and an angle of attack of 9«7°« Local failures of the fabric attachment to the elevator ribs occurred. By moving the elevator vent holes from the vicinity of the trailing edge to the leading edge, the bulge was eliminated for these test conditions at the expense, however, of some increase In fabric depres- sion on the pressure side of the elevator. Marked increases in the elevator hinge-moment coef- ficients occurred as the test Mach number was increased. For the elevator with )4-inch rib spacing the hinge-moment oarameter C-u (rate of change of hinge-moment coef- ficient v/ith elevator deflection) increased from a value at low speed of -0,005 to a value of -O.OO9 at a Mach number of 0.68. The effect of fabric deflection for the NACA ARH llo. LRPOla eleva'cor witli S-inch rib spacing caused an additional adverse increr'ent in hing'6-iaoiasnt coefficient as the speed ".vas increased^ xlie effectiveness of tlie elevator \/ith ]}.-inch rib spacing did not change apprsciably v/ith I.iaoh number, As a result of fabric deflection, hov/ever, the effectiveness of the elevator -^ith 8-inch rib spacing decreases sharply at Mach nujiibers above O.56, ihe adverse effect of fabric deflection on elevator hin^^e r.ioinent was decreased slightly by locating the vent hol3s in ohe leading ed.,';;^ rather than at the trailing ed^e of the elevator . INTi^ODUGTIOlJ Tests we±^e made to detorrnina the effects of elevator- fabric deflections at high speeds and of compressibility on the aerodynamic chai'actei'j.stics of a full-scale hori- zontal tail surface, fhe necessity of sucn an investi- gation has been der.onstrated by the excessive and irregu- lar hinf^e moments cncoiuitored during high-speed maneuvers and .by numerous instances of control-surface failure on some of the more recent high-speed airplanes equipped v/ith fabric -covered control surfaces, fhe present report giv^^a the result? of tests on three elevators with idonticiJ. external dim.ensions, hle- vators 1 and 2 were fabric covered, had i-ib spacing? of approxim.ately I;, and 8 inches, respectively, and v/cre used to determine the fabric deflection. The third elevator .v;as made of solid mahogany, included tv;o rov;s of pressure orifices, and was used to determine the extei'nal pressure distribution, liach elevator was tested throvigh I'anges of I.I&ch number of 0,2 to 0,68, elevator angle of "9° to"-9°, and stabilizer angle of 0° to 9°. Tests of any combi- nation of the aforem.entioned variables v/ere limited hj the maxlmuiii allov/able loads. In addition, elevator 2 was tested with the original vents sealed and vents at the leading edge or at 10 percent elevator chord c^ to determine the effect of vent location on fabric deflection. The tests were conducted at the Langloy l6-foot high- speed tiinnel, Langley I.Iemiorial Aeronautical Laboratory, NACA ARK No. 1,57 Ola COE?FICIZ?TTS AND SYMBOLS C-r. drag coefficient (D/qS) C^x hinge -mcinent coel'ficient (E/qCg'^b) C-r lift ccefficient (L/qS) C^ -oitchlnp-ir.orient coefficient / — l ""ra - ^ \qSc' J D drag of entire inocel F hinge r.otnent L lift ^f entire nodel M^ , A pitching morent abo'it quarter-chord point of rr^ean aerodynamic chord b spsn,feet c chord of horizontal tail surface exc-ept when designated otherwise by subscript, feet c' mean aerodyna.mic chord of horizon-ta] -tall "cg root-nean- square of elevator choice behJ.nd hinge line /'l 2\ dynamic pressure ( —oy'~ ] p mass density of air, -slugs p-er- cubic foot V velocity, feet per second S total model area,- square feet M Mach num^ber P p static pressure at any point a angle of ■ai:"tack of" s-ta.bilLzer-, -degree-s- / \ /P - Po\ pressure coef f icaent \ ■ } 5 angle of elevator chord with respect "to stabilizer chord, deg:ree3 -rr elevator— effectiveness -oar-ameter- T -) h NACA APR No. L5F01a Parameters : Cl a ^L5 Cha Ch^ The subscripts outside the parentheses represent the factors held constant during the measurement of the parameters . Subscripts : b balance e elevator f flap (elevator + balance) i internal o free stream APPARATUS AND METHODS Test model . - The model was a full-scale left-hand horizontal tail surface of the SB2D-1 airplane. The air- foil section used was based on the NACA 0020-6i| airfoil profile modified to have a maximum thickness ratio of 10.7 percent and a straight taper behind the 65-per cent- chord station. Since a sem.ispan model was used, it v/as necessary to locate the center line of the airplane in the plane of the tunnel wall to produce air-flow con- ditions corresponding to those of flight. This result was accomplished by adding a 20.5-lnch stub wing to the tail surface. Figure 1 shows the model Installed in the tunnel and figure 2 presents the physical characteristics of the model. FACA ARR l^Q. I5^01a The stabilizer was metal covered and included a fabric seal to prevent air from flowing between the rear part cf the stabilizer and the elevator leading edge. (See fig. 3') 'i'^^s model was not aerodynanically smooth. Brazier head rivets, access and inspection doors, and considerable waviness characterized the stabilizer surface. Each ^^levutor had a modified elliptical nose and a straight taper behind the hinge line ending in a trailing- edge angle of 12°. The coordinates for the elevator con- tour are pr-esented in figure J. Elevators 1 and 2 were of metal construction and fabric covered. Details of the rib locations are shown in figure [j_. The average rib soacings are approximately I], and 3 inches for elevators 1 -7 and 2, respectivelv. Both elevators had one — -inch- o diameter drain hole in each elevator panel on the lower surface approximately 1 inch from the elevator trailing edge. Since each elevator panel hart one hole, these openings also served as air vents. Elevator J was made of solid mahogany and was dimensionally equal to ele- vators 1 and 2. Two rows of pressure orifices on the upper and lower surfaces, J 5 ^'^^ 7^ inches from the longi- tudinal center line of the airnlane, v\/ere built into this elevatox" . Einge-mor.ient measurement,- Figure 5 is a schematic vievif of the model installation and illustrates the apparatus used to measure the elevator hinge moment. This sketch shows the extended elevator torque tube passing through a hole in the side of the tunnel and into two self -alining bearings mounted on the tunnel balance frame. The elevator hinge moment was transferred through the elevator torque tube to a IC-lnch crank and then through a jackscrew to the scale platform. The jackscrew was also used to vary the elevator ai^gle. The platform scale was attached rigidly to the tiinnel balance frame and. since all other related parts were also attached to the tunnel, balance frame, hinge-moment mea.surem.ents could not interfere v^ith the m.easurements of lift, drag, and pitching moment. 411 force and moment data were recorded s imu 1 1 ane oi^. s ly . Fabric-deflection measurements.- Stripes j^-inch v/ide I^" were painted chordwlse on both surfaces of the fabric- covered elevators to perm.it the m^easurement of the fabric ITACA ARR No. L5F01a defleoLlon. (See fig. 1.) cclid stripes were painted over each rib and broken stripes inidway between ti-.e ribs, on the upper and lower surfaces. These stripes are straight and parallel i'cr tne sttttic condition (see fig. 6 ) but because of air loaos the fabric deflects and the stripes bend. Cameras in fixed positiors ■'^'ere provided to photograj^h the elevator surfaces siinultaneously and thus provide records of the fabric deflection. ihe deflection of the painted stripes v;as measured from enlargeiuencs of tiie photograpr.s. Fs br i c - 1 e ns ion :nc a s ur '3 inen t each elevcitor pane designed by the Fl A detailed descrip of measure-uent are tensions '-vere .ne ceteriTiine any chan repeated stresses testing. Table I and indicate c that vator 1 is witiiin 1 WciS measure ight Research tion of the i given m ref ured before a ; in I auric that ¥vere app presents a su the chanre i tne accuracy Elevator 2 had a slightly lov;er testing, but this difference ma humid ity effect. s . - The fabri ~d vvitn an ins Divisi on of nttruaent and erence 1 . Tl- no afte r t he tension resul lied to the f .fk.:ary of ti.o n laori c tens of tne .Tieasur fabric tensi y le a temper c tension for trument the iiaboratory. the tecimique e fabric tests to ting I'rom abric during xTieasurements ion i'or ele- e.r.ents . on after ature or Pressur e mga s i re i aent s . - The orescure distribution over the"'eievator was obtained with elevator 5> which contained t\vo rovs of orifices. The external pressures over the 'jjper surface of the stabilizer were obtained by the i;se of two pressure belts located at the JJ-inch and 70~ii^ch stations. All stations were measured in inches fro..-i the longitudinal center line of the airplane. Two O.O^O-inch-diameter tu^es were installed in ele- vators 1 and 2 at the iiY-inch and 97~i'^^2h stations to measure the elevator internal oressure. TEST PROCEDbRE The .-..eneral procedure in conducting tbe tests was to set ti^e desired angle or attack and elevator an^le at the beginning of each test. Dcta were then recorded at each of tne. foliowin^j, speeds; ■■!ach number = 0.20, O.J^j 0.1+5, 0«,j3, O.-;,''^, O.oO, 0.6^, ana 0.68 or until the maximum allowable load on the tail surfc^ce was attained. mCA ARE No. 157 Ola The stabilizer root angle remained fixed during the test. The elevator root an^le was measured, and recorded at each test point, since it varied slightly because of twist of the torqae tube and deflection of the scale platform. The angles of abtac'-c ei^d elevatoi^ angles are believed to be accurate ivithin ±0 , 1"^ . DEDUCTION OP DATA Force data.- The lift, drag, and pit ching -moment presented in this report are based on bhe vving area of the complete model (see fig, 2) including the stub vifing. All data "were taken with the elevator seal in, the elevator vents at the trailing edge, and the trim tab neutral, unless specified otherwise. The force data were corrected for tunnel-vvall effects by the use of the reflection-plane theory given in reference 2, The model thickness vas such a small part of the tunnel diameter that tiunnel blockage correc- tions v/ere negligible. Since the elevator 'torque tube could 'twist and the scale platform deflect, the elevator angle changed with hinge moment. Calibrations of the twist of the elevator torque tube and the deflection of the seals platform with elevator hinge moment were use'd to correct the indicated elevator angles to actual angles The corrected data -.vere cross-plotted and the values at selected "angles of attack and elevator angle were then plotted against Mach number. Since a large part of the' data presented is plotted against Mach n-ujiiber, figure 7 has been included to show the average dynatr.ic pressures and the average Reynolds nimibers corresponding to the test Mach numbers. The Reynolds n'-jmber is based on the assuraed moan ac;rodynaraic chord of i+oli-l fee't. It should be mentioned that the changes \vhich occur with speed are not pure Mach number effects out include effects due to distortion of the model loncer load. The effects shov;n therefore apply only to the particular com.bination of dynamic pressure and Mach number tested herein. The resiilts, however, are plotted against Mach nuraber , and the dynamic pressure at any Mach number m.ay be obtained from figure 7 • Fabric deflection .- h special film vievi-er v/as used to enlarge the oho'tographic negatives of the elevator surfaces. Vertical scales were attached to the elevator surfaces at each broken stripe and photographed for all 8 NACA ARR No. L5F01a model configurations to obtain fil.ris of the etatic condition (zero aeflection). A quantitative .rxeasure of the faji-'ic deflection was ovitamed by comparing a photograph for the static condition (ze^'o deflection) ivith one made durint^ a test. i'he displace.rient of any strij^e wab then measiired and recorded. RESULTS iiND DISCUSSION Fabric Deflection Elevaj:or_l.- Figure 8 is a photograph of the fabric defleotT5n~on the upper surface of elevator 1 (i4--inch rib spacing) at a = 0°/ 6 = k.2'^, and M = 0.66. The fabric deflection is not appreciabie at any point elong the ele- vator except for a s.aall bul^e occurring near the inboard hinge. No ether photographs are snovm for this elevator because the fabric deflection was not ser^-ous during any of the tests with ti.is elevator. Eleva tor 2 .- Fia,\ire 9 is a photograph of the fabric def lec"tib~n "of both surfaces of elevator 2 (6-inch rib spacing) at a = 0^, 5 = 5'3°j ^"^'^ ^'^ - ^•y5« Considr- erable bulge occurred on the top surface behJ.nd the hinge line. This bulge- changed to depression on the rear part of the elevator. Since the fabric v«as sewed to the ele- vator ribs, the solid ' stripes should show no deflection. A number of solid stripes, however, are deilected. (See fig, 9('^)») Deflection of the solid stripes Indicates failure ol' the fabric attach. uent at these points and is the oeginnlng of a condition tiat woi Id result in coi.ii-'lete failure of the surface if the air loads -vere increased. 'Figure 10 is u photOj.raph oi tx.e fabric deflection at a = 5°, 5 = -0.7'^,^ and M = 0.62. In general, the up^jer surface -s slightly bulged just bei^ind the nmge line. The aiost serious bulge occurs at ti.e inooard hinge and IE believed to oe a result of weak faoric attachment around tne hmge-pocket cut-out rather txi&n of local- suction peak pressures. Figure; 10 wlso shows the fabric pulled a.vay froiii tiiC ribs. (Note solid sti'ipes.) Figures 11, 12, and IJ are plots shov<'ing the vari- ation of the fa'cric deflection witix percent of elevator chord and include only the portion of the elevator chxord for which the faoric v;as deflected; therefore only the end points of zero deflection are shown. These data are FA1A ARP Fo. L^^Ola for a representstive spanvis'? ?tation (77«l-inch station). Pi-T:\^.re 11 presents the fabric deflection for various ""lach n-arril e r 3 f. t elevator aii'^les avp-rsging -1-5° cJ^d a-= 0^ . \Althcugh the elevator angle changed slightly (0.5°) with soeed, it is apparent from figure 11 thet Increasing the speed increases the fabric deflection. The inaximum fabric deflection of the lower' surface has been plotted separately for each speed in figure llj. and shows that the fabric deflection varies linearly with dynamic pressure for elevator 2 at a = 0*^ and 5 ^ -l.^'^.: Figure 12 presents the fabric dex'lectlon for various elevator angles at a = C° and M = 0.55* Increasing the elevator angle negatively increases the fabric bulge on the low^er surface while the deflection of the upper surface changes from bulge to depression. Figure IJ presents the fabric def]ection for various angles of attack at M =0-55 and W, - . 53 • The rriaxlmujn fabric deflection' attained during these tests was a 0.6-inch bulge en the lower sur- face of elevator 2 at a = 7."^, ^ =^-^.7°, and M = 0.S5 (fig. 15). Pressure distribution.- Febrlc bui.ge tends to be unstable since it causes an increase in the local negative pressures, which in turn cause an increase in the fabric balge. This adverse effect is rragnified at high speeds and has been observed to result in failixre of the fabric attachments to the elevator structure and finally complete failure of the fabric. An invesitgation to determine the external pressure distribution over the elevators and the location of air vents that would result in negative inter- nal pressures and a reduction in elsvacor fabric bulge was therefore undertaken. Flevstor 3j w^hich was dir.en- tionally equal to elevators 1 and' 2, was tested for this purpose. The tests of elevator 5 indicated that the pressure:." distributions at the '^^-^XioSx and ^Q-\tiq,\\ stations were very nearly the sa;ne on the elevator but differed appre- ciably near the stabilizer leading edge. This difference may be attributed to surface irregularities. Removing the elevator seal Increased slightly the positive pres- sures on the lower surface of the elevator balance area for positive elevator angles but had little effect on the pressures ever the other portions of the elevator. The external nressui-e distributions at M = 0.20 3n and M = 0,53. Comparison of figures 20 and 9 shows that the upper-surface bulge is changed to depres- sion with vents at the elevator leading edge, except for a small local bulge at the upper surface near the inboard hinge. It is apparent from figures I9 and 20 that loca- tion of the vents at the elevator leading edge will eliminate the danger of the fabric pulling loose from the ribs and failing for elevator angles up to ut least l^.'^ IIACA ARR ITo. L5F01a 11 Aerodynamic Characteristics Basic data .- The lift, drag, pi tchirig -moment , and hinge-mo.aent coefficients are plotted against Mach number in figures 21 and 22 for elevators 1 and 2, respectively. These data are presented for a = 0"^ , 5°, 6°, and 9'^, and a maximum range of 5 = 6'-' to -9°- The fact that the Cl, Cm, and Ch values for a = 0° and 5=0^ are not zero Is due either to asymmetry of the model or to small errors in setting the neutral angle of the stabilizer, elevator, or trim tab. The increase in the lift or pltchlng-moment coeffi- cient with Mach number for both elevators is less than the / 9 1-1/2 increase predicted by Glauert's factor (1 - M^/ This difference ' is believed to be a result of the twisting of the stabilizer and elevator toward their zero angles due to the aerod\-namlc loads. The drag-coef f Iclent curves show the usual large Increases in the vicinity of the critical Mach n^ombers. The data show pronounced Increases in elevato.r hlnge-monent coefficient with Increasing Mach number. Integration of the elevator pressure-distribution diagramis showed Increases of approxi- mately the same magnitude. The rate of Increase of hinge moment vvith Mach nLuaber was more than twice as great as would be predicted by the use of the Glauert factor. In general, the changes in the aerodynamic coefficients with Mach nuraber were gradual and consistent. The critical Mach .numbers for the various model configurations could not be greatly exceeded In these tests and consequently the abrupt and drastic changes that have been iioted In tests of smiall models at high supercritical speeds were not encountered. The only indication of such changes occurred for elevator 2 near the highest test speeds. (See figs. 23 and 2L, ) Variation of lift v/lth a and 5 .- The variation of the lift- curve-slope narameter CLq, with Mach number for elevators 1 and 2 is presented In figure 25. The slopes were measiored from plots of C^ against a in the region of a = 0^ to 3"-*. The values at low speed of Ci,|^ are considerably lower than the x^alue estimated from two-dimensional data for a v/ing of this section and plan form, principally because of the discontinuity of the airfoil contour at the stabilizer trailing edge and the elevator leading edge. 12 NA.CA ARR No. L5?01a The change m Ct ^ vn'.th Mach mimber for elevators 1 and 2 is shown in figure clL\. and indicates good agreeraent between the two elevators at low speeds, por elevator 1, Ct,p^ anc-^esses graduell:/- with speed. At the maximnm Mach number attainable (0.6S}, the data indicate that Cr^ was ■ beginn: ng to decrease. The vai^iation of Cj^c with R'ach number for elevator 2 indicates a marked adverse effect of fabric dsf^'ection at Mach numbers above 0.60. Eleva tor e ffectiveness.- The variation of the elevator- effectiveness parameter with Llach number is shown in figure 25 for elevators 1 and 2. The- curves show a small decrease in effectiveness as the sneed is increased from M = 0.20 to ¥• = 0.1+5' Beyond Mach numbers of . l^S the • effectiveness for both elevators increases. The effec- tiveness of elevator 1 is still increasing at M = 0.69 ' but falls off sharol;- beyond values of M' = O.56 for elevator 2. Since elevator 1 had negligible fabric deflection and elevator 2 had serious fabric deflection, the adverse effect shewn is a result of fabric deflection.. The theoretical effectiveness for a plain flap/hingea at its 'leading edge has been computed according tc the thin-. aii''foil theory (see. ref ei-ence , 5). and is- shown in figure 25 • The actun.l elevator effectiveness is aoproximately 7I P'sr- , cent of the theoretical value for a plain flap at moder.ate spe.c-ds . . ' . pitching moment .- The variation of the pitching- moment parameter '^Crp/dCT with Mach. number is shown in figure 26 for elevators 1 and 2. The value of this parameter is a':r)roximately the position of the aerodynamic center of the airfoil with respect to the. quarter-phord point of the assuineri mean aerodynamic, chord (fig. 2).. The chan;-"e in the center-cf-lif t position caused by ele- vator d'^fl.ftctlon is. given by the parrmetcr { ^C-^/^Gj^} q^. The vari'^.lioxi of chls para.mettvr with F'ach n^jmb^r was about the same for both elevators; that is, the center of lift was shifted rearward, The change in the conter-r of -lift Dcsition ca^ised by angle of; attaci" is given by the paramo"ter (^^111/^0^)5. Increasing the Mach number caused a greater increase in this prrameter for ' ele- vator 2 than for elevator 1, probably as a result of the fabric deflection on elevator Z. i NAG A ARR IIo . L5F01a ,■ 15 Einge moment .- The change in ' Cy^^ with ?.'lach number is shown In figure 27 for elevators 1 and 2. In general, the agreement of the cata for the two elevators is good, although an aliucst constant small difference exists between the values for the two elevators. Sm.all differ- ences in contour between the t'vo elevators could cause this difference. The small low-soeed value of Cv, (-0.001) decreased about (0 percent between , M = 0.20 and M = Q.oO^ The variation of C|-i=- with Mach number is shovifn in figure 28. Large increases in the negative values of Ch5 occurred v;ith increasing speed for elevators 1 and 2. The value of- Ch5 . fo^ elevator 1 (Li.-inch rib spacing) increased from -0.005 to -O.CO9 between M = 0.20 and ■ M = 0.68. The difference betw.een the low-speed values of Ch!^ for the two elevators is believed to be caused by minor physical differences such as a small bum.p that existed on the upper surface of elevator 2. This bump was 5*5 percent of the elevator thickness, was located at 6.5 percent of the total elevator chord from the nose, and tapered to zero at the elevator leading edge and at the hinge line. Figure 28 also shows" ci;rves for elevators having zero and lOO-psrcent aerodynam.ic balance. The curve for zero aerodynamic balance was -calculated according to thin-airfoil theory (reference 5) for a plain flap hinged at its leading edge. Elevator 1 had 50-percent aerodynamic balance at M = 0.20 but^ becaiise of the adverse Mach number effects, the balance was reduced to 8 percent at M = 0.68. The control forces required for such an elevator v/ould thus aoproach those that would be obtained with an ordinary unbalanced flap, when it is assumed that the value of C^ for such a flap does not change with Mach number. In the absence of boundary-layer changes, it might logically be assumed that the elevator hinge moment would increase y/ith speed according to Glauert ' s factor. The Icw-soeed value has 2 -1/2 been Increased according to this factor (1 - M ) and the data are "plotted in f.l-ure ^io . A comparison of the two curves shows that the rate of increase in Ghe- with Mach number is about double the rate of increase pre- dicted by Glauert 's factor. Elevator 2 had l^J) -per cent aerodynamic balance at M = 0.20 but zero aerodynamic balance at M = . 60 . The increase in Chs is m.arkedly greater for elevator 2 than for elevator 1 because of the adverse effect of fabric deflection. The Ik KACA ARR No. L5P01a c'ifference in t^e increases of C'^r rith Kach number for the tv'o elevators ao-peers to be an' effect of fabric deflection, since fabric deflection was the princi"oal dif'f^erence betv/een the tv/o elevators. This difference is :ilctt?a at the top of figure 23. The effect of fabric cef lection on Cy,^ v/s s to cauce an increase of -0.C02 from M = C.20 to tt = 0.6O. This increase ^vas about [j-O percent of the lov;-Epecd value of Cv^^ for the ele- vator tasted. An increase of this magnitude would be still more serious for a highly balanced tail surface for v/hich the initial Cv,^ misht be of the order of -0.001. E ffect of vents on hi nge moment.- Xs was shown in figiu'e 19, the fabric deflection varies with vent location. The best vent location from a consideration of safe fabric deflection was found to be at the elevator leading edge. Figure 29 sbov.s the. variation of the hinge-moment coefficient with elevator angle, at , M - C.55> i'o^ -^-^e three vent locations tested and with all vents sealed. The data presented in this figi.ire shew that the vents located at the elevator loading edg'^ produced the smallest value of Cv- - . The beneficial effect of vents at the leading edge (reduced internal pressure) is -probably a result of changing the asymiretrical elevator-surface deflections, which resulted in acprociaole elevator camber, to more syiijretrical deflections with less cambei-. (See fig. 19 and reference i|.) Effect of elevator seal .- A limited amount of data v^ith the elevator seal removed was obtained over a sr^all range of elevator angle at a = 0°. These data indicated no appreciable effect of the seal on the elevator hinge mom.ent . Tab effectiveness.- The effectiveness of the elevator trim tab through the speed range is shown in figure JO. The data for a tab angle of -10'^ shov; a gradual decrease in effectiveness with increasing soeed - the ACh decreasing 'from O.OJ^ at M = 0.20 to O.G5O at M = O.65. The effectiveness reruains approximately constant from M = 0.20 to I.I = 0.6c for a^ab angle of 6.3°. C0NCT,U3I0-NS An investigation of the characteristics of a full- scale horizontal tail v>'ith fabric-covered elevators at MCA Ar?R Fo. LSFOla 15 Mach numbers ranging froiu 0.2.0 to 0.6^? has led to tie following conclus ions J 1. Elevator 1 ([i.-inch vih spacing) had no apx':;::'ec iable fabric deflection in the speed range of the^e tettc. Elevator 2 (8-inch rib spacing) nad a maxl:nu:Ti fabric bulge of G.6 inch between rlbj: at a Mach number of 0,.35j an elevator angle of -5'7°j -I'^c; sin an;;,le cf attack of '...7°<> Local failures of the fabric attachment to tiie elevator ribs occurred with elevator 2. 2. Vent holes located at the e]evator leading edge on either side of the seal, rather than in their original position on the lower surface near the trailirig edge,, eliminated the bulge for a aioderate rahge of elevator angle at the expense, hov/cverj of some , jncri;aEe in fabric depression on the pressure side of the elevator. 5= Elevators 1 and 2 produced very large increases in elevator hinge-.:joment coefficient as the Mach nuir±er was Increased. T;-e value of Cv (the slope of the curve of hinge mo.nent against elevator deflection) for elevator 1 (4--inch rib spacing) increased froia -0.005 to -0,009 between Mach nurabers of 0,20 to 0.68. L. c In addition to the increase in iiinge aoiue nt resulting fro.ri inert, asing speed, elevator 2 (B-incji rib spacing) had an increase in hinge .itiOLuent due to I'abrlc deflection. The fabric deflection for this elevator increased the value of C-,. by -0.002 at a Mach number of 0,60. Fabric defl-ctlon also caused an early loss in elevator effectiveness. Elevator 1 rriaintaineG its effectiveness up to the :Tiaxirfiu:n test speed (a Mach number of 0.68) but tiie effectiveness of elevator 2 decreased sharply at Mach numbers above 0=56. C m ) • he adverse effect of fabric deflection on ele- vator hinge moment was decreased sligiitl^r by locating the vent holes In the leading edge rather than at the trailing edge of the elevator -o^ Langley .Memorial Aeronautical Laboratory ITational Advisory Committee for Aeronautics Langley Field, Va . l6 NACA ARR No. L'^FOla REPER^x^CES 1. TIelhouse, A. I., and P^emp , 77. B. : Effect cf Fabric Deflection on Rviader Hinge -Moment Charac- teristics as retevmined by V/ind-Tunnel Tests. :J\CP. ARR No. T5"-^-4, 19^^^. 2. Swaneon, Robert £ . , and Toll, Thomas A.: Jet-3oundary Corrections for Rei lecticn-^lsne Models in Rectan- gular "-Vind Tunnels. NACA ARR No. ?E22, 19'+3 . J. ?err?.ng, 'V. G. <-. . : The Theoretical Relationships for an Aer^ofcil with a T.'^ultiply Ringed Flap System. R. & M. No. 1171, British A.R.C., I928". I|.. Llathews, Charles 7/. : A-i Analytical Investigation of the Effects of Elevator-Fabric Distortion on the Longitu.dinal Stability and Control of an airplane. NACA ACR No. L/^EJC , I9lih' IJACA ARR ^To. r5F01a TABLE I ELE\/.VxCH ?><.ERIC TENS 10 KS [Accurucy of n.easiir ement , ^0.2 lb/in.] 17 Test condition Tension (lb/in.) 1 Upper siarface Lower surface i j Max , i:i n . Max . Min. i Elevator 1 ! 1 1 Before testing After testing 7.I^ 7.6 6.5 6.7 8.6 8.8 6.8 ( . t> 1 Elevator 2 Before testing After testing 1 7.2 5.2 7.0 6.0 7-5 6.7 6.8 6.0 NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS L NACA ARR No. L5F01a Fig. «3 nJ c QJ o C N C ■H 3 u ■P o x: Id 01 c q; a p. (X CO CO 1 •H jq e bo 0) •H ra XI i-i ■P 1 O Q o W V-i OQ 1 CQ ^D «—( V-i O P>5 (U s -— 1 0) bo •H C > td J r-( 03 C ;j •H (U c t3 QJ 0) O .— 1 -H 1 cd • ■P 1— 1 to C 0) •H u 3 bo NACA ARR No. L5F01a FiR. 2 Z uj ii 8 o 1- _ cj m M o o 0> O O 3 c o Isl Q ^ o - c 3 I i £ I T64!IJ*A ^ - S - (r < < H < o a> f c ■o 5 ? U. P LU Q llJ U. U_ O O O i _l _J f? ^^M IC < < 1- i UJ o CJ Q o i. ro i NACA ARR No. L5F01a Fig. 4 torque tube denotes fabric covered metal ELEVATOR I Denotes fabric- Denotes ribs covered metal ELEVATOR NATIONAL ADVISORY COMMITTEE FOt AEIIONIUTICS FIGURE 4 .—RIB SPACING FOR THE TWO FABRIC - COVERED ELEVATORS. NACA ARR No. L5F01a Fig. 5 NACA ARR No. L5F01a Fig. 6a, b :'4 ,^r ^^^^ NACA ^^^ LMAL38341 (a) Upper surface. ( b ) Lower surface . Figure 6.- Static condition for elevator 2. NACA ARR No. L5F01a Fig. 7 a- 3 9 E o o> o « > < 520 480 440 400 360 320 280 15x10" 240 200 160 120 .2 .3 .6 M Si E 3 CO S o c >> 01 o o» o k. > < Figure 7. -Variation of the overage test Reynolds number and dynamic pressure with test Mach number. NACA ARR No. L5F01a ■ig. 8 u o ■fj Id > .— 1 W • rH "=)* (U O oa di a ■' bD NACA ARR No. L5F01a Fig. 9a, b (a) Upper surface. NACA LMAL 3834& — ., — V \ -:h ^zia k 1 L^^^^^^^H ¥" "' ' \ ( b ) Lower surface . Figure 9.- Fabric deflection of elevator 2 (8-inch rib spacini M = 0.55; a = 0"; S = vents at trailing edge. 3.3^ elevator seal removed; elevator NACA ARR No. L5F01a Fig. 10a, b (a) Upper surface. (b) Lower surface. Figure 10.- •7 o . a = o , Fabric deflection of elevator 2. M •= 0.62; S = -0.7°; elevator vents at trailing edge. NACA ARR No. L5F01a Fig. 11 I tn 10 c c o o (> o •o (> o J3 5 ^ £ n 01 o o 1 3 » 0) n o =) -1 — , in — O 1 n ■ 2^ t \ A "~* * r / / + 1 1 ~^ — — \-- - i 1 1 r i i J / 1 / i 1 1 1 1 1 1 \ ( \ +• \ _j \ \\ \\ < m lo St n n v= ^ 1 1 1 1 1 ^ 1 +) » 4 n T r~ ' \ f 1 1 t \- 1 X 1 \ \ \ \ \ \ \ \ > \ \ _4 ' ■ o d 1 ■ ■ i 1 1 1 / 1 ; y 1 I i n j 1 1 '' • 1 \ ) \' \ i \ \ \ f \' \> o o® cvj — 2^» — — ._ — 1 1 1 1 1 - — 1 1 I 1 ^ 1 .1 ' , + CM CVJ r «* CSj O CM r CM O CM O o £ Z o o c C ■A o a> (0 Sj V^ "O o> Sg M Q C i| -=- 00 ^^ .£ o> 00 gg si if o CO o !2 o CMih ■s o No o « > 0> o> o ».E o2 2 £ <0 V c £ o £.E «) "O _ fl> « o> a. c 0) a> s» o O "O w u ■o 3 'C o S «"^ « .2 O O o »- O O c o> CMo „5-o to* ); <» ■S <" c E £^.9 2 «t3 ••- 0) ^ 'i- -o-^S CMa> u O. c *^ 11 o o " a, u> «> 10° O u, o •P " s? o 0) 2 •i:^ c oE-r, > 3 ^ c c CO 0, ° £ £ -Js. jO o u> _ 3 O a> i_ ^ ^ o o S. > *- o> 'ui 'uoj(38||3p 0!jqo^-Jo;oA3|3 NACA ARR No. L5F01a Fig. 11 Cone. v> in fe c o 13 o Oi 0) «»- <^* « QT •o ■a u c ?i o n II I. lO CO •o c o — *^ o o > CVJ «> to E p CM tt o c o M '5 (0 00 CM CM f * CM O CM * CM O •uj 'uonoaipp 3!jqD^-J0iDA3i3 CM r ■o 3 o o 3 NACA ARR No. L5F01a Fig. 12 in c o I/) c o o 0) o 0) ■o o o o ^ £ 00 oo o u _c ' 1 Q> o |v^ U) 00 o o 0) £*» 71 CM T3 N- S 4- « f o o> W > c *- * o (0 a> o c ^ o (0 c « » o o E o (0 £ in « o c 0) T3 a> a> •o f —

« c o « ■o o Pi 9 o s ^ w « s o > TJ * •^ 0) If) in .1 CM « o c H- o 11 s. 1;; o s c c T) (0 u o o o C o o > u 3 00 « c o o •4- o CM o w 4 « 0) o c « 3 a> T) a> u. NACA ARR No. L5F01a Fig. 12 Cone. r (/) M c C o o <) o a> Q) H— H— 0> 0) ■o TJ () O k. o X> o o 8 o 0) u o T 3 w Ifi w w a> CL 5 a o Z» _l \ \ \ T < I / / i +/ / / t , 1 1 1 1 1 ^i 1 1 , 1 -4 \ ^ J \ \ \ V \ \ ^V \\ \ \ t \ 1 t I I "^ ■ J , ==; tzz 1 r T ^ r 1 t I t i^ t / ^ i_ J L 7 1 T t / 4_ 1 > t 4 t j \ ^ ■^ \ T -:^- L_ ^ — 1 1 \ \ \ 1 \ J \ ] \ / 1 o ro- ii J 1 . I 1 / 1 i 1 J 1 / / Q A / i / f 1 ., 1 / / ^1 ' \ ^ + s \ ^ V \ OJ.OCM'* 3 u c o o 1 ^ r c VJ C 3 CM r 3 NACA 'ARR No. L5F01a Fig. 13 I u> in c o c o o *•- 0) 0) T3 •o U u ^ ^ n o n s- a> u o 3 3 (0 u> V- J. a> V Q.5 a. O 3-1 ( ( 1 \ 1 \ \ \ 2 1 o una e S 1 o u n a \ ^\ -- o o / / / V V c / / 1 / ^ t + / / ' ' ♦ a / 1 / ( i 4 / 7* 1 1 1 / 1 1 V 1 f / 1 t / 1 1 [ 1 1, \ \ * *"" \ \ \ i, \ ^ ' ^ > \- \ \ \ — ( (_ \ \ ^ \ — 6 V. Kiin lO 1 d una/ i o 1 * ( ( V / t J 7 / 1 ■> 1 (1 ' 1 ^ 1 S 1 1 \ \ I \ r \ \ ■K i 1 1 \ \ o z — (rt S > S - 8 V_ V i 1 \ 1 1 V i \ 1 \ ; ' II U u 1 o + ; ; G h t 1 1 O r 1 t ) / / c U / '^ f 1 / 1 i 1 1 \ 1 V c 1 \ ^ \ \ A _j * CVJ CVJ OJ CM CJ CM r CVJ O O (0 o 00 ^ I? > o .^ a> « 5 o a> £ CJi o -a T3 O i: 2 «> -2 E O M- ■o o u c a> u w O a. •o o .2 ^S i E p to I CO CM I" 'Ul 'U0!i39|;3p 3UqD;-J0iDA»|3 *- c » - c ■o — .9 .. 0) <" ■^ o> -S O c > c > . o to x: (u ^ S Q) NACA ARR No. L5F01a Fig. 14 <, » \ \ \ b \ ■ \ \ > CO ^ CM O -U! '930^ J nS J9M0| 9l|j ^0 U0IP9|^9P oinujixo^ CM g (O mil ro w o > a Z ui p Is 8 — —CO in —in — CVI c o o 0) a> "O o I > a> c o Cl If) S 0) o. T ■4 u — "^ £ o c -« ^ O >s ^w Q ro o II 8 UJ CVJ — o NACA ARR No. L5F01a Fig. 15 -.6 -4 -2 c - 2 0) o u 9> m a. .6 .8 1.0 Tailed symbols denote lower surface NATIONM. ADVISORY COHMirrtE FO* AEMNAUTICS Figure 15 . — Pressure distribution of elevotor 3 for three elevator positions. 1 = 0"; M=0.20; gap sealed ; 33- inch station. NACA ARR No. L5F01a Fig. 16 -1.2 -1.0 -8 -6 Q- -4 c ■^ -2 0) o o I 2 .8 1.0 I Tai ed symbols denote lower surface NATIONAL ADVISORY , COMMITTEE F« AEDON/UJTICS Figure 16.— Pressure distribution of elevator 3 for three elevator positions, a- 3°; gap sealed; 33-inch station; M = 0.20. NACA ARR No. L5F01a Fig. 17 \ -1.6 -1.4 -12 -1.0 -8 ^ -6 c "o O U -4 .2 .4 y \K oo— o- _>- _^ -o — - o- 0) C A *3 O '/ S (deg) " -0.4 * - 3.8 3.1 Q I Toiled symbols denote lower surface \ k\ ^^ \ ^ ^^ \ V N \ + N ^ '^ + A V ^J \ ^* \ X ^ ^"l \ + \ 1 V V I 1 N 5^ \ \ f ^ i^ 1 1 / /* / CC 1 1 1 NATIONAL ADVISORY HMITTEE rot AEMNUITICC Figure 17. — Pressure distribution of elevotor 3 for three elevator positions. flc«6"*; M=0.20 ; gap sealed; 33-inch stotioa NACA ARR No. L5F01a Fig. I 00 c 0) u «^- «4— ID sure coe 2. ^ — Q) OJ or interna VI = 0.20;el <".) o> ^■v ^^" 0) 9 - •o 0) o ^ oiO II CVJ 1 of the tion. a 1 Variation tor posi CO 1 o > £.€ §.» d Uuepj^^doo ajnssdJd NACA ARR No. L5F01a Fig. 19a-c I - < 1 ■♦• 1 1 o o eflection surf pper ower ■i \ \ \ ■CD -■ o 1 1 1 1 1 1 1 1 * 1 / ; f 1 i 1 1 1 I \ \ \ \ \ ( t ■o S E O 0) O o I c o v> c 2__ 1 ■ ► i . (A O 3 c o Q. O. 3 0) u O d 1/1 1 5^ o| !^ ^ ^ig gst *iT n 3- 1 (c) Vents -^-inch from elevator leading edge above and below fabric seal. \ \ I \ A M \ \i !l ( i li / ) / ^ / - - o o ^ g) ID Od CVJ I' OJ CM r OJ I" 1 uj 'uojpaipp 3uqD^-J040A8|3 o c o o a> o Vi 1 c o o 1 ID Q> oc In o 3 U ■D CVI k- o o ^ -8 t^ o o 1 ^^ a> 8 o c u KD a. ■o c o o o o JC ih C3> c ■o •D c 5 q o a> o ■fi O '3- o o .1 Q. > u in a> in b II o b O S r fO >4- f o 0) o ^ 0) c <3) II >- o ^ H> u. ( > >t ^ CM (-) O b II i3 .o b ^ o o ^ ? c 0) O ^ o fi- s 3 c lA c o o 2? 8 o Q) r > ■o NACA ARR No. L5F01a Fig. 20a, b (a) Upper surface. ( b ) Lower surface . Figure 20. a = 0°: Fabric deflection of elevator 2. M = 0.55; 8 = 4°; elevator vents at leading edge. NACA ARR No. L5F01a Fig. 21a, b 08 cT \deg) '9^ .04\-6- s (deg) -4" -2- -2— -2-^ O— -4~ 2 — -6= = - 4 — -04 .10 -10 - — - 1 -4A 4\ ^ 0^ -^ ^ '"^ :|^ = — -< I' ir 0— — o -4A 2- -2- .4-z -6- -4~ ~ 1 2 -4^ ■ — -— .6 2 — 4 -Q- -2 4 2 /7 -4 1 4- -9 .2 1 -6 NATIONAL ADVISORY -9 OMMCn^F mt AFMUiAHTirc .2 .5 M M (0)00 = 6''. (d)(x:=9°. Figure 21 . — Conoluded. NACA ARR No. L5F01a Fig. 22a, b JL/<^ r / o (deg) (deg) 04 -9 -|-| -4- = = = = = = - -= -2 2 ^ .0 _ ^ 9 _ ■* - -= H^ 04 5 jl - — . 1 JO -10' f - - - - - - - - - - - ;ii - = - = - t = = = = = = E = E = - * E _ ^ - = f -4^ = = = = E = = — — — — 4-^ — = = - — — — 6-~ & 1 1 .03 .02 .01 4 .2 -2 6- -9^ -%< 4-. 2 ^ Co -5 -— ■ — 1 1 ' — 1 _--^ ? 6 - 4 2 6 4 — ■ U -2 c, 2 -2 -4 -4 -s -9 -6 NATIONAL ADVISORY -d — CO HMITT " tt l-U R UM HI«UT C5 6 .2 .3 4 .2 .3 4 .5 M M (0)00=0". (b)ac=3°. Figure 22. — Variotion of the aerodynamic characteristics with Moch number for ranges of elevator angle and angle of attack ; elevator 2. NACA ARR No. L5F01a Fig. 22c, d P8 .04 -.04 (deg) -5-|=f= r (den ; _ ^ II ^' ^ _ -^ ■»/ _ ,-- -- -6 -- — — =■ = — - _^ " — — — ==.- -■= ^4 = - — — -^ -6 - -— _- — — -4 - -- = = = — - - - -2--^ -0-- = 2 — .4 — - - = - — — = = = ~2- 4 - - 'm ,/^ 1 > _ _ __ — F= ~ _ — _ 4" — — — — — — — "^ -T- — -" _ — — = = ' _ _ -^-= = = = = = = = ir _ _ — "?^ = = = = = = = — — — 'iz - ~ = = — — — — — — — — ^^ 4 .06 .05 04 .03 .02 .8 .6 .-^ .2 4 ^ " 2 .-^ ^ / ^ • ' / ^0 -2 __, y / 4 — ^ / / 2 / -4 , ^ V/ -6 _____^ ^ ^ -2 — ^ y -9 — ^ 4 - — ■ — 2' ■ 0, 4 -2~ -2 -4 — -2 -4 _ -6 -9 ' ' NATIONAL ADVISORY J -6 --,9- CO HHITT Et FM AEM NtUI CS .2 (c)a=6' M M (d)(t=9''. Figure 22 .— Concluded. NACA ARR No. L5F01a Fig. 23 .07 .06 .05 6 Ideg) 0) E o o Q. .07 ^. .06 .05 Elevator 2 I I L M .6 .7 NATIONAL ADVISORY COMMITTEE FOft AERONAUTICS Figure 23. — Variation of the lift parameter 0. with Mach number for elevators I and 2. a NACA ARR No. L5F01a Fig. 24a, b .040 .030 .020 0- a (deg) -\ ^ y / -9 ^ ^ • r "^ ' \ 6 (a) Elevator I . o >.040 .030 .020 .010 (deg) 6 and 9 NATIONAL ADVISORY COMMiTTEE FM AEIIONMmCS 4 .5 M (b) Elevator 2. Figure 24. Variation of the lift parameter Cl^ with Mach number for elevators and<^2. i. L NACA ARR No. L5F01a Fig. 25 I.-, I 1 CVI 1 ' 13 O o o o si 11 \ UJ > Q> llJ §9 ^5 \ ^ ^^ ^ \ / \ I ^ 8 1 u o o 2 o 1 1 M- 1 ^ 1 ♦- 1 ^ J IV. f— k^ vo lO ro CVI /Wop:) * SSaU9Aip8^^9 J0iDA9|3 V0 ID ro CVJ O Vi c o Q> H— *^- 0> > q> Q> c o E 3 II CVI C o_ O) *♦— LJ w O > 3*4- NACA ARR No. L5F01a Fig. 26 ^ (^opAp) c 5 1 - CVJ 1 • X NATIONAL ADVISORY COHMITTEE FOB AERONAUTICS \ 1 \ \ \ -\ I 1 v 1 t \ 1/ Elevator 1 Elevator 2 \ \ \ ; 1 [ \ \\ If 1 { J p 1 p" o 3 o IS ID lO rO CVJ GO Q Q Q CVJ Q CVJ Q I 9 a a> E o o Q. C a> E o fcvi c c I- 0)2 ^ o ••- > •ft "55 c o JO O' o. E o o C3» (■•op/ ""OP) NACA ARR No. L5F01a Fig. 27 Z lu ( CVJ o .o . 1 / / Al / / i / / ^ 1 1 o o > \ UJ o o iD lO ro CJ 1 o a> o» c o • .Co oo II M-OJ C w o Q cvi O O o (MX) |i NACA ARR No. L5F01a Fig. 28 100-percent aerodynamic balance -.002 X O -.004 -006 -.008 -010 -.012 Eevator I Z Effect of fabric deflection Theoretical values based on the factor (l-M*)-'A: Zero aerodynamic balance (Theoretical, reference 3), .3 4 .5 M .6 7 NATIONAL ADVISORY COMMITTEE FOA AERONAUTICS Figure 28 . — Variation of the hinge-moment parameter Chr with Mach number for elevators land 2 ; oc =0" or*^3^ NACA ARR No. L5F01a Fig. 29 Fabric seol^ o c o o o .04 .02 c E o E -02 c -04 -06 'Vents 1/4 in. fronr^ leading edge Wents at trailing edge, lower _ surface only _ Vents at O.IO ce — ,z^ Vents at trailing edge NATIONAL ADVISORV COMMITTEE FOR AERONAUTICS Vents at, leading edg^ ^ Vents at 0.10 Ce ^ All vents sealed -4 -2024 Elevator angle, deg Figure 29 . — Effect of elevator vent location on elevator hinge moment; a=0**; M=0.55; elevator 2. 1 NACA ARR No, L5F01a Fig. 30 la H z S c c o a } 9 ' 1 c :> 00 GO p / / flO <£> lO w w a> c o Q> o E c o "O Q. ro »«r CM o M— LU 00 O O o 00 o o II roo 0) II '0\7 UNIVERSITY OF FLORIDA 3 1262 08106 456 9 UNIVERSITY OF FLORIDA DOCUMENTS DEPARTMENT GAINESVILLE. FL 32611-7011 USA