Sei sulla pagina 1di 7

The Analytical Design of an Axially

Symmetric Laval Nozzle for a


Parallel and Uniform Jet
KUNO FOELSCH
Consulting

Physicist

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

ABSTRACT

The equations for the nozzle's contours are derived by integration of the characteristic equations of the axially symmetric flow.
Since it is not possible to integrate these equations mathematically in an exact form, it was necessary to find a way to approximate the calculations. The approximation offers itself by considering and comparing the conditions of the flow in a cone with
those in a nozzle, as a linearization of the characteristic equations.
The first part of the report deals with equations for the transition curve by which the conical source flow is converted into a
parallel stream of uniform velocity. The equations are derived
by integration along a Mach line of the flow in the region where
the conversion takes place. A factor / is introduced expressing a
relation between the direction and the velocity of the flow along
a certain Mach line. / remains undetermined and is not involved
in the final equations.
In the second part of the report, the spherical sonic flow section is converted into a plane circular section of the throat.
The nozzle's contour adjacent to the throat is formed by the arc
of a circle connected with the transition curve by a straight line.
The gas dynamic properties of the boundary Mach line are calculated in Table 1, the use of which shortens the calculations considerably.
SYMBOLS

M
a
w
wp
6
co
r
D
s
R
A
\p
ix

= Mach Number of the flow


Mach angle of the flow
= velocity
= mass flux of the flow
angle between direction of flow and axis of symmetry
= semicone angle
= radius vector from source point
= diameter of nozzle at exit
= radius of the throat area
= radius of the nozzle's wall curvature at the throat
= length of a straight part of nozzle's contour
= expansion angle of a Mach line
a small increment of Mp
= a small increment of dp
T = r/ro
L = length of the supersonic part of the nozzle
7 = ratio of specific heats; / and k constants along a Mach line
Subscripts
P refers all quantities to a variable point along the Mach curve
A fixed values at point A
E final values of all quantities at point E
0 = values at the critical arc in a cone flow

and

Aerodynamicist

often required for jet propulsion units. The design of a


two-dimensional nozzle (plane flow) has been treated
in the past by several authors. In 1929 Prandtl and
Busemann 1 worked out their famous graphical method,
which has been much in use since. Later, Puckett 2
modified their graphical method to a semigraphical
one. The author 3 of the present report proposed at the
J.P.L, Cal-Tech symposium in 1946 an analytical design method which is based on the fundamental ideas
used by Prandtl and Busemann in solving the problem
graphically.
However, the case of a three-dimensional nozzle
with axial symmetry represents a more difficult problem.
Up to date there are known in the literature only graphical design methods as proposed by Sauer,4 Guderley, 5
and de Haller. 6 All these proposed methods for designing an axially symmetric nozzle are, for practical
use, tedious and complicated. They require a great
deal of graphical work, since two charts are made and
used simultaneously and involve errors of uncertain
magnitude, particularly when the graphical construetion approaches the axis of symmetry.
It is the purpose of this report to show that a good
approximation for an analytical solution of the problem may be obtained. In other words, the coordinates
of the contour of an axially symmetric nozzle, as well
as of any streamline in the nozzle's flow, may be immediately determined from two simple equations.
The first part of the report deals with the derivation
of the equations for the transition curve by which the
conical source flow emanating from the sonic section
is converted into a parallel and uniform flow. In the
second part of the report the spherical sonic flow section
is adjusted to a plane flow section at the throat of a
nozzle.
PART I.

CONVERSION OF A RADIAL FLOW INTO A


PARALLEL AND UNIFORM FLOW

Fig. 1 shows a section through the upper supersonic


part of a three-dimensional nozzle with axial symmetry. The flow in the nozzle may be divided into three
regions as to the character of the flow.

INTRODUCTION

(I)
(II)

parallel and uniform flow


are a necessity for wind-tunnel testing and are

AVAL NOZZLES PRODUCING

(Ill)

Received July 6, 1948.


161

Region TREA: Radial flow.


Region AEB: Conversion of flow into parallel
stream by transition curve AB.
Region BEX: Parallel and uniform flow.

162

OF T H E

AERONAUTICAL

SCIENCES MARCH,

1949

locity vector along PK will not differ much from the


velocity vector at point P as to its direction and magnitude. This statement justifies the below performed
integration of Eqs. (2a) and (2b) in a linearized form.
For the analysis of Eq. (3) we make the assumption
that, along a continuous Mach line PK, the velocity w
is uniquely determined by the direction 0 of the flow.
FIG. 1. Conversion of radial flow into parallel flow.
Since we are dealing with an expansion and potential
flow and since the Mach line PK is continuously intersected by Mach lines of the second family, we conclude
Analysis of Flow in Region I
that, along PK, dw must be positive and dd either posiThe flow enters the nozzle at the arc TR with sonic tive or negative but not zero. However, along a streamvelocity and proceeds in radial streamlines until the line down the flow in Region II, both quantities apMach line AE is reached. The expansion of the flow proach zero, a value that is attained at the straight
along the Mach lines in that region is given by the Mach line BE.
equations [see Eq. (24)]
We consider for Eq. (3) the possibility of two extreme
cases with the actual case somewhat between
dd = =*= (dw/2w) cot a
(1)
them.
At point E the expansion is completed, and the flow is
The assumption that dd is positive along PK suggests
at the state the nozzle is designed for.
to consider Eq. (1) with the upper sign as one extreme
case for the differential form of Eq. (3). Substituting it
Analysis of Flow in Region II
into Eq. (2b), we obtain
Region II is bounded by the curved Mach line AE,
dy/y = (1/2) (dw/w) cot a (cot a + cot 0)
(4)
the straight one EB, and the transition curve AB.
A left Mach line PK, starting at any point P along AE, Eq. (4) indicates a fast expansion along the curve PK,
is described by the characteristics equations of an since it is the equation for a Mach line in a conical
axially symmetric potential flow:
source flow, as can be verified immediately from the
dy/dx = tan (0 + a)
'(2a) properties of the flow in a cone (Fig. 2).
\S

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

JOURNAL

/ -^_K

dy/y = [(dw/w) cot a dd] (cot a + cot 0) (2b)


The above equations represent two relations between
four quantities w, 0, y, and x, and the functions w{x),
6{x), and y(x) along a Mach line may be found only if
there is another equation stating, for example, a certain relation between w and 0 for a certain Mach line as
F(w, 0, eP) = 0

(3)

Substituting Eq. (3) into Eqs. (2a) and (2b), we obtain


the three above-mentioned functions by direct or stepby-step quadratures.
If the nozzle problem is solved by the methods of
characteristics, then there is no need of Eq. (3), since
the geometric and gas-dynamic properties of a Mach
line are established by using all four characteristic
equations simultaneously in a slow graphical integration process.
Before we analyze Eq. (3), let us first consider the
flow in the nozzle after passing the reflected Mach line
AE and compare it with the flow in a cone. At point
A the straight cone wall has been bent into the nozzle
wall AB (transition curve) by which the radial streamlines at AE are being turned parallel. The magnitude
of the velocity that would have been acquired by the
flow in a cone along the circular arc LE will be attained
by the flow in the nozzle only after passing the line EB.
I t is obvious that the process of expansion in the nozzle
has been considerably slowed down and that the ve-

y = r sin 0,

dr/rdd = cot a

The assumption that dd is negative along PK leads to


the other extreme case for Eq. (3):
dd = -(1/2) (dw/w) cot a

(la)

Eq. (la) shows a much slower expansion along PK than


in the first case, as can be seen by substituting it into
Eq. (2b):
dy/y = (3/2) (dw/w) cot a (cot a + cot 0)

(5)

It is of interest to notice that Eq. (la) represents the


characteristic equation for the curve AE, and, by applying it to the curve PK as the differential form of Eq,
(3), each point of PK is given an image point along PE
as to the magnitude and direction of the vector w.
If K\ is the corresponding point to K (Fig. 1), the
entire curve PK is gas-dynamically reflected in Curve

FIG. 2. Mach line PK in a conical source flow.

AXIALLY SYMMETRIC
PKu and, further, each point of the transition
curve A3 has gas-dynamically an image point along
AE.
We assume now that Eq. (3) has the differential
form
dO = k(dw/w) cot a

(6)

with +(1/2) > k > - ( 1 / 2 ) , and, by substituting it


into Eq. (2b),

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

dy/y = f(dw/w) cot a (cot a + cot 0)

(7)

with (3/2) > / > (1/2). / is assumed to be constant


along a Mach line but may vary from Mach line to
Mach line as a function of the boundary conditions along
AE.

LAVAL

dy _
y ~f

[(M2 - 1) + VM2 - 1 cot 6]dM


M{1+ [ ( 7 - 1)/2]M2}

( a)

wp = (wp)o M / + ^ M 2
(12)
\7 + 1 7 + 1
/
Substituting Eqs. (9), (10), and (12) into Eq. (11) and
observing the rule we have set up for Eq. (8), the integration of Eq. (11) gives
r02(cos Op cos co) =
2fA(M-MP)
[yp

= dP + e

AI,

(8)

and consider p. and e as small quantities so that their


higher powers and products p2, e2, pe . . . may be neglected in our calculations. Observing that rule for
the integration of Eq. (7a) from point P to any arbi-r
trary point N on the curve PK, we obtain
=

yP^M~Mp)

A =

VMP

- i (VMP

MP{1+

- I + cot eP)

[(7-

1)/2].MP2}

A line element drj of the curve PK is


drj = Jv/sin (a + 0)
or with Eq. (9)
dri = fAy(MdM/cos

0+ W

sin
i
- " dp(VMp

- 1

- 1 + cot dp)rP2

/
Tp'

V 1

\ <T + l)/2(7-D

\7 + 1

7+1

- 1 sin 0) (10)

To compute the y coordinate for the intersection point


K of the Mach curve PK and the transition curve AB,
we apply to the flow in Region II the law of conservation of mass, which states that the mass flow that passes
the surface of the spherical zone at critical section between the angles co and 0P and with the radius r0 must
also pass the surface formed by rotation of PK about
the X axis.
2ir(wp)tf<? I sin 0 dd = 2TT / (wp) sin ay drj (11)
J eP
Jp
The above equation determines also the y coordinate

(13)

rP has a geometrical meaning. It represents the ratio


of the radius at point P to the radius r0 of the critical
arc (reference 4, Eq. 33).
rP = rP/n

(14)

yP = r P sin dp

(15)

With Eqs. (14) and (15), Eq, (13) is simplified to


^ pyj
efA(M-MP) s[ndp

(16)

with
F(dP) =

V s i n 0P + 2(cos 0P - cos co) ( VMP

- 1 sin 0P + cos 0P)

(16a)

(9)

with
2

with

Since it was shown above that the values of M and 0


at any point of curve PK do not differ much from those Furthermore,
of MP and 0P, respectively, at point P , we set
M = MP +

163

for any streamline if co and K are changed correspondingly.


The mass flux (wp) at point N may be expressed by
the Mach Number as (reference 4, Eq. 28):

Calculation of the Coordinates of the Transition Curve


Expressing the velocity w and the Mach angle by
the Mach number M of the flow, Eq. (7) is

NOZZLE

and by Eqs. (9), (14), and (15) the y coordinate of point


Kis
y = r0TPF(dp)

(17)

r0, the radius of the critical arc TRt is determined by


applying the law of conservation of mass to the exit of
the nozzle with the diameter D (Fig. 1). The flow at
the exit is gas-dynamically of the same state as in point
E.
4 W sin2 (co/2)(*e;p)o =

(TTD 2 /4) (Wp)E

(18)

and by Eqs. (12) and (13a)


(wp)E = (^p)o(l/r^ 2 )
Substituting-it into Eq. (18), we obtain
r0 = D/4:TE sin (co/2) -

(19)

which transforms Eq. (17) into


y = [D/4t sin (U/2)](TP/TB)F(6P)

(20)

y, the ordinate of the transition curve, has been obtained as a function of 0P, since rP and MP are deter-

164

J O U R N A L

OF

T H E

A E R O N A U T I C A L

S C I E N C E S MARCH,

1949

TABLE 1
P R O P E R T I E S OF BOUNDARY M A C H L I N E AE

( F I G . 1)

y = 1.40

V^ 2 - i

oTi

0.00788
0.06188
0.2021
0.4581
0.84729
1.09387
1.3749
1.68968
2.0369
2.41491
2.82200
3.25612
3.7151
4.19662
4.6987
5.21910
5.75559
6.30609
6.86885
7.44175
8.02319
8.61138
9.20503
9.80251
10.4026
11.0041
11.6059
12.2072
12.8068
13.4039
13.9980
14.5883
15.1743
15.7553
16.3311
16.9011
17.4651
18.0228
18.5737
19.1178
19.6549
20.1848
20.7073
21.2225
21.7303
22.2306
22.7233

0.2
0.3
0.4
0.5

0.55

0.6
0.65

0.7
0.75

0.8
0.85

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

0.9
0.95
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
1.55
1.60
1.65
1.70
1.75
1.80
1.85
1.90
1.95
2.00
2.05
2.10
2.15
2.20
2.25
2.30
2.35
2.40
2.45
2.50
2.55
2.60

VM2

1.0000
1.0002
1.0008
1.0024
1.0055
1.0077
1.0106
1.0141
1.0182
1.0231
1.0288
1.0352
1.0425
1.0506
1.05965
1.0696
1.0804
1.0921
1.1048
1.1184
1.1330
1.1485
1.1650
1.1824
1.2008
1.2202
1.2406
1.2619
1.2842
1.3075
1.3318
1.3571
1.3834
1.4106
1.4389
1.4682
1.4985
1.5298
1.5621
1.5955
1.6299
1.6653
1.7017
1.7393
1.7778
1.8175
1.8582

2.65
2.70
2.75
2.80
2.85
2.90
2.95
3.00
3.05
3.10
3.15
3.20
3.25
3.30
3.35
3.40
3.45
3.50
3.55
3.60
3.65
3.70
3.75
3.80
3.85
3.90
3.95
4.00
4.05
4.10
4.15
4.20
4.25
4.30
4.35
4.40
4.45
4.50
4.55
4.60
4.65
4.70
4.75
4.80
4.85
4.90

mined by the conical source flow (see Table 1). co and


TE are constants; they will be analyzed below.
The x coordinate of point K is computed from Eq.
(2a) in the following way:
dM
dy
= fAy
tan
(6
+ a)
tan (0 + a)

dx

efA(M-MP)

I dx = fAyP I

JMMP

J*p

xP = yP

VMP2

1 -

tan (0 + a)
tan dp
2

1 + tan dpVMp

dM

rjA{M-MP)_

1]

- 1

and by using Eqs. (9), (14), (15), (16), (19), and


xP yP cot Op
we obtain the x equation as
x = [Z)/4sin (U/2)](TP/TE)

X
2

1 + (cos SpVMp
sin 6PVMP2

- 1 - sin eP)F(dp)
-

1 +

cos dP

(21)

- i
~"~

23.2085
23.6861
24.1563
24.6189
25.0742
25.5221
25.9626
26.3959
26.8220
27.2411
27.6532
28.0584
28.4568
28.8480
29.2336
29.6122
29.9814
30.3503
30.7100
31.0638
31.4115
31.7534
32.0894
32.4198
32.7449
33.0644
33.3786
33.6876
33.9914
34.2903
34.5841
34.8732
35.1576
35.4373
35.7125
35.9832
36.2497
36.5117
36.7696
37.0234
37.2733
37.5192
37.7611
37.9994
38.2339
38.4648

r
1.8999
1.9428
1.9867
2.0317
2.0778
2.1251
2.1734
2.2229
2.2734
2.3251
2.3780
2.4320
2.4871
2.5434
2.6009
2.6596
2.7194
2.7804
2.8427
2.9061
2.9707
3.0366
3.1037
3.1720
3.2462
3.3124
3.3844
3.4577
3.5224
3.6082
3.6854
3.7639
3.8436
3.9247
4.0071
4.0907
4.1758
4.2621
4.3498
4.4389
4.5293
4.6211
4.7142
4.8106
4.9047
5.0020

Eqs. (20) and (21) determine the transition curve by


which a supersonic conical flow emanating from the
spherical surface TR is converted into a parallel and
uniform jet. We notice that the unknown factor /
in Eq. (7) is not involved in the final equations for x
and y.
The parameter 8P has values from zero to co; TP and
1 are determined from these values as will be
shown later. For 6P 0 we obtain from Eqs. (20) and
(21) with
[F(0p)]ep>=Q ~ 2 sin (w/2) and TP = rE
Ly]*,-o = D/2
D
D
[x]6p^ =
.
.
+
cot aE
f
4 sin (co/2) 2
which is in agreement with the geometrical picture
of the nozzle, since, according to Eqs. (14) and
(19),
OE = rB = rEr0 = -D/4 sin (co/2)

AXIALLY SYMMETRIC

LAVAL

165

NOZZLE

"The possible shortest nozzle is obtained from


(27)

% ^-^NOA

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

FIG.

3.

Definition of an expansion angle in a conical source


flow.

which, gas-dynamically interpreted, means that the


boundary Mach line starts at the intersection point of
the critical arc and the axis of symmetry and is reflected once at the wall, at point A> as it was assumed
in Fig. 1. The values of \p and r computed from
Eqs. (25) and (13a), respectively (for y = 1.40), are
given in Table 1 as function of VM2 - i, which is a
more convenient term than M since it is involved in
Eqs. (20) and (21).

Properties of a Conical Source Flow, Relation Between


M, r, 0, w, and $
In a conical source flow a Mach line starting at any
point of the critical arc (point N, Fig. 3) runs across the
flow and is reflected from the walls. A point moving
along the Mach line from the start of it to any arbitrary
position will have covered a certain angle, which we call
the expansion angle of the Mach line (at the position
of the point) and denote it by \p as it is shown in Fig. 3
for point P (see also reference 3). The expansion angle
for a certain point of a Mach line can be expressed by
the Mach Number. The logarithmic differentiation of
Eq. (13a) gives
dr
r

(M2 - 1) dM
2M{l + [(7 ~ l)/2]M2}

PART II.

It is usually assumed that the flow section at the


throat (M = 1) of a nozzle is plane and not spherically
curved as at the critical section of a conical source flow.
The conversion of the spherical flow section into a
plane circular flow section may be accomplished by
bending the portion of nozzle wall adjacent to the
throat into any smooth convex curve QTD tangent to
the rest of the wall at D and having at the throat a
tangent parallel to the axis of symmetry (Fig. 4).
For continuity reason the cross-section area at the
throat must be equal to the area of the spherical section.
ITS2

= dr/r #

= 47rro2 sin2 (co/2)

(22)

or by Eq. (19)

from Fig. 3 and Eq. (14) it is


dr/r #

PLANE FLOW SECTION AT THROAT

$ 2

= VM

- 1

D/2TE

(28)

(23)

The change in the geometry of the sonic flow section


will certainly affect the source character of the flow in
By substitution of Eq. (23) into Eq. (22),
Region I of the nozzle. However, by inserting a straight
dip __
VM2
- 1
(24) line into the nozzle contour ahead of the transition
dM~.2M{l
+ [(T - l)/2]Af 2 }
curve, the source character of the flow may be restored
and
secured in the portion of the nozzle adjacent to
and by integration of Eq. (24) along a Mach line from
the
Mach
curve AE. This is a necessity for our nozzle
the start (\p = 0, M = 1) to an arbitrary point with the
design,
since
the calculation of the transition curve was
Mach Number M,
based on the existence of a radial flow along the Mach
curve AE, We now refer all coordinates to the new
7 - 1
(ikf2 - 1) origin 0i, at the center of the throat and choose for the
7+ 1
*

[V^T--V

tan-1 VM% - 1

(25)

It is, by definition of \[/ along the boundary Mach line


AE of the flow in Region I (Fig. 3),
6P = \pE \pP

(26)

\pE is the final (maximum) expansion angle reached by


the flow at point E and is computed from Eq. (25) for
MEi the design Mach Number of the nozzle. \pP varies
along AE from $A to \}/E. \J/A determines the length of
the nozzle and is chosen by the designer from the relation
FIG.

4.

Conversion of a spherical flow section into a plane flow


section at the throat.

166

JOURNAL

OF T H E

AERONAUTICAL

curve QTD the arc of a circle having radius R and the


center at O2 (Fig. 4). The equation of the circle is
- (x/R)2

y = s + R - RVl

(29)

and the coordinates of the point D are


xD = R sin co;

yD = 5 + i<!(l cos co) (30)

Through the transformation of the coordinate origin


from 0 to 0i, the x values for the transition curve must
be reduced by x0j which, derived from Eqs. (29) and
(28), is

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

Xo = (D/2rE) cot co - R tan (co/2)

(3D

The y values of Eq. (20) are not affected by the above


transformation.
It can be shown that between R and the length X
of the straight part DA of the nozzles contour exists
the relation
X + R tan - =
2
in (co/2) LA
4rE sin

cos (cc/2)J

(32)

R must be so chosen by the designer that X will stay


positive. If we set R = X, X and R are then calculated
from
R =X

rA cos (co/2) 1
>
(33)
4:TE sin (co/2) cos (co/2) + sin (co/2)

Given: Mach Number ME and diameter D.


(1) Determine the fundamental constants \{/E, rB
from Table 1 or by use of Eqs. (25) and (13a). Accuracy of one ten-thousandth of unity is recommended.
(2) Determine the third constant co from

co = y 2 f e = oA = fa
Determine rA and

rA cos (co/2) 1
co/2)]/
cos (co/2)]

{cot co 2 cos (co/2) [sin (w/2) +

(34)

The length of the supersonic part of the nozzle can be


found from
D
2

All equations derived above for the nozzle design may


be divided by the diameter D of the nozzle, since D
is involved in them only as a first power multiplication
factor. As dimensionless quantities these functions may
be used then for a nozzle of a certain Mach Number ME
and of any diameter.
The change of M and 6 along a Mach line from point
P to point K may be calculated from Eqs. (16), (6), and
(7) as
M - MP =

VMA2

1 from Table 1 for fA =

(3) Calculate the coordinates of the transition


curve from Eqs. (20) and (21) by varying VMP2 - 1
from VMA2
- 1 up to Vikftf2 - 1; use Table 1. 0 P is
determined by
eP = \l/E ypp

is found from Table 1 for VMP2


- 1.
(4) Calculate x0 by Eq. (34) and subtract it from all
x values obtained in 3.
(5) Calculate X and R from Eq. (33) and determine
the coordinates of point D (Fig. 4) by using Eq. (30).
X/SP

APPENDIX

Eq. (20) may be obtained in a shorter and mathematically more general way than in the above analysis.
The only assumption we make here is that the ordinate
y of the Mach curve PK is uniquely given by
(36)

y = yMM, Op)

D
4 sin (co/2)

1949

SUMMARY OF STEPS IN ANALYTICAL DESIGN OF A T H R E E DIMENSIONAL LAVAL NOZZLE WITH AXIAL SYMMETRY

Eq. (33) substituted into Eq. (31) gives


xo = D/2TE

SCIENCES MARCH,

M changes along the curve, but the way it changes will


remain undetermined. We are concerned here only
with the values of the function in Eq. (36) at the
boundary points P and K. It is obvious that for point
P
<j>{MPi dp) = 1
For point K the value of the function $(MK, 6P) is
given by the continuity equation as applied in Eq. (11).
Substituting into Eq. (11) the line element
<!>'{My6p)dM*
sin (6 + a)
and performing the integration in the same manner as
above, we have
4>*(MK,ft.)- 1 =
2(cosflp - cosco) [sin 6PVMP2
sin2 dp

- 1+

COS 6P]

log F(6P) - log sin 6P


or
4>(MK, Op) = F(6p)/sm 6P

resp.

log sin dp

*- m^m

+ cot dp

(37)

<' is the first derivative of <f> with respect to the argument

M.
(Continued on page 188)

188

JOURNAL

OF

THE

A E R O N A U T I C A L

In the case of flat-bottom planing surfaces and hulls,


errors can be kept to a minimum by a judicious choice
of the factors rj and f.
I t is believed that there are numerous engineering
problems connected with low aspect ratio lifting surfaces in which the errors described above would be
tolerated in exchange for the ease of calculation.

S C I E N C E S MARCH,

1949

10
Tsien, Hsue-Shen, Lower Buckling Load in the Non-linear
Buckling Theory for Thin Shells, Quarterly of Applied Mathematics, Vol. 5, No. 2, p . 236, July, 1947.
11
Wagner, H., Remarks on Airplane Struts and Girders Under
Compressive and Bending Stresses;
Index Values, N.A.C.A.
T.M. No. 500, 1929.

CONCLUDING REMARKS

Downloaded by VIRGINIA TECHNICAL UNIVERSITY on November 12, 2014 | http://arc.aiaa.org | DOI: 10.2514/8.11758

On the basis of this empirical study, it is concluded


that the lift of low aspect ratio airfoils, planing surfaces,
and hulls may be represented by a simple power function of the form:
CL = v K an
where K and n are functions only of the aspect ratio
and are shown graphically in Figs. 4 and 5. y is a
function only of the type of lifting surface and, if
assumed to be unity for airfoils, becomes 2/3 for hulls
and 1/2 for flat-bottom planing surfaces.
This simple relation should be a useful engineering
tool for making quick estimates. Much work remains
to be done to obtain reasonably accurate means for
determining the lift of various kinds of planing surfaces
and different type of yawed hulls. It is hoped that this
study may provide a foothold on which to expand.
REFERENCES
1

Weinig, F., Lift and Drag of Wings with Small Span,


N.A.C.A. T.M. No. 1151,1947.
2
Bollay, W., A Contribution to the Theory of Planing Surfaces,
p. 474 of the Proceedings of Fifth International Congress of Applied Mechanics, 1938.
3
Winter, H., Flow Phenomena on Airfoils and Plates of Short
Span, N.A.C.A. T.M. No. 798, 1936.
4
Zimmerman, C. H., Characteristics of Clark Y Airfoils of
Small Aspect Ratios, N.A.C.A. T.R. No. 431, 1932.
5
Yates, C. C , and Riebe, J. M., Effect of Length-Beam Ratio
on the Aerodynamic Characteristics of Flying Boat Hulls, N.A.C.A.
T.N. No. 1305, 1947.
6
Sottprf, W., Experiments with Planing Surfaces, N.A.C.A.
T.M. No. 661,1934.
7
King, D. A., and Hill, Mary B., General Tank Tests of a Vio
Size Model of the Hull of the Boeing XPBB-1
Flying Boat,
N.A.C.A. T.N. No. 1057, 1946.
8
Parkinson, J. B., Tank Tests to Show the Effect of Rivet Heads
on the Water Performance of a Seaplane Float, N.A.C.A. T.N. No.
657, 1938.
9
Davidson, K. S. M., Some Experimental Studies of the Sailing
Yacht, Transactions of the Society of Naval Architects and
Marine Engineers, 1937.

Axially Symmetric Laval Nozzle


{Continued from page 166)

Eq. (37), substituted into Eq. (36) and by use of Eqs.


(14), (15), and (19), gives
y = [D/4 sin ^/2)]rP/rEF{dP)

(20)

The exact analytical form of Eq. (36) is not essential


for computing the transition curve AB. This appears
to be strange, but it is explained by the fact that the
mass flow equation was integrated along PK in a linearized form. '
' REFERENCES
1

Prandtl, L., and Busemann, A., Stodola Festschrift, p. 499;


Zurich, 1929.
2
Puckett, A. E., Supersonic Nozzle Design, Journal of Applied
Mechanics, Vol. 13, No. 4, pp. 265-270, December, 1946.
3
Foelsch, K., A New Method of Designing Two Dimensional
Laval Nozzles for a Parallel and Uniform Jet, Abstracts of Papers
Presented at the Guided Missiles and Upper Atmosphere Symposium, J P L G A L C I T , California Institute of Technology,
Publication No. 3, pp. 156-169, March, 1946.
4
Sauer, R., Theoretische Einfuehrung in die
Gasdynamik;
Julius Springer, Berlin, 1943 (reprinted by J. W. Edwards, Ann
Arbor, Mich.).
8
Guderley, G., Die Charakteristischen Methode fuer ebene und
achsensymmetrische
Ueberschallstroemungen, Jahrbuch 1940 d.
deutschen Luftfahrtforschung, pp. 1522-1535.
s
De Haller, V., On a Graphical Method of Gasdynamics, Technische Rundschau, Sulzer. No. 1, pp. 6-24, 1945.

Recent N.A.C.A. Research on High


Length-Beam Ratio Hulls
{Continued from page 183)
5

Principles of Structural Design


for Minimum Weight
{Continued from page 149)
9
Tsien, Hsue-Shen, A Theory for the Buckling of Thin Shells,
Journal of the Aeronautical Sciences, Vol. 9, No. 10, p. 373, August,
1942.

Davidson, Kenneth S. M., and Locke, F. W. S., Jr., Some


Systematic Model Experiments on the Porpoising Characteristics
of Flying-Boat Hulls, N.A.C.A., A.R.R. No. 3F12, 1943.
6
Bell, Joe W., and Willis, John M., Jr., The Effects of Angle of
Dead Rise and Angle of Afterbody Keel on the Resistance of a
Flying-Boat Hull, N.A.C.A., A.R.R., February, 1943.
7
Olson, Roland E., and Land, Norman S., Effect of Afterbody
Length and Keel Angle on Minimum Depth of Step for Landing
Stability and on Take-Off Stability of a Flying Boat, N.A.C.A.
T.N. No. 1571, 1948.

Potrebbero piacerti anche