r y-e, _ "--'-',_,',w'.,'mlllW.

lk_,

J,,

rC

,i NASA TECHNICAL NASATMX-74018

MEMORANDUM

(,AsA-T_-x-7401_)Low-sp_._.oWIND_u.._._. ,79-2496o
[_ oc ._suz.Ts Fo. A MoDIFIeDI_-P_RC_.._-T.ZCK
_ A:rRVOIL (NASA) 41 p HC AO3/MF A01 CSCL 01A
: m-O [!llclas
!; _ G3/02 26857

li I-- LOW-SPEEDWIND-TUNNEL
RESULTS '

i
F:
IU
•
"
_
_
""
FORA MODIFIED13-PERCENT-THICK
AIRFOIL

RobertJ. McGheeand WilliamD. Beasley

if:

L"

lu/A
Nahonal Aeronautics and
Space Admm_strahon

Langley Research Center

HamPton•V, rgmJa 23665

} :
 this researchshowed that the 13-percent-thick
airfoil providedthe best

performancefor this.i_tial thicknessfamily of airfoils. This airfoil,which

i
is designatedas the NASA LS(1)-0413airfoil,has been modified in an attempt ,

to furtherimprovethe low-speedperformance. The airfoilcontourwas changed

!_ to reduce the aft upper-surfacepressuregradientand hence delay_boundary-

layer separationat typicalclimb lift_oefficientsfor light.generalaviation

airplanes. This report presentsthe ba-_iclow-speedsection_characteristics

of this modified airfoil-andevaluatesthe effectson performanceresulting

_i from the change iD airfoilshape.

The investigationwas performedin__tbe_Langley

low-turbulencepressure

i.: tunnel at Mach numbersof 0.15 or less. The chord Reynoldsnumber varied

_' from about l.O x lO6 to 9.0 x l_6 and the geometricalangle of attack varied

_ from about -l_ to 20°.

i,!

SYMBOLS

Values are given in both SI and U.S. CustomaryUnits. The measurements

: and calculationswere made in the U.S. CustomaryUnits.

Cp pressurecoefficient,PL " P®
q_

c airfoilchord, centimeters(inches)

.
° cc /
sectionchord-forcecoefficient, Cp d

cd sectionprofile-dragcoefficient,_' d d(_)
wake

c'd point drag coefficient (ref.5)

cI sectionlift coefficient,cn cos _ - cc sin _

• • ....... ............. -J
 cm sectionpitching-momentcoefficientabout quarter-chordPoint,

._.
'iC, cn sectionnormal-forcecoefficient,- s Cp d

: h v.ertical
distancein wake profile, centimeters(inches)
i: M free-streamMach number
i:!
p static pressure,N/m2 (lb/f_t2),

_"' q dynamicpressure,N/m2 (Ig/ft

2)

R Reynoldsnumber based on free-streamconditionsand airfoilchord

S separationpoint

_' t airfoil thickness,centimeters(inches)

x airfoilabscissa,centimeters(inches)

z airfoil ordinate,centimeters(inches)

_ zc mean line ordinate,centimeters(inches)

i."

i',
i,
zt mean thickness,centimeters(inches)

! _ geometricangle of attack,degrees
ii Subscripts:

.i
!
' L
max local point on airfoil
maximum
i

[. _ undisturbedstream

Abbreviations:

LS(1) low-speedfirst series

[, Mod modified
I

AIRFOILMODIFICATION

The airfoilcontourwas changedto reduce the aft upper-surfacepressure %

gradient (fig. l) and hence reduce boundary-layerseparationat typical climb

i:_ lift coefficientsfor light generalaviationairplanes (cI = l.Eto 1.2). The

maximum thicknessratio, trailing-edgethickness,and design lift coefficient

I = 0.40) of the originalairfoilwere ret_,ned_

_' The modificationto the surfacecontour of airfoilLS(]_)__0413
is

illustratedin figure I. The upper surfacemodificationwas accomplishedby

adding material from approximatelythe 2,5 percentchord stationfairing with

the originalairfoilat the 40 percentchord stationand removingmaterial

_ from this station to the airfoil trailingedge. The lower surfacemodifica- .

: tion was accomplishedby adding material from approximatelythe 50 percent

chord stationto the airfoiltrailingedge. The maximum thicknessof the

_,:
r_
modifiedairfoilwas moved forwardabou_,-5-_cent chord. Figure 2 compares

_ the change in mean thicknessand camber distributionsfor the two airfoils

:_ and figure 3 comparesthe changes in surfaceslope distributions. Coordinates

: for both airfoilsare given in tables I and II.

The theoreticalviscousanalysiscomputer programof reference2 was

used to predictthe pressuredistributionsand boundary-layerseparation

points for the airfoils. Boundary-layertransitionwas specifiedat x/c = 0.03

for the theoreticalcalculationsto ensure a turbulentboundary-layerdevelop-

ment on the airfoils. Figure 4 shows the theoreticalresults for both airfoils

at Reynoldsnumbersof 2.0 x lO6 and 4.0 x lO6. At a lift coefficientof 0.40

and a Reynoldsnumber of 2.0 x lO6 (fig.4(a)) both airfoilsare separation

Iii_ free. At a lift coefficientof 1.20 the theory indicatesa decrease in upper-

4
 surface•separation
of about 0.05c for the modifiedairfoil (reduced.pressure

_ gradient). At thi-ssame lift coefficient(cI = 1.20) and a Reynoldsnumber

of 4.0 x 106 (fig.4(b)) a decrease in separationof _nly about 0.02c is "

shown for the modifiedairfoil. Basedon these theoreticalresults,improve-

I. ments in per-f-o_mance
for the modified airfoilat climb lift coefficientswould

!.
• be exl_ected, particularly at a Reynolds number of_2.0 x lO6. Since the

theoreticalmethod is only valid for attached or boup.da_y-layers

with small

amountsof flow separation,the maximum lift coeffi_for the airfoils

could not be determined from the theory,

:_ MODELS,APPARATUS,AND PROCEDURE

:'

:' Models

The airfoil-m_delswere constructedutilizinga metal core around which

• plasticfill and two thin layers of fiberglasswere used to form the contour
ill
_.'
of the airfoils. The models had chords of 61 cm (24 in.) and spans of

g-I-.44
cm (36 in.). The models were equippedwith both upper and lower surface

_ orificeslocated5.08 cm (2 in.) off the midspan. The airfoil surfacewas

,:_ sanded in the chordwisedirectionwith number 400 dry siliconcarbidepaper

to providea smooth aerodynamicfinish. The model contouraccuracywas

generallywithin +.lO mm (.004 in.).

'_i Wind Tunnel

'_ The Langley lJw-turbulencepressuretunnel (ref. 3) is a closed-throat,

single-returntunnel which can be operatedat stagnationpressuresfrom l to

IO atmosphereswith tunnel-emptytest sectionMach numbersup to 0.42 and 0.22,

respectively. The maximumunit Reynoldsnumber is about 49 x lO6 per meter

 F

_L
¸¸ _

. (15 x 106 per foot) at a Mach number of about 0.22. The tunnel test section

is 91.44 cm (3 ft) v,_deby 228.6 cm (7.5 ft) high.

Hydraulicallyactuatedcircularplates providedpositioningand attach-

ment for the two-dimensional

model. The plates are I01.60 cm (40 in.) in

il diameter,rotatewith the airfoil,and are flushwith the tunnelwall._ The

airfoilends were attachedto rectangularmodel attachment•

plates (fig.5)

and the airfoilwas mounted so that the center of rotation of the circular

plates was at 0.25c on the modelreference_line. The air gaps at the tunnel

il walls betweenthe rectangularplates and the circular plates were sealed with

flexibleslidingmetal seals, shown in figure 5.

P.
F

Wake Survey Rake

,.r

A fixed wake survey rake (fig.6) at the model midspanwas cantilever

_ mounted from the tunnel sidewalland located,

one chord length behind the

trailingedge of the airfoil. The wake rake utilizedtotal-pressuretub_s_

0.1524 cm (0.060 in.) in diameter,and static-pressuretubes, 0.3175 cm

(0.125 in.) in diameter. The total-pressuretubeswere flatteredto O.lOl6 cm

(0.040 in.) for 0.6096 cm (0.24 in.) from the tip of the tube. The static-

;i pressuretubes each had four flush orificesdrilled 900 apart and located8

i_. tube diametersfrom the tip of the tube and in the measurementplane of the
total-pressuretubes.

Instrumentation

Measurementsof the static pressureson the airfoil surfacesand the wake

rake pressureswere made by an automaticpressure-scanningsystem utilizing

variable-capacitance-type
precisiontransducers. Basic tunnel pressureswere

measuredwith precisionquartz manometers. Angle of attack was measured with

f
_i̧¸ '

1 a calibrateddigital shaft encoderoperatedby a pinion gear and rack

attachedto the circularmodel attachmentplates. Data were obtained by a

" high-speed acquisition system and recorded on magnetic tape. :

i
TESTS AND METHODS

i The modified airfoilwas tested at Mach numbersof O.15_.oz_less

over an

i angle-of-attackrange from about -lO° to 20°. Reynoldsnumber based on the

airfoilchord was variec from about l.O x lO6 to 9.0 !O6 The airfoilwas
i:! x .

upper and lower surfacesat 0.075c. The roughnesswas sized for eecn
i tested both smooth (naturaltransition)and with roughnesslocatedon both
i:_ Reynoldsnumber accordingto reference4. The roughnessconsistedof granular-

i .
type strips 0.127 cm (0.05 in.) wide, sparcely distributed, and attached to the
!.

: airfoilsurfacewith clear lacquer.

The static-pressure
measurementsat the airfoil surfacewere r_duced-to

t standardpressurecoefficientsand machine integratedto obtain-section

k "
!: normal-forceand chord-forcecoefficientsand section pitching-momentcoeffi-
cients about the quarter chord. Section profile-drag coefficient--was computed

i:_ from the wake-raketotal and static pressuresby the method reported in

reference 5.

An estimate of the standard low-speed wind-tunnel boundary corrections

(ref.6) amounted to a maximum of about 2 percentof the measured coefficients

and these corrections have not been applied to the data.

PRESENTATIONOF RESULTS

The resultsof this investigationhave been reduced to coefficientform

and are presentedin the followingfigures:

7
f
 I

, Figure
ii.

Effectof Reynoldsnumberon sectioncharacteristics

for

LS(I)-0413
Mod airfoil
........................ 7

_ Comparison
of sectionchaz:acteristics
for LS(I)-0413
and

i, LS(I)-0413
Mod airfoils...................... 8
_ Effectof Reynoldsnumberon chordwisepressuredistributions
F.

{ for LS(1)-0413
Mod airfoil
..................... 9
_i Comparisonof chordwisepressuredistributions--for
LS(1)-0413

Variation of .maximumlift coefficient with Reynolds number for

" and LS(1)-0413 Mod airfoils.................... lO
_,:
,..
LS(I)-0413and LS(I)-041 3 Mod airfoils ........ .ooo.o.
11

i:_
• DISCUSSION
i'

, The airfoilcontourmodificationproduced the theoreticallypredicted

/
_ decrease in aft upper-surfacepressuregradient shown by the experimental

if pressuredata comparisonfor both airfoils in figure lO. Note (fig. lO(a))

I_ that alteringthe shape of the LS(1)-O413airfoilto reduce the aft upper-

i': surfacepressuregradientand retain the design lift coefficientof 0.40

i, removedthe characteristicflat-typepressure distribution. Thus, the modi-
fied airfoilexhibitsa gradualpressurerecoveryof nearly uniformslope

over approximately50 percentof the upper surface. Figure lO(b) shows the

decrease in upper-surfaceboundary-layerseparationat _ = lO° and R = 2.0 x lO6

for the modified airfoilas predictedby the viscousanalysismethod of

reference2 and discussedunder "AirfoilModification." Boundary-layer

_, separationis indicatedby the lack of pressure gradienton the upper surface

near the trailingedge of the airfoils. At o,= lO° and R = 4.0 x 106,

8
i'
F
• (fig.lO(c)) the pressuredistributionsfor both airfoils indicate]ess but

" about equal amountsof upper-surfacetrailing-edgeseparation. This trend

of decreasedseparationat higher Reynoldsnumberswas also indicatedby the

ii,

, theoreticalmethod. Figure lO(d) comparesthe pressuredata for the two

i airfoilsat _ _ 16° and R_ 2.0 x IOG. For this angle of attack airfoil
I
LS(1)-0413has reachedc and upper surfaceseparationextends from about
I Imax
x/c : 0.65 to the trai_ingedge. The LS(I)-0413Mod airfoil is fully stalled
L
at this angle of attack and separationextends from about x/c = 0,25 to the
C

_ trailingedge• This differencein behaviornear stall is attributedto the

!_i absenceof the reducedpressure-gradientnear the airfoilmid-chordfor the

!.
,, modifiedairfoil• (See fig• lO(b))• This reduced pressure-gradientretards

the rapid forwardmovementof upper-surfaceseparationat high angles of

_ attack.

The sectioncharacteristicsfor both airfoils are compared i=-figure8

for Reynoldsnumbersof 2.0 x lO6, 4.0 x lO6, and 6.0 x lO6. For a Reynolds

number of 2.0 x lO6 (fig. 8(a)) and angles of attack from about 40 to 13°

i_ the modifiedairfoilgeneratesmore lift and less drag comparg__tothe

I originalairfoil. This result is attributedto less upper-surfaceseparation
ii i
_ for the modified airfoilwith the reducedpressuregradient. Thus, the

: lift-curveis more linear at high angles of attack compared to the lift-curve

for the originalshape• An improvementin maximum lift-dragratio of about

L.

12 percentis indicatedfor the modified airfoil. However, the angle of

attack for maximum lift was reducedabout 30 and hence c decreasedabout

Imax
I 0.04 for the modifiedairfoil• The stall characteristicsfor both airfoils
• were similar. At the higher Reynolds numbers (figs.8(b) and 8(c)) the

capabilityfor improvementin performanceover that obtainedat R = 2.0 x 106

9
 _"
L'T
for the modified airfoilwas not availableand thereforenone occurred. How-

ever, the same earlierairfoil stall and decrease in cI were exhibited

_i m_LX

i
by the modifiedairfo_las was previouslynoted at the lower Reynoldsnumber. _

The absence of the improvementin performancefor the modifiedairfoil at the

I_
i
higher Reynoldsnumbers is not surprising,since the turbulentboundary-layer

i_ thicknes_i._decreasedat the highe_ Reynoldsnumbersand thereforecan with-

standCncreasedpressuregradientsbefore separating. Figure II compares the

values of Clmax for both airfoilsfor a Reynoldsnumber range from about

t_ 2.0 x lO6 to g.o x lO6. The modified airfoilexhibits a loss in Clmax of

_ about 0.04 throughoutthe Reynoldsnumber range. The less negative values

i_i
_ of pitching-momentcoefficientsfor the LS(1)-0413Mod airfoilcomparedto

the LS(1)-0413airfoil (fig.8) are associatedwith the reductionin aft

_ camber which resultedfrom alteringthe upper-surfacepressuregradient for
r

_ the modified airfoiland which is illustratedby the comparisonof camber

.,:
I:
_ distributionsof figure 2.
li:i
:/_' CONCLUDINGREMARKS
!

,. Low-speedwind-tunneltests have been conductedin the Langley low-

turbulencepressuretunnel to evaluatethe effectson performanceof modify-

ing a 13,percent-thick
airfoil. The airfoilcontourwas altered to reduce

the aft upper-surfacepressuregradientand hence delay boundary-layer

separationat typicalclimb lift coefficientsfor light general aviation

airplanes. The testswere conductedover a Reynolds number range from

about 1.0 x 106 to g.o x lO6.

The resultsshow that the modificationto the airfoilcontour increased

the maximum lift-dragratio about 12 percent at a Reynoldsnumber of 2.0 x 106

I0
il.

_:: but that essentiallyno improvementwas obtainedat Reynolds numbersof

)' 4.0 x 106 and 6.0 x 106. Also, the resultsshow that the modificationto the

C airfoildecreasedthe maximum lift coefficientabout 0.04 throughoutthe

, Reynoldsnumber range tested. The theoreticalviscous analysismethod employed

proved to be _ valuabletool in predictingthe airfoil pressuredistributions

and boundary-layerseparationpoints.
_-

REFERENCES

I. McGhee, Robert J.; and Beasley,William D.. Effectsof Thicknesson the

AerodynamicCharacteristicsof an InitialLow-SpeedFamily of Airfoils

.... for GeneralAviation Applications. NASA TM X-72843, 1976.

2. Bauer, F.; Garabedian,P.; Korn, D.; and Jameson,A.: SupercriticalWing

Section II. LectureNotes in Economicsand MathematicalSystems, __

iil M. Beckmannand H. P. Lunzi, eds., Springer-Verlag,

c. 1975.

iii_ 3. Von Doenhoff-,

Albert E.; and •Abbott,FrankT., Jr.: The LangleyTwo-
_ DimensionalLow-TurbulencePressureTunnel. NACA TN 1283, 1947.

4. Braslow,Albert L.; and Knox, Eugene C.: SimplifiedMethod for Determi-

nation of Critical Height of DistributedRoughnessParticlesfor

Boundary-LayerTransitionat Mach NumbersFrom 0 to 5. NACA TN 4363,

1958.

5._ Pankhurst,R. C.; and Holder,D. W.: Wind Tunnel Technique. Sir Isaac

• Pitman and Sons, Ltd., London, 1965.

6. Pope,Alan; and Harper, John J.: Low-SpeedWind-TunnelTesting. John

Wiley and Sons, Inc., New York, 1966.

II
 _.,_ ....
,_I,_,T. _ ......................
!i

_i
I TABLE I,- LS(I)-0413AIRFOIL COORDINATES

_ x/c z/c, upper z/c, lower

[

0.00000 O.OOOO0 O.OOOO0

! .00200 ,01040 -.00500
.00500 .01590 -.00940
lil .01250
.02500 ,02420
.03320 -.01450
-_OlglO
! .03750 .03970 -.02230
.05000 .04480 -.02500
.07500 .05260 -.02940
.I0000 .05860 -.03280

I;ii .12500
.15000 .06350
.06760 -.03560
-.03790

. oooo .o __
oo -.o4 4o
.22500 .07650 -.04270
i i .25000
_27500
.o oo
.07860
.08030
-.04370
-.04430
_ .30000 .08180 -.04480
!; .32500 .08300 -.04510 i
_ .35000 .08380 -.04520
I: 37500 08430 .04500
'i .40000 .08460 -.04470
!/ .42500 .08460 -.04420
!/ .45000 .08440 -.04350
.... 47500 .08380 -.04260
.50000 .08290 -.04140
i .52500 .08170 -.03990
.55000 .08020 -.03810
I_!. .57500 .07830 -.03590
);) .60000 .07610 -.03330
.62500 .07330 -.03050
.65000 .07020 -.02740
.67500 .06670 -.02420
.70000 .06290 -.02100
.72500 .05870 -.01770
.75000 .05420 -.01440
.77500 .04950 -.Oll3O
.80000 .04450 -.00830
.82500 .03930 -.00570
.85000 .03400 -.00350
.87500 .02840 -.OOl8O
.90000 .02270 -.00080
.92500 .01690 -.00060
.95000 .OllO0 -.00130
.97500 .00480 -.00340
l.O0000 -.OOl60 -.00710

12
'"4

i,i- TABLE II.- LS(I-).=.(2_/2LM[D_AIRFOIL

COORDINATES

:_, x/c z/c, upper z/c,lower '

I i
L.

O.00000 0.00000 0.00000

•00200 .O1040 -.00500
,,i_ .00500 .O1590 -.00940
i: .O1250 " .02440 -.01450
!:' ,02500 ,03420 -,O1910
' .03750 .04180 -•02280
.05000 .04760 -.02550
E_ .07500 .05640 -.02990
I_ •lO000 .06290 -•03330
'I .12500 .06800 -•03600
•15000 .07220 -.03820
: .1.7500 .07560 -.04000
•20000 .07830 .. -.04150
i.. 22500 08050 •04270
,_. .25000 .08220 -.04360
;" .27500 08350 .-04430
•30000 .08440 -.04480
I; .32500 .08490 -.04,510
_ .35000 .08500 -.04520
_:: .37500 .08470 -.04500
•40000 .08410 -.04470
•42500 .08320 -.04420
'i. .45000 .08200 -.04350
,_ .47500 .08050 -.04260
": .50000 .07870 -.04150
•52500 .07660 -.04010
•55000 .07420 -.03850
_:,. .57500 .07150 -.03660
[ •60000 .06850 -.03440
•62500 .06530 -.03190
•65000 .06180 -.02910
•67500 .05810 -.02620
•70000 .05420 -.02320
•72500 .05010 -.02020
.... .75000 .04580 -.O1720
'... .77500 .04140 -.01420
•80000 .03680 -.O1120
•82500 .03210 ..00840
. • 85000 • 02730 -, 00590
• 87500 ,02240 -, 00390
•90000 ,O1740 -,00260

i.-i.'
• ••95000
92500 ,O1230
.00700 -, 00210
-.00280
' ,97500 ,OO150 -,00510
1•00000 . 00430 . 00940

' r

L
" _'; ...."q--:_' '_ ;' _';_,
,_,_g NOT F!I,I_L.:'*
Figure3.- Chordwlsedistributionof slopes.
 (a) R = 2.0x I06.

Figure4.- Theoreticalchordwisepressuredistributionsfor LS(I)-0413and LS(!)-04!3Mod

i.. airfoilsat M = O.15.

i ,
(b) R = 4.0 x 106;q. = I. 2.

Figure4.- Concluded.
 I i,
_ /-Tunnel side walls

" Diam.= 1.67c _ ' --

i_ C_-_fi i l i i i i lllliZ,' lll l i t l i i / l ');_l _........ L-L_ _ "---_-'_-- _-, %_%-_ _. %

,_:_.. _ L_
A -I ___
A I.so:
_ Airflow ,

End view ,section A-A

Figure5.- Airfoil modelmountedin windtunnel, c : 61cm (24in.).

 .189c
(typ.)

i Figure 6.- _,_ke survey rake, c -- 61cm 124in.),

J
 (al I,_ - 2.0 x I0 6.

' Fi(I.)e 9. I-Um:l ol k.vnohls n.mhor on chordwis_ pres<,urP distrihulion_ _or l.S(I)-O/II3 Mod
,-i ,lirfoil for t,1 (). 15 ,ll'd r,),lHIiI;i,_,, i)ll ICollirr(,{l _.yHiholg lov,,or <,,)rfa{p).
_k ..
,..I
(a)R = 2,0 x 106, Concluded.

Fig.re 9.- Continued.

i.... Ih) R = 4.0 x 106.

i ifl_lrr 4.- (}l)ntil_llerl

IlL '
 , ey_-'m'W_._ '" - "_T ¸ - _

!,

,!: - 1LI C[ ct cd cra .

- ............................ o 12.2 1.701 .0200 -.082[_
i;_ -t3 [] 13.7 1.796 ,0250 -.0731
_ 0 15.1 1,872 .03u_2 -.0699
,_ Z_ 16.2 1.836 -,0710
t -12 '
i.
(c}R = 6.0xI06,

Iiqure 9.-- Continued.

t,i

6
[' -3.6 .....
r Airfoil
i::- o LS(I) - 0413
-5.2 t:l LS(I)-0413 Mod
i •

• -2.8 ,_ i - 5

-2.4 1 -4 _
! _ _ "_

t,
'i) -2.0 ___ ..... _ - 5 l "2
!' -I.6 _ -2 -
_.; , _ .... 0 .2..
x/c
,:SI °

i..
i -.8 _
[.,. ,.,

i.r "_k,,_k_

0 ,

.4 _ "-"4_ _....x_---4_'-" " - :'-----(_---4__.<L._{_._

_

. I _-
I

, V7
0 .I .2 .3 .4 .5 .6 .7 .8 .9 1.0
' x/c
,.. (b) a :_ I0°; R = 2.0 x I06.

Figure I0.- Continued.

_- -I0 Airfoil
o LS(I)-0413
-9 [] LS(I)-0413 Mod

_ 1
-8