A critical assessment of wind tunnel results for the NACA 0012 airfoil

~~ ~~ ~~~

USAAVSCOM Technical Report 87-A-5

NASA Technical Memorandum 10001 9

A Critical Assessment of Wind

Tunnel Results for the NACA

0012 Airfoil

W. J. McCroskey, AeroflightdynamicsDirectorate, U.S. Army Aviation Research and

Technology Activity, Ames Research Center, Moffett Field, California

October 1987

US A National Aeronautics and AVlA Space Administration SYSTEMS COMMAND AVIATION RESEARCH AND Am= Research Center TECHNOLOGY ACTIVITY MoffettField, California 94035 M0FFEl-r FIELD. CA 94305-1099 1-1 A CRITICAL ASSESSMfNT OF WIND TUNNEL RESULTS FOR THE NACA 0012 AIRFOIL.

U. J . McCroskey U . S . Army Aerofl ightdynuics Directorate (AVSCOW) NASA Arrs Research Center. N258-1 Moffett Field, California 94035. USA ABSTRACT A large body of experimental results, which were obtained in more than 40 wind tunnels on a single.

An assessment of well-known two-dimensional configuration. has been critically examined and correlated.

some of the possible sources of error has been made for each facility. and data which are suspect have been identified. It was found that no single experirnt provided a conplete Set of reliable data.

although one investigation stands out as superior in many respect$. However. f r m the aggregate of data the representative properties of the NACA 0012 airfoil can be identified with reasonable confidence over wide ranges of Mach number. Reynolds number. and angles of attack. This synthesized information can n w be used to assess and validate existing or future wind tunnel results and to evaluate advanced Compute- tional Fluid Dynamics codes.

I . INTRODUCTION Reliable determination and assessmnt of the accuracy of a W O d y n M i C data generated in wind tunnels remains one of the most vexing problems in aeronautics. Aerodynamic results a n seldom duplicated in different facilities to the level of accuracy that is required either for risk-fne engineering develop- ment o r for the true verification of theoretical and numerical methods. This shortcoming i s particularly acute with regard to today's rapid proliferation of new Conputational Fluid Dynamic (CFD) codes that lack adequate vel idat ion 11 I.

On the other hand. the NACA 0012 profile is one of the oldest and certainly the most tested of all airfoils: and it has been studied in dozens of separate wind tunnels over a period of lore than 50 years.

Although no single high-quality experiment spans the complete subsonic and transonic range of flow condi- ' tions. the combined results of this extensive testing should allow S O W conclusions to be drawn about (2-0) testing.

wind-tunnel data accuracy and reliability. at least for tW-dilcnSiOMl This paper attelpts to extract as much useful, quantitative information as possible from crltical examinations and correlations of existing data f r m this single, well-known configuration. obtained in over 40 wind tunnels and over wide ranges of Mach number. Reynolds number. and angles of attack.

A preliminary caparison by the author 121 in 1982 of results f m rbart a dozen widely-quoted inves- tigations for the NACA 0012 airfoil revealed significant and unacceptable differences between wind tunnels. and subsequent examinations of more data sets merely colpounded the confusion, as Indicated in Figs. 1 and 2 . Therefore. a major part of the present investigation was the development of a filtering process for screening the available data and classifying the experimental sources into broad categories of estimated reliability. This process is described in the next section. Detailed caparisons. correla- tions, and uncertainty estimates are discussed in subsquent sections, where the the following results are considered: 1. Lift-curve slope versus Mach and Reynolds number 2 . Minimum drag versus Mach and Reynolds number 3 . Maximum lift-to-drag ratio versus Mach and Reynolds nuder 4 . Maximum lift versus Mach and Reynolds number

5. Shock-wave position versus Reynolds nunber at W - 0.8

As this list indicates. the present study deals mostly with the integral quantities. lift and drag.

Despite the large number of references available on this most popular of all airfoils, it was found that there is insufficient overlap in the experiments to make many meaningful, direct comparisons of more detailed quantities. such as pressure distributions, in the transonic regime. It is acknowledged that pitching moment is also a sensitive integral parmeter that displays interesting transonic behavior. but is not considered in this paper.

C , 11. THE FILTERING A N 0 ANALYIS PROCESS The main objective of this section is to c o d i n e the critical. relevant information that is available on airfoil testing and on airfoil aerodynamic behavior into a systematic screening. or "filtering." pro- cess that can be used to assess the quality of individual experimental Sources of data. This process will then be used to classify each data set and to weigh the accuracy of those data against the quantitative or qualitative infomation that they can provide about the aerodynamic characteristics of the NACA 0012 d i rfoi 1 .

+Presented at the AGARD Fluid Dynsnics Panel S T o s i u m on "Aerodynamic Data Accuracy and puality: Requirements and Capabilities in Wind Tunnel Testing, Naples. Italy, 28 September-2 October 1987.

1-2

0 Group 1 data, Y < 0.55

0 Group2 A Group3 .16 0 Group4 M other NACA 0012 experimenb ---Homer. corralation o f 12% airfoil arp.

.14 ------hVirdd T h 4 0 ~

............ p l , I 2 g r

... B ............

.............

.

A .10 -. 8

Y

a

.m -

I I Fig. 1 . Lift-curve slope at zero lift a. Reynolds n u m b : all data. A( c 0.55. L q j m d explained in

T h l 8 S 1 - 4 , 0 fl

.a-

ctoup2 A A Croup3 0 Group4

____-_

BClr - 2=

-

.20

-

.15

.---

-

- a .lo

Y

-

.05 0 - 0 .2 .4 .6 . E 1 .o 1.2 MACH NUMBER Fig. 2. Lirt-curvr slope va. Mach number: all data. L e g a d explained in Tables 1-4.

1-3 ORPGmm PAGk I,.,

A. Develowent of the Process OF POOR Q'FJALETY

The c r i t i c a l information used i n the development of the process i s derived frm four broad categor- ies, as follows: A very large c o l l e c t i o n o f wind-tunnel data f o r the NACA 0012 which varies widely f o r many 1.

possible reasons.

2. A modest c o l l e c t i o n of "facts." 1.e..

a. well-established theories and s i m i l a r i t y laws b. generally-accepted empirical laws c. recent advances i n identifying, analyzing. and correcting f o r wind-tunnel w a l l effects.

3 . A fuzzy c o l l e c t i o n o f " f o l k l o r e " about a i r f o t l behavior, t e s t techniques. and wind-tiinnel character i s t i c s .

4. Recent CFD r e s u l t s f o r a few standard a i r f o i l cases i n both simulated f r e e - a i r conditions and combined airfoil/wind-tunnel i n s t a l l a t i o n s .

This aggregate o f information f i m l y establishes some i l p o r t a n t sources o f wind-tunnel errors and c e r t a i n properties o f a i r f o i l s such as the NACA 0012. This knowledge can be sunnarized as follows: f i r s t , a l l four wind-tunnel walls generally i n t e r f e r e w i t h the flow around the a i r f o i l . and t h i s p h e m - non i s generally more acute than f o r three-dimensional ( 3 - 0 ) bodies. The top and bottom walls particu- l a r l y a f f e c t the e f f e c t i v e angle o f attack. the shape o f the pressure d i s t r i b u t i o n (and hence pitching- moment c o e f f i c i e n t ) , and the shock-wave location. and t o a lesser extent, lift. drag. and effective Mach number. S o l i d walls increase the e f f e c t i v e o and Mach number. but these effects are considered t o be e a s i l y correctable. a t least i n subsonic and m i l d l y transonic flows. Slotted or porous walls lower the e f f e c t i v e a; attemps are o f t e n made t o correct f o r t h i s . but It i s d i f f i c u l t .

Second, side-wall boundary layers have been shown t o lower C , , cde and the effective H. and t o move the shock forward. Flow Separation a t the a i r f o i l - w a l l juncture affects the shock location and reduces . The effects can be reduced substantially by the application of suction on the side walls, and corrections can be applied i f there i s no separation i n the corners.

Third, free-stream turbulence and boundary-layer t r i p s increase cd and o f t e n a f f e c t C , . f. and shock location. Many a i r f o i l s , including the NACA 0012. nay be p a r t i c u l a r l y sensitive t o Reynolds number variations i f M t r i p i s used: however. extreme c a n must be exercised i n t r i p p i n g the boundary layer t o avoid causing excessive drag increments and erroneous changes i n The e f f e c t s o f and shock position.

C , both t r i p s and turbulence are d i f f i c u l t t o quantify.

Concerning a i r f o i l behavior. two important "facts" have been established about the behavior o f l i f t and drag i n subsonic flow a t s l l a l l angles o f attack. A t high Reynolds numbers, both cd a t zero l i f t and the quantity m C i are independent o f M and are only wcakly dependent upon Re. Unfortunately.

most other aspects o f a i P f o i l characteristics are not as f i r m l y established, and even these two quantities are not well defined I n transonic flow. However. measurements of general trends and q u a l i t a t i v e behavior are generally accepted. even i f the absolute values of C , . cd. and &. f o r example. are uncertain.

To inprove on t h i s situation. the following f i l t e r i n g or screening process i s proposed. F i r s t , an attempt w i l l be made t o i d e n t i f y the highest-quality experiments i n which the aforementioned wind-tunnel problems were c a r e f u l l y controlled, corrected for. or otherwise ameliorated. Second, the r e s u l t s of these tests w i l l be used t o establish the quantitative. "factual.' behavior o f the c r i t i c a l parameters Cd and eC, , where B - m, as functions o f Re i n the subsonic regime where they are e s s e n t i a l l y indgpendent of H. This information comprises the f i l t e r s t h a t are necessary, although not s u f f i c i e n t , screening c r i t e r i a f o r judging the c r e d i b i l i t y o f the remaining data. Third. these f i l t e r s w i l l be used t o help i d e n t i f y obviously erroneous aspects o f a l l the data sets and t o c l a s s i f y each experiment accord- ingly. Fourth, a l l the data w i l l be c r i t i c a l l y examined outside the range o f Mach and Reynolds numbers f o r which the f i l t e r s were developed. Finally. a subjective extension o f the f o u r t h step w i l l be nude.

The " f o l k l o r e " correlations and other information referred t o above. and established transonic s i m i l a r i t y laws. w i l l be used t o combine selected NACA 0012 and other a i r f o i l data i n order t o estimate the transonic properties of the NACA 0012 over a range o f Mach numbers. 0.85 < M < 1.1. f o r which v i r t u a l l y no r e l i a b l e data e x i s t .

B. Application of the Process Table 1 l i s t s and summarizes the experiments which c l e a r l y stand out as having been conducted w i t h the utmost care and/or as most nearly eliminating the important sources o f wind-tunnel errors. These sources are referred t o throughout t h i s paper as Group 1 . It w i l l be noted frm Table 1 that. unfortu- nately. only one of the experiments extends s l i g h t l y i n t o the transonic regime. and t h a t the turbulence level i n t h a t t e s t was r e l a t i v e l y high. Also, f o r the present purposes. it i s unfortunate t h a t the only data reported from t h a t experiment were obtained w i t h -a boundary-layer t r i p , although some unpublished data were also obtained without a t r i p .

1-4 The r e s u l t s f o r 6C f r o n Group 1 are p l o t t e d versus Re i n Fig. 3. It i s clear that the r e s u l t s ' a shown i n t h i s f i g u r e represent a major laprovemen over the l a r g 1 . A good f i t o f the scatter i n Fig.

l i f t - c u r v e slope data i n the l i m i t e d range 2 M 10' < Re < 2 x 10 9 i s given by 6 C 1 , = 0.1025 + 0.00485 Log(Re/106) per degree (1) w i t h an ms standard e r r o r o f 0.00024 and a maXiUm e r r o r of 0.0029 f o r the 30 points shown.

Similarly, the r e s u l t s f o r C are p l o t t e d i n Fig. 4. The meaning o f the various groups i s d0 explained below. The drag data from Group 1 without a boundary-layer t r i p , i.e. thc open circles, can be approximated w e l l by = 0.0044 + 0.018 (2) cdO w i t h an rms standard e r r o r o f 0.00005 and a maxlum e r r o r of 0.0007 f o r the 36 points frm Group 1. The data w i t h a boundary layer t r i p show a greater s e n s i t i v i t y t o Reynolds number.

In accord w i t h the approx- imate v a r i a t i o n o f f u l l y turbulent s k i n f r i c t i o n w i t h Reynolds number 131. a good f i t t o the Group 1 tripped data i s given by

, = 0.0017 + 0.91/(Log Re)2-58 (3)

cdO where the constant 0.0017 was chosen t o optimize the curve fit shown i n Fig. 4 .

For reference, i t i s estimated t h a t the individual ValUeS o f SC and cd can be detennined o r LlJ 0 calculated from the individual Group 1 data points t o an o v e r a l l precision o f about iO.0005 and iO.0002.

respectively. It may be mentioned t h a t Ref. 4 l i s t s the desired accuracy o f cd from wind tunnels as 0.0005 f o r the assessment o f configuration changes and 0.0001 f o r the v a l i d a t i o n o f C F O codes.

The i n f o r m t i o n i n Eqns. 1-3 can n o w be used t o assess the accuracy o f the data from the remaining sources and t o group the data i n t o separate categories. After much deliberation, it was decided t o define Group 2 as comprising those data which generally agree uith the l i f t and drag c r i t e r i a expressed i n Eqns. 1-3. t o w i t h i n iO.0040 f o r eC and t o w i t h i n t0.0010 f o r t d . These e x p e r i v n t s are l i s t e d i n 1, Table 2. Foremost i n t h i s group i s the experiment of C . 0. Harris 157. Although t h i s experiment was c a r e f u l l y conducted and offered the advantage o f a large aspect r a t i o , l i f t - i n t e r f e r e n c e corrections on the order o f 15% are r q u i r e d f o r the angles o f attack. These were & a a j o r concern h i t i a l l y . but i n the subsequent discussions and figures i t w i l l becolc evident t h a t these r e s u l t s are comparable i n accuracy t o those o f Group 1 .

.14 .13 .12 .

............................................................. ..- _.... ".B" ....._..

$? .ll

v -o----&B ---y-#--$----

a

--P 0

.10 .08 I 106 1 o7 1 08 Lop. Ro Fig. 3. L i r t - c u m slope at zero lift vs. Reynolds number: Croup 1 data, M < 0.55. Expanded vertical s a l e .

1-5 0 Group 1 Data, no trip . 0 1 ! i a Group1,rithtrip '

0 Group 2. Y < 0.7. no trip

e Group 2. Y < 0.7, with trip

A YAW& GD HSWT (Croup 3 ) ; Y < 0.7, no trip

.01#

------

-fit Croup 1 . no trip

-.- curve-fit croup 1, with trip

.OlW 'do .W75 .w50 .W25 I 1 1 Fig. 4. Drog coefficient at zom l i f t vs. Royndds number.

Several sources provide data t h a t agree w e l l w i t h the Group 1 r e s u l t s f o r 6C E C but 'a do' not f o r both. I n some cases, only one o f these key quantities WM wasured. These are c l a s s i f i e d as Group 3 and are, l i s t e d i n Table 3. A n exanple o f t h i s group i s the essentially interference-free experi- ment o f Vidal e t a l . 161. which provides good lift data, but which used a large t r i p t h a t evidently pro- duced excess drag.

A feu sources provided data t h a t generally s a t i s f y the basic lift and/or drag c r i t e r i a outlined above, but f o r which other major problems have been identified. I n addition, a s i g n i f i c a n t nunber o f t e s t s f a i l t o s a t i s f y cithcr o f these two c r i t e r i a . but they do cover ranges o f Mach number where even q u a l i t a t i v e i n f o r n a t i o n i s helpful. These sources are referred t o as Group 4 and are b r i e f l y su*l*irlzed i n Table 4. f i n a l l y , s t i l l other sources were exmined t h a t f a i l e d t o satisfy the c r i t e r i a . and which d i d not appear t o o f f e r any s i g n i f i c a n t additional i n f o r m t i o n relevant t o the present investigation. For information purposes these are l i s t e d i n Table 5, but t h e i r r e s u l t s are not used i n t h i s paper.

111. RESULTS AN0 DISCUSSION I n t h i s section, the r e s u l t s from Groups 1-4 and from the other sources alluded t o Section 1 I . A are used c o l l e c t i v e l y t o establish the p r i m a l j characteristics o f the NACA 0012 a i r f o i l over a wide range o f Mach number. Reynolds nulber, and angle o f attack.

A. Lift-Curve Slope. dC,/do f i g u r e 5 shows the data frol Groups 1-3 f o r 6C as a function o f Reynolds wnber, f o r M 0.55.

'n H a r r i s ' r e s u l t s 151. a t Re = 3 and 9 lo6, are highlighted by s o l i d symbols. and t h i s convention w i l l be followed i n most o f the remaining figures. The scatter i n the Group 2 data i s s l i g h t l y greater than .

seems t o be established now over the that o f the Group 1 results. but the q u a n t i t a t i v e behavior o f 6C 'n range o f most wind-tunnel' t e s t s f o r aeronautical purposes.

i s i l l u s t r a t e d i n Fig. 6. where the relevant Group 3 data The c a p l e x transonic behavior o f C , have been added. This f i g u r e c l e a r l y repre&?nts a major Iaprovcwnt over f i g . 2. For these conditions, the good agreement between Harris' r e s u l t s 151 and those o f Green and Newan 171 c o n s t i t u t e further v a l i - dation o f the f o m r . The largest discrepancies t h a t rcnrain occur w i t h the data frun Vidal e t al. 161 below M = 0.8. which seem t o be mostly a Reynolds-mder effect, and Sawyer 181. who reported large values a t M = 0.8. It Is unclear whether t h i s i s due t o side-wall interference, o r smething else. But i n a l l cases. the peak i n C occurs a t M = 0.80 3.01.

'a 1-6

-

.14

0 Croup 1 Data, M < 0.55

0 Harrir, LaRC 8TPT (Group 2)

0 Remainder of Croup 2

.13 0 Croup3Data

------ m e r i t of croup i

.....- I^.. CpCk = 2n

I- I .12

.................. o @

^................................._..................

-eg---$-Q _--- -0------

5 .llppl&

I I 4 x 105 106 lo7 Log. RI ' Fig. 5. Lilt-curve slope vs. Reynolds number. Same r o l e s at Fig. 1.

0 Harris,IaRC8TPT(Boup2) 0

, -

.25

0 Group 2 Data, Re > 1.5.10'

0 Group 3 Data. E ; e > 2x10'

e vidal. CALSPAN 8' ( Croup 3 ), Re-lo'

-

.20

------ cume-fit of Croup 1. Re = 5x10'

.15

L----aeeqy e*-%

a l **

o 0

.05 0 .

-.a I I I I 1 I I 1

0 .2 . 4 .6 .8 1 . 0 MACH NUMBER Fig. 6. Lift-cwve sIope vs. Mach number.

The data i n Fig. 6 indicate rapid variations w i t h Mach nulber I n the narrow range 0.8 < M c 0.9.

Unfortunately, the Group 2 and 3 data are very sparse i n t h i s region. and are nonexistant above M = 0.95..

Therefore, an attenpt was made t o extract selected additional i n f o m a t i o n from the Group 4 data and from other sources, as discussed above.

Three points are relevant here. F i r s t , I n the transonic p o r t i o n o f Fig. 2. the r e s u l t s of Scheitel & Wagner 191 can be argued t o be the most r e l i a b l e o f the Grwp 4 measure- ments, because side-wall suction was used and because t h e i r r e s u l t s are lore nearly consistent w i t h the Group 2 and 3 data where there i s some overlap.

Second. a l l of the supersonk data points o f Group 4 are i n good agreement w i t h one another and w i t h the s i m i l a r i t y c o r r e l a t i o n given below which encompasses other synnetrical a i r f o i 1s I 10 .ll I , = 0.0551(y + l)M2t/cl-"3 210% ( 4 ) c,o 1-7 It must be noted t h a t t h i s simple r e l a t i o n i s only v a l i d i n the low supersonic range. 0 . 1 < < 1 . where R = (M2 - l)[(r + l)d;/~I-"~, and although i t i s based on transonic s i m i l a r i t y . the thickness correla- t i o n breaks down f o r M < 1 1101.

A t h i r d i r p o r t a n t aspect o f Figs. 2 and 6 i s the behavior around M - 0.9. There i s a wide v a r i a t i o n i n the minimum value o f C and i n the Mach number a t which t h i s occurs; and Refs. 9 and 12 o f Group 4 .

'a and Ref.

C , . This phenomenon was investigated b r i e f l y i n

13 o f Group 5 reported negative values o f Ref. 1 4 . wherein Navier-Stokes calculations a t M = 0.88 and o = 0.5' produced a l a r g i M l l y - S t a b l e solu- t i o n w i t h C = 0. These calculations were repeated recently w i t h a time-accurate code, and t h i s t i m e they producet an unsteady s o l u t i o n w i t h periodic o s c i l l a t i o n s w i t h an m p l i t u d e o f AC, = 0 . 1 around a mean value of approximately zero. This behavior appears t o be q u a l i t a t i v e l y the s m e as the transonic - self-induced o s c i l l a t i o n s reported on a biconvex a i r f o i l by Levy 1151 and i n several subsequent investiga- tions. On the other hand. only 'steady. r e s u l t s have been reported i n the NACA 0012 experiments. and t h i s nay have been overlooked. F u r t h e m r e . It i s not known what e f f e c t the wind-tunnel unsteady behavior W d l l S may have. Considering these factors. it i s the author's subjective opinion t h a t the correct value o f C i s a m i n i l u m value somewhere between 0 and -0.05, occurring a t behavior f o r the 'a M = 0.88 20.02. This area needs f u r t h e r investigation.

Figure 7 shows the collective. * f i l t e r e d . i n f o m a t i o n described above i n the Mach number range from 0.6 t o 1.2. including the author's judgement o f the upper and l o w r bounds o f the correct transonic l i f t characteristics o f the NACA 0012 a i r f o i l a t moderate Reynolds numbers and small angles o f attack. I n sm- mdry. the m s t inportant points are the following: 1. I n the subsonic range M < 0.5, C i s given by Eqn. 1 t o w i t h i n 62%.

'a 2. The maximum value o f C i s 0 . 2 1 r5% and it occurs a t M = 0.80 60.01.

'a 3. The minimum value o f C i s -0.025 t O . 0 2 5 and i t occurs a t M = 0.88 20.02.

' a 4. A secondary m a x i m i n C occurs near M = 1 . w i t h a value o f 0.09 210%.

'a 1.05 c M < 1 . 2 . C i s given by Eqn. 4 t o w i t h i n 210%.

5. I n the low supersonic range ' 0 These estimates represent the maxilun precision that can be extracted from the e x i s t l n g information, and i s probably the best absolute accuracy t o which interference-free l i f t can be measured they represent what on a i r f o i l s i n wind tunnels today f o r an a r b i t r a r y angle o f attack.

9. M i n i n u m Drag, cdo The baseline i n f o m a t i o n for t h i s fundamental quantity i n subsonic f l o w was discussed e a r l i e r i n connection w i t h Fig. 4 . Although the data f r o m Groups 1 and 2 are self-consistent. the scatter i n the from Groups 3 and 4 (not shown), owing t o free-stream turbulence, surface roughness and/or bound- r e s u l t s ary layer t r i p s , w a l l interference, and measurement errors. would a l N S t t o t a l l y mask the v a r i a t i o n o f drag w i t h Reynolds number. Numerical r e s u l t s congiled by Holst 1161 i n h i s recent v a l i d a t i o n exercise f o r cd l i e s between the values given by transonic viscous a i r f o i l analyses, suggest that f u l l y - t u r b u l e n t Eqns. 2 and 3, but t h i s has not been validated adequately.

Another i n t e r e s t i n g s i t u a t i o n i s the transonic drag rise, Fig. 8. f o r which only a l i m i t e d number o f high-quality sources are available. Here the scatter i s excessive, but below M = 0 . 7 , each individual data set seems t o be essentially. independent o f Mach nuber.

This suggests subtracting out an average o f the subsonic values f o r any given data set, as follows:

A c = c (M) - td (M)

do do 0 i s the average o f the measurements f o r M < 0 . 7 .

where The r e s u l t s o f applying t h i s procedure are shown i n Fig. 9. which I s an obvious i l p r o v c c r n t over The drag-divergent Mach nullber fig. 8. Remarkably. even the Group 3 data are i n good agrement f o r Acd .

can now be estimated a t Mdd = 0 . 7 7 t O . 0 1 , w i t h a m a l 1 cwunt o f drag creep f o r M > 0 . 7 2 .

The behavior a t higher transonic Mach numbers i s nrch more d i f f i c u l t t o establish. A l l o f the data frm Groups 1-4 are p l o t t e d I n Fig. 10. along w i t h estimates based on transonic s i m i l a r i t y correlations o f data frola many other symmetrical a i r f o i l s 1 1 0 . 1 1 . 1 4 . 1 7 - 2 0 1 . These l a t t e r sources indicate t h a t a i r f o i l behavior i n the low superonic region I s given by from source t o source. but which i s bounded by about 4 . 0 and 5.6 .

where a i s a "constant' t h a t varies The dashed l i n e i n Fig. 10 i s f o r a = 4.8.

Data from Groups 1-4 do not extend beyond M = 0.95. Between M = 0.8 and 0.9, when cd I s r i s i n g rapidly, there i s a large amount o f scatter, and the uncertainty i n the measurements i s v i r t u a p l y inpossi- b l e t o assess. The s o l i d l i n e s represent the author's subjective judgement o f the probable upper and 1-c ' 0 H a d , LaRC 8TPT ( Croup 2) CB Vidal, Calapan 8' ( Croup 3 ), Re-10' 0 0 OtherCroup3Data 0 ScheiteldcWagner. 'I" (Group 4)

X Other Croup 4 Data, Y > 0.97

Ransonic iimilarity arr. Y > 1

.20 .15 I

-.os

.6 .7 .8 .9 1 .o 1.1 1 . 2 MACH NUMBER Fig. 7. L i l t - c w e slope US. Moch number, including u t i m o t u d upper ond lower bounds.

@= GreeniWerman, IaRC 0.3m Tcr; trip

-

O= Harris, Re=3x1Om, no trip

.016

e

H = Harris, Re=3xlO0, trip

e= Harrim. R S 6 x l O ' . trip A

-

.014

I= Harris, Re=QxlOo, trip

A = Coethert. DVL (Croup 21, no trip V = L 0 . m . GD HSWT (Croup 3), no trip

-

.012

-

.010

a

.006 c

-

.004

-

.002 I I I I 1 I 1 0 1

.35 .45 .55 .a .75 .85

MACH NUMBER Fig. 8. Minimum drag vs. Moch rurmber: 2 L IO6 c Re < 4 I I O ' .

lower bounds of the correct transonic drag characteristics for t h i s a i r f o i l . III brief. the most important m l n i n u m drag may be surnarized as follars: points concerning 1. The subsonic phavlor withou a boundary layer t r i p I s given by Eqn. 2 to within a b w t tO.0003 I n

!

the range 1 0 < Re < 3 x 10 .

bmGmAL PAGE 1s

1-9

OF POOR QUALITY

T k subsonic behavior w i t h a fully-developed turbulent boundary layer over the e n t i r e a i r f o i l i s 2.

The uncertainty i s d i f f i c u l t t o estimate f r a the available data.

given a p p r o x i m t e l y by Eqn. 3.

but the value 60.0005 i s proposed.

3. The drag-divergence Mach nulnbcr I S between 0.76 and 0.78. Above Mdd. cd r i s e s r a p i d l y t o 6 maxi- value o f 0.11 610%. which occurs between M = 0.92 and 0.98.

<2)j2* 'do 1s given by Eqn. 6 t o w i t h i n 610%. I n t h i s I n the low supersonic range 1.05 < 4.

regime. both C and C vary as M- .

do La @ QeendcNewman. LARC 0.3m TCI: trip 0 .014 ~ u r k ~ o ' , n o t r i p

0 Harri.. Re-3-BXlO@, trip I

.012 0 kethert, DVL (Croup 2). no trip 0 vidnl, CAlSpAN (Group 3 1 , trip 0 Sawyer. ARA (Group 3 ) . no trip .010 sawyer. (croup 3 ) . trip V Fwada, NAL (Group 3). no trip A Lore, GD HSWT (Group 3 ) , no Wp

.....-. . . +0.0005 %

-

00" .m

(I .004 .002 ........ .................................................................. .......... ....................

........ Q O O B a - *dA

f ..................................... ......... $.A...e....& ... -" ..........................

1 1 -.w2 . 3 5 .45 .55 .65 .75 .85 MACH NUMBER Fig. 9. Incremmtal drag vs. Mach numbor: Croups 1-3.

e Group 1, trip

.16 0 Group2notrip

w Group 2. trip

.14 0 Group 3 . no trip 0 Group3,trip A Group 4, no trip .12 )c Group A, trip

------Cd=O.Ol + 0.105

.10

" 0 " .08

.06 .04 .02 I

.4 .5 .6 .7 .8 .9 1 .o 1.1

MACH NUMBER Fig. 10. Minimum drag VJ. Mach number: all data, including estimated upper and lower bounds.

1-10 C. Maxinun L/D Ratio This quantity has i l p o r t a n t p r a c t i c a l consequences f o r both fixed-wing a i r c r a f t and r o t o r c r a f t , and it also represents a r a t h e r d l f f e r e n t and sensitive check on wind-tunnel accuracy and f l o w quality. On the one hand. i t compounds the uncertainty i n both l i f t and drag, but does SO under t e s t conditions t h a t are less severe than C , f o r exanple. O n the other hand, errors i n angle of attack o r uncertainties h X i n the a-corrections are not a t issue here. Therefore, sane experiments i n which C i s suspect my 1, s t i l l provide useful information on ( L / O ) , .

Reynolds-nunber e f f e c t s on (L/D)- Can be isolated f o r examination i f the Mach nuaber I s less than a b w t 0.5. This i s i l l u t r a t e d in Fig. 11. which shows an increase i n (L/D)Mx by about a f a c t o r o f two b e t w e n Re = lo6 and 1 0 5 . I n Fig. 11. the Group 1 r e s u l t s generally show the highest values o f (L/O)mx.

consistent w i t h the o v e r a l l high q u a l i t y o f these investigations. Several o f the' Grwp 2 experiments extend the Reynolds nuaber m g e t o lower values than those o f Group 1 . In additlon. the Gmup 3 r e s u l t s and three sets o f data frm Group 4 are i n f a i r agreement. Unfortunately. Harris 151 d i d not provide l i f t and drag polars f o r untripped conditions, but It i s i n t e r e s t i n g t o M)te t h a t h i s r e s u l t s y i t J a boundary- layer t r i p are i n f a i r agreement with the other data shown. This was not the case f o r any other tripped data.

A t higher Mach nuabers the variations i n (L/Cl)max w i t h Mach and Reynolds number are a l m s t ingossl- b l e t o separate from one another. As a c o l p n l s e between the l l m i t a t i o n s of so feu data available a t a given Reynolds number an the large changes i n (L/O)llax w i t h Re. Fig. 12 shows the available r e s u l t s fur the narrow range 4 . 10' < Re < 9 = IO6. The data from Groups 3 and 4 are o f i n t e r e s t here, because they . are the only available r e s u l t s without a t r i p that extend i n t o the transonic regime. However, they are suspicious because they l i e s i g n i f i c a n t l y below the tripped data o f Harris 151. Additional transonic data would be p a r t i c u l a r l y valuable t o c l a r i f y the quantitative behavior o f ( L / D ) .

D . Maximum L i f t Conventional wisdon holds t h a t three-dimensional separated boundary-layer e f f e c t s are almost lmpossi- b l e t o control a t the s t a l l conditions, and there i s s o w questton as t o whether t r u e two-dimensional s t a l l exists. even f o r extremely high aspect r a t i o s . Parenthetically. the accurate p r e d i c t i o n o f C Lmax f o r the M A 0 0012 a i r f o i l also remains one o f the greatest challenges t o CFD. Therefore, t h i s quantity needs t o be establ ished e x p e r i r n t a l ly.

@='&up 1 data, no trip

e= - , L ~ R C e * m Y < 0.5, trip

B=crOup 2 data, no trip 0- Sawyer, 8"x18"; no trip

05: UTRC 8'; no trip

A- LaRC 6x28; TM X-73890; no trip

12S l o a n .

I

z

I"

I I 0 I 3 x lo5 1 06 107 Log, Re Fig. I I . Maximum lift-to-drq rotio us. Rqwoldr number: M < 0.5.

1-11 0- IaRC no trip

. = Iiarri8, IaRC 8Tp1: Re=!Mo', trip

0- Sawyer, 8"xlB"; no trip

V- IaRC 6.28; no trip, Ty X-73990

A= LaRC 6r28; no. trip, TP-1701 0- Ohio shta 6x22"; no trip

O O O

0 0 0 10a .

-1 I

z n

A

R

I"

A 0

b o

43.

m

0 .2 .4 .6 .8 MACH NUMBER Fig. 12. Marimum lift-to-drog ratio vs. Moch r u m k r : 4 IO6 < Re < 0 s I O ' .

figure 13 shows thc v a r i a t i o n o f C vs Re f o r the available data from Groups 1 and 2. at Mach

hx

numbers less than 0.25. A monotonic increase i n m a x i u l i f t w i t h Reynolds number 1s evident. These p a r t i c u l a r r e s u l t s a r e surprisingly consistent. whereas the values frol Groups 3 and 4 (not shown) were found t be s i g n i f i c a n t l y lower, i n general. Also. it should be mentioned t h a t the data shown a t R e < lo8 are somewhat higher than the values o f t e n quoted (e.9.. Ref. 3 ) . based on older sources.

.25 t

01 I 1 I 2 x 105 1 06 107 Log. Ra Fig. 13.

Maximum l i f t u s . Reynolds number: Croups 1-2. no trip: M < 0.25.

1-12 The e f f e c t o f Mach number on C i s shown i n Fig. 14. f o r R e > 2 I lo6. The scatter below %ax M - 0.25 seems t o be p a r t l y due t o Reynolds n u b e r and p a r t l y due t o wind-tunnel wall effects. However, local transonic e f f e c t s i n the leading-edge region evidently play an increasingly dominant r o l e in the s t a l l process a t M = 0.25 and above, where the maxinun l i f t s t a r t s t o monotonically decrease w i t h increasing M. It i s i n t e r e s t i n g t o note t h a t most o f the Group 4 data are only s l i g h t l y below the data fron Groups 1-3 a t M > 0.4. and the scatter i n t h i s regime i s surprisingly w a l l .

0- Group 1 Data, no trip 2.00 P- xiarrb (croup 2). trip 0- other Qoup 2 D a t a no trip 01 Group 3 Data, no trip 1.75 e=croup3Data,trip

+- Group 4, with & w/o trip

1.50

z

1.25 I I I 1 I I I 0 .2 .4 .6 .8 MACH NUMBER Fig. 14. hlarimum lift vs. Mach number; all data. 2 x IO6 < Re < I O ' .

E. Shock-Wave P o s i t i o n As noted i n the Introduction, there i s so l i t t l e overlap i n the specific transonic t e s t conditions o f However, the myriad experiments, t h a t nost colparisons are necessarily l i m i t e d t o force and moment data.

some i n t e r e s t i n g comparisons can be made o f the measured shock-wave positions, as t h i s quantity appears t o and t o errors i n Mach number.

be p a r t i c u l a r l y sensitive t o wall-interference e f f e c t s Data from 17 experiments a t M = 0.80 and a = 0 are p l o t t e d i n Fig. 15, where X, i s defined as I n t h i s figure, the open diamond the approximate midpoint o f the pressure r i s e across the shock wave.

symbols represent data obtained a t s u f f i c i e n t l y - l a r g e aspect r a t i o s t h a t side-wall boundary layer e f f e c t s should be minimal, and the s o l i d diamond i s a data p o i n t corrected by W. 6. Sewall i n a p r i v a t e conwnica- t i o n using h i s theoretical analysis o f side-wall effects 1211. (The p r i n c i p a l e f f e c t i s t o increase the e f f e c t i v e Mach number by about 0.01). The squares denote expriments i n which the side-wall boundary layer was e i t h e r removed or i t s e f f e c t corrected f o r . The c i r c l e s represent the refmining sources, f o r which no p a r t i c u l a r a t t e n t i o n appeared t o be given t o side-whll effects.

The grouping o f the data i n Fig. 15 i s inspired by recent numerical analyses 122.231. which s h o e d the tendency o f three-dimnsional viscous e f f e c t s on a i r f o i l s i n wind tunnels t o move the shock wave f o r - This explanation i s tempting f o r some o f the data w i t h unreasonably ward o f i t s tWO-di~nSiOMl position.

X,. but data from several other Swrces without side-wall treatment appear " n o m l . " small values o f Neither does there seem t o be any systematic e f f e c t o f other factors, such as boundary-layer t r i p s or the amount o f tunnel s l o t or perforation openness. Althwgh the majority o f the r e s u l t s seem t o l i e between X s = 0.44 and 0.48, the o v e r a l l scatter i s disturbing. and the actual reason f o r i t remains a mystery.

Therefore, t h i s i s yet another area where the key experimental information that would be valuable f o r C F D code v a l i d a t i o n i s not satisfactory.

ORIGINAL PA-GE P S

OF POOR QUALITY 1-13

.31 P i g . 15. Shock-wave position vs. Reynolds number at It = 0.80 and o = 0 : all data.

I V . SUMMARY AND CONCLUSIONS Results from more than 40 two-dimensional wind-tunnel experiments have been c r i t i c a l l y examlned and analyzed.* Sadly, the scatter I n the t o t a l ensemble o f data i s unacceptable i n the author's view, and it i s not r e a d i l y apparent which o f these r e s u l t s a r e correct. It i s clear, however, t h a t the requirements f o r flow q u a l i t y and data accuracy set f o r t h I n AGAR0 Advisory Report 184 141 are seldom met i n a i r f o i l testing.

The r e s u l t s o f t h i s investigation also suggest t h a t no Srnple e x i s t i n g experiment i s adcquate e i t h e r NACA 0012 a i r f o i l , o r f o r v a l i d a t i n g C F D f o r defining the complete aerodynamic characteristics o f the codes.

Nevertheless, the aggregate o f available data i s extramely useful. A systematic screening process has been used t o help define the r e l a t i v e merits o f the various experlmnts and t o f i l t e r considerable useful, quantitative information from the confusion. Correlatlons o f key parameters w i t h llach and Reynolds number have also narrowed the uncertainty i n the a i r f o i l section characteristics t o acceptable levels, and the judicious use o f a i r f o i l theory and numerical calculations permits extrapolatlons t o be made i n t o regimes where hard evidence i s sparse. This combirkd i n f o r n a t i o n serves three inportant func- i t allows individual experiments t o be c r i t i q u e d with more confidence than heretofore: tions. F i r s t .

second, it allows the complete NACA 0012 a i r f o i l characteristics t o be estimated m r e precisely. Third, i n the figures and equations can be used t o establish the c r e d i b i l i t y of the synthesized r e s u l t s presented indlvidual a i r f o i l f a c i l i t i e s .

On the basis of both corpleteness and accuracy. the expcrlment of Harris 151. chosen by Holst 1161 i n h i s recent v a l i d a t i o n exercise f o r viscous transonic a i r f o i l analyses. emerges as the' most satlsfactory *Tabulations of the data presented i n t h i s paper are available from the author upon w r i t t e n request.

1-14 single investigation of the conventional NACA airfoils to date. Harris' range of flow conditions is not nearly as complete as desired, and the accuracy of the data was not evident a priori, as lift-interference corrections on the order of 15% were proposed for the angles of attack. However, the present study indi- in fact, adequate, at least for low angles of attack, cates that Harris' estimates of this phenomenon are, and that most other major sources of errors were minimized. On the other hand. the author is persuaded by 1211 that some side-wall boundary-layer interference existed. Therefore, the arguments of Mr. W . G. Sewall it is strongly recommended that this be corrected for before using Harris' data for CFD code validation.

As discussed in Section 111. the values of lift-curve slope and u i n i w drag 1 subsonic flow can now

?

be established with high confidence in the Reynolds nulber range lo6 < Re < 3 I 10 . The behavior of these key quantities can also be estimated throughout the transonic regime and up to low supersonic Mach The issue of self-induced oscillations nulnbers. but with rapidly-deteriorating confidence above M m 0 . 8 .

and the possibility of negative values of C in the range 0.85 < M < 0 . 9 0 need further ' . a and above M = 1 would be useful for CFD code investigation. A better definition of the behavior at validation.

The variations of C I M X with M and Re can now be specified with a moderate degree of confidence, and the data from most of the available sources are surprisingly consistent above M = 0 . 4 . This conclu- sion appears to contradict folklore, conventional wisda. and recent nunerical studies of wall interference. I On the other hand, the behavtor of the maxirum lift-to-drag ratio and shock-wave position i s not nearly as well defined, and both these quantities appear to be particularly sensitive to wind-tunnel wall effects and turbulence. Therefore, additional studies under carefully-controlled conditions are strong.ly It is also suggested that both of these quantities w w l d be especially inportant criteria reconmended.

f o r CFD code validation. if they could be reliably established by well-documented experiments.

Finally. the results of this investigation indicate that measurements. corrections, and/or treatments For all Four walls of the test section are essential for any reasonably-sized model under transonic flow Although results from some facilities appeared to suffer more than others from wall- conditions.

interference effects. 2 facility that failed to address the potential problems on all four walls provided data that could be judged entirely Satisfactory.

V. ACKNOWLEDGEMENTS The author is extremely grateful to the many people who generously shared stiarlating ideas and insights. background information, reference sources, and unpublished results during the course of this investigation. The manifold contributions of Mssrs. Charles Ladson and Willlam Sewall of NASA-Langley, including extensive unpublished data. were truly Invaluable. Grateful acknowledgement is also extended to Dr. Terry Holst of NASA-AWS, Mr. Frank Harris of Bell Helicopter Textron. and M r . Ray Prouty of McDonnell-Douglas Helicopters. for their helpful comments. suggestions, and unpublished information. Mr.

Lars Ohman of the National Aeronautical Establishment and Mssrs. 8 . F . L . Hanond and T . E . B . B a t m a n o f the Aircraft Research Association, Ltd. provided Mach-nunber corrections and other useful information concern-' ing their respective facilities. Also, Mssrs. Lawrence Green, Clyde Gumbert. and Perry N e m a n of NASA- Langley. Herr D . Althaus of the Univcrsitiit Stuttgart. Prof. Siegfried Wagner of Universitat der Bundeswehr Munchen. Mr. Kazuaki Takashima of the National Aerospace Laboratory, and M S . Mary Berchak of Ohio State University kindly provided explanations and tabulations of unpublished data, and their generous assistance is deeply appreciated.

VI. REFERENCES 1. McCroskey, W. J. "Technical Evaluation Report on 'AGARD FDP Symposiua on Applications of Computa- tional Fluid Dynamics in Aeronautics," AGARD Advisory Report No. 240. 1986.

2. McCroskey. W. J. " R w n d Table Discussion on 'Wall Interference in Wind Tunnels,'" AGARD Conference Proceedings 335. May 1982. ' 3 . Abbott, I . H., and von Doenhoff. A . E . Theory of Winq Sections. including a Summary of Airfoil Data, Dover Publications, New York. 1959. pp. 124-187.

4. Steinle. F . , anJ Stancwsky. E . "Wind Tunnel Flow Quality and Data Accuracy Requirements," AGARD Advisory Report 1 8 4 . 1982.

5 . Harris, C . 0 . "Two-Dimensional Aerodynamic Characteristics of the NACA 0012 Airfoil in the Langley 8-Foot Transonic Pressure Tunnel.' NASA TM 81927, April 1981.

6 . Vidal. R . J . . Catlin. P. A . . and Chudyk. 0. W. "Two-Dimensional Subsonic Experilnents with an NACA 0012 Airfoil." Calspan Corporation Report No. RK-5070-A-3. 1973: also, Paper No. 11. AGARD Conference Proceedings CP-174. Oct. 1975.

1-15 Green. L . L . , and Newmn. P. A . "Transonic Wall Interference Assessment and Corrections for Airfoil 7.

Data from the 0 . h TCT Adaptive Wall Test Section." AIAA Paper 87-1431, 1987.

Sawyer, Mrs. J. "Results of Tests on Aerofoll M102/9 (NACA 0012) in the A . R . A . Two-Dillcnsional 8.

Tunnel ,* Aircraft Research Associates Model Test Note MlO2/9, 1979.

Scheitle, H. "Messrelhen zur Bestilung stationarer Profilbeiwerte der Profile NACA 0012, H1-Tb und 9.

H3-TbSY Inst. fur Luftfahrttechnik und Ldchtbau, Universitat der Bundeswehr Munchen Institutsbericht Nr. 87/2. 1987; also private c a r r n i c r t i m s from 5. Wagner. 1987.

Ladson, C . L . "Two-Dimensional Airfoil Characteristics of Four NACA 6A-Series Airfoils at Transonic 1 0 .

Mach Numbers up to 1.25.' NACA RM L57F05, 1957.

11. McDevitt. J . B. ' A Correlation by Means of the Transonic Similarity Rules of the Experimentally Determined Characteristics of a Series of Sylwtrical and Calbercd Wings of Rectangular Planform." NACA TR 1253. 1955.

"Aerodynamics." Rotor & Wing International. Aug. 1984. pp. 17-22; also private c a u n i c a - 12. Prouty. R.

tions 1982. 1984, and 1987.

Feldnan. f . K . "Untersuchung von symetrischen Tragflugelprofilcn bei hohen Untetschallgeschwindig- 13.

keiten in einen geschlossencn Windkanal." Mittellungen aus d e n Institut fur Aerodynamik. No. 1 4 . A . 6 .

Gebr. Leeman I Co., Zurich, 1948.

14. McCroskey. U. J . . Baeder. J. D.. and Bridgerrn. J. 0 . "Calculation of Helicopter Airfoil Characteris- tics for High Tip-Speed Applications.. J. k r i c a n Helicopter Soc., Vol. 31. No. 2. pp 3-9. April 1986.

15. Levy, L. L.. Jr. "Expertmntal and Cocputatlonal Steady and Unsteady Transonic Flows about I Thick Airfoil," A I M Journal. Vol. 16, No. 6 . pp. 564-572. June 1978.

Holst. 1 . L . "Viscous Transonic Airfoil Workshop - Conpendim of Results." AIAA Paper 87-1460. 1987.

1 6 .

17. Crane. H. L . and M a s . J. J. "Wing-Flow Investigation of the Characteristics of Seven Unswept.

Untapered Airfoils of Aspect Ratio 8 . 0 . ' NACA RM L51024a, 1951.

18. Daley. 8 . N. and Oick. R. 5. "Effect of Thickness. Cuber, and Thickness Dirtribution on Airfoil Characteristics at Mach Numbers up t o 1 . 0 . " NACA TN 3607. 1 9 % .

Hoerner. 5 . F. Fluid-Dynamic Draq. publishkd by the author, Midland Park, N . J . , 1965. pp. 17-7 19.

to 17-12.

20. Hoerner. 5. F . and brst. H. V . Fluid Dynamic Lift. publishcd by Mrs. L . A . Hoerner. Brick Town.

N.J.. 1975. pp. 2-12 tO 2-14.

21. Sewall. W . G . 'Effects of Sidewall Boundary Layers in Two-Dimensional Subsonic and Transonic Wind Tunnels." AlAA Journal. Vol 20. N o . . 9 . pp. 1253-1256, fept. 1982: also private communications 1985. 1986.

and 1987.

22. Obayashi. 5. and Kuwahara. K . 'Navicr-Stokes Silulation o f Side-Wall E f f e c t of Two-Dimensional Tran- sonic W i n d Tunnel," AIAA Paper 87-037. 1987.

23. Obayashi. 5. and Kuwahara. K . mSide-Uall Effect for a Wing at High Angle of Attack," A I M Paper 87-1211. 1987.

1-16

Table 1. NACA 0012 - Sumnary o f Experiments -- Group 1

SOURCE MACH Re (10') TRIP ? TUNNEL CHAR. REMARKS range range X t 1. Abbott e t al.; 0.07-0.15 0.7-26 yes It no s o l i d walls l i n e a r wall corrections; "Std. R" AR = 0.75-6 very low turbulence; Langley LTPT h/c= 1.9-15 excessively t h i c k t r i p ; possible minor side-wall boundary-layer e f f e c t s

data available: C , , &. Cd. (L/D)max, CLmX

2. Ladson; 0.07-0.36 0.7-19 yes L no s o l i d walls l i n e a r wall corrections; Langley LTPT AR = 1.5 very low turb. a t low M; Xt.0.05 h/c = 3.8 possible minor side-wall boundary-layer e f f e c t s data available: C , , C , , , . cd, (L/O)max, Ctmnw 3 . Gregory and 0.08-0.16 1.4-3 yes L no s o l i d walls l i n e a r wall corrections; AR = 3.6 w i t h & w/o side-wall 0 ' Rei 1 l y : NPL 13'x9' varying h/c = 5.2 boundary-layer control data available: C , . C , . cdn 1.e. Cp. (L/O)na,t Cllnax 4. Green & Newan; 0.5 - 0.8 9 yes adaptive walls four-wall corrections: O . 3 m TCT AR = 2 moderate turb. level Langley X t = 0.05 h/c = L d d t a available: C, , C , (low R only) References f o r Table 1: la. I . H. Abbott and A. E. von Doenhoff: Theory o f Wing Sections, 1959.

l b . A. E. von Doenhoff and F . T. Abbott. Jr.: NACA TN 1283. 1947.

IC. C. C. Critzos. H. H. tleyson. and A . W. Eoswinkle. Jr.: NACA TN 3361, 1955.

2. C. L. Ladson: NASA-Langley, p r i v a t e connunicetion.

3. N. Gregory and C. L. O'Reilly: NPL Aero Report 1308 (ARC 31 719). 1970.

4. L. L . Green and P. A. Newnan: A I M Paper 87-1431. 1987. and p r i v a t e c o m n i c a t i o n s .

1-17

Table 2 - S u m r y o f Experiments -- Group 2

SOURCE MACH Re (lo6) TRIP ? TUNNEL C H A R . REMARKS range range X t 5. Harris: 0.3 - 0.86 3 - 9 yes L no s l o t t e d w a l l s large a corrections: Langley 8 ' TPT AR = 3.4 possible side-wall boundary Xt10.05 h / C = 3.4 e f f e c t s on X s & C d

data available: cn. c , . Cd, cp. (L/D)mx. x,. l i l i t c d c

%ax 6. Goethert: 0.3 - 0.85 2 - 6 no s o l i d w a l l s w a l l and end-plate corrections: OVL 2.7m W.T. AR = 2.6 turbulence l e v e l ~ 1 % ; h/c = 5.4 some flow a s m e t r y data available: C , . C , . cd. Cp 7 . Sheldahl (1 Klimas 0.1-0.2 0.35-1.8 no s o l i d w a l l s l i n e a r wall corrections: Wichita St. 7 ' x l O ' AR = 2.4-6 same flow asymnetry: h/c= 5.6-15 0 < a < 180 data available: cp. Cd. (L/o)mx, %lax 8. McCroskey, e t a1 0.1-0.3 1 - 4 yes 11 no s o l i d w a l l s l i n e a r wall corrections; Ames 7 ' x l O ' No.2 AR = 3.5 continuous. dynamic data X t = 0.01 h/c = 5 data available: C , . C , . l i m i t e d cd* Cp. (L/D)Mx 9. Eevert: Poisson 0.06-0.11 1.1-2.2 no s o l i d w a l l s l i n e a r wall corrections; Ouinton & de Sievers: AR = 1.3 Tu < 0.2% S1.Ca kn h l c = 4 ddtd available:

c , , c , . Cdr c

p. ( L I D ) 10. Wortmann & 0.07-0.17 0.3-2.5 no s o l i d w a l l s Side-wdll suction: AlthduS: Techn. AR = 1.5-3 very low turbulence Ilochs. S t u t t a a r t h/c= 5.5-11 e a r l y C suspect La Lam. H.T.

dJ+a available:

c,, Cdr (L/D)Mx. Clmax

References f o r Table 2: 5. C. 0. Harris: NASA TM 81927, A p r i l 1981.

6. E. H. Goethert: NACA IM-1240. 1949: Nat. Res. Council (Canada) 11-27. TT-31. 11-38. 1947; RAE TN Aero 1684. 1945.

7. R. E. Sheldahl and P. C. Klimas: Sandia N i t . Labs Report SAND80-2114. 1981.

E. W. J. McCroskey. K . W. McAlister. L. W. Carr. and 5. L. Pucci: NASA TM 84245. 1982.

9a. A. Bevert: ONERA Doc. 76/1157.AN. 1972.

9b. Ph. Poisson-Quinton and A. de Sievers: AGAR0 CP-22. Paper No. 4. 1967.

loa. F . X. Wortnann: AGAR0 CP-102. 1972.

l o b . 0. Althaus: I n s t i t u t fur Aerodyn. und Gasdynamik. Stuttgart. p r i v a t e ccinmunication, 1987.

1-18

Table 3 - Summary o f Experiments -- Group 3

SOURCE MACH Re (lo6) TRIP ? TUNNEL CHAR. REMARKS range range X t 11. Bernard-belle; 0 . 3 2 5 3.5 no(?) s o l i d walls side-wall suction. care- ONERA R1.Ch AR = 0.67 f u l study o f side-wall h/c = 3.3 e f f e c t s data available: l i m i t e d C , . C , . cd

12. Sawyer; 0.3 - 0.85 3 - 6 yes & no s l o t t e d walls a. M. and curvature

ARA 8"x18". AR = 1.6 corrections; poss.

side-wall boundary Trans. W.T. Xt.0.07 h/c = 3.6 layer e f f e c t s 13. Vidal e t al. 0.4 - 0.95 1 yes porous walls thick t r a n s i t i o n strips: CALSPAN 8' AR = 8 s l i g h t f l o w angularity: xt=O.l h/c = 16 m i n i m interference

d d t d available: c , , C , . Cdr cp. (L/o)mx, l i l l l l t e d clmx, x s

14. McOevitt & 0.72 - 0.8 2 - 12 no s o l i d walls contoured walls, wall

Okuno: AR = 2 pressure meas. ; Ames Hi-Re Channel h/c = 3 side-wall suction: unsteady measurements data available: CIu, Cp. X s (low G only) 15. Gunbert & 0.7 - 0.8 3 - 9 yes & no s l o t t e d #ails o corrected: Neman: AR = 1.3 side-wall boundary-layer

Langley O.3m T L T Xt.0.05 h/c - 4 corrections

data available: ccU. cdo (low Only) 16. TdkaShima. 0.6 - 0.8 4 - 39 no s l o t t e d walls wall pressure-rail meas.: Sawada e t al. AR = 1.2 - 2 poss. side-wall b.1.

N A L Transonic U.T. h/c = 4 - 6.7 effect on shock position: data available: C , , cd. Cp. X s (low a only)

17. Sewall: 0.3 - 0.83 4 - 9 yes & no s l o t t e d walls a and side-wall

Langley 6" x 28" A R . 1 - 2 b. 1. corrections (revised) Xtm0.08 h/C= 4.7-9.3 data available: C , . C , . Cd. C * xs Emax 18. Lowe 0.63-0.82 15-38 no perfor. walls 22% perforation, side-wall AR = 1 suct ion; General uyn. Hi-Re 4 uncertain u corr.

20 Test Sect, HSWT h/c= data available: C , . Cd. Cp. X s ~~~~ 19. Jepson: 0.3 - 0.9 2 - 6 no s o l l d walls l i n e a r w a l l corrections: Lizak; Carta: AR- 1.7-5.8 m u l t i p l e entries: various UTRC 8 ' h/C-4.7-5.8 models and end plates data available: C , . C , . cd. Cp. (L/O)mx, CLmaX. Xs

20. Uang e t al. 0.7 - 0.9 -3(?) yes perfor. walls porosity adjusted f o r

Chinese Aero. Inst. AR= 3.2-6.4 min. interference Transonic W.T. X t=O .06 hR-2.6-5 .2 data available: l i m i t e d C , . c , . X , ~~ ~ References f o r Table 3: 12th Applied Aero. Colloq.. E N S W C E A T (NASA TT-F-17255). 1975; also 11. R. Bernard-Guelle: J. P. Chevallier: ONERA TP 1981-117. 1981.

A i r c r a f t Research Associates Model Test Note M102/9. 1979.

12. Mrs. J. Sawyer: Calspan Corporation Report No. RK-5070-A-3. '1973.

13. A . J. Vidal. P. A. Catlin. and 0 . W. Chadyk: 14. J. 8 . McDevitt and A.F. Okuno: NASA TP 2485. 1985.

A I A A paper No. 84-215!. 1984.

15. C. R. Gumbert and P.A. Neman: 16a. H:Sawada. 5 . Sakakibara. M. Satou. and H. Kanda: NAL TR-829. 1984.

1-15 Table 3 - Concluded.

I C A S Paper 82-5.4.4. 1982.

16b. K. Takashima: National Aerospace Lab. also p r i v a t e c o u n i c a t i o n s . 1985 and 1987.

16c. K . Takashima: 17. W. G. Sewall: NASA TM 81947. 1981. also p r i v a t e c o l u n i c a t i o n s 1985. 1986. and 1987.

18. W. H. Lone: General Dynamics Report HST-TR-74-1, 1974.

Sikorsky Report SER-50977. 1977.

19a. W. 0. Jepson: 19b. A. 0. S t . H i l a i r e . e t al: NASA CR-3092, NASA CR-145350. 1979.

19c. W. H. Tanner: NASA CR-114. 1964.

19d. A. A. Lizak: Amy Trans. Res. Con. Report 60-53. 1960.

presentation t o Sino-U.S. Joint Symposium on 20. 5. Uang. V. Chen, X. Cui. and 8. Lu: "Fundamental Experimental Aerodynamics," NASA-Langley, 1987.

Table 4 - Sunary o f Experiments -- Group 4

SOURCE MACH Re (lo6) TRIP ? TUNNEL CHAR. REMARKS range range X t

21. Sewall: 0 . 5 8 - 0.92 3 - 4 yes s l o t t e b walls data corrected f o r t h i c k

LaRC 6"x19" AR = I side-wall boundary Xt=O.Oa h/C = 3.2 interference but not data available: Cna. Cdo. Xs l i f t interference & 0.35 - 1.0 1-10 yes & no s l o t t e d walls (I corrected: 22. Noonan Bingham: Ladson: AR = 1.0 side-wall b.1. effects on LaRC 6"x28" X t = 0.1 h/c 4.7 Shock p o s i t l o r dnd C h X data available:

Cn. C , . cd, Cp. (L/D)llaxr CLnax. Xs

23. Ohman. e t al; 0.5 - 0.93 17-43 no porous walls 20% porosl ty; NAE 5 ' x 5 ' AR = 1.3 side-wall suction: w i t h 20 i n s e r t h/c = 5 data s l i g h t l y asymmetric; Mach No. corrected herein data available: Cdo, Cp. X s a t a = 0

24. Thibert. e t al: 0.3 - 0.83 1.9 - 4 no porous Walls large wall corrections. but

ONERA S3.Ma AR = 2.7 wall press. measured: h/c= 3.7 t h i c k side-wall b.1.

data available: C , . cd. Cp. X s 25. Scheitle & 0.36 - 1.6 3 - 10 IK) s l o t t e d walls suction on a l l four walls.

Wagner: TWT niinchen AR = 1.5 variable w i t h M t o U n i v . Bundesuehr h/c = 3.4 match other f a c i l i t i e s : mderate turb. level data available: Cto, C d m i n , (L/D)mx.

'lmax

26. Jepson: 0.3 - 1-08 2 - 5 no s l o t t e d walls large l i f t interference

NSROC 7 ' x l O ' AR = 7.5 h/c = 5.3 data available: C , . C , . Cd. (L/D)Mx. C %ax

27. Lee, e t al: 0.2 - 1.06 2 - 12 no porous walls independent plenums f o r

Ohio State 6"x22" AR 0.5 - 2 top and bottom walls

Trans. A i r f . Facil. h/C= 0.9-7.1 data available: C , . C , , cd. (L/DImX, C i m u . X,. l i m l t e d Cp 28. Prouty; 0.34-0.96 3 - 7 no s l o t t e d walls large l i f t interference: LAC 15"x48" AR = 1.5 poss. side-wall boundary h/c = 4.6 layer effects; some flow asynmnetry data available: C , . C , . td. (L/D)max, %ax 29. Gregory & 0.3-0.85 1.7-3.8 yes s l o t t e d walls probable w a l l effects W i lby: AR = 1.4 on a l l data

NPL 36*x14" Xt-0.02 h/c - 3.6 f a i r l y large roughness

data available: C , , C , . C I C , (L/O)max. Char, xs

d P 1-29 Table 4 - Concluded.

30. K r a f t C 0.8 - 0.9 2.2 no adaptive walls variable porosity and

Parker; AR = 2 hole angle; AEDC 1-T h/c = 2 no side-wall treatment data available: C , . X , ~~ 31. Triebstein; 0.5 - 1.0 1 - 3 no porous walls no corrections applied; DFVLR l m TU1 AR = 5 unsteady measurements h/c = 5 data available: XS. Cp

32. Ladson; 0.5 - 1.1 1.5 - 3 no s l o t t e d walls a corrected f o r l i f t

LaRC 6”x19” AR = 1.5 interference but not h/c = 4.8 side-wall boundary layer Cn. C , . C , . surface o i l flow. schlieren data available:

33. Ladson; 0.8 - 1.25 2.7 no s l o t t e d walls no corrections applied

LaRC ATA 4”x19” AR = 1.0 h/c = 4.8 data available: Cn References f o r Table 4: 21. W. G. Sewall: AIM Journal. Vol 20. No. 9. pp 1253-1256. 1982; also p r i v a t e comunications 1985. 1986. and 1987.

22a. K. W. Noonan and G. J. Bingham: NASA TM X-73990, 1977.

22b. K. W. Noonan and G. J. Bingham: NASA TP-1701. 1980.

23. J. Thibert. M. Grandjacques. and L. Okan: AGARD AR-138. Ref. A l . 1979: a l s o p r i v a t e communication from L. Oman. 1987.

24. J. Thibert. M. Grandjacques. and L . Ohman: AGARD AR-138. Ref. A I . 1979.

25a. H . Scheitle: Inst. f u r Luftfahrttechnik und Ldchtbsu. Universitat der Bundeswehr Munchen I n s t i t u t s b e r i c h t N r . 87/2. 1987.

2%. 5. Wagner: Universitat der Bundeswehr Munchen. p r i v a t e comunications. 1987.

26. W . 0. Jepson: Sikorsky Report SER-50977, 1977.

A I A A Paper No. 78-1118, 1978.

27a. J. 0. Lee. G . M. Gregorek. and K. 0. Korkan: Ohio State University. p r i v a t e comunications, 1987.

27b. M. J . Berchak and G. M. Gregorek: 28. R. Prouty: “AerodyMmics.” Rotor C Uing International, Aug. 1984, pp. 17-22; also p r i v a t e communications 1982. 1984. and 1987.

29. N. Gregory and P. G. Uilby: ARC CP-1261 (NPL Aero Report 017). 1973.

30. E . M. K r a f t and R. L. Parker. J r . : AEOC Reports TR-79-51. 1979. TR-60-83, 1981.

31. H. Triebstein: J. A i r c r a f t , Vol. 23. No. 3. pp. 213-219. 1986.

35. C. L . Ladson: NASA TO 0-7182. 1973.

33. C. L. Ladson: NACA RM L57F05, 1957.

1-21

Table 5 - Experiments examined but not used -- Group 5

34. J. Stack.and A. E. von Doenhoff: NACA Report 492. 1934 (NASA-Langley 11" HST: s o l i d wdlls, severe blockage effects).

35. R. Jones and 0. H. Wllllsns: ARC R&H 1708, 1936 (NPL W n s S e d A i r Tunnel: e f f e c t s o f surface roughness and Re on wings: AR = 6).

3 6 . E. N. Jacobs and A. Sherman: NACA Report 586. 1937. and Report 669. 1939 (NACA-Langley VDT: AR = 6; high turbulence level).

37. H. J. Goett and W. K. h l l i v a n t : NACA Report 647. 1938 (NASA-Langley 30'x60' Full-Scale UT; AR I 6; low turbulence).

38. J. V. Becker: NACA Wartime Report L-682. 1940 (NASA-Langley 8' HSWT: t r a n s l t l o n and s k i n - f r i c t i o n measurements a t high Re).

39. A. E. von Doenhoff: NACA Wartime Report L-507, 1940 (NASA-Langley LTT: boundary-layer and minimum-drag measurements vs Re).

40. F . K. Feldman: Techn. Hochsc. Zurich M i t t e i l u n g m aus d m I n s t i t u t f u r AerodyMdk. No. 14, 1948 (Ackeret's High-speed Wind Tunnel: transonic measurerrnts on wings: AR = 3.3).

: I . 1. K. L o f t i n and H. A. Smith: NACA TN 1945. 1949 (NACA-Langley LTT: low lift values. not synrnetrical f o r p o s i t i v e and negative angles o f attack).

42. J. Stack and W. F. Lindsey: NACA Report 922. 1949 (NASA-Langley 24" HST: s o l i d walls. variable AR).

43. L. K . Loftin: NACA TN-3241, 1954; P.J. Carpenter: NACA TN-4357. 1958: C.L. Ladson: NASA TO 0-7182, 1972 (NASA-Langley LTPT using freon).

44. J. Ponteziere and R. Bernard-Guelle: L'Aero. e t I'Astro. Vol. 32. 1971-8: (ONERA R1.Ch before side- wall studies).

45. A. G. Parker: A I A A Journal, Vol. 12. No. 12. pp. 1771-1773. 1974 (Texas A&M 7 ' x I O ' : large a l r f o i l , comparison o f open and closed t e s t section).

46. N. Pollock and 8. 0. F a i r l l e : ARL Aero Report 148. 1977. and Aero Note 384. 1979 ARL Variable- Pressure UT w i t h s l o t t e d and s o l i d walls: large corrections, but pressures measured on s o l i d Malls).

4 7 . K . w. McAlister. W. J. McCroskey. and L. W . Caw: NASA TP 1100. 1978 (NASA-Ames 7 ' x l O ' #2: large d i r f o i l ; unsteady measurements: w i t h and without end plates).

48. F. W. Spaid. J. A. Dahlin, F. W. Roos, and L. 5. Stlvers: Supplement t o NASA TM 81336. 1983: L.

Stivers. NASA-Alas. p r i v a t e connunications (NASA-Arms 2'x2' TWT: large l i f t interference: incomplete resu I t s avai lable).

presentation t o Sino-U.S. J o i n t Symposiu on "Fundamental Experimental Aerodynamics."

49. Q. Zhang: HSWT: detailed study o f a l t e r n a t i v e intrrrerence corrections).

NASA-Langley, 1987 (Nanjing 0.6xO.h 50. R. J. Hansman and A. P. Craig: A I A A Paper 87-0259. 1987 ( M I 1 l ' x l ' LTWT: comparative study of the e f f e c t s o f t r i p s and r a i n a t low Re).

Report Documentation Page

Nlh.rll-nn. tu/\sA <.Vd

- 3. Recipient’s Catalog No.

1. Report No. 2. Government Accession No.

NASA TM-100019

USAAVSCOM TM 87-A-5

-

5. Report Date

October 1987

____________

A Critical Assessment of Wind Tunnel Results

6. Performing Organization Code

for the NACA 0012 Airfoil

__ .

-

8. Performing Organization Report I 7. AuthorM

W. J. McCroskey

A-8732 1 __ _ _

10. Work Unit No.

992-2 1-0 1

9. Performing Organization Name and Address 11. Contract or Grant No.

Ames Research Center, Moffett Field, CA 94035-5000

and Aeroflightdynamics Directorate, U.S. Army

Aviation Research and Technology Activity, Ames

13. Type of Report and Period Cove~

Research Center, Moffett Field, CA 94035-5000

G.-Sponsoring Agency Name and Address

Technical Memorandm

National Aeronautics and Space Administration

14. Sponsoring Agency Code

Washington, DC 20546-0001 and U.S. Army Aviation

Systems Command, St. Louis, MO 63120-1798

--- __-.- --

15. Supplementary Notes Point of Contact: W. J. McCroskey, Ames Research Center, M/S 258-1,

Moffett Field, CA 94035-5000 (415) 694-6428 or

FTS 464-6428

\

-

16. Abstract

A large body of experimental results, which were obtained in more than 40

wind tunnels on a single, well-known two-dimensional configuration, has been

critically examined and correlated. An assessment of some of the possible

sources of error has been made for each facility, and data which are suspect

have been identified. It was found that no single experiment provided a

complete set of reliable data, although one investigation stands out as supe-

rior in many respects. However, from the aggregate of data the representative

properties of the NACA 0012 airfoil can be identified with reasonable confi-

dence over wide ranges of Mach number, Reynolds number, and angles of

attack. This synthesized information can now be used to assess and validate

existing or future wind tunnel results and to evaluate advanced Computational

Fluid Dynamics codes.

18. Distribution Statement 17. Key Words (Suggested by Authorls))

Wind tunnel testing Unclassified-Unlimited

Airfoil characteristics

Subject Category - 02

I

- 22. Price 20. Security Classif. (of this page) 21. No. of pages 19. Security Classif. (of this report)

Unclassified Unclassified

23 A03

I

IASA FORM 1626 OCT 86