1、REPORT No. 823EXPERIMENTAL VERIFICATION OF A SIMPLIFIED VEE-TAIL THEORY AND ANALYSISOF AVAILABLE DATA ON COMPLETE MODW WITH VEE TAILSBy PAUL E. Pussms snd JOHNP. CAMPBKLLSUMMARYAn analyti has been made of acailable data on tee-tail twr-faw. Pretiowly pubk%hed theoretical du the other two investigati
2、ons were wind-tunnd testsof compIete models with various taikrface arrangements.The present paper extends the theory of reference 3 toinclude ccntrcd effectivmess and control forces sa well asstability, summarizes the results of the two complete-modelillVeStiiOIIS, and lepOrpositive when trailing ed
3、ge is down8 controkurface deflection mcasurd in phmc noFmal to ohord plane of tail surhwc, dcgmea1, tail length; distance from ccnlcr of gravity tohe line of control surfacec angIe of downwash, dcgrcceangle of eidewaeh, dcgrcce;/da rate of chauge of downwaeh angle at tail withangk of attackratio of
4、tip chord to root chordF stick or pedal formThe symbols used in the development of the thcoqy 01vce taiIs are defined as follows:lift coefficient of tail meammd in plane ofsymmetryangle of attack of tai mcaeurd in pkinc ofeymmetryp degrclateral-fome coefficient of tail measured normalto plane of sym
5、metryangle of sidcelip of plane of symmrtryelevator deflection or clmudder Wlcction whenelerudder surfaces am dcflcctrxi upward ordownwati togetlwr, degrcoerudder deflection or clcruddm Mlcction whenelerudder surfaocs are dcfkkl equal andoppoeitc amouuta on thr two eidcs, dcgrccsdeflection of single
6、 clcruddcr surface, degrees;subecripta R and L denote right and lrf t elc-ruddcr surfaws reepcct ivelydihedral anglo of tnil surface measurw.1 fromXP-plane of vcc hail to mch tail pmml, drgrccstail lift coefficient mmeurcd in planu norm.1 tochord plant of Cach tail panrlsum of changes in tail lift c
7、ociki(nt normal 10each tail panrl when tail is yawed; cqwd andopposite span Iond distributions ovmhip sothat CL= :.,orba = “Sm at =: .-.or ca- _ .BiIl aIf as and aN are smallbca= .al aNor c aN57But=COS rTherefore=COS ra;or a=aCOS r-(1)(2) For mull angks of sidedip, the ohangea jn angls ofatti measur
8、ed in the planes normal to eaoh panel of thevee tail are equal and opposite in. sign and are equal to theangle of sidaalip multiplied by the sine of the tail dihedralangle (. 3(b). Thw ,orororBut$=sh rThflreforosin rBtorqppl sin r (2)(3) The lift coefficient measured in the piano of symmetryis equal
9、 to the lift coeffio,hmtmeasured in tho piano normalto each panel of the vee tail multiplied by tho cosiuo of thetail dihedral angle (fig. 3(c), ThusCL,=CLNcos r (8)(4) When tho ves tail is sidsalippcd, the changes in liftooef6-r= KCL=Msins r(4) Directional control as measured by Yt,:(5)(0)(7)=KC=Mr
10、 sin I (8)Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-(I3)VERIFICATION OF WE-TAIL THEORY AND ANALY818 OF DATA ON COMPLELCEMODEM WITH WEETAIM. .CL= c+“ I.Iw#%? l%g Cb(d) Vwaillnmcksup.Provided by IHSNot for ResaleNo reproduction or networking perm
11、itted without license from IHS-,-,-242 . REPORTNO. S 2 8NATIONAL MWISORT.COMXE FOR AERONAUTICSThe relation between the stability parameters CL=, andC,for vee tails maybe obtained asG18FIIKW S.Vtittm d M meflMeatwlth mngIetialtack #xImWmikaixmItuaa mIme ML Tall mrfam B;.44.iIZ A-oS.taiI was greater t
12、han the eEective aspect ratio of the weetailin side.dip, even though the verticaI tail was t-ted in theisolated condition and did not have the beneficial end-pIate e.Y3E&-:,:.dFbt-F=-. - . . . . . . -.-. &u 6.1 a41ToM . . . . . . . . . . . . . . . . . . .E-i I* - -Ve M-=- - .-ikme7W *!4_ - -. ,i-.-1
13、1The curves for rmdta of teata of the vcc tail in grnmal arcmore regular and are smoother than tho curves for rcdta oftda of the conventional tail, partidady at CY=O.lo. Fora= O.l”, the conventional ftn stalls rather abruptly atangles of yaw of + 16 and then regains effcctivcncss whereasthe yawinf-m
14、oment curves for the vcc tail form a rclntivclystraight line for valu of # up to +40. This oharactcr-istic remdta probably” because the sect.ion angle of attack ofthe vee tail is a function of the sine of the angIo of yaw andthus the vee tail would bo expected to stall at grcati anglesof yaw than th
15、e wmvontional tail. The inlmrcnt tcndrncyof the vee tail toward later stalling is atso iIhsfratcd inflgur8and9. “7!0 -B 76-4 ”+02-.=%.-.-.-.-Z-.$.,A *- k-.+. . -,-! .-,- . . - 4 -. -7-. - . . .FIOUSB12.-lhe +- model ofhkuslrrknewfth vm tulland wmuotlcml titwtedhlbngh?yfreeillahttollwl. wlngue&Msquar
16、ofw.!n PItdiof a71 wmmtedrIMoo model bted Intla hll8Y 7. by I$fwt tunneL(m-(wm-r.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-(a)Vatln.FIOUnX14FMmhhh#m daEefIUMonldwwmkla InPItaboram-2 I I I I I I I I I I I I I I I I 1(b) OmlvwtbmlW.ImpwSkphDllmm
17、tOllInUuImlKbgT-byIbbttmmd.(lJb-uk+r.EiidIProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-I I40 R I I I I I 4P*.,&$I It! I I I I I I 1-A5RountM.-cmdud,d
18、mProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-254 REPORT NO. 828+ATIONAL ADVISORY CObiMIITEE FOR AERONAUTI08mlc?o-m-.!0 78 -.6 :4 -2 0 “.2 .4 .6 .8 .40 “B=t-tfWIT+PPIP, ,-.1 mI lfJll 191 IA* of Ot roci, q li&%f%e”OUIU 17.M, . and tCh t dmrncwmim C
19、4aghtm?JrM *1tdcd IO the Idagky rra341tmnnel. Fwe retmdd 7,-* h-a”.The plot of tho lateral-stability and direction.al-stibilityderivrkivea in figure 16 indicates that neither the vee tail northe conventional tail appreciably affecte W variation ofthwe elopes with lift cocient. One intermting point i
20、athnt the vee tail contributed about 1%0 more tiectivedihedral than the conventional tail although the values ofC% and CFwme appmxhnately equal for the two tails.Teata in langlayfree-ht tunnel,-A three-view drawingof the oomplete fighter-airplane model teat.ed in the Langley&flight tunnel is shown i
21、n &we 12. Dimensions of thelMont36 vee tail and the mnvmt.ional taiI tested on homodel arc nlaogiven in figure 12. The model wmtested with v-t-ail dihcclral angha varying from32,4 to 46.The redta of force ksts to detwminc thelongitudimd stability characteristi of h modelwith the conventional taiI an
22、d t,hc 3t3 vcc tailare ahown in figure 17. The data in figure I7exhibit no unumud characterialiq and the fligllt-teat dnta presented in figure 18 provide anolhcrquantitative indication thaL the static longitudi-nal stability characteristics were eescntially equalwith tho we and convent ioual tails.
23、The wc-tail arrangement showed lass change of trim withpower find flap deflection, probably bccawe of ibhigher location. During tho ilights of h model,the pilot could detect no dihcnc&. in. he dy-namic stabdity and hrmdlhg characteristics withthe two tails.A summary of the sttibility and control cha
24、rm-teltitics measured in force teab of 1110variousvee-tail arrangements is presented in fiiuro 10.The “scatter of the data in figure 19 is caused partly bythe slight variations in arm, aspect ratio, find pcrcen(tigc ofmovuble arm for the. different we tails as well as in dihe-dral angle. These resul
25、ts imhcate fairly good agrccmcnlbetween experimental and theoretical results except for thevalues of (C)$ at dihedral anglce greater than 36. Sim-ilar results were noted previously for h iaolald-tail tesE4.(d 4-0”.(b) #/-w.Provided by IHSNot for ResaleNo reproduction or networking permitted without
26、license from IHS-,-,-VERIFICATION OF JXI+TAIL THEORY AND ANAITSIS OF DATA ON COMPLETE MODEIJ3 WITH VEE TAILS 255The f.er+irplane model was also tested in the Langleyfree-fIight tunnel on a test stand on which it was free to yawbut was restrained in rolI and pitch. An indication of therudder-force-re
27、versal characteristics of the modeI with con-rentionaI and 43 ree taiIa was obtained with this setupfrom the trim angles of yaw produced by diflerent tiredrudder deflections. The rmdte of these teata me presentedin figure 20. The tests showed that. with the vee taiI thenmdeI would trim only at. fair
28、ly small angles of yaw evenwith full rudder deflection. WKh the conventiomd tad,however, the model yawed to huge angles with Ieft rudderdeflections greater than 13 indication that rudder-forcerevemal or rudder lock probably exists for the airphme withthe conventional tail. From these data, therefore
29、, rudderlock appears to be IS Iiiely to occur with a Tee tail thanwith a conventional vertical taiI. The previoudy notedfacts that the Tee tail stalls at a higher angle of sideaIip andmay require a control surface of smaller chord ratio thanthe conwmtiontd vertical td also indicate less tendencytowa
30、rd rudder lock with the we taiI.l%Awioieci fmm fortnul%s(24 to qQ “Expeninentol &h.00?-c.&#o - m%a) -.01 tt? 20 “ 40 m 80 Imoihed-of cTlgkq OlqFIODEE 10.-VfitIm of tihULty aud omkd mltlltSIldIMrnIencorllallteMirpmQ lmleltemed lo MISkrkeMtEb:knuMLu Ilfllt rltl 1 i t 1 1GENERAL EEMARE8STASUJIT AND CO.
31、NMLOLACI%MSTIThe foregoing andysii of we-tail theory and test data hasindicated that a Fee tail can have the following charuc-teristica rdative to those of a conventional tail produc thesame dues of stability and contrcd parameters:.(1) Apprcmimately equal area unless the conventionalvertical tail i
32、s in a bad canopy wake, urdess the usuallyhigher Iocation of the Tee tail pIaces it in a region of greatlyreduced dowmmsh, or unkas the Yee tail has a highereffective aspect mtio than the conventional horizontal anderticaI tails.(2) Possible inadequacy of controls and interaction ofcontrol forces wh
33、en qimuhneous ffl deflection of bothcontrols is required. This difficulty is IikeIy to be en-countered if the Tee taiI is designed to give values of C4.and C, r equal to those provided by a conventional Massembly. It is apparent that, if maximum rudder andelevator deflections of 25 or 30 are myd wit
34、h the con-cnntional taiIs, elerudcler deflectio of at let 50 or 60.would be required with the vee taiI. At such large deflec-tions, the elerudder wmdd be operating in the nonlinear.range of control tiectivenem against deflection and mightpossibly be in the range where the control effectiven perunit
35、deflection either remained constant or deoreased. withinmeasing deflection. Ona method of avoiding tl cond-ition is to use a large balanced elerudder surface thatDroduces larger ahma of C-L and C=,rthan the conventional-.-.-Provided by IHSNot for ResaleNo reproduction or networking permitted without
36、 license from IHS-,-,-256 REPORT NO. S2 3NATIONAL ADVISORY COMMITTEE FOR AERONAUTICStail control surfaces and therefore. produces the requiredpitching or yawing momenta with smaller deflectione-notover a total of 30 or 400with simultaneous fulI deflectionof both rudder and elevator controls.(3) Poib
37、le interaction between longitudinal and direc-tional trimmkg when tabs are at fairly largr deflections.(4) Less tendency toward rudder lock,(6) Possible reduction in control forces or in amount ofbalance required.(6) AJorc dihedraI eflect due to taiI.(7) Larger adverse rolling momente with rudder co
38、ntil.(8) Lees change in trim with application of flaps or powerbecause of the usually higher location of the vee tail.Additional points not previously considered are that thehigher location of the vee tail may deereaee the groundeffect on tho elwrator control required for take-off and lancl-ing and
39、should aIeo make it simpler to keep the tail out ofthe spray for takff and Ianding in flying boats.DRAG CHAEACTBB18TKXTIM nats from tests in the Lmgley 7- by lo-foot tunnelshown in figure 13 indicah a decrease of 0.0016 -in dragcoefficient from usc of the vee tail: tests of the same modelin the bngl
40、ey two-dimensional low-turbulence pressuretunnel indicated approximately the iame drag reduction.For the model tded, a large part of. the reduction wasprobably caused by a dwxeased fuselage-tail interferencewith the we-tail installation. A vee tail, howver, has onlytwo fuselage junctures instead of
41、three and some reductionin drag thus is usually obtained.COMPltSSSl_Y PECTSFor high-peed flight because the vse tail can be installedwith a better fueelag-tail juncture, the tieck of compres-sibility on tail drag should be reduced. This advantage,however, tends to be canceled by the fact that, for v
42、ee tails,the individual surfaces will probably be operating at higherlift eoe.flicientsfor.trim and will almost certahdy be canceledif the tail. is so installed on top of the fuselage that a sharpvee is formed at the junotnre. The location of the vee-tailarrangement should place the surfaces farther
43、 from the wakeof the wing and canopy and thereby should tend to reducethe powibilitiea of tail btieting or mughnesa at. high speed.SFiN-EECOVERYCEABACTEEISTJU3Tests in the LangleT 20-foot free-spinning tunnel of a modelof the same fighter airplane that was. testd the Langleyfrea-flightt tunnel indic
44、a.ti that the Yee-fi arrangement hadslightly better spin characteristics than the Conw.ntiond tailassembly. The improv spin chacterietice might haveoccurred because, with the vee tail, there was no horizontalsurface to blanket the vertical tail .The data presented inreference 2, although inconclusiv
45、e, indicated approximatelythe same spin charactmistice for.the two types of tail.At present no general conclusions can be drawn concmingthe relative merits of the vee tail and conventional tails forspin recovery. Although available twt dan indicate hatthe vee tail may have better spin-rccovcW clmnwl
46、erislicsthan the conventional tail, it is possible tlmt if simultaneousfull deflection of both rudder and elevator is rcquirml kspin recovery the vee tuil might havo lass desirable spin-recovery characteristic than the conwmtional lttil tmcmbly.STEUCTUML CONSIIM?RATIOMhfarmfacture and maintenance sh
47、ould be sinqh for thevee tail than for conventional surfaces, sinm no vertiral tnilsurface must be manufactured, stored, or repaired. Themechanism requird to opmato the control surfaces both aaelevators rind as rudders, however, is eomohnl con:pliratcdand naturally tends tQ offset this advantage,The
48、 vee tail, because of its configuration, must curry loadsthat do not contribute to tlJLIstabiIity and control. Thisfactor will result in higher tail and fusdagc loads in bothpitching and yawing manruvcrs, and inrrraacd st ruct.uralvvcight will be rcquirwl. to carry tim grcatm loads.CONCLUSIONSThe followiug conclusions wrc drown from the rcsulte ofthe analyzia of available datti on vm-tail surfams, from anextension of previously preeentwl vce-tsil theory, nml fromgene
