You are on page 1of 72

I

, NASA
TN
D-7428

I
0

.)

N74-11821

LOW-SPEED AERODYNAMIC CHARACTERISTICS


OF A 17-PERCENT-THICK AIRFOIL SECTION
DESIGNED FOR GENERAL AVIATION APPLICATIONS
R o b e r t J.

McGhee,

et a1

Langley Research Center


Hampton, Virginia

LOAN COPY: RETUR TO


AFWL TECHNICAL LIE !ARY
"KIRTLAND AFB. N. n.

D e c e m b e r 1973

DISTRIBUTED BY:

National Technical information Service


U. S. DEPARTMENT OF COMMERCE

w
4. Title uxl Subtitle

. . . .

D-7428

..

--

6. R e p a t Dam

December 1973

LOW-SPEED AERODYNAMIC CHARACTERISTICS OF A


17-PERCENT-THICK AIRFOIL SECTION DESIGNED FOR
GENEFf&- A-WTION APPLICATIONS
. . . . . .

8. F'orkrning Dmniniim cod.

8. Worming W n i u t i o n R.pwt

7. Authodt)

L-9132__ _ __~

Robert J. McGhee and William D. Beasley


- ..

___._

10. Work Unit No.

9. Worming Orpization Name nnd Addrsl

501-06-05-02
_ _
11. C o n t m or Gnnt No.

NASA Langley Research Center


Hampton, Va. 23665
..

~
~~

__

_--_____I.--

12. Sponsoring Agency Name and Addrarr

National Aeronautics and Space Administration


Washington, D.C. 20546
.

1s.

s u w m riotes
~

~ _ _ _ _

_____
~
__
. .

-.-

_..___m____

_-__I

.___

_
J
_
E

16. Abrtnct

Wind-tunnel tests have been conducted to determine the low-speed two-dimensional


aerodynamic characteristics of a 17-percent-thick airfoil designed for general aviation
applications (GA(W)-l). The results were compared with predictions based on a theoretical method for calculating the viscous flow about the airfoil. The tests were conducted
over a Mach number range from 0.10 to 0.28. Reynolds numbers based on airfoil chord
varied from 2.0 x lo6 to 20.0 x lo6. Maximum section lift coefficients greater than 2.0
were obtained and section lift-drag ratio at a lift coefficient of 1.0 (climb condition)
varied from about 65 to 85 as the Reynolds number increased from about 2.0 X lo6 to

PRICES SUBJECT TO CHANGE


-

- _ ._I_-

Low-speed airfoil section


Reynolds number effects
Experimental-theoretical comparison
GA(W)-l airfoil
General aviation aircraft
- - -- __
.- - _ _ _ _
19. SIcurity Urcsif. (of this nport)

Unclassified

Unclassified

- Unlimited

--I

20. Security Clanif. (of thus prqs)

._

Unclassified
I

.
_-

I
_
L

*For ah by tho Notion01 Tochniclll InformationService,Springfiold, VirOinio 22161

-____

18. Distribution S m m n t

17. Key Words (Suggatsd by Author(r1)

. ..

LOW-SPEED AERODYNAMIC CHARACTERISTICS OF A


17-PERCENT - THICK AIRFOIL SECTION DE SIGNED
FOR GENERAL AVIATION APPLICATIONS
By Robert J. McGhee and William D. Beasley
Langley Research Center
SUMMARY
An investigation w a s conducted in the Langley low-turbulence p r e s s u r e tunnel to
determine the low-speed two-dimensional aerodynamic characteristics of a 17-percentthick airfoil designed for general aviation applications. The results a r e compared with
a typical older NACA 65 s e r i e s airfoil section. Also, a comparison between experimental
data and predictions, based on a theoretical method f o r calculating the viscous flow about
the airfoil, is presented. The tests w e r e conducted over a Mach number range f r o m 0.10
Reynolds numbers, based on the
to 0.28 and an angle-of-attack range f r o m -loo to 24'.
airfoil chord, w e r e varied f r o m about 2.0 X lo6 to 20.0 X lo6.

The results of the investigation indicate that maximum section lift coefficients
increased rapidly at Reynolds numbers f r o m about 2.0 X lo6 to 6.0 X lo6 and attained
values g r e a t e r than 2.0 for the plain airfoil and g r e a t e r than 3.0 with a 20-percent-chord
split flap deflected 60. Stall characteristics were generally gradual and of the trailingedge type either with o r without the split flap. At a lift coefficient of 1.0 (climb condition)
the section lift-drag ratio increased f r o m about 65 to 85 as the Reynolds number increased
f r o m about 2.0 X 106 to 6.0 X lo6. Maximum section lift coefficients were about 30 percent
g r e a t e r than that of a typical older NACA 65 s e r i e s airfoil section and the section lift-drag
ratio at a lift coefficient of 0.90 w a s about 50 percent greater. Agreement of experimental
results with predictions based on a theoretical method which included viscous effects w a s
good for the p r e s s u r e distributions as long as no boundary-layer flow separation was p r e s ent, but the theoretical method predicted d r a g values greatly in e x c e s s of the measured
values.
INTRODUCTION
;

Research on advanced aerodynamic technology airfoils h a s been conducted over the


last s e v e r a l y e a r s at the Langley Research Center. Results of this r e s e a r c h have been
applied to the design of a 17-percent-thick airfoil suitable f o r a propeller driven light
airplane.

3
I-

The subcritical characteristics of thick supercritical airfoil section r e s e a r c h of


reference 1 indicated performance i n c r e a s e s over conventional airfoil sections. Some
of the features that produce these favorable aerodynamic c h a r a c t e r i s t i c s have been
applied in the design of a new low-speed airfoil section. T h i s new airfoil is one of several being developed by NASA for light airplanes and h a s been designated as General
Aviation (Whitcomb) -number one airfoil (GA(W)- 1).
The present investigation w a s conducted to determine the basic low-speed twodimensional aerodynamic character'istics of the NASA GA(W)- 1 airfoil section. In addition, the results are compared to a comparable NACA 6 5 series airfoil section. Such
sections are presently used on some light airplanes. Also, the experimental r e s u l t s are
compared with results obtained f r o m an analytical aerodynamic performance prediction
method.
The investigation was performed in the Langley low-turbulence p r e s s u r e tunnel
over a Mach number range f r o m 0.10 to 0.28. The Reynolds number, based on airfoil
chord, varied f r o m about 2.0 X lo6 to 20.0 X lo6. The geometrical angle of attack varied
f r o m about -loo to 24O.
SYMBOLS
Values a r e given both in SI and the U.S. Customary Units.
calculations were made in the U.S. Customary Units.

a
cP

mean- line designation


p r e s s u r e coefficient,

- pco

PL

qco

chord of airfoil, c m (in.)

CC

section chord-force coefficient,

forward (t/c),,

Cd

The measurements and

max

section profile-drag coefficient determined f r o m wake measurements,

'd'

point d r a g coefficient,

Cn cos

Cl

section lift coefficient,

c2, i

design section lift coefficient

CY

cc sin

CY

section pitching-moment coefficient about quarter chord,

cn

section normal-force coefficient,

vertical distance in wake profile, c m (in.)

l/d

section lift-drag ratio,

f r e e - s t r e a m Mach number

statfc p r e ss u r e, N/ m2 (lb/f t2)

Cp d($)
L.S.

Cp d(5)
U.S.

Cl/Cd

dynamic pressure, N/m2

(lb/ft2)

Reynolds number based on free-stream conditions and airfoil chord

airfoil thickness, c m (in.)

airfoil abscissa, c m (in.)

airfoil ordinate, c m (in.)


mean line ordinate, c m (in.)

CY

angle of attack of airfoil, angle between chord line and a i r s t r e a m axis, deg

density, kg/m3

(slugs/ft3)

Il1II Illll

Subscripts:

local point on airfoil

max

maximum

thickness

tunnel station 1 chord length downstream of model

tunnel station downstream of model where static p r e s s u r e is equal to


f r e e - s t r e a m static p r e s s u r e

undisturbed s t r e a m conditions

Abbreviations:
GA(W)- 1

General Aviation (Whitcomb)-number one

1. s.

lower surface

U.B.

upper surface
AIRFOIL DESIGN

The airfoil section (fig. 1) w a s developed by employing some of the favorable characteristics of the thick supercritical airfoil of reference 1, which indicated performance
i n c r e a s e s over conventional airfoils at subcritical conditions. In o r d e r to expedite the
airfoil development, the computer program of reference 2 was used to predict the resuits
of various design modifications. The final airfoil shape w a s defined after 17 iterations on
the computer. The airfoil is 17 percent thick with a blunt nose and a cusped lower s u r face n e a r the trailing edge. The design c r u i s e lift coefficient w a s about 0.40 a t a Reynolds
number of about 6 X 106 . In defining the airfoil emphasis w a s placed on providing good
lift-drag ratios a t cl = 1.0 for improved climb performance, and on providing a maximum
lift coefficient of about 2.0. Several key design f e a t u r e s of the airfoil are:

1. A large upper surface leading-edge radius (about 0.06~)


was used to attenuate the
peak negative p r e s s u r e coefficients and therefore delay airfoil stall to high angles of attack.

2. The airfoil was contoured to provide an approximate uniform chordwise load


distribution near the design lift coefficient of 0.40. To account f o r viscous effects this
airfoil incorporated more camber in the rear of the airfoil than the NACA mean camber
line (fig. 2).

3. A blunt trailing edge w a s provided with the upper and lower surface slopes
approximately equal to moderate the upper surface pressure recovery and thus postpune
the stall.
1

The airfoil thickness distribution and camber line are presented in figure 2. Table I
presents the measured airfoil coordinates.
APPARATUS AND PROCEDURE
Model Description
The airfoil model was machined f r o m an aluminum billet and had a chord of
58.42 c m (23 in.) and a span of 91.44 c m (36 in.). The airfoil surface was fair and
smooth. Figure 3 shows a photograph of the model. The model was equipped with both
upper and lower surface orifices located a t the chord stations indicated in table II. A
base p r e s s u r e orifice was included in the blunt trailing edge of the airfoil (x/c = 1.0).
In o r d e r to provide data f o r a simple flap deflection, an aluminum wedge w a s installed
'
0
6
Orifices w e r e installed on this
on the model to simulate a split flap deflected .
simulated flap as indicated in table II.

Wind Tunnel
The Langley low-turbulence p r e s s u r e tunnel (ref. 3) is a closed-throat singlereturn tunnel which can be operated a t stagnation p r e s s u r e s from 101.3 to 1013 kN/m 2
(1 to 10 atm) with tunnel-empty test-section Mach numbers up to 0.46 and 0.23, respectively. The maximum unit Reynolds number is about 49 X 106 per m e t e r (15 X 106 p e r
foot) a t a Mach number of 0.23. The test section is 91.44 c m (3 f t ) wide by 228.6 cm
(7.5 ft) high.
Circular end plates provided attachment f o r the two-dimensional model. The end
plates are 101.6 c m (40 in.) in diameter and a r e flush with the tunnel wall. They are
hydraulically rotated to provide for model angle-of-attack changes. The airfoil w a s
mounted s o that the center of rotation of the circular plates w a s a t 0 . 2 5 ~on the model
chord line. The air gaps at the tunnel walls were sealed with flexible-sliding metal
seals (fig. 4).

Wake Survey Rake

A fixed wake survey r a k e (fig. 5) at the model midspan w a s mounted f r o m the tunnel sidewall and located 1 chord length r e a r w a r d of the trailing edge of the airfoil. The
wake rake utilized 91 total-pressure tubes and five s t a t i c - p r e s s u r e tubes 0.1524 c m
(0.060 in.) in diameter. The total-pressure tubes were flattened to 0.1016 c m (0.040 in.)
f o r 0.6096 c m (0.24 in.) f r o m the tip of the tubes. The static p r e s s u r e tubes had four
flush orifices drilled 90' a p a r t and located 8 tube d i a m e t e r s f r o m the tip of the tube and
in the measurement plane of the total-pressure tubes. T h r e e tunnel sidewall static
p r e s s u r e s w e r e also measured f r o m orifices located in the measurement plane of the
total-pressure tubes. One static orifice was located on the center line of the tunnel and
the other two orifices w e r e about 0 . 3 5 ~above and below the center line of the tunnel.
Inst rumentation
Measurements of the static p r e s s u r e s on the airfoil s u r f a c e s and the wake rake
p r e s s u r e s w e r e made by an automatic pressure-scanning system utilizing variable
capacitance type precision transducers. Basic tunnel p r e s s u r e s w e r e measured with
precision quartz manometers. Angle of attack w a s measured with a calibrated potentiometer operated by a pinion gear and r a c k attached to the circular plates. Data were
obtained by a high-speed data-acquisition system and recorded on magnetic tape.
TESTS AND METHODS
The airfoil was investigated at Mach numbers f r o m 0.10 to 0.28 over an angle-ofattack range f r o m about -loo to 24O. Reynolds number based on the airfoil chord was
varied f r o m about 2.0 X lo6 to 20.0 X 106, primarily by varying the tunnel stagnation
pressure. The model was tested both with the wake rake installed and removed to determine its influence on the flow over the airfoil. Figure 6 shows typical lift coefficient and
pitching-moment-coefficient data and no effects were indicated. The p r e s s u r e distribution data also indicated no effect of the wake rake on the flow over the airfoil. The airfoil w a s tested both smooth (natural boundary-layer transition) and with roughness located
on both upper and lower s u r f a c e s a t 0 . 0 8 ~ . The roughness was sized according to reference 4 which indicated a nominal roughness particle height of 0.0107 c m (0.0042 in.) at a
Reynolds number of 6 x 106 and 0.0257 cm (0.0101 in.) at a Reynolds number of 2 X 106.
The corresponding commercial grit numbers required a r e number 120 and number 60.
The transition s t r i p s w e r e 0.25 cm (0.10 in.) wide. The roughness was sparsely spaced
and attached to the airfoil surface with lacquer. Several different roughness s i z e s were
used f o r the s a m e test conditions and these results are shown in figure 7. F o r several
runs the standard NACA method of applying roughness (number 60 grit wrapped around
leading edge on both surfaces back to 0 . 0 8 ~ )w a s employed (ref. 5). F o r s e v e r a l test-runs
6

oil w a s spread over the airfoil upper surface to determine if any local flow separation
w a s present. Tufts w2re attached to the airfoil and tunnel sidewalls with plastic tape t o
determine stall patterns on both the airfoil and adjacent sidewalls.
The static-pressure measurements at the airfoil surface were reduced to standard
p r e s s u r e coefficients and then machine integrated to obtain section normal-force and
chord-force coefficients and section pitching-moment coefficients about the quarter chord.
Section profile-drag coefficient w a s computed from the wake rake measurements by the
method of reference 6. The wake rake static-pressure measurements indicated some
influence on the static p r e s s u r e s due to the presence of the rake body; therefore, the
tunnel sidewall static p r e s s u r e s w e r e used i n computing the section profile-drag
coefficients.
An estimate of the standard low-speed wind-tunnel boundarp corrections as calculated by the method of reference 7 is shown in figure 8. These corrections amount

to about 2 percent of the measured coefficients and have not been applied t o the data.
An estimate of the total head tube displacements effects on the values of Cd showed
these effects to be negligible.
RESULTS
The results of this investigation have been reduced to coefficient f o r m and a r e
presented in the following figures:
Figure

Tuft photographs of NASA GA(W)- 1 airfoil

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

Effect of Reynolds number on section characteristics, model smooth


Effect of Mach number on section characteristics, model smooth,
R=6X106 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.......

.......

9
10
11

Effect of Reynolds number on section characteristics, roughness located


at0.08~

12

Effect of Mach number on section characteristics, roughness located


at O.O8c, R = 6 x l o 6 .

13

Effect of

14

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

..............................
roughness on section characteristics . . . . . . . . . . . . . . . . . . .

Comparison of section characteristics between NASA GA(W)- 1 and


NACA 652-415 and 653-418 airfoils

15

Variation of maximum section lift coefficient with Reynolds number


f o r various airfoils without flaps

16

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

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

llIlIllIllllIlll llllll llllll

Figure

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

Section characteristics for 0.20~simulated split flap deflected 60'

17

Effect of angle of attack on chordwise p r e s s u r e distributions

18

....
Comparison of experimental and theoretical chordwise p r e s s u r e distributions . .
Comparison of experimental and theoretical aerodynamic characteristics

19
20

DISCUSSION O F RESULTS

Experiment a1 Results
Lift.- Figure 10 shows that with the airfoil smooth (natural boundary-layer transition) a lift-curve slope of about 0.12 p e r degree and a lift coefficient of about 0.52 at
a = 0'
w a s obtained f o r all Mach numbers and Reynolds numbers investigated, Maximum lift coefficients increased f r o m about 1.64 to about 2.12 as the Reynolds number
was increased f r o m about 2 X lo6 to 12 X lo6 at M = 0.15 (fig. 16), with the most
rapid increase occurring between Reynolds numbers of 2 x 106 and 6 x 106. Increasing the Reynolds number above 12 X lo6 had no additional effect on maximum lift coefficient as shown by figure 1O(b) (M = 0.20).
The GA(W)- 1 airfoil section encounters a gradual type stall (fig. lo), particularly
in the lower Reynolds number ranges. Tuft pictures (fig. 9) indicated the stall is of the
turbulent o r trailing-edge type. (See also p r e s s u r e data of fig. 18.)
At a Reynolds number of 6.0 X lo6, increasing the Mach number f r o m 0.10 to 0.28
had only a minor effect on the lift characteristics as shown by the results presented in
figure ll(a). The stall angle of attack was decreased about 2O and maximum lift coefficient about 5 percent.
The addiiion of roughness at 0 . 0 8 ~(figs. 12 and 14) did a l t e r the effective airfoil
shape because of changes in boundary-layer thickness, particularly for R = 2.0 X 106 as
shown in figure 14(a). For example, the angle of attack for z e r o lift coefficient changed
No measurable change in lift-curve slope was indicated; theref r o m about - 4 O to -3.6'.
fore, the lift coefficient at (Y = Oo decreased f r o m about 0.52 to about 0.43. These

effects on the lift characteristics decreased as the Reynolds number was increased above
2.0 x 106 as might be expected because of the related d e c r e a s e in boundary-layer thickness. Figure 13(a) indicates that the effects of Mach number with roughness applied to
the airfoil were s i m i l a r to those with the model smooth.
Comparisons of the values of ( C Z ) ~for
~ the NASA GA(W)-1 airfoil with other
NACA airfoils without flaps a r e shown in figure 16. Substantial improvements in (c,7kax
for the GA(w-')-l airfoil throughout the Reynolds number range are indicated when compared
8

t o the NACA 4 and 5 digit airfoils and 65 series airfoils. Both the GA(W)-l and 653-418
airfoils have the s a m e design lift coefficient (0.40) and figure 2 indicates both airfoils
have roughly the s a m e mean thickness distribution in the region of the structual box
( 0 . 1 5 ~to 0 . 6 0 ~ ) . At a Reynolds number of 6.0 X lo6, a 30-percent improvement in
(C2)max is shown f o r the GA(W)-l airfoil over the comparable 653-418 airfoil. Typical
operating ranges of Reynolds numbers f o r general aviation airplanes are from about
2 X lo6 to 6 x lo6. These improvements result from two primary considerations of the
design; first, attenuating of the peak negative pressure coefficients on the upper surface
near the leading edge by use of a large leading-edge radius, and second, the attainment
of increased aft loading by using the greater aft camber. Figure 15(a) shows a comparison of the lift characteristics of the NACA 65 series airfoils and NASA GA(W)-l airfoil
a t a Reynolds number of about 6 X 106 with roughness located near the leading edge of the
airfoils. Even when the large wraparound roughness w a s employed on the new airfoil,
it exhibited superior lift characteristics to the older NACA 65 series airfoils although
it stalled about 4' e a r l i e r than with the narrow s t r i p roughness now usually employed.
In o r d e r to obtain some preliminary information on the new airfoil section in a
high-lift configuration, a simple 0 . 2 0 ~split flap deflected 60' w a s installed on the model.
The increment in cz a t a Reynolds number of about 6.0 X lo6 between the basic airfoil
and flapped airfoil w a s about 1.46 a t (Y = Oo and about 1.2 a t ( c ~ ) (Compare
~ ~ .
figs. lob) and 17.) The data of reference 5 indicate an increment in c2 of about 1.40
at CY = Oo and about 1.2 at ( ~ 2 f)o r ~the ~NACA 653-418 airfoil. Comparison of the
values of ( ~ 2 for
) the
~ ~two~ airfoils with the simulated split flap deflected 60 shows
about a 17-percent increase f o r the GA(W)-l airfoil (3.16 compared to 2.70). Similar
improvements a r e also indicated when compared to the NACA 4418 airfoil with simulated
split flap deflected 60'.
The stall characteristics of the GA(W)-1 airfoil with the simulated flap were gradual as indicated by the lift characteristics and tuft studies.
Pitching moment.- The pitching-moment-coefficient data (fig. 10) w e r e generally
insensitive to F?,eynolds number in the low angle-of-attack range. However, f o r angles
of attack greater than about 4' the low Reynolds number data indicate less negative
values of cm. Increasing the Reynolds number, which results in a decrease in boundarylayer thickness, caused negative increments in c m up to airfoil stall. At a Reynolds
number of 6.0 X lo6 increasing the Mach number from 0.10 to 0.28 (fig, ll(a)) caused no
effect on the pitching-moment data up to about 12'.
At higher angles of attack a positive
increment in c m is shown.
The addition of roughness (figs. 12 and 14) at 0 . 0 8 ~resulted in a positive increment
in cm at Reynolds numbers of 2 X lo6 and 4 X 106. However, at a Reynolds number of
6 X lo6 this increment had essentially disappeared.

Comparison of the pitching-moment data of the GA(W)- 1 airfoil with that of the
653-418 airfoil (fig. 15(a)) at a Reynolds number of about 6 X lo6 indicates a more
negative Cm of about 0.04 for the GA(W)-l section. This is expected because of the
aft loading of the GA(W)-1 airfoil section as illustrated by the camber distribution of
figure 2.
Drag.- The profile d r a g data of figure 10 generally show, at moderate lift coefficients, the expected d e c r e a s e in Cd with i n c r e a s e s in Reynolds number. This drag
reduction is associated with the related d e c r e a s e s in boundary-layer thickness and
accompanying reduction in skin friction drag. An increased amount of laminar flow is
indicated at a Reynolds number of about 2 X lo6 (fig. lO(a))by the low values of Cd
obtained in the low lift-coefficient range (typical laminar bucket). In practical applications no laminar bucket, such as shown here, should be expected since the design
velocity characteristics w e r e not selected for this purpose. Laminar flow designs are
generally impractical f o r general aviation airplanes since transition is usually fixed
near the leading edge of the airfoil by the roughness of construction or insect remains
gathered in flight.
F o r general aviation application, the d r a g data of most practical interest are
those obtained with a turbulent boundary layer over most of the airfoil chord in the
Reynolds number range f r o m about 2 X lo6 to 6 X lo6. Figure 12(b) illustrates the
d r a g data with fixed transition at 0 . 0 8 ~f o r this Reynolds number range. The d r a g
coefficient at the design lift coefficient (cl = 0.40) a t a typical c r u i s e Reynolds numb e r of 6 X lo6 is about 0.0108. However, figures 7(b) and 12(b) indicate a l a r g e liftcoefficient range where the values of Cd remain approximately constant. This is
of particular importance f r o m a safety standpoint f o r light general aviation airplanes
where large values of section lift-drag ratio at high lift coefficients result in improved
climb performance. Thus, at cl = 1.0 section lift-drag ratios vary f r o m about 65 at
R = 2.1 X 106 to about 85 at R = 6.3 X 106 (fig. 12(b)).
A comparison between the section lift-drag characteristics of the GA(W)-1 airfoil
and the older NACA 65 s e r i e s airfoils is shown in figure 15(b). For the older type airfoils a c o a r s e size grit was extensively applied ( 0 . 0 8 ~ over
)
the airfoil in o r d e r to
achieve transition in the wind tunnel. This older method of applying l a r g e wraparound roughness (NACA standard) results in an increment in cd of about 0.0010
at cl = 0.40 for the GA(W)-l airfoil when compared to the narrow roughness s t r i p now
usually employed (NASA standard). However, in o r d e r to obtain a direct comparison
of the d r a g coefficients between the airfoils the comparison is made with the older NACA
standard method of employing roughness.

On this basis figure 15(b) shows at R =: 6.0 X lo6 that the section d r a g coefficient,
at a c r u i s e lift coefficient of about 0.40, f o r the 17-percent-thick GA(W)-l airfoil is about
10

-_

-. .. . ... ..

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

. . .. .. . ..

0.0010 higher than that f o r the 653-418 airfoil. However, comparison of the section liftdrag ratios at a lift coefficient of 0.90, the highest' ct for which data w e r e available,
indicates about a 50-percent improvement for the GA(W)-l airfoil section (Z/d = 47 f o r
653-418 airfoil compared to Z/d = 70 for GA(W)-l airfoil). The figure also indicates
that even greater improvements would probably occur at higher climb lift coefficients.
P r e s s u r e distributions.- The chordwise p r e s s u r e data of figure 18 illustrate the
effects of angle of attack f o r a Reynolds number of 6.3 X lo6. The data at CY = '0
(cl = 0.47) indicate approximately constant values of Cp from about x / c = 0.05 to
x/c = 0.55 f o r both the upper and lower surfaces of the airfoil (design condition). Upper
and lower surface p r e s s u r e coefficients at the airfoil trailing edge are slightly positive.
Some upper surface trailing-edge separation is first indicated at a n angle of attack of
about 8' by the constant p r e s s u r e region on the upper surface of the airfoil and is also
indicated by the nonlinear lift curves above t h i s angle of attack. Increases in angle of
attack above 8' resulted in this constant p r e s s u r e region moving forward along the airfoil and a t maximum lift coefficient (CY= 19.06O) trailing-edge separation was present
from about x/c = 0.70 to the airfoil trailing edge. The airfoil stall is of the turbulent,
o r trailing edge, type as indicated by figure 18(k) (a = 20.05O) and as observed by means
of tuft studies.
Comparison of Experimental and Theoretical Data
Predictions of the aerodynamic characteristics by the viscous flow method of reference 2 are compared with t h e experimental results at R = 6.3 X lo6 in figure 19. AS
previously mentioned, this viscous flow method w a s employed during the development of
the GA(W)- 1 airfoil shape. The theoretical method predicts the lift and pitching-moment
data well f o r angles of attack where no boundary-layer flow separation is present (up to
about 8'). However, the theoretical method overpredicts the d r a g data throughout this
s a m e lift-coefficient range. Examples of p r e s s u r e distributions calculated by the theoretical method a r e compared with the experimental p r e s s u r e s in figure 20. The agreement between experiment and theory is good over most of the chord length of the airfoil,
as long as no boundary-layer flow separation is present.
I

CONCLUDING REMARKS
Wind-tunnel tests have been conducted to determine the low-speed two-dimensional
aerodynamic characteristics of a 17-percent-thick airfoil section designed f o r general
aviation applications. The r e s u l t s w e r e compared with a typical older NACA 65 series
airfoil section. Also, the experimental data are compared with predictions based on a
theoretical method f o r calculating the viscous flow about the airfoil. The tests w e r e

11

1l1l11 III I l l l l Il l 1

conducted over a Mach number range f r o m 0.10 to 0.28. Reynolds number based on the
airfoil chord was varied f r o m about 2 X lo6 to 20 X lo6. The following results w e r e
determined f r o m this investigation:

1. Maximum section lift coefficients increased rapidly a t Reynolds number f r o m


about 2.0 X lo6 to 6.0 X lo6 and attained values g r e a t e r than 2.0 f o r the plain airfoil and
g r e a t e r than 3.0 with a 20-percent-chord split flap deflected 60.
2. Stall characteristics w e r e generally gradual and of the trailing-edge type either
with o r without the split flap.

3. Section lift-drag ratio at a lift coefficient of 1.0 (climb condition) increased


f r o m about 65 to 85 as the Reynolds number increased f r o m about 2.0 X 106 to 6.0 X 106.
4. Maximum section lift coefficients were about 30 percent g r e a t e r than a typical
older NACA 65 series airfoil and the section lift-drag ratio at a lift coefficient of 0.90
was about 50 percent greater.

5. Comparison of experiment with predictions based on a theoretical method which


included viscous effects was good for the p r e s s u r e distributions as long as no boundaryl a y e r flow separation w a s present, but the predicted d r a g values w e r e much g r e a t e r than
measured values.
Langley Research Center,
National Aeronautics and Space Administration,
Hampton, Va., October 16, 1973.

12

REFERENCES

1; Ayers, Theodore G.: Supercritical Aerodynamics: Worthwhile Over a Range of


Speeds. Astronaut. & Aeronaut., vol. 10, no. 8, Aug. 1972, pp. 32-36.
2. Stevens, W. A.; Goradia, S. H.; and Braden, J. A.: Mathematical Model for TwoDimensional Multi-Component Airfoils in Viscous Flow. NASA CR- 1843, 1971.

3. Von Doenhoff, Albert E.; and Abbott, Frank T., Jr.: The Langley Two-Dimensional
Low-Turbulence P r e s s u r e Tunnel. NACA TN 1283, 1947.

I
'

4. Braslow, Albert L.; and -ox, Eugene C.: Simplified Method f o r Determination of
Critical Height of Distributed Roughness Particles f o r Boundary- Layer Transition
at Mach Numbers F r o m 0 to 5. NACA TN 4363, 1958.
5. Abbott, Ira H.; and Von Doenhoff, Albert E.:
Inc., e. 1959.

Theory of Wing Sections. Dover Publ.,

6. Baals, Donald D.; and Mourhess, Mary J.: Numerical Evaluation of the Wake-Survey
Equations f o r Subsonic Flow Including the Effect of Energy Addition. NACA
WR L-5, 1945. (Formerly NACA ARR L5H27.)
7. Allen, H. Julian; and Vincenti, Walter G.: Wall Interference in a Two-DimensionalFlow Wind Tunnel, With Consideration of the Effects of Compressibility. NACA
Rep. 782, 1944. (Supersedes NACA WR A-63.)

13

TABLE I.- NASA GA(W)- 1 AIRFOIL COORDINATES


= 58.42 cm (23 in.)]

(z/c)u,

0.0
.002
.005
.0125
.025
.0375
.05
.075

.loo
.125
.150
.175
.20
.25
.30
.35
.40
.45
.50
.55
.575
.60
.625
.65
.675
.700
.725
.750
.775
.800
,825
.a50
.875
.goo
,925
.950
.975
1.000

14

-6

0.0
.01300
.02035
.03069
.04165
.04974
.05600
.06561
.07309
.07909
.08413
.08848
.09209
.09778
.lo169
.lo409
.lo500
.lo456
.lo269
.09917
.09674
.09374
.09013
.08604
.08144
.07639
.07096
.06517
.05913
.05291
.04644
.03983
.03313
.02639
.01965
.01287
.00604
-.00074

(Z/c)lOWer

0.0
-.00974
-.01444
-.02052
-.02691
- .03191
-.03569
-.04209
-.04700
-.05087
- .Os426
-.05700
- ,05926
-.06265
-.06448
- .06517
- .06483
- .06344
-.06091
-.05683
-.05396
-.05061
-.04678
-.04265
-.03830
- .03383
-.02930
- .02461
-.02030
-.01587
- .01191
-.00852
-.00565
-.00352
- .00248
- .00257
-.00396
- .00783

TABLE 11.- AIRFOIL OFUFICE LOCATIONS

Upper surface

x/c
0.0
.00630
.01248
.01730
.02461
.03713
.04961
.06222
.07522
.lo013
.14970
.ZOO04
.24991
.29965
.34991
.39978
.44974
.50004
.55035
.60017
.64996
.70004
.75000
.79987
.a5004
.go004
.95026
.99004

Lower surface

__

- ..

x/c

Z/C

0.00678
.01204
.01722
,02596
.03726
.04970
.06196
.07422
.09957
.14961
.19943
.24965
.30004
.34983
.39991
.45009
.49983
.54970
.59983
.65022
.70022
.75000
.BOO13
.8 5004
.a9987
.94970
.99004
1.0

0.00030
.00228
.03083
.03543
.04143
.04957
.05583
.06109
.06570
.07313
.OB409
.09209
.09778
.lo170
.lo409
.lo500
.I0457
.lo270
.09917
.09374
.08609
,07643
.06522
,05296
.03987
.02643
.01287
.00204
.-

Z/C

~-

0.01635
-.02035
-.02326
-.02683
-.03187
-.03583
-.03909
- .04200
-.04700
-.05426
-.05930
-.06265
-.06452
-.06517
-.06487
-.06348
- .06100
-.05691
-.05070
-.04270
-.03387
-.02483
-.01596
-.00857
-.00357
-.00261
-.00613
-.00430

__

Lower surface orifices f o r 0.20~simulated


split flap deflected 60'

x/c
-0.03913
-.06087
-.07696
-.09565
-.11413
-.13304
-.15217
-.17065

0.81304
.a2609
.a3478
.a4565
.a5652
.a6739
.a7826
.a8696
.

15

.2

.I
z/c

0
-.I

.I

.2

.3

.4

.5
x /c

.6

.7

Figure 1.- Section shape for NASA GA(W)-l airfoil.

.8

.9

I .o

.I 0

NASA GA ( W ) - l airfoil
NACA 6S3-018 airfoil

,O8
.O6
Zt/C

.O4

.o 2
-

Thickness distribution
-

.I

.2

.3

.5
x /C

.4

.6

-rll---

1 1

I.o

N A S A . GA (w)-l airfoil

-.020

.9

.8

.7

NACA mean line, a=l.O]

Camber line

.I

.2

.3

.4

.5
*/c

.6

.8

.7

.9

Figure 2.- Comparison of camber lines and thickness distribution.

17

..

, _

T
I .57c

L-.

__t

Airflow

,- Tunnel

sidewalls

balance attachment

Seal detail "z"

Tunnel center line

Fnrl u
iow
sprtinn
-...."
.. .""".
.-..

't.-

Zero incidence
reference

A.- A. .
,

in wina runnel. mi aimensions in terms


of airfoil chord. c = 58.42 cm (23 in.).

IUL-LULL

IIIWUIIL~U

19

11111ll1ll1llIl l l I I

Static- pressure probe

7-D--'

--J

.022c

Static -pressure probe

=I
__t

Air f low
Tunnel

-r

%--

.I 9 6 c
[ ty P o l

Total Dressure probe


(tubes flattened)

-3

-3-

-L
Figure 5.- Drawing of wake rake. All dimensions in terms of airfoil chord.
c = 58.42 cm (23 in.).
20

2 .o

I .6

.a
Q
.4

-.4

CIn

-.I

- .-2

Figure 6.- Effect of wake rake on section characteristics.


M = 0.20; R E 6 X 106.

Model smooth;

21

..

Roughness
0

No. 80

No.100
NO,120

20

(a) Variation of

cl and c,

with a.

Figure 7.- Effect of roughness size on section characteristics.


M = 0.20; R = 6 X 106.

22

24

Roughness

I
I:

No. 80
0 No. 100
0 No-120

-03
1

.02

.o I
0
-1.2

-.-

-.8

-.4

0
(b) Variation of

.4

1.2

I .6

Cd with cz.

Figure 7 . - Concluded.

23

ii

No boundary correction
Boundary correct ion

it

i!

I .6
I .2

.8
Cl

.4

- .4
- .8
.I

il
'

0
CmI

-.I
1

- .2-12

-8

-4

-!

12

16

20

(a) Variation of el and c m with a.


Figure 8. - Effect of standard low-speed wind-tunnel boundary corrections on section
data. Number 80 roughness located at 0 . 0 8 ~ ; M = 0.15; R 'I 8 x 106.

No boundary correction

Boundary correction

.03

.02
d

.o I
Q

-. 4

.0

(b) Variation of

Cd with cz.

Figure 8.- Concluded.

25

CY=

00

CY=1
4'

a = 40

C Y = 16O

a! =

100

C Y = 18'

L-73- 6898

Figure 9.- Tuft photographs of NASA GA(W)-lairfoil. M = 0.20; R = 2.70 X 106.

2.4

. I

0
A

I .6

..,..

R
0

2 .o

, .

1 1

I ,

I '

1.9x106

3.9
5.7
9.2
2.3

i;
I

1i
i

I .2

+ I

iI

II

.0
=1

.4

I
+
-

- .4

I!

- .0

C
Cm

-.I

-.2
-12

-8

-4

4
Q

0
,deg

12

16

20

24

(a) M = 0.15.
Figure 10.- Effect of Reynolds number on section characteristics.

Model smooth.

27

lllI llIll l 11l1l1l I l l l l

Il 1l1l11llI

.04
.03

'd

I
II

.02

.o I
0

I .6

(a) M = 0.15. Concluded.

Figure 10. - Continued.

28

2.4

R
0 2.7~10~
0 4.1
0 6.2

2 ,o

9.3

I .6

n 12.2
16.5
!

I .
I .

-.4
I

4 .
I .

- .E

I
I..

:::I:
...

.I

,
I

I
I

..
I

I . .
. II ./

-8

-4

12

(b) M = 0.20.
Figure 10.- Continued.

16

20

R
2.7~10~

.03
.02
'd

.o I
2.0
0.20.

Concluded.

10. - Continued.

30

2.4

3.5 x IO6
6.0
0 9.I

0
0

2 .o
I .e

1.5

3.6

I .2
I

.E
=1
1

- .8
.. .. .. .

.I . ... .
. I

1;:
t

-.I

-.2

(c)

12

16

20

24

M = 0.28.

Figure 10. - Continued.

31

___

.....

.03

mtu

.02

'li
il

Cd

.o I
0 ,2

- .8

- .4

.4

.8

C1

(c)

M = 0.28. Concluded.

Figure 10. - Concluded.

32

7.
....

. ......_. .

I .2

I.6

2 .o

Variation of

cz and cm with a.

Figure 11.- Effect of Mach number on section characteristics.


Model smooth; R = 6 X 106.
33

..

II Il1 III1

l l l lll I I

.04

II

Id

.03
d .02

1
t

tI-

.o 1

I
I
I

0-

1
1.2

-.8

-.4

.4

.0

(b) Variation of

Cd

with cz.

Figure 1 1. - Concluded.

34

h
.. .. ....

1.2

2.0

2.0

'

I.6

I.2

.0
Ca

.4

-.4
t i

j
iI
ij
i

-.2 I

-12

-8

-4

12

16

20

24

(a) M = '0.10; R = 3.7 x 106.

Figure 12. - Effect of Reynolds number on section characteristics.


Number 80 roughness located at 0 . 0 8 ~ .
35

.03

002

.o I
0
-,1.2 -.8

.4

-.4

Cl

(a) M = 0.10; R = 3.7

lo6. Concluded.

Figure 12. - Continued.

36

H
2 .o

2.1x IO6

4.3

6.3

I .6

!I
1

I .2

it

.e
c1

IT

.4

I
I
I

- .

cm

-. I

P
t

-.2
-

i
2

-0

-4

20

24

a,deg

(b) M = 0.15.

Figure 12.- Continued.

37

:i.

. ,

1111111l1l11l1lI Il Il 1 Il 1 l l l Il l

(b) M = 0.15.

Concluded.

Figure 12. - Concluded.

38

, .
, , : I

..

.- .

M
I

2.o

i!

+
!

0.15
.20
.2

I.6
I
t

I.2

.a
=1

i
i

II
I
t

.4

i
I

12

16

a ,deg

(a) Variation of

c l and cm with a.

Figure 13.- Effect of Mach number on section characteristics.


roughness located at 0 . 0 8 ~ . R = 6 X IO6.

Number 80

39

II

I I 1
l11111
l111
llIl1
lIl

-0E

.05

.20
.28

.04

M
0.1 5

i
d

.02

1
I

.o I

i
I
I
I

-0

i
.4
c1

(b) Variation of Cd with


Figure 13. - Concluded.

40

1.2

I.6

2 .o

2.o

o o n

-.4

-.I

-.2

- 12

-8

-4

I'

16

2.o

Figure 14.- Effect of roughness on section characteristics. M = 0.15.


Number 80 roughness located at 0 . 0 8 ~ .
41

,
..

I I I1 111l1l 1l11l1l1l 1

1
l1
l1
l1
l1
l1
lIll l l l l l l

I1
C

-1
-I
i
i

I
I

I
-0

(a) R = 2 X lo6. Concluded.

Figure 14. - Continued.

42

2.o

I
2 .o

I .6
12

=1

12

16

20

24

43

li

!I

.04

II

If

't

.03

i.

L
Y
A

fl

d
Fh

-02

v
f.
I

.o I

f1

II
f

0-

I .6

(b) R = 4 x 106. Concluded.


Figure 14.- Continued.

44

. .
Roughness
...

2.41

Off
On

0
0

1.61

I.2

.8
Cl

.4

- .4
- .a
I

Cm
-.I
,. . ..

-0

-4

(c) R

=:

12

16

20

24

6 X 106.

Figure 14. - Continued.

45

11l11l111ll1l1l1l1l1111ll Il 1 Il1 l Il l l Il 1 I l l l l I l I I I I

005

.04
.03
Cd
.02

.o I
.4

-8

=1

(c) R

= 6 X 106. Concluded.

Figure 14.- Concluded.

46

1.2

I .6

2.o

N A S A G A ( W ) - I airfoil
NASA standard roughness
0
NACA standard roughness

NACA airfoil ,NACA standard roughness (ref.5)


6 5 2 - 415
653- 418

2 .o

I .6
I

I .2

i
i

.e

,!

Cl

.4
0

- .4

- .a
..,
I

,I
I

:it:.-cm data for model

t!: :

smooth

I . , ,
I

I . .

1'

l...

... .

cm

. ..

..
..

...

..

+\-TII

-.I

- .2
-

. /

-0

-4

4
Q

12

16

20

24

,deg

(a) Variation of cz and Cm with

CY.

Figure 15.- Comparison of section characteristics of NASA GA(W)- 1 airfoil and


NACA 652-415 and 653-418 airfoils. M = 0.20; R =: 6 X 106.

47

Il11111IlIIIIIIl l l IIll Il I I I I Il 1

NASA GA(W)- I airfoil


o
N A S A standard roughness
0
N A C A standard roughness
NACA airfoil , NACA standard roughness (ref.5 )

- - - 652- 415
.04
-653- 418

.03

'd

.02

.o I
0
-I 2

-.e

-.4

.O

(b) Variation of

Cd with c z .

Figure 15.- Concluded,

4%

NASA G A ( W 1 - I airfoil
0
Roughness off
0
Roughness on

2.2
2.0

I.8

PZh

1.6
I.4

1.2

I .O

6 R

IO

12

14~10~

Figure 16.- Variation of maximum section lift coefficient with Reynolds number
for various airfoils without flaps. M = 0.15.

49

.-.w

..

..,

3.2

. . ,,
I

R
Roughness
o 2 . 7 ~ 1 0 ~ Off
4.1
Off
0 6.1
Off
A 6.1
.08c

2.8

2.4

It

it1 l l i r

2.o
c~ 1.6
~

ct ,!

I.2
,

.8

tJ
.I.I..

,J
t

9-

I
I

: 1

.4

0'
0
I

.
Cm

-.2

-.4-I

12

16

20

Figure 17.- Section characteristics for 0.20~simulated split flap


deflected 60. M = 0.20.

50

-9

Upper surface
Lower surface

-8

-7

-6

-5
cp -4

-3

1-

II
cp

.I

I I

II I I I

.2

I I

I 1

I I

-3

II

I I ! I ~I I , I

.5

.4

(a)

I ! I '

a!

0
!I

I II I II

I1

' 1

I I

I !

I I I

g ii
I II I I Ill

.6

= -7.99'.

Figure 18.- Effect of angle of attack on chordwise pressure distributions.


Number 80 roughness located at 0 . 0 8 ~ . M = 0.15; R = 6.3 X lo6.
(Flagged symbol indicates base pressure orifice.)

51

I.

-jli

- 2.0

- 1.6

__

0
0

0
0

1.2 h I 1 I I I I 1 t

i I I I I ' I I

(b)

(Y

IIIIIIIIIIIIIIIII

= -4.11'.

Figure 18.- Continued.

52

Upper surfoce
Lower surface

2-8E

..

o
-..

..

Upper surface
Lower surface

. .

0
0

Ll

0
Cl

. .

*41
.8

- -

IIIIIIII

I I I I I I ~IIIlIl:I

(c)

cy =

IIIIIII

00.

Figure 18. - Continued.

53

'

1:

.
,.

..

.. ... .,.

- ..

......

-2*8F
.
a&---- -.

.-.

-1.6.

cP

.......

. . . .

.......

.........

0
.

..

.,.

.....

. .- .-

- ....

_....._. . . .

. . .

.-

......

. -

..

__

......

Upper surface
Lower surfoce

. . . .

. . . .

....

. .

.....

...

.-

0
0

.....

......

......

.Q
0

-.4

F
i

0 ..

. . . . .

....

--

0 ..

.....

E.

. . . . .

1.2 LL1.1, 1.1

II

I LI.Il I I

. . . . ._
.

..

.....

-.

.u. .

. . . .

cl
0

...

I I I 1.1 I L , , I I i I I I I I

' II

I I

I II1I

IIIIIII

x/c

(d)

CY

= 2.06'.

Figure 18. - Continued.

54

. . .

1 1.1. I LL.1

1.111 1.1 I

' I I 1 It I I

-2.8
0

Upper surface
Lower surface

-2.41

-1.6

- 1.2
cp -.8

-.4

0
0

0
0
11

IIIIIII

.I

'2

I IIII

llllllll

.3

.4

.5
x /c

.6

.?

1 1

I1 I I l l 1 1 1 L

.8

I .o

(e) a = 4.17'.
Figure 18.- Continued.

55

l1l111 IlIlllIll l l l I 1l 1l 1l 1l 1l 11l11l 11l11l11l l !

0
0

Upper surface
Lower surface

,
0

" 4
I I I 11.1 I ! 11. l.l!J

'0

.I

.2

II

l.lJ 11.1.I I1 I t
.3
,4

.5

c-1

01

= 8.02'.

Figure 18. - Continued.

56

U.U! L

.6

X /C

(f)

.7

-8

.9

I .o

(g) a! = 12.04'.

Figure 18. - Continued.

57
2

;.' ;

..

1 1 III lll111l I I l11ll 11111ll1l l l l l l l I I

.. - . . . . .

-gE-

0
0
.

. . . .

....

.-

- ...

. . . .

. .

_ .

. . .

. .

. .

....

Upper surface
Lower surface

. . .

. .

0
.

ij
I

0
.

c
.

LII.L.I.I I I

.I

.2

. .

'

IIILIII

.3

n l

0
IIII 11 I1

I I I I I ! I1

$6

x/c

(h)

CY

= 16.04'.

Figure 18.- Continued.

58

.......

( I I I I I I1

I I I I I I I I ~I I I I I I I

.5

.4

.7

.0

.9

I .o

(i)

(Y

= 18.25'.

Figure 18.- Continued.

59

11 111llI l Il 1 1lI1 1l1l1l1l1111ll1l11l1Il I

Upper surface

Lower surface

0
(
Q

, I (.I1 1 I I

.3

IIIIJII

.4

IIIIIIII

.5

.6

X/C

(j)

a! =

19.06'.

Figure 18. - Continued.

60

IIIIIII

.7

.0

.9

I .o

0
0

.I

-7
-6

Upper surface
Lower surface

i'
o

-5k

I
I
tI

'f'

[J

.3

.4

.5

,O

.6

x /C

(k)

CY

= 20.05'.

Figure 18.- Continued.

61

.i

Il1 l l l

IIIIII

. . .

.-

. . . .

0
0
.

...

_.

..

. . .

.~

. .

. .

Upper surface
Lower surface

. . . . .

---....

. Q .

....

'

. . . .

I I I I I. I I

.I

.2

.3

.c1

0 -

'

111.11 I I

.5

,4

.6

X/C

(1)

= 21.14'.

(11

Figure 18.- Concluded.

62

.I 1.1 I .I I.

.7

.8

.9

I .o

>

~ . -

2.8

. ...........
Ex perlment
0
Theory

. .

I' I
, , ..

,.

,'

'I

ii i

2.4

jj

surface trc

2.o

separatic

12

.4

-.4
-.a
.I

C
cm
-.I

-.s

-12

-8

-4

(a) Variation of

12

16

20

24

c1 and cm with a.

Figure 19.- Comparison of experimental and theoretical aerodynamic characteristics.


Transition fixed at 0 . 0 8 ~ ; M = 0.15; R = 6.3 x 106

63

1l1l111I111l11

II I Il

.05
0

Experlment
Theory

.04
.03
'd

.02

.o I
0-

-.O

..

-,4

,L

(b) Variation of

Cd with c i .

Figure 19.- Concluded.

64

I .6

2 .o

..

Experiment

-081

0
0

-2.41

--4

-.

Theory.

---

Upper surface
Lower surface

i
I

I
If
JX
.8

I.

1.2 --

.I

.2

.8

.?

.6

.3

.9

I,@

(a) a = -4.11O.
Figure 20. - Comparison of experimental and theoretical chordwise pressure
distributions. Transition fixed at 0 . 0 8 ~ ; M = 0.15; R = 6.3 X lo6.
(Flagged symbol indicates base pressure orifice.)

65

. -

. .:

Experiment

-2.4

Theory

---

Upper surface
Lower surface

cP

3I
L

1.-

1
I
I
1
I
.5

.4

.6

X /C

(b)

cy=

oo.

Figure 20. Continued.

66

.7

.0

.9

1.0

:'

(c)

(Y

= 4.17'.

Figure 20.- Continued.

67

--.-.

- 3 0 6 1 7 Experiment

Theory

.a

(d) a = 8.02'.

Figure 20.- Continued.

68

.9

I .o

-9
Experiment

Theory

-8

_--

Upper surface
Lower surface

-7

-6

-5

cp - 4

-3

-2
-I
0
I

Ep.

.6

(e)

a = 12.04'.

Figure 20.- Concluded.


I

NASA-Lurgleg. 1973

-1

L-9132