I

[
\ [
1
I

I
I
I
f

I I
FOR AERONAU
1
1

TECHNICALNOTE
( No. 1779
i

EFFECTS OF ANTISPIN FILLETS AND DORSAL FINS

I

ON THE SPIN AND RECOVERY CH.ARACTEHSTICS

OF AIRPMNES AS DETERMINED FROM

I
FREE -SPINNING-TUNNEL TESTS

By Lawrence J. Gale and Ira P. Jones, Jr.

Langley Aeronautical Laboratory

Langley Field, Va.
1

I
)

I Washington
[
December 1948
1
.!
-,. -J7
I
I

,
-’ : ---
I ‘~” ,/.- J.
/,-.
I
\ -- .. . . . . . .. .+. ., . . . . . . . ... . . . . . . .. . . . . . .. ... . ... -.,-.
,, ,.
 ( TECH LIBRARY KAFB, NM

NATIONAL ADVISORY COBM3TEE

TECHNICAL NOI!ENO. 1779

EFFECTS OF ANTISPIN FII&ETS AND DoRsALFms

ON THE SPIN AND RECOVEKY CHARAC?TERL5TI@

QFAIRP-AS ~ FROM

l?REE—sP~ TUNNEL TESTS

By Iawrence J. Gale and Ira-2. Jones, Jr.

SUMMARY

The effects of antispti fillets and dorsal fins on the spin and
recoyeq characteristics of airplanes have been determined from an
_sis of we retits of @.nning breatigations of a large nuniberof
models tested in the Langley 15-foot ad 20-foot free-sphning tunnels.

Tb [email protected] ‘ b.diCtiiXId _&M.twhen antisp~ f~ts were ~~~ed

on au airplane, the fuselage area below the fillets became more effective
in damping the spimning rotation (higher tail-damping ratio). Whether
or not fillets satisfactorily improved recovery characteristics of a
given desigp depended, with few exceptions, upon the tail-ikmping power
factor of the design with fillets ~tallecl and upon the mass distributim
end relative density of the airplane. The results indicated that dotial
fillSgenerally had.little effect on spin * recovery c=ctefistics.

INTRODUCTION

_ aPPr~~Q 13 Years of operation of the I.angley15-foot

end 20-foot free-epindng tunnels, model tests have been maie for appra8i-
mately 200 different military airplane designs to determine their spti
and recovezg characteristics. Durhg these tests the various fQing
conditions of the airplane were usually investigated, and when the
results hlicated that the spin and recovery characteristics’would be
Unsatisfactoqj dimensional modifications were made to the model end
recommended for the airpkne such that the f5nal design would possess
satisfactory spin and recoveq characteristics. The recommmded
modifications, in most cases, consisted of increasing the tail length,
raising the horizontal tail, or adding a ventral fin. For some cases,
however, these m.odificat-ions
were not considered feasible and other
modifications were studied. One such modification that was found
effective in improving the spin-recove~ characteristics was the installa-
tion along the fuse2&e of n.&’rowexte=ions of the horizontal.stabilizer
desi~ted as antispin fillets. An analysis of the results of tests

-----..
-_,_.____,____--er.— —....-.-..,-
......._ -. ..——— .-_. ..— .—. .—
,-
 I

2 NACA ~ ~0* 1779

of such fKLets has been made h order to detemine the hportsnt factors .1
govern3ng their action.

On the Wsis of very meager data, it was indicated in reference 1 that

the action of antispin fillets was de~endent upon making the fuselage
area lelow them effective in dsmping spin rotation (increasing tail-daqdng
ratio) and it was assumed that the unshielded rudder area was unchanged.
Data from 21 diffemmt models have been used in the present paper to
detemine the acticn of fillets as regsmis damping of the spin rotation.
Consideration was also given to the possf%ility that the fillet may in
some cases shield parts of the rudder snd3 consequently, reduce the
rudder effactiveness and that the wing end fusek.ge ?MLYshield the fillet
and, thereby, reduce fillet effactiveness.

The blependent effect of dorsal fins on the spti and recove~

characteristics has also leen obtatied from available data’for 30 models.
Dorsal ftis have usually been hsta12ed on spin-tunnel models when, in
the course of development of the alrplsne, their Installation was
deemed necess~ from considerations of normal-flight sta%ility charac-
teristics.

SYMBOIS

P air a~f3ity at a given altitude, slug per cu%ic foot

s W3ng area, square feet

II wing “span, feet

w weight, pounds

f3 acceleration of gravity (32.17 ft/sec2)

m mass, slugs (w/g)

w airplsne relative-density coefficient

moments of imrtia about X and Y airplane bdj- axes,

respectively, slug-feet2

inertia yawiMg-mommt parameter

.

‘lDR tail-dsqdng ratio (reference 1)

URvc unshielded rudder volume coefficient (reference 1)

,.-. -,..-., —..

.,....-.., .. . .. ..— -——---— .. ...- . —.. —..
. . .. . ‘.
NACATN NO. 1779

tail-damping power factor (product of tail-demping ratio and

unshielded rudder volume coefficient, reference 1)
●

@ of attack, degrees

~velocity about spti axis, revolutions per second

rate of vertical descent, feet per second.

spin coefficient .

-betie~ ~axis and horizontal.,degrees

inner tig up

inner wing down

-c pressue, pounds per square foot ~

()
rollinn moment coefficient ‘0 moment
(w q% )]

pitching-moment coefficient ‘itc*

( moment
)
Wnents measured
about body axes ‘

yaw5ng-moment coefficient ‘atiq~mmnt

( ‘) )

MErmDs

For the analysis of the effectiveness of antispin fillets, the models

were’separated into groups based on the effect on the recove~ charac-
teristics. The tail-dm@ng power factors were computed for each model
with @without fillets installed in accordance with the method indicated
in reference 1, modified as a result of the present analysis, andplottai
as a function of the inertia yawing-moment ~arameter. ~ order to
separate the models h these plots according to fillet effectivm.ess,
various syaibolswere employed to indicate the degree of fillet effectiveness
on the model recovery characteristics. Plots were made for three relative-
density ranges.

Because the data indicated that dorsal fins had little effect, no
detailed @sis was made to determine their action during the spin or
recovery.

Tests

The steady-spti end recovery data used.for the analysis b this

paper were obtatied from imestigations of specific airplane models

- —,-. ,_-. .——.-——.-.— -—. —.——._ . . ..-_ ___ __ . ... ._. _ .-___
 4 NACA ~ NO. 17’79 “

in the Iam.gl.ey 15-foot end 20-foot free-spinning tunnels. ,1

The methods used for making spin-tunnel tests are described in

reference 2, althou@ in recent years the model launching technique
has been changed from hunching frcm a sp&le to launching by hand.
Briefly, a model ballasted by me- of lead wei~s to obt~ -c
simikrity to a full-scale airplane at some &Ltitude is launched by
hand with rotati~ into a vertically ristig air stream with the
controls set in a desired Positionq After a nuniberof turns, tie model
assures its spin attitude and is maintaipd. at a s~cified level in the
tumnel by admsting the airspeed so that the model dmag equals its
weight. After a nuder cf turns h the established spin have been
photographed smd timed, a recovem atteqt is made by morlng one or
more controls by means of a remote-cmtrol mechanism; if recoveq is
effected, the model dives or glides into a stigm net. The data Obt.ai~a
f- the t.e8ts= c~~aa tO CO=SP* --s* -W bY mWcdf3
descriled in reference 2. Msximum and intermediate control settings
are investigated. Airplane ~covery characteristics are considered
satisfactory if the moclklrecove= in 2 turns or le8s from the steady
spin when in the normsl spinning catrol configuration (aileqns neutral., ,
elevator up, and rudder full with the SPti) - if tie model recovers ~
2$ twos or less even with small deviations frm this control configuratim.
\ ,
A conia?olconfiguration designated.as the criterian spin indicates the
effect of small deviatims from the nomsl spinning control configumtion.
For the criterion spin, ailerons are deflected 1/3 of their fuJJ.deflection
in the direction leadhg to slow recoveries, the elevator iS qet to ~
2/3 of its fall-up deflecticm, smd recovery is attempted by reversal of
the rudder to only 2/3 full against the spin. The qmibol _ indicates
that the model required 10 turns or more for recove~ or did not recover
at all.

Factors Considered

lh order to detemdne the effactiveness of antispti fil.letson a

given design, the spti-recovery data were comP~ for we ~el ~th ~
without the fillets installed. This comparison was made for recove~
by full rudder reversal from the normal SPU conti configuration
and for recovery from the criterion spin.
/

—— -—-——.—...—, ,- . .._ —.——. —..

,,’ .,
.’
 NAC!ATN NO. 1779 5

The models were separated into groups on the foll.owingbasis:

Turns originally required Turns re-a for recovery %H?fect of fillet

for recoveqy with fild.etsinstalled on recovery

5 or more + or more None

5 or more 3 Slightly favoralile

3 or more Slightly favorable

+
~ or more 2 or less Satisfactory

l? or less Satisfactory

1+ or less Satisfactory

3/4 or leaa , Satisfactory

lAny recoveries within 1/2

, turn of one enother were conside~d as indicating
—
no effect inammzch as this is tithh the range of experhentd error.

After the models were separated into grouTs indicated by the effect
on their respective recovery characteristics of emtispin fillets, the
tail-damping power factor was computed, as previously indicated, for each
model with the fillets installed by use of the method described in
reference 1 whereby the fuselage area under the fillet is considered.
effective h damping rotation.

~ an attempt to obtain a more complete picture of the action of

antispin fillets h the spin, however, it was considered that:

(a) For steep spins, the wake of thewdng may shield part or all
of the fillet and consequently reduce or ekbninate the axea of the
fuselage under the filLet that is effective in damping the spti rotation.

(b) For certain fuselage cross sections, the wake of the fuselage
may shield the fillet and consequently reduce the area of the fuselage
. under the fillet that is considered effective in damping the spti rotation.

(c) For certain positions of the fil.letin rel.gkionto the rudder,

the fiUet may shield.part of the rudder that was pretiouel.yunshielded.
and thus reduce the unshielded rudder volume coefficient if angles of
attack and the sideslip amgles at the tail of the spinnhg model me
taken into account.

...—— .....— .._ —_— —— .... . -—.._,_.————. —.—. ----- -— —-.. . .. . . .—. .
.,
6 WA ‘1’NNO. 1779

. I
(d) When the fillet was faired into the fuselage in such a manner that
the fozward end of the fillet was very narrow, this end would probatly
be ineffective in ticreashg the demping ability of the fuselage area
under the fWLet.

RESULTS ANDDISCUSSION

‘I’d& I lists acme of the mass and dinwmional paranete~ for the ,
modeb conside~d b the investigation as well as their recovery
characteristics before and after filkt installation. For some models,
data are presented.for more them one antispin fillet tested and, in
some cases, one entispti fillet was tested on a model for several
loadhg conditions. Sketches of antispin fillets that had a satisfactory
effect on spti-recovery characteristics are presented in figure 1.

b investigating the possible shielding of the fillet by the wing,

a wake Me was drawn from the trai~g edge of the wing at tie wing-
fusehge juncture emd made en [email protected] with the wing chord which was 15° .
less than the an&jleof attack. The value of the tail.-dmupi.ngpower
factor was computed (see refer~ce 1), based on fie area below the fillet
and outside the wake line. For seveml of the models for which fill.ets .
had a satisfactory effect on spin recoveries, consideration of possible
shielding of the fillets by the wing reduced the velues of the tail-
-~ Power factor to such = extent that the value was below the
minimun value d TDPF recmmuended to insure satlefactory recovery as
presented in refemmce 1. It thus appeared that shielding of the
fill.etsby the w5ng was unldkely and for further calculatims of TDl?F,
this effect was disregazxied.

b considering possible shielding of the fillets by the fuselage,

when located above the statim of me.dmmm thickness, and also possible
shieldhg of the rudder by the fillets, use was made of the emgle of
attack of the spin amilof an avemge value of the eidesl.ipangle at
the tail of 12°. Calculations were made of the tail-damping power factw
baaed on tk possible s~e~g of the fillets by the fuselage (causinga
reduction of the kil-demping ratio) and of the possible shielding of ti
rudder by the fil.lets(caudng a reduction of the unshielded rudder vollme
coefficient). Considemtion of these factors reduced the value of TDPF
to such an extent for some models, for which fillets led to satisfactory
recovery characteristics, that the value was below the mininmm value of
TW’F recommended to insure satisfactory spin recove~ pzwsented in
reference 1. Fusel~ shielding of the fillets and fillet shielding of
the rudder were unlikely and, therefore, these effects were disregarded
for further calculations of TDPF.

—— --. — .- T-- -. —,- ,—. ..— —.-..——— —-— ., —-—.,. ——

,,, -
“,
NACA TIiNO. 1779 7

It was reco~zed that if the fillet was faired into the fuselage in
such a manner that the fo~ end of th6 fillet was very narrow, this
faired part wOtia pro’balily
be ineffective in increasing the damping ability
of the fuselage axea under the fillet. Accordingly, it was beLLeved thd
some mtnimum angle in the plane of the fillet, at which the fillet joined
the fuselage at the forwahi end, should be used to determine the effective
length of the fillet. Tnasmch as the minimum value of this angle was
12° for fillets-which, in the present study, indicatedsatisfactory effects
on spin recoveqy, this angle was arbitrarily selected. For a fillet that
made an sngle of less than 120 with the fuselage at its fomard end, the
area of the fuselage under the fKLLet considered as contributing to tail
damping was only that erea under the largest possible fillet within the
contour of the original fiUet which faired into the fuselage at an angle
of l@. (See fig. 2.) Values of TDPF were recalculated for all models
having fillets Jo* the fuselage at angles less than 120 and a better
separation between models for which fillets had a satisfactory effect
and models for which,fillets either exhfbited no effect or a small effect
(slightly favoralle) was evident. This factor should, therefore, be con-
sidered tn calculatim of TDPF when fillets are tistalled.

Figures 3 to 5 indicate the effects of entispin fillets on the

recoveq characteristics of the modeti for three rel.ative-derisityranges
and for various values of tail-damping power factor end inertia yawing-
moment parameter. The regions determined in refe~nce 1 for satisfactory
and unsatisfactory recovery characteristics are indicated in the figures.
The plotted values of tail-damping power factor were computed by consider-
ing all the fuselage area under the fillet as contributing to tail demptig
with the exception of the area under that part of the fiUet making an
angle of less than 120 with the fuselage; for these fiKLets, the method
previously described and recommended for future use was employed. It
appears from figures 3 to ~ that whether or not antispin fillets w5U satis-
factorily @prove recovery chamcteristics of a given design mill generally
depend upon the tail-damping power factor of the design with fillets
instaJJ-edand upon the mass distributim and relative density of the a~
plane.

me restits presented in figure 6 iniiicatethat the titition of anti-

spin fil.lets,for the models consid&ed in this investigation, usually
caused the angle of attack of the spting model to steepm so that better
recoveries were generaUy made.

A few tests were made for a low-wing fightie~me airplane model

(model 5A) attached to a rotary balance mounted in the Langley 20-foot
free-spinrdng tunnel. The rolling-, pitchhg-, end yawing-moment coef-
ficients presented tn figure 7 were measured with end without the fiIlets
which had previously indicateila satisfactory effect upon recovev charac-
teristics during free-spinning tests. The tests were made for em angle

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

8. NACA TN NO. 1779 .

of attack renge up to 90°, fL’

b/%?V was kept cmstant at a’typical value of .
0030, emd the wing tilt *e and the spin r&Uus were maintained at zero.
The results tnticated that antispti fillets generally had Uttle effect on
rollhg and pitching mments, ~though at very high angles of attack,
fULets did indicate a small nose-down pitching moment. Installation of
finlets genen311y created, at mOdemte and high angl.es.of atteck, an auti-
SPin yaw3ng poment which for the particular model tested was enough to
elimlnate the flatter of the two types of spti originally obtained without
the fULets and thus insure rapid recoveries.

An tivestigation of spin results obtained with the hstallatim of

dorsal.flas hdicated that generally dorsel fins had Httle effect on the
spin and recoverg characteristics,of the models. I&unmch as dorsal fins
had such a small qffec+ m the spti recovery, data are presented only far
two typicel models (one of which spins steeply and the other of which sp~
flat) for which dorsel fh were @tslled. These data are presented in
table II.as are also sketches of the dorsal fins.

Collusions

Based on an analysis of the results of free-spu-tunnel investi= o

gations on numerous models for which antispti fillets and dorssJ.fins were
tested, the follmnlng conclusionE were made:

1. The effectiveness of antispin fillets for spti recoveq a~earei

to depend primarily upon the fact tit We fuselage area ‘belowthe fillet
became effective h daqhg the spin rotation. The portion of the fuselage
area effective in da@ng the rotation was all area lelow the fillet, except
that forwanl of the station at which the fil16t joined the fuselage at en
angle less than 12°.

2. Whether or not sntispin fillets satisfactorily 3mproved recovezy

characteristics of a given design generalQ depended upon the tail-damping
power factor of the design with fillets iustalbd and upon the mass
distribution snd relative density of the airplane.
●

3* Dorsal fins generally had ltttle effect a spin and recove~

.
characteristics.

Langley Aeronautical Laboratoq

Nation&l Advisoq Committee for Aeronautics
WY Field, Vs., October U, 194$

.- -- .- ..—. .—
., —.. — ..
!.’-.. ,. ”-’..”.
—T . . . . . . . . .

.-
.-. .- ,.’.
. l’TACA
TN NO. 1779 9

0 REFERmm

1. Neihouse, Anshal I., Lichtenste5n, Jacob H ., and Peyocm, l?hili~W.:

Tail-besign Req&ments for E&tisfacto& S@n Reco&y. NACA
TN No. 1045, 1946.

2. Zhunemm, C. H.: Prelimimry Tests in the I’?.A.

C.A. lEree-Sphning
Wind Tunnel. NACA I@. No. 557, 1936.

-...—— ———.—. . ... —-— .——. ..—. —..-. .— ----- .- --—. -. —-----—- ---—--——-.-. -.— —---- --
 u &&Ix UEJm ~y:%

.
1 .lm x ld o.ol@5 o.o&Q a57x Icr6 0.0363 3% x lo~ 13.2.9 .-...- ----- 4,, l+,$

a -lea .m .023 &4 .OW qf 17.7 ----- ------

4. “ %
E3 -w, .Qm .Wc9 elk .** l-in 2.8.
Q . .... ___ .
I
% “ +
m -* .Ow .@3 m .@@ WY la Q --.—.. ..r..- .
“ + +
N -1.42 .W39 .- 40 .02% 79 a.s - &l ----- .....-

‘1 w -ti .W9 I.u

G30 -* .O11o .Oim

----
-------
“!& OA

I M
I

.C%7
I

a
I

17.65
I .4 --

1, 1~”
‘* “

----
146 O&7 ‘2a
17.@%$ ~,y“’>”,% –-...
,i
1.46 .Ow w 17.@ a- “+ ‘-- ‘-–
, , I , ! I

145 [email protected]! a. 17.65 em 7 l— ------

1

. . .
.
I

i %tmwdmln.
%WYeq .tkmpted W. find. utaq attita.hiawag a~.
%k+ttw of @o * cd * obt.nird uithmti fill.eta m elMmtq viti rillut.q
hdalhi.
 G

-Tmn8r or —V

P@rd for orit-mim

hlth
fuJd

.!
.W O.- @t x 1o-6 0.0373
*
------- —-.--
“1
,. J
I
----
TEa -28 k63 -----
I
r9

lo-&9
-IJ.’r

. m~o SW w .03@
0%

m rf.M I ‘+,3 I >,

‘?
-----
‘*
--.—-
!

I I I

1-
1

IJA -63 493

bl
b’?
%

-m 4M lJ3.g6 --.—- -.—.

+ +

F
.,.,
1

flm -7 I I .@, 15.3 ------ -—-- ------- 1 .---.-

I I
.0146 .013, p 32

1 ----- ------
51-2--- .014.6 .’W15 W’1 mm V.3 I
-----
I ----

%u0*af6pln.
~ -tmatd hfom firal *P atu+aie w attailwi. u
%.mveryiatilad n’dbM.OfOr~OmtMl amf@m-ntim for @.nning or arltd.m 6pinj anal.yaia ‘w d9~ flJJ.4t
@fret
m4eult.Mi@ cuOt&.rlmplwultsd dab.

. . . .
—
 .

I
i
I

I
 .—

NACA TN No ● 1779
14

Falddm Snu with

C.r.naiti.n Hi*ont ml@l .,

dmael ~

m a8a-t*afiti’Q-k
fnm2/3uPto~3dwLI.
%eCamv by rweraiw
attammea
thenuldarfmmfdlvfth the
Spinto 2/3agdnet.
m 1

I
I
.“

.
___ —- —

_,_
--. :-: ...- — “-’ .-. — —-— -
.._- ——- ,.-— !.
,, .-. -,
..:
. . . . . . . .
+(+$!7 TP4 Model I Model 4D

y ‘ a T ,.=

Models 2A and 2B Model 5B

T “d + i‘ ‘

Model 3C Model 6B

mgllra l.- Sketohes of antispin fil.ldm that had a satisfactory effect on the spin and reoovery
o_terii3ticf3. (Model -era refer to those given in tablm 1.)
 16

— . / -.
P

bngth of fiUet
considered
effectiveb
providing&mp@
in rotation
[
●

r
I
(

I1
i
I
‘c
I 1
I .
area under
fillet
Fuselage area
lmaer fillet I Fueelage
area under
horizontal

Iaw~=gz=
considered
ccmBMerea
not pro-
Tiding damp-
ingin
providhg dqp- tail con-
ing in rotation sidered
provmiing
“1 .
rotation rotation
Figure 2.- Sketch of fill-etfor which not all the fuselage area below
.
the fillet is comidezwd effective in dampimg the spin zv’cation.
—

—. —-— — —. . . . . .= =.. ...—-..—---- .. —------ . ..-— — ——

,’, .. ,’ ..
.,-.
NACA TN NO. 1779 17

-6
; 10

~ OrlglnaluosatlsfaotorgoondltionOr
model without fllletn installed

o Q Flrletm Installed on model had no

effeot on reoovery ohaxzioterletios

~ Filletsinstalledon mcdel bad satls-

faotoryeffeoton reooveryoharaote+

Moore

Unsatlcfaotoryregion .
13
~ 1
I
o -240 -200 -160 -120 -m -40 0 40 go x ld
IX - Iy
rnb2

Figure 3.- Effect of antispin fillets on the recovery charactimistics

of airplanes with relative densities of 15 or less u related to
requirements for tail design for satisfactory spin recovery.
(Ntiers placednearqmo~ refer tomodeb Hsted in table 1.)

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

. . ,.. .
 18 NACA TM NO. 1779 .

2000
.

MOO

1600

1400
.

1200-

10W

.
ho
v

600
I gion for sat lsfsotoryreo-
wem layrudder rewraal
aloti I
I Y P I I I
“I I 1

o
A / .

Onaaticfaotory region
4
200-

I 1 1 ,

!2S0 -2Q0 -200 .s0 - -120 -m -46 0 w (10x 10+

IX - Iy
ml?

l?i~ 4.- Effect of antispin fillets on the recovery characteristics

of airplanes with rela.%ivedensities ~eater than 15 and as much
as 20 as related to requirements for tail design for satisfactory
spin recoveq. (Nmibers placed near smob refer to models Mated
intatd.e I.)

....__— — —- ,--—- : — ——-..—

,.:,:: ,. -.
 NACA TN No. 1779 19

20C0 x 10-6

~ OriginalunaatisfaotoryoondltionOr
model without filletsinstalled f
Moo ❑ FXlleta Inatel.hdon modal had no
effeot on reoovew oharaoterletlos

1600 1
r ~ Filletslnstelledon model had satls-
faotoryeffeoton reooveryoharaote~
latioa

1400

1200

4~

Region for ●atlafaotoryreo-

overy by xwdder ra~ereel
alone

/
/ ‘

400
5

Uneatlefaotoryregion
200 m
17

“=5=
0
-280 -240 -200 -160 -120 -80 -40 0 h ao x Id
IX - Iy .

nb2

Figure 5.- Effect of antispin fillets.on the recovery characteristics

of airplmes with relative densities greater than 20 as related to
requirements for tail design for satisfactory spin recove~.
(Numbers placed near symbolE refer to modeb listed in talle 1.)
r

-. -—;— ——.- —-----—, -- —-- —..— -.,. --— --.——

—-— .-—. --—-
.,
..,-
 20 NACA TN NO. 1779 .

e-a
.

70

60

*5O

20

10

0
0 10 20 30 w 50 so 70 m
c, deg (without
rllletn
)
.

Figure 6.- Effect of antfspti fillets on spin angle of attack. (Num3ers

@aced near s~ols refer’to models Usted h table 1.)

.
I
I
—.. . ——=
,., __ . .—— ——.,, ——.. . .— _ _____ ___
.,, ,. ,, ,.- ,.
 NACA TN NO. 1779 21

Angle ofattack, a,deg

I?igtu’e 7.. Effect of antiepin fillets cm the rolling-, pitc~-, a

yawing-moment coefficients of a low-wing fighter-t~e atrplane
Inoael(mcdel ~) . The wing tilt emgle and the spin radius were

‘bt?a at‘em; % = 0“30“

— ---- -—–.
—... ._. ----—. -. —... ..7 . . . ..— —. ._ . ___ ___ _.

--—- .. ______ ____ . ___

.,