeller slipstream on transition or airfoil drag; the data that are available do not show consistent results. | This inconsistency may result from vatitions in lift coefficient, surface condition, airstream turbulence, propeller advance-diameter ratio, and number of blades. Tests in the Langley 8-foot high-speed tunnel indicated trensition occurring from 5 to 10 percent of the chord from the leading edge reference 40). Drag measurements made in the Langley 19-foot pressure tunnel (fig. 35) indicated only moderate drag increments resulting from a windmilling propeller. Although the data of figure 35 may not be very accurate because of the difficulty of maki wake surveys in the slipstream, these date seem to preclude very large drag increments such as would result from movement of the transition to e position close to the leading edge. ‘These date also seem to be confirmed by recent NACA flight data (Gg. 36), which show transition as far back as 20 percent. zu « s ie ‘ e é ae © Right wing sect Sisice cissireom Left wing section in alpiveom I Orordwiee position of transition a 7 a 2s 4 Séétion UR coottioart e ‘Fceae aught moncwremeats often ex em NACA Otenrien wine witha and ‘tlle thepaeam,288 of the chord in the slipstream. Other unpublished NACA flight date on transition on an S-8,14.6 airfoil in the slipstream indicated that laminar flow occurred as far back as 0.20. Even less date are available on the effects of vibration on transition, Tests in the Langley 8-foot high-speed tunnel (reference 40) showed negligible effects, but the range of frequencies tested may not heve been sufficiently wide. Some ‘unpublished flight data showed small but consistent rear ward movements of transition outside the slipstream when the propellers were feathered. This effect. was noticed even when the propeller on the opposite side of the airplane from. the survey plane wes feathered and was accordingly attributed to vibration. Recent tests in the Ames full-scale tunnel showed premature adverse sealo effect on drag coefficients measured by the wake-survey method when's model-support strut vibrated, MPT CHARACTERISTICS OF SMOOTH AIRFOMLS ‘Two-dimensional data,—As explained in the section “ of Zero Lift,” the anglo of zero lift of an airfoil is largely determined by the camber. ‘Thin-sirfoil theory provides a means for computing the angle of zero lft from the mean-line date presented in the supplementary figures. ‘The agreement between the calculated and the experimental angle of zero lift depends on the type of mean line used. Comparison of the experimental values of tho angle of zero lift obtained from the supplementary figures and the theoretical values taken from the mean-ine data shows that the agreement is good except for the uniform-load type (a=1.0) mean line. ‘The angles of zero lift for this type mean line generally have values more positive than those predicted. ‘The experimental values of the angles of zero lift for a number of NACA four- and five-digit, and NACA 6-series airfoils are presented in figuro 37. ‘The eirfoil thickness appears to have little eect on the valuo of the angle of zero lift regardless of the sirfoil series. For the NACA four-digit-series airfoils, the angles of zero lift are approximately 0.93 of the value given by thin- airfoil theory; for the NACA 280-series airfoils, this factor is approximately 1.08; and for the NACA 6-series airfoils with uniform-lotd type mean line, this factor is approximately 0.74, ‘The lift-curve slopes (fig. 88) for airfoils tested in the Langley two-dimensional low-turbulence pressure tunnel are higher than those previously obtained in the testa reported inreference 8. It is not clear whether this difference in slope is caused by the difforence in air-stream turbulence or by the differences in test methods, since the section data of reference 8 were inferred from tests of models of aspect ratio 6 ‘The present values of the lift-curve slope were measured for a Reynolds number of 6X10* and at values of the lift coefficient approximately equal to the design lift coefficient of the REFORT NO, 824—NATIONAL ADVISORY COMMPTTEE FOR AERONAUTICS irfoil section, For the NACA G-series airfoils this lift cocff- cient is approximately in the center of the low-drag range. For airfoils having thicknesses in the rango from 6 10 10 percent, the NACA four- and five-digit series and tho NACA GA-series airfoil sections havo valucs of liftcurve slope very lose to the value for thin airfoils (2x per radian or 0.110 per degree), Variation in Reynolds number between 3X10" and 9X10" and variations in airfoil eamber up to 4 percent chord appear to have no systematic effect on valucs of lift-eurve slope. The airfoil thickness and the type of thickness distribution appear to be the primary variables. For the NACA four and five-digitscries airfoil soctions, the lift curve slope decreases with inerease in airfoil thickness, For the NACA G-series airfoil sections, howover, the lift curve slope increases with increase in thickness and forward movement of the position of minimum pressuro of the basic thickness form at zero lift. Soma NACA 6-series airfoils show jogs in the lift curve at the end of the low-drag range, especially at low Reynolds numbers. This jog becomes more pronounced with inerease of camber or thickness and with rearward movement of the position of minimum pressure on the baste thickness form. ‘Dhis jog decreases rapidly in severity with increasing Rey- nolds number, becomes merely a change in lift-curve slope, and is practically nonexistent at a Roynokls number of 9X10® for most airfoils that would be considered for practical application, ‘This jog may bee consideration in the selection of airfoils for small low-speed airplanes. An analysis of the flow conditions leading to this jog is presented in reference 28. The variation of maximum lift coofficiont with airfoil thickness ratio at a Reynolds number of 6%10¢ is shown in figure 39 for a number of NACA airfoil sections. ‘The airfoils for which date are presented in this figure have 2 range of thickness retio from 6 to 24 percent and eambers up to 4 percent chord. From the data for the NACA four- and ies airfoil sections (Bg. 39 (a)), the maximum lift cocflicients for the plain airfoils appear {o be the greatest for a thickness of 12 percent, In general, the rato of change of maximum lift cocfficient with thickness ratio appears to be greatest for airfoils having a thickness less than 12 percent, ‘Tho date for the NACA G-series airfoils (igs, 89 (b) to 39 (e)) also show a rapid increase in maximum lift coefficient with increasing thickness ratio for thickness ratios of less than 12 percent. For NACA O-scries airfoil sections cambered to give a dosign lift coefficient of not more than 0.2, the optimum thickness ratio for maximum lift coefficient appears to be between 12 and 15 percent, except for the airfoils having the position of minimum pressure at 60 peréent chord. ‘The optimum thickness ratio for tho NACA 66-series sections cambered for a design lift eoeli- cient of uot more than 0.2 eppears to be 16 percent or greater.SUMMARY OF AIRFOT: DATA 289 ts zg § series s we, tice y oo 24 fF = Oe . = 44: ws S230 (e-igit) $ o8 = na g 8 Ro i j oo 1 EE 5 a ye into thtxness|Sarcent ot chord Yictit tetas, percent Of ced (Shokan red (9 AoA eta, aR £4 2 T, © T g &% 8 & °0 tte’ tl tee eT t ea lalice, 2 4 ‘ ° 8 sP[e 4 } ba § = & j pr | g 2 2 } 2 le a) ig ze % 15 53924 & Airfoil thickness, percent of chard Airfoil thickness, percent of chord @ moto to Wack tet & A 9 (el 7 Measured section angle of zero ltl, dog fy red nemrner peau fs es 0 EE ES Surtoit thickness, percent of chord (@ NACA Gesu, ‘rocas 9 —Meesurd seston agi of aro fo anata of NACA abl tons of varonsthlcknemesend mbar, RAK,290 REPORT NO, 824—NATTONAL ADVISORY COMMITTEE FOR AERONAUTICS Bog Rv, . Tiopeel syntbie tedbots rho) etoalioy o &,, 1% be $e 8 : wd a g Sees g ® oa 37a : ed 3.2 2 Tel 9, 240 Gata : ol s%e 8 0 i 6 18 20 22 248° 6 0 i 4 6 16 20 2 #4 s firtoll thickness, percent of chard S ‘Aictoll richness, percent of chord (9 NACA har andrei, ) NACA Bane Ree, Rie, ¢ - i. L bl “ & : Xd Sd 8 g e ia S00 eee eo ae aS ee eee ea “ictoil thickness, pencent of chord (9 NACA bere, aa| rome Litt-curve slope per ogres 4 (0) NACA cose, te, ‘2. 4] aoe (seme edge eae oe) Airtoll thickness, percent of chord () NACA costa, ‘looms —Voratn of tauve slope with all hicks rao end cule fr amb of NACA sl alos oth tbe emt and rua enon, eA, ‘The available data indicate that a thickness ratio of 12 percent or less is optimum for airfoils having a design lift coefficient of 0.4. ‘The maximum lift coffcient is least sensitive to variations in position of minimum pressure on the basic thickness form for airfoils having thickness ratios of 6, 18, or 21 percent. ‘The maximum lift coefficients corresponding to intermediate thickness ratios increase with forward movement of the position of minimum pressure, particularly for those airfoils ‘having design lift coefficients of 0.2 or less. ‘Tho maximum lift coefficients of moderately cambered NACA G-series sections increase with increasing camber Gg. 39 (b) to 39 (@)). The addition of eamber to the symmetrical airfoils enuses the greatest increments of maximum lift coefficient for airfoil thickness ratios varying from 6 to 12 percent. ‘The effectiveness of camber as a means of increasing the maximum lift coefficient generally decreases fas the airfoil thickness increases beyond 12 or 18 percent, ‘The available data indicate that the combination of a 12- pereent-thiek section and a mean line cambered for a design ift. coefficient of 0.4 yields the highest maximum lift cocfficient,SUMMARY OF ATRFOM: DATA 291 & & ata 2 d ie “24 ‘4 zaps i Ml aga airtoiwn = 24S Aicfot wth i a eel a tee P foo Spi top S20} Bsc Sade @ 3 14) 4-con| Ee : # . 1 Sua} £ , § Pron giz poe Bu aifoit fl i j Bagh 5 Snot 2 oo ee, oe 2 16 oe 6 Abo thekress: percent of chord Airfot inches, percent of chard (9 S408 oerand treat st, (0 NACA eat, F 5 i bed od BaP J p% a5 a 7 S2sL9 2 Sea? 408 al PALE Airfoil wath g7*Ts 4 bea Aur foil with gle Z ‘9H tap Bd S22 ‘lt top 30 Oy Si Veer eos Li S202 % 2 Sie pei 22! 53 fi, = A100, orf Bu. ~ oe Pat * ough 2° ~ Rough [ Tirta i al Sean a oa ee 8 2 AvloN tictress, percent of chord Mittal tikross, percent of herd NACA oat, (eo Naca i, Se i ee peat Woted tt ep Gehected OO seal 3 Alon with g ‘ott top S29 4 a, Pio ofan § £ aaa Regt = Bn 48 e160 Airfoll thickness, porcert of chord (9 NAOA deans, rovas9 variate of mse secon oekaki eso (cues ei and camber fr axel NACA sel seins with and witheat simulated spt fps aod stuart morte Rati292 ‘The variation of maximum lift with typo of mean line is shown in figure 40 for one G-series thickness distribution. No systematic data are available for mean lines with values of aless than 0.5, Itshould be noted, however, that airfoils such as the NACA 230-series sections with the maximum camber far forward show large values of maximum lift, Airfoil sections with maximum eamber far forward and with thickness’ ratios of 6 to 12 percent usually stall from the leading edge with large sudden losses in lift. A more desirable gradual stall is obtained when the location of meximum eamber is farther back, as for the NACA 24-, 44-, and G-series sections with normal types of camber. B 4 |__| i ct Tt i | : } § 6) number 8 | Peal : St i i | | i. Oa te Type of camber, a ‘loons d—Vvtion of maxim it oxdent wits typeof ober for sme NAOA (cet setons rom taf be Laney tvedlzeatnn! low Srbuene esas taal A comparison of the maximum lift coefficients of NACA 64-seriea airfoil sections cambered for a design lift coefficient of 04 with those of the NACA 44- and 230-series sections (Gg. 39) shows that the maximum lift coefficients of the NACA 64-sories airfoils are as high or higher than those of the NACA 44-series sections in ell cases. ‘The NACA 230- series airfoil sections have maximum lift coefficients somewhat higher than those of the NACA 64-series sections. ‘The sealo effect on the maximum lift coefficient of e large number of NACA airfoil sections for Reynolds numbers from 3X10* to 9X10 is shown in figure 41. The scale effect for the NACA 24-, 44-, and 230-series airfoils (figs. 41 (a) and 41 (b)) having thickness ratios from 12 to 24 percent is favorable and nearly independent of the airfoil thickness. Increasing the Reynolds number from 3X10" to 9X10 results in an increase in the maximum lift coofficient of REPORT NO, 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS approximately 0.15 to 0.20, ‘The scalo effect on the NACA 00- and 14-series airfoils having thicknoss ratios lees than 0.12¢ is very small ‘Tho scale-effect data for the NACA G-series airfoils (Gge. 41 (c) to 41 (D) do not show an entirely systematic variation In general, the scalo effect is favorable for these airfoil sections. For the NACA 63- and G4-series airfoils with small camber, the increase in maximum lift coefficient. with inerease in Reynolds number is gencrally small for thickness ratios of less than 12 percent but is somewhat larger for the thicker sections, ‘The character of the scalo effect for the NACA 65. and 68-scries eirfoil sections is similar to that for the NACA 63- and 64-series airfoils but the trends are not 50 well defined. In most cases the sealo effect for NACA 6-gerics airfoil sections cambered for a design lift coefficient of 0.4 oF 0.6 does not vary much with airfoil thickness ratio, ‘The data of figure 42 show that the maximum lift coefficient for the NACA 63(420)-422 airfoil continues to increase with, Reynolds number, at least up to a Reynolds number of 26X10 ‘The values of the maximum lift coefficient presented were obtained for steady conditions. ‘Tho maximum lift coefli- cient may be higher when tho angle of atteck is increasing, Such e condition might occur during gusts and Janding maneuvers. (See reference 41.) ‘Tho systematic investigation of NACA. G-series airfoils included tests of the airfoils with a simulated split fap do flected 60°. It was belioved that Wiese tests would serve as an indication of the effcctivencss of more powerful types of ‘railing-edge high-lift devices although sufficient data to verify this assumption have not been obtained. ‘Tho maximum coefficients for a large number of NACA airfoil sections obtained from tests with the simulated split flap are presented in figure 39. ‘The data for the NACA 00- and 14-series airfoils equipped with split flap for thickness ratios from 6 to 12 percent show a considerable inerense in maximum lift coefficient with increase in thickness ratio. Corresponding data for the NACA 44-series airfoils with thickness ratios from 12 to 24 percent show very little variation in maximum lift coefficient with thickness, For NACA 6-serios airfoils equipped with split flaps the maximum lift cocfficients increase rapidly with increasing thickness over a rango of thickness ratio, Une rango Deginning at thickness ratios between 6 and percent, depend= ing upon the camber. ‘The upper limit of this range for the symmetrical NACA G4- and 6-scries airfoils appears Lo be greater than 21 percent and for the NACA 63- and GO-scries airfoils, approximately 18 percent. Between thickness ratios of 6 and 9 percent the values of maximum lift coefficient for the symmetrical NACA G-series airfoile are essentially the samo regardless of thickness ratio and position of minimum pressure on the basic thickness form. ‘The maximum lift coeffciast decreases with rearward movement, of minimum pressure for the airfoils having thickness ratios between 9 and 18 percent,SUMMARY OF AIRFOIL DATA 293 20) 20, AOA 24-ceries (4- f | _ ss [ I I A As aalvarea lo As ceiivered’ Base 3h 8. ae ona et a hk at ele 3 Ge ar reer deo Per test tas | ° (Oar rast 32 T a | { { 7 @ @ I () O86 8 5g a 8a Section angle of attack, &, ce (o waoa ma @) Mack 208 (o waca mo, {one Ami arte ofthe NACA O86 I i ea at hy ral nae, NH a. ‘The maximum lift coefficient may be lowered by failure to maintain the true airfoil contour near the leading edge, but a no systematic data on this effect have been obtained. Ex- amples of this effect that were accidentally encountered are 1 presented in figure 44, in which lift characteristics are given for accurate and slightly inaccurate modcls, The model 12 inaccuracies were so small that they were not found previous to the tests. ‘Three-dimensional data —Tests of soveral airplanes in the Cee Lift coothicient, G Capra (ZEeamiC. Jngle of offack, « deg Moons 4—Too ess of sues snttons oo thn Mt characters of when ‘epaoe, RoAMI Zoe Langley full-scale tunnel (reference 42) show that many factors besides tho airfoil scctions affect the maximum lift coefficient of airplanes. Such factors ns roughness, leakago, leading-edge air intakes, armament installations, nacelles and fuselages make it difficult to correlate the airplano maxi mum lift with the airfoils used, even when the flaps aro retracted. The various flap configurations used make such a correlation even more difficult, when the flaps aro deflected, ‘When the flaps wero retracted, both the highest and tho lowest maximum lift coofficionts obtained in recent tests of airplanes and complete mock-ups of conventional configurations in the Langley full-scale tunnel were thoso obtained with NACA 6-series airfoils. ‘Results obtained from tests of a model of an airplane in the Langley 19-foot prossure tunnel and of th airplane in the Langley full-scale tmnel are presented in figure 45. Both tests were made at approximately the same Reynolds number, The results show that the airplane in the service condition had & maximum lift coefficient more than 0.2 lower than that of the model, as well as a lower lift-curvo slope. Some improvement in tho airplane lift characteristics vwas obtained by sealing leaks. ‘Those results show that airplane lift cheracteristies are strongly affected by details not reproduced on large-scale smooth models.SUMMARY OF AIRFOTL DATA 297 16 tab * #7 RI VA | area setiars ‘Rook, NACA 631420) 422 Ties Modified NACA SBiga-517 Litt cootticient, O NY Airfoil sections, Lilt coetticiont, bs & Wie toe ee parcent) Eee 2 Nofur!transilion Natural transition © Transition fixed of. Ie: Transition Tied af 106 NO a ecru ee 0 ae Joga of oftock, deg ‘Pcous sh The eet onthe Mt shart of fang the tation ox ode nthe Tangy Toot pease Ce RoAT%I0 (Mol wilh Devi el eto) Lift characteristics obtained in the Langley 19-foot pressure tunnel for two airplane models in the smooth condition and with transition fixed at the front spar are presented in figures 46 and 47. In both cases, the lift-curve slope was decreased throughout most of the lift range with fixed transition. The maximum lift coefficient was decreased in one case but was inereased in the other ease. ‘The attempt to obtain low drags, especially for long-range airplanes, leads to high wing loadings together with relatively low span loadings. ‘This tendency results in wings of high aspect ratio that require large spar depths for structural efficiency. ‘The large spar depths require the use of thick root sections. ‘This trend to thick root sections has been. encouraged by the relatively small increase in drag coefficient with thickness ratio of smooth airfoils (fg. 12). Unfortunately, airplane swings are not usually constructed with smooth surfaces and, {in any case, the surfaces cannot be relied upon to stay smooth under all servico conditions. . The effect of roughening the leading edges of thick airfoils is to cause large increases in the oO 20 a a8 16 ‘angie of otfack, te, deg ‘isewe 4—~Theet onthe it carters of faag tb tron ona mote fa tbe Tanger oot pears tel. eA TOO, (adel wid NACA ail stone) drag coefficient at high lift coefficients. ‘The resulting drag. cocflicients may be excessive at cruising lift cocficients for heavily loaded, high-altitude airplanes. Airfoil sections that have suitable characteristics when smooth but have excessive drag coefficients when rough at lift coefficients corresponding to cruising or climbing conditions are classified as. ‘uuneonservative. . "The decision es to whether e given airfoil section is consery- ative will depend upon the power and the wing loading of the airplane. The decision may be affected by expected service ‘and operating conditions. For example, the ability of a multiengine airplane to fly with one or more engines inoperative in icing conditions or after suffering damage in combat may be a consideration, ‘As an aid in judging whether the sections are conservative, the lift coefficient corresponding to a drag coefficient of 0.02 was determined from the supplementary figures for a large number of NACA airfoil sections with roughened leading edges. The variation of this eritical lift coefficient with aim foil thickness ratio and camber is shown in figure 48. ‘These data show that, in generel, the lift coefficient at which the drag coefficjent is 0.02 decreases with rearward movement of298 sere] | 2} a4 Lb a-cigi 10) 1230 (6.21 ieee eet SF 5S 5 & ze 5 Secton lit coefficient atwhich section dhag coathcient is O02 ro} a od 22 a a a Airfol thickness, percent of chord (@) NACA four and roc ate, 0) NACA coat @ NACA tomrae @ NACA domen, (@ NACA cba ‘Fovne salto ofthe if nate coreapanng 1 8 dug cet of 0.2 with (hides and ember fora utr oN ACA al oct wits eagbebod eng ae Zamaase [REPORT NO. 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS position of minimum pressure. ‘The thickness ratio for which this lift coefficient is a maximum usually lies between 42 and 18 percent; variations in thickness ratio from this optimum renge generally cause rather sharp decreases in the critical lift coefficient. ‘The addition of camber to the symmetrical airfoils usuelly causes an inercase in the critical lift coefficient except for the very thick sections, in which ense increasing the camber becomes relatively ineffectual and may be actually harmful, All the data of figure 48 correspond to a Reynolds number of 6X10'. As shown in figuro 49, the drag coefficient at flight values of the Reynolds number may be considerably lower than the drag coefficient at e Reynolds number of 6X10* if the roughness is confined to the leading edge, PrreMIxG MOMENT ‘The variation of the quarter-chord pitehing-moment, cocfficient at zero angle of attack with airfoil thickness ratio and camber is presented in figure 60 for several NACA airfoil sections. ‘The quarter-chord pitching-moment coefficients of the NACA four- and five-digit-scries airfoils become less nogative with increasing airfoil thickness. Almost no variation in quarter-chord pitching-moment, coefficient with air- {oil thickness ratio or position of minimum pressure is shown, by the NACA 6-series airfoil sections. As might be expected, inereasing the amount of camber causes an almost uniform negative inerease in the pitching-moment coefficient, As discussed previously, tho pitching moment of an airfoil seetion is primarily a funetion of its camber, and thin-airfoil theory provides a means for estimating the pitching moment from the mean-tine data presented in the supplementary figures. A comparison of the experimental moment cocfli- cient and theoretical values for the mean lines is presented in figure 51. The experimental values of the moment coc- cients for NACA 6-series airfoils cambered with the uniform Toad type mean line are usually about three-quarters of the theoretical values (figs. 50 and 51). Airfoils employing mean Jinog with valuos of a less than unity, however, have moment coefficients somewhat more negative than those indicated by theory. The use of a mean line having a value of a less than unity, therefore, brings about only a slight reduction in pitching-moment coefficiont for a given design lift eoofficieut when compared with the value obtained wilh « w oad type mean line. ‘The experimental moment coefficients for the NACA 24-, 44, and 230-series airfoils are also less nogative than those indicated by theory but the agreement is closer then for airfoils having the uniform-load type mean Tine. ‘The pitching-moment data for the airfoils equipped with simulated split flaps deflected 60° (fg. 60) indicate that the value of the quarter-chord pitching-moment, coefficient becomes more negative with increasing. thickness for all the airfoils tested. For the thicker NACA G-series scetions the magnitude of the moment coefficient increases with rearward movement of the position of minimum pressure.8 5 28} Section itt costticient, 6 & Sechon dog coefficient 6 [airfoth AACA 651229)-42° (noarfiogh Karone $8 2416 2 16 eB SR ae oe 2 @ Section angle of ottach, ty, deg Section 11M coettcient, 6, Miovns @.—Litand dng chatritonctan NACA W242 (mda) al wih macdard regina appellee ae, 2 1 Viva TosuIY 40 AEYEEOS 662300 REPORT NO, 824—3 Single flogged symbols are far 60" simulated spit Flop] é z g Series 3 ea fa Sepa a a 44. 3 v, 230 B-agity 2 16 24 eS 20. Alcfollthichoses, percent of chard (9 NAOA tor and rede sates, ani Moment cootticiea buy ermal EE near ld sears a0 ees) Airfoil thickness, percent of chord (@ NAO state, 4 Moment costticiont bayg (or o NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS TEM Saag a — a Meonertcoote | 2 te ren a aaa ere Airfoil thickness, parcent of chore! (©) NACA sar, coettislert, Guay Mement Ae 2 a a Airfoil thickness, percent of chard (@) NACA costa i ae Airtoll thickness, pertent of chord (9 NACA Stes ‘Moons #.—Vartlonotaetion qa pltshing omen entice! (menmre atenanglcfatac ef with eri tices rate seat NACA ae eto often amb BoxSUMMARY OF AIRFOIL DATA SSS | at design itt costficrent (e19)-216- 24 peee/a274 a-as 1 e012 Moment coefticion', See 06 BB 1G, J] 16 Theoreiice! manent coefficient for jhe airtol ‘ean ine at uerter= chord paint rocax Comparten of thereto! and manuel pitelng moet eoeticlet or sme ‘NAGA atte, O00. POSITION OF AERODYNADHIC CENTER ‘The variation of chordwise position of the eerodynamic center corresponding to 8 Reynolds number of 6X10" for & large number of NACA eirfoils is presented in figure 52. From the date given in the supplementary figures there ‘appears to be no systematic variation of chordwise position of aerodynamic center with Reynolds number. The date for the NACA 00- and 14-series airfoils, presented for thickness ratios less than 12 percent, show that the chordwise position of the aerodynamic center is at the quarter-chord point and does not vary with airfoil thickness. For the NACA 24, 44-, and 230-series airfoils with thickness ratios ranging from 12 to 24 percent, the chordwise position of the aerodynamic center is ahead of the quarter-chord point and moves forward with increase in thickness rati ‘The chordwise position of the eerodynamic center is behind the quarter-chord point for the NACA. 6-series airfoils and moves rearward with increase in airfoil thickness, which is in accordance with the trends indicated by perfect-fuid theory. There appears to be no systematic variation of position of minimum pressure on the basic thickness form for these airfoils. ‘The date of reference 43 show important forward movements of the aerodynamic center with inoreasing trailing-edge angle for a given airfoil thickness. For the NACA 24-, 44-, and 230-series sirfoils (fg. 52) the effect of increasing trailing-edge angle is apparently greater than the effect of 301 inoreasing thickness. For the NACA 6-series airfoils, the opposite appears to be the case. ‘migH.UT DEsICcEs Lift characteristics for two NACA 6-series airfoils equipped with plain flaps are presented in figure 53. These data show that the maximum lift coefficient increases less rapidly ‘with flap deflection for the more highly cambered section. Lift characteristics of three NACA 6-series airfoils with split flaps are presented in reference 44 and figure 54. ‘The maximumift increments for the 12-percent-thick sections were only about three-fourths of that increment for the 16-percent- thick section, ‘The maximum lift coefficient for the thicker section with flap deflected is about the same as that obtained for the NACA 23012 sirfoil in the now obsolete Langley variable-density tunnel (reference 45) and in the Langley ‘7-by 10-foot tunnel (reference 46). Tests of a number of slotted flaps on NACA G-series aicfoils (supplementary figures and reference 47) indicate that the design parameters necessary to obtain high maximum lifts are essentially similer to those for the NACA 230- series sections (references 48 and 49). Lift deia obtained for typical hinged single slotted 0.25¢ flaps (fig. 55 (a) on. the NACA. 634-420 airfoil are presented in figure 55 (b). A maximum lift coefficient of approximately 2.95 was obtained for one of the flaps. Lift characteristics for the NACA 65-118 airfoil fitted with » double slotted flap (reference 47 and fig. 56 (a)) are presented in figure 56 (b). A maximum lift coefficient of 328 was obtained. It mey bbe concluded that no special difficulties exist in obtaining high maximum lift coefficients withslotted flapson moderately thick NACA 6-series sections. ‘Tests of airplanes in the Langley full-scale tunnel (reference 42) have shown that expected increments of maximum lift coefficient are obtained for split flaps (Bg. 57) but not for slotted flaps (fg. 58). This failure to obtain the expected. maximumift increments with slotted flaps may be attributed to inaccuracies of flap contour and location, roughness near the flap leading edge, leakage, interference from fisp sup- ports, and deflection of flap and lip under load. LATERAL-CONTROL DEVICES An adequate discussion of lateral-control devices is outside the scope of this report. ‘The following brief discussion is therefore limited to considerations of effects of airfoil shape on aileron characteristics. ‘The effect of airfoil shape on aileron effectiveness may be inferred from the data of figure 59 end reference 50. Th section aileron effectiveness parameter Aay/AS is plotted against the aileron-chord ratio e,/e for a number of airfoils of different type in figure 59, Also shown in this figure are the theoretical values of the parameter for thin airfoils.302 REPORT NO. 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS 24 - NACA 2310-xerieo 2a = “HAGA ab-series 24 &R x 38 ge Chordwise position of aerodynamic center, so iS x & & in 0d ond Oi Symbole with Vlogs eerrespord to 6,06 s 4} 2d 28} od lato “O42 “eis 20242838 Airfoil thickness, percent of chord (w) NACA fear-and red seen (@ NACA toa 28; 26| gun O4 ond Qi fs wi comnespord 10,64 = 2 a th flags comnempond 10.6, = 24 2d 3 3 sk x & 8 Chorawive position of geodon center, fe & & 24| 24| 24| 2} 20 ai “4818162028 Airfoil iickness, percent of chord 0) NAGA toast, {@ NACA sore, (@) NACA tba, ‘Pome Verano secon chert pooner cit wit fl hoe far seis NACA el sels cf deen cuinet. RABIoe SUMMARY OF ATRFOIL DATA 303 28; 28 —— 2 = 24} A eo} é iol NACA 65,3-68 é 7 - 4 WA ? 4 \ ie 8 ee g ge i V g RA B42 8 3 , i 3 ‘ 2 Maka cola -dio dox104 q 3, tig Si TY 35 a mMersne” Js 8 hac oxcere So 4 d a 2 7 20, 4 Pip defection, Gy dog Frocax Marin it osteo forthe NACA 69-60 ané NACA #606 ae os itt with Oabalsacert pia ps. BEX. ‘The data show no large consistent trends of aileron-effectiveness variation with airfoil section for a wide range of thickness distributions end thickness ratios. In order to evaluate aileron characteristies from section data, a method of analysis is necessery that will lead to results comparable to the ustial curves of stick force against helix angle 5/21" for three- dimensional date. ‘The anelysis that follows is considered suitable for comparing the relative merits of ailerons from ‘two-dimensional data. ‘Two-dimensional data are presented in the form of the equivalent change in section angle of attack Aag required to maintain constant section lift coefficient for various de- Aections of the aileron from neutral. This equivalent change in angle of attack is plotted against the hinge-moment parameter Aca’, which is the product of the aileron defection from neutral and the resulting increment of hinge-moment ° 27 30 700 7 20 Flop Gettection, dy deg ‘rocas H.-Darinam Mt coticents fo rome NACA alos ad ith OaDalchot ou ps, coefficient based on the wing chord. This method of analysis: takes into account the aileron effectiveness, the hinge moments, and the possible mechanical advantage between the controls and the ailerons. ‘The larger the value of Aas for a given value of the hinge-moment parameter, the more advantegeous the combination should be for providing a large value of pb/2¥’ for a given control foree. ‘The essump- tion thet the aileron operates at constant lift coefficient as the airplane rolls is not entirely correct, however, end involves an overestimation of the effect of changing angle of attack on the hinge-moment coefficient. In addition, the span of the ailerons end other possible three-dimensional effects are not considered. In spite of these inaccuracies, the method provides e useful meens of comparing the two- dimensional characteristics of different ailerons.| aa J \ 2 | i Re gL |_ : = Bush - Faxhinoe § eat ; ey gal Q 4 4 | i o | 1 a a Fiep anection Sy c09 nent, obtener ons Pap contrat tin and mn Wea fr the NACA 6-0 al with x OBballtcort Mowe oteé Gap, txt woe SOILAVNOWSY OK TELLINGTOS AUOSIAGY IVNOLLVN—F28 “ON LOTTESUMMARY OF AIRFOT: DATA 305 TTT? Trig ‘ihe ct oplaenlnt apwhen meade 1a “and predicted results| OOS alee : Feo detected 05° g _= a a “ a te Lf 3 toa 2s) A, predicted ‘Pao &7—Cempartea betes mearered a ofthe leremests a ft eoxfeets doe 0 Dap debactiag and alas peed fom twodlnessecs! dats. Spt 4p. nee A Taal : teal = = | toe ct encseore nae maa e 5 b a & Naximum section lift coefficient, g | Sa 4 > Oo. 2 4-3 WE Alig, preciicted ‘Racas st—Comparon Detroc eared vale ft increments ceca de we to tap deflation snd ros predicted em two-dimensional ata tied Bap. o 20 30 0 6 Flop Ceflection, by, clog (0) Fup eatgnnton, (9) sexier it arc, ‘Proves t4—Flapcoatgustion and maim Ut oefBns fr fhe NACA 8 alt rth cochlea Bp. R=xI, '306 REPORT NO, 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS SUPPLEMENTARY INFORMATION REGARDING TESTS OF TWO-DIMENSIONAL MODELS ‘Aledow caren ymbet Tymetmp ese 7 ™ z ° ie | ow aoe + ue | ow * x | xaca ua. um | ln Oo | waca marae ue | ow © | xaceonaes «| cn @ | Nacasyeacrn ope) o |x ® | aca ngsan-ne enna. o | 2 D | RAca wenie o | ow ® | raca mac. a2. © 7 | xack 288 ese. ° s | Raca wane oop. © | ACA meu9-8 ent o |e o 4 | NACA io. re | ow a © | aca ens-ai enn @ |e o o | os o of @ Q o | os ® 6 o | os o | i a 2 o |x ° NACA 843-1088) appx) AES alse ae i ; i j Hf coo ale 1 iG ri 7 2 - 2 * Aileron chord ratio, eg/e Alteron chard ratio, tle © snap toae toi, (0) sang tomo to. ‘Figore @—Vrton f section ron eectrenus with allron cord rflo or trastotnstor sles without expend orerbang Delaney on a number of ail meen ‘Onpeeaed amdSUMMARY OF AIRFOIL DATA 307 = p= = ‘| = ‘Nagae ee, ry oy oa plan taxa 5 ot Aileron dettaction deg |W NACA 63, 4-4 (17.8) por om} 4 SKE mat adage | PN & a Sroight sided bk Tad Getatertor a < has ara ata [ {+ HA -al eS Bie mee DED aes Teas TT HE ae 4 _ =2=—er——t—t—— Be EE TIE TB TO Eo en ae Fee Sate ee en ‘gees Vasa ofthe hlngesementparaceer Sas with te equfalent cage tn sro angle ttn ee fo ails coustant seen Ut eeeat ection ‘the alton oo the NACA Oud, NACA OLE), and NACA BHI 26 «= 08 ‘rt aetiss: Cape led. For the purpose of evaluating the effect of airfoil shape on the aileron characteristics, it is desirable to make the” comparison with unbelanced ailerons to evoid confusion. Plots of the parameters for plain unbslanced flaps of true airfoil contour on three airfoil sections are shown in figure 60. ‘The characteristics of the NACA 66(215)-216, o=0:6 section are essentially the sume as those for the NACA. 0009 airfoil within the range of deflection for which data are available. ‘The NACA 64.2-(1.4)(18.5) airfoil shows appreciably smaller values of Acué for a given value of Aas than the other sections presented. No explanetion for this difference can be offered, although some of the difference may result from the slightly smaller chord of the flap for this combination. ‘The effects of using straight-sided ailerons instead of eiler- oas of true airfoil contour ere shown in figure 61 for to NACA G-series airfoils. One of the two combinations for which data are available was provided with en internal balance whereas the other combination was without balance. ‘This difference prevents eny comparison between the two combinations but does not affect comparison of the two contours for each case. For the NACA 66(215)-216, a=0.6 airfoil, the straight-sided eileron hes more desirable characteristics for the range of deflections for which data are avail- sed me NACH rl sects. On ae able. It appears, however, that the straight-sided sileron would be less advantageous than the aileron of true contour for positive deflections greater than 12°. In the case of the NACA 3,4-4(17.8) (epprox.) airfoil, the straight- sided aileron appears to have no advantage over the aileron of true airfoil contour. The advantage of using straight- sided ailerons appears to depend markedly on the airfoil used but sufficient data are not evaileble to determine the signif- icant airfoil parameters. Figure 62 shows that in one case the effect of leading-edge roughness on the aileron character istics is unfavorable. ‘The problem of designing satisfactory leading-edge air intakes is to maintain the lift, drag, and criticel-speed characteristics of the sections while providing low intake losses over a wide range of lift coefficients and intake velocity ratios. ‘The data of reference 65 show that desirable intake and drag characteristics can easily be maintained over a rather small range of lift coefficients for NACA 6-series airfoils. ‘The data of reference 65 show that the intake losses increase rapidly at moderately high lift. coefficients for the shapes tested. Unpublished data taken at the Langley Laboratory indicate that shapes such as those of reference 65 have low maximum lift coefficients. Recent data show308 { I { JAieron defection, deat “6 al LOE DOIG DOTA DOE DOD 0008 = dopd, rectors ‘roves @.—vatatln of ths hingesneret parueter se witht eutrlect abc fa ‘een anal of tat raged to mala cotant secant or deeton ‘te lero oa tis NACA 842-(L.0034) iol nctino, eaoth nd wih ogo {he lenge bo at. (Poe Seaton a arn, 8), “D004 “OO REPORT NO, 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS that air-intake shapes can be provided for such airfoil seo- tions with desirable airinteke characteristics and without Joss in maximum lift-eoefficient. (fg. 63). Somo pressure- distribution data for the air intakes shown in figuro 63 indicate that the critical speed of the section has been lowered. only lightly and that falling pressures in the direction of flow were maintained for some distance from tho leading edge on both surfaces at lift cooficients near the design lift coefficient. for the section, Sufficient information is not available to permit such desirable configurations to be do- signed without experimental devclopment. INTERFERENCE ‘The main problem of interferonco at low Mach numbers is considered to be that of avoiding boundary-layer separation resulting from rapid flow expansions caused by the addition of induced velocities about bodies and the boundary-layer eecumulations near intersections. No recent systematic investigations of interference such as tho investigation of reference 66 have been made. Some tests have been made of airfoil sections with in- tersecting flat plates (reference 67). Theso configurations may be considered to represent approximately the condition of @ wing intersection with a large flat-sided fuselage. In ir.) Giron chord 2 a eal ia TL sale [ , Lat Ga isl | al . 1 sp j j t Sse} ~ 4 { 7 | i. f ae § & a § ify Yi Ol | Bes Apt e 4 ise Feat | 2 aust rer AGH Poet | 24] 4, ps3: | | * al | J a 4e- 2 ta a 16 Section angle of ettock, @, deg Oe. ‘section ‘tit eoefelent, & ‘ove 6.—Lit und ow charactor ofan NACA Tae type all selon with adage enh,SUMMARY OF AIRFOTL DATA this case, the interference may be considered to result from the effect on the wing of the fully developed turbulent bound axy layer on the fuselage or flat plate end the accumulation of boundary layer in the intersection. These tests showed little interference except in cases for which the boundary layer on the airfoil alone was approaching conditions of separation such as were noted with the less conservative airfoils at moderately high lift coefficients. Some scattered data on the characteristics of nacelles mounted on airfoils permitting extensive laminar flow are presented in references 68 to 70. The data eppeer to indicate that the interference problems for conservative NACA 6-series sections are similer to those encountered with other types of airfoil. The detail shapes for optimum interfering bodies and filleta may, however, be different for various sections if local excessive expansions in the flow ere to be avoided. Some lift and drag date for en airfoil with pusher-propeller- shaft housings are presented in reference 71. These results indicate that protuberances near the trailing edge of wings should be carefully designed to avoid unnecessary drag increments. Another type of interference of particular importance for high-speed airplanes results in the reduction of the eritical ‘Mach number of the combination because of the addition of the induced velocities associated with each body (reference 72). This effect may be kept to a minimum by the use of bodies with low induced velocities, by separation of interfering bodies to the greatest possible extent, and by such selection and arrangement of combinations thet the points of maximum induced velocity for each body do not coincide. APPLICATION TO WING DESIGN Detail consideration of the various factors affecting wing design lies outside the scope of this report. ‘The following discussion is therefore limited to some important aerodynamic features that must be considered in the application of the date presented. APPLICATION OF SECTION DATA Wing characteristics are usuelly predicted from eirfoil- section data by use of methods based on simple lifting-line theory (references 73 to 76). Application of such methods to swings of conventionel plan form without spanwise discon tinuities yields results of reasonable engineering accuracy (reference 77), especially with regard to such important characteristics as the angle of zero lift, the lift-curve slope, the pitching moment, and the drag. Basically similar methods not requiring the assumption of linear section lift characteristics (references 78 and 79) appear capable of yielding results of greater accuracy, especially at high lift ‘coefficients. Further refinement may be made by consideration of the chordwise distribution of lift (reference 80). ‘Wings with large emounts of sweep require special consideration (reference 81). 309 ‘The usual wing theory assumes that the resultent air force and moment on any wing section are functions of only the section lift, coefficient (or angle of atteck) and the section shape. According to this assumption, the air forces and moments on any section are not affected by adjacent sections ‘or other features of the wing except as such sections or features affect the lift distribution and thus the local lift of the section under consideration. ‘These assumptions ob- viously are not valid near wing tips, near discontinuities in deflected flaps or ailerons, near disturbing bodies, or for swings with pronounced sweep or sudden changes in plan form, section, or twist. Under such cireumstances, cross flows result in a breakdown of the concept of two-dimensional fiow over the airfoil sections. In addition to these cross flows, induced effects exist thet are equivalent to 2 change in camber. Such effects are particularly marked near the wing tips for wings of normal plen form end for ‘wings of low aspect ratio or unusual plan form. Lifting- surface theory (see, for example, reference 81) provides & means for calculating wing characteristics more accurately than the simple lifting-line theory. ‘Alihough span load distributions calculated for wings with discontinuities such as are found with partial-span flaps (roferences 82 and 83) may be sufficiently accurate for structural design, such distributions are not, suitable for predicting maximum-lift and stelling characteristics. Until sufficient data are obtained to permit the prediction of the meximumdlift and stalling charecteristics of wings with discontinuities, these characteristics may best be estimated from previous results with similer wings or, in the case of unusual configurations, should be obtained by test. ‘The characteristics of intermediate wing sections must be known for the application of wing theory, but data for sue sections are seldom available. Tests of a number of such intermediate sections obtained by several manufacturers for ‘wings formed by straight-line feiring heve indicated that the characteristics of such sections may be obtained with reasonable accuracy by interpolation of the root and tip cheracter~ istics eccording to the thickness variation, SELECTION OF ROOT SECTION ‘The cheracteristics‘of a wing ere affected to a large extent by the root section. In the ease of tapered wings formed by straight-line fairing, the resulting nonlinear variation of section along the span causes the shapes of the sections to be predominantly affected by the root section over e large pert of the wing area, The desirability of having a thick wing that provides space for housing fuel and equipment and re- duces structural weight or permits large spans usuelly leads to the selection of the thickest root section that is aerodynamically feasible. ‘The comperetively small variation of minimum drog coefficient with thickness ratio for smooth airfoils in the normal range of thickness ratios and the maintenance of high lift coeficient for thick sections with flaps deflected usually result in limitation of thickness ratio by characteristics other than maxiraum lift and minimum drag.310 ‘The critical Mach number of the section is the most serious limitation of thickness ratio for high-speed airplanes. It is desirable to select a root section with a critical Mach number sufficiently high to avoid serious drag increases resulting from compressibility effects at the highest level-Bight speed of the airplane, allowance being made for the increased velocity of flow over the wing resulting from interference of bodies and slipstream. Available data indicate that a smell margin exists between the critical Mach number and the Mach number at which the drag increases sharply. As airplane speeds increase, it becomes increasingly difficult and finally impos- sible to avoid the drag increases resulting from compressibility effects by reduction of the airfoil thickness ratio. In the eases of airplanes of such low speeds that compressibility considerations do not limit the thickness ratio to valuos Jess than about 0.20, the maximum thickness ratio is limited by excessive drag coeficients at moderate end high lift coefficients with the surfaces rough. In these cases, the actual surface conditions expected for the airplane should be considered in selecting the section. Consideration should also be given to unusual conditions such as ice, mud, and dainage caused in military combat, especially in the case of multiengine eirplanes for which ability to fly under such conditions is desired with one or more engines inoperative. In eases for which root sections having lange thickness ratios fare under consideration to permit the use of high aspect ratios, a realistic apprnisal of the drag coefficients of such sections with tho expected surface conditions at moderately high lift coefficients will indicate an optimum aspect ratio beyond which corresponding increases in aspect ratio and root thickness ratio will result in reduced performance. Inboard sections of wings on conventional airplanes are subject to interference effects and may be in the propeller slipstream. The wing surfaces are likely to be roughened by access doors, landing-gear retraction wells, and armament installations. Attainment of extensive laminar flows is, therefore, less Likely on the inboard wing panels then on the outboard panels. Unless such effects are minimized, little drag reduetion is to be expected from the use of sections permitting extensive laminar flow. Under these conditions, the use of sections such as the NACA 63-series will provide advantages if the sections ere thick, because such sections are more conservative than those permitting more extensive laminar flow. SELECTIOX OF TIP SECTION In order to promote desirablo stalling characteristios, the tip section should have a high maximum lift coefficient’ and a large range of angle of attack between zero and maximum lift as compared with the root section. It is also desirablo that the tip section stall without e large sudden loss in lift, ‘The attainment of a high maximum lift coefficient is often more difficult at the tip section than at the root section for tapered wings because of the lower Reynolds number of the tip section, For wings with small camber, the most effective way of increasing the section maximum lift coeficient is to increase the camber. ‘The amount of camber used will be limited in most cases by either the critical-speed requirements or by the requirement that the section have low drag at the high-speed lift coefficient, : REPORT NO, 824—NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS ‘The selection of the optimum type of camber for the tip section presents problems for whieh no categorical answvers ean be given on the basis of existing dais. ‘Tho use of a typo of camber that imposes heavy loads on the ailerons compli- cates the design of tho lateral-control system and increases its weight. ‘Tho use of a type of eamber that earries the lift farther forward on the section and thus relieves the ailerons will, however, have little offect on the maximum lift cocll- cient of the section unless the maximum-camber position is well forward, as for the NACA 230-scries sections. In this case @ stidden loss of lift at the stall may be expected. ‘The effects on the camber of modifications to the airfoil contour near the trailing edge, which may bo made in designing the ailerons, should not be overlooked in estimating the charae- teristics of the wing. If the root sections are at lenst moderately thick, it is usually desirable to select a tip section with a somewhat reduced thickness ratio. This reduction in thickness ratio, together with the absence of induced velocities from inte fering bodies, gives a inargin in critical speed that permits the camber of the tip section lo be increased. ‘This reduetion i thickness ratio will probably be limited by the loss in maximum lift coefficient resulting from too thin a section, A small amount of aerodynamfe washout may also be ‘useful as an aid in the avoidance of tip stalling. Tho permissible amount of washout may not be limited by the increase irinduced drag, which is smell for 1° or 2° of washout (reference 73). The limiting washout may bo that which causes the tip section to operate outside the low1. Considerable uncertainty exists regarding the correct numerical yelue of the coefficient occurring in equation (41). If e row of sources, rather then the Rankine ovals used in the present analysis, is considered to represent the effect of the wake, the value of the coefficient in equation (41) would be approximately twice the value used. Fortunately, the correction emounts to only about 2 percent at maximum lift for an extreme condition with @ 2-foot-chord model. Further refinement of this correction has thereforenot been attempted. COMPARISON WIFH EXPERIMENT A check of the validity of the tunnel-wall corrections has been made in reference 87, which gives lift and moment curves for models having various ratios of chord to tunnel height, uncorrected and corrected for tunnel-wall effects.318 REPORT NO, 824—NATIONAL ADVISORY COMBIITTEE FOR AERONAUTICS T 1 | | | | | y : +4 | iV eal 14 e i ASH f 7 y i ! i : | | 1 3 7 f / : L. 5 t CE pei ; T teat 6/8 ‘ at I o a ee ate tat (© Compton rat A. “ im | i : yal et Lf lA ul Z| | ECE i f | Bem wl, ea 68 OB Section ongie of attack, a, deg ‘igoas 7-Comparon between te edt fem Maloes ewreneats nd Teta onthe for and algo te tana 18.—Comparta beim ts obinne rm presse dtsbtonmessursents ands ebaled (2) Comperon fact . ren uct on the or and ogo tunel Ohare, in uncorrected for block corrected Yor blocking 2 | E} 13 ai t “ oe £ 7 4 le ‘Fovns 7,~Compaton baton sarc and unsorted prasure bute fe 9 hordes ef ymmael NACA Gace il of tbprcent tena ast?SUMMARY OF AIRFOIL DATA ‘The general agreement of the corrected curves shows that the method of correcting the lifts and moments is velid. ‘A comparison is made in reference 87 between the theoretical correction factor (equation (40)) and the experimentally derived corrections of reference 88. ‘The theoretical com rection factors were found to be in good agreement with those obtained experimentally. Tn order to check the validity of the r-factor, e comparison hhas been made of lift values obtained from pressure distributions with those obtained from thé integration of the floor end ceiling pressures in the tunnel. A comparison for two eirfoils given in figure 69 shows thet the two methods of ‘measuring lift give results that are in good agreement. ‘The rfactor bas also been checked by comperison of th lift obtained from belence meesurements with the integrating manometer values in figure 70. Finally, a check has been made of the method of correcting pressure distributions (reference 87) for NACA 6-series airfoils of two chord lengths at zero angle of attack in figure 71, in which the pressure coefficients are plotted against chordwise position 2/e. ‘The agreement between the corrected pressure distributions for both models verifies the method of making the tunnel-wall corrections. REFERENCES 1, Jacobs, Eastman N., Kenneth E., and Pinkerton, Robert ‘M.: The Charscteristios of 78 Related Alsfoil Sections from Tests in the VatiableDensity Wind Tunnel. NACA Rep. No. 460, 1983. 2. Jacobs, Eastman N., and Pinkerton, Robert M.: Teate in the Varlable-Dencity Wind Tunnel of Related Airfolla Having the ‘Masimum Cember Unusually Far Forward. NACA Rep. No. 587, 1985. 8, Jacobs, Eastman N., Pinkerton, Robert M., and Greenberg, Harty: Tests of Related Forward-Camber Airfolls in the Variable-Density Wind Tunnel, NACA Rep. No. 610, 1087. 4. Stack, John, and Von Doenhof, Albert E.: Tests of 16 Related Airfolls at High Speeds. NACA Rep. No. 402, 1934. 5. Jacobs, Eastman N., and Sherman, Albert: Airfoil Section Chatacteristies as “Afected by Vaviations of the Reynolds Number, NACA Rep. No, 586, 1987. 6. Pinkerton, Robert ML, and Greenberg, Harry: Aerodynamle Characteristics of « Large Number of Alsfolla ‘Tested in the Variable-Density Wind Tunnel. NACA Rep. No. 628, 1038. 1. Jones, B. Melvill: Plight Experiments on the Boundary Layer. Jour. Aero. Sel, vol. §, no. 8, Jan. 1988, pp. 81-01. 8. Jacobs, Eastman N,, and Abbott, Ira H.: Alrfoll Section Data Obtained in the N.A.C.A, Variable-Dencity Tunnel as Atected bby Support Interference and Other Correstions. NACA. Rep. No. 668, 1939. 9. Theodorsen, Theodore: Theory of Wing Sections of Arbitrary Shape. NACA Rep. No. 411, 1931 10, Stack, Joba: Tests of Aisolls Designed to Delay the Compress Ibllty Burble. NACA Rep. No. 768, 1948. 11, Jacobs, Eastman N.: Preliminary: Report on Laminar-Flow Alrfolls ‘and New Methods Adopted for Airfoil and Boundary-Laser Tavestigations. NACA ACR, June 1980. 319 12, Von Deenhof, Albert, and Stivers, Louis S,, Jr.: Aerodynamic ‘Characteristics of the NACA TATASIS and TATASIS Airfolls from Teats in the NACA Two-Dimensional Low-Turbulence « _ Pressure Tunnel. NACA CB No. 14125, 1044, 13, Naiman, Inven: Numerical Evaluation by Harmoate Analysis of the eFunetion of the Theodorsen Arbitrary-Airfoll Potential ‘Theory. NACA ARR No. L5HIS, 1945. 14. Theodorsen, Theodore: AirfolContour Modification Based on ‘Curve Method of Calculating Pressure Distribution. NACA ‘ARR No. L4G05, 1944. 16, Allen, H. Julian: A Simplifed Method for the Calculation of ‘Alrfoll’ Presaure Distribution. NACA TN No. 708, 1080. 16, Munk, Max Mf: Elements of the Wing Section Theory and of the Wing Theory. NACA Rep. No. 191, 1925. 17, Glauert, H.: The Elements of Aerofoll and Airserew Theory. Cambridge Univ. Press, 1926, pp. 87-08. 18, Theodorsen, Theodore: On the Theory of Wing Sections with Particular Reference to the Lift Distribution, NACA Rep. ‘No. $88, 1981. 19, Von Kérmés, Th: Compressiblity Effects in Aerodynamics. Jour. Aero. Be., vol 8, no. 9, July 1041, pp. 887-856. 20, Von Doenhoff, Aibert E.: A Method of Rapidly Estimating the Position of ‘the Laminar Separation Point. NACA TN No. 671, 1038, ‘21, Jacobe, E. N., and Von Doeahot, A. E.: Formulas for Use in ‘Boundary-Laer Calculations oi Low-Drag Wings. NACA ACR, Aug. 1941. 22, Von Decahot, Albert E., and Tetervin, Neal: Determination of General Relations for the Behavior of Turbulent Boundary Layers. NACA Rep. No. 772, 1048. 22, Squire, H. B,, and Young, A. D.: The Calculation of the Profile ‘Drag of Astofoils. Rd M. No. 1835, Britith AR, C., 1938, (2%. Tetervin, Neal: A Method for the Rapid Estimation of Turbulent ‘Boundary-Layer ‘Thfcknessea for Caloulating Profile Drag. NACA ACR No. LiG!4, 1944. 25. Quinn, Jobn HL, Jr, and Tucker, Warren A.: Scale and Turbulence Effects on "the Lift and Drag Characteristios of the NACA 65-418, 2= 1.0 Atoll Seotlon. NACA ACR No. IAE11, 1944, 26, Tucker, Warren A, and Wallace, Arthur R.: Sosle-Eteet Tests ina Turbulent Tunnel of the NACA 85-118, a=1.0 Airfoll Section with 0.20-Airfoll-Chord Split Flap. NACA ACR No. — TAI22, 1944. 27, Davidsoo, Milton, and Turner, Harold R., Jr: Effects of Meas Line Loading on the Aerodynamle Charucteristics of Some Low- Drag Airfoil, NACA ACR No. 3127, 1943. 28, Von Doenhofl, Albert E., and Tetervin, Neal: Investigation of ‘the Variation of Lift Coofisient with’ Reynolds Number at Moderate Angle of Attack on a Low-Dmag Alrfoll. NACA CB, Nov. 19:2. 29. Oswald, W. Bailey: General Formulas and Charts for the Caleula- tion of Airplane Performance. NACA Rep. No. 408, 1932. 80. Millikan, Clark B.: Aerodynamics of the Airplane. John Wiley & Sons, Yno., 1941, pp. 108-109. 81. Hood, Manley J.:’The Efects of Some Common Surface Irregularities on Wing Drag. NACA TN No. 605, 1930. 82. Loftin, Laurence K., Jr.: Effects of Specie Typea of Surface Roughness on Boundary-Layer Transition. NACA ACR No. 153200, 1046, 88, Charters, Alex C,,Jr.: Transition between Laminar and Turbulent ‘Flow by Transverse Contamination. NACA TN No. 691, 1983, ‘4, Braslow, Albert L.: Investigation of Efeets of Various Camoullage Paints and Painting Procedures on the Dreg Charactevisties of an NACA 6542-420, a= 1.0 Alrfoll Seotion. NACA CB No. TAGIT, 1044.320 85. Jones, Robert T., and Cohen, Doria: A Graphical Method of Determining Prosaure Distribution ia Two-Dimensional Flow. NACA Rep. No. 722, 1941. 36, abbott, Frank T., Jr, and Tumer, Harold R., Jr.t The Bates ‘of Roughness at High Reynolds Numbers on the Lift and Drag, Characteristics of Throe Thick Airfoil, NACA ACR No. TAH, 1945, 87. Jacobs, Eastman N., Abbott, Ina, H,, and Devidson, Milton: Investigation of Extreme Leading-Kage Roughness on Thick Low-Drag Airfols to Indioate Thoso Critieal to Separation. NACA CB, June 1942. 88, Zaloveli, Jobn A: Profile-Drag Coefiients of Conventional ‘and Low-Drag Airfolle a2 Obtained in Flight. NACA ACR No. 1AES1, 1084, 30, Zaloveik, John A., and Wood, Clotaire: A Flight Investigation of the Effect of Surface Roughness on Wing Profile Drag with ‘Transition Fixed. NACA ARR No. 14125, 1044, 40. Hood, Manley J., and Gaydos, M. Edward: EAects of Propellers ‘and of Vibration on the Extent of Laminar Flow on the N. A.C, A. 27-912 Airfoll, NACA ACR, Oct. 1089. 41, Sliverstein, Abe, Katzofi, 8, and Hootman, James A.: Com- parative Flight and Full-Scale Wind-Tunnel Messurements of ‘the Maximum Lift of an Airplane. NACA Rep. No. 618, 1988. 42, Sweborg, Harold H., and Dingeldein, Richard C.: Summary of ‘Meastrements in’ Langley FullScale Tunnel of Maximum Lift Cooficiente and Stalling Characteristis of Airplanes. NACA Rep. No. 828, 1966. 48, Purser, Paul E., and Johnson, Harold §.: Effects of ‘Trailing. ‘dge Modifications ‘on Piiching-Moment Characteristice of ‘Aisfols, NACA CB No, 14130, 1044. 44, Pullmer, Felicien F., Jr.: Wind-Tunnel Investigation of NACA (86(216)-216, 60,1-212, and 65;-212 Airfols with 0.20-Airfol- ‘Chord Split Flape. NACA CB No. AGO, 1044. 45, Abbott, Ira H., and Greenberg, Harry: Tests in the Variablo- ‘Density Wind Tunnel of the N. A C. A. 28012. Airfoil with Plain and Split Flaps. NACA Rep. No. 661, 1989. 46, Weozinger, Carl J, and Harris, Thomas A.: Wind-Tunnel Invest ‘gation of N. A’ C. A. 23012, 29021, and 29090 Alefoils with Various Sizes of Spllt Flap. NACA Rep. No. 608, 1989. 47, Bogdonoff, Soymour M.: Wind-Tunnel Investigation of » Low- ‘Drag Airfoil Seotfon with a Double Slotted Flap. NACA ACR No. 8120, 1943. 48, Wenzinger, Cart J., and Harris, Thomas A.: Wind-Tunnel Investi- ‘ation of an N. A. C. A. 29012 Ainfll with Various Arrangements of Blotted Flaps. NACA Rep. No. 664, 1088. 40, Weazinger, Carl J, and Harris, Thomas A.: WindTunnel Tnvesti- ‘gation of an N. A.C. A. 28031 Airfoil with Various Arrangements of Slotted Flaps. NACA Rep. No. 677, 1080. 50, Swanson, Robert S., and Crandall, Stewart M.: Analysis of Avail- ‘able Data_on the Effectivenoss of Allerons without Exposed Overheng Balance. NACA ACR No. LAEO1, 1944. 51, Street, William G., and Ames, Milton B., Jr.: Pressure-Disteibu- ‘ion Tnvestigation of an N. A.C. A.”0000 Airfoll with» 50- Pereent-Chord Plain Flap snd Three Tabs. NACA TN No, ‘a4, 1089. 52, Amos, Milton B., Jr, and Soars, Richard I.: Pressure-Distribution investigation of an N. A. C. A, 0000 Airfll with an 80-Pereent- Chord Plain Flap and Three Tabs. NACA TN No. 761, 1040. 58, Ames, Milton B., Jr, and Seare, Richard I.: Prossure-Distribution Tnveatigation of an N. A. C. A. 0009 Airfoil with 80-Peroent- Chord Plain Fisp and Three Tabs. NACA TN No. 759, 1940. 54, Sears, Richard I.: Wind-Tunnel Investigation of Control-Surface Characteristies. I—Effet of Gap on the Aerodynamic Chara teristioe of an NACA 0000 Alrfoll with a 80-Peroent-Chord Plain Flap, NACA ARR, June 1941. REPORT NO, 824—NATIONAL ADVISORY COMMOTTEE FOR AERONAUTICS 55. Jonos, Robert T., and Amos, Millon B,, Jr: WindTunnel Invos- tigation of ControlSurfaco Characteristics, V—The Use of Beveled ‘Trailing Hago to Reduce the Hing Moment of a Controt Surface. NACA ARR, March 1942, 50, Soare, Riohard I., and Liddell, Robert B.: Wind-Tunnol Invostiga- ton of Control-Surface Charaetorkities. VIA. $0-Poreent- Chord Plain Flap on the NACA 0015 Alrfoll. NACA ARR, June 1942, 817, Wenzinger, Carl J,, and Delano, James B.: Pressure Distribution overan N. A. C. A, 22012 Airfoll with a Slotted and a Plain Flap. NACA Rep. No. 038, 1988. 58, Gilli, Clarence L,, snd Lockwood, Vemart B.: Wind-Tunnel Tavestigation of Control Surface Characteristies. XITI—Varloun Flap Overhangs Used with a 80-Peroont-Chord Flap on an NACA 66-000 Airfoil. NACA ACR No, 8620, 1043. 50, Rogallo, F. M.: Collestion of Balanced-Aileron Test Data. NACA ‘ACR No. 4411, 1044. 60, Densei, H. G., and Bird, J, D.: Wind-Tunnel Taste of Ailerons at ‘Various Speoda. I[—Allerone of 0.20 Airfoll Chord and True Contour with 0.00 Aileron-Chord Sealod Internal Balance on the NACA 06,2-216 Airfoll, NACA ACR No. 3F18, 1943. G1. Purser, Paul B,, and Riebe, John M.: Wind-Tunnel Investigation of Control-Surfacs Chatacteristion. XV—Various Contour Modifications of a 0.80-AisfoilChord Plain Flap on an NACA 66(218)-014 Aisfoll. NACA ACR No. 8120, 19:3, 62, Braslow, Albert L.: Wind-Tunnel Investigation of Aileron Fites- ‘tiveness of 0.20-Alrfoll-Chord Plain Allerons of Trug Atrfoll Contour on NACA 65,-415, 65-418, and 08-421 Airfoll Sections, NACA CB No. L4H12, 1944. 68. Sears, Richard I, and Purser, Paul B.: WindTunnel Investigation ‘of Control-Surface Characteristen, XIV-—NACA 0009 Airfoil with a 20-Percent-Chord Double Plain Flap. NACA ART No. $F20, 1043. 64. Crane, Robert M,, and Holtzclaw, Ralph W.: Wind-Tunul Invos- tigation of the Eecta of Profle Modifications and Tabs on the Characteristics of Allerons on a Low Drag Aitfoll. NACA Rep. No. 808, 1048 86. Von Doenhoff, Albert E,, end Horton, Elmer A.: Prolinunary Investigation in the NACA Low-Turbulenee Tunnel of Low- ‘Drag-Alrfoll Sections Suitable for Admitting Alr at the Leading Edge. NACA ACR, July 1042 66. Jacobs, Eastman N., and Ward, Kennoth E.: Interference of Wing ‘and Fuselage from Tests of 300 Combinations in the N. A. C. A. Variable-Density Tunnel. NACA Rep. No. 540, 1995, 67, Abbott, Ira H.: Interferenoo Effests of Longitudinal Flat Plates on Low-Drag Airfoil. NACA CB, Nov. 1942. 68. Ellis, Macon C., Jr.: Some Lift and Drag Measurements of & Representative Bomber Nacelle on a Low-Drag Wing—tT. NACA CB, Sept. 1942. 60, Eile, Macon C., Jr.: Effects of e Typical Nacctle on the Charae- teristies of a Thick Low-Drag Airfoil Critially Affected by Leading-Bage Roughness. NACA CB No. 327, 1048, 0. Allen, H, Julian, and Prick, Charles W., Jr: Experimental Investi- gation of a New Type of Low-Drag Wing-Neeelle Combination. NACA ACR, July 1952, ‘71, Abbott; Frank, Jr.: Lift and Drag Data for $0 Pusher-Propollor ‘Shaft Housings on an NACA 65,8-018 Airfoll Soetion. NACA ACR No, 2X13, 1043. ‘72, Robinson, Russell G., and Wright, Ray H.: Estimation of Crittea! ‘Speeds of Airfolls and Streamline Bodies, NACA ACR, March 1940. "8. Andoreon, Raymond F.; Determination of the Characteristics of ‘Tapered Wings. NACA Rep. No. 572, 1935. ‘14, Jacobs, Eastman N., snd Rhode, R. V.! Alrfoll Section Charac- ‘terlatios as Applied to tho Prediction of Air Foreos and Thelt Disteibution on Wings. NACA Rep. No. 031, 1938,

Abbot and Doenhoff Naca Airfoil Data

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Abbot and Doenhoff Naca Airfoil Data

Uploaded by

Copyright:

Available Formats

You might also like