100
ARR July 1949 b NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS J P L LIBRARY liioiitd/£ . AtrllRTlMl, RI: Pi)RT ORICa_NALLY ISSUED July 19h2 as Advance Restricted Report WIND-TUNNEL STUDY OF THE EFFECTS OF PROPELLER 0PE_ATION u_ND FLAP DEFLECTION ON THE PITCHING MOMENTS _D ELEVATOR HINGE MOM_NTS OF A SINGLE-_GINE PURSUIT-TYPE AIRPLANE By H. R. Pass Langley Memorial Aeronautical Laboratory Langley Field, Va. L ...... Y WASHINGTON NACA WARTIME REPORTS are reprintsof papersoriginallyIssued to provide rapiddistributionof advance research resultstoan authorizedgroup requiringthem for thewar effort. They were pre- viouslyheld under a securitystatusbut are now unclassified.Some ofthese reports were not tech- nlcallyedited. All have been reproduced without change inorder toexpedite general distribution. L- hll

liioiitd/£ . AtrllRTlMl, RI: Pi)RT

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

ARR July 1949

b

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

J P L LIBRARY

liioiitd/£ .

AtrllRTlMl, RI: Pi)RTORICa_NALLY ISSUED

July 19h2 as

Advance Restricted Report

WIND-TUNNEL STUDY OF THE EFFECTS OF PROPELLER 0PE_ATION

u_ND FLAP DEFLECTION ON THE PITCHING MOMENTS _D

ELEVATOR HINGE MOM_NTS OF A SINGLE-_GINE

PURSUIT-TYPE AIRPLANE

By H. R. Pass

Langley Memorial Aeronautical Laboratory

Langley Field, Va.

L

...... Y

WASHINGTON

NACA WARTIME REPORTS are reprintsof papers originallyIssued toprovide rapid distributionofadvance research resultstoan authorizedgroup requiringthem for the war effort. They were pre-viouslyheld under a security statusbut are now unclassified. Some ofthese reports were not tech-

nlcallyedited. All have been reproduced without change inorder toexpedite general distribution.

L- hll

41,

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

ADVLNCE RESTRICTED REPORT

!

W!_TD-TUI_NEL STUDY OF THE EFFECTS 0Y FROPELLER 0PERATI0_!

A_TD FLAP DEFLECTION ON TI_E PITC_IiNG MOMEI_TS AND

ELEVATOR HINGE MOMENTS OF A $1_GLE-ENG!NE

PURSUIT- TYP_ AIRPLA_E

P Y H. R. Pass

SUMMARY

A mock-up of a pursuit s.irple, ne hc, s been tested in

the _ACA full-scale wind tunnel and the effects of pro-

peller operation and flap deflection on the aerodynsmlc

che, racteristics of the wing-fuselage combination and ofthe horizontal tall have been determined. The results

of these tests havo been compared _ith thG results of

previous tests and with available theori_s and. in general,

satisfactory comparisons hove been obtain_,d. These re-

sults have also bean used to develop empirical procedures _

for det_rmining the effect of propcll_r operation on the

llft and on the pitc}_ing moments of a fls_pped _ing and to

evaluate e_pirical fs,cto_'s for calcul_ting th_ down_r_sh

an,_<_s at the t c,il ",:_th the propeller oporatin_. The

general _pplics,_ility of the-so empiricisK_s has not been

determined. The elc_vator hin_-moment charact_?ristics

have also been detormine_ from _ests on the mock-up and

indicat, _. the Inadcqus.cy of av_ilable data on the hinge-

moment p_,ramotors. The procod_r_ for _alculating s_ick

forces from _ind,tunncl d_ta h_s b_en outlined.

•INTRODUCTION

Extensive lonzitudinel-stability and control tests

have been con_ucte_i in the NACA full-scale wind tunnel on

a mock-up of a _ursuit airplane. The res_].ts have been

analyzed to evaluate the variou_ factors that effect the

pitcLing moments of the airplane and the _tlck forces.

A comparison of th_se results with the results of previous

work indicates the limitations of available informationfor preliminary design purposes.

The study is considered in four parts. In part I,the effect of the fusel,_ge on the wing characteristics isconsidered. Part II is a study of the effect of the tailon the pitching mo_nent and includes an estimate of theisolaLted tail characteristics and of the effective down-wash and velocity acting on the tail. Surveys of airflo,_ in the region of the tail are also included. Theresvlts of parts i and II are combined in part Ill inwhich the pitching-moment curves for the complete airplaneare developed. Part IV deals with the elevator free-floating and stlck-force characteristics of the airplaneand indicates the interdependence of the various factorspreviously considered, The effects of flap deflectionand propeller operation arc considered in all sections,

SY_4BOLS

W

CL

CD

CN

M

Cm

Che

Cp

C m

c%

P

T

gross weight

llft coefficient

drag coefficient

normal-force coefficient

pitching moment

pitching-moment coefficient

elevator hinge-moment coefficient

power coefficient / P )\pn 3 05

section pitchlng-moment coefficient

section lift coefficient

power input to propeller

axial propeller thrust

hinge momcnt)qSec" e

L

--= = ;

F

D

V

11

P

T c

CT

q

qo

(q/_o) _ff

(q/qo)av

(<U%) aa

£

ceff

S

stick force

propeller diameter

airspeed

propeller blade.angle

propeller rotational _pe_d, revolutions per

second; also, dist_ncs from c ent,_r of gravity

forwar& to a_._odynamlc center of wing-fuselage

combination (measured parallel to thrust llne)

propul_ive efflclency

air dehsity

r T hthrust coefficient _pVeDe #

absolute t_ru_t coefficient

local dync, mlc 1_ressure (&eV_h\:" Y

free-_troam dynamic pressure

effective dynamic-pressurefactor, ratio of

measured dCm/dee to value corresponding

to free-_tream dynamic pressure at _ail

ratio of average dynamic pressure at tail,

as founl from air-flow surveys, to free-

stream dy_amic pressure; the average is

weighted according to chord

ratio of arithmetical-average dynamic pressure

at tail, as found from air-flow surveys, to

fr_o-stream dy_smic pressure

local dcwnws_sh an_._le

effective downwa:_h angle at tail, as found by

comparison of pitching moments with various

tall settings and without tail

_v

S

_0

S

b

C

C

_2

_3

c_t

i t

8

T

A

A t

average downwash engS.e at tail, as found from

_ir-flow _urveys; the average is weighted.

according to both chord and local dynamic

pressure

the arithmetical average of downwash anglo

across tail, as found from _ir-flcw surveys

velocity-increment factor back of propeller

di_k

ii _ s!_t-curv_ ope for infiuite _sloeot rs.tio

span

chord

moan geometric chord

distance from propeller disk to center of

gravity of nirpln, no (me_sured pmr_llel

to thrust line)

dis._a_e fr center _f gravity t_ elevator

hinge line (messured parallel to thrust

line)

distance from trailing edge of root chord to

el_vator hinge line (measured parallel to

thrust line)

angle cf attack ef thrust axis

angle of attack of tail

angle of tail setting relative to thrust axis

control-surface deflection

rel_tive elevator effectiveness factor

empirical factor in formula for 8etermining

increase in ].ift due to slipstream velocity

theoretic_l factor used in determining .increase

in tal-'_ lift _ue to slipstrcam

m U

mm=

r

L

r-g

r-W

_s stick length

u,v hinge-moment part, motors

A¢I, A¢ 2 empirical corrections u_ed

downwmsh s ngles at tail

v&lu_ _

A denotes chan_,_c, usually due

operation

Subscripts:

o propeller-remove_

p propeller-operating

P propeller

w wing

f flmp

f + w wlng-fuselag_

fw flapped wing

t horizontal tail

A airplsnc

e elevator, beck of

i portion immersed

Is isol_ted

s slipstre_m

b balance

tr trim

ff free floating

_.c. aerodynamic c_nter

cal calculated

condition

condition

combinstion

hing_

in slip, s tream

to obtain effective

from c_Iculated

to propeller

TESTS

The tests were m_de in the NACA full-scale windtunnel (reference i). The usual wind-tunnel correctionsto the anglo of attack and the drag, obtained from refer-ence 2, and the additional correction dun to the "blockingeffect" (reference 8) have bee%_appli:_d to the experimentaldata. The pitching moments h_ve not been corrected for thewind-tunnel interference on tue downwash at the tail (refer-ence 4); the interference was, however, considered in thediscussion of the results.

The mock-up represented a single-engine, tractor-type,low-midwing airplane design (fig. 1). All parts of thecooling system snd the carburetor scoop were removed forthe tests. The elevator _,ras controllable from the cock-pit during the runs. The wing flap was of the slottedtype and was deflected 40o for _ll flap-deflected con-ditions. A 25-horsepower electric motor installed inthe mock-up oporated a Curtiss electric controllable-pitch propeller whose blade-angle setting could be con-trolled end detc_rminsd during the runs.

The force tests consisted of m_asurements of lift,drag, and pitching moment on the mock-up _,_ithout the tail

i %surfaces and _Tith tn_. tail surfaces with 7e_rlous settings

of the stabilizc_r and the elevator. For the elevator-

effectiven_ss and h inge-momeut tests nn operator in the

cockpit _anipulated the elevator control stick and. using

a conventional i_ACA control-f0rcs indicator, m_ssured the

stick forces. All tests included the effects of flap

deflection and propcll,J_r operation. The propeller char-

acteristic_ (fig. 2) were det_:rmined from propulsive-

efficiency tests of the complete mock-up. The accuracy

of the stabilizer and elev2,tor s_ttings _,_as estimated to

be within _-0.20 °. In the ano, lysis of the data, extensive

cross fairing was performed.

With the horizontal and vertical tails removed, air-

flow surveys _ere made in the region of the tail. The

surveys were made by means of a survey r_ck consisting

of 15 pitch-ya,_ tubes,

At each angle of attack: the propeller was operated

over a range of blade _.ng](_s s_nd advance-diameter ratios

to obtain _ rs.nge of thru._t coefficients. AIs, rge range

i

__=

L

_-r_

r-4

of possible operating conditions _ras thereby covered;

the greater part of the measurements, however, were madefor conditions that approximated full-power operation

of the mock-up as 9. typical pursuit airpl_:_ne _,ith 1000

brake horsepo_:rer (fig. 3). Propeller charts for a nearly

similar propeller were used for the preliminary calcu-lations. In order to obtain desired values of V/nD,

the tunnel speed was varies between _0 and 60 miles per

hour.

As previously noted in reference 5, it was found

that the lift and the pitching moments were relatively

unaffected by reasonable variations of the propeller

blade angle _ if the same thrust coefficient Tc

was maintained° Th_ results of reference 5 indicate

that, for the cases with flaps retracted, the use of

the lift coefficient for the propeller-removed condition

in determining th_ propeller-operating conditions is

barely satisfactory as a first approximation. For the

cases with flaps deflected, however, this procedure is

entirely unsatisfactory and the effect of propeller

oper_'.tion on the lift must be estimated. The propeller-

operating conditions must then be recalcul_:_ted, the

new lift coefficient being used.

I. WING-FUSELAGE COMB I_TAT i0_T

The addition of a fuselage to an isolated wing gen-

erally shifts the aerodynamic center forward (reference 6);

the lift and the pitching moments for a conventional

combination, however, are practics lly the same as those

of the isolated _ling (the pitching moments being taken

about the corre_pondlng aerodynamic centers). The wing

and the fuselage can therefore be conveniently treated

as a unit.

Lif t-Cu1've Slope

Lift, drag, and pitching-moment curves for the tail-

less mock-up with flaps both retracted end deflected are

presented in figure 4, For the retract_d flap the ex-

perimental slope of the lift curve is 0.071 per degree.

The slope for thu isolated wing as calculated by the

met!ods of reference 7, estimated section characteristics

being used, is 0.073 per degree. The results of previous

tests of similar wing-fusel3ge combinations (reference 6)

8

also show practically negligible _ffect of the fuselageon th_ slope of the wing lift curve.

The experimental slope of the lift curve for thecase with flaps deflected is 0.0?2 per degree. Reference8 also indicates only a slight change, in general, in theslope of the lift curve due to flap deflection.

Aerodynamic-Center Location

The experimental aerodynamic-center locations havebeen determined for the wing-fuselsge combination fromfigure 4 following the methods of refer_once 9.

2k_ttracted flaos.- With the flaps retracted the aoro-

d:Tn_v_ic center is 0.32 foot in front of and 0.89 foot be-

low th_ centaur of gravity. The calculated locetion for

the wing s lone, by reference 7, is O.10 foot in front of

the center of gravity. The forbear@ shift of the aero-

dynamic center caused by th_ fuselage is, therefore,

&n = 0.040_w, which is in approxim_%te agreement with the

experim,_ntal results of reference 5. This value is also

in excellent s gr_emont _Tith the t_eoretical value of

0.04Z-g w for An ca_ulated__ from the formu_ _s_2 glven" inr_f ercnce i0.

The vertical location of the aerodynamic center is

primarily a function of the drcg characteristics of the

mock-up.

_eflected fl_- The position of the aerodyns_mic

center for the wing-fuselage combin_tion _ith flaps de-

flected is 0.60 foot in front of end 1.55 feet below the

center of gr_vity. This position i_ considerably forward

of the location with retraoted flaps. The theory of re-

ference l0 indicates that p_rt of this additional forward

shift is probs_bly due to an increm_e in the effect of

the fuselage when the flsps are deflected. The further

downward movement of the aerodynsmic center is due to

the increased _ing drag.

Effect o_ Fropoller 0pers_.tion

Propeller op,_r_ttion hs,s t_m_O separate effects on the

lift and the pitching moments of the wing-fuselage com-

bination, The fire, t, des'_-_nated the direct effect, s rises

-3

9

from the forces on the propeller itself _nd may beestimated from the results of tests of isol2ted pro-pellers in yaw. The second, designated the slipstreameffect, results from the increased velocity and thechan_:e in the direction of the air flow at that partof the wing immersed in the slipstream.

_etractedfla_- The experimental effect of propel-ler operation _t v_rious angles of attack and thrust con-ditions on the lift and on the pitching moments of thewing-fuselage combination with flaps retracted are pre-sented in fi_Ires _ and 6, respectively. For comparison,the effects calculated by the zethods of reference 5 arealso shown in th_ figures. The agr,_ement bet_vecn the ex-perimental and the calculate& lift values is consideredsatisfactory. The agreement for the values of pitchingmoment, however, although satisfactory, is not quite sogood as for the lift values; the effects of the slipstreamon the wing and the fuselage pitching moments, which haveb2en neglected in reference 5, may possibly account forpart of the discrepancy.

Deflected flaps.- The experimental off,._'cts of pro-

poller operation on the 1}ft and the pitching moments of

the w_ng-fuselage combination w_th deflected flops are

presented In figures 7 _nd 8, respectively. The lift in-

crements due to propeller operation _re much !erger than

those obtained for the correspon,iing condition with flaps

retractf_.d _n:1. the pronounced .Jiving moments in:licate the

con_iderab]e effect of the sli]}_t_eam on the wing pitching

moments for the fl_p defl_:cted. An attempt ,_as made to

apply the methods of r,_fe]:ence 5, heretofore used only for

uuflapp_._d _iucs , to the present case, in order to indicate,

if possible, the applicability of these methods to flapped

wings. It was found that, except for the nece3s_ty of

changing one parameter, the effect oi_ the lift calculated

by these methods was in reasonably satisfactory agreement

with the experimental results. Those methods are sum-

marized as followers:

The calculated lift values (fig, 7) were obtained from

CLp CL o,_ (1)

= + ACLp + _ L w

where ACLp wa_ determined from the formulas and charts

of reference 5 and

lO

m

w

A CLw Sw o hClo

This formula is similar to the corresponding for_.ula

(reference 5) for the plain _,ing; it was found, however,

that k should be 1.6 in_te_,_d of 1.O. According to ref-

erence ]i, this value indic_tes the _ marked effect of the

slipstream on _he flappe_-wing vortex system. Tho term

Cio is the estimated local llft coefficient, without

slipstream, at the center of the flapped win_ rather than

the average lift coefficient of the _:;ing.

The calculated pitching-moN_cnt coefficients, presented

l_i figure 8, have been oOt_tined by consideration of the

direct effect of t_.e propeller forces and of the slipstream

effect on the win_ pitching moment. The slipstream effect

is much ];Brger than the direct effect of the propeller

forces, as indicated in figure 9, in which the direct effect

ha_ been calculated from the formul_ of reference 5.

The slipstream effect on the wing pitching moment

h_s bccn taken as the sum of two components. The first

coml encnt is due to the _ing-lift increment, which is

assumed to act at the fusclc_gc-wing aerodyne.mic center.

The second, e_nd l:,.rgest, component is the i_c_-ease in the

actual pitching moment of the _ring center sections about

their aerodyn_mic c_;nters. The s_.cond component is a

function of the incre_tse in velocity of the slipstream

and of tl_e imm<_rsed wing area. If it i_ assumed that the

section pitching-moment coefficicnts are not ;_ffectcd by

the slipstream, th_s increment racy be expressed _s follows:

= qo) sw (;)AMa c Cma. c. i i

The factor Cmo.c. is the pitching-moment coefficient of

the flapped sections and i_ assumed constant across the

flapped y,ortion Sfw of the wing aren. It is closely

approxlm_ted, from the data for the propeller removed, _.s

cm = Cm_. Sw&oC. _ C. Sf w

, =-

,-4r-4

ll

Dividing equation (S) by Swqo _w gives

SAC m = c _J.'_i(q/qo I) _'gwi Swi _8 Tcs.c. ma.c. _w Sw - = Cm_.c. Cw Sw

(4)

The final expression for the effect of the slipstream on

the wing pitching moment is

ACmw = Cma, c, cwj____--.--Swi --8Tc + _ ACLwCw Sw _ c"w

(5)

If the effect of the slipstream on the fuselage pitching

moment is neglected, the total c_lculated effect of pro-

peller operation is given by

_=

AC m = ACmp + AC m (6)p vr

The value of gCmp is, as for the condition with flap

retracted, determined by the charts of reference 5.

II. TAIL CONTRIBUTION

The study of the tail contribution to the pitching

moment of th_ airplrne involves consideration of the

isolated-tail parameters and of the effective dynamic

pressures and effective downwash angles at the tail. Thecharacteristics of the isolated tail, although an important

link in the analysis, were not available, because no tests

were mP_de of the tail alone. For purposes of this devel-

opment, these charact_ristics were estimated by analysis

of the data for the propeller removed; methods that havereceived some verification in previous studies (reference

12) were followed. The effective dynamic pressure at the

tail is defined by the elevator effectiveness dCm/d8 e and

is equal to the average local dyn:mic pressure at the tail

for the low-angle propeller-removed conditions but, forthe propeller-operating conditions, it is loss than the

average loce.l dynamic pressure me inly because of the finite

extent of the slipstream. The effective downwash anglo is

z £

12

defined by the tail incidence for which the contribution

of the tail to the pitching moment is zero.

The data on the elevator effectiveness was found to

be in good _grcoment with the theory of reference 5; the

dJ_tm on the downwash angles s_ppeared, in general, to be

less satisfactory and exhibited some apparent inconsist-encies.

As a check on the over-all applicability of the

v:_rious assumptions, empirical factors, and formulas,

the total tail contribution to the Ditching moment has

been calculated with their aid and compared with the

experimental values.

Air-Flow Surveys

Some surveys of air flow in the region of the hori-

zontal tail are presented in figures l0 to 25. ?_ith the

propeller removed, the wing wake is considerably below

the horizontal tail but approaches it ._ith increasing

angle of attack. The fuselage boundary layer is clearly

evident in al _ cases. _'_ith the propeller operatin_o, the

limits of the slipstream and the effects of propeller

rotation are readily determined. As is apparently char-

acteristic for single-en,3in_ _irplanes (references 5 and

12), the slipstream is not circular. The marked increase

in dynamic pressure, especially evident at the h_gh angles

of attack and the large thrust coefficients, on the side

of the downward-moving propell:_r blade hs_s been attributed

to a shifting of the controid of the thrust, as discussed

in reference 1.. The w_ry strong local downwash fields for

the case with flaps deflected should be noted. It should

also be observed that the downtrash :_ngles do not appreci-

ably vory with distance from the elevator hinge line.

All the surveys were evaluated to determine the

avers_ge dynamic pressure and the downwash of the air flow

at the horizontal tail. The results are presented in

table I for the case with flaps retracted and in table II

for the case with flaps deflected. Two different typesl,of avers go arc sho_rn in t_ t_bies. The values with the

are straight _trithm_tic averages, definedsubscript aa

&S

! f btl_(q/qo) .a= dx (7)bt/_

!

r-4

c- bt

bt/_

_,p ¢ dx

-bt/2

13

(8)

The vo.lues with the subscript av ar weighted _ver(o=_s,

th,_ local dyn2mlc pressure being weighted according to

the locus chord _nd the lozal downwssh anglo being

weighted according to both local dynamic pressure _nd

loc _l chord:

bt/s

(qlqo)aV - St '. (q/qo) c dx

- t/_

gaV z_

I

st (q/qo)av

bt/2

_._bt / 2 .

Tables I and II indicate that, in most instance,._, either

_lod may be used to evaluate surveys VTei_;ill.ed surveysme,_ , •

have been used exclusively herein.

Isolated-T%_il Parameters

The isolated-tail parameters _,re the slope of the

normal-force curve dCNt/da t and the relati_ze elev&tor-

effectivenesa factor T. From tests '_Jith the y ropeller

removed and 1_ith the horizontal tall at various settings

(fig. 26) and from the formula

dCi_ t

dc_t

(dCm/dit) Sw _'w

(q/qo)o St _2

(ll)

the average experimental value for dCNt/dG t was found

to be 0.051. (Values of (q/qo)o • _ere taken from Surveys.)

This value is in excellent agreemeat with the value of

0.052 taken from figure 21 of reference 14. The average

value of T, determined in this rc]?ort by the ratio

14

'di is 0.59 and is in excellent agreement witht,the value of 0.58 obtained from figure 26 of reference 14.

It should be mentioned that previous comparisons of

experimental data with figures 21 and 26 of reference 14

hsve not always given such excellent agreements as in-

dicated in the foregoing paragraph.

El@vator Effectiveness and nf_ective Dynamic Pressure

The experimental variation of the elevator effective-n_ss with thrust coefficient with the fl_ps retracted and

with the flaps deflected, is shown in figures 27 and 28,

respectively. With the propeller removed, the elevator

effectiveness is a.gproximately proportional to the average

dynamic pressure at the tail; accordingly, for these con-

ditions, the effective dynamic pressure approximately equals

the average dynamic pressure; that is,

k-J

St 12,(dOm/d6e)o = (dCfTt/dc_t) T (q/qo)o _ --

Sw Cw(12)

The proportionality no longer exists at the higher thrust

coefficients; for such conditions the effective dynamic

pressure is lass than the average found from the surveys

(tables I and I!).

The difference is due mainly to the finite extent

of the slipstream, which is taken into account in the

following equation (simplified from reference 5):

'" ,,<. = (q/qo)o +P

where

jbtistCti kt Ss (dCm_d-_c/i"s

(is)

and bti

stream,

(dCm/d_,,) e

(qlqo)o(i4)

is the span of the tail immersed in the slip-

k t is a function of this immersion and may be

L

L

=_

L_

=

r-_

,--4

A

15

obtained from reference c andt2 ,

:-i + vd +

The effective dynamic pressure is thus given by the

factor

( /qo)e:f =bt i _t i

+

Stk t s s (14a)

For comparison with the experimental results, the

elevator effectiveness was calculated by formula (l[_) for

a range of conditions. Experimental values of (dCm/dae) °

and (q/qo)o were used. For the condition -,rith the flaps

retracted, the surveys and also the computations made by

the methods of reference _. indicate practically complete

immersion of the tail in the slipstream; accordingly, a

value of 2 for k t, as indicated by the analysis of refer-

ence 15, was _sed for these cases. For the condition with

the flaps deflected, the tail immersion was calculated to

vary between 8.5 and 9,0 feet (also approximately verified

by the _urveys), giving an average value of 1.64 for k t

(fig. 41 of reference S). The values of elevator effec-

tiveness calcuio.t,_d with the_e two values of k t are

shown, together with the experimental results, in figures

27 and 28. Satisfactory agreement is observed in beth

CaSe;S,

Down,_ash

As previously mentioned, the average do,.,,nwash at the

tail £_:tv has been evaluated from the air-f!o_,f surveys.

For these same conditions, the effective down_,#ash ¢cff

has been determined from figures _9 and 30. The dis-

agreement between those two experiments,1 downwo$h anglos

(sho,,.;n in figs. [51 and 8_ and in tables I and ii), OSlOO -

cially in the lower angles, has been previously observed,

notably in reference 12. The reasons are uncertain. The

discrepancy, Acl = Ceff - Car, is apparently n'ainly a

function.... of 6av and is in_e_-,;ndentp of flgp deflection

and propeller operation, as shown in figure BS. The curve

of this figure was us,_.d to sufq.ly a downwash-angle correc-

16

tion in the calculations to be gi_zen later; its generalapplicability, however, is obviously very questlon_ble.

Pr___o2elle_rremovedo- The aver uge dew,wash angle ofthe air stream at the tail for all propeller-removed

conditious has been calculated reflecting the methods of

references 16 and i7. The agreement between the calcu-

lated and the experimental average down_ash angles,

indicated in figures 3i _nd S3, is considered satisfactory,

especially for retracted flaps. The calculated values

include %he effects of @ing twist (references 16 and 18)and the wind-tunnel corrections.

p_rov,eller olLLrati_g_.- The _ver<ga downwash at the

tail with the propeller operating has been calculated by

the procedure given in reference 5. Briefly,

f_

whore cp

£p = {wp + Cp = 6ca I

is obtained from charts in r<_.fc_rencc 5 and

g_'Tp "- gW 0

o v!

_0

This rather elementary procc_dure giv_s fairly satiG-

factory checi<s with t_,_e my _ ..... i _ a!ues (t_bles......g_. cxp_r m_nt_l v

I and If). A comp_rison of th:.:s<_ results with the results

of som_ r<_cent _ritish tc_>ts indic,_:_t<_s theft the methods

used giv_ values of Cp that, for the flap-deflected con-

ditions, are too large, l_s_much as _ven small increments

of downwash may consider,_Jbly affect the pitc}_ing moment

contributed by the tail, the discrepancy, At2 = (av- (cal,

was computed _nd plotted as a function of _cai in fig-

ure 34, Different curves were found for the cases with

flaps d_fi_cted and with fl_.ps rctr_ct_d; p_-opeiler

oper_tion, however, had no definite effect. Without fur-

_ _hc gencr._l applicability o_ thether cxparimen_ study, ....

specific values given i_ figure 24 is very question._b!e.

17

,-q

_CI_.

!

III. COM_AR!SON OF CALCULATED WiTH

EXPERIMENTAL PITCHING MOMENTS

Parts I ,-__ndII have summarLzed the available methods

for calculating the pitching mom_ _ _ _-_ __ _n.._s of _.ing!e _no_ne

airpl_mes and have derived the necessary parameters. The

purpose of part Ill is to compare the pitching moments

calcul_,_ted by these methods _rith experimental pitching

moments, in order to show the gen:_ral applicability over

the entire r_nge of operating conditions of parP_meters

dJrived as avLrag_ values from p;_rticul_r sets of tests.

The comparison is first given for the contribution of the

. 0 °)fixed tail (tall-setting ,_ngle, 1 _o; clev,_tor angle,

to the pitching moment; the co_p_.rlson is then o xt0ndod

to the complete mock-up.

Tail Contribution

The experimental tail contribution has been obtained

as the difference between pitching moments of the mock-up

with the tall attached and with the tail removed. The

calculated tail contribution Is obtained by the following

formula:

r-

Om - _ _o.a_._--_. (cl/qo)o+tp = Cl_'i" _/Sw Cw bti-gtiht ssl (_T+it_Ceff> (16)

St

In equation (16)

+ &el + Ace (17)ceff = £cal

in which

Coal obtslned from theory of reference 5

A ¢_ and A Ee

dONt/d_t = 0.051

(q/qo)o

At

s s = + -- T c - 1

f_iven by figv.r_s [_3 and 34

values obtained from surveys

1.64 for flaps extended and 2.0 for

flaps retracted

(zero for propeller removed.)

18

The _xperimental and calculated tail contributionsfor the Gandition with propeller removed are in satisfactoryagreement (figs. 35 and 36). For comparative purposes, thetail contribution has also been cnlcul_ted with experimentalvalues of ceff and with ceff = Ccsl (figs. 35 and 36).

For the propeller-oper_tlng _ondition, the agreementbetween the experimental and th calculated values is notentirely satisfactory (figs. 37 and 38). Calculations ofthe tail contribution using experimental values of Ceff(as obtained from cross plots of tables I and II) are givenin figures 39 and 40. These cclculations indicate that alarge part of the discrepancy in figures 37 _n_d38 occursbecause the methods used in the estimation of the down_ashangles _re inadequate. The discrepancies at low thrustcoefficients for the highest angles of attack may be, atleast partly, attributed to the fact that few experimentaldata in this range were t_ken and to the fact that atzero thrust, with the propeller operating, the conditionsare not quite equivalent to the conditions with the pro-peller removed_ Calculated values of the tail contributionwith Ceff = Coal are also included for comparative pur-poses in figures 39 and 40.

Pitching-Moment Curves for Complete Mock-Up

The experimental and the calculated pitching-momentcurves for the complete mock-u]j are presented in figure 41for the case with retracted fl_ps and in figure 42 for thecase with deflected flaps. The calculated curves wereobtained by the folloving formulas:

For retracted flaps,

= Cm(f _V)o tp mpvmA + + Cm + AC (18)

For deflected flaps,

= + Cm + ACmp + ACre (!9)CmA Cm(f + W)o tp w

Experimental values of Cm(f + w) ° were taken from fig-ure 4; the other terms have been previously evaluated. As

F.

E"E

L

F

ezi_eLed, the" _ 'r.._a:'_nt is not enS_rely _atisfactory;

the disagreement i_ n-erely duc to tile aco,zmulation of

errors incurred in ,_stimating the v_rious co_ponents.

The effect of _t.....landing gsar on the pitching

momcnt is presented in figure _&o As the landing geK_r

is loc_t_:d outside tLe _lipstre_m, the increment of

pitching moment due to the landing gear is probably

unaffected by propeller operation.

IV. ELEVATOR HI_E-MOi_,_SNT CHARACTERISTICS

The stick-force _at_ have teen analyzed with regard

to the hinge-zo_e_t para}_eter_ of the tail surfece, the

elevator free-floating angles, and the stick forces re-

quired to trim the _!rp!ane.

Hinge-Moment Param_ters

Some typical curve_ of the v_,riatiou of hinge-moment

coefficic_nt with angle of elevator deflection s,re shown in

figure 44. Thes_ coefficients are based on free-stream

dynamic pressure. Th_ increase in slope at a value of 8 e

of approximately =_o occurs for all conditions and Is

probably due to the i,rojection of the l_ading edge of the

elevator. _h_ followln Z analysis app]iec only to elevator

angles within the linear rs.nge that, although limited, in-

cludes most fli_ht conditions. Extcnding the m_thods to

the larger elev_:,tor angles th'_t are used in certain ma_

neuv_rs may serve to sho_r no nor< _. than the order of magni-

tude of the hi_,_ge moment. '

The basic equation for hinge moment, taken from

reference 19, is

Ohe = u CNt + v 8 e (_o)

where the coefficients Ohe and CNt are based on the

local dynamic pressure acting _t the tail.. The hinge-

moment parameters u and v should be functions mainly

of the area ratios Se/S t anl Sb/S e.

¥

2O

The v¢..uesn_ o+_ u and v were determined experimen-

to,]ly by using the follot Jing relations, based on equation

(2o):

u = (SChe/_CNt)8 " L

The parameter u _,,_,,s obtained as the moan slope of the

curve obtained by plotting Che against C!_Tt for an

elevator an_zle of 0°; v was similarly obtained from

an interpolated curve of Che (based on local dynamic

pressure) against 8 e for CNt = 0. Specifically, the

factors were obtained as follo_?_s:

(21)

(Cho)_ e=O

from curves simil_..r to those sho_,_Jn in

figure 44

oNt) 8 e=0from the experimental values given in figures

_w Sw

37 aud 38 by equ_.tion CNt=-Cat -_12 St

_Nt =0

from figures _7 and 38 and figures 27 and 28

by the eq_1_tion 8c = Cm$dCm/d6e

Che)ONt=0from curves similar to those shown in figure

44 for values corresponding to (Se)oiTt=0

(q/qo)av by cross-fairing the values given in tables

I and I I

The average experimental value of u is -0.022 and

of v is -0.004_. The generalized charts of reference 20,

which were based on tests of a large number of isolated

tail surface_, indicate a v_iue of u = -0.067 and v

-0.0084 for the horizontal tail surface.

21

,-4

--I,%

The disagree'_nt b_tween the values of u and v

determined from the_e tects _nd from the generalized charts

of reference 20 is considerable. References 21 and 22

indicate, however, that details Of elevator plan form and

trailing-edge profile may considerably affect u and v;

other factors, such as _cale effect and the cut-out,

probably affect the pressure distribution ov,_r the eleva-

tor. For those r_a_ons it is not unlikely that charts

ba_ed on a large number of tests with various uncontrolled

factors _¢ould be unsatisfactory for any particular tail.

E

The P_ate of Ch:_nge of Hinge Moment

with Elev:-_tor Deflections

The rate of changa of hi_ge moment with eievator

deflection at co__i:ant angle of attack dChe/d8 e has been• - f i

determined by measuring the slope at 8 e = 0 ° ef curves

similar to those _Xo_n in fidure 44. The experimental

variation of thi_ fsctor with &ngle of attack and with

thrust coefficient is given in figure 45(_) for the case

with retractad flaps end in figure 45(b) for the cs_e

with def!ccted fl:_ps. It should be mentioned that the

hinge-moment coefficient Che is based on free-stream

dynamic pressure.

The formula _o.__- calculating &Che/d8 e may be obtainedby differentiating, equation (_0). If ,the difference between

the effective end the averh_e dynamic pressures at the

tail is neglecterS, ti-e final _--_ression is

F -rdO_'r "+ (2s)] (_m_

where, if desired, (q/q.o)av for the propeller-operatlng

conditions may be calculated from

- -A"

"--'b=(c-;b\qo av o St

For comparison, d" /d8 values ,_ere calculated, experi-ui_ e e

mental values being _]sed for a!l factors, and are also

presente_, in figure 45.

22

The experimental and the calculated valuss are inexcellent agreement for the case with flaps retracted.The agreement, however, is not entirely satisfactory forthe case with flaps deflected; the discrepancy probablyarises from the very marked variation in dynamic pressureacross the elevator span.

Elevator Free-Floating Angle

The elevator free-floating angle is important withregard to stick-free stability characteristics of anairplane. The formula for calculating it is derived bysimultaneously solving equation (20), with Che O, andthe normal-force equation. If the difference betweenthe effective snd the average dynamic pressure at thetail is neglected, the solution is

_.u dC Nt/dat) (C_T + it - Ceff)

8 = - (24)

elf u<dCNt/d_t) • + v

F_F=#

By the substitution into this equation of valuos

previously derived, the elevator free-floating angles were

computed for a number of conditions. The results are

plotted in figure 46, together with experimental values

for the same conditions. There appears to be an almost

constant differenco of about 2o in 6ef f between the two

sets of curves. The discrepancy is possibly due to dis-

symmetry of the tail surface. _,_easurements showed that

the elevator hinge line was slightly above the chord line;it is uncertain, however, whether this error in construc-

tion can account for the entire observed discrepancy.

Stick Forces

The stick forces required to trim the airplane at

any given condition can be determined from these tests

after the corresponding elevator hinge-moment coefficients

have been evaluated. The usual method of determining these

coefficients is to use the basic equation for hinge moment

z

f w!".ere

Chetr(nON t + VSet r \qOJav

etr dgm

d._e

23

(25)

and

dC_t_CYt = _t _ - )(_'T 4- it - c elf + TSetr

Inasmuch a_ the elev_tor free-floating angles an'J the

rates of change of hinge-moment coefficient ,vith elevator

deflection have been experimentally determined (figs. 45and _6), the hinge-moment coefficient at trim has been

obtained more simply from

d e_

Chetr tr elf d8 e

Values of Chctr are precented in figur_s 4? and 48 for

the conditions with flaps retracted and with flaps de-

f!octed, respectively° .Experimental values of CmA and

(dCm/dSe) have been taken from figures 29 and 30 (for

i t = 1,2 °) and figures 27 s.nd 28.

Neg].ecting the effects of friction in the control

system allo-v,rs the stick forces for trim to be calculatedfrom

Ftr =

Ch qo Se _eerr

_s

24

where

W

.._no =.CL_.p

S:.:

or, if the e.irpl_ne is climbing or diving at a largeangle,

Sw Te cos _- C

CLAp

qo =

C SwT;Ap

(_9)

At high angles of attack and large thrust coefficients

equation (PS) gives values of qc the.t are about 12 per-

cent greater th_.n those obtained from equation (29).

Sufficiently accurate values of CLA p may be obtained

from figures 5 and ? _nd values of CD may be obtained

from figure 4.

SUMF.ARY OF __I)TDI_TGS

The follow!r_g remarks, although applying directly

to tLe mock-up tested, iorob_.bl_',possess v_rying degreesof gener_l applicabilit3-o

i. For cs,se_ t,..,ithfl_ps deflected, the propeller-

operatiug co]ditfons c_.nnot be directly determined from

the _.ropeller-rezove_ lift coefflcicnt.

2. The ,'}lope of the lift curve of the tailless

mock-up can bc accurately calculated by the uF_e ofre.ferenees 7 and 8.

3. The forward shift of the aerodynamic center of

the plain vri.:_gce.used by the fuselage can be estimatedby the use of references 6 aud 10.

25

r_

4. The effect of propeller operation on the lift

and on the pitching moment Of the tailless mock-up with

retracted flaps con be satisfactorily estimated from theprocedures given in reference 5.

v

5. _ith the flaps deflected, the increments of llft

due to propeller operation _re much larger than thos_

obtained for the corresponding condition with the flaps

retracted• The differcnc_ is probably due to the effect

of the slipstream on the fl_pped--wing vortex system.

6. Tile slipstream markedly increases the flapped-wing divin Z moment.

7. The isolated-tail param_ter_, as determined from

those tests, compare satisfactori'w _y ith those given bythe _c_larts of reference 14.

8. The effective dynamic pressure at the tail for

the propeller-operating conditions can be accurately

estimated __rom __efercnces 5, ll, and 15.

9. The downwash antics at the tail determined from

different tail settings are not _qual to _hose determined

from air-flow surveys, especially at low angles of attack•

]0. The average downw}_sh angles of {he air flow at

ta. tail, with the propeller removed, can be closely cal-

culated from references 16, 17, and 18.

i! The meth = o alca!c_tin,_• o_s f c _ the propeller-oporatinz down,rash angle _t the tail from reference 5

sre barely satisfactory a_s first apprcximatlons unless

empirical correction factors rare used. It is believed

that most of the discrepancy, for the flap-deflectedcondition, zgj be attributed to th_ methods of calcu-

lating the dow:_w,_sh due to the propeller.

13. Pitching-moment curvets fo," very nearly similar

_.irplanes can prob_bly b_ sati._factoril_ estimated by the

_se of the propeller-remov_d pi%cning moment of the tail-

less airplane and the empirical do_nwmsh correction fe.ctors.

13. The use of the chL_rts of refer?nee 20 for

determining u and v, which are based on the results

of a l_r_e number of te_ts of horizontal tails, is un-

satisfactory. References 21 and 32 indicate tl_at details

of the elevator ple.n form end trailing-edge profile areimportant considerations.

36

14. The climb o7 the d_ve angle of an airp!_ne inpowerad flight should bc_con_idar_d in calculating thefree-_tre_ d._namic p:_ssure.

C01_CLUDI_GREMAPKS

Most of the basic, factors affecting the pitchingmoments and the stir. forces o* _n airpl'_ne can besatisfactorily estiJ__.ted by use of the av_L_lable theoriesr_nd procedures; further s_-stem_tic experiments and relatedtheories, ho_evcr, _re neccss _z_...... ore the do_nwash _t the

_l_er ing rtail with prop_ _ operat m:_y be eli_bly predicted.

Experimental data a_d charts of the hinge-moment parameters

should be used v,,ith ext_'eme care, and du_ consideration

sh __o_._ be given to the vazlous factors, affecting these

pars_meter s.

Langley Memori:_l Aeror_au_leal Labora_tory,

_Tatio_al Advisory Co_n_ittee for Aeronautics,

Lo_n:_ey Field, Ta.

R_FERE_TCES

,-4

r4

--i"

• DeFrance, Smith J. : The }?.A,C,A, Full-Scale %'_rind

[;ur,nei. Rep. No, 459, }TACA, 19 _'_,.)_) ,

_o Silv,:_rsteln, Abe, and White, James A,: Vind-Tunnel

Interference with P_rticular p_eference to 0ff-Ocnter

Positions of the %:ing and to the Do,.,,nwash at the

Tail. Rep. ![o. 547, NACA, 1955_

So Theo_orsen, Ti_eodor3_ and Silverstein, .&be: Experi-

mental Verification of the Theory of _'gind-Tunnel

3oundary In i;erference. Rep. }[o. 478, 17ACA, 1934.

o Silvsrstein, Abe, an_ .Xatzoff, S0 : F.xp'crimental.

Investigation of ",'_Jnd-Tv.nnel Interference on the

Dewnwash behind an Airfoil_ Rep. No. 609, }TACA,

19 370

5o Goett, Harry J,, and Pass, H. R. : Effect of Propeller

' S ,,g!e0per'_tion on tae Pitching Mom,:_nts of i- -EngineI (D" •14onop!anes. NACA .A.C.R._ May _,:_1

o Jacobs, Esstms, n N., and Ward, Kcun,._.th E,: Inter-

ference of Wing s,nd Fuselage fret4 Tests of 209

Combinations in the hToA.C.A. Variable-D_:nsity

Tunnel_ Rcp, No, 540, I,_ICA: 1935,

o Anderson, Raymond F, : D_t rmination of the Charac-

tcri-_tlcs of Tapored Uings, Rcp. No. 572, IfAOA,

1936.

• pearson, Henry A. , and Auderson, Raymond F. ; Calcu-

culation of the Aerodynamic Characteristics of

T.%percd Wings with Partial-Span Flaps. Rep, No.

6_5, NACA, 1939.

¢ Thompson, M. J.: A simp].c _,_.:_t'_odfor Determining the

A:_rodynamlc Center of an Airfoil. Jour. Aero. Sol. ,.

vol. 5, no. 4, Feb. 1938, p]:,. i_8-140.

I0. _,.ulthoFp, H : :_erodyngmic_ of thc .... ei_"e

Ti No. IOY3, 1242.

17.'_.CA

Ii. Smelt, R , _ _ ....... _ of Increase• ,:.n_ Davies, _° : _s_.mation"_o. 1788i_', Lift due to Slipstre_m. 7{. & M. _ .

_r- tisi% A.P..C , 19[:,7_

28

12. Katzoff, S.: Longitudinal C_ability and Controlwith Special Reference to Slipstream Effects.Rep. No. 690, _ACA, 124C.

13.

14.

Olauert, H.: Airplane Pro_eliers. Vol. IV, !i_. L,

_ec. 5, ch. XiI of iercrynamic TheorF, _. F. Du_'ond,

ed., J_!iius Springer (Zerlin), 1935, _p. 351-357.

Silverstein, Abe, and Katzoff, S.: Aerodynamic

Characteristics of Horizontal Tail Surfaces.

Repo _[o. 688, ZfACA, lb40.

15.Koning, C.: Influence o/' bhe Propeller on Other

Parts of the Airplene Struoture, Vol. IV, dlv. M

of Asrodynam_c Thecr_, _. F. Durand, ed., Julius

Springer (3erlin), 12J5, pp. _61-430.

16.Silverstein, Abe, Xatzoff, S., and Builivant, We

Kenneth: Downwa_h and Wake behind Plain and

Flapped Aid'foils, Rep. No. 651, NACA, 1939.

_7.Silverstein, Abe, and _Lzoff, S.: Dcsi_2n Charts

fo _ Predicting Do,ffn_ash ' __ _n_le_ and "_a;:_e Ch_rac-

te1"istics behind Plain and Flappe:.] Wings. Rep.No. 648, NACA, 1989.

Pearson, Henry A., and Jones, Robert T. : Theoretical

Stability _4 Control 0h_racteristics of Wings with

Various Amounts of Taper an& T_Tist. Rep. No. 635,NACA, 193_.

19, Glauert, H. : Th:_oretical ]_.e!ationship_ for an

A_rofoil with Kin_:ed Flap, R. & M. No. 1095,B_'it!sh A.R.C. , 1927.

Pass, H, R.: Analysis of Wind-Tunnel Data on

Directional Stability and Control. T. N. l_o.?75, NACA, 1940.

21. Jones, Robert T., _n& Cohen/ Doris: Determination

of Optimum Plan Forms for Control _mrf_c_s.

E_C.\ RoD. No. 7ZI_ 194_.

22. Jon,_s, _obert T., _n-i Aze-, Milton _., Jr. : Wind-

Tunnel Investi-_'_t_'on_.. of Control-Su_'face Charac-

toristlcs. V - The Use of a Beveled TrailingEdge to Reduce the ;{_nle Moment of a Control

Surface. h_AOA A.P,R., Znrch 194_.

Br4

,4,I

_ACA TABLE I

COMPARISON OF EXPERIMENTAL AND CALCULATED

DOWNWA_ ANGLES FOR MOCK-UP WITH FLAPS RETRACTED

iJI

(deR_

-0.21

_.d. i

6.9110.9 I-2.8 I-2.5 I-1.5 i--I.0 I

I #'% !.K.%2 |

5.11•,a !

,,.#..A !

5.01

5.116.81

6.8i8.9t

I0,7 I14.5l

14.71

r

T c P

(dog)

(a) (a)

(a) (a)

(a) (a)(a) (a)

O _ 41

.02 ! 34

0 56.02 41

.01 55

.04 50

.08 I 29

.03 l 50

.O9 I 37

.16 I 26

.01 26

.II 30

.18 37

.05 26

.51 55

.46 29

.12. 29

&Propeller removed

(q/qo)aa

0,87e_

.84

.82

.94

.93

.93

.99

.87

.97

1.08

.93

1.041.29

.961.09

1.44

1.021.62

2.01

1.17

(q/qo)av

0.84

.79

.81

.79

.91

.92

.90

.96

.8,5

.951.07

.91

1.051.26

.93I. 07

1.46

1.00

1.632.01

1.15

TABLE I I

i

aa av

(deg) ideg)

1.2

2.64.6

6.6.i

.4

.8

.7

1.31.5

2.8

5.i

5.55.2

5.2

5,9

5.]6.8

effl

(deg)l

1.2 2.5 I2.6 4.2 I4.7 5.6 I6.6 6.4 I

.i 1.01

.4 1.1 I9 -I w_ i. .L. , i

6 _ i.,, n

1.3 2.5 I1.4 2.5 12.8 3.5 I3.0 4.b I3.3 4.7 I5.3 6.1 I

4.6 5.5 I

6.0 7.1 I5.4 I 7.] I7.0 7.8 I

i

8._ 9.1 9.9112. i 12.9 12.71

13--_'7 _ 111--!_ 11.7 I

COMPARISON OF EXPERIMENTAL AND CALCULATED

DOWNWASH ANGLES FOR MOCK-UP WITH FLAPS DEFLECTED

ccal "

(deg)

0.9

2.4

4.26.5

-,5

-.3

.2

.2

.9

.91.9

2.6

2.84.5

5.8

5.56.1

6.39.9

14.2

i0.9

_S

uT T c

(deg) (deg)

7.8 (a) (a)

15.1 (&) (a)

5.8 0.22 23

6. I .07 18

7.3 .52 288.6 .34 23

9.6 .35 329.7 .17 18

9.7 .19 18

12.8 .46 2313.1 .21 18

14.2 .58 28

15.0 .08 28

(q/qo)aa

0.88

.83

I. 551.28

I. 921.67

1.60

1.341.36

1.80

1.552.001.03

(q/qo)av

0.85.81

1.61

1.272.03

1.721.66

1.36

1.381,89

i .35

2.011.03

apropeller removed

Iaa _ av

(deg) 1(deg

i0.71 II.

13.0 13.14.01 15.

13.0 12.

16.3 15.

16.6 16.16.6 16.15.2 15.16.3 16.

i 19.4 19.

! 18,0 17.

i 21.4, 20.

_16.55 .... 16 .'

¢ eff i c cal

I l(deg) !(deg)

11. i0.5

13. 14.614 • 12.7

;I 12. n.o16.4

_/l 18". 17.2 16.5I 17.5 17.6

it 15.6 14.

19.3 !21.917.1

20.8 , 24.8

16.4 _ 18L]_

r4",4"!

v

NACA Fi_. 1

F

r-i,.-i

.--.1"!

,--1

,-I,-4

!- ,._i

c !

, \!

\_" ....\

\\

i

\\

\

!

- \

I ...........

Q ' ' 0 _ '

\'h,\\

4-_'_.L.

! ! : :

\

e,l o

ol.

_ _ D_ _

_:_ ,_:

0

•-.a _.j _ _.

._ o

rO

_O /_:._:_!_o_ r:,_j O

I I

"D c _,lglll_O_ 44,n

_,, C)

.....3-__,

-- ,4_-

JI

0

qJ

i __

T'.I _• l,_kg. _

e-I

!

NACA

u

,.*,_!

!I

!

III

I!I

I

I

I

, _I

I o"I

' i.' 5

I II

it "_: ?;0-

\, ,

°r-._0

- °o.-

\i" I,\

\\

\

m

_6I

!!

I

I

I

0I"

0 -

_ "_'_

' _I.

.__

_r--4,-4

I

NACA

ii

ii

i

I"

w

2//I

//

t;;/p!

,-? / /////////Ll//

%

om

\\ -\

\ \

.\- .... _.\•,\ \\ \

\ ,

• \

"..,.\ \

\ \

//

\

\

\

\

\,\ \

-4-

o

I

__i___ t __

\\\

0

°_

I

O3

L=

I

f--I

. -,,,,+!

NA CA oi

T

I

L. _

!

i

=,

! o

|

=::

=

NACA

-- I

r o

S

%

¢

-A e --

ID

_ -3--

II1 -4.--

• I I

Fi?ure I0.-• Concluded

8/

'I-1 I I I I I3 2_ I 0 I P- 3

Disf_nce f'r'om center line, {t,

(e) DynQmic-pressure (R/q) COntOurs,

_igure I0 .--Air flow in She plane c_ +,he

erewtor Iqinge l;ne.

vTew IooKincj forword; O_T,-O Z°_

m _ m

• it

m

w

Ir

I

_. o'; propellerI

/

-,, ]/ .

I

I

removed.

I I I I I_. o 0 I P-

Disl;_ce from oent, er li_ _

(_ Incllna_ion of f,he air $_ream.

4-

4,

I I I I'

rig. _o

I

NACA£-

_l-

C 1.0_

_o

L

O

o

oE

_3-¢--

.9 • 0 o

.0

Fi 9. .

.9

I 1 I I 1 I 1 I I I

Distance from center line_ ft

I

.E

t:: |-

E_- _

o

'_-3--

Ca) Dynamio- pressure (o/'clo) contours.

Fi_ur_ _ I! .--Air flow In the 101_ne of the

elevator hincje line.

View looKir_ forw_rd;c_T,5.1";

6?, 0°; propeller removed.

"x

i

m

\/ / J

I _"

w #

I I I• 5

Figure I1.Concluded

,gCale Of veotor_

Deviation, deg

I I I I I2 I o I R

Distance _'rom center line, ft(b) Inclination of the oir streorn.

I I i 13 4 5 It

r4

i_q

NACA

q-

B

l:: I E

L

> o

.qv

q-

o

I n8 _

,I/) )/ -.'--

.7 II

° •

,@1o0

I 1 I I I I tI o I 2 3 4

DIst_nce from center line, ft

Fig. 12

R]ure

¢c- 1_ I

#3

_= i mI:

_o--

E- ! --

@

I3

_3

#

I I I6 _ 4

Figure I_.. --C_clud_4.

(a)

12.--Air flow in the plone o# _he

elevo_or hinge line.

View IooKin_ l_orw(3r_;(_T, _,_f;

_-_, 0"; propeller removed.

Dynamic-- pressure ('ci/_la) contours,

i

\ \\ \

3 Cal e _Yi_ c_or@o I_ _o

l # o I P..Distance from center line+ _t

/ / 1

# lI ,f /

I I I 13 4 5

(_ lnclirl_ion Of lh_ olr stfellm.

r-I" ,--I

"4!

,4

NACA

E

"_" 3--

(b

_o - __ "__//

o-2--

E

O_E

J_

-;]p--

16

Figure

4_

_w

p t

--4--

I I I I I I I I I I.5 4 ,3 2. I 0 I _. _ 4-

Dmtance from center" fine, f't(0) D)'nom/c--pressure (q/qo) contours.

i3. --A;F flow in the plane 0? the' elevator hir_5_9 4_r?e. View

Ioc_in 9 FOrW_rd'joCr,lO.q _ 4_0; propeller removed.

\\

\\(1) 0--- _

E ' J 1 'o_,. I _ \

0

c5

%

t

/

/ / l _ i 1

/ I I6 5 4

F;9 ure 13.-COmclud¢4,

\ /g

I 1 I I I I• _ 0 I 2 $

Distance from center line_ _'t('b) lmClirnot!On Of" +.he olr StreOr_.

I I I

4

I

NACA

3o>

S

US

I°o )

I I I I I I I I I I I I

Distance from cen_er line_ ft

3--

,E 2_w

R-

0

;_-s--

Ficj ure

-4-

(c_) Dynomic-pressure Cq/qo) contours,

_4.--Air flow rn ghe plane of" the

-5--

I I I6 5

F_gur_ 14,- Concluded,

\ \

/

f

Sc_16 o_" vect.ors

o io ,!o

Deviotion= degI I I I I

Dist, ance from cenger line=f_

(b) I nclina_,ion

3

o£ the air st, re_m.

\

-J-- i-

I I IS 6

I_4

I/aCA pig. 15

_T 4--E-- ,0

.E

_5

I-- .0

O--q-

aa

.8

I I I I I I I I I I I I8 _ "1- 3 P- _ o I & s 4 S 6

Distance £rom center line, ¢t

CO) DynamJc pressure(q/%)contour_.

F c .re 15.--_ir {Io_ in, the _;ur, e cfChe ele'/o,_or b_ngel(ne. View looking

¢LT, 15. I°_ d'f_ 40°; pr'opelier" _emoved.

5 m

4_4_

@

:5 2u

-r--

L

_rw

-3--

-4w

I

I

\

I

/

/

_-orw3_d;

I I I_5 _ 4

Fi jr'F Lf-£ " _ udu

I

..Scaleo{' vector3

0 I0 _0

l DCvi0tion_lI de'I I

2 I 0 IDiSt,3ngC from c8r_er lilge_ ft(b !/nc_Fr, 9t ior_ ¢{+.he e;r 5±regrn

I

L

NAEA

j_ j /I.o

_° ......... k \X[/2,./3 N

-s- hi,_ (=o / \

I.I

I I I I I I I I I I I I I6 _ 4 8 P- I 0 I P_ _1 41, _ 6

Distance from center line, ft

(a) Dynamic-pressure Cq/qo) contours.

Figure 16.- Air flow in 1::he plane o£ ±he elevdtor h_nge line, View Iook;n 9 1%r-

wordldT,--O 2 °" @. C°- %, 0.014.

6 r-- _

c_

G

g

E

g-

I._ -3--

c

/

II• .ip

iF a,

, /

t

_COIe Of v_otors

r_r_r_m

Devi_tion_ dle?i I I _ I I I - I6 _ 4- .3 2. I 0 ! P_

Distance from center line, ft

Figure 16.-Conc}uded (b) fncl;notion of eir _trearn,

I t I I3 4- _ 6

-4I

NACA Fig. 17

£-

ZO-- -

_4-- -7 "

o

E

LO

I I I I I I I I I I f I-$-- _ 5 4 3 & I o I 11 _1 4 l_ 6

Distance from center line, ft

(o) Oyr_ornic-pressure Cq/qo) conLom-s.

Ffgul-e 17.-Air flow inthe iolone or-the elevo-{;or h;ncje Ii'ne VJ'ew Iookin 9 for-

W'or'd]O(T,--0.2°_ dr, 0°) T c, 0.04.

E

_Z

¢

E

o

@

2

-3--

--4--

I I I6 ,5 4

Figure 17.-COl;C!,JJed

J • /

P B

I,3

Scale of vectors

0 I0 _0

D_viotion, deg

w

I I I I I

Distance from center line, ft

(b) Inclina_r/on of the olr _±r-e _'_,.

I I I,'_ 4, .5

-#I

NA6AI w

-l -- ,q _I.7

®

N_ I_

_-_$__

_'ig.Je

),,,,,.,,.,,,,_I. i l.ll

f?Z/

i.l I,! f.o

\\,-oI,E I.!

I / I Is 4 sG

I I I I I IP. ! 0 I I_ 3

Dlotence from oen%er line, ft

2--

di m -w= _

L

t= -#_--

°9.r,,,_

-3--

-4--

DJnamlo-pressure: (ql_lo)contours.

--Air-Flow In "the pfene of" _he

elevcf+.or hln_je line.

View IooKincj ?orword_ e_,

" - t I

I

I I I I

8caler._C_cor3O I 0 P-O

Deviotlon., cIe(:j

I 1 t I I_- I 0 I P

Distance from center line, t't(b) Inclination o£ the oir 3treom.

F_gur6 18. --Concluded.

rdr_

-_t

NA_

4:

c° 1-0

-'_- I I L I I

i0__

I I I i t 1o I P_ 3 4, ,5 i_

Distance from center line, ft

Fi 9. te

I.O

p_

e-

l-

L.-I--

o4.11

}_,,-

Q

Figure

(o) Djvnamic- pressure. ('q/qo) contours.

19. --Air flow In the plone of the

elevator hinge line.

View Iookln_j forward; oLT, 3.1";

df, o'; %, 0.03.

/

I

I4,

G 11

• ,,,,i,

J

w

a

_:ale of vector_

o io £oDeviation, de_!

I I I I I2 I 0 I P-

Distance _'rom center line_ Ct(b) Inollno_ion of the. air .?{._-eam.Fi(jvre Ig . -- Concluded.

i I5

I"

I8

!_q

.8

_.0

Fig:

(a) Ddnamlc-pressure (q/qo) contours.

Figure Zo. --Air flow in the plene'of the

3m

c

c

_'-\ X " \

% % • -.

E

q- \ / JI1)o

{z-2_ fO

g °

elevator hincje line.

View looKin 9 forward; _T, 6.8"I

_f, o'_ T,, o.18.

t

L

/ /

/ // // l

/ s

I I I I6 5 4 3

_cole of' v_ctoFs

o Io 20

D evkltion_ de9

I I I I IE I 0 I l}

Distance £rom center line_ Ct(b) Inclination of" th_ (_tr stream.

Figure 2.0. -- Concluded.

I I I I

Ilq'ACA

+;

,qI I

Lq L J I.Z 1.3

!o

-4--

I I I I6 5 _- 3

F io_ z)

1.4-__13 ';'4

I._-----------

,.,_"°

I I It ¢.. 3

1,3

4 5 6Distance from (:enter line, ft

_Sm

E

c

_o--

Eoq- -| I

g

o

-Sm

-4m

Fiqu('e Zl.

\\ \

t /

(a) O_.ynem¢- pressure . (q/qol Contours.-Air _'low ih _he ploge of t.h6^ elevaf.or hinge, line.

V;eWd,lOol(in 9 ¢or_verd{ (_r, 6.B36f, ; To, 0.11. .

/

.D

't'" /\\ i

- x ",. 1" " " / / 1 1

/ " " _ " / 1 1t - \ , / / i \

ir

/

%

I

\

i /

/

I I I6 5 4

Scale of vectors

o io 20Devidtion_ deg

I I I I I I I I3 _- | o I 2 3 4-

Distonce from center line_ ftCb) Ir_lina{ion Of the. oir S'creom

Fi3ure. 2.1 . -- Concluded.

I

-t-1.,t

Distance from center line, f_,

Ca) Dyno.mic-pressure cq/qo)contours.

Fiqure 2 2.- Airflow in %he plane of the elevator h;ncje line. View t_l_(n_ _rwo_t_c_T_ 10.7 °) 8f, 0°; To, 0.31.

4_ 3--q-

c

C

t- I--

\

_o--_

1EO-1--

t_

-3--

-M--

\

\

\ , \ N \

/ / " _

" \I

\

\

i//

I I

Figure 22.-Concluded.

\

//

4 3

//

11W

t /

1 /_cal_ of vector_

o Io P.O

Deviation, dQgI t 1 I II 0 I P. $

I)tstLanee-from _levator hlt_e IIn_, _td_ Inclination _f _he air s_reorn.

/

\

I I' t4 8 •

I

Fig23

\ \ ,

\1

\

\

Stole of vectors

F_F_0 IO PO

1 jDeviati°n'J de_ [

I o !

Distance from center Dine, ft

(b) IncJl'no-I:ion Of the oirsfream.

/I/,

dc

C

&z

0 -I_9-

0

I I I I6 .5 4 0

rig,_r_ 2_5.-Concluded.

I I I I3 4- .5 6

=

i

,--I

I ,,f,'_

I-

I I I I I I I I ! l._ I I

II ._ 4 3 I_ I 0 ! I_ ,II + 5 G

Distance ?_m _w_i_er tlt_I, ft

(0) D_/nClm_c-l_ure ('_t%) Contour._.FigureZ4- Air#low in_hee evotor hinge line.View _ook{nq forward; (_T, 15.C_l

6f,4c?lT_,0.0&

i

NACA

\

\

I I I I6 5 4 3

-6--

\k

Figure 25.-C;)ncluded,

.Scole o£ vectors

0 I0 PO

D eviotion, de 9

t I ] I I2 I 0 I

Distance from center line_ £t;

(b) Inchnot_on of "t;he oft streom.

I I t3 4.- S ,6

k_

4

NACA

I-f

I" |.

-6

_. u_l 0

I.

o9

0

I'!!I

I

I

I!

d

/I

.... Jl

i;t t

iT1......i/t

,/

/i t

/w

C) oi _ |,

Ficj_s. 2.7_Z8

_2

' rL _. ,,

_t

w 0

I

b.

1.J

!

@

mP

NACA .2.0

f ==_ _ j 3_

_---_-- __P

.... _ :_-_'____'_-2__ ,.-_...__ .11 --.I0 ...._ _ __.____ _

H ____ "--- _____. ._.

o C_ _,-Z 6 ° _

4 t " °................ _ ..... ¢o ......... _- --

i (t_} Lt l._, , / -'__:i -';'s ! I

e _ l 0 I-- 'I --I r ...... _ " I [ ................ 4

I

................. L-4 .........._ ° .

_=1_ .zs,._zs, 4o ,.,_as

:_° i ('-3 "t • t.2. ° _.__" .oq , ',0% i _"_" .07.- ' I

i /1o I } ! _ --

- 18 9.l_ i i

) ......._I _ ..... * - q• .--_ 7 05 : ,

_/ J;:5- •o_ I i i

- 4 0 4 8 12.. 16 2._

_3- _,ff¢_nt _,_ s_ffm{_ _ fl&ps _a_{4, Doffe4 curve_ &re _rom e#rapol&T_8 wl_es

-- i--Ii-4

I,4

NACA Fi 9, 30

0

0

0

-4"

,,S

NACA

,9

......

,_\

0

,,,,.

°

.#-

I.t'4 __ N

fr

_%'_ e

.-- .._ ed

3_

o_..?

g..a -_E o uo

{.

g

Fi cj,s. 3_:,3.z

i-Ir'-I

I

NACS

I I

.... T---7 £' II

t

r

' 4- !

i I

Q

_4 -.. O

I

I

i

Figs. 35,34-

3

.j

!

,@

I

r

F

NACA

__ _ +

I

:1:1

/ j

j /

f /

//,///

13 " I

:_ __-_ _---

_ _ _

' I I _ /

(/'5 °

o _

,0

, Ii

!

!

It ....

__.J. .........

t"

_g r_

'_. _ --

_ _ ..._ _

, _,_I.

, _"

la_

_i9_. 3_,36

Ib

--t!

NACA FInis.

.I

w, -I.S° _?.o

g-:.i

I

ExI_I_| I I IC_Icul_le.cl,w ifh (_+a_ +aez

-Z -5.0°

Co_°

J

.-_ 8._°

I0.I °

_..---_ 14.fo°

3-/,38

0 .I Z .3 ._- .5Th_sl- coeffi_m_,'F_

,4r----

I,3I_ •

--4 ....

I

• I ....

I IExp • _ i_ e_ta I -

t _'_+"_'+_ ___

0 t .2 .3 .4- ,5 .6

Thr=St co,ff,k,'evT_T=

F_re 38 .- Co_npa_i6c_ oy _xl_rlme,i'_,l &_ calc_l_t_l_o61

i'la_)5dcYlec_fe_l"_g:--l&°.Ca_¢_l_idValwe$/lave_ ob_a_"_eawi"l'h 6ca_+A_,+a_ z.

lIP

,-I

I

NACA

.___

!

:-- ;I+

t

F/ ]

• ',I !i

_ ., 11'_ I_ '

/

! __ ":1_ _kP',''_"i

tl

L_ ....... i .....J I(_ --'. ,.

[ •- _ _

._

p_

0 _

1

'-I-I

!

L

E

F

i1'

F_

L

ii

NACA FicJs. 4-1:,4.2

t-I

"4"I

,.q

N,_C/- _r" _-rigs. 4-3,44

-'_ 1- ,,,_ l r- --I _ IP, ._

....... ,:, .... ,. _ __]o H <l.-c,I "I3 • _ -0-" o

!o _ o _

'::' o _V" _, o o _. o o ' '_' '_ -- "- "

<>_ _ / o_- _. _:_: I I o,.,I J//_. I I I ._/ 1/,_ .,,

I I /I I ; , ._ , '. ! I', ,, _ iZ._.

! t/--_.....4 t;_ 4--....i--t/, .ti _

a_._, 4._t!_)jj._o ) .I_U_,_oLu. - it_lt, i.!H '

cJ

_ .... _.i ' i

_01 ' i --

2 , it3_

:2,

i ----....... L_,_,D $,_. _,<,_,.t,._....

, _ ! ! I

i i _ i : ,, _j_ i ...... +-- i

_--l- ....

-- _£ .... L_........

- 4 0 .'_ B 12- 16 20

An_le of _."ffa.cK71" iCle_

F,_._-_,_5 -

i=:

I

NACA

"+.S,I::,p _1_)p

J' ",4"!

NACA

I II! ]

-- I r_vJI

!

°!JP

cO

¢.

Fig_. 47_4a

--1 .... _ ! _. _

--t ....t ÷-f ....i

I

t

............. __L_ 1 __

/ i

', _L

I"

4:

o

|

4,-

_ ,.,...

u

o

6*"

.f "_ ,'T"_ _ ".-4

.,,=_

I

h,..