/'-"

fr
^pristf
NATIONAL ADVISORY COMMITTEE
FOR AERONAUTICS

TECHNICAL NOTE. "". : ~~

No. 1169 ~~~

THE EFFECT OF GEOMETRIC DIHEDRAL ON THE

AERODYNAMIC CHARACTERISTICS OF A 40
SWEPT-BACK WING OF ASPECT RATIO 3
By Bernard Maggin and Robert E. Shanks

Langley Memorial Aeronautical Laboratory

Langley Field, Va.

^0^
Washington
December 1946
 ]
3 1176 01425 7597

NATIOJSAL ADVISORY COMMITTEE FOR AERONAUTICS

TECHNICAL NOTE NO. H9

THE EFFECT OF GEOMETRIC DIHEDRAL ON THE

AERODYNAMIC CHARACTERISTICS OF A J*0

SWEPT-BACK WING OF ASPECT PATIO 3

By Bernard Maggin and Robert E. Shanks

SUMMARY

Force tests at low Reynolds numbers were made to determine

the effect of changes in the geometric dihedral an the aerodynamic
characteristics of a wing of aspect ratio 3 having an angle of
swoepback of 40 measured at the quarter-chord line. The results
of the tests for the swepWback wing of aspect ratio 3 indicated
that, for low and moderate lift coefficients, changes in geometric
dihedral from -10 to 10 resulted in a change in the effective
dihedral that was about 75 percent as great as that obtained
for an unswept wing of aspect ratio 6. For dihedral angles out-
side the range of -10 to 10, changes in geometric dihedra.1
produced about half as much change in effective dihedral as for
dihedral angles between -10 and 10. At a lift coefficient
above a value of 0.8, the maximum values of effective dihedral
obtained with large negative geometric dihedral angles were
greater than those obtained with 0 geometric dihedral. Over the
linear range of the lift curve, the directional-stability param-
eter generally increased with increasing negative dihedral
and increasing lift coefficient, but did not change appreciably
with increasing positive dihedral. Increasing positive dihedral
resulted in an increase in the nosing-up pitching moments
(destabilizing) at the stall, and increasing negative dihedral
resulted in an increase in nosing-down moments (stabilizing)
at the stall. Increasing positive or negative dihedral caused
a decrease in the lift-curve slope and an increase in the
variation of lateral force with sideslip.
 NACA TN No. II69

IMKOEOCTION

One undesirable characteristic of highly ewopt-back wings

is the large variation in affective dihedral with a variation
in lift coefficient. This variation tends to give excessive
values of effective dihedral at moderate and high lift coefficients,
In order to limit the maximum value of- off active dihedral to a
value that will permit attainment of satisfactory dynoaic luteral
stability and control characteristics, it airy bo necessary in
many cases to use negative geometric- dihedral. In order to obtain
some indication of tho effects of changes in geometric dihedral
on the aerodynamic characteristics of a swept-baclc wing/ an
Investigation has been made at low Reynolds "numbers in the Langloy
free-flight tunnel. This investigation consisted in force tests
of a '+0 ewept-back wing of aspoct ratio 3 wiLh geomotric dihodral
angles ranging from 20 to -3O0. The results of the investigation
are presented herein.

3YWBCLS

The forces and moments wore measured with respcct^to the

stability axes, (See fig. 1.)

CL lift coefficient (H$\

r
Cy,
'D drag coefficient /(P Itoag\
3 3
^

Cm pitching-moment coefficient

0, rolling-moment coefficient f _k_]

Cn yawing-moment coefficient (~^;)

Cv lateral-force coefficient /Lateral^orgjA

L rolling moment about X axis, foot-pounds

M pitching moment about J .axis, foot-pounds

NA.CA TN ITc. H69

N yawing moment cJbout Zaxis, f >ot-pounda

I
S wing area (0 dihedral wing), square feet
b wing span. (0 dihedral wing), feet
c wing chord, measured in plane parallel to plane of
symmetry, feet "
c wing mean aerodynamic chord measured in plane parallel
to plane of symmetry, feet
y lateral location of wing mean aerodynamic chord measured
from axis of symmetry, inches
z vertical location of wing mean aerodynamic chord measured
from lower surface of the wing' (0 dihedral), inches
wing asnect ratio 1 _. ]
\S /
dynamic pressure, pounds per square foot
\2 /

p mass density of air, slugs per cubic foot :_...._ . ._

T airspeed, feet per Becond
0 angle of roll, degrees
angle of sideslip, degrees
ty angle of yaw (-), degrees
p geometric dihedral angle, measured with respect to
under-surface of wing, degrees
a angle of attack at the lower surface of the wing, degrees
X taper ratio, ratio of tip chord to root chord
C, effective-dihedral parameter, rate of change of rolling- , \
moment coefficient with angle of sideslip, per degree [ ~2LJ
GJV, direcbional-stability parameter, rate of change of yawing-
moment coefficient with angle of sideslip, per degree

N
 NACA TN No. II69

Cy lateral-force parameter, rate of change of lateral-force

coefficient with angle of sideslip, per degree ^ft

l_ rate of change of effective-dihedral parameter with

d" r geometric-dihedral angle, per degree
CT lift-curve slope, per degree (..tfyXi)
cc, \ da,;

APPARATUS AND TESTS

The force tests to determine the aerodynamic character-

istics of the various wing configurations were made on the
Langley froe-flight-tunnol six-component balance which rotates
in yaw with the model so that all forces and moments are measured
with respuct to the stability axes. (See fig. 1.) I: complete
description of the balance system 1-& given in reference 1. All
the- trfsts wore run at a dynamic pressure""f^3^CTp^nas per square"
foot corresponding to a test EoynolAs number of 2h9,OOQ based on
a mean aerodynamic chord of 0.77 foot.
A sketch of the tapered wing (X, = 0.5) of aspect ratio 3,
with a sveepback of U0 measured at the quarter-chord line, is
presented in figure 2. The wing has a Ehode St. Genese 33 airfoil
section parallel to the plane of symmetry. This wing section was
U3ed in accordance with the free-flight-tunnel practice of using
airfoil sections that obtain maximum lift coefficients in the
low-scale tests more nearly equal to those of full-scale wings.
The wing was constructed of pine in three sections: a 0.10b
center panel, and two O.V?b outboard panels. The outboard panels
were hinged to the center panel and faired wedge blocks were 0used
to give a range of geometric dihedral angles from 20 to -3O
measured at the under surface of the panels perpendicular to the
plane of symmetry.
A series of force tests were made for dihedral angles of
0, 5, +10, 20, and -30 to determine the aerodynamic
characteristics of the wing over the lift-coefficient range for
yaw angles of 0 and 5. A few force teats were made over a
range of yaw angles of 3O0 to -3O0 at an angle of attack of
2 to determine whether the values of the lateral-stability
parameters obtained from angle-of-attack tests at 5 yaw were
reliable over a reasonable yaw-angle range.
 NACA TIT No. 1169

RESULTS AND DISCUSSION

All the t st data are "based on the area, span, and mean
aerodynamic chord of the zero-dihedral configuration and are
measured with respect to the same moment-reference axes unless
otherwise stated. The origin of the moment-reference axes shown
in figure 2 is the quarter-chord point of the mean aerodynamic
chord when the wing is set at 0 geometric dihedral. Since the
mean aerodynamic choi'd moves upward with respoct to the moment-
reference point as positive dihedral is increased and downward
as negative dihedral is increased, it is necessary to correct the
"basic data presented with respect to the moment-reference axes
when moment data are desired ah out some point on the mean
aerodynamic chord of the wing.

The results of the angle-of-attack tests at angles of yaw of

0 and 1-5 are presented in figure 3. The results of the yaw
tests at a = 2 are presented in figure k. A comparison of the
data of figure 3 with data of figuro k indicates that at least
over the linear range of the lift curve, the lateral-force, parameter
CYR, the directional-stability parameter Cno> sxid. the effective-
dihedral parameter C, obtained from the tests at yaw angles of

5 give reliable values of these parameters over a range of yaw

angles of approximately 15. "

The longitudinal-stability characteristics of the wing over

the dihedrtil range have been summarized in figure 5 and the
lateral-stability characteristics in figures 6 to 8. Symbols^
which have been used in some of the summary plots to aid in
distinguishing the curves, should not be taken as test points,

Lift Characteristics

The data of figure 5 indicate that the lift-curve slope

decreases with increasing positive or negative geometric dihedral.
The investigation of reference 2 showed that the decrease in
lift-curve slope with geometric dihedral can be expressed as
.2
<x>
S'^-o"
This theoretical relationship is presented in figure 5 and is
in good agreement with the experimental rosulta.-
 HACA TN No. II69

The low-scale lift data in figure 3 indicate that the maximum

lift coefficient generally decreased with increasing positive or
negative geometric dihedral. This decrease in maximum lift
coefficients resulted from the fact that the lift coefficients
were "based on the area of the wing with zero dihedral and not
the projected- wing area, which decreased with dihedral.

The data of figure 3 alao show an increase in the angle of

attack of maximum lift with increasing positive and negative
geometric dihedrals. As pointed out in reference 2, this increase
is caused by a reduction in the angle of attack measured in the
plane normal to the wirig surface'as the geometric dihedral is
increased either in the positive or negative direction.

Pitching-Moiiient Characteristics
The data of figure 5 indicate that, when the pitching /
moments are referred to the momont^reference axes, an apparent
increase in longitudinal stability dC^/dCL with increasing *
geometric dihedral results. When the pitching moments are
referred to the mean aerodynamic chord for each dihedral angle,
however, only a slight change in longitudinal stabilitywith
geometric dihedral results.

The pitching-moment data of figure 3 indicate that goometric

dihedral affected the low-scale pitching-moment characteristics
at high lift coefficients. These data obtained at low Reynolds
number indicated that-increasing positive goometric dihedral
resulted in an increaso in nosing-up pi telling moments (destabili-
zing) at the stall and increasing negative dihedral resulted in an
increa3e in nosing-down momenta (sbabilizing) at the stall. The
changes In pitching moment at stall are not so pronounced when
these data are corrected to the mean aerodynamic chord for the
corrospending dihedral configuration.

Rolling-Moment Characteristics (Effective Dihedral)

The data of figure 6 indicate that up to a lift coefficient"

of O.80 the offective-dihodral parameter -Cj increases with
lift coefficient and posit!vo geometric dihedral and decreases
with negative geometric dihedral. The wing with 0 geometric
dihedral reaches a maximum value of -Cj at a lift coefficient of
1.0. With increasing positive geometric dihedral the maximum
 NACA TN Wo. 1169 7

value of -Cj occurs at increasingly lover lift coefficients

(O.90 for 20 dihedral). With negative geometric dihedral, maximum
values of -C^ would "be reached at some lift coefficient "beyond

maximum lift. This phenomena can be explained "by the fact that the
change in Cj with lift coefficient near maximum lift is governed

by the nature of the wing stall. With positive geometric dihedral

the angle of attack resulting from sideslip is such as to increase
the angle of attack on. the leading-wing panel and to decrease the
angle of attack on the trailing-wing panel. With negative geometric
dihedral the opposite effect takes place and the leading-wing
panel has the smaller angle of attack. As maximum lift is approached
in a sideslip, the wing panel with the higher angle of attack
therefore begins to stall first a.nti a decrease in lift (and rolling
moment) produced by that panel results.

The data of figure 7, which is a cross plot of figure 6,

indicate that, for any lif t coefficient in the linear portion of
the lift curve, an approximately linear variation of effective-
dihedral Tiarameter C, with geometric dihedral occurs for a
t t
range of geometric dihedrals from -ID to 10 . The variation of

Cj, with dihedral -^_"-

ar decreases for geometric dihedral angles

outside the range of -10" to 10". The curves of figure 7 are

cross-plotted in figure 8, which show3 that the dihedral effectivo-

neso parameter for geometric dihedral angles between -10

and 10
*r -
varies over the lift range from about -0.OOOI7 to -0.00012
z
and is more than twice the value of for geometric dihedrals

in the range outside 10. For an unswept wing of aspect- ratio 6,

which is representative of wings on many present-day conventional

airplanes, the value of l&- is 0.00021 (reference 3). The

average value of E over the low and moderate lif t-coefficient

range for dihedral angles between tlO for the wing tested was
0.00016 or about 75 percent of the value for the unswept wing of
0 NACA TN No. II69

aspect ratio 6. This reduction in the value of z2l> for the

oi'
ewept-back wing a3 compared to the unswept wing is attributed in
part to the lower lift-curve elope of.the sWept-back wing.

The data of figures 6 and f indicate that at any lift coefficient

up to a lift coefficient of 0.8, increaaing the negative geometric
dihedral causes a reduction in the value of -C. . At a lift

coefficient above 0.8, however, increasing the negative geometric

dihedral increases the value of -C^ . For example, for lift

coefficients above 1.0, the values of -Cj are greater for

goomotrlc dihedral angles of -20 and -30~ than for zero geometric
dihedral.

Yawing Moment K&A. Lateral Force

The cross-plots of figures 6 and 7 indicale^thttt the directional-

stability parameter C generally increases with increasing

negative dihedral and Increasing lift coefficient over the linear

range of the lift curve hut is not appreciably affected by incruaaing
positive dihedral. An analysis of tho forces acting on the wing
indicated that at any lift coefficient in the linear range of the
lift curve the directional-stability parameter C^ ahould

increase with an increase in negative dihedral and decrease with

an increase In positive dihedral. The discrepancy between tho
analysis and the teat data for wings with positivo dihedral h^3
not heen explained. The data of figures 6 and 7 also Indicate
that, over the linear portion of the lift curve, the variation
of lateral force with side3lip Cyfl increases with increasing

positive or negative dihedral but does not vary with lift coefficient.

CONCLUDING REMARKS

The effects of varying the dihedral angle of a wing of

aspect ratio 3 having an angle of sweepback of kO meacured at
HACA TN No. II69

the quarter-chord line were determined 'by force teste made at

low Reynolds numbers and the results arc summarized as follows:

1. For low and moderate lift coefficients changes in geo-

metric dihedral from -10 to 10 resulted in an effective dihe-
dral change approximately 75 percent as great as that obtained
for an unswept wing of aspect ratio 6. For dihedral angles
outside the range between -10 to 10, changes in geometric
dihedral produced only about half as much change in effective
dihedral as for dihedral between 10 -and -10. At lift coef-
ficients above 0.8, the maximum values of effective dihedral
for wings with large negative dihedral angles were greater than
the maximum value obtained for wings with 0 geometric dihedral,

2. Over the linear range of the lift curve, the directional-

stability parameter generally increased with increasing negative
dihedral and increasing lift coefficient but showed no appreciable
change with increasing positive dihedral.

3. Increasing positive dihedral resulted in increasing

nosing-up pitching moments (destabilizing) at the stall and
increasing negative dihedral resulted in increasing nosing-down
moments (stabilizing) at the stall.

h. Increasing positive or negative dihedral resulted in a

decrease in the lift-curve slope and an increase In the variation
of lateral force with sideslip.

Langley Memorial Aeronautical Laboratory

National Advisory Committee for Aeronautics
Langley Field, Va., September 19, 19^6
10 NACA TN No. 1169

1. Shortal, Joseph A., and Draper, John V.: Fre-Flight-Tunnel

Investigation of the Effect of the Fuselage Length and the
Aspect Ratio and 3ize of the Vertical Tail on Lateral
Stability and Control. HACA APE'No. 3D17, 19^3.

2. Purser, Paul E., and Canrpbell, John P.: Experimental

Verification of a Simplifled Vee-Tail Theory and Analysis
of Available Data on Complete Models, with Vee Tails.
NACA ACS No. L5AO3, 19^5.

3. Shortal, Joseph A.: Effect of Tip Shape and Dihedral on

Lateral-Stability Characteristics. NACA Hep. No. 548, 1935-

*N.

/
\

N
NACA TN No. 1169 Fig. 1

Wind
direction

Wtnd
direction

NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS

Figure 1.- The stability system of axes; arrows indicate

positive directions of moments, forces, and control-
surface deflections. This system of axes' is defined as
an orthogonal system having the origin at the center of
gravity and in which the Z-axis iB in the plane of
symmetry and perpendicular to the relative wind, the
X-axis is in the plane of symmetry and perpendicular to
the Z-axis, and the Y-axis is perpendicular to the
plane of symmetry.
Fig. 2' NACA TN No. 1189

Moment- axe<s origin

27
Dihedral
angle (mJ Cmi
o ,99 0.00
10 5 9Z Ql
20 571 1.59 Rhode St Geneee*
-10 $.92 -81 33 airFoif
-20 ?-7l L-I9
-30 P-37 2.32 A -A

NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS.

Figure 2.- Drawing of the 40 swept-back wing tested.

Aspect ratio, 3; taper ratio, 0.5.
NACA TN No, 1169 Fig. 3a

1
1 *5

A 7* rt f
!
I | i
1 i
i
c v

<? JV9S3[J/3CO
\ ^ r^^-1-^ l
O 'jU?tp/JJ7XZ>

?
c
ag^on
& L Q 13
OOD
1
o
v.'?*
>

T-T
%-

lill
5S -I

CO "b
1
i

CO i
^J 05 sq ^ cv} Q

iuo'jusqijjsco ^j. i5 c\J ^. Q

'<?? 'juatotjjaco 6oj(j
Fig. 3b NACA TN No. 1169

^^T
\

TT

Ti 1

TT
i 1
1
/
D
i
1 1
*>-

| H
i 1

u
7 1
i

6-
Hc3' -k ?0 'juaioupoo
tU3Ukul-6uiMC>k 4U3UJ0UI-(XJIIP&
<v
Kl -S

p01c>

OOD $

1
I
iUMJOui- Ouu/ifid
'7j '. jiiaiyjjaoj fjij
A-V CVJ \ O5
 P
A
8=?
>-H
IV
,AY\Y
L
O
>
~w~ --- ~^jP*-"

T~
< y~ c ) c!>
It
<s M
OS
*s
CO

$0* /
IX.

y
A-< > ?$ 0 -A u ^=<i=^ --< fcll
][ '3^
to
,/
3
Y J1L
.8 /
/

p ~r
j
/
if c
IA
/
i
O
,/ /
I / /
YA
O
X
y
// A V . 1
rC-
1

4 O 4 a 16 so 0
ftfcSyng-moment
Ang/e of a/tacK.jx, dcg coe/r/djent,Cm

20 &
Angle of attackt<c, dbeg
(c) r=-/o: I-"

Figure 3'.- Continued.

CO
o
 Fig. 3d NACA TN No. 1169

, ^F
\\ /
-< -G3-
1 /

1
1

h
1
1
tr*
1
H
i

\\
, f,

J i n

"^ "s5 >.

%
.1
e|r
: s
>-< b-ei
-^o'p
g
OOD
*!1 i
s
^^
l \
h

K
f\ *n_
\
\ \
K \S k I1
V.
A
\ o
T
VV ^
^x
1
'A
\V
\ I4
\,
\
^ 4
i
'St S} 3.
Q
j

\?iU3S
K
SCO
c

'{UP/o/SjOCo 25fe/(7
A
NACA TN No. 1169 Fig. 3e

^
-

0 >- 1

1
1

i
i

<

8 8
Aj '4U3I01JJ3OO <? '4USDUJ3O0
aajc^-iaiafoj juaukut-GjiMO^ jusuxui-Eut/iay

c."

^fw 3T
I
OOD

i'T
ff
o
i > -^ s
r
<&? (.us/omaca
juauxw-a/xpH;/
 --*
J -o
-s era
5 p^ 1=5
r* ^^^"
M
a-~n a_ ~y-
=?n
J** f *j>
f

1 1
k r-=4^-^1V~- lY^<\
t-~
1

:
/
J
/
HfcS
^

& /
. !

4
? , /" .oz
\jfs

L L-ihLL
\ Jr
V--*".

O
-T
A v k A
ri

tf Jf
r y I
J^I
t f^-.
*\K^
\
\\ _n
f
* >\

^ y V4 "Y- r^L_
<M< ^ ^s^* Y 1 \\
</
< r*
0 L
Tl
&>
1
cxmrnt fI FM AENMUTKS .
U..
0 4 8/2 /6 ZO I 1 1 O
/*/)/<? of aftactL,oc,deg Pitching -moment o 4 e> iz 16 JO <& >
cceft/c/snt,Cm
Angle of CLttocKjO^cteg

(f) r = s.
Fiqure. J- Continued.

CO

i,
NACA TN No. 1169 Fig. 3g

f -&
I/ ^^S

* V/ >-
7\J>-
t /

1
1 1

i
11
1
i vjL
>3'
i rf
i i
i l
i

1
i
i

uI K1

OJ
~~. Q T* ^v, P. i o p
-tU3it/O/-/SU/M/H, ?uswau/-ou 1//0& o 3

$vV<a<i C = ^ 7

OOD >5 I
\
"i .J

fr
*
S3
X 1
Vs D
\
\
>, 3
"s
\
\ I
V,
s s
1
/
T <H 4
<
c
<
<
5 c0 ua >k c^ c:

4USLUOUI- butyo jjy -*-. q Bj -i

 o--r
o o <jq
i
D S
y
f*-*
0 ,1 2=<>-l
<,p
-K-
<>^
=fc
Y Y""v '
"vl

.04

V /*>
"^ X , f\ vr.
, et^or^^w^^r- /L
i '"N.Y\P^
MN?
\c

-n A

O
>

6 IZ /6 20 M
NATWKM, AWtSORY
Anfr tf attack, c, ofe*? ^S!f
|W T= 20.
/*/<?<VAH 3' Concluded.
CD

I I

'J
NACA TN No. 1169 Fig. 4

r
Cdeg)
o- o
O JO
o -20

j
O
rdfc^
MvJ;
S -./ r=f- -V/-"^

.2,
JOZ
Or-

^L
"~0
o
^"*
^>

-Q2.

-30 -2D -/O O JO 20 SO

Figure 4.- Variation of lateral-stability parameters

with, angle of yaw for a 40 swept-back wing of
aspect ratio 3. a = 2.
 Fig. 5 NACA TN No. 1169

.06 O/Ieasared s/o/>es.

_ Ca/cu/<zfe.d frorr, t/ieorj/
S* y/reference. ?, C^fa^) cossP
*

c1

,66-h

.. .i

Z. Moment dafa.

-30
1
xbauf o.Asc

-eo vo o
Qjbour/noc/c/ reference
JXK/S
/o
if eo
T
D/hedra/ asig/e., /3 c/e.p NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS

Figure 5.- 'Variation of lift-curve slope Cj.^ ai*d

longitudinal stability (dCm/dCL) with dihedral
angle for a 40o swept-back wing of aspect ratio 3.
 3
o
>

I [ OTWITTH f MHNAWnct
.* .6 J~~^"
Angle of aftckcff^^deg WV- coef>/cien1jCL

Figure 6.-. Effect of an^le of attack, and lift coefficient on the lateral-
stability parameters Cv, Cn, and Cr V for a 40 swept-bacfc wine of 3
~P* ~P' P
aspect ratio 3 at various dihedral angles. ca
Fig. 7 NACA TN No. 1169

SO

S?
u

I -.ooa
-20 -io o /o
Dihedral angfcj r, deg
Figure 7.- Effect of dihedral angle on the lateral-
stability parameters Cy , Cn , and -C| of
a 40 swept-back wing of aspect ratio 3.
 >

52!
/hedra/ angles befveen
I* .oooz and ~/0-
c
CD

~
vK .000/ I IrP/n ectra/ oungie^-tjo-ajcange.
I os> L AJLA. -0 oa/sjct& /Ofo -/0

NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS
I I I I
,x2 .4- .6 .6 i.O
Lift coefficient, Cu

Figure 8.- Variation of -J& witli lift coefficient for

a 40 swept-back wing of aspect ratio 3.

*l
aq
co
 SECURITY INFORMATION

ATI-71 182 CONFIDENTIAL

Langley Aeronautical Lab., Langley Air Force Base, L Maggin, Bernard

Va. (NACA RM L6117) H, Shanks, Robert E.
THE EFFECT OF GEOMETRIC DIHEDRAL ON THE
AERODYNAMIC CHARACTERISTICS OF A 40 SWEPT-
BACK WING OF ASPECT RATIO 3, (RESEARCH MEM-
ORANDUM), by Bernard Maggln and Robert E. Shanks.
8 Oct 1948. 27 pp. CONFIDENTIAL

(Not abstracted)

DIVISION: Aerodynamics (2)

SECTION: Wings and Airfoils (6)
PUBLISHED BY: National Advisory Committee for
Aeronautics, Washington, D. C.
DISTRIBUTION: U. S. Military Orgaizations request ARMED SERVICES TECHNICAL INFORMATION AGENCY
copies from ASTIA-DSC. Others route re(y to DOCUMENT SERVICE CENTER
ASTIA-DSC thru AMC, Wright-Patterson k^Jorce CONFIDENTIAL
Base, Dayton, 0. Atta; NACA. XSfJ./