1、NASA Technical Memorandum 81912 Low-Speed Aerodynamic Characteristics of a 14-Percent-Thick NASA Phase 2 Supercritical Airfoil Designed for a Lift Coefficient of 0.7 Charles D. Hrrris, Robert J. McGhee, and Dennis 0. Allison L,atigIey ResearA Ceriter Hamptoir, Virgttiia National Aeronautics and Spac
2、e Administration Scientific and Technical Information Branch 1980 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Airfoil characteristics varied systematically Over the range of test condi- tions and behaved in a conventionally accepted manner. lift
3、coefficient was about 2.2 and occurred near M = 0.15 ber of 12.0 x lo6. Mach numbers and at all Reynolds numbers exhibited behavior generally associ- ated with gradual trailing-edge separation in the prestall angle-of-attack range, but the stall itself was more abrupt than would be expected of class
4、ical trailing-edge stall. The stall became more gradual at the higher Mach numbers. Drag remained essentially constant over a lift range which extended frtn near zero to beyond the design lift coefficient of 0.7 for a constant Reynolds num- ber with fixed transition. The maxirun aeasured at a Reynol
5、ds nut The variation of lift with angle of attack at the lower INTRODUCTION Continued development of supercritical airfoil technology has resulted in the design of family-related phase 2 supercritical airfoils. 10- and 14-percent-thick, designed for a lift coefficient of 0.7 have been tested at tran
6、sonic speeds in the Langley 8-Foot Transonic Pressure Tunnel, and the results were reported in references 1 and 2. Two such airfoils, The 14-percent-thick airfoil has also been tested at low speeds in the Langley Low-Turbulence Pressure Tunnel, and the results are presented herein. Included are the
7、effects of varying Reynolds nunber from 2.0 x lo6 to 18.0 x lo6 at a Mach number of 0.15 and the effects of varying Mach number from 0.10 to 0.32 at a Reynolds number of 6.0 x lo6. SYMBOLS Values are given in the International System of Units (SI) and in U.S. Cus- tomary Units. Measurements and calc
8、ulations were made in U.S. Customary Units. cP P - P, pressure coefficient, .hare 1) -0.7 design lift coefficient. 14-percent-thick SC (2) -071 4 supercritical (phase 2) -0.7 design lift coefficient, 14-percent-thick A mrted effort within the National Aeronautics and Qmce Achiniatra- tion (MA) over
9、the past several years has been directed toward developing practical trro-dirsnaional airfoils with good transonic behavior while retain- to as the supercritical airfoil. acceptable law-sped characteristics and has focused on a concept referred An early phase of this effort was successful in signifi
10、cantly extending drag-rise characteristics beyond those of conventional airfoils. (see ref. 4.) These early supercritical airfoils (denoted by supercritical (phase 1) prefix), however , experienced a gradual increase of drag (drag creep) at nacb numbers juat preceding the final drag rise. This gradu
11、al buildup of drag was largely associated with an intermediate off-design Mcond velocity peak (an mlera- tion of the flar over the rear portion of the airfoil just before the final recompression at the trailing dge) and relatively weak shock waves above the upper surface at these speeds. (See, for e
12、xample, ref. 5.) Improvements to these early, phase 1 airfoil. resulted in airfoils with significantly reduced drag creep characteristics. (See, for example, refs. 6 and 7.) These early phase 1 airfoils and the improved phase 1 airfoil8 of ref- erences 6 and 7 were developed before adequate theoreti
13、cal analysis codas were available and resulted frolr iterative contour rodifications during wind-tunnel exper inents. The process consisted of evaluating experimental pressure di8trf- butians at .on-carbide paper to provide an aerodynamically smooth finish. kO.10 lllp (k0.004 in.). 1 f The rodel con
14、tour accuracy was generally t hin Wind Tunnel The Langley Law-Turbulence Pressure Tunnel (ref. 9) is a cl red-throat, single-return tunnel which can he operated at stamation pressures fraa 10.1 to 1010 kPa (0.1 to 10 atm) with tunnel-empty test-section Mach numrbers up to 0.42 and 0.22, respectively
15、. per meter (15.0 x lo6 per foot) at a Mach number of about 0.22. test section is 91 cm (3 ft) wide by 229 cm (7.5 ft) high. The maximm unit Reynolds number is about 49.0 x lo6 The tunnel Hydraulically actuated circular plates provide posit .ning and attachment for two-dimensional models. The plates
16、 are 102 ca (40 in.) in diameter, rotate with the airfoil, and are flush with tbe tunnel wall. ment plates in the circular plates hold the nodel in such a way that the center of rotation for angle-of-attack adjustnent was at 0.25 on the nodel reference line. Rectangular model attach- A sketch showin
17、g the relationship between the ends of the model, the tun- nel walls, the model attachment plates, and the rotating circular plates is shown in figure 2. With this mounting system, the model completely spans the tunnel with each end anfined in a recess in the tunnel sidewall. At the side- walls, ins
18、ide joint sea- between mating pieces were sealed and faired smooth with model plastic and silastic rubber to minimize the effect of air leakage. Measurements Surface-pressure measurements.- Static pressures were measured on the surface of the model and used to determine local surface-pressure coeffi
19、cients. The surface-pressure measurements were obtained from a chordwise row of orifices located approximately 5 cm (2 in.) fran the tunnel centerline. Orifices were concentrated near the leading and trailing edges of the airfoil to define the pressure gradients in these regions, and a rearward-faci
20、ng orifice was included in the trailing edge. In addition, three spanwise OWR of orifices, located at 1-, lo-, and 75-percent chordwise stations, were included to establish the two dimensionality of the flaw over the model. 4 Provided by IHSNot for ResaleNo reproduction or networking permitted witho
21、ut license from IHS-,-,-Wake measurement.- Drag forces acting on the airfoil, as determined by the mentum deficiency within the wake, were der:ved from vertical variations of the total and static pressures aeasured across the wake with the wake survey rake shown in figure 3. positioned in the vertic
22、al centerline plane of the tunnel, one chord-length behind the trailing edge of the -1. The total-pressure tubes were 0.15 a (0.060 in.) in diameter and the static-pressure tubes were 0.32 cm (0.125 in.) in diameter. * The fixed rake, aounted from the tunnel sidewall, was The total-pressure tubes we
23、re flattened to 0.10 u (0.040 in.) for 0.61 Q (0.24 in.) from the tip of the tube. flush orifices, drilled 900 apart, located eight tube diaraeters from the tip of the tube in the measurement plane of the total-pressure tubes. The static-pressure tubes each had four Instrumentatior Measurements of t
24、he static pressures on the airfoil surfaces and the wake pressures were made by an autoaatic pressure-scanning system utilizing variable- capacitance-type precision transducers. with precision quartz mananeters. brated digital shaft encoder operated by a pinion gear and rack attached to the circular
25、 el attachment plates. tion system and recorded on magnetic tape. Basic tunnel pressures were measured Angle of attack was measured with a cali- Data were obtained by a high-speed acquisi- Reduction of Data and Corrections Calculation of cc, cn, and %- Section chord-force, normal-form, and pitching-
26、manent coefficients were obtained by numerical integration (based on the trapezoidal method) of the local surface-pressure coefficient measured at each orifice multiplied by an appropriate weighting factor (incremental area). Calculation of ca.- To obtain section drag coefficients, point drag ooeffi
27、cients were computed for each total-pressure measurement in the wake by using the procedure of reference 3. summed by numerical irtegration across the wake, again based on the trapezoidal method. These point drag coefficients were then Corrections for wind-tunnel-wall effects.- In the linear portion
28、 of the lift curve, corrections for lift effects and solid and wake blockage based on references 10 and 11 are small, on the order of 2-percent or less, and are usually neglected. As the model approaches maximum lift conditions where the lift characteristics become nonlinear and the viscous effect8
29、become signifi- cant, the assumptions underlying the corrections based on references 10 and 11 begin to break down and become inadequate. For these reasons, the data presented herein is uncorrected for tunnel-wall effects. however, to indAcate the effect corrections would have on the aerodynamic cha
30、r- acteristicc if they were applied. Figure 4 is presented, 5 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-?ZST CONDITIONS Tests were conducted over a range of Reynolds numbers based on the wdel chord from about 2.0 x lo6 to 18.0 x lo6 at a Nach n
31、umber of 0.15. was also varied frola about 0.10 to 0.32 at a Reynolds number of 6.0 x lo6. airfoil was tested smooth (natural transition) and 4th transition trim fixed along the 5-percent chordline on the upper and lomr surfaces. The traneition trips consisted of sparsely distributed 0.13-cla-wide (
32、0.05- In.) bandp of car- borundum grains, sized according to the technique in reference 12 and attached to the surface with clear lacquer. Mach number The PRESENTATION OF RESULTS The results of this investigation are arranged in the following figures: Figure Effect of angle of attack and Reynolds nu
33、mber on the up;?er-surface spanwise pressure distributions. M . 0.15 . Tuft photographs of the effect of angle of attack on the upper- Effect of fixing tranuition on section characteristics. M = 0.15 Effect of fixing transition on chordwise pressure distributions. M=O.lS Variation of pressure coeffi
34、cient on upper surface at particular chordwise stations with angle of attack. M - 0.15, model smooth Variation of pressure coefficient on upper surface at particular chordwise stations with angle of attack. M = 0.15, transition fixed . M - 0.15, transition fixed surface flaw pattern. M . 0. 15; R .
35、2.0 . 106 Effect of Re-ynolds number on section Characteristics. Effect of angle of attack and Reynolds number on the chordwise Effect of Reyr.olds number on base and upper-surface pressure Variation of minimum drag coefficient with Reynolds number. Variation of maximum lift coefficient with Reynold
36、s number. Effect uf Reynolds number on chordwise pressure distribution. Variation of sectlon drag coefficient with Reynolds number. Effact of Mach number on section Characteristics. Variation of maximutn lift coefficient with Mach number. Effect of Mach number on the chordwise pressure distributicc
37、for R = 5.0 x lo6, transition pressure distribution. M . 0.15, transition Fixed . coefficients. M . 0.15, transition fixed . MtO.l5eoeo M . 0.15, transition fixed . M . 0.15, a . 8O, transition fixed M . 0.15, cl . 0.0 ?nd 0.7, transition fixed . R . 6.0 . lo6, transition fixed R = 6.0 x 196, transi
38、tion fixed . angles of attack near maximum lift. fixed . M . 0.32, Rffef-t of angle of attack on the chordwise pressure distribution. R . 6.0 . lo6, transition fixed . . 5 . 6 0 7 . 8 . 9 . 10 . 11 . 12 . 13 . 14 . 15 . 16 . 17 . 18 . 19 . 20 . 21 6 Provided by IHSNot for ResaleNo reproduction or ne
39、tworking permitted without license from IHS-,-,-Figure Bffect of Nach nwrb.r on the chordwire prerrur, dirtxibution. R - 6.0 x lo6, a - 8O, tranrition fixd . . . . . . . . . . 12 Effect of Reynold8 number on cl,mX. N (I: 0.15, modal smooth . . . . 23 DISCUSSION Two Dimensionality of Flow The spanwir
40、e prerrure dirtributionr shown in figure 5 indicate that the flow over the model renained two dimenrional to angles of attach at which muti- mum lift coafficientr occurred. For example, the flow at R = 2,O x 106 and a = 16O figure 5(a) to be two dimnrional in character even though a considerable reg
41、ion of reparstion over the rear upper surface is rhown in the tuft phot+ graphs of figure 6. the other test Reynolds numbers (fig. 11) with the corre8ponding rpanwi8e prer- sure distributions in figure S indicate similar rerultr. (shown in fig. 11 to be near maximum lift conditionr) is Jnlicated in
42、Correlation of the angles of attack for mxiraw lift for In addition to the flow over the airfoil remaining tm, dimenrional to angler of attack beyond maximum lift, the tuft photograph8 for (fig. 6) show %it the flow on the tumel sidewall above the rodel remains attached until an rnsition was obrerve
43、d. The low Reynolds number and the near set0 prerrure gradient8 wuuld encourage a long run of laminar flow before natural tranrition would OCCC. for the smooth mode!. 7 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-i These effects generally decreas
44、ed with increases in either angle of attack or Reynolds number. The reduced effects with increased angle of attack are associated with the steep adverse pressure gradient of the leading-edge suction peak at high angles which causes natural transition to accur near the leading edge on the upper surfa
45、ce. The reduced effects with increased Reynolds number are due to the thinner boundary layers, the tendency for natural transition to occur nearer the leading edge, and the higher wind-tunnel turbulence level which would also induce earlier transition. The small differences at the higher angles of a
46、ttack may not be completely explained because of the inability to determine the relative positions of boundary-layer transition between free and fixed transition on the lower sur- face. This inability to determine lower-surface boundary-layer transition at high angles of attack is due to the favorab
47、le pressure gradient over a large portion of the lower surface and to the fact that the lower-surface transition strip was ahead of the stagnation point at high angles and, therefore, MY not have properly trippet the lower-surface boundary layer. The general effect of fixing t:ansition and its trend
48、 with ihcreasing Reynolds number reverses at high angles of attack for in figdre 7(f) are increased lift, increased pitching moments, and reduced drag with fixed transition. These results are believed to be associated with the elimination, or reduction in size, of a small laminar-separation bubble o
49、n the airfoil upper surface by the introduction of roughness near the stagnation point on the lower surface. Tabulated pressure distributions (not presented) indicate that at a = lZ0, where the lift curves of figure 7(f) first diverge, the stagnation point has moved to a lower-surface location corresponding to the trip location (0.05). The presence ,f a laminar-separation bubble can only be ass?tmed since there ere no discernible dismntinuiti
