NACA - General Airfoil Theory

,...— ...
.4
,, -. -i
.
.—,. —,
.,
...— ------- -. . -.. ..— —, .,

i.
TECHNICAL MlilMORANDUIUIS
,.
NATIONAL ADVISORY COMMIil?TE13FO# AERONAUTICS
# ‘
,!.
,-. %;
m“,
,- ,
.,.. G33:NI?JIIAL
AIRFOIL THEORY
,!, ,.”
., By H. G. Kiissner
,.
..J’-
Iiuftfahrtforschung
,,
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
vol. 17’; No. 11/12, December 10,, 1940

Verlag Ton R. Olde,pbour&’, Mii.nchen und Berlin
;:
>’
:.
J.
i
Copyright National Aeronautics and Space Administration (NASA)

Provided by IHS Licensee=Boeing/5910770006
No reproduction or networking permitted without license from IHS Not for Resale, 01/16/2015 11:45:54 MST
/: w.’
y,
$. NATIOitAL ADVISORY COMMITTEE 303 AERONAUTICS
““l!E”CHNIC”AL’”
MEMORANDUM No. 97”9-”””-”
GENERAL AIRFOIL THEORY*
By H. G. Kiissner
On the assumption of infinitely small disturbances

the author devekclps a generalized integral equation of
airfoil theory which is applicable to any motion and com-
pressi-ole flv.id. Successive specializations yield vari-
ous simpler inte$ral equations, such as Possiofs,
Birnbs.um~s, and i)randtl is integra,l equations, as well as
new ones for the wing of infinite span with pericdic down-
wash distribution and for the oscillating wing with high
aspect ratio. Lastly, several solutions and methods for
solving these integral equations are given.
.:. INTRODUCTION
There are a number of airfoil theories which hold

true in two or three dimensions, are stationary or nori-
stationary, and allow or disallow for the compressibility
of air. All these theories have one thing in common:
They are, strictly speaking, valid only for infinitely
small disturbances; hence the airfoil must be assumed as
infinitely thin “and with infinitely small deflections from
a regulating surface , the generating line of which is
parallel to the direction of flight. Then the regulating
surface itself:,can be approximately considered as the
place of the wing and the ar”ea,of discontinuity emanating
from its trailing edge. Up to the present time, a plane
has been commonly chosen as a regulating surface, but
this restriction is not necessary.
Following the temporary interest attaching to the

v“ortex theory, the introduction of Prandtlls >acce.le.r.?t.i...u
.—.—...
-
potential rnadq the ol,d,,~o,$eg$iql,t~eo.ryapplicable to air-”
foil theory. The particular advantage of this method over
,.
..”
* ‘lAilgemeine Tragf15chentheorie .II Luftfahrtforschung,
vol. 17, no. ‘ll/12,.December 10, 1940, pp. 370-78.
..
.
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

Provided by IHS :, Licensee=Boeing/5910770006
No reproduction or networking permitted without license from IHS Not for Resale, 01/16/2015 11:45:54 MST . -- —
2 NAGJV Techuical Mtirnbraridui’
No. 979
. .
the vortex method is that ‘the [email protected],S,~.~..$~~,$.Y

of air can
be taken into account. ..BY th.is,method the wing iS re-
placed by an arrangement of acoustic ‘Ir?i.a.tors.11 The
sole essential restriction of thzs theory consists in the
assumption of moderate fields of sound, that is, small
inter.ferenees. Then the classical wave equation
(1)
is ,applicable for the velocity potential and the sound.

pressure of a quiescent source distribution. Their solu-
tions have already been explored in all directions. Its
application to the moving airfoil is achieved with the
aid of the well-known Lorentz transformation, the sole
invariant,, trig the speed of radiation c, which, in the
P
case in question, is equal to the velocity of sound.
The setting-up of the integral equations of the air-

foil theory is a preliminary task, which is definitely
achieved by the subsequent expositions. But this pre-
liminary work alone accomplishes little without attacking
the purely mathematical main problein, namely, the solution
of these integral equations without entering into new
discussions every time regarding the method of derivation
and its physical significance.
2. THE VELOCITY POTENTIAL Ol? THE ELEMENTARY RADIATOR
The wave equation ❑ @ = O has, as is known, a very

simple solution for’s spherical wave that sprciads ,out ra-
dially at speed c from its source. The solution reads:
,,
(Do’*f t
(’-:) , ~ ‘2)
,“ .,
., where r denotesthe radius, t the time, and f an ~
arbitrary function. Such,a spherical wave is produced
by an elementary radiator of zero order, ,w~ich represents
a“ simple ‘point source. The velocity potential of radia-
tors of higher order fo3.lows from (2) ‘by partial deriva-
tion alongqny coordinate directions.
The airfoil which is to manifest a pressure differ-

ence between its two surfaces is best replaced by a su-
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

\
. NA,CA Technical Me.mor:an*dti-

No’: .9?9 3
perposition” of. radi~tors of the first order (so-called

doublets) whose axes “are normal to the air foil.” lf.
ith n
denoting the direction of the normals, the potential of
a doublet of the superposition iS
:(3)
The sound ,yressure of the. field of? sound is ,
(4)
hence it satisftes equations DXJ = O also because p-

constant for very small disturbances.
3. THE LORENTZ TRAUSFORMATION .
The arguments sofar have dealt with the radiator

at rest at infinit’y. j,To reach-the pressure field of the
radiator moving at constant speed V<c past the X
axis,, the Lorentz transformation is resorted to. With
the “pritie 1 indicating the transformed system, the
transformation formulas for a motion,aloqg, th,e negative
. x a“xis of the quiescefit reference system read (refer- “ * “
.. ence 1):
~.. = xt+’vt~ t-r ,+ v xt/c2

r= v2/c2
;Y= Y’; ~ = z’; t =
~ ‘5) --`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
On passing to the coordinate x~ = Xl+ vtf in the

‘entrained reference system and posing Mach’)s number
P = v/c, ‘equation (5) “gi’ves: ‘“
‘=Z%;YY
= ’;‘=z’;t=t’~+ .;+3 (6)
Acco”rdin& to (6) , the change to the moving reference

system ,entalls the following su%stitut iotis:
2 ~f2
r =—+y12+zt2 (7)
pz
1-

4 liACA Technical Memorandum No. 979
Inasmuch as the airfoil is to be situated on a regu-

lating surface with generating line parallel to flow direc-
tion, that is, along x direction, the Lorentz transforma-
tion does not alter the direction of the normals n.
Later, the knowledge of the sound pressure of a mov-

ing doublet is necessary. According to equations(2), (3)3
and (4), the sound pressure for the quiescent double%, the
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
sequence of the differentiations being changed, is:
(i- :)) (8)
Since f is an arbitrary
af
function, ~ can be
replaced by another arbitrary function ?; and constant
.. .. 1
factors such as Jl -$2 can then be omitted or added.
After completing the Lorentz transformation, equations
(7) and. (8) give for the rnovins doublet the pressure
.,
~a
y~.+
( x$
C(1 -’pa)
‘ -
a)(
p’)
~2+ ~2))
)
‘
p(x,y,z,t) =- — (9)
4m h
x’ + (1 - 82) (Y2 + “)
where the transformation prime can then be omitted.
4.
-* The Acceleration Potential
According to Prandtl (reference 2) a new approach to

the airfoil theory is afforded by proceeding from the
field of the acceleration vector
where ~ is the velo’oity vector.
Assuming frictionless, homogeneous fluid gives the

Euler equation:

NACA Technical Memorandum No. 9?9 5
dp
—=
~+$gradp = ~ + grad. o (lo)
P
J.
Hence Q is also a gradient, and an ltacceleration
potent ial[l ~ can be introduced, so that ~ = grad ~.
l%en integration of (10) gives
n dp
y+ ~=f(t) (11)
I
-.
If the flow is undisturbed at infinity, then f(t) =
const. If, in addition, it is restricted to small dis-
turbances, then p- Coilst. and the acceleration
q= - ~ -i- const (12)

P
follows from (11).
On the airfoil, pressure p, and hence V, manifests

a jump (or discontinuity). At any point outside the
lifting surface the value p, and hence cp, is continu-
ous for V<c.
The acceleration potential CP can be approximated,

in the case of compressible fluid, from a d~u–blet super-”
position. This procedure would be rigorously executed for
&compressible fluid, but for iarcompressible fluid the
pressure amplitudes would have to le small. This super-
position can also be assumed on the regulating surface in-
stead of ‘on the lifting surface, provided the interferences
are very small.
Through the superimposed doublets, the intensity of

vhich at any point corresponds to the magnitude of the
pressure jump on the lifting surface, the acceleration
potential ~ is defined and follows from (9) and (12).
Then the velocity field must be defined from cp. On the
previous assumption that the flying speed v is constant
and that the lifting surface moves along the negative x
a~is, the acceleration vector is ,
(13)
the terms of the second order leing ignored.
posing Q = grad q, ~= grad @, followed by in-

tegration (13), gives
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
Provided by IHS
,-
Licensee=Boeing/5910770006
I .. —— -—— . .- -.—. ..-
6 I?AC4 Technical .Memo~apdum No. 979
(14) ,
Gtven Cp(x,y,z,t), the desired acceleration poten-

tial finally follows from (14) ,at
x
XI dz~1
@(x,y,z,t) = + ,CpXt,y,s,t -~ (15)
r( )
“J_~
One” particular advantage ofthis method of represen-

tation is that the so-called”area of discontinuity behind
the airfoil plays no part because, though it manifests a
velocity jump, it has no pressure jump and hence no dis-
continuity in cp.
An element d c of the surface covered with doublets

furnishes, according to (9), (12)8 and (15)> the Potentia~’
The analyzed element da lies”in the zero point of

qur coordinate syste,m~ that is~ at x, y, z = O. . .
.. .
5. “THE GENERAL INTEGRAL EQUATION OF AIRFOIL THEORY
The location of our elementary radiator is now shift-

ed to, any.point of the airfoil with the coordinates ~s .
Y(n), z[n)* Coordinate ~ is measured parallel to the
,generating ,liqe of the regulating surface; hence along
the. x axis. Coordinate o is so chosen that, after de-
velopment of the regulating surface .in a P~anes ~ and
n form a system of Cartesian coordinates, and the sub-
stitution
x Y.. z
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
x-g Y- Y(q) z - z(m) {17)
must be effected in (16).

..”.
The element of the surface of the airfoil is da =
ti~da. QrdinariJy Y .’will.still lie a fu~ction’of ~ and

?-”-
,., , ,., ,
NACA !Jechnical Memorandtirn No. 979 ?

. .
,.
-- . . .
q because of ““-
the variation of doublet intensity on the
airfoil. The “differentiation along the direction of the
normals “can be divided in %
a .——a dy+ G dz .
—— sin a(~) -$ + cos a(i)’% (18)
z- ay, dn az dn
where angle a(q) between the direction of the normals

and the z axis is a function of the curved coordinate ‘n.
Posting (17) and (18). in, (16), followed by integration
over the airfoil, gives the complete term of the velocity
potential
,. of the airfoil at:
~I=x-f
r’ a\
@(x,y,z,t)=*
‘J]J ‘F1
“d~!d~dq”sin
‘m
( ,
a(n)a$+cos a(~) w)
)
> X1-x+f Xla+(l-pa) [(y-y(q))a+(z-z (1-1))2]
y’ t.,a’?t+ — X1$ --
\. ~ v c(l-@2) C(I - @2) (19)
. .
+(1- P2) L-(Y - Y(n).)2 + (z - z(d):]

--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
On the other Band, the p,ressure jump TT on the air-

foil is proportional to the intensity of the doublets.
The constant factors in (9) are precisely so chosen that
n(g,n,t) = pvY(5,n,t) “ (20)

This conforms to usual practice: lift’, positive upward,
downwash, positive ”downward. Literature at times quotes
the more abstract? downwash, po”sitive upward, in which
. case the prefix of the right-hand side of (20) must be
reversed., ..
‘“ Equations (1.9) and (20) represent the most general

integral equ,atio.n of the airfoil theory for small dist-
urbances that can be used for computing th’e pressure
jump -@ for a given down”wash. E“quatio.n (19) represents
a boundary value problem. Its soluti’ori-rests on” the fact
that. the ,downwash,on ‘the airfoil it’sel”fis given by the
type of,motion and forq change of the” airfoil in first
approximation. Assume that q = n(x,.n,t). is a small de-
flection of the<airfoil ‘$,r,omthe regulating surface, in di-
rection of the normals. !l!hen:.thedownwa~on the airfoil -
the terms of the second order being neglected’- is

,.
8 NACA Technical .M?vor.andVm No. 979
,.. Zln ih
,,. . ‘W(x, y(n), z(”n),t) =“= + ~ ~ (21)
When solving (19), the Kutta flow-o,?f condition must

be met. It is met when Y = O on the trailing edge. A
frequently employed. formula,which corresponds to this con-
dition, is:
wg,l-l,t) = g(n,t)
[
aocot$+~an sin n e
1
where
~
COS 6 = - i
and 1 = l(n) , half the wing chord. For special pur-

poses, such as airfoils with circular contour, for in-
stance, coqb’inations of spherical functions of the first
and second kirdare used, whioh also satisfy the Kutta
condition,
It may be mentioned that the v’ortex.method also

arrives at the quantity ‘Y defined by (20). .It indicatfs
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
then the density of the bound vortices. This quantity

has, however, a much more general significance than doub-
let intensity indication since the vortex concept is re-
stricted to” incompressible fluid, while (19) and (20) ap-
ply equally to compressible fluids.
6. SPECIAL FORMS OF INTEGRAL XQUATIOIJ (19)
The general integral equation {19) can be specialized

in several directions.
a) For solving the boun~ary value problem, the down-

wash on the airfoil itself is used; that is, for points
of origin with theooordinate y = Y(n~): z = Z[n,).
Sirice solution (2) of the”wave equation UpOIA which our
‘integral bases applies only “oiitside the singular point
r’= O in view of the ,linearization, one may not integrate
through the doublets beqause then “the integral over X1
becomes divergent. Hence the downw~sh oxi“the airfoil
itself can be’obtairied onlyby a limit transition from
‘(19) . “ The method nearest at “hand consists in carryitig
out the integration with respect to ‘two surfaces a% tiis-
tance ‘+s’ and -c fromthe airfoil, in forming”the”. aver-
age value and proceeding to c-+’). This method, applied

NACA Te,chn.ic,al.,
Mem~.r,and,~,No. 979 ‘9
:.. .
‘-t--o
,(19) , leads” to cornpl”i”tiatedfor,mul”as. On the other ~~
hand, the same result. i.s achieved if, after the differ-
I entiations, y = Y(ml) and z = Z(ql ).,,, are. formerly
1 written in (19.) and the divergent integral is then so
‘$ ‘transformed by partial integration that. Cauchy’ts:princi-
~~ pal yalue canbe formedi This change makes the .formul,a
.longer ant!, in. general, less comprehensive. ,So.in the.
following we confine ourselves in most cases to the.
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
,st.atement..of, the simpler divergent integral while tacit-
ly presuming the further transfor,mati.,on in, the Cauchy.
principal value. .If.”the integration with respect to X~
is, made numerically, the divergence can be avoided quite
simply, ,by either integrating from x - ~ to -m or
from +m .,to x -t. depending upon whether x - g< O
or > 00 For in any event the downwash must disappear at
infinity. In the divergent method ,of yriting the down-
wash on the airfoil ultimately assumes the form:
b) It can be assumed that the regulating surface is

a plane, say, the xy plane, for instance. Airplane wings
are usually. flat structures. ,Then a= O, y(~)= ~, and
z(q) = o. Nith specialiaztion a) z = O also. This as-
sumpti~n itself effects a substantial simplification in
the,eguation form and has been practically always int&o-
duced in the airfoil theory so. far. .“
...
c) Some special” assumptions “regarding’ the type of

time rate of change ”can be made. The linea~i’ty of the
integral equation affo~ds. especially simple forms on the
assumption of’ harmonic processes with respect to time;
heqce,harrnoqiq oscill.ations,,of the airfoil. The preemi-
‘hent importance of the harmonic solutions of the wave.
equation in.,physicq is an es”tablishedf act. Once harmonic
solutions of the integral equatio,n, are known, solutions
for any time ra’te of change of downwash can be found by
superposition whereby the r~La@lace transformation
plays a pro~inent part. ‘Proceeding to the limiting case
,of”’”verysm?ll oscil,lati.o~.and putting. .y = y(~t~) results

.. 1111111 II ImIn I I I.,.. I . . . .... .. ... .I. m m,,.. ,.... . .. ,. .,., .- . .. ,-.., I . .
. . . . ,, ,. .- —-—.. .-. —-—.-.—
— — _ --
10 ““NACA !l!ec”hni”cal
Me”mci%rid.umNo. 979
in the stationary ‘form of the fntegral equation.. ‘Enc”um-

b’ered with further r,estrictiona, this special case ‘has
been, treated most so “far. ‘
,. J
d) “Simplified “assumptions regarding the shape of the
airfoil can be made,; Allowing the span of the airfoil to
increase to infinity and assuming the downwash to be in-
dependent of rI gives the same flow process in all the
~ planes n = const, so that the integration ‘with respect
to V can be effected, The form ‘of the integral equation
1 so obtained represents a two-dimensional flow process,
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
This form is of particular intere”st because it is solved

for any downwash function under the restriction 9 = O
anii for stationary flow also with @#o. ‘The significance
bf this solution in the so-called vortex fi”lament theory
w*13 be explained elsewhere. .,
.,
On an airfoil of constant chord and infinite span
both the ,downwash distribution periodic in q
.
and that independent of q can be taken into considera-

tion.. ‘This defines a Y distribution pe,riodic i~ q. ‘
The integration with res,pect to q can be effe..cted.in
various cases. From these” harmonic solutions superposi-
tion affords solutions for any finite airfoil of constant
chord, in motion. The method is therefore applicable
only to airfoils having a parallelogram (especially a
rectangle) as contour and pressu-res that no flow “passes
around the” lateral edges of the parallelogram.
The ultimate aim of the simplifications is the

change of the surface integral in (19) to a line integral,
which naturally is more promising” for a solution of the
integral equation. This aim can also be reached by as-
suming airfoil contou”rs representing coordinate lines in
simple s’ys”tems ‘of curvilinear orthogonal co-ordinates. “On
airfoils of elliptic plan’ form the “nonfocal” elliptic co-
ordinates permit” integration with’ the aid of the Lam~
functions’. For circular contour there arethe spherical
functions of the first a“fidse’eond type. Examples’ can be
found in the. reports by Kinner (reference), Krien8s
(r.eference. 4), andSchade (reference 5).
‘..’” ‘..
e“) It is readily apparent from (19) that the assump-
tion of incompressible fluid affords”a stibstantial sim-
plification. in the form of the integral equation beeause .

—--
y-“
)
\ .,, NA-CA T,echni.cal Mem~;ran&u”m No. 979 11

)
\
y-
i, P“’=o and” c’”= cl”.’”> ‘“O&e’”< xce~t’<on” ‘is the stationary f low ,
~ where, m~reover, t,he complicated argument of V, which
starts with t, ‘disappears. Allowance for thk cornpressi-
1 bility then consists merely in the reduction of the x
( uJ~,
f coordinate in the ratio the so-called Prandtl
\ contracti.om..,..which represents the exact counterpart of the
Lorentz contraction” in the special relativity theory. The
d“ownwash therefore should “be computed for an airfoil with
less chord, which,thus affords by given pressure jump a
smaller downwash and$ conversely, by given downwash a
higher pressure jump than for P & O.
In the case of stationary flow it forthwith affords

solutions F$+o, as soon as solutions for ~ = O are
available.
The’ oscillating airfoil for. 0< B < 1, has been

treated in only one report (reference 7), and then, as
a two-dimensional problem. All other investigations of
nonstationary airfoil motions are restricted to the assump-
t~oil of incompressible fluid,. The reason for this is to
be found in the fact that the general case @ # O was not
approachable by the vortex method solely employed hereto-
fore until Prandtl introduced the acceleration potential.
7. THE AIRl?OIiI OF INFINITE SPAN
The integration with respect to q on an airfoil of

constant chord and infinite span in (19) is dictated by a
special assumption for the relation of function V ,~ith
time T. We put
@T);= g(t) exp ivT

,.
that is, study the harmonic oscillations of the airfoil.
This assumption implies no loss in general validity since
any others can be built up from the ha,r”monic solutions by
superposition. As the ultimate result a closed solution
b’
“ev@fi’cafi b’~ indicat’ed ’which”’is Applicable to” any downwash
fqpctionwithbut resorting, to harmonic analysis? With the
specializations a)’; b), and c) we obtain from (?2) ‘bY
derivation according to z the downwash slang the z axis
at :
,, ,,:
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

I,~},,
No reproduction or networking permitted without license from IHS
,, l,#mmmm
m,,,,
m,m—---.-m .mmm.—m-m s 8.-ss . , ,
Not for Resale, 01/16/2015 11:45:54 MST
. .. .. ....... ........
—. ——... .-.-.——— —
. .-
12 ..NACA Technical, Memorand@ No? 979
-----
+m.
“’”:.
t(P-wf
,: +!
,,
X’:kx-f
W(x, y,z?t) = — d X1 d q ii{:-”~”(~,

t)
{’
=. =- ,/ -m
~2 GXp i v L+ “
v(l-~n) -
xf2+(l.-@2) ((y-n )a+z2) 1
I
[ v - . —r.
.. , . .-:C(l - P2)
. —, —,.. — -.— J (23)
.——.
l~hen the new variable

-~2)(y-n)2+l
u’ = J-~-—-———
x’~ -t (1 - pz)zz
is introduced into (26), bearing in mind that the part in-

--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
tegrals with respect to q tctweem the limits + m to y

and y to -m must be of ideniical magnitude, the inte-
gration with respectto m can be effected with the cylin-
drical function (reference 6)
03
‘2)(x)
. ! ‘1
J.
e.p(-iXU) ‘_j..~~
= --E~ O (24)
and (23) and (24) give the Possio int:egral equation
W(x,z,t) =
*~fl ‘:’7’ ‘x’d’’(g”) ‘Xp$[’ - x
=- -m
1
x ~ ~ (~) vJx,~2 + (1 - 62) 22
+— (25)
1- P2 1 az2 o
( C(l - $2) )
Possio (reference 7) does not give” this equation explicitly’,

but’ m”erely i,ts kernel in t“he complex real fo m in order to
comp’ute numerical approximate solutions. Ho t)2 denotes
Hank’ells cylindrical function of the “second type, 1 half
the airfoil chord. The origin of the. coordinate
., system is
at airfoil c“~nter. ‘
After differentiation and tnsertion of z = O,

equation (25) gives the downwash at

..—.—..—.———-—.
.,. ,,...-—. . ... . . . .... ... .. , , . ,.,,., ,,, ,,, ,,., ,,
’13
(26)
wherein
Y .’
03
=- &Y’ exp iuHl (a)(p ~) * for y>o

[ .
c 3r
m
=~PY r exp (-’iu]H1(2)(@ u) & for Y < 0 (27)

d!.,
.}
The absolute value Iul appears in (27) and farther on
because the argument has the significance of a radius.
With
lim x HI (2)(X) = ~
X+o
it affords for the function R the special values.

/
R(p,()) = 1 (28)
~Y
R(O,y) =“-y /“expiu$ i29)
/
:.-m
In stationary flow v
. O
=. and (26) and (,28) afford
W(x,o,t) = q ‘:’f(g) _u_ (30)

,,
J -[
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
‘-t””x’
The root factor represents Prandtlls law, according
to which the compressibility raises the pressure jump in
the ra~io’ l:d~. .Integral equation (30) was origi-
nally solved’ clos”ed by Munic (reference 8) for P = O.
In compressible fluid ~ = O and (26) and (29) give,

)$
I
Wmm II.1A II II ■ m-- m-

’14 NACA Te&hnical’ Mdmoraritlurn No, 979
with another integration variable ~ II=.,x - + u, the

Birnbaum integral equation,. .
Birnbaum (reference (9)) writes his equation in the

convergent form following from (31) after partial inte-
gration and formin-g Cauchyts principal value, whereb~the
sequence of the integrations can be changed.
Equations (30) and (31) could also be deduced fron

(23) direct, bi putting ‘v = O and c = b before inte-
grating.
Birnbaum.fs ?quation was solved closed by Ki.issner and

Schwatiz [’reference 10) with allowance for Kutttifs flow-
off condition. I?or more convenient representation of the
solution it is expedient to introduce the new variables
,.
‘x’= -1 COS”6 ; [ =’ -1 Cos d
and the parameter

iv t
w= iz=~
Then the solution, onthe premise of harmonic time func-

tion, reads
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
m..
Y(e,t) = $ .[1+COS19 + T(-iw)(l-cos 79)] cot :

.
[{.
0
2 sin e 1-COS(6+’8) :
-b +wsin tln w(~,t)d~ (32)
cos e - Cos & l-cos(e-o) }
:,
HO(2)(U) + i Hl“(+(-J)
T(m) = (33)
-Ho( ?J(@ + f Ei(2] (~)
.,,
The complex.function T can be computed with the
aid of tabulated functions and is given in table 1.
.,, . .

~7.. .—
l?” ‘“”- “ ““ ‘--
NAOA Technical Memorandum No. 979 15
Thik result’ lias been: applied’by the author (reference

11) witht lie aid of the [email protected] principle ,to any
“downwash function W(ags)”; ‘s, is’’the p’ath of’ the airfoil.
,. ..,..~+~n+,
. . -.+
Let w(d,s) =0 for s <’O.Then the. solution reads:
.,,. :,.. ,,, ..
l-i
.Y((j,s)
f
‘m0
= *: (I+cos
~~‘““.2’c
Os”e- ‘
a)cot
1
w(~,s)
~~
+sin71 in l-cos(ei.$), aw.($,s)
l-cOs(e-d) a9 ,
s“ ‘“:
aw($,o-)
+(1 - Cos 73) cot :“ Ul(s -5) (34)
r ~do } ‘a’
The real function UL(S) can”be represented by

series or a pure real integral over tabulated functions
Numerical .values are given in
[~~~~rj~ces It 11
should be noted that (34] holds for variable
and 12).
flying speed too, but (32) for constant flying speed only.
ilext we shall analyze the yawing airfoil of infinite

span and constant chord. The angle between leading edge
and axis y, the so- ca,lled angle of yaw, is denoted with
6. The half airfoil chord Z is always measured parallel
to axis x. Putting c = w..~nd s’ubstituti~g
x- ‘y tan 8 = s; ~-I-I tan 8 = 0

,, ,.
gives, .frim (23)$ the downwash -
W(s,o,t) =
+1” m
r.
21TCOS2
1
&
--~
0--
h :.,
.’,
xll=(J-
,
‘Y(~,t)”‘exp-+(0- X“) ‘X”do
(S.-X11)2
(36)
This result differs from (31) onlyiti the constant fac-

tor COS26. ~iit~ given downwash the pressure jUMp of the
yawing airfoil of’ infinite span is ’;there~ore smaller by the
factor COS26 than that of the normally flying airfoiJ of
,..
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

16 NACA Te’cfinical Memorandum No.” 979
equal chord measured in. direction of flight. This result

is synonymous with that, in that only the normal component
of the flying syeed, i.e., that located’ in the plane nor-
mal to the airfoil, is aerodynamically effective. It iS
readily apperent that in frictionless fluid the tangential
component, which corresponds to a displacement of’ the air-
foil of infinite span along its generating line, must be
devoid of aerodynamic effect, and therefore not ascertain-
al)le by pressure measurements outside the airfoil.
But , if the pressure is measured in a reference sys-

tem that is solidly connected with the transversely moving
airfoil, the Lorentz transformation is dictated if the
fluid is compressible. With V~ denoting the transverse
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
velocity measured contrary to direction y, it is neces-
sary, according to (5) to substitute
in the integral (26).
Next comes the case of the airfoil with periodic y

distribution across the span. Let
The study is restricted to ‘the special case c = ~,

that is’, in compressible fluid. Putting
v -Y
‘= JX12 + Z2
while noting that the part integrals from +~. to y and

from y to -m’ must be of the same magnitude, affords
the determinate integral (reference 13)
m
(38)
where N is Neurnann$s cylindrical function and Z the

Struve ,function. JLbbreviating
GO(X) = Zo(i x) - No(i x)

(39)
GI(x) = Zl(i x) - lJL(i x) - #
Copyright National Aeronautics and Space Administration (NASA) 1
— — .— —
.NACA Technical Memozaqdum ~o. ,979 17
.,
..-. -.+.l-=xl=x-.~.
.,, ~~ “’“’
,. .... . ..—--, -.’,,”
(40)
d GO(X)
=- i Gl(x)
ax
the differential follow,ed by insertion of z = O, x“ =

x - Xt’ gives the downwash at
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
Integral equation (41) is similar in structure to
(26) . Its further integration” requires a special assump-
tion for the time function. If harmonic oscillations are
~ .-involved, a real formula ‘must: be used or (41) must be,
I Writt,fjne Since. ,
1
t 2i
lim’ XGI(X) =-— (42)
1
,’
; the
..
downwash for
x+” o
K ->0
.,,
n
follows from (41’) and (42) at
\ .,.
1}
8. THE AIRFOIL WITH HIGH ASPECT RATIO

,.-, .,
According to (23) the downwash of a-flat oscilla ting
!
I airfoil in incompressible’ flti”id is
L
‘1:
~Provided by IHS Licensee=Boeing/5910770006
>]No reproduction
-,-,,, or networking permitted without license from IHS
,,-,m--_m,,,,, ,,,.. ,.,.--,,.- Not for Resale, 01/16/2015 11:45:54 MST
—
,18 I?ACA Technical ~Memor.anciurnNo. 979
,..
Xl=x-g
i
W(x, y,z, t) = ~ dxtd~d~~(~,~’,t) exp ~ ‘(xt-x+t)
:,
!/v‘l?l’
-m :
a2 1
(44)
x m Jxt2+(y.@+z2
The introduction of the new integration variable

~1 = - W/(y -.)2 + !Z2 (45)
in (44) affords
a2 ‘J’
‘F1
x— exp(-$JY-~)2 +z2~) *q- (46)
az2
L
-x’+.
“J(y.-q ) 2+22
On an airfoil with high aspect ratio we find for most

pointso”f the surface
~.
(47)
I x - E<<
y -~
:..
Only in the closer proximity”of the starting point,
which itself may be situated on the airfoil, does inequa-
tion (47) fail to hold unconditionally. The more slender
the contour the smaller the error introduced when the lower
limit of the integrals approximately put at u = O in
(46) ●
~ol~owing this,the integration can be carried out
according to (38) with known functions, ultimately yielding~
after differentiations the downwash
W(x,y,o,t) :“*
n ,, d~dri y(~,m,t) exP + (-x+f)
(48)
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
IY - q
,, ,.
.. . .,.
Equation (48) is substantially simpler than the orig-
inal (44). Admittedly, it holds only for special downwash
distributions in x direction; namelY~

— .
.NACA Techn”icalM&m”orand.um No. 979 “19
,, W(x,”y”;
o, t)” = FTy) “ex~ iV ‘t (49)
““ (“:””+) ‘“
a.s is readily seen from a comparison of (48) and (49). It
affords the downwash function
flpy(~,n) -p
~G’~,
v
“ d@q
;(y) +-
J
/j
., ~1
(
lv,ly-
“ !) IY-111
(50)
1
!.
f,
This equation can be solved implicitly
supplementary
next assumption
assumption for function
it is established
only with a
Y(ts.ll;. d:;t:;:”u
that the . .
I tion as a function of Et uP to Z faCtOr c(n) d~pend,~
ent on q, shall be such as if two-dimensional flow ex-
isted in every section. This is the second essential ap-
proximate assumption. The downwash being given by (49),
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
the desired Y distribution can be computed from (32)
and the integral
TI(T)
K(~) = ivedt
r y(~,n) exp ~ (51)
d
t~(’q)
formed. Entering (49) in (32) affords with
f= to(n) - l(~) Cos a;ti(q) =vl(q)/p
Y($,q) = C(n) ~ (~) exp - ~~ fo(n)) cot j [Jo(m)

(
+ i Jl(~) + (Jo(@ - i J1(U3)) T (0)] (52)
to(n) is the coordinate of the airfoil medial line. Yor

I airfoil contours that are symmetrical to the q axis$
ii go = O*
~~
‘,
i Jo are Besselts functions; T is given bp
,1 (33)’ as q~~~ie~~ of: Hankells func.tions~ Entering (52) in
\ (51)., followed by integration with respect to ~, gives
1:
= c:(;) n.w(n): I(n) f’(fiin))
K(o),,,
p
(53)
f(~) = J02 (m)+ J12(ti)i-(Jo(fi)- i Jz(~))z T(m)
‘}
,
If the flow were actually’ plane, ~(q) would be equal to 1.
I
,.
,, Next (51) is entered in (50). Since (50), for rea-
‘$ sons outlined previously, is divergent, partial integration
II
~~
/1Provided by IHS
Licensee=Boeing/5910770006
!’
,J,!
20 NA.OA T&hnical Memorandti No. 979
is necessary for ,..,

forming Cauchyls principal value. Then
.,
By unlimited increase of aspect ratio, dK/d~ tends

toward ze’ro. Then exact plane flow prevails and
K (y)
W(Y) = F(S,K) = —— (56)
n?(y) f((qy))
according to (53) and (56). The final form of integral

equation for the oscillating airfoil of large aspect ratio
follows from (55) and (56) at
+b
w(y) ~
K(Y)’”””
nl(y)f(~(y))
-
+ 4~
1
,
[S(3Y+!WSL
(5,)
Putting u = O,
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
s(c)) = 1; f(o) = 2
gives, from equation (57), the stationary

+b
(58)
This is the well-known Prandtl integral equation of

“vortex filament theory’. ‘1 One usually substitutes in the
first term of (58) for n the measured stationary circulat-
ion constant of a real airf.o.il section C1<TT, although
the whole theory dealt with here applies” only to airfoils.
Quantity K ,.in (58) has a very elementary significance;
namely,
i(m) = r y(~jn)dg
J,
that is, the .so-calledcirculation ,of the wing. The sig-

NAC~ Technical Memorandum
.. I~o.,,979 21
nificance ,of K ,.,.,.,”,...

in. (57)x.=i,5
.le.
s.s:
,e.lemen$ary. . .Cicala (ref-
; ‘e~en@e’’~4],=-id others have attempted to extend the vortex
filament theory by means of vortex concepts to the case
of the oscillating airfoil and actually arrived at an
equation .agreea33e.with (5’7)., But Cicala arbi,t.rarily
identifies K with the ’total “circulation (sum of bound
and free vortices al,ong wing chord) and accordingly finds
AS concerns ihe permissible downwash distribution

in x direction no data are give”n ’at ‘all. SO, in this in-
stance, it is necessary to state, the vortex method leads
to inaccurate results. Equation (58) ca~.,notbe general-
ized to (57’) by vortex theory, althoug’h (57) can be spe-
cialized in (58).
Equation (57) ‘first gives quantity K(y). But to

compute the lift requires the bound circulation
3
r(n) = r Y(5,q) d~ (59)
$J
Putting (52) in (59), followed”by integration with respect
to f yields
+ i Jl(m) + (Jo(ti) - i Jl(~))T(~)] (60)

--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
Comparison of (53) and (60) gives the bound circula-

tion
.
‘K(y) exp (,- ~~(Y)
)
r(y) = ....! (61)
Jo(T(Y)) - i Jl(ti(y))
~’ It is again pointed out that (57) obtained from (44)

by two approximate assumptions applies only to downwash
I functions of the form (49), where ~(y) can be a complex
B. function. Other downwash functions cannot he treated by
~ (57) unlessone,is satisfied .to subtract. a downwash func-
tion w in the form of (49) from the given downwash ~
I .,function,. Wg and. to ,inte,,gratethe, smallest possible dif-
ference w - w on th”e assumption qf, tw~-dimensional
flow~”by m~an$ of (32). This method has beenused until
the present time $n’the practical application of the sta-

— ,...——
. ..
22 “’ NAC’A ‘Te”chni”d’al;
M&rnor&tidUm No: 979
tiohhry” ,form ‘(58) without” bei,ng.at all Clear $h.a~ it in-

., clude”s a suppl.e”mentary”’‘assumption going beyo’nd” the vor-
tex filament theory. ~ ..
When an attekpt is made” to a~ply (57) under the
Cited assumptions” approximately to any other downwash
distribute’o”h W(a, y)$ the” question of what mean value
to employ for %7(Y) in. (57) arises. BY assuming plane
flow quantity Y can be computed with (32) for any func-
tions W(?9,Y) Equation (4’9) is then so chosen that it
●
a,ffor,ds the same characteristic quantity K On the as-

sumption” of” plane flow” “as’‘the ~iveti do”whvrash w(ij,y) in
plane flow. Then . “i-r.
,n
K1(y)=l (y)eXp&~o(Y) “f(8*Y)e
xp(-~cosO)”sine de ““ ~
J
“o
rr{
~(y) ‘n,-
= exp$~-o(y)~ [l+cosO+(l-cos$ )T(U)]cot$
.,<
:+ Jo
2 sin l-cos(e’+ a)..
-1- bitisin~ in w(.d,y)xexp(-ificos 9)sin(3klBd8
cOse-cOs O l-cos(6-a) }
.
=I(y)exp (+to(Y) ~ (l-costl)w( 8,y)d~[Jo(~)+iJ1(fi)
“J’.
+ (Jo(m) -. ~ Jx (a) ) T (~) ‘1 (62)
On the other” ”hand~’ equation, (49) gives,’ according to

(53)
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
K2(Y) = ITl(y)fi(y) [’JO(=): - i Jl(~)~
[Jo(~) -I-i Jl(~) + (Jo(m) -’ i Jl(@), T“(~)] (63)
Putting Kl(y) =. K,z(y), equations (62) and (63) give
(64)
:.’, ,,
,’.. !?~e numerator o.f ttiis.fraction represents the known

integral, term W. that is decisive for the circulation
.,” in pl’ane-flbw. By translati~nal’ and rotary oscillatitins
o.f a’flat ’airfoil:,wo:’is ‘equal to the downw’ash at 3/4
.
wi’ng chord (rear’ n@ut”ral point)’. ‘For<’st&tiohary flow the
.! ,..

.,,,,. !.
,.—-.—.
!.
NAGA Technical Memorandti No* -979 23
\ .. .. . ._
~
.. ... denominator- of -the f-ra”c-tlvn (64) is ‘eqii’ al”
to l-; then w =
T w. But this does not hold for nonstationary flow.
J t!ng, as a check, (49) in (64) identically satisfies ~%
~~ .,, .
I
‘d Allowing for (53), (61)s and (64) , equation (57) can
be written in the f.orm
I
,.. r(y) “
il~
~ (1-co sa)w(’a,y)afi = ,
i n r Jo .tiJ1 HO(2)+ i HI(2)”
~, -1 o ,,-
!’ TTt(,y] .
1 Jo -iJ1 HO(2) -i Hl( 2)’ 1
+b
“J”-
yJ
(65)
where J and H are functions of m(y) outside the in-

tegral and of m(n) ~nsi.de of the integral ; ,r is the
customary bound circulation. In consequence of the ap-
proxiinake assumption regarding the mean value of ~, equa-
tion (65) no longer gives the circulation of the plane
flow exact for infinite aspect ratio, which makes it nees-
sary to put the firs”t quotient of the first brace equal
to one. This is apparently as justified as in equation
(56). ,.
Visualizing a stationary’ periodic downwash, field,
while discounting for the time being the e’ffect of the
vortices already contained in this vel”ocity fi,eld on the
airfoil, a flat airfoil moving rectilinearly and at con-
stant speed through this downwash fteld, is continually
“ subject to’ a downwash of the form (.4’9),whence (57) is
.f~’.rthwithapplicable. Any s“tationary ’d”ov~nwashfields can
‘be presented b,y linear superposition of several such
~~fields of different periods. Equation (57) is therefore
especially suitable for treating gus-t stresses of airfoils
ii) with large aspect ratio. .
The complex function s(x) occurring in the kernel

of the integral (57) is given by (39) and (54). i’Jumeri-
cal”values. are given in table 3. Because of
., :
*
~im X2 Gl(x) ’,= - 2 (66)
x —>m “ --`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

Provided by IHS *Probably an error. .. Licensee=Boeing/5910770006
..—— .
24 ITACA’Te”chn”i’cal’
Memorandum No. -979
‘ i im x, S (“x”)=.- +”
x,+- m
follows from (54) and (66). 4

In rapi,d oscillations, that is,” large values of
v
—Y- I the induction
n} , effect given by the second term
v .
of (57) is therefore very small, and in the extreme case
of high frequency an almost plane flow must be counted on
even on an airfoil of finite span, if the aspect ratio is
high. This is an important result.
Theoretically the solution of (57) can be effected in

the same manner as the much discussed equation (58)S al-
though the addition of the complex function S makes it
more protracted. The result, which is, moreover, usually
encumbered by the assumption (64) in partial cases, con-
sists on airfoils” with large aspect ratio and at the prac-
tical values of G in the order of magnitude of 1.0 only
in a small correction relative to plane flow.
In case such refinements of the solution are not

deemed necessary, it is more appropriate to apply an iter-
ation method which ties in with the exact integral equa-
tion (44). A somewhat correct approximate solution
Yi[,q,t) for a given downwash w is afforded from (32)
on the premise of plane flow. Entering VI in (44) gives
a downwash til # w. The difference w - WI is then en-
tered again in (32); A’Y is computed and this added to
Y. as correction factor, etc. The convergence of this
method needs to be checked of course from one case to the
other, although it should be sufficient in general fop
slender airfoils.
There is nothing to prohibit the application of this

method to compressible fluid (s+0), once the general
solution of (26) in form of an integral representation of
the type of solution (32) found for @ = 0, is available.
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
9. SYSTEMS OF SEVERAL AIRFOILS
,“
The general form (19) of the integral equation of the

airfoil theory comprises the possibility that the surface
integral d~dq can be extended over severalspatially sepa-
rated regions of the airfoil. A case in point is the
I
NACA Technical, ~e”tiorsn&w;;~o.,.9,?9 25
!.
biplane or the wing with .spl.it;i:f~:a~pOf .cours.e, .Kuttals

~ flow-off condition mus-t-b~satisf-ied for the trailing
m. -
;) edges of each part of the airfoil.
,,
,..,..
....
.. . . ., .>- ,;:, ,.
‘i
I
A Give: a general”’ solution Y = &(w) .“o’f.,[email protected]
‘)
equation for a single airfoil, the, problem of two airfoils
11 can be sglved by successiv.e:a.pprox’imati’on accor:ding.%;o the
~-
ii
! superposition prin i le”~ .As-.sume the. given downy s: of the

. airfoil (1) as” w ?? ,1 .and~of airfoil (2) as Wg:a” t? = O.
t
First &ompute:as Solution, i“n fir’st approximation’
.
,,.,
...”. .
.,.,.
.
As a result of this pressure distribution, there is in

.conformity with the” general integral equation. an”.alwais.:
computable downwash fj..el.d .,,
,.. . .. . ‘,
.
which induces bn aififoil. (2) the pressure distribution
which in turn creates the downwash field

*VT(l) = qAY(2))
,— .
, ,. ‘,:, .:, .
,.,. . . .
,..which induces on.!airfoil (1) the. pressure distribution

Ay(d = ~(l) (Aw(l))
.,. .’,
..-:+.?..:,..%
. . . . ..
,.
., .In the second approximation- the’prei-u e dis r bution of

airfoil (1) %s there.f,ore,g vqn by V ??~ -i A’Yt!1 , and
that of airfoil (2) by AY t)2“. This method of iteration
can then’:be continued. ‘:Iu-cases where the distanoeof
the airfoils is not very small in’ r’~latioh td’thewing
chord, it converges very well, as proved by Kleinwschter
(ref.erenee 15.) for stationary -flow. ~~... . .
\
~ ,. Theiteration method is “chiefly opportune when the
solution & in continuous form is known. Thus for t,h,e
present it,is restricted to two-~im.ensional flow- with
m c.=?, for which the ..solutions- (32) and (34), ‘respec-
tively,. are availabSe. : .
Translation by J. Vanier, ,$
National Advisory Committee .,
for Aeronautics.. . ., ,.
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

IL
26 NACA Technical Membrandun No. 979
REFERENCES
1. Joos, Go: Lehrbuch der theoretischen Physik. Leipzig,

1934., p. 210. 4
2. Prancltl, L. : General Considerations on the ?1OW of

Comi>ressible Fluids. T!.M..
No. 805, .NACA, 1536;
Beitrag zur Theorie der Tragenden l?l~che.’ Z.f.a.M.M.,
vol. 16, no. 6, Dec. 1936, pp. 360-61: Theorie des
Ylugzeugtragfliigels in Zusamnendrflckbaren Mediun.
Luftfahrtforschung, vol. ,13, no. 10, Oct. 12, 1936,
PP 313-19. ●
3. Kinner., W.: Die Kreisf5Fmige Tragfl~che auf potential-

theoretischer Grundlage. Ingenieur-Archiv, vol. 8,
no. 1, 1937, pp, 47-80. illus.
4. Krienes, Klaus: The llllli~tic Wing Based on the Poten-

tial Theory. T.M. NO. 971, .NACA, 1941.
5. Schads, Th. : !Theorie der schwitigenden kreisf8rmigen

i!ragfl~che auf potentialtheoretische r Grundlage.
Luftfahrtforschungi vol. 17, no. 11/12, Dec. 10, 1940,
387-4d0.
6. Yrank-Mises : Die ~ifferential und Integralgleichungen

der Mechanik und Physik. Braunschweig, 1930, vol. I,
P. 418.
7. Possio, C.: L~azione Aerodinamica SU1 Profilo Oscil-

l.ante in. un Fluido Compressibil,e a Velacita Iposonora.
LIAerotecnica, vol. 18, no. 4, April 1938, pp. 441-58.
8. ‘Munk, Max M. : General ~he,ory of Thin Wing Sections. Rep.

iiO., 142, NACA,, 19226
9. Birnbaum, W. : Die ,tra,g~nde Wirbelflache als Hilfsmittel

zur Behandlung des ebenen Problems der Tragflfigel-
,, theorie. Z.f,a.M.Me, vol. 3, no. 4,, pp. 290-97.
10. Kiissner, He .G. und Sohwar.z, L.: Der schwinge.nde Fliigel

mit aerodynamiscn ausgeglichenem Ruder, Luftfahrt-
forschung, vol. 17, no. 11/12,, DecO 1.940, pp. 337-54,
11. Ki.issner, H. G.: Das zweidimensionale Problem der be-

liebig bewegten Tragf15che unter Be.riicksichtigung von
Partialbewegungen der Fliissigkeit. Luftfahrtforschung,
vol. 17, no. 11/12, Dec. 1940, pp. 355-61. --`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---

,-
NACA Technical Memorandum No. 979 27
~,, 12. Schwarz, L.: Berechnu.ng der Funktionen ,Ul(s) und

u~ (-s) fiir gr~ssetie Wert’e “von s. Luftfahrtforsch-
~ ‘ “- ung, vol. 17, no. 11/12, Dec. 1904, PP* 362-69.
iI
+
j
13. Nielsen, N.:
tionen.
Handbuch der Theorie
Leipzig, 19C14, p. 224.
der Cylinderfunk-
:,,
J
\\ 14. Cicala, P.: Comparison of Theory with Experiment in
f the Phenomenon of Wing Flutter. T.M. No. 88?,
)- l!rAcA, 1939..
15. Kleinw3chter, J.: Beitrag zur ebenen Leitwerkstheorie.

Luftfahrtforschung , vol. 15, no. 3, March 20, 1938,
PP 127-30+ ●
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
.,
.,L

—
,, .. - _...-F . ..v-...-..wwA& -.
ti
,!
Table1;Function
T(ti). Table
IT.
Function
~1(s). Table
IItFunction
$(z).
——
s .U,(s) s u, (s) Z I@S(z) J~5 (X) .% Jg s (x)
Yfl T G ~e T &IT
0,ooo l,m 0,72 0,12481 — 0,248650,0 0,oooo 3,0 0,4391 1,00ooo 1,0 — 0,37626
0,002
0,010
0,99342 — 0,74
D,96525 — o,091’35 0,76
MW6 0,12034 — 0,244580,1
0,11611 — 0,240610,2
-0,0244
0,0476
3,5
4,0
0,4799
0,5159
0,96969 — w%
0,94104 —0,119191;2
— 0,35560
— 0,33362
4,5 0,5479 0,06 0,91373 -–0,15537 — 0,31091
0,02 Q,92745 — 0,15042 0,78 0,11210 — 0,236760,3 0,08 0,88759 — 0,18537;$ — 0,28905
0,04 D,85340 — 0,23200 0,80 0,4
0,10829 — 0,233041 0;0910 5,0 0,5764
0,1113 5,5 0,6020 0,10 0,86252 — 0,210912,0 — 0,26845
0,06 D,78408 — 0,28519 0,82 0,10467 — 0,22935 0,12 0>S3842 — 0,23302 — 0,24933
0,08 D,72086 — 0,32080 0,84 0,10123 — 0,22580w 0,1308 6,0 0,6251
0J494 6,5 0,6460 0,14 0,81522 — 0,25235?4 — 0,23177
0,10 D,66386 — 0,34460 0,86 0,09795 — 0,222340,7 0,16 0,79286 — 0,269402,6 — 0,21574
0,12 0,61266 — 0,36015 0,88 0,09483 — 0,218980,8 0,1674 7,0 0,6650
0,6824 0,18 0,77129 — 0,28451 — 0,20118
0,14 0,56674 — 0,36978 0,90 0,09186 — 0,215700,9 0,1846 7,5 0,20 0,75046 — 0,29795
0,2012 0,6983 a — 0,18798
0,16 0,62554 — 0,37513 0,92 0,08902 — 0,21251 1,0 8,0 0,22” 0,73033 — 0,309943,2 — 0,17603
0,18 0,48851 — 0,37735 0,94 0,08631 — 0:20940 1,1 0,2172 8,5 0,7128 0,24 0,71087 — 0,320663,4
0.2326 0,7263 -–0,16521
0,20 0,45516 — 0,37725 0,96 0,08372 — 0,20637 1,2 0,26 0,69203 — 03302b 3,6 — 0,15542
0,22 0,42504 — 0,37645 0,58 0,08124 — 0,20342 1,3 0:2476 u 0,7386 0,28 0,67380 — 0,338843,8
0,2618 10 0,7501 — 0,14655
0,24 0,39778 — 0,37239 1,00 0,07887 — 0,20055 1,4 0,30 0,65615 –-0,3iG524,0 — 0,13851
0,26 0,37303 — 0,36841 1,1 0,06843 — 0,18721 1,5 0,2757 11 0,7706 0,32 0,63904 — 0,353394,2 — 0,13120
--`,`,,,,```,,,,,,,`,,,`,`,,```,-`-`,,`,,`,`,,`---
0,28 0,35050 — 0,36376 1,2 0,05991 — 0,17542 1,6 0,2891 12 0,7883 0,34 0,62246 — 0,359534,4 — 0,12455
0,30 — 0,35864 1,3
0,3!!!394 0,05287 — 0,16493 1,7 0.3021 0,8296 0,36 0,60638 — 0,364S94,6 — 0,1.1849
0,32 0,31114 — 0,35319 1,4 0,04699 — 0,15555 1,8 0:3146 :; 0,s733 0;38 0,59079 — 0,369854,8 — 0,11295
0,34 0,29389 — 0,34752 1,5 0,04203 — 0,14713 0,3268 25 0,9003 0,40 0,57566 — 0,374155,0 — 0,10787
0,36 0,27804 — 0,34172 1,6 0,03780 — 0,13953 Yo 9,3386 30 0,9183 0,42 0,56098 — 0,377946,2 — 0,10322
0,38 0,26344 — 0,33585 0,03417 — 0,13264 2,1 0,3500 ho 0,9405 0,44 0,54672 — 0,381276,4 — 0,09894
0,40 0,24995 — 0,32997 ;:: 0,03103 — 0,12636 2,2 0,3611 50 0,9535 0,46 0,53288 — 0,384165,6 — 0,09499
0,42 0,23748 — 0,32411 1$9 0,02830 — 0,12063 2,3 0;3719 100 0,9781 0,48 0,51943 — 0,3S66G5,8 –-0,09134
0,44 0,22592 — 0,31831 2,0 0,02591 — 0,11538 2,4 0,3823 0,9959 0,50 0,50637 — 0,38879 6,0 — 0,08796
0,46 0,21518 —0,31258 2,5 0,01749 — 0,09459 2,5 0;1924 J;; . 0,9980 0,52 0,49368 — 0,360597,0 — 0,07425
0,48 0,20518 — 0,30695 3,0 0,01256 — 0,08001 UJ 1,0000 0,54 0,48135 — 0,3~20T8,0 — 0,06431
0,50 0,19587 — 0,30142 0,00944 — 0,06924 0,56 0,46936 — 0,393269,0 — 0,05677
0,62 0,18718 — 0,29600 ?0 0,00734 — 0,06099 0,58 0,45770 — 0,3941910,0 — 0,05084
0,17905 — 0,29071 4,5 0,0d587 — 0,05447 0,60 0,44637 — 0,3948712,0 — 0,04214
2: 0,17144 — 0,28554 0,00479 — 0,04920 0,70 0,39416 — 0,3951114,0 — 0,03602
0,58 0,16430 — 0,28049 1?0 0,00124 — 0,02489 0,80 0,34858 — 0,3013220,0 — 0,02510
0,60 0,15760 — 0,27557 20 0,00031 — 0,01249 0,90 0,30866 --0,38475 w 0
0,62 0,15130 — 0#7078 30 0,00014 — 0,00833
0,64 0,14537 — 0,26611 40 0,00008 --0,00625
0,66 0,13978 — 0,26156 0,00005 — 0,00500
0,68 0,13450 — 0,25714 .R 0,00001 — 0,00250
0,70 0,12952 — 0$5284 m 0 0
I-J
“
.w
CA


NACA - General Airfoil Theory

Uploaded by

Copyright:

Available Formats

NACA - General Airfoil Theory

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NACA - General Airfoil Theory

Uploaded by

Copyright:

Available Formats

,...— ...

...— ------- -. . -.. ..— —, .,

vol. 17’; No. 11/12, December 10,, 1940

Copyright National Aeronautics and Space Administration (NASA)

GENERAL AIRFOIL THEORY*

On the assumption of infinitely small disturbances

There are a number of airfoil theories which hold

Following the temporary interest attaching to the

Copyright National Aeronautics and Space Administration (NASA)

the vortex method is that ‘the [email protected],S,~.~..$~~,$.Y

is ,applicable for the velocity potential and the sound.

The setting-up of the integral equations of the air-

2. THE VELOCITY POTENTIAL Ol? THE ELEMENTARY RADIATOR

The wave equation ❑ @ = O has, as is known, a very

The airfoil which is to manifest a pressure differ-

Copyright National Aeronautics and Space Administration (NASA)

. NA,CA Technical Me.mor:an*dti-

perposition” of. radi~tors of the first order (so-called

The sound ,yressure of the. field of? sound is ,

hence it satisftes equations DXJ = O also because p-

3. THE LORENTZ TRAUSFORMATION .

The arguments sofar have dealt with the radiator

~.. = xt+’vt~ t-r ,+ v xt/c2

On passing to the coordinate x~ = Xl+ vtf in the

Acco”rdin& to (6) , the change to the moving reference

Copyright National Aeronautics and Space Administration (NASA)

Inasmuch as the airfoil is to be situated on a regu-

Later, the knowledge of the sound pressure of a mov-

(i- :)) (8)

According to Prandtl (reference 2) a new approach to

where ~ is the velo’oity vector.

Assuming frictionless, homogeneous fluid gives the

Copyright National Aeronautics and Space Administration (NASA)

q= - ~ -i- const (12)

On the airfoil, pressure p, and hence V, manifests

The acceleration potential CP can be approximated,

Through the superimposed doublets, the intensity of

the terms of the second order leing ignored.

posing Q = grad q, ~= grad @, followed by in-

6 I?AC4 Technical .Memo~apdum No. 979

Gtven Cp(x,y,z,t), the desired acceleration poten-

One” particular advantage ofthis method of represen-

An element d c of the surface covered with doublets

The analyzed element da lies”in the zero point of

5. “THE GENERAL INTEGRAL EQUATION OF AIRFOIL THEORY

The location of our elementary radiator is now shift-

x-g Y- Y(q) z - z(m) {17)

must be effected in (16).

Copyright National Aeronautics and Space Administration (NASA)

NACA !Jechnical Memorandtirn No. 979 ?

where angle a(q) between the direction of the normals

+(1- P2) L-(Y - Y(n).)2 + (z - z(d):]

On the other Band, the p,ressure jump TT on the air-

n(g,n,t) = pvY(5,n,t) “ (20)

‘“ Equations (1.9) and (20) represent the most general

Copyright National Aeronautics and Space Administration (NASA)

When solving (19), the Kutta flow-o,?f condition must

and 1 = l(n) , half the wing chord. For special pur-

It may be mentioned that the v’ortex.method also