Top Banner
1 .. NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS TECHI\ICAL :\OTE No. 1208 OF AN AIRPLANE MODEL WITH SKEWED WING IN THE LANGLEY FREE-FLIGHT TUNNEL By Jo hn P. Campbell and Hubert M. Drake Langley Memorial Aeronautical Laboratory Langley Field, Va. Washington May 1947 https://ntrs.nasa.gov/search.jsp?R=19930081890 2018-03-18T21:32:43+00:00Z
38

Investigation of Stability and Control Characteristics of an Airplane ...

Jan 11, 2017

Download

Documents

NguyenDat
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Investigation of Stability and Control Characteristics of an Airplane ...

1 ..

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

TECHI\ICAL :\OTE

No. 1208

OF AN AIRPLANE MODEL WITH SKEWED WING IN THE

LANGLEY FREE-FLIGHT TUNNEL

By John P. Campbell and Hubert M. Drake

Langley Memorial Aeronautical Laboratory Langley Field, Va.

Washington

May 1947

https://ntrs.nasa.gov/search.jsp?R=19930081890 2018-03-18T21:32:43+00:00Z

Page 2: Investigation of Stability and Control Characteristics of an Airplane ...
Page 3: Investigation of Stability and Control Characteristics of an Airplane ...

~-~-------- - --

NATI ONAL ADVISORY COMMITI'EE FOR AERm;AUTI CS

TECTINICAL NOTE no . 1208

INVESTI GATI ON Of STABI LI TY AND CONTROL CffiLRACTEnI STI CS

OF AN AIRPLAIIJE MODEL WI TH SKEI{ED WING I N TEE

LA}'I~GLEY FREE -FLIGID' TUNNEL

By J ohn P . Campbell and Hubel't M. D:cake

SUi: 1M/illY

_'ill j.nvestigation to determine the otabili ty and. control char ­acteristics of an ai;:-plane model with a skewed 1vlng has been made i n the Langley f r ee -fligh t tunne l . 'J..'hG 'vinC of the ni.odel 11"ae pi voted in such a .ray tha t it conl d be r otated as a uni t 1{i th respe c t to the fuselege so tha.t one sic.e of the ving \Tas Sl-Tept fonTard. and. the other sid.e svrept back . Hi th an d.rrnngement of this type the wing of an airpla..11.e could. be set at right anGl es t o the fuselage for take -off , lanclin3 J and lo~," -E::;?eed flight ana COUld. be rotated to some lar e angle of Ske1-l t o perltli t flight at high speedS .

I n the investigation, fl i ght teats , foX'ce t '3sts , and demping­in -roll tests "rere made on the mod.el vri th the wiag set at angles of skei'l from 00 to 600 • This investigation iias 0 an ex-.rloratory nature and was intended to provide only a preli minar: e.:1cl quali ta ­ti ve indi ca ti on of vThethe::.' such a o.eeign coulet b e f l own .

The resul ts of the invGstigation ind.i catecl that it was possible to sleevr the wing as a lUli t to angles as g!'eat as 1~00 ,,,i thout encOlmterlng serious sta,b ili ty EW."l.d control difficulties . At an. angJ.e of skev of 600 , hovrever , the aileron control be came l..ms8,tisfactOl'il:r weak . The ailero rolling effecti veness was not l'educec. by s::8I'Ting the wing fTom 00 to 400 because the damping i n roll decreased app:toximately the srune amount as the aileron rolli ng moments . The force t e Ats s~Wvred ·t..l-J.at for a skew- angle of 400 the ailerons produced l ar Ge pitching moments , but i n the flight tests no pitching tendencies were observed in aileron 1'011.'3 , apparently because the lift fo!'ces on the wi n produced by rolling introduced pitching moments that were equal and o~?osite to the ai l eron pitchi ng moments . The model did not eJ~ibit the undeSi rably l arge variation of effective dihed.::.:'al vTit h lift coeffici ent. that is characteristi c f vlinGs 1-11 th large amounts of svTeepoack or sveepforward . Skewi n,S

the .ling as a unit, hOi',evel~ , did introduce large changes in l ateral trim '11hi ch varied. 1fi th lift coeffi cient and skew angle .

Page 4: Investigation of Stability and Control Characteristics of an Airplane ...

2 NACA TN No . 12 a

INTRODUCTION

Theoretical and er'perimental investications hove sho'WIl that c0mpressib 11 ty effects on a '\-."in£:; can bEl deleyef:. 0~- slTeepinG the wing fOl'vrard or' backl-;ard. I n orc'er to obtai:l a l a:.'ce incrense in the Mach numbor' 3.t w'hich ~om:pressibility effect s occur; the nse of angles of svreep of 1}00 or more is necessary, bu_t these large ansles of sweep introduce s erious stability and. control problems at moderate and hj.e..h lift coei'ficieI~ts . For ex:!nple) 181'136 e.nGles of BI'Toep produce undesil'ably la1:'Ge variations of effocti ve dih3dr31 and pi t chins -momont coefficient "lith 11ft coe~ficicnt . Also a rapid increase in drag OCCU.1'S at motierate lift coofficients ",hich is detrimental t o take - off and climb perfOy·w.1l:lCe a...l1U which compli­cates tho l anding probl om .

In ordor t o cain the advantages of S~'Toe:l at hi6h speeds without experiencing the d.iff ict'.lties introduced 'b~" 8':1Oup at 10'l'! speeds J it has bee::1 proposed t ha t an airplano be 6q,uiPP00_ v;:l.th a ",-:tng pivotally att a.ched to the fuselage so that it can be s e t ot rie.,ht angles to the fuselage for take -off J landing) and lOi~-spoed flight and a t some angle of 31.,rOOP for flight at hi{;h speeds . In one suggested design the Hing is s}:ewuc1. or pi voted 8S a unit so t hat oni3 sid.o of the wing is s\·wpt fon/erd and the other sicte SI-TOpt back . In order to QscortDin the low-speod stability and c')ntrol character ­istics of such a ctOSir:;rl ::m inv )stigation hes boon concLucted in the Lanelcy froe-flig..~t tunnol. This invostigntion c oneistod in flight tests, forco t osts, and dampinG- in-roll tests of a IDodel equipped '\-ri.th 1:1 pi voted vTing that could bo set at variol'.s 81161 s of skew from 00 to 60 0 .

This invosti ga tion vTas of an oxplorator 3' na turo and "JaS intended to I'rovide only a prolimi.nary and qu<-< li tati.ve in'~.icQtion of whether such a desien could be flOl'T!1 . No attempt is thol~ofore made in this paper to give a complute and cOID::?rehonsi vo c1..isc~s:Jion of tho. sta ­bility and control problems involvod in SkO'Ioloc'··vTinc; designs . Tests at higher sca le of skm'Tod-vrlng moduls more r opros.:;ntat1ve of high­speed airplano dosigns will pr obably bo n ooded poforu an acclrrato and detailed analysis can be 1llCdo of tho stabil:i:ty ;IDd control characterj.stics of this typu of airplane .

Sl11BOLS AND COEFFICmNT8

Th0 forcos "ond coofficionts i1OrO moasu.ror.l 'vi t} roforonco to tho stabili ty axes . A diagram of those GIGS Sho -yri!1: t"_c posi ti V0 diroc­tions of foJ;'cos and momonts is presented QG fi sv.re 1) "Thich sho,\·TS the vTing in the symmetrical (ullokovcd) condition . For Gkowod condition tho axe s arc determinod by the fusolage r athcl' 'c!l(m -DY the wing .

Page 5: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No . 1208 3

M

L

N

s -c

b

v

5 a

(_L~ft \, lift coefficient -\, qS /

drag coefficient (Dra g\ . qS /

p'itching moment

rolling moment

yai-ring moment

pitchinG-moment coefficient

rolling-moment c r)of fj cient

ya~~ng-moment coeffi c i ent

/ Lateral fo:~c o\ l a t era l-force coeff ici ent .-----.----. )

\, ' CIS I

vr1.ng area> sqnar e feet

mean aerodynamic chord of t1Ilske1ved irinG, feet

span of lLTlskel{ed wing, feet

d~mami c pressure , pounds per square foot

airspeed, f eet per second

mass density of 3ir , slug per cuM.c foot

anel e of sidesli p: decroes

anr,le of yai", c_egrees ; for force -te~t cl.J:::to equa l s -~

skevT anGle ( anel e throu:-;h vh1ch ",i<; io rot8tec: vlith respect t o t he fusel aGo )) deGreos

aileron def15ctlon) degrees

el evator deflection, degroos

ruc:der defl ction) degroes

Page 6: Investigation of Stability and Control Characteristics of an Airplane ...

~.

x y

z

pb

2V

p

NJ\.CA TrJ No. U:oG

angle of attack, d.egrees

longitudinal axis

lateral axio

normal 0. xi S

ra .ius of g)rrution of model about X- ax-ts , feet

r a d.h1S of gyration of mod.el about Z- 8.yis, feet

aileron-roIling-effectiveness fa ctor or helix angle generated by ,·rinS tip in rol l , r aCLi.ans

rolling fl n{jl.uar velocity) r a dians per second

effective- dihedral par aI!leterj r ate of change of rolling­moment coefficient with angle of sicle slip, per d.egree (COL/ dB)

On d.irectionaJ.-stabili ty parameter; rat e of chance of Y8l-ring-f, moment coefficient '"I'li th angle of Did.esli]? , per

degree (60n /d13)

d.':lllIpinc;- in-roll par ameter ; r ate of change of rollinB-moment coefficient I ,d th rollinG··an c:ular-veloci ty f actor

/ 0C~') / -\ d~~ .I

A1?PAl:1P.TUS AND HODEL

The flight tests and f orce tests \,'ero conducteo. in the Lun~ley free-flight tunnel, a complete d.escripti on of whi cb :L s given in reference 1 . A photograph of the tunnel tes t section in. th the model in flight is prosent ed. as fieure 2 . All force .tests were ma de on the fl'ee- flight - tunnel ba l :mcc (reference 2) i,'hich mea surGS forces and moments about. tho stabi l ity axes .

The v;:.: lues of tho d.amping- 5.n -roll o.eri v2.ti VG c~ were p detormined by rotation t esta in tho LElnl~ey 15- foot :free-spinnin.s tu~~el by tho method d.escribed in r oferenco 3,

- - - ~-------~-

Page 7: Investigation of Stability and Control Characteristics of an Airplane ...

NAeA TN No . 1 208

A s ~etch of the model used :i.n the t ests is pr esented as fi l::>lITe 3, and [3 photoer .::lph of the model is presented a s fi gm-e lj. . The moo.el fusel2.{se consisted of a boom on which was mounted the rotDt able w:i.ng .snd fi~:ed horizontal and vertical t ails . The 1vine vas mounted by a pivot at t he 50 -percent - chord point so that it could "be skewed t o anY':lngle up to 60° . L'l all tests the i·ring \\'8S

s kewed. so t:1at the right 3io.e "res Sr18pt b~ck and the left side Si'18pt f orward, a s s ho"\-IIl in fi 81lre 3. The dimensional and mess char a cteri stl cs of the model are rsi ven in te blo 1.

TESTS

Force tests 'Here macle to determine the stati c stability nnd a iler on control characteristics of t he model ,\Tlth ske,·; s n r les of 0°, 200

, 400 , an.d Goo. For ce tests ':-rerc also made t o det81w.ne the 101lcHudinal stp,bility of the model vn:t,h horizontal t ail off for skei; an{Sles of 00, !}.()o, and 60° . ro control- effectiveness tests i-TOre made for clevator ana. rl-'.ddcr . All force tests • .. rere made at a dynhl:li.c pressure of 1.9 pounds pCr 8~1J.are foot "Thich corresponds to a t est ROuTIlOIds nUrlDCl' of 179,000 <c! t 0 0 s kow' based on the mean aer odynamic chord .

All r olling-mom.ont and m-ling-moInent .B t u obtr..: i::-lCd in the

5

t es ts ilith d.ifferent skev! angleo are ~asod on tho sp"'n of' tho l.ms;:m·!od l;inS and all pitchi.ne -moment d.a t.a '.r e bQrJod on the mean aerodymunic chord of the Ullskouod ,ving . AE. t est da t a are r oforr ed to t he nor:nal contor of gravi t y at 20 p ercent of the mO"Jl 30rodynoDLi.c cho~~d of tho UJlskCl-TGd .vinc; unless othenr.!. S 0 notocl all the figure s . If it is desired, t he momonts can l)e based on the spFm ana. chord of the sk81'led wing by using t ho valuea g.i ven i n t ilble I f or tho differont skew an ·los.

Rotation tosts vore IDede to detormine the damp: .ns in rnll £'o!' the modol 'Hi th ske,\.; an glos of 0°, 20 0 , 400

, end 60° &t D Uft coeffic~.ent of 0 . 5 . All d3IlIpine;- in-:coll t.osts "Toro made a t a dJ~malc prossure of 3 .0 pounds per square foot which corrosponds to a Reynol.:1s ll'l.l.l!!.l)er of 226;000 at 00 sko,-l.

FliGht tests of tho modol ,vi th tho centor ("If c,ravi ty a t 0 . 20 mo::m aerod.y.namt c chQ}:,d of the 1..mslmved l~nG 'Ivere Inacl0 at a lift cooff'::' ciont of a-p})ro::dTIlr':ltely 0 . 6 for skou angle s of 0°, 10°, 20°: 300 , :.0 0

, 50°; ana. 600 • I n addition, for the same center ­ol- (;!:'crvity location, fli g;l1t tests ,vcre 1lUlde OV01~ a lif t - coefficient range fro:n 0 . 3 t o 1.0 f or 0 0 aml 400 ske"l. Furth,:;r flight t ests "lor e Il13de 'I·Ti t h 40 0 skev at lift coeffici ont s from 0 . 6 t o 0 . 9, Qnd tho center Oi Gr avity vres movod suc cessively back t o 0 . 25, 0 . 30,

Page 8: Investigation of Stability and Control Characteristics of an Airplane ...

6 NADA TN III o. 120':

and 0 . 35 mean aerodynamic chord of the unekewed Wlng . In the fllgpt t ests, abrupt deflections of approximately ±18° aileron ( total 360 ), ±5° rudder, and ±5° elevator were used for controlling the model. P.eference 1 describes the flight - testLTlg technique used in the Langley free-fli~lt tunnel.

BESUIJTS .AND DISCUSSION

Force -Test BestLlts

Lon3itudinal stability.- The results of the force tests nwde to c.e'feriiilIi.e- the loneitudinal stability char acteristics of the model are presented in fiG~0s 5 t o 7 . The data of figure 5 show that as the skei" angle of t ho model '·7~~ S i ncreas ed from 00 to 60°, the static l ongitudina l staljilitY (63 i '..1dicated by the s lope of the pitching-moment curve ) progres sively decrea sed u..l1til at 600 skev the mode l "Ha S longi t udi nally Ul1stable at lif t coeffi ­cients abo\'e 0 .'7.

The data of fi gure 6 show th1;l't "\vi th the horizonta l t ail off the r eduction in longit udinal stability with increasing skew lJaS

even more pronounced. A cOl.'lJ;Parison of the dat a 01 figures 5 and 6 indicates that the add.i tion of the hori zor;.tal t3.il groatly reduced the voriation in 10nc:Ltudinal stability over the lift r engo fo:c the 400 and 600 ske"\\red 'dngs .

Tho data of fi gures 5 and 6 ShOVT ·the e:x;pecte(l r eo.uction in lift - curve slope vTith increasing slec"\". The slopes of the lift curves are ap:proximateJ.y proportional to the cosine of the angle of skew.

The force - test da ta of fi gure 7 show the longitudinal sta­bility and trim chara cte:ci ctlcs of the model 1dth J+Oo sk(3"\v for the most f Ol'"W3rd and r oar ·rv. rd center-of - gravi ty locations usod in fli ght tosts . These data sho'\', that for the most r earward center ­of- gravity location tho model vms either statically longitudinally unstable or about neutrally stablo over the entire lift-coefficient range .

Later a l stabllity.- The results of t~e force tests made to dete:r.mine the later3l - stability chara cteristics of tho model at a lift cooff _Gient of 0 . 6 at z ero anGle of ya \'T ar e shov711 in fi f,'Ure 8 . l 'he offects of eke" on tho directional-stability parameter Cn13 and on the effecti v o dihedral .para.meter C~ 13 as determined by the

slopes of tho curves bati-Teen ±100 in flcvxe 8 can be summ<:;rized as follows:

Page 9: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No . 1208

81m if angle

0 .0034 -0 .0006

I .0033 ' - .OOO~

__ ,-_ __ • 0_0_2~ ___ 1_-=.' 0006

The values of C ~ p f'or the model are consJd0rallly smaller

than the values norma ._ly encolli"1tered. a t 11 lift coefficient of 0 .6 on wings "liti. l ar ge 8!lountS of S1{eep~8ck .

The lateral-tri.m changes c/':used . y sko'Fin,g the i·dnB from 0 0

to 40 0 ena. 60 0 can be seen from the dcta for zero c.n~le of ;yavi in figure 8 ~d f r om tl 6 laterel - con~onent dat n of f::.gu;'es 5 and 6. 'L'he data of figu.re 8 for a lift coofi'icient of 0 . 6 sllo'\,- t h ·.., t u:i. th inc:ceasin3 SkSI-T an increasing positive (right ) l E:lteral force and an increasing negat,j.ve (lef t) r ollins ornent occUJ:-red . Changing t:t.e ske\~ from. 0 0 to 400 cAus ed a neE,ative ( left ) ':..i·ring moment but a f rrther i ncrease in ske".T t o 600 :proc.uc'3d a posi ti vo (rieht) ya'\dn g nomont. The 6..ata of figul'cS 5 ... ,nd 6 sho';7 t· mt th3 c an3es in lat eral t r im varied considcral ly .lith engle of att6ck .

I f the pi vat point of t he \. nl} ,lers shIfted forvT~rd the chan ge in rollins moments vr.i. th ske.; at lOll lift coefficiOl ts ifould be rcd wed because the ar ua of the l e: t '\Jin3 '·.'ould bo i ncr eased vrlth increasin,3 skOi,~ .

LnterEll .control. _ - T 10 results of the for ce tests B :ldo to det ermine the a ileron effectj.vencss aro shov71l in figures 9 and 10 . These results show t ha t t he a iloron effectiveness in producing rollins mOIllent was somowhut reduced by slcovTine; tb.o ,d.n6 to 1~00 ane:. "TaS e:coatly reduced by Sb:liv:1.ng tho w1,nc to 600 • "w110n deflected the ailerons at the 1\.0° and 600 olco~·; aJlt;les also ? T O ucod. s1z:3ble pi tchin13 moments . Thei telling m.omont :p:roduced by the a.ile:..~ons ,,;as rJ.uch greater at .00 SkOlv Lngie t han a t 600 , appar ently, bocm:so of the reduced effectiveness of tho ailerons in :producing l :lft at 600 ako'li.

The da ta in f i (S1.1.re 10 J which 8hm'T tho independont contrj.1JU­tions of t he r i8-ht EU1d lol't a ileron s t o tho cor d;ynamic m.omonts, incliceto that the l oft ( loading) aileron was most affocted by skew 811g10, and th:'!t a t )~OO and 600 skew its offecti venoss ::'n producing r olling moment was somewh:3 t l ess than that of the r:tght (trGiling)

Page 10: Investigation of Stability and Control Characteristics of an Airplane ...

8 N!-\.CA TN No, 1208

a ileron. This difference can be attributed to the increase in "i:-lnG area ahead of the trail.ing a~ leron.

Rotation-Test Results

The resllts of the damping-in-roll tests are shown in figure 11 tOlScther ,d th the ro11 inG ef'f'ecti veness of the ailerons based on the rolling-Llomcmt data of figure 9 . These d.ata show' t hat the damping in roll "?as r oa.uced by skew t o such an extent t hat the ailerons remai:!lec eff ecti ','e to eng.les of skew gr'eater than 400 •

The clalJl1)ing-in-roll ' c.a t a b i.A s ecl 0:1 the projocted span at ea.ch IJke~T anglG (rather than on the spqn of the i·'J. ~lv .. 71th 00 ske .. ,) are presented in fi g'JrG 12 for the rmrpose of com:purinr, the data 'in th ce lcclations made 'oy the simple r e1.::.":.ion I'~efJe~ltGti. in reference 3. This fl cut'e ShOH8 that JC:1G dan;ping in roll v[.ried cpproximately a s the cosine of tho S};:O"'l·l an ::;le, as WDulct be expected freID. the data of r(;ference 3 for cO!lventional1.y svTOpt -back idnga .

FUght-Test Rosults

In tho fU ght t ests of the model .. ;ith t he contor of gl.'uvity at 0 . 20 meon aorodJ:1a:.m.c chord the genoral flight cho.racteristics VTero sCltlsfactory and. rcrnn.1ned oss oriGially unc.13n30d. a s the "rlne ,vas SkOi-T3d froLl 00 t o 400 by 10° -inc:;:'oL'lonts . \-lith the ir:ing skewed. 500 tho flight chara ct eristics of tho mecLol i'Tor o se.tisfactory ey.cept t hat the ileron effectiveness , :as noticeably reduced and somo diff iculty "'I·TaS conso~uently oxper:tcnced in controlling tho model. ii Hh 600 SkO"'li the aileron offectiveness "\',8S even further reduced C).nd .. ras inado~uatG for :rnaintain1ng sustained :flights of sufficiont l ength to POJ:'!l1it j udginG the ot or stabiHty and control charectoristico of the model. Thes e flight-tost r esults nr o in agreoment with tho f orco-test r osults of ficures 10 to 12 in regard to the reduction in aileron offecti veness with 600 ske • .,.

Io prono'.mced chc:m~o s in stability and. control vore apparent with ske"1 cm glos up to 1.:00 but sizable ch8nges in l ateral tri m -v!Bre noted, 1-:hon t ho 8kew angle "as increased f r om 0° t o 1~00 while tho fli Cllt lift coefficiont 1-18S held cO:J.sJwnt at ab01..t 0 . 6, uso of a t0ta l of about 170 riCht aU.cr on trim and 30 l~i eht J.'ud.dcr £lotting was necossary to rnaint3in l.r:Ger al trim, (t .... lOt i s, to koep tho vTing loyel and. tho fuselaGe at z oro sideslip ). As the fli&~t lift coef­fici ent ,':as increCt seo. at 400 ekev!, howver, :progl.~esfJivel.y 8Il1:ll10r amount.s of [.dleron and rudd.or trim were re~uircd 1.1Iltil at a lift coeffici ent of 1.0 no trim "WUG needed. Thoso trim ch3nec13 lIDrD

indicated by t ho forco ... tost rosults of' figures 5) 6, anc:. 8 .

Page 11: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No . 1208 9

lihen the skew angle was chanGeQ from 00 to 40°, only slight changes in elevator setting (not over 30 ) 'vere required to maintain the same fUght lift coefficient . itli th 40° s1;:ew , hm-rever, the glide -path angle "Tas 2 0 or , 30 higher than wi th 00 skew over' the lift -coefficient "-'ange from about 0 . 3 to 1.0 . Thi s result is substantiated by the fo"-'ce -test Qata of figure 5 "Thich indicate

'that the Qrag at 4c o slee", is higher than that at 00 ske',T at a gi ven lift coefficient . The force -test results in fignre 5 8hmr that at lift coefficients Greater than 1.0 the drag at a giv'en lift coefficient (and hence the glide -path angle ) f or the 40° skew angle became increasingly. gl'ea ter than that of' the unslceiveQ ,ving .

In the flight tests of the 400 skevred 'lorinG satisfactOl'y flights were made over a l ift -coefficient range from 0 . 3 to 1 .0 with static margins (static 10ng5. tudinal stability ) from lar ge values t o very small values . The fli@1t characteristics appeared to be slightly better at the ImTer lift coefficients . No pronounced changes in the longitudinal stability characte"-'istics viel'e noted as the static margin was progressivel y decreased by moving the center of gravi ty from 0 .20 to 0.2) and. 0030 mean aerod:Jl18Jl1ic chol'Q . 'I{i th the center of gravity at 0 . 35 mean aerodynamic chorQ, however , the model appeared to be lengituc.inal ly unstable and continv.ou8 application of elevator control Ifas r6quired t o keep the mOQel flying . The fo"-'ce-test d.ata of figure 7 indicate static longi ­tudinal instability for the forogoin6 conQition .

In the fli sht tests no pitching motions 'with aileron control wel~e noted for any sko>-r angle , lift coefficient , or center -of ­gravity location. This result appears to disagree "lith the force ­test results of fiGUres 9 and 10 vThich showed 8i zable pitching moments ,vi th aileron deflection for lj·Oo skeYT . This apparent Qiscrepanc;y- is explaineQ by the fact that Quring a steady ail eron roll , lift forces Que to rolling are proQuced "'hich are equal and opposi te to the lift forces proQuceQ by the ai1e:.:,ons . These lift forces due to r olling produce pitching moments that arc equal and opposite to the aileron pitching moments and hence oliminate the pi tching tendencies in a steady a iler on roll . No flights were made to determine the effects of these a ileron pitch:i,ng moments in a steady sideslip .

CONCLUSIONS

The results of the investigation in the Langley free -flight tunnel to Qetermine the stability and control characteristics of an airplane mOQsl wi th a skevled wing are surmnarlzed as fol1mrs :

Page 12: Investigation of Stability and Control Characteristics of an Airplane ...

10 NACA TN No . 1208

1 . In general, the results indicate that an airplane vdng can be skewed as a lUlit to angles as great 88400 w1.thout encountering serious stability and control difficulties .

2 . Longitudinal stability and control:

( a ) The longitudinal stability and control charactor- . i stics were satisfactory--in flights mad.e ,vi th 40° skew over 8 lift­coefficient ranee from 0 . 3 to 1.0 even for very l ow values of static margin .

(b) Only a slif.~t chullge i n longi tudinal trim occurred. 11rith increasing skew' but an appreciable increase occurred :i.n the glide angle reqvJ.red at a g:tven l:lft coefficient .

3. Lateral ste bili t :r:

(a ) The values of effective d.:i.h~al for the wing skewed as a unit were consideraoly l es8 thon those encountered on 1-lings "dth largo amounts of sweepforward. or sweepback .

(b) Sleewlng tho wing caused. sizable Cr.8ngeS in t he latoral trim ,<lhich varied. with lift coefficient and skew angle .

4. L~tGral control :

( a ) The aileron control effectivoness Wa S only sligJ.tly red.uced by skew for angl es less than 400 because the dampinS in roll decreased opproximately tho se.me amount as the aileron rolling moments. At 50° skew, however, tho ailo:ron coritroi effectiveness "res noticeably r educed, and e t 60° it was so weak that sustained fliGhts could not bo !Jl.&de .

(b) The force tests indicated that for 400 Sltew angle the ailerons pr~:)(luced large pi tching moments . In the flight tests , howeve r , no pitching tendencies were observed in aileron r oJ.ls, appa~Gntly because the lift forces on the wing produced by rollin3 introc1ucedpitching moments that were equal cmd oppOSite to the aileron pitchil13 moments .

La.Tlgley Hemorial Aeronautical Laboratory National Advisory Committee for Aeronautics

La.Tlgley Field , Va ., July ~3 , 1946

Page 13: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No . 1208

REFERENCES

1 ~ Shorta l, Josep 1 A. , and Osterhout , Clayton J .: Preliminary Stability and Control Tests in the NAC/\. Free-Flight Wind Tunnel and Correlation with Full-Scale Flight Tests . NACA TN No. 810, 1941 .

2 . Shortal, Joseph A . ) and D-£'aper , John \{ .: Free -Flight-Tv.nnel Investiga.tion of the Effect of tlle Fuselage Length and the Aspect Ratio ancl Size of the Vertical Tail on latoral Sta ­bili ty .and Control. NACA AIm No. 3D17 , 1943 .

11

3. Bennett, Char les V., and Johnson, Joseph L.: E:>..-perimental Detel~nation of the Damping in Roll and Aileron Rolling Effectiveness of .Three Wings Having 2° .. 42°, o.nd 62° Sveepback. NACA TN No. 1278 , 1947 .

Page 14: Investigation of Stability and Control Characteristics of an Airplane ...
Page 15: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No . 1208

TABLE I

DD1ENSIONAL AND MASS CP'AHP.C'l'EHISTI CS OF MODEL USED

I N SKEWED -1trING I NVESTIGATION

Weig...ht, Ib • • 4.73 to 5 .03

Wing: Area , sCi ft Span , ft

0° ske1'T 40° skew • 600 skeVT •

Mean aerod;ynamic chord , ft 0° skew 1~00 oke'" • 60 0 Skel{ •

Aspect ratio (0° skew ) •

SI'ieepoack of 0 .25 -chOTeL line, deg Dihedral, deg • Taper ratio (ratio of tip chord to Root chord, ft (0° skmoT ) Tip chord, ft (00 sJ:ew ) Loading , 10 per sq ft .. , .

Radii of gyration (for 0 0 skev ) : , ft KX' . kz ' f t .

Ailerons: Type . Al'ea

Sq ft Percent S

Sp<:l.n, percent 0 '.

• ,

.root. ChOl'OJ

2 .6'7

l~ .00 3·07 2 .00

·70 ·91

1.40 6.0 3·0

o 0. 50 0 ·90 o . 1~5

• 1 . '(8 t o 1.69

nATI ONAL lllVISORY COHHI 'l'FKE FOR AERONAUTICS

0.625 0 . 844

Plain

0 .19 '7

44

Page 16: Investigation of Stability and Control Characteristics of an Airplane ...

I

Page 17: Investigation of Stability and Control Characteristics of an Airplane ...

NACA 'I'N No . 1208

TABLE I

DII~NSIONAL AND MASS CHARAC1~RISTICS OF MODEL USED

IN SICE'w"ED -HI NG Il!TVESTI GATIOllJ - Concl uded

Horizontal Tail : Area · . . . . . . . . . . . . . . . . . . . . . . . . . . . . · . . . .

• • • • f • • • • • • • • • • • • • • • • •

Sq ft ••• Percent S •

Aspect ratio Tail length , hine;;e line t o center of gravity , ft ••••

Verti cal tail : Area

Sq ft ••• Percent S •

· . . . . . . · . . . . • • • • . . . . . . · . • • • • . . . . . . • • . . .

Aspect ra tic .. • ...•...•...••...... Tail l ength , h inge line to cen tel:' of gra v i ty , f t • • • •

NATIONAL ADVISORY COMMI1~ FOR AERONAu~ICS

?O

2

o .!~ 15

2 2

13

Page 18: Investigation of Stability and Control Characteristics of an Airplane ...
Page 19: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208

x

Wind dJrect,on

y

\

I

~¢ MI

~ - --_ ----r_-rl

-."

Fig. 1

y

z

z NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

Figure I. - The sfabJldy system of axes IS defmed as an orthogonal system of axes having fhelr ongm at th~ cenfer of qrovlfy and In whIch the Z- aXIs IS In the plane of symmetry and perpendlcu/ar to the relafwe wind) the X-axIs /s In Ihe plane of symmefry and perpendIcular to fhe Z-axIs, and fhe Y-axIs /s pl!rpend1cuJar ta fhe plane Df symmefry. Arrows indicate post/we dlrechons of momenfs) forces J and contro/-surfoce deflecflons.

Page 20: Investigation of Stability and Control Characteristics of an Airplane ...
Page 21: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208 F ig . 2

Figure 2.- Model with skewed wing flying in Langley free-flight tunnel. 40° skew.

Page 22: Investigation of Stability and Control Characteristics of an Airplane ...
Page 23: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208

I

Skj-V C1.n9 /e~/\ (vane d from O'to 60°) ",

3: " 0.50 chord ",

_~"~r-~

/Vormal Cenfer-oF-gra VI t)' location ---_ __ --=.../

1"-----Z4---~

NATIONAL ADVISORY COMMITTEE fOIl AERONAUTICS

F lgure3. - Three-vlew s lr e fch of modC?/ used In ad)tJs ta b/e-.5 tewed-wl/ly I nye5!190!IOIl. (~/I d/!l1el!s/ O//5 /// 1/)C~tf>5 .)

Fig . 3

J

Page 24: Investigation of Stability and Control Characteristics of an Airplane ...

-- -- -- -- ~----

Page 25: Investigation of Stability and Control Characteristics of an Airplane ...

Figure 4.- Three-quarter front view of model used in skewed-wing investigation. 0 0 skew.

~ o >-1--3 Z

~ ,...... C\:)

@

~ 1-'. aq

f.P.

Page 26: Investigation of Stability and Control Characteristics of an Airplane ...

- ~.--~~--.~--~-

Page 27: Investigation of Stability and Control Characteristics of an Airplane ...

L

NACA TN No. 1208

.2 -+-

~ ~ ~'-5 s:::5 ""' 0 C:-+-.. I~ ~D .c:::~

0 :- _

~:-~ ,-;:-~, y - ----< , :...,-~ ~~

"t ~ -', ,y-_A' , [J~~ '~': ::-' ;r- -'

~ ?\

'~ r.l. , " -c~ ~

,-2 ~

--..;, - --<: '

ct(J

.8

& ,6 ""' -+-.. ~ (j

~4 § ~.2

C3

o

~'

-,4

1.6

(jI.Z 0 / $

~-) (,ynl . --~ :~.;;J

.~

.~~~~- .

+-§3 (j,B

s;::: c;..::

.If /~~, V" /)~

PI /1 J ~/ II / ,

~ 10

~ .4 ~) I /

~) /

-.......J (V/o/' ,~ .~ r-( ,

~:r=u o

o

_ 'U' .~~ .

8 /6 ' 24 32 Ary/e of attock) CX, dt:9

.r o fd7J) 0 - - 20 o ----- 4(J L'::.- -- 60

F ig . 5

NATIONAL ADVISORY COMMITTEE FOR AERONAUT ICS

40 .2 0 -:-2 P/tch/[Iq -moment coeffICIent) em

-4

fiq(Jre 5 - Effed of sAew anq/e on the aerodynamIc character/sties of the comp7ete sket--/ed - NIf)9 mode/

Page 28: Investigation of Stability and Control Characteristics of an Airplane ...

Fig . 5 cone.

4-sue ~--... ~~ 0 I CJ fS'-~~

~~ -.4

.0 4

1::>.::: §~

0 §a3 <:)

'e:+::: g''t-:0-.. 0)

~8 -:04 ::>::

-:08

.C6

-.08

-fl.~ .s=-~~ .>=-= ~ .):" - -l'r -, .:.-- - "t.

~ .~ --<->- - -> ..:.' L:.J

.y--<r-D_"'~ ' , . F='Q ~~ ~-Q F~! ~ '\ - ' . ..:c&:-o

'~ }, 1\ '~>-~

f'-- - ...

--<: -- y - :>--.._.., t/ ~

/ /'

~~ /i> I-----i )---;j ' -I--==(:J==(

v..:..., 0 I-O~ .~ r\ < ~-=

NATIONAL ADVISORY C01414ITTEE fOR AERONAUTICS

o 8 16 24 32 Af!q/e of ot/ock) ct) deq

hr;ure 5. - Concluded.

NAC A TN No. 1208

40

r (deq)

0 - -- 0 0----20 0------ - 40 6--- - 60

Page 29: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208

o 8 AS 24 32 .4 Angle 01' attack." ~ deg

.t (deg)

o---() 0------40 t:,. ----60

Fig. 6

. 2 0 -.2 PlfchlnQ -momed ca:llie;-ent~ em

Dgu re 6. - E lfeel 01 sKe W' angle on tfJe aerodynamic chara.cter/,s/Ics 01 the sl:ewea-wlng model with the: hon zontal tail removed . NATIONAL ADVISORY

COMMITTEE fDR AERONAUTICS

Page 30: Investigation of Stability and Control Characteristics of an Airplane ...

Fig. 6 conc .

4

,08

-.08

.~-~ ~--=-< :~-=< 'r--< . - --; 'r - -..? -4. -~ . =='. ~

'r- -_( '7-- - .>- ~ ~

Ir -~ '>, .-(. .-

-:~ / /-l'

~

d /

~

.~-~ / < ~=- .--'j ~~ ",/

~.

NATIONAL ADVISORY CO ...... ITTEE FOR AERONAUTICS

I I

o 8 16 24 32 Angle of' a tlac/:., CC, dc:g

figure 6. - COl7clud ed .

NACA TN No. 1208

j (deg)

0---0 0---40 6- - - - - 60

-~-- -- -- --- -- -- ---.-- ---.. ----_____ ~ ~ ___ J

Page 31: Investigation of Stability and Control Characteristics of an Airplane ...

.c.

J .1 ,--. ~ ~ \J

~ ~ 0

B ~ ~

~ -.1 ()

~ I

~ ~ :-.2 ~ ct

-;3

n a ......... ~

1......--

o

1---l-o-f---_ fa c.g. J L.--r.l

-.......... I ..... 0e (percent:

1-0 -

I-- tJ (deg) M,A.C.): - I- - \ 0- 0 20 ........

~ ~D--O 35 1'-0 , .F1-'" 0----- /2 20 ---1-0. C>---/Z 35

-0.. t---

~ ~ ~---

1-'---V~

fb" p

r:-. .... - --, v \ \A 1\ tv' \ ........ ...... ......

-- r--_ \

1"<>- -- <> Ih 1'- .........

0--' ~"

Q, .......

" I. ' ,(::

<> ) .......... /

.2 .4- .6 .8 /.0 /,2 /.4- 1.6 Lift coeFr/cleht) CL NATIONAL ADVISORY

COMMITTEE FOR AERONAUTICS

F/qvre 7. - £.ffecr of" ~/e J/a tor de,CJecf/on and cei?t~y- of'-9ra.vlfy loca;lon on th~ stahc long;/vdlna.J sta.bllity ctncl tn/?1 of tnt!. complete .s/(ew~d-wtr)9 model with 40 0 slrew_

~ o ::t> t-3 Z

~ I--' i:'V

@

f"Ij ~.

aq

-..::J

Page 32: Investigation of Stability and Control Characteristics of an Airplane ...

Fig . 8

.4 Q)G ~-I-..~

~~ 0 , G 1L...l.-

~~ ,f.\ ~Qj '§8

-.4 "-l

.08

10'k

-:08

-h.. .04 c: §() ~1€ 0 IQ:! ~<j ~~ -,..;qj

~8 -.04

,0

If

-.08 -30

(.

'1 L ----~

,;7->- ) --- ~--- --- tJ-

NACA TN No . 1208

r a (deQ) (deq)

0 --- 0 3 o -- -- -- -- 40 7 /::,. ---6D 14

~ ........... r- -6._·1 --<_6

.r ~ ...(A\

~;~ P'

NATIONAL ADVISORY COMMITTEE FOIl AERONAUTICS

-20 -10 0 /0 20 30 Arqle of YON I 0', dey

fJqure 8 - Effect of skef...J anqle on the loterol stobillty and trim characteristics of the complete skewed - J,Jtn9 mate/. 0. oppronmofely 0. b.

Page 33: Investigation of Stability and Control Characteristics of an Airplane ...

NA CA TN No. 1208

~

~ ~\0~ .J C)'1-..' ~~ 1<U

\j-,'-.. C:::\J 0 ~~ ~<u ct8 -;-J

/./ I~

/ / 4' / /' L ('h

/'

-H

6-- -y-- -/

V <t-

v

J \.-~ .~--

?t- ~

/~ R -6

/ff I~ t-- .

- .~

l;s-- - . r-- ;; L-A / . /

k /

/.

c7 NATIONAL ADVISORY COMMITTEE FOA AERONAUTICS

40 20 0 20 40 Tot-a/ aileron de-fJecflon,deq

Leff oheA RIghi sf/ex

t (degJ

8=---4~ 6.----60

Fi.1..ur e 9. - Erreo! of' oxew anqle on aIleron effecf/I/c?/7ess of' The okewed-w lnq m o de!. Bofl; al Jero17o deflecfed wllh ec;.ua/ up atfd dowl7 -{ravel. CL ::: 06.

Fig. 9

Page 34: Investigation of Stability and Control Characteristics of an Airplane ...

F ig . lOa

-+--. .I ~~ ~~ §t:' a 19.)

~~ ;:,,;~

~§ -I

-+..... .01 ~~ ~<:.J ~-+-::' ~c:: a IS:!

~ ~ -;01

.oz G 4-..' c: .01 S!J ~ ~ ~ Cj

0 -+--c:::

~ § ~Ol I

~ ::3:::: 15 Cl;2 :-0.2

9 9------ -k-Y -9 cr .... C}-- ---0- 6

:., .... ...{ .) !:;:f ,... ..... ' L'J

~¢ n/ .6

1/ /' V

/ / :J lV

/L V ;'

/1 / L /0

t¥-- NATIONAL ADVISORY COHHITTEE fOR AERONAUTICS

20 /0 a 10 20 AIleron deflectIon"" deq

Left s lick kJ9hl sltck

(a) J' ::: 0".

NACA T N No. 1208

AIleron O--Rlght O--Left

Figure 10. - Effect of she /AI onq/e on the OJ leron efPecf/ye-!leSS of the skeNed - v./Jnq mixiel. A!/erol15 defleded mdlVldually. CL =0.6.

-----

Page 35: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208

'k

~ ~~~ .I

~'" I ~ ~ 0

~~ :-k~ ct\j --:/

"."'- " tr= t~ r--t;1

-~ A/~

/ ,/

./ ':/ (\/ W l:)

NATIONAL ADVISORY COHHI'TTH fDQ AERONAUTICS

20 /0 0 /0 20 Aileron def/ecflo/l) deq

Leff she k. Rlj'hf 5f!ck

(b) b= 40 ~ . Figure /0. - COl7fl l7Ued

Fig. lOb

Aileron O-R/qhf 0-- Leff

Page 36: Investigation of Stability and Control Characteristics of an Airplane ...

Fig. lOc

.2 ~ /

/ [l

/ v>-.tV fY ~'T

_R r- '1 ~

,~ -t;:J

r ~ r===8 a----: ~ v T \.

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

20 /0 0 /0 20 A;/C'/,o/7 de/l{?ch ol7) deq

LeH c5 fJcx R'9hf She/<.

(c) J = 60°,

flrjure /0. -Coflc/ud C?d .·

NACA TN No. 1208

AIleron -RIC/hi --Lc:f'f

------~ --~

, I

Page 37: Investigation of Stability and Control Characteristics of an Airplane ...

NACA TN No. 1208 F ig . 11

..

..

(r-I---~

~.~ , ~ ~ ~

~ ~.~

~ ............

~ ~

()

- U r--r---I--:-:--6

./6

-~ ~ I---- ---...... ~

10 .!t) 3'J 40 50 60 SAe hi angle ) 1; deg

(C)Ca./calafe..d Troff) I.:Jh = C? - NATIONAL ADVISORY ~ e) f) (r COMMITTEE fOA AERONAUTICS

FIgure //. - £/'/ec.l or SAeW a.ngle on damping )n /01/ an d aileron roilIng eTTectIYe/)ess.

"---'o_~ __ ~~ ____ ~_~ __ ----------

Page 38: Investigation of Stability and Control Characteristics of an Airplane ...

Fig . 12 NACA TN No. 1208

-.5

(.'\..

o o

- r--- -=:::-

/0

LColculal-ed LExoerJmenfa/

~ --- .~~ --...

~ --- t:::--~ \.~

~ ~ ...........

.......

~.

20 dO S kew angle)

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

40 r ) deg

so 60

H9l1YeI2.- Var;arlon Or damplng- 1/7-ro/1 paramekr CZp w;/i, sfew angle. £ypeYlmenfa/ dC/fa ,cor eoc.h skew Cln91e IS based 017 projecled span(b co).r J. C alculof-ecl Cvrye waS" oil-cun ed be; use of fOrmu/a Trom rererence 3 ) C~ == ell CS =:ofoS t.