https://ntrs.nasa.gov/search.jsp?

R=19930083341 2018-10-08T15:03:28+00:00Z

FOR AERONAUTICS i-

TECHNICAL NOTE 2579

1
APPROXIMATE METHODS FOR CALCULATING THE FLOW

ABOUT NONLIFTING BODIES OF REVOLUTION AT 1

1
HIGH SUPERSONIC AIRSPEEDS \

By A. J. Eggers, Jr. , and Raymond C. Savin

Ames Aeronautical Laboratory

Moff ett Field, Calif.

,
1

Washington
December 1951 I
I
I

,---- .- —- — . .__. .. . .. .- ...- —— .-..-. . .. . . ...=- ——.4

 (‘—”iECHLIBRARY
.—— ~FBI ‘M
1

: lllBIBI1l’lluNllllB
oOb553b
._—

NA!HONAL ADVISORY COMMITTEE FOR AERONAUTICS ‘

,TECHNICAL NOTE 2579

APPROXIMATE METEKIDSFOR CALCULMUNG TEE FLOW AEJ3UTNONLIFTING

BODIES OF REVOLUTION AT HIGH SUPERSONIC ~

By A. J. Eggers, Jr., and Raymond C. Savin

SUMMARY

Flow at high supersonic speeds shout a body of revolution is

investigated analytically. With the assumption that the flow at the
vertex is conical, it is found that algebraic solutions can be obtained
which yield the Mach numbers and pressures at the surface over a con-
siderable range of free-stream Mach nmibers and apex angles. In the
special case of cones, these solutions define the entire flow field with
good accuracy, and may therefwe provide a useful adjunct to the well-
knmm M.I.T. tables.
.
The investigation of flow downstream of the vertex reveals that
when the value of the hypersonic similarity parameter for the flow
(i.e., the ratio of the free-stream I&h number to the slenderness ratio
of the body) is large compared to 1, the Mach number along a streamline
(downdmeamof the nose shock) varies with flow inclinat~on angle in
approximately the same mamner as for two-dimensional (Prandtl-Meyer)
flow. In the special case of streanddnes near the surface, it is wg-
gested that this parameter may approach 1. This result and the soltitions
obtained for flow at the vertex are combined to yield what might be
ca12ed a conical-shock-expansionmethod for calculating the Mach nuuiber
and pressure distributions at the surface of a body. T&se calculations
are shuwn to be Tarticulsrly simple in the case of slender bodies.

Surface Mach nmiber and pressure distributions calculated with the

simplified methods of this paper for a mmber of ogives are found to be
in good agreement with those obtained ‘withthe method of characteristics
, at values of the hypersonic similari@ pammeter greater than 1. In the
case of the conical-shock-expansioncalculations, the agreement is within
the order of accuracy of the characteristics solutions When the hyper-
sonic similarity ‘parameterhas a value of only 2. Eecause of the rela-
tive simplicity of these calculations, the methods for detemdning the “
flow at the surface of a body may prove useful.for engineering purposes.

.—-. — ..—. ..-. ——— — .—— --- - -- ——---. — —..—-. —.—..— ...- .
 2 IiACATN 2579

INTRODUCTION

Determination of supersonic flow fields about nonlifting bodies of

revolution by means of the method of characteristics (see, e.g., refer-
ence 1) is generally accepted as an accurate but tedious and time-
consuming operation. Because of the latter features of this method,
recourse is often made to simpler methods which, although less accurate,
can be applied with relative simplicity and rapidi~.
I
Perhaps the most widely used simplified theory of axially symmetric
supersonic flow is the linear theory, which was ftist employed by K&m&n
and Moore (reference 2) to study supersonic flow about a body of revolu- I
tion. This theory has a limited range of applicabil.i~, however, due to
the assumption in its development of potential flow with infinitesimal
disturbances. In particular, it is only applicable with good accuracy
to bodies of practical.slenderness ratios operating at low supersonic
airspeeds.

More recentlyj Van Dyke (reference 3) has ‘developeda second+rder

supersonic flow theory which yields results which sre generally more
accurate than those obtained with the linear or first-order theory,
altho@ the calculations are, of course, more lengt~. The range of
applicability of Van Dyke?s second-order solution is specifically
4
limited, however, to cases for which the product of the free-stream
Mach number and maximw slope of the body is less than 1; hence, it is
not generally suitable for calculating the flow about bodies of usufil
proportions at high supersonic, or hypersonic,airspeeds.

Flow at hypersonic atispeeds in the limit as the free-stream Mach

number approaches infinity and the ratio of specific heats of the gas
downstream of -a shock wave approaches 1 has been studied by Busemann
(reference 4). It was found that under these circumstances a very sbple
expression is obtained for the pressures actihg on a nonlifting body of
revolution. This eqmession provides in general, however, only qualita-
tively accurate pressure distribtiio~ at high but finite free-stream
Ifachnumiberswhere the ratio of specific heats is ~eater than 1 and is
usually closer to the ideal diatomic gas value of 1.4.

Perhaps the first step in the direction toward providing simplified “

methods for determining axially sxtric flows in the high supersonic
speed range was made by Tsien (reference 5). Tsien demonstrated that a
similarity law (which is analogous to the wen-kno~ la.wapplicable at
low supersonic speeds) exists for slender pointed bodies in this range.
Thus, with the aid of this law, the flow about a fsmily of shapes alJ
having the ssme thickness distribution cam be easily determined, provided
the ratio of the free-stream Mach nuniberto the slenderness ratio is the
same for all shapes, and provided the flow about orieof these shapes is
known. This result materially reduces the net labor associated with

————— ——c .—-— ...—.

 NACA TN 2579 3

csJ.culatingflows about a number of related bodies of revolution opera-

ting at high supersonic airspeeds; however, the probl= r-m of
calculating the flow about a representative body for each value of the
shilsrity parameter. At present, this problem can onlybe solved with
good accuracyby means of the method of characteristics.

There appesxs, then, to be a need for a simplified theory which

provides, with engtieering accuracy, the Mach nuniberand pressure distri-
butions a~outpointed nonlifting bodies of revolution operating athigh
supersonic airspeeds. The development of such a theory is undertaken
in this re~ort, and it is found that by treating the flow in two parts,
namely, the flow at the vertex and the flow downstream of the vertex, a
theory of the desired s~licity and accuracy can, in fact, be obtatied.
In the specisl case of flow about cones, the theory demonstrates sur-
prising accuracy over a considerable range of supersonic airspeeds.

EwME!ms

c1 first fmlily cliaracteristicline

C2 second family characteristic line

d maximum diameter of a body of revolution

E entropy .

K hypersonic similari* parameter + .

()
2 characteristic body length (measured from vertex to most forward
point of maximum diameter)

M Mach number (ratio of local veloci~ to local veloci~ of sound)

n distance measured normal to ray passing through vertex of cone

s
P static pressure

9. -C pres-bme ‘

P yressm coefficient l?-PQ

() %3’

R gas constant

r radial distance from vertex of the cone

.
s distance along a streamline

---- .--- ———— —-— — ——-. -— —--- ——-- . ..—. .

 4 “WA I’IIT
2579 I

0 ,

U,v velocity components parallel and normal, respectively, to ray

passing through vertex of cone

.e.titit .e,ociW(Z)
msximumveloci~ obtainable by expanding to zero temperature

X,y I

Mach angle
‘(=C ‘+)
ratio of specific heat at constant pressure to specific heat at
constant volume

angle of flow inclination with respect to the body axis

i
P mass density

U angle between axis of cone and ray passing through vertex of cone

Subscripts ‘
L I

o free-stream conditions
1
,,

c conditions on the surface of a cone

N conditions on the surface at the vertex of a body i

s conditions immediately behind the shock wave at the vertex of a

body

ANALYSIS OF ltLOWAI!OUTA NONIXFTING I?ODYOF REVOLUTION

,
This investigation is concerned with a simplified method of calcu-
lating axia13y s~etric flow about a body of revolution traveling at I
high supersonic &speeds. It is assumed throughout the analysis that
the disturbed flow is everywhere supersonic, md thusj of co~se~ that
the body has a sharp nose or vertex. With these restrictions on the
free-stream Mach number and body shape, it iS e~dent t~t a tYPic~
flow.field will be characterized by the bow shock wave lying close to I
the surface of the body. For the purpose of analysis, it is convenient
to study the flow field in two parts, namely, the flow at the vertex
and the flow downstream of the vertex. The results of these two phases
of the investigation will then be applied .tothe determination of the
lkch number andyressure distributions over the body surface.

.
I
_——. _—. —.—— ——— ---- —..— ———---— . .. J
 \

Flow at the Vertex of a Body

It follows from the assumptions basic to this analysis that the

nose shock is attached, and co~sequently that the initial conditions of
the flow at the vertex will.be the same as for a cone tangent to the
body at the vertex and operating in the same free stream. In the classi-
cal paper of Taylor and Maccoll (reference 6) the basic differential
equations were developed for steady, axially symmetric supersonic flow
about cones.1 The essential assumption in their analysis was that fluid
~roperties were constant along radial lines passing through the vertex;
hence the flow was irrotational. This assumption wSll be employed here
to redevel~ these differential equations in a form more suitable for
this @ySiS.

A schematic dia@am of axially symmetric supersonic flow about a ‘

COIE iS shown ti fi~ 1. Considering ftrst the flow downstream of the
shock wave, the equations of motion and continui~ are written (after
reference 6) in the forms

(1)

and

&(pvsin U)+2pusinu. O (2)

respectively. The condition of irrotational.ityis given by the

expression

-v= &“ (3)

WI
Now it is evident from figure 1 that

u= v Cos (@a) (4)

and

v= -V sin (U-5) (5)

I
Thereforej equations (1) and (2) maybe combined to obtati the ~lation

lAS pointed out in reference 6, Busemnn had previously suggested a

graphical solution of the problem of supersonic flow past “acone.

I
,_ .- .--— ---’
. —-. -. . —-- - -. , -—.- —- -— - ——-——. .— . . . - —— —-.— ---
 6 NACA TN 257!3

~ ($2 -V2) V cos (w-5) l+% -sin (u-b)

[ ()

(
v Cotu+ ym
)1
+V2yu sin (u+) = o
(6)

\
Similarly, the &rotationali@ equation may be’written

dv
—=-tan((d+)m (7) I
v o
Combining equations (6) and (7), there results, then, the eqmtion of
motion I

which will be employed in the following development.

Equations (7) and (8) are smenable to numerical integration follow-

,
ing methods similar to those employed earlier in reference 6. In order
to obtain algebraic solutione to flows at the vertex, however, it is
necessary to simpl~ these equations. To this end, since high Mach
number flows are of principal interest, an assun@ion which has proved -1
useful in studying other aspects of these flows (see, e.g., reference 7)
is employed, namely, the magnitude of the resultant velocity vector is
essentially constant, the change in velocity being prharily one of
change in direction. With the aid of this assumption it is possible to
first determine a rehtion between 5 and (IIfor a given V using I
equation (8), and then to determbe more accurate the nature of the
dependence of V on 5 and u using equation (7 T . This procedure
will be employed for the follow5ng two cases: (1) slender cones for
which 5, the angle of flow inclination~ is smaSL compared to 1 radian
throughout the flow field, and (2) cones for which u+, the difference
between the ray angle and the inclination angle, is small compared to
1 radian throughout the conical part of the flow field.’

Analysis of cones for which 5<<1 slender cone

restriction 3mposed upon 5 in this study, equation maybe reduced
to the form .

2~ gemr~> ‘+ will be small for blunt cones; however, at Mach num-

bers very large compared
— to 1.
. w-5 will be small for relatively
slender cones as well.
1
I

.— .— -. —
.—.——
 NACATN257’9 7

5 cotu+ QQJ1-M%in2w (1% cot w)2]=0 (9)

where M is considered constant. This relation is still nonlinear in

both 5 and w , and is not readily smenable to exact algebmic solution.
An ayyroximate solution may be obtained, however, in the following man-
ner. Near the surface of the cone, equation (9) reduces to the linear
equations

which has the solution (10)

Now 5 cot u decreases rapidly with increasing w away from the cone
surface; hence it is suggested that without appreciable loss of ac=uracy$
the solution for 6 (near the surface) givenby this equation csm be
substituted into the coefficient of dS/dw in equation (9). Performing
this operation, equation (9) becomes

bcotu+ ~[1- M’%in2W (l-kl CSC m COtw)2]= O (IL)

which is linear in ?5 and can be integrated to yield (to the order of

accuracy of this anal.Ysis)
1
2M2(sin%c- sin%) */W412sin%c
1+
1 +~44M2sin2wc
6
—= (12)
5C 2M*(sin%c- Sinzw)
1+
L] 1- +4M%in2wc

which satisfies the boundary condition at the surface of the cone

(i.e., 5 =5C at w =LOc).

Since M was assumed constant throughout the conical part of the

flow field, it can convenientlybe taken as the Mach number.just down-
stream of the shock. In this case, flow conditions at the shock
(i.e., Ms, us, amd 86) are obtained by the sjnmltaneous solution of
equation (12) and the oblique shoti”-waveequations

. 31t is clear that cot w could be replaced with l/u in this equation;
however, the trigonometric operator is retained for consistency with
the rest of the analysis in which such operators must, in general, be
,> retained (rJ.I
is restricted to be small only near the surface of the
cone).

f -——. - . ..—— .- . .
. . ...-. ..— —— ---.—— --—- ------— —-
 I
8 NACA TN 2579
I

%2= (7+l)2M04sin2%- 4(~2sin~-1) (7~2sin2~+ 1)

(13)
[27 ~2sin2ms- (7-1)] [(Y-1) ~2sin%~ + 2]

and

cot bls(M&sin%ds -1)7

8+an~7+1 (14)
yM02-(Mo2sin2fds
-1)
[ ~
I
Having determined the flow conditions at the shock, the variation of 5
with W throughout the entire flow field about a cone is known from I
equation (12).

It now remains to more accurately determine the magnitude of the

velocity throughout the conical part of the flow field. For this pur-
pose, equation (7) must be employed in cotitiation ~th ~ -ression 4
relating (L)and 5. Equation (12) provides such a relationship; how- I
ever, it is unnecessarily complicated for detem~ the smal-1-
ch~ges
in v. A much simpler expression of acceptable accuracy is obtainedby 1
neglecting 8 cot ; in the coefficient of db/du appearing in equa-
tion (9),4 thus yielding ,

~= -5 Cotw
l-M’%n% (15) . I

or

5 = k#&c%M= (16) I
I

In this equation, k2 and M2 are determined by the requirements that

I
5C at w=uc

I
8 = 5S at u=us I

{
where the latter quantities are known from the previous analysis. Sub- 1
stitution of equation (16) into equation (7) provides the fofiowing I
relation for V; namely, I
I

4As will be shown later, in the cases where this simplifying assumption .
introduces significant error in the variation of V with (JI(viz.,
when M2sin% becomes of the order of 1) the results of the subse-
quent analysis of cones for which u-5<<1 maybe applied with good v I
accuracy.
1,

1
—...———— —. — --- .— -—— .—. J
?H NACA TN 2579 9

dV -kfl k22
—=
v
+ 1+
/52-k22(l-M2) [ 52-k22(l-M2) 1 bdb (17)

which can be integrated to yield

lnV= -~ln
[
6+Y6-)
1
where ka is determined by the requirement that V = V~ when 5 = as.
Replacing the constants in this expression with their values given by
the imposed boundary conditions, there is then’obtained the relation

[J ‘
5s cotzwc-cot%l~ + ~8c%s2) cot%~ --2J cot~uc-:ot*M*

b Cotzwc-cotzus + fiba%~z) Cot=wc +(5C2 -52)cot=w~

J (19)

Knowing the velocity, the Mach number may, of course; be detetined from “
the relation

M2 . ()
(20)
1 - ;%;&)=- .]

and the pressure coefficient anywhere in the flow field may be obtained
with the aid of the e~ression

(21)

s
.. . .. . . . . . -..—. . . . . . .--. .— .— - ..— — ———. -..—- .-. —–. —-- .– ——-.. -.--.——- ——-—.—— -——-.—-—- -
 10 NACA TN 2579

where

P* 27 ~’ Si112@~-(7-1)
.— = (22)
7+1

and

(23)

The Mch number and pressure distributions (as well as the orientation
of the conical shock) throughout the flow field about a slender cone are
now known. ‘Iheflow field about a cone for which u~ is small compared
to lwill be considered next.

haly-sis of cones for which u-5<<1.- In this case, equation (8)

maybe reduced to the form

1 - (W-5) cot m+ ~ [1 -M2(w-5)2] =0 (24)

4

where M is again considered constant. This nonltiear relation will be

solved in a manner analagous to that employed in the solution of equa-
tion (9). Thus, near the surface of the cone, equation (24) reduces tos

m
aG=-l
or

F5+(I)=25C

Conibiningequati&s (24) and (25) yields

~=2(w- (JIC)
Cotm -1
au (26)
. . ~
~ -442 (W+)2

which can be integrated to field (substituting in the boundary condition ‘

at the cone surface)

~t should be noted that -1 is the exact value of dS/d@ at the surface,

so long as the shock is attached. Consequently, in this analysis and
the slender cone analysis both the magnitude and rate of change of 5
with w are satisfied at the surface.

——.c — —.—.
 .
U
I!IAC!ATIT 2579

.J

t~ (IIc-t-M(
2hft= wc-1)
‘1 (27)

Again M is chosen as the Mach number just downstream of the shock.,

thus enabling M (or Ms), US, and 8s to be determined by solving eqya-
tions (13), (14), and (27) simultaneously. ~owlng these quantities,
the relation between 8 and u throughout the remainder of the flow
field may be obtained tith equation (27).

In order to determine the small variations in V throughout the

conical flow fteld, it is sufficiently accurate (since W- b<<l) to
assume a linear variation of 8 with u, namely,
.

● 8 = k~u + k~ (28)
✌

where k~ and k~ are fixed by the requirements that

=5catj W=Wc

and .
5 =bsatu=u~

Combining equations (28) and (7) there is then obtained the .relatione

(29)

-. %e retention of the tangent operator in this relation is, to the

accuracy of this analysis, optional.
.

.-. — ..-— ——— —--

..—. — .—— -. —.- — .— -.---.-—--.--———
 .
12 NACA TN 2579

which is easily integrated to yield

( 30)

t
from which the Mach number and pressure coefficient anywhere in the flow
field canbe obtained byemplo~ equations (20) through (23).

There is swell-defined restriction on”the range of applicability

of the results of this analysis. In particular, when Ms(w~ac)>l/2,
an imagimmy value of as is obtained from equation (27). This result
yermits the establishment of a boundary in the (%,5c) Plae, given W

M~(ws-~)=1/2, separating the area in which equation (27) applies from

the area in which it obviously does not apply. This lIomdarY iS sh~
in figure 2,7 and it is clear that there is a minimum value of SC for
w Mo, below fiich the cone solutions just presented do not apply. In
the area below this boundary, the slender cone solutions must be employed.

I
●

Flow Downstream of the Vertex of’s Body .4 ‘,

Simplified expressions have been obtained for calculating the flows I

about cones operating at high supersonic airspeeds, and as was pointed
out previously, these expressions can be
employed to determine the fluid properties 4
at the vertices of pointed bodies of revo-
lution other than cones. The investiga-
tion of the flow downstream of the vertices
of such bodies is now undertaken.
this purpose it is convenient to first
obtain an exact expression describing the
‘or y *S
fluid motion along a @pical streamline in
axially symetiic flow. To this end, con- 1 c~ (

o x
sider sketch (a) showing the first and
(a)
second family characteristic lines passing
through a point cm such a streamline. The compatibility equations defin-
ing the variation of fluid properties along these characteristic or l.lath

7The boundary curve was obtained by using the results of reference 8. If

the equations just developed were used, a boundary curve slightly below
the one shown would have been obtained. Consequently, the indicated
area of applicability of eq’uation(27) is slightly conservative. .

.——— -— ——. - ---- ———— ..

 NACA TN 2579 13

lines maybe written (see, e.g., reference 1)

-COS2p a$ m sinpsin5 1 dE
—+ +—— sinsp (31)
~ + sti2B x = acl Y’ yR ()d.n

along Cl, and

-COS2 p ap
—=. —ab+stipstib-—-—
I dE
sinsp (32)
7-1 +
~ sinz~
3C2 ac2 Y yR ()
dn

along c~ . Combining these equations, and noting that

aJ _afbcl +2_&
as aclas ac2as
and

where

.’ ,_ac2_ 1
%$a s 2 Cos p

there is obtained the rela~ion

COS3 f3 & sinj3sinb ~

d$ = Cos p= - (33)
y + Sinz j3 ( G+ y )

It is evident that the,troublesome quantity in this equation is ~1%>

the rate of change of flow inclination along a first family Nkch line. “
An insight into the behavior of this quantity in the flow field about a
body of revolution operating at high supersonic airspeeds maybe obtained,
however, when the value of the similarity parameter K (the ratio of the -
free-stream Mach nuniberto the slenderness ratio) for the body is large

.,

-..—- . ..-. -—.. -— –.—-———---- . .- —- ----- .. —.—. ——.- ----- _ . .

14 N!ICATN 2579

0(
i
compared to l.s To illustrate, cobider first the flow between the
shock wave and the surface in the region of the vertex of the bcdy.
Since the flow is conical.in the limit as the vertex is a~roached, it ,. I
seems reasonable to assume that the flow remains predominantly conicsl
in type (see, e.g., the streamline pattern shown in sketch (b) for some
distance downstream of the wertex. I

(b)

With this assmtions we have approximately, in the region under con- I

sideration (see analysis of cone for which @<<l),

c1 A

and n is measured normal to

the ray making the angle m
with the x axis as shown in
sketch (c). However, dn may
& n
P

6
v
s

be related to dcl by the -x

following expression 4
(c)

‘When K is less than 1 the second-order theory may ordhszil.y be

qloyed to analyze the flow field about a body.
‘It is tacitly assumed throughout this and subsequent analyses that 7
of the fluid is greater than 1. At the extreme hypersonic speeds
where, for example, 7 of the air flow downstream of the bow shock
ayyroaches 1, the streamline pattern must ~ as po~ted out by B~e~
in reference 4, have the shape of the body in the infinitesimsJJ_ythin
region between the shock and the surface. In this limiting case,
Busemannts analysis wiIl apply.
.

—.— 4
 NACA TN 2579 15
.

hence, ccmibiningthese equations yields (in the notation of equation (33))

i% -sti (P + b-fd)sinw
z~ = Y
(34)

To the accuracy of this analysis, however (see equation (25)),

Introducing this relation into the above expression for &~Cl, and
recalling that 2(5C-5) =w=<Q, there is then obtained

ati -[sinpsin5- 2(8C-5) sin (b-p) - 4(6c-@2 Cos p Cos 5]

(“35)
q= Y

Consistent with the assumption that K is large compared to 1, this

equation may be further sh_plifiSd, since the last two terms on the
right in the numerator are negligibly small compared to the first term.
In this case equation (35) reduces to the form

a8
—= - sin P sin 5 (36) ‘
acl Y

With the aid of this expression, equation (33) reati~ reduces to the
more tractable form

Cos=’p
d~== (37)
7-1 .
~+sinq3

relatimz the chanze in Mach number with flow inclination along a stream-
line in the region of the vertex. Equation (37) is recognized, of
course, as the differential equation for Prandt14eyer flow.

-.. - . . —’r—— .–- — -------- ----—- --——.

. — .— . -——— ——. -- .-— — —--— ..---— — --------
 I
NACATN2379 I

Consider now the flow downstream of the region of the vertex (just
considered) where the slopes of the streamlines are relatively small~”and
the ordinates are relatively large. F1OW in this region is, in genersl, I
certainly not predominantly of the conical typd, and hence the pretious
analysis cannot be expected to apply; however, for the tiues of K
under consideration, certain observations can be made regarding the rela-
tive order of magnitude of the terms in equation (33). For example, it
is noted that as K is increased for a given body (which is tantamount
I
to increasing lb), the term ‘in P ‘h S dS decreases while the term
Y I
cos @ increases. AS K takes on large values compared to 1, the
former term must become, in fact, small.co~med to the ~at~r term” It
is thus indicated that for flow of the type under consideration equa-
tion (33) maybe simplified to the two&nsionsl form

cos~ P
d$ =cospdb——
a8 ~
~+ sin2 f3 acl

● Similarly, the comparability equations reduce to their two~mensional. “

forms, which implies; of course, that the flow field itself is approxi-
mately two-dimensional. !l?hisbeing the case, however it my be sho~rn
I
.
(from the results presented in reference 9) that &/$cl fs generallysd
compared to &~c2 and hence that (%~cl)ds is likewise small com-
pared to cos $d~. l’husessentially two-dimensional (Reandtl-Meyer)
I
flow is found to prevail downstream of the region of the vertex. This
result, although remarkably simple, is not entirely surprising since it
would be anticipated that three+llmensional effects’would be reduced as
the local Mach nuders become lsrge and hence the local Wch angles
become small. In any case, we have the result that when K is large
compared to 1, the Mach numiberand flow-inclination angle along a I
streamline in the flow field about a body of revolution are related by
the familiar isentropic expansion relation (obtainedby integrating
equation (37)).

bB - 5A = sin-l~ - sin-l~+ I
MA MB
y+l
-tan-L/ (y-l)(MA2 -1) (38)

1°In this discussion it is assumed that &5/&<O on the body surface.

In cases where &5/&lX on the surface, it is not evident that the
argument will apply. It is suggested, however, that in these cases . ,(

the previous arguments concerning flow in the region of the vertex

will apply over a larger region downstream of the vertex.

——— ---- ——- .— — -.— —-.— .— .—.

NACA TN 2579 17
.

where A and B are different points on the same streamline.

If the streamline flow pattern were known throughout the flow field,
then the Mach number distribution could be readily obtained with equa-
tion (38). Perhaps the most useful application of this and previous
results of the analysis is, however, to the determination of Mach nmiber
and pressure distributions at the surface of a body. Attention is there-
fore turned to this calculation.

Flow at the Surface of a Body

The yrocedure for determimn

“ g flow conditions at the surface is
entirely analogous to that employed in the application of the so-called
shock-expansion method to airfoils, and hence might be cslled a conical-
shock-expansion method. For example, the Mach number on the surface at
the vertex is obtained with equations’, (14), and (20) h conibi-
nation with equations (12) and (19), or (27) and (30) depending on
whether 5C <<1, or the values of ~ and 5C are such that u-5<<1,
respectively.11 The variation of Mach number downstream of the vertex
is then obtained with equation (38). Wowing the Mach number distri-
bution, the pressuxe distribution (in coefficient form) on the surface
is readily obtained with equations (21) through (23).

Simpltiied Expressions for Slender Bodies

In the case of slender bodies the above described calculations

become so simple as to warrant special attention. This additional sim-
plification arises from the fact that now, not only is M-5 small.
throughout the flow field, but also both u and 5 are small through-
out the field. Thus, too, M is everywhere large compsred to 1, and it
results, as shown in app-endixA, that explicit solutions can be obtained
for the Mach number and pressure at any petit on the surface of a body.
These solutions maybe summarized (from appenti A) as folhws. The
local Mach nuniberat any point is givenby the relation

M“-j (39)

lLIt is clear, of course> that the results of reference 8 may also be

used for this pn?pose. It is preferable, in fact, to use these
results when interpolation is not necessary. .

-.-—— - --- -- -—. .— _— .—.. — —..———.-. —.

—Z -—. .—-— . .— ---- -.—. ——.
18 NACA TN 2579

I
where

MF

(40)

and
I

The pressure coefficient at any ~otit on the surface may be obtained

from the expression

(41) .

It is interesting to note that these expressions predict the ratios of

local to free-stream llachnumbers, and local to free-stream static pres-
sures to be the same at corresponding points on related bodies, provided
the flow fields about these bodies are related by the same value of the
hypersonic similarity parameter. These predictions are identical to
those of the hypersonic similarity law (reference 7) and consequently
they provide a necessary check on the validity of the assumptions under-
= the devel~ment of the simpl~ied methods of this paper. It
remains now to evaluate the accuracy of these methods by comparing their
Predictions with those of more exact theories.

Comparison of Approximate and Exact Calculations of Flow

About a Body of Revolution and Discussion of Results

The flow fields about a number of nonlifting cones operating at

supersonic airspeeds have been computed numerically by Kopal (refer- )

ence 8) follcn@ng a procedure similar to that first employed by l%ylor

-1
and h@CCO~. The results of these calculations provide an accurate check
on the approxhate solutions developed in this paper for cones.

.—. .. .——-
 NM!A TN 2579 19

In-figures 3 through 6 are shown the variations of 5 with (JIfor

cones having semiapex angles of 5°, 10°, 20° and hOO. The predictions
of the slender-cone solutions are observed to be in good agreement with
the results of reference 8 forhoth the 5° and 10° cones (figs. 3 and 4) “
even at Mach ntiers as low as 1.5, and in from fair to good ~eement
for the 20° cone. The agreement improves with increasing Mach number,
as would be qected. At the higher &h numbers, where the cone solu-
tion for &ich @<<l applies, b is accurately predicted as a func-
tion of w for all four cones, the ~ge of applicability (in tem of
Mach number) being the largest, of course, for the kO” cone. It is
interesting to note that fair agreement is obtained for the latter cone
down to a free-stresm Mach number as low as 2, which is only slighfiy
above the Mach nuniberfor shock detachment.

The variations of velocity tith ray angle tmoughout the flow-fields

about the above described cones are shown in figures 7 through 10. The
agreement between the cone solutions of this paper and the results of
reference 8 for these variations psrallels that found for the flow incli-
nation angle as a function of the ray angle. In general the velocity
variations are small, decreasing percentagewise with increasing Mach
number, as was assumed in the analysis. It is encomaging to note, too,
that either or both of the two approxtite solutions gener&K& provide
an accurate prediction of the variation of 8 and V with u through-
out the flow field (e.g., figs. 4 and 8). It appears that a tie of
throb for choosing the preferable solution is to employ the cone SOIU- “
tionfor which u-5<<1 whenever it is applicable (fig. 2).

The pressure coefficients,on the surfaces of the four cones have

been calculated using the two approximate solutions, and the results of
these calculations, along with the predictions of second-order theory
(taken from reference 3) and the results obtained from reference 8 m
presented in figure 11. It is observed that the two approximate solu-
tions overlap to predict pressure coefficients for the 5°, 10°, and 20°
cones that are in good agreement with those of reference 8 at Mach num-
bers fromappr~tel.yl.5 to infinity. Comparable accuracy is obtained
from the cone solutions for which u~<<l for the ~“ cone at Mach
numbers above about,3. In general, of course, the agreement improves “
with increasing Mach nuniber. The second-order theory yields more accu-
rati results than the slender cone theory for the ~ and 10° cones at
the,lower Mach nunibers,although the reverse seems to be the case for
the 200 cone. Neither second+rder nor slender-cone theory is appli-
cable to the b“ cone.

I!Yomthe preceding comparison of the conical flow’cal.culationsof

reference 8 with the predictions of the simplified solutions of this
paper, it is indicatid that the latter solutions maybe employed to
predict the properties in the flow field about a cone with from good to

‘1

. .. ---- . .. .. . .. ... ._. _ .__. ..—_.. __ ~. .--. —— -— ..- ..---- _. .._ ..-. . . .. . ..__
 20 NAC.ATN 2579
I
.
excellent accuracy, depending on whether the Mach nwiber of the free
stream corresponds to intermediate or high supersonic airspeeds. ItiiS
therefore suggested that these solutions may, for exsmple, %e particu-
larly useful at high supersonic speeds for accurately.determiningthe
conical flow fields about cones at Mach numbers not treated in the M.I.T.
tables (reference 8).

It remains now to determine the accuracy with which the solutions

for flow about cones in combination with the isentropic expansion equa-
tions predict the flow at the surface of bodies of revolution other than
I
cones. The Wch number and pressure distributions at the surface of a
family of ogives operating at a free-stream Mach nuniberof 6 were calcu-
lated using the simplified methods of this paper in the manner described
in the analysis. These distributions are presented in figures 12 and
13 for values of K vaz@ng from 1/2 to 2. Also shown are the results
presented h reference 10,12 obtained with the method of characteristics
including effects of rotation. In general it is observed that with
increasing K, the conical-shock-expansiontheory of this paper yields
Mach numbers and pressures that are in better agreement with the predic-
tions of characteristics theory. Indeed at K as low as 2, the pre-
dictions of the two theories wee within the accuracy of the character-
istics solution on the ogival nose.la The several assumptions made in
the development of the conical-shock-expansionmethod are thus justified
in part. It is somewhat surprising, however, that the method works as
welJ as it does at values of K near 1. It is suggested that this
result may be peculiar to stresmilinesnear the surface of a body, for it
may easilybe shown (tith an argument analogous to that presented in the
,,
_sis) that in the region of the vertex where three-dbnensional
effects are perhaps most pronounced, flow in stream tubes adjacent to
the surface nevertheless shows agreement with the two-dimensional
(FYandtl-lkyer)relation between stream inclination andlkch number or
pressure for values of K approaching 1.

The slender body solution also displays increasing accuracy with

increasing K, although it consistently predicts too high Mach numbers
and too high pressures on the surface near the vertex. This result is

12The Mach nmiber distributions presented here were calculated from the
pressure distributions given in reference 10.
laThe frequently suggested method of determining surface pressures on a
I
body of revolution by assuming the pressure at a point will be the
same as on a cone tangent to the body at that point was al?sotried
an~in general, the pressures were too high. The error was less at
larger K, however, although it was greater than for the conical.-
shock-e~ansion theory. As would be expected, the Mach number distri-
butions were generally in considerable error.

——
— — 1 ..—- -
NACA TN 2579 21

pr-ily a consequence of the approximate nature of the conical flow

solution employed in the development of the theory. In any case even
at a K of 1, the slender %ody theory displaye sufficient accuracy for
many engineering purposes. Indeed, upon inspection of the pressure dis-
tributions, it is evident that, although doubtless fortuitously, the
slender body theory will.yield more accurate drag coefficients than the
conical-shock-expansiontheory at the lower values of K. -

Compm”isons similar to those just discussed were made for the other
ogives considered in reference 10, and in general the same results were
obtained; namely, for K>l, the s~ltiied thedries of this paper were
in good agreement with the predictions of the characteristics solutions,
the difference between the predictions of the latter theory and the
conical-shock-expnsion theory being of the order of accu&acy of the
characteristics solutions at K as low as 2.

.CONCLS-JDING
REMARKS

With the assumption that the flow at the vertex of a body is coni-
cal, it was found that simple approximate solutions can be obtained
which yield the Mach number and pressure at the surface over a consider-
able range of free-stream Mach numbers and apex angles. In the special
case of cones, these solutions define the entire flow field with good
accuracy, and may therefore protide a useful adjunc!tto the well-known
M.I.T. tables for flow about cones.

The investigation of flow downstream of the vertex revealed that

when the hypersonic similarity parameter K (the ratio of free-stream
Mach number to slenderness ratio) is large compared to 1, the Mach num-
ber varies with flow inclination angle along a streamline in approxi-
mately the same manner as for two-dimensional (Prandtl-Meyer)flow. In
the specisl case of stretiines near the suface, it fis s~ested that
this parameter may approach 1. This and preceding results obtained for
flow at the vertex were combined to yield what might be termed a conical-
shock-expansionmethod for calculating the Mach number and pressure
distributions at the surface of a body. In the special case of slender
bodies, exceedingly simple explicit expressions were obtained for these
quantities.

Surface Mach number and pressure distributions were calculated with

the simplified methods of this paper for a family of ogi.vestraveling at
a Mach number of 6. These distributions were in good agreement with
those obtained with the method of characteristics at values of K
greater than 1. In the case of the conical-shock-expansioncalculat-
ions, the agreement was within the order of accuracy of the chmacter-
istics solutions at K as lowas 2.

.—. -.——. .-.-..—..—. —-. .—-.. .— -.— -. -—...-.-.—

 22 NACATN 2579

I
Because of the relative shpliciti of the proposed methods of deter-
I
mining the flow at the surface of a body operating at high supersonic
airspeeds (K>l), these methods should prove useful for engineering pur- {
poses . It is also suggested that the same general approach may be
~pplicable (again for-K >l)”in developing m&thods fo;-calcillatingflow
at the surface of yointed bodies of revolution at smsll angles of attack.

Ames Aeronautical Laboratory

National Advisory Commtttee for Aeronautics
Moffett Field, Calif., Sept. 14, 1951 i

-,
{

.-.— ——— . —.— ———...

.-—. —-. — . -—,
—_—— .— -—
 NACATN257’9 23

APPENDIX A

FLOW AT!I!EESURFACE OF A SLENDER BODY OPWTiNG

AT HIGH SUPERSONIC AIRSPEEDS

,e., a body on the surface of which the slopes

If a slender body (i.
are everywhere small compared to 1) is operating at free-stream Mach
nuubers very lsrge compared to 1 (again, of course, K>l), the local
Mach numbers will likewise be large compared to 1. It follows, then,
that the inclination of the nose shock wave will be small and, conse-
quently, that w will always be small. In this case the relation
between 5 and u at the vertex is extremely simple;~4 namely, (see
equation (lo))

(Al)

Combining this expression with equation (7), the relation

(AZ!) .
.

defining the velocities in the flow field is easily obtained. Hence,

the Mach number on the surface at the vertex, ~, ~Y (to the order of
accuracy of this analysis) be related to M~ by combining eqwtion (20)
with this expression to held

%2=Ms2 ~+~ * (Ms6N)2[l+Zn(*)’-

{

Now the oblique shock-wave equations for flow of the t~e under
consideration reduce to

14The equations developed in the following analysis could be obtained

from the final expressions previously developed by reducing them
to confom to the assumptions of this slender~ody theory. It seems
simpler, however, to proceed, as indicated, from the b=ic-flow
equatiois.

_ —.
-—-----
I . _. —.. -——— ._.._. . . . -——-—.- ——— -—-— -.+. ..-———-—
. . , . . ... . .
 2h NACA TN 257$3

.J

~~2 .
(7+l)2m2(M3b.d2
(A4)
[27(MOWS)2- (7-1)1 [(7-1)
(Mofl.),)2+
2]
and

(JU)

where

(A6) ‘

Combining equations (Al) and (A6) there then re’suits

(A7)

.
and
,

(A8)

Equation (A4) can now be written (co&idering equation (A7))

(A9)

Equatio~ (A3), (A8), smd (A9) provide the ~ch’ n~ber on the s~ -
face at the vertex. ntiu ~, a simplified expression for the Mach
nuniberanywhere on the surface of the body may be obtained from equa-
tion (37). Stice the local Mach number is assumed very large, this
equation may be reduced to the form

.—— .—. .—— —---- –—. . . . ..

t

4H NACA TN 2579 25

which, in turn, can be written, upon integration, to yield the local

Mach number

(Ale)

Now the pressure coefficient is given by the expression

2 PSQ -1
P=— (All)
YMo2 ( Kg )

The pressure rise across the shock may be obtained by conibiningequa-

tions (A5) and (A7) to yield

~f .“ 1 -j- 7 (M&)2

Similarly, the ratio of the pressure at the surface to the ~ressure at

the shock can be ~ressed (to the order of accuracy of this agalysis)
in the form

Iliththe aidof these expressions, eqution (All) &an now be written

(AM)

yielding the pressure coefficient at any point on the surface of the

body.
.

I r~ -- . . .. ------ ----- .—-— —- __. ..— . .. —- :.___ . .._-. —.. —–— -- —-— --—— -.— - - -
 ,-

26 2579
NACA ‘IT?

REFERENCES
I
1. Isenberg, J. S.: The Method of Characteristics in Compressible
Flow. Part 1 (Steady Supersonic Flow).Tech. Rept. F-!12R-l173A-ND, I
t&il?,Air Materiel Command, Wright Field, Technical Intelligence
(Brown University, Graduate Division of Applied Mathematics,
A-g*II/l), Dec. 1947.

2. mnK&&n , Theodore, and Moore, Norton B.: Resistuce of Slender

Bodies Moving with Supersonic Velocities, @th speci~ Reference I

to Projectiles. A.S.M.E., vol. 54J no. 23, DeC. 1!5, 1932,

pp. 303-310. ., I

3. Van Dyke, MLlton D.: A Study of Second-Order Supersonic-Flow

Theory. NACA TN 2200, 1951.Q

4. Busemann, A.: Flussigkeits- und Gasbewegung’. Handworterbuch der

Naturwlssens~hsften, Zweite AufI-age,Gustav Fischer, Jena, 1933,
pp. 275-277.

5. Tsien, Hsue-Shen.: Similarity Laws of Hypersonic Flows. Jour.

Math. andphy., VO1. 25, no. 3, Oct. 1946, Pp. 252-259.
.
6. Taylor, G. IVand Maccoll, J. W.: The Air press~e on a Cone
Moving at High Speeds - I and II. Proc. Roy. Sot. of London,
Ser. A, vol. 139, No. 883, Feb. 1, 1933, pp 278-w.

7. Hsmaker, lR&nkM.
, Neice, Stanford E., and Eggers, A. J., Jr.:
The Similari~ Law for ~ersonic FlowAbout Slender Three-
Dimensional ShaTes. NACA TN 2443, 1951:

8. M.I.T., Dept. ofElec. Engr., Center of Anal.ysis:Tables of

Supersonic Flow Around Cones. Sttif of the Computing Section,
under direction of Zdenek Kopal: Tech. Rep. No. 1, 1947.

9. Munk, M. M., and R&n,R. C.: Surface Pressure Gradient and Shock-
lRrontCurvature at the Edge of a Plane Ogive with Attached Shock
IYont. Jour. Aero. Sci., vol. 15, no. 11, No% 1948, pp. 691-695.

10. Rossow, Vernon J.: Applicability of the Hypersonic Similarity

Rule to Pressme Distributions Which Include the Effects of
Rotation for Bodies of Revolution at Zero Angle of Attack.
IT#.C.&
TN 2399, 1951.
.

,,

-. .—.—— ..
 /
//
/
/
1 +Ow

S@’
/
/
,/
I / u

‘\
‘\
\
\
\
\*4
--+4

\%\

\
\\

,,

I1.
Figure 1 – Schematic dbgrum of supersonic flow ‘ about u nonh’fting cone.
 28 ‘ ItACATN 2579

/
/
l’”

-shock detachment
z — — — — — — -

SoI’ufion hr~-8a/ mdion upphwble

i t
\

W’tion &r 8< K/ radian appkobfe

=W=–
I
2345678.9/0///2 131415/6 I

Freeksfream Mach number, MO 1

{
Figure Z.+ariation of cone ha/f-ongle with free-stream MQCh
number showing range of applicabilh’y of cone sohflon ?br -1
u-a <(/. /

—. —.— .—. ..— —— —— --- --- — —.—.——-——— --

NACA TN 2579 - 29

4 ‘
, —- Equotion (72}
-l\l Ill
a
+777--I”
2

/ — 1 1 1 I

I
I ~ I

“-1 I I I I I I ~_/
, 1Y I
‘5 10 15 20 25 30 35 40 4k

4
---–Eqution (72)
—Refenwce 8
3

/ /

05 6 7 % 9 /0 II 12 /3
(W 4? ’’=4.8602

5\

2
----Equation (12Y
/ —– Equation (27) –
—Reference 8
1 1
%.0 5.4 5.8 6.2 6.6 ZO Z4 Z8 8.2
Roy angle,+’egrees
(C) M“=13.196 v

Figure 3 .-Variofion of flow inclination angle with ruy angle for

u 5“ half-angle cone operating at IWch numbers of /.5058,
4.8602, and /3./96.

. .. . . . . . .. --—--— ._ ______ .—-. ..— __ .__ .._. . .. . ...— —.——.. —._._ ,_ —._ . .. ____...._
 30 NACA TN 2579

10

8
—- E~otion 02)
\ — Aiw%mce 8
6

4 <

2 /
/
I
90 15 # 20 25 30, 35 40 45 50
k!! 4g=L5957

~ —Rv%n?nm 8
Q2
< . I

;0
~ /0 // /2 /3 /4 /5 /6 /7 /8
(b) MO=5.4223 1,

+
/0 —

8 ?
shock -“

4
--- Equation (/2)
2 ––Equution (27) –
— Reference 8
Q~ 1 1
/(20 10.2 /0.4 10.6 /0.8 /lo //2 /14 IL6
Roy angle,u, degrees
(b) ~=/5.146 w

Figure 4.- Variufion of flow inclination angle with ruy angle for
u 10° half-angle cone operuting ut Much numbers of /.5957, .
5.4223, and 15./46.

.—.—.— _— . — —. -.—
.
NACA TN 31

20 ~
I I I
16 k
---– Equo??on (/2)

/2 I I I I — l%%
“- ‘?ti?nce 8

~
----
8 “ ------

01 I I I I I I I I
20 24 28 32 36 40 44 48 52
(u’ I&k4672
20 ~

-+ - - -
16 . ‘ -

/2
—Eqution (la ,
8 ––Equotion (2ZJ –
—Refenme 8
4

020 2/ -22 23 24 25 26 27 28
c!) Mf55457
20 —
—

16
Shock- -“

/2

8
---- Equafion (/2.)
4 ——Equation (2?) –
—Refenwce 8
I !
!?0.0 20.4 ~.8 212 2L6 22.0 22.4 22.8 232
Roy ungle, U, degrees
==$=
(C) M~9.5928
Figure 5.- Variufion of f/ow inc/inutlon angle with ray ungle for
u 20° half-angle cone operating d Mach numbers of 1.4672,
5.545Z und 9.5928.

—.—— ..—. — —...—..—. . .. —--—.—-——— .—. .— —–—.—. .—— _ .—. — . . . .. .—. .

 32 NACA TN 2P~9

4
1
40 \

I
30

20

/0 I

GO 44 48 52 56 60. 64 68 7P

30

20 ““

/0 I
.

~0 41 42 43 44 45 46 47 48 .
b) 1%=58336
40—
— —
~ shock
30

20
I

—— Equotlon (27)
/0 — Reference 8 .–
1
1 I

$0 4/ 42 43 44 45 46 47 48 I

Roy angle, U, degrees

w
(Ij M..=I3JO1

ffgure 6 .-Variation of flow inclination ungle with ray angle for ,

. u 40° holf-ungle cone operating at Much numbers of 2.0242,

5.8336, and /3.10/. ‘

.—. _ .. -.— ——— — -- .-— —-.—— —

 m 2579 33

.570

.566
---- Equ&on (I!9)
— Refen?nce 8
.562

.558

.554

.550 5A
//0 /5 20 25 30 35 40 4r5
(b) /%=15058 , ,

.9/0 1

LI=. .908 --– Equoffon (/9)

—Rethwce 8
$ .906

~ .904
*
g .902
---------- --------
.90
8 9 /0 // /2 1
(b) IUO=4.8602

.98/0 - 1 1
-–– Equation (19)
.9808 – –—Equation (30)
—Reference 8 .
.9806 ‘

.9804 - #-- \

.9802 ‘ /

w
“980~ =5>5 550 575 6.00 6.25 6.50 6.75 zoo .
Roy ungle,w,degrees
k) II?.=/3.196
Figure 7 .—Varlatlon of the ratio of local velocity to /imff~n9
velocity with ray angle for u 5° h olf -angle cone opertilng ot
Mach numbers of 15058, 4.8602, and /3./96.

-.. .-— .-. ..-— .. --— ----- .. ... ..—.———__ -.-. _—.- ----- ----- -.— .——.- -~ .. .. . .. . . . ----
 34 NACA TN 2579

. I

.590

.582 I
–– Equollon (79)
— A@fwce 8 /
.574

.566

.558

.25 30 35 40 45 50

. 9/0

‘IsA .908
.

$ .906
I I I I I I
I
t

‘-

l—————
-i--~
,
“gool~ 3 /4 15 16 17 A . .
{

.9710 I I
— Equafion (19]
.9708 – —Equuflon (30)
I
—Reference 8 t,

.9706

.9704

I
— ----
=@=
.970
?~
000 10.25 10.50 10.75 ILOO 1/’25 1150 IL75 /200
Roy ongle,~, degrees.
I
k?) M“ =/5.146

Figu)e 8 .— Variation of the ratio of local velocity to IAWing

“,
I
velocl& wtih ray angle for a 10° half- ungle cone opemting at I

Mach numbers of 1.5957, 5.4223, and /5.146.

—— —— .--— ..—
 NACATN 2379 35

.495

.485
—--Equ&on (79) ,0’
-Re?%nce 8 /
.475 ..-

.465

.455
----
--------

“44%0 25 30 35 45 50 55 60
(b) M%4672

.855
\
.854

.853 /
/--’
.852 /

.85/ — — — += -- -

,850 —
20.0 20.6 212 218 22.4 23.0 23.6 24.2 24B
(b) M=5.5457
.9025 I I
–--@uotlon (!9)
.9020 – —– Equation (.30) .
— Reference 8
.90/5
./
~.. ***
.90/0
—. —-- - .-— —---- . / -------

.9005
‘ v
.900
h—”
.0 20.4
‘ 20.8 212 216 220 224 228 2.32
Ruy ungle, U, degrees
(c) I% =$?5928
Figure 9 .-Vurlotlon of the ratio of local velocity to limiting
velocity with ray ongle for o 20° half-angle cone operating
Qt Afuch numbers of /.4672, 5. 545~ and 9.5928.

-.. . . . . . . . ..- ------- .------ —- .--— —

-—--— .——-.. .-— —. -———. -—— ___ ,._. _____ .-
 36
I
.,
.450

.430

/
.4/0

.390

.370
~
“35~0 44 48 52 56 60 64 68 72
(b) /%=2.0242
.660

+> .658 ,/
/
ci-
a . ./. /
~ .656
* / /
‘5 .654 /
*
g .652 /
/ .

.65$0 4/ 42 43 44 45 46 47 48
(.U /%%=5.8336 ,,

. 7/0 1 I
—Equufion (30)
.708 –
—Reference 8

.706 /
~“~ i
.704 /
@ /
.702 F
/9

“43 44 45 46 47 48
Ray ungle, w,degrees
w
h) 1%=/3./0/

figure /0 .+arlatlon of the ratio of local velocity to limiting

velocity With ray Ungi8 for a 40° h olf-angle cone Op8~thg .
at Mach numbers of 2.0242, 5.8336, and /3.101.

—. — . —. __ .. ——--— —.. —.
 EACA TN 23’i’g 37

L04 \

.96 i

40°
.88 — — — ~ = — — — -

.80

.72
——Soh%m tiraccl
——Sbh+on I& u-3 ~1 /
.64 — Reference 8
— Reference 3
.56

.48

.40

.32

.24

./6

.08

./2

./0

.08

.06

.04 \
\ _ 5°
— — u. 1
.02
luo=w- 7

~ 23456789/01/ /2/3/4/5/6
Free-stream Much number, IWO
=E=

Figure // .— Vffrlation of pressure coefficient with free-stream

Mach number for 5°, /0°, 20°, and 40° hu/f - angle cones.

I .
(

I
..-. . .. . . . . .—-—
-., ... —-—-.— — ———————-.
—-- -———
.—. ._-._-...—--------
 38 NACA TN 2579

I
6

“5
b] K = 0.5 .
7
I

6
I

5
[b) K=/
I
I

t-H.
4
‘rmml-, , , , , ,
Method of churactenkfics I
---– Slender body theory
I I I
30 10 20 30 40 50 60 70 80 90 /00 1
Longh’uo7nol coordinoie, percent nose length.
(o’) K=2
v (

Figure 12 .– Vuriation of local Much number olong ogives (

corresponding to four values of the similarity parume terJ .

I
!

K, for u free –sfreum Mach number of 6.

.

.— -— —— -— -— .-A -—.—— ——..—- ———–

. .
NACATN 2579 39

.024
\

—Conical-shock-expansion theory
—--Slender body theory
—Method of characteristics
.0/2

.Ow

.004
\ -N ~
%, \
o‘ Y
‘%,
-.004 \ A_ _ —
\ ,\
‘%,
K w ?\ -. ------
-.008 .
>’ \ – .——
g -.012
k (d K=O.5
Q
Q
Q .07 -
Q
3
w
w
$ .05

.04 ‘

.03 “

.02 -

.0/ -

0 ‘

1
-. 0/ I

–.oz~ I I I I I I I I I I I f I I I

O /0 20 30 40 50 60 70 80 90 /00 1/0 /2?0/30 /40 /50

Longituo’ino/ coordino~, percent nose length -
. (b) K=/

Figure /3 .-Voriotion of pressure coefficient olong ogives corresponding to fmr

voh?s of the slmh’arity porvmeter, K, fir o free -streom IUoch number of 6.

.—. ..— —— --— .--. ——— ,----— .—-— —— - --——— ——. . .-—
. . . . . . .. . .—. - ...-. .. —-----
 .i6

I
—Conical-shock-expansion theory
—– Slender body theory
—Method , of chamcferisfics
.10

1
.08

I I I I I I
.06
I

.04 A

h
.02
I
1 L
o :
I
%- --- .r --- .—. .
I
; -.02
Q
~
I u-na.- .
Q
I
u . k) K=I.5
Q1
I ~
g .24

,
.16

1 ./2 \

.08
I

.04
I

\
-.04
0
I
0 /0 20 30 40 50 60 70 80 90 /00 //0 /20 /30 /40 i50
Longitudinal coordinate, percent nose length w
I
&) K=2

I
Figure /3.-Gonc/uded.

NACA-L.an@6y -12-10-51- 1000

—— —- .._