c sa:: FiLE - NASA · c sa:: F"iLE COpy TE CiINI CAL NOTES NATIONAL ADVISORY COMMITTEE FOR...
Transcript of c sa:: FiLE - NASA · c sa:: F"iLE COpy TE CiINI CAL NOTES NATIONAL ADVISORY COMMITTEE FOR...
c sa:: F"iLE COpy
TE CiINI CAL NOTES
NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS
No. 417
WIND -TUNNEL TESTS OF A HALL HIGH- LIFT WING
By Fred E. Weick and Robe r t Sanders Langley Eemorial Aeronautical Laboratory
Washingt on Hay , 1932
https://ntrs.nasa.gov/search.jsp?R=19930081213 2020-04-12T04:40:44+00:00Z
. NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS
. " '
TEC}INI CAL NOTE NO, 417
WIND-TUNNEL· ~ESTS OF ~ HALL HIGE-LIFT WING
By Fred E. Weick and Robert Sanders .:' w ••
SU M"AARY
Wind-tunnel tests have been made to find the lift, drag, and center-of-pressure characteristics of a Hall hiGh-lift wing mode l. The Hall wing is essentially a split-flap airfoil with an internal air pas sag e. Air ent~rs the passage throug h an opening in the lower surface somewhat back of and parallel to the leading edga, and flows out throug~ an o~ening made by deflecting the r ear portion of t ~ e unaer s u rface downward as a flap. For ordinary fl i ght condi tions the front opening and the rear flap can be closed, providing in effect a conventional airfoil (the Cl ark Y in this case). The tests were made ' w~th various flap s ettings and with the ~ntrarice to the pasiage both open an~ clo s ed. ~he hi ghes~ lift coefficient found, CL = 2.08, wa s ob ta ined with the passage closed.
IN'TRODUCTION
The present Hall high-lift wing is She result of a development which started with a s tudy by Theodore P. Hall of the possibilities of converting the upper and lower surfaces of the conventional wing into sepa~ate airfoils, forming in effect a biplane combination wi:h a s mall gap • . (Re~ erence 1.) rhe development was continued with the construction of an airplane incorporating the wing shown in Figure 1, which was entered in the Guggenh eim safety competition, (Referen ce 2.) The flaps on this airplane ~ere later made automatic in operation. More recently, further wind-tunnel tests have been made on a small model of the Hall wing in its latest form (similar to that . 1"n fig. 2). In both the wind-tunnel and flight tests substantially higher lift coefficients were obtained with the hi Eh-lift arrangements than with conventional wings, but the a~proximate nature of the fl~ght . tests and the low scale of the wind-tunnel tests made further experiments desirable.
2 N.A.C.A. Technical Note No. 417
The present tests have been made as part of a series on various high-l ift devices in the N . A.C.A. 7 by 10 foot wind tunnel. The latest form of Hall hiBh-lift wing was t es ted ~ith the flap set at various angles , and the entra n ce to the internal passage both open and closed .
APPARATUS AND ~ETliODS
The Hall wing model, which was constructed of laminated mahogany (figs. 2 and 3 ) had a chord of 10 inches and a span of 60 inches. Nar r ow ribs in the internal passage ~ ere sp a ced about 3 inches apart . In the center of the wing a 4- inch cell was made solid for mounting. The rear flap ~as s upported on hing es ahead of its leading edge. Thus when deflected there was a gap bet ween the flap and t~e lo we r surface of the wing, which is not the case with the wing on the Cunningham- Hall airplane . (Fi g . 1 . ) In the model the flaps were hing ed at eight points along the span and provided with small quadrants having holes drilled 10 0 apart for deflections from 0 0 to 50 0 • For one test the slope of the upper surface of the passage just above the front portion of the flap was increased from 26 0 t o 45 0 by me e ns of Plasticine . (Shoun in fig. 3 and in an insert of fig. 2 . ) The front entrance to the passa g e was closed by means of a cover plate , making the win g in effect a conventional airfoil with a split flap. The cover plate was used o n the model to replace a forward vane or valve; with the plate removed the inside form of the p assag e simulated that with the vane displaced 30 0 upward.
The 7 by 10 foot wind tunnel is o f the open-j"et type and is described in detail together with the balan c es and standa rd test procedure in reference 3 . The Hal l wing model , whi ch lacked rigidity be c ause of its hollow construction, wa s supported by a fine wire at each wing tip in addition to the usual center support . The tests were made at an air s p eed of 80 mi les per hour, correspo ~ding to a Reynolds Number of 609,000.
RESULTS AND DISCUSSI ON
T b. e val u e s 0 f C l' "CD, and c . p • are p lot ted a g a ins t ang le of attack for t he various flap settings with the passage op en in Fi Gure 4 and with t h e p assage closed in Figure
----------------------~--~----~-~-------------------~
N.A'.C.A. ' Technical Note , No. , 417 3
5. No " tunnel-wall ' correcti~n was made. !' . • ,
Lifi- - From the cross p lots of the maximum lift coeffi c i en t s a ga inst flap setting given in Figure 6 , it c a n be s een t hat the highest value of CL max with the passag e op e n is 2.05 and occurs with a flap ang le of ~bout 4 5 0 • Th 'is ' i-s a '60 per cent increase over the maximum lift c o c ffi 61 ~nt of ' l.28 obtained with the basic Clark Y ~ w1ng . Wit ll 't !:J. e passage closed the highest value of 0L "' is 2 . 08 with a flap setting of about 480. max
Changi ng the slope of the upper surface of the pass a Ge j u s t a b ove the front portion of the flap fro m 26 0 to 45 0 ma de no ap p reciable difference in the characteristics o f t h e wing.
It is interesting to cOlnpare the highest value of C obt a ined with the Ha ll for m 'of flap (2.08) with L max th e v a lue obtained in tests of a conventional tr a iling-edg e f lap on a Clark Y airfoil. , (Reference 4.) The tests wh ich we re ma d e at the same air s p eed and with models of the same chord , gave a value of CL max of 1 . 95 with the plain f la~
The c h or d of the conventional flap , was 30 per cent of the wi ng ch ord, and the flap was depressed 45° . , Th~ actual cho r d o f the Eall flap was 34 per cent of t h e wing chord b~ t th e over-all distance from the hinge axis to the tr a i lill~ edge was 41 per cent of the wing chord.
Dr~. - The minimum drag coefficient of, the Hall wing mo d el u ith the cover plate in place and the flap neutral wa s 0.01 61, as compared with the value 0 . 0150 obtained und er th e same test conditions with a solid Clark Y airfoil. Tln:!,s th e ext ra drag due to the flap supp ort sand to the s light imper f ections of section accompanying the split f orm o f fl ap was approximately 7 per cent.
A point of interest is the high drag at maximum li f t wi t h th e flap do wn. At the peak of the lift curve the va lu e of LID ~ ith the flap in the position g iving the hi ghe s t maximu m lift was less than one-half that of t h e plain wi ng a t the peak of its lift curve , making possible a mu ch ste ep e r ap p roach for a landing .
Ce~i~~_Qf_~~~~~~. - Th e ce n t e r of pressure with th e f l a p d own 20 0 or more u as b a ck a b ou t 10 per cent of the ch ord from its location with the flap up. This was appro~
4 N.A.C.A. Technical Note . No. 417
i mat ely true for ~ny angle of attack above 0° phether the passage th r ough the wing was open or c l osed.
CONCLUSIONS
1. The highest lift coefficient obtained with the model of the Hall h i g~-lift wing tested . waS of the order of that obtained with a conventional trailing-edge flap.
2. S l ightly higher values of the max i mum lift coefficient ~ere obtained with the passage through the wing closed than wi th it open.
Langley Memorial Aeronautical Laboratory , National Advisory Committee for Aeronautics,
Langley Field, Va., March 31 , 1932 .
REFERENCES
1 . Hall, ~heodore P .: Hall Conve r tible Air fo il. Thesis . . t ~ as sa chuset ts Ins t itute of Te chnology (Cambridge),
1928.
2. Anon.: Final ReIort , The Daniel Guggenheim In ternational Safe Aircraft C~mpetition , January 31, 1 930 .
3 . Harris , Thomas A. : The 7 by 10 Foot Wind Tun n el of the N . A. C. A . T.R. No . 412, N. A . C. A., 1931 .
4 . Weick, Fred E . , and Shortal, Joseph A.: The Effect of Mult iple Fixed Slots and a Trailing-Edge Flap on the Lift and Drag of a Clark Y Airfoil . T .R. No . 427, N. A . C . A ., 19 32 .
I l O(}fo c -1 ~--== ==-- I I - ----0.-====== L-- l !% c-----J
~I'I .! - . --------- -- - ----:~- I
~ I ------- ~ ------ ' . - _1. _- -_ . _---- _. __ -- - ~
--10;0 c ,!" 1 5~ c ---~ .~' - -r------------·· • I I . ~ I 1 : . ~ 3i o
r -8 ~ . ~ -"", :J • vfo C >/ ~ ~
~ ......... I . :;g
c
~" ....... Section is Clark Y wHh f l ap s neutral
"'" '-y
Fig.l Section of wing used on Cunningham- Hall airpl ane .
~ . il> o il>
f-3 (;) o p-' ::; f-'. o III r-'
~ o cT (!)
'-' ",. , o fi'> r-' -,J
I->j f-"
C1tl
t--'
1/ . ~ 1 ~oo ------ ;< - G . tj;:)
\ .12511 ~ I y! 26 0 1 2 . 37 11 -1.
I _ ---,,.- 1 2~ 11 ~ I
:<--- 1. 3211
_1___ " ~ • 0 _~ l , I f 1.78 '1 I ::--'- ~-- -- - ~I .51 r 5 92" Flap h i nge./'kr:-68'. ' - "--'-'-' =-~
• ! t- ---- 3 ~ 1I _=_ 1 c . . ;r -- --------1
",.' 1
C-~upper !Lower : I ~s id
ISta- lo r di- 'or d i - Ior ditio!?-~ nat e \ nat4~ate II ~c9-+I nch ! I nch lI nch 0 . 0001 0 . 35010 . 350
. 1251 . 545: .1931
. 2501 . 650 I .147 i
. 500, . 790 i . 093 ! \ .750! . 885 \ . 063 t 11. 000 1 . 960 , . 042 1
\1. 500 1. 069 I . 0151 12 . 0001 1.136 i . 003
i ~Jpper '\ Lower I I n side I I
Sta- lo r di- ordi- i or d i- ~ Fl an ! tion \nate \n a t e \n a t e :Sta-!Or di- l 3 . 000! 1.170'0 . 000!0 .~ionlnate I ; . 0001 1.1;0 \ . 000 I . 92 I I nch! I nch i' ;:) . 0001 1 . 0 ;:)2 . 000\ . 84 ,0 . 25jO .16, 6 . 000 I . 91 5 I . 000 I • 72 , . 50 . 22 I 7 . 000 • 735 II' . 000 I . 58 I' . 75\ . 26 'I
8 • 000 • 522 . 000 \ 1. 00 I • 29 1 I' 9 . 000 • 280 I . 000 . 2 .00. 21 'I
9 . 5001 .1491 . 0001 \3 . 001 .13 0 . 000 . 01 2 I . 000 I :3 . 40, . 01 ,
L. E . r adi us = 0 .15 i nch
, , , , ,
, , , , ,
, , , \ ,
, , , , \ , , , , \
' ,~
F i g,2 Section of Hal l high-lift wing model . Dotted lines indicate flap set at 500 and c losed passage .
~ · tJ> o · lJ> · 1-3 Co
g. ;:! r" ()
~ h ' ~I
o c+ o ~ o .p. r' :'l
b;j r"
O'll . 1:\)
N.A.C.A. Technical Not e No. 417 Fig. 3
. ~ .... ~
~ IH .... M
1: ~ ....
• .s:: M
i IH 0
M CD '0 0 ::e
~ . ItO ....
IZt
•
· 1'>1
H
S o H 'H
()
.p s:: (l) ()
H (l) p~
]'lap setting
-- - _ .+-.-l----+-
1. 6 1---+---+
1. 4 ---+----+-
0.4
Fig. 4 30 ° -- -- ------v 40o---X
500- -- .-- -- - - +
~:~ -,-=-4 ~
~~T-
1---l---'~+----+·2 •
+
15 20 25 30 c , degr ee s
Fig . 4 Char acteristic s of 10 i n . by 60 in. Hall wing with passage ouen and f l ap at various angles uncorrected f or tunnel wall ef~ect.
N.A .C.A. Technical Hote No.417
o 00
'\7 --30 0
x- --400
0.2.:,......c:==!--
Flap setting 500 --- + 590 - - - - - - - - --f:::,
Fig. 5
I __ -i- - I . I 1 F -!.-- t r-- ..-0" __ -f-'-~--T-' -+----+----I
o I LJ-~:-r- 1_1 ___ --'-_---L--1-1 ~----'-----' -10 - 5 o 5 10 15 20 25 30
a., degree s
Fig.5 Characteristics of 10 in. by 60 in. Hall wing with passage closed and fl ap a t various angl e s uncorrected for tunnel wall
effect.
N .A.C. A. Tecl nical Note ~Jo . 417 Fig.6
0------ Passage open !J- -- - - - -- Passage closed
---~-T-'---- -l--'-T--' --~I----' 2 . 1J0 r---
1
-t--l-~~-· - ~+-r ~ +f -- -r-- -- -- 1--- - - - - - : f----t--+--I---
2.40
I--r-I---+ ._-- i - - -- -[--2.20 . Il--I---- --~---t--t -1---, --
1----+ _ _ 1 ____ 1 _ 1-__ _ __________ - e-- 1 ___ -L--.j' t- -j ---- -- -,~ - - I
I L 0:~ - 61 2 . 00 f----I--+I-- --1--~71--r----- 1--- --t--
CLmax . 1- ./ / , i -'--r' ---I V // I !
1,80-1 7:7-"- -~~I-~--- ---- --t-~~ / / t- +-+--
1// / I ~ / I I
1. 6C
l, W ! / ·-·-:---r-~-- -----~--- -+- -- +--I +--r- i - --I---t--t---+----- 1----1-- -L 1 I I I '
:':l_...1.-__ -_ -~t ____ -~~u_--~-:~ =-1----1 o 10 20 30 40 50 60
Flap angl e ,degr ee s
Fig.6 Variation of maxi mum lift coeff icien t with flay &~gl e .