Document
f
RESEARCH MEMORA~
t Kenneth Margolis Langley Idem o r ial A e r onautic z l Labor at ory Langieg Field, Va.
NATIONAL ADVISORY COMMTTEE
F O R AERONAUTICS
I l'?ACA RM No. L';7fIE .
An a n a l @ . a has b w n made of a"irf'oil data taken on several IiACA l&eriee propeiler a i r f o i l s from test6 of F-incfi-chord modelo irr the Langley 2!Liach hfgh-speed tuonel and l2-inchrchard m o d e l s in the.La#ey &foot higiGepeed t;umel, . T h l s analysis has shown tbat%he sambined e f f e c t s of Reynolds mxtiber changes and vertatioae in a i r f o i l characterietics resultiw from differences i n models and tunnels are such that when >inch- chord and l2-inch-chord data are applied 60 full-scale propeller design a t or near the design condition, differences of l e s s than 1 percent i n efficiency w i l l . be involved.
The design of j?reaent-day propellera is u s u a l 3 ~ Sa138d 3pon data obtained under conditions of scale which difgor frou! those of o p m t i m .
These propellers a r e made up t o a great degree of high-speed a i r f o i l 3ectlon8, data for which a r e Dbttsined f'rm teerts of models of 2- t o I n addition, met of the t e s t s of m d - e l propellers us- >inch chord.
WACA l h e r i e s airfoil sections heve been conducted on blades of this same width.
The questionthsrefca-e has arisen as t o the v a l i d i t y of a p p l y i q these test data d i r e c t u t o larger m a l e rlesigu.
I n order t o provide a t least E qualitative answer to these
I -
n r , a n + 9 A*- yLrvuUIV;4), ~ i i -lpIcj h a t t v a i&e or" 13me irate zvaiiable on A several NACA &series a i r f o i l s of both 5- and l.2-inch chord.
comparioon of data from K r and E-inch-chord a i r f o i l s has a d d i t i o n a l sigaificance because a 12-inch chord is rspresentative of blade widths ccmmonly used on full-scale propellers.
&+ 2 NACA RM No. L7Iil2
APPARATUS A ~ V D ‘METHODS
The t e s t s were mde iii the Langley %foot high-speed tunnel and i n the Langley 24-inch high-speed tunnel. The Langley &foot high-speed tunnel i s a closed-thrcat single-return tunnel and a t the timc of these t e s t s the speed w a 8 continuously controllable up t o a Mach number of approximately 0.70. The Langley 24-inch hieh- speed tunnel is a nonreturn induction-type tunnel with the speed continuously controllable t o a Mach nuuber of approximately 0.30 for a F i n c h , l%percont-thick a i r f oj 1, Bot,h tunnels hav\.
degrees of turbulence which a r e sniall though s l i g h t l y higher than that of f r e e a i r . I n both tunnels the mddals completely spanned the jet; thus, the r e s u l t s a r e e m e n t i a l l y twdimenaional.
The chord of the models tested i n the Langley &foot; high-speed tunnel was 12 inches; t h a t of the models tested In t h e Langley 2 b i n c h high-speed tunnel was 5 inches. The a i r f o i l s tested were tho follow- ing NACA & s e r i e s sections : 36-209, 1 - 1 7 , lG509, 16,515, 16-709, and 16-715, t h a t i a , sectYons havinc thicknem r a t i o s of 0.09 and 0.15 and having design l i f t coefficients of 0.2, 0.5, and 0 . 7 .
The data Gbtainbd wesc l i f t , drag, and pitching moment. The data on the Finch-chord a i r f o i l s were obtained by neans of force measurement& i n the Lzangley &-inch tunnel .I For the 12-inch-ohord a i r f o j l s the l i f t and moaent data were obtained fram pressure-, distribution measurements and the drag data were obtained by mean6 The average variation of Roynold&nnumbcr wlth Mach of wake s u p e y s .
nlxmber f o r the a i r f o i l s a s tested i s shown i n figure 1.
SYia3OLS , ..
M Mach number R Reynolds number U angle of attack, degreos section l i f t coefficient c Z section q mr t cr-chord p i t c hing-moment c oe f f i c i erit %/4 dc2/6a lift-curve slope C d section drag coefficient NACA RM No. LITH32 Ql0 = t a n - 1 v, mnn.
. .
.- .
. .
V f & d velocity . .
zotational speed n . , .
D propeller diameter X radius mtio ai .induced s q l e of attack .y = tan - 1 9 cl The chengss which occ~m in airfoil characteristics such as dmg and maxirm;lo f i f c c m f f i c i e n t with changes i n t h e valde of Rejmolde nmber are dlrectlg- connected wlth the action of. the bwq€er3 layer on the f l c w ov9r tLe airI"ol1.
A discussion of the mechanics of these flow chmges is contained i n refersnce 1.
The variations of l i f t coefffcient w i t h angle D f attack f o r the a i r f o i l s tested a r s cmparcd In f p p r e 2. Die t o the f a c t that the teste w e m made with different sized models of the same a i r f o i l sections and because the mcrdels were t e s t e d in different t u n n e l s , variations In the data e r e t o be e q e c t e d as a r e e u l t of iridividual m o d e l irregularities, f a i l u r e t o exactly duplicate model e f i n a e n t , and slightly dffferent wall effects.
For these reasons, oal;r the shape and character of the curve^ i n ffgwe 2 should be ccanpared.
The most notjceable effects of difference i n Reynolds number a r e a l i g h t changes i n l i f t - c w v e slope and differences i n the character of the brezk i n the l i f t : m e correspcndjng t o t h e end of the Low-dra~ region.
These effects are more marked f o r tlze thicker airfoils.
The v ~ r i a t i c n ::lth W c h iluca'ber or^ tne lift--curve slope, tnken in the desi* l i f t mnge, f o r the a i r f o i l s of different 3 - i : ~ 13 presented i~ figure 3.
Tho differences i n slope are generhlly mall althou& marked differences occur for tihe NACA l6-2Cg and 1 6 7 1 ' j air- foils above a Fich num3er of 0.60.
4 NACA FM No. L7H12 If the moment coefficients of two a i r f o i l s a r e cmpared a t a l i f t coefficient, an indication of differences i n given value of load distribution is obtained. When t h i s procedure is applied t o data f o r t w o geometrically similar a i r f o i l s t e s t e d under different conditions of scale, an indication of fundamental-flow changes is obtained. Therefore, the variation of pitching-ent coefficient w i t h l i f t coefficient for several a i r f o i l s is presented i n figure 4.
Analysis of these data indicates that the fundamental-flow changes, which may be due t o scale e f f e c t s o r model i r r e g u l a r i t i e s , a r e small.
The most noticeable differences occw a t l i f t coefficienta corre- sponding to the end of the region of low drag. These differences indicate that the l i f t coefficient a t which t r a n s i t i o n occurs decreases a8 the Reynolds number is increased, as has been pointed cut i n reference 1. This effeot is apparent f o r the thicker a i r f o i l s . The differtinces between t h o data f o r the 5- and E-inch NACA 16-509 and 16-5Pj a i r f o i l s suggest individual model i r r e g u l a r i t i e s .
Because boundary-lager changes a r e involved it i e t o be expected that with changes i n Xeynol.ds number t.ho drag characteristics w i l l be affected t o a greater d e g e o than t h e l i f t and moment character- i s t i c s . The variatioiis of drag cocff?:cient w i t h 1 i f t . c o e f f i c i e n t f o r f o u r a i r f o i l s , tlis HAC-! 1&20g, 16-215, 16-709, and 16--r(13, a t two valum of Mach number a r e S!I.GTE ir! figure 3.
In figure 6, curves cl“ t h : : v 9 r 7 it ions of skin-frict9on drag coefficient wit11 R e y m l d s xmbcr 707 3 f l a t p l a t e are presented.
These curves a r e baEed e-1 t ~ e l a ~ ~ . ~ ~ ; i ? r and turbulent laws f o r skin- drag (reference 2) and snow ‘now the drag coefficiont f r i c t i o n decreases as the Reg’nolda nu-%%* i o lncrcasod, for a given r e g b e .
For the combined drag coefficient of both surfaces, . .
the laminar l a w i s .,the turbulent law is ( R ) The pcints plotted on figure 6 a r e the values of miiiimim drag These data a r e general-j.y T:!*,hin the coefflcient tzken from f i s r c 5 .
limit8 of the’ laninar and t-ubulent curves. It has be?-7 ?.-lp>:. a, i n reference 1, +,lint wlAgh a i r f o i l surfaces give drag co::”-~.:bLcl;~:: w e l l above the tu&uien$ skin-friction curve. The r e l a t i v e pcb25iiun of the .
N data f o r a given a i r f a i l betwsen fhe lmdnar and t u r b u l u n t ~ c ~ e s depends on scale e f f e c t s and factors such as pregkure gradients and surfat% r'bizghne'ss -nhl&it agfwt: $oun@rq-hqer tramttion-.
- 6 -4 ' 7 , ' I , * , . . , a * . . ; - .
^.. .
&c&ieti in Ti@'e' 6 I& the 'vvarAat&a of' minitmum drag coefflcienk fop th& %?CA L L A 9 81iLlffii Ae s e p d e d Sln.-re??erqce 3. Theee data we:-e ob'talule& Pn one %tunnel under c'onditicma of l o w turbulence.
The similarity Wetxe&'the treniis of tho variation with'Repclds number shown by the dsta-'talren frm reference 3 and those reported herein, indicate the% t h e rri%her large diffarence i n the drag as o b t a i k d 'fn thg LaI&ey &foo$ high-speed tunnel End the Langley 2 b i n c h high-qeed tunnel is a c t u a l 3 a $scale effect, find is not caused by differences i n tunnel t e s t t e c h n i q u e s or model suzface conGition.
The differences in slope of the 'vaz-iation of drag coefficient with Reynolds number for thcl Finch-chord a i r f o i l s a s cmpared with the-lQO9 data frm reference 3 are ascribsd t o c o q x e s s i h i l i t y e f f e c t s which result 'in ghe ctee_oerClng of t h e prezsure-recovery w a d i e n i s &er the sfifSS*h. X i % : I s to be expected t h a t these e f f e c t s will be%&% prmounce'd f o r th'?,ck airfoils as is ilLwfsoted'by the relatively s l i g h t v a r i a t i w of drag coefficieat w i % h Riynolde riumber f o r the NACA 1 - 1 3 and 16-715. The adverse e f f e c t s of increassd rgcovery p h d i e n t s i n critioalBe,-nolds number ranges e r e further i U d s t r a t e d by t h e Increases i n value of ?Tag coefficient a8 the thickness and caberbare increzsed.
In considering the application of these data t o Eropeller design, it should be pointed oizt t h a t changes i n drag coefficient of the order of those f o m d i n f i g m e 6 w i l l have ordy a small e f f e c t on propeller conditions the perfoAmance a t design conditions.'becauee a t these l i f t - d r a g r a t i o is high and, aislce.the elemental efficiency is
t h e changes in'effzciency will be af smil oraer.. -
Uind-t-qnnqlmodeis e r e c-full;. prepared and nqintained; whereas i n a c t u a l cperation manufkctming i r r e g u l a r i t i e s and surface roughoese w i l l probskly produce values of drag coefficient sloaer t o those Thersfore, altho@ the >inch- obtained on the '/-Inch-chord models.
chorcl c l ~ t e do c o t r z p c r ; e n t t r - i c XiGltiyiiS ~ . f Ecaie, tney m y be safely used t o est-te propeller performance. For example, t h e difserenceein efficiency canputed by t h e above r e l a t i o n based upon t k s differences i n drag coefficient for the 5 and 12-inch-chord airfoils w i l l be of the order of 0.6 percent if it is assmed that for a t y p i c a l propeller t h e reqresentative sections over t h e &portant L 6 IQACA RM No. L7Ef.2 0 . 0 9 o r less and are area ofthc h l a d e have thickness ratloo of cambered to give a design lift coefficient of 0.5. This difference of efficiency will hold for a range of lift coefficient f O . l from I f thlcker or lower design at trahnes of Mach number up to 0.50.
cembered aections are used or the blade is operated away f'ramthe design condition, differences greater than 1 percent may be expected. If, however, operational drag coefficients are higher than those presented fir the l&incbchord a i r f o i l s , the differences I in efficiency will be Bmaller. Moreover, use of the 5inch-chord data gives a more conservative estimate of efficiency.
CONCLUSION Differences of less than 1 peroent in propeller efficiency .1.
at or near the design condition will be involved in applylng data from Finch-chord and lsinch-chord airfoil testa to full-scale propeller des Ign.
Langley Memorial Aeronautical Laboretory Natlonal Adviaory Comaittee for Aeronautico Langley Field, Va., August 21, 1947 I?EEEENrnS 1. Abbott, Ira H . , von boenhoff, Albert E . , and Stiqers, Louis s., Jr.: W C A A m No. L5C05, 1945.
Ehmmxy of Airfoil Data.
and Thompson, Milton J . : Fluid Mechanics.
2 . Dodge, Russell A., First e d . , McGraw-Hill Book Co., Inc., 1937, pp. 322-325.
Preliminary Report on LamimFloyq Airfoils 3. Jacobs, Eastman I ? . : and New MethoCts Adopted for'Airfoi1and bunday-hyer Investigations. NACA ACR, June 1939 .
NACA RM No. LEI12 . 7 .
NACA R M No. L7H12 .
NACA RM No. L7H12 9
./ -2 -3 4 .5 b -7
Mach number,/W
f34re 3. - Vir/'cYf/bn of //Y+cut-ve slope w/'+h
Much / 7 u r n k r fir 3 1 % N A C A /6-.ser-/;es cy/>fm/s o f two di'fferenf chord /en+k.
NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS NACA RM No. LEI12
I I I r 1 I I 1 1 I I I
/- / - . / - 4 - c- _/ -- c-
- --
.08
J .3 .4 .5 4
.2
Much r,umber, /M
NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS f / g ure 3. - Conduo'ed.
NACA RM No. L7Hl2 11 . I
\ cu
I' I.
12 NACA RM No. L7H12 .
v) c I I
d-
I NACA RM No. L7H12 NACA RM No. L7H12 Q
E
NACA RM No. L7H12 .