\l(\^(^l'^^ If! NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WARTIMK REPORT ORIGINALLY ISSUED August I9U5 as Advance Eestricted Report L5F30a HUTCEE MOMENTS OF SEALED-lN'i'JilHHAL -BALAKCE AERAMSEMENTS FOE COBTROL STIRFACES H - EXPEEaMEUTAL INVESTIGATION OF FABRIC SEALS HI THE PRESENCE OF A THITI-PLArE OVEEmAIKJ By Jack Flschel Langley MemDrlal Aeronautical Laboratory Langley Field, Ta. ^'^ NACA "^ WASHINGTON NACA WARTIME REPORTS are reprints of papers originally issued to provide rapid distribution of advance research results to an authorized group requiring them for the war effort. They were pre- viously held under a security status but are now unclassified. Some of these reports were not tech- nically edited. All have been reproduced without change in order to expedite general distribution. i'-^2 DOCUMENTS DEPARTMENT Digitized by tine Internet Arcliive in 2011 witli funding from University of Florida, George A. Smathers Libraries with support from LYRASIS and the Sloan Foundation http://www.archive.org/details/hingemomentsofseOOIang f 17- i<20 (f f NACA ARR ]\To. L5P30a NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS ADVANCE RESTRICTED REPORT HINGE MOI\ENTS OF SEALED-INTERNAL-BALANCE ARRANGEIVISNTS FOR CONTROL SURFACES II - SXPERILSNTAL IN\^STIGATION OF FABRIC SEALS IN TEE PRESENCE OF A THIN-PLATE OVERHANG B5' Jack Fischel Tests were made in a seal test chamber to deterinine the hinge rtioments contributed by the. fabric seal in an internal-balance arrai-igement employing a thin-plate over- hang. These tests were performed mth various widths of fabric sealing various widths of flap-nose gap, vifith a horizontal, a vertical, and a circular type of wing structure forward of the balance, and 'with various heights of balance chamber. This investigation is an experimental verification and extension of a previous analytical investigation. The present investigation indicated that the moment of the seal may be a balancing or an unbalancing moment and may be an appreciable part of the total balancing moment of an internally balanced flap, depending on the overhang deflection and the configuration of the internal balance. Variation of the width of the fabric seal, the sealed gap, or the location of the seal attachment to the wing struc- ture affected the seal moments through most of the over- hang deflection range. The shape and size of the balance chambei'' affected the seal -moment characteristics in the deflection range v/here the seals contacted and were con- strained by the chamber walls; the values of the seal moments were usually/ reduced when the seals were con- strained. The results indicated also that sn optimura balance configuration would employ a seal width such tb^t the seal vjould barely touch the chamber ceiling when maxim\xm overhang deflection Is attained. NACA ARR Fo. L5F30a INTRODUCTION One of the devices employed for balancing control surfaces, especially ailerons, of ir.odern high-speed air- planes is the sealed internal balance (fig. 1). Much of the experimental work that has been done in the develop- ment of the sealed internal balance has been summarized In reference 1, v/hich includes a brief consideration of seal effects. In reference 1 (and in all previous internal- balance work), the balancing effect of a seal in an internal-balance arrangement was accounted for by an approximate m.ethod - that is, by ass-'om.ing the balance chord to be the distance from the control-surface hing-s axis to the center of the sealed nose gap, regardless of seal vv'idth. An analytical investigation to determine the contri- bution of the seal to the balancing moment of a sealed- internal-balance arrangem.ent Indicated that the approxi- mate m^ethod was generally in error over most of the flap- deflection range (reference 2). , This analysis also indicated that variation of either the seal vjidth, the sealed-gap width, or the shape of the wing structure forwfard of the overhang has an important effect on the balancing moment contributed by the seal. The present investigation is an extension of the investigation reported in reference 2 and was begun in order to check experim.entally the fabric-seal analysis reported therein. The present Investigation covered a larger deflection range than was covered in reference 2? in addition, the effect on the seal mom.ents of a con- fining balance-cham-ber roof (the contour of the airfoil) and an off-center attachm.ent of the seal to the vv'ing structure were determined. The moment contributed by the seal in each of the configurations tested is presented as a fraction of the moment of the thin-plate overhang over the complete deflection range. A com.parison is m^ade of the seal m.oments obtained experim^entally and analytically (refer- ence 2) with seals and sealed nose gaps of similar size in the presence of each of the three different types of internal wing structure with no vertical restrictions. In addition, a comparison is made between the experi- mentally determined and the computed hinge-mom.ent coef- ficients of an internally balanced aileron, and an NACA ARR Ko. L5F50a example is presented to show the relative accuracy of hinge -moment data com.pnted by the approzimate method and those computed from the results of the present investiga- tion. SYMBOLS mg seal-moment ratio (M^/M. ) ■jp pileron section hinge-moment coefficient (h /qcg_ j a Ac-j^ increment in aileron section hinge-mom.ent coef- ficient produced by an internal-balance arrange- ment ^a Pp pressure coefficient .across balance (pressure below balance manus pressure above balance divided by dynamic pressure) 5, thin-plate overhang deflection, degrees; positive when deflection is increased from neutral b^r pressure across balance 5-. limiting deflection of thin-plate overhang, 7 degrees (deflection just prior to contact betv^een leading edge of overhang and. roof of test chamber (contour of airfoil)) .6„ aileron deflection v/ith respect to airfoil chord line, degrees M. moment of thin-plate oi^erhang (used with subscripts exp and comp to indicate experimental and com- puted, respectively), foot-pounds M_ moment of seal, foot-pounds g width of sealed gap between points of attachjuent of seal when Sj^ = 0", fraction of c^j s seal i,vidth., fraction of Cv, h aileron section hina:e mcm.ent , foct-nounds t thickness of overhang at hinge axis, fraction of c k NACA ARR Fo. L5P50a c airfoil chord, feet except when otherwise indicated c aileron chord behind the hinge line, fraction of c a ^ Ci_ overhang chord from flap hinge line to leading edge of overhang, fraction of c q dynamic pressure, pounds per square foot ^— pV^^ V absolute velocity of air stream, feet per second p mass density of air, slugs per cubic foot M Mach number (V/a) a velocity of sound in air stream.^ feet per second a angle of attack, degrees APPARATUS AND METHODS The seals were tested in a specially prepared seal test cham-ber that simulated the construction of an internal-balance chamber ahead of the flap hinge line (figs. 2(a) and 2(b)). The span of the overhang was 2i| inches and the chor'd was 10 inches from the hinge line to the leading edge. The thin-plate overhang was rigidly attached to a torque tube that was, in turn, attached to a dial outside the test chamber and deflected the over- hang thi'ough the test langs. A clearance of 5/6Ll inch was allowed at each end of the overhang span to prevent contact with the side walls through the deflection range. A small clearance behind the hinge line between the torque tube and the test-chamiber structure was sealed v;lth a small fabric seal; the m.oment produced by the seal was considered in the calculations. The difference in normal pressure existing across the seal and overhang of an internal-balance arrangement was simulated by the con- trolled pressure produced in the part of the test chamber below the overhang by a blower, while atmospheric pres- sure existed above the overhang. The pressure across the overhang and seal was indicated by a microm-anom.eter and this pressure was m.aintained at approxim.ately 17 pounds per square foot b^r a door on the blower intake. The distribution of pressure in the region below the overhang FACA ARR No. L5F30a and the seal vias deterrrined from a trief survey to be unlfor-j^i viiithin appi'oylmately ±1 percent; a pressure drop of approxirrately IS to ^0 percent was found to occur within 1/16 inch of either end of the overhang- span, about which a flow took place, bat the effect of this pressure difference on the hinge moments is believed to be negligible. One of the chordwise chamber walls was made of plexiglass through v-hich photographs were taken of the seal profile under various conditions. The hinge moments of the balance arrangement were determined 07 means of a calibrated torque-rod system built for this setup (figs. 2(a) and 2(b)). The over- hang deflection was determined by the reading of the overhang-deflection dial v/i th respect to a pointer attached to the outer wall of the test chamber (fig. 2(a)) Three types of balance-chamber structure ahead of the overhang were used in the Investigation. These surfaces are shown in figure 2(a) and are referred to herein as backplates. The backplates were of uniform. height and span; their chordwise positions were varied during the tests to give various sizes of gap. The seals tested were made of Koroseal, which is an air- tight, flexible, fabric material, and had a varying chord width and. a span equal to that of the overhang. A thin metal strip was fastened along the span at both ends of each seal to attach the seal to the backplate and to the leading edge of the overhang (fig. 2(a)). The vertical balance -cham.ber restriction (contour of the airfoil) Awas slmiUlated by a horizontally held plate; the distance of this plate above the o^^'erhang and seal vjas varied to give the proper value of 6-^ . For b unlformiity and agreement with reference 2, all the linear dimensions of the balance configuration are expressed as a fraction of the overhang chord. A list of the balance configurations tested is given in table I. Since the pressure difference across the balance of a control surface does not always reverse when the control is neutral, tests were m^ade at negative deflections up to -12° and the tests were run with the overhang deflection varying in 2° and li° Increm.ents up to 50'^ C!^ -he miaximum deflection allowed by the seal or TAG A ARR No. L5?50a overhang. The results are applicable to both negative and positive flap-deflection ranges, however, and the change in direction of the balancing inoinent is deter- mined by the deflection at which the sign of the pres- sure difference across the balance changes. In obtaining the hinge moment due to the seal, the moment of the overhang had to be subtracted from, the total moment of overhang and seal measured by the torqxie system. The moment of the overhang alone, without any seal present, v;as obtained over the deflection range by closing the gap betv^-een the overhang nose and the back- plate to a very small value . Because of the large leakage area for this condition, a pressure difference of only about 10 poiands per square foot, considerably less than the normal test pressure, could be maintained. The overhang moment thus obtained experimentally was com.pared with the overhang m.oment computed from the thin- plate dim.ensions and the pressure difference across the overhang. The experim.ental moment was found to be approxi' mately 1 percent higher than the computed moment. The seal moments were therefore obtained b;/ subtracting the computed overhang moment, corrected for the 1-percent discrepancy, from the total m.oment m-easured in each test. The com.putations for obtaining the seal-m.oment ratio are indicated in the following equation: ^-s ^s ^^total exo " I'^l K^, omn M^ 1.01 M>. c om^p RESULTS AND DISCUSSION. Seal-Frofile Photographs Som.e typical profiles and positions of the seals with a pressure difference across the balance are shown in figures 5 to 7 with a vertical backplate, in figures 8 and 9 with a circular backplate, and in figures 10 to 13 with s horizontal backplate. The constraining effect on the profile of the seal caused by the vertical and circular backplates in the positive d.ef lection range is shown in figures 5j ^j ^'^^ ^3 whereas figure 10 indicates MCA ARR Fo. L5F50a that the horizontal backplate had little or no confining effect i Similarly, figures 5, 7, 8, 11, and 12 indicate the confining effect of an overhead restriction siiriulating the top or bottom of the balance chamber. When a seal was free to billow xinrestricted by backplate or overhead limit, the seal generally tended to assume a circular shape. This fact, which formed the basis for the ana- lytical work of reference 2, is illustrated in figure 10(d) by a circle superimposed on an enlargement of the photo- graph of figure 10(b). Desired Seal-Moment Characteristics A desired variation of the seal-mom.ent ratio m with overhang deflection is one that offers a positive value of the slope dm^/b5-^ with deflection. (See refer- ence 2. ) This variation of mg Vv'i th deflection would tend to compensate for the decrease in the variation of pressure coefficient across the balance P with deflec- R tion (that is, the decrease in dPp/dS]-,) as the deflec- tion increases and to provide more nearly linear balance hinge moments and control forces. Experimental Seal-Mom-ent Characteristics The se-al-m.oment characteristics over the deflection range for various sizes of fabric seal and sealed nose gap, without and with overhead limits, are shovm in figures li^, to 21 vjhen the seals were tested with a vertical backplate, in figures 22 to 26 with a circular backplate, and in figures 27 to 5^ with a horizontal backplate. With all three types of baclcolate, the mom.ent exerted by the seal in the balancing configuration is appreciable, particularly with large seals and large sealed gaps. This seal moment may. be a balancing moment amounting to as much as I4.O or 50 percent of the overhang balancing moment, or it m.ay be an unbalancing m^oment amounting to as much as 50 or i|.0 percent of the overhang balancing moment, depending on the overhang deflection and the configuration of the internal balance. Negative values of m. were somietimes obtained with all three types of backplate in the negative deflection range and over a small portion of the positive deflection range ne-ar zero 8 MCA ARR No. LSFJOa deflection (figs, ll; , I5 , 22, 27, anc 28). In these configurations the seal overlapped a part of the over- hang, which equalized the pressure on "both sides of this part of the balance and tended to reduce the amO'Xat of effective balance available. A configuration illustrating this tendency vas not photographed, but this condition is approached in figure 9'^)> except that the seal actually lies flat against the overhang when m^ is negative in the discussed deflection range. Except -when a sealed-gap width of O.5 is used, the slopes of the seal-moment curves at small positive values of deflec- tion are usually positive and indicate an increasing balancing effect in this range. This increase in balancing effect is independent of that obtained by the increased pressure difference across the balance V'ihen the flap is deflected or vi/hen the angle of attack of an air- foil is increased and is a function of the seal. At high positive values of overhang deflection, as the seal became taut, the seal-moment ratio of the unrestricted seal decreased and became negative and the direction of its force was opposite to that of the overhang balancing force (figs. 5^0-) ^-^d 10(c)). Effect of the Balance Configuration on the Seal-Moment Characteristics Effect of seal mdth .- The effect on the seal- moment claracteri sties of varying the seal width, with other variables kept constant, is shown in figure ^5 and is also evident from fig;.ires lli to 19, 2.2. to 2li, and 27 to 52. For a given sealed-gap width, v: depends on seal width. The curves indicate that m^ decreases in the negative deflection range and generally at small positive deflections and ixsually increases at large positive deflections as the seal width increases. The curves indicate also that, for a given sealed-gap v/idth and as the seal width increases, the maximum value of m.g generall^r increases, except when the circular backplate is used, and the maximum value of m.g occurs at an overhang deflection that increases with seal width, regardless of the backplate. NAG A APR Fo. L^FjOa 9 Effect of S3al8d-g ap -width.- Ths effsot of sealed- gap w-idth on seal-moment ciaappicteri sties is indicated in figure 36 and in figures ik to I9, 22 to 2li, and 2? to 32. For a given seal width, the seul-iroraent ratio generally increases with seal8d~g?p width at small deflections and decreases at large deflections. gffect of backplate .- The seal-moment characteristics obtained with the three backplatos for given sealed-gap and seal widths generally differ only in the deflection range in vdiich the seal lies against the backplate; hence the values of m in this range depend on the type of backplate contscted and the effect this backplate has on the seal profile. A comparison of the seal m.oments obtained over the deflection range v/ith constant gaps and seals for the three different backplate arrangements Is given in figure 37 for the condition in v.hich no vertical restrictions are used. The characteristics exhibited by the seal with a vertical backplate and a circular backplate are quite similar (see also figs. II4. to 21 and 22 to 25) and indicate some linearity over a part of the deflection range. The horizontal backplate has little or no effect on the m.oments produced by the seal; these m.oments are usually larger at small deflec- tions and smaller at large deflections than those obtained v/ith the circular or vertical backplate. The effect of the backplate appears to dimiinlsh with an increase in size of the sealed gap and the seal, and the character- istics obtained with the three backplates differ by only a small amount vidth medium, and large gaps and seals (figs, 57 s^<^^ '^h to ^h) . At sm.all gaps, the effect of a backplate constraining the seal is to cause the maximum m.om.ent of the seal to be reduced and to be developed, at a higher deflection (fig. 37(9^))" Sffect of vertical restriction .- Limiting the height of the balance cham.be r reduced the values of rr.^ over that part of the deflection range in which the seal con- tacted the roof of the chamber (figs. 11 and 1I4. to 34)' Increasing the seal width or the overhang deflection in this condition or decreasing the value of the limiting overhang deflection 5h caused a greater reduction in mg 7, as more of the seal contacted both the chamoer ceiling and the backplate. The deflection at which the seal con- tacted the chamber ceiling and the seal m.om.ent started to decrease was greater v;ith the horizontal backplate 10 iVACA ARR No. L5F50a than with the two other types tested. Vhen no vertical restriction was present, the iraxiinuru values of m„ usually occurred at higher values of overhang deflection and the values of mg vrere greater over the deflection range affected than when a restriction was used to simu- late the roof of the balance chamber. It appears, therefore, that an optimum balance configuration would employ a seal width such that the seal would barely touch the chamber ceiling when maxim^um deflection is attained. (See figs, p and 8.) O ff-cente r sea l attachm ent . - The seal-moment char- acter! sTircT'~obraTnFd~wil:h~a C.3i4-C|_) off-center attachment of the seal to the backplate are shov/n in figures 20, 2.1, 25 J 2.6, 55, and ^l\. when tested with the three types of backplate used In the ir-vestigation. (The value of O.jkcjj of the off-center seal attachment corresponds to an offset of the position of seal attachm.ent to the backplate to the top or bottom, of the balance chamber v/hen 5vj = 20'^.) In addition, a com.parison of the seal- m.oment claracteristlcs obtained in the presence of the circular backplate mth the seal attached to the wing structure at the center and off-center positions is shown in figure 25- The seal-moment characteristics obtained with the off-center seal attachjnent were gener- ally unfavorable over the deflection range because a decreasing balancing tendency or an inbreasing unbal- ancing tendency is indicated, regardless of the type of backplate or of the point of attachment. Vi/'hen attached above center, the seal invariably lay against the back- plate and overhead restriction at positive deflections (figs. 7 ^nd 12); when attached below center (figs. 6 and 15)5 however, the seal had a mom.ent vector that would decrease positively, then increase negatively with over- hang deflection. These effects account for the unbal- ancing characteristics of this type of seal attachment. The effect of attaching the seal off center to a circular backplate when the seal did not contact the balance-chamber ceiling was to shift the seal-mom.ent curve by an angle the sine of y/hich was equal to the off- center dlsplacem.ent (expressed as a fraction of the over- hang) divided by the radius of the backplate arc (also expressed as a fraction of the overhang). The curves of figure 25 approximately verify this conclusion; the com.puted offset angle was sin" ° '" ■ ' - = 15.2'^, and NACA ARR No. L5F50a 11 the data of figure 25 indicate that the offset angle is about 16 . Fince an off-center seal attachment produces an unbalancing effect that may be greater than the increased balancing effect on the overhang caused by flap deflection or the increase in the angle of attack, this type of seal attachment is believed to be undesirable and should be Comparison of Analytical and Experimental Seal -Moment Characteristics The similarity between the analytical (reference 2) and experimental seal-momient characteristics is evident in fig\ire 37 > v.hich shows the agreement between these results. The analytical and experimental results were compared for several configurations in addition to those herein and slm.ilar agreem_ent Vv'as obtained. Comrputation of Seal-'vloment Characteristics by Approximate Kethod The seal-moment ratio computed by the aDproxim.ate method (reference 1) for various sizes of sealed gap is shovv-n in figure jS. This miethod of computing the seal- moment ratio is independent of the seal width and assumes a balance chord equal to the overhang chord plus one-half the width of the sealed gap measured when the flap is in the neutral position. The Inaccuracy of this method is apparent by a com;parison of the values shown in figure 38 with the data of figures iLj. to jii.. As indicated in ref- erence 2, the error involved in using the approximate method may be a considerable part of the hinge-m.oment coefficient or control force of the balanced control surface" it is therefore believed that this method should not be used. Application of Seal-Moment Data Inasmuch as the investigation reported herein was made v;lth the various limitations and configurations of an internal-balance arrangement and the data presented are representative of the seal effects in such configura- tions, the figures are believed to be applicable for design purposes In calculating the hinge-mom^ent character- istics of various balanced flaps with thin-plate overhangs. 12 FACA ARR He. L5F50a if the pressure difference across the talance and the plain-flap hinge -i-noment coefficients are known. In order to obtain the seal characteristics of those con- figurations not tested, it is possible to interpolate between the seal-troirjent curves for an intermediate gap- or seal-width conf igurt^tion. As indicsted in reference 2 these results are believed to be applicable also to over- hangs having nose shapes with fairly srall angles, but the liirilting nose angle has not yet been determined. SXAI^LES Comparison of experi^.en t al and computed hi n ge -rcoment a bs.lsnoed flap . - The aileron configura- tions sho^^nn in figure 59 were selected to illustrate the computations necessary- for a practical application of the seal-rrioment data to th^e oaiancing moment of an internally balanced flap. The plain-sealed-flap hinge-mioment data for the 0.20c aileron were used in the computations, together with the nressure coefficient across the balance P-^ . rt (from unpublished data) given in figure 4O, rhe seal- moment data presented herein, and the geom.etric dim_ensions of the balance. In figure 1.0, the computed balanced- aileron hii'ige-moment coefficients are com^pared with the balanced-aileron hinge-moment coefficients obtained e xpe r im.e nt al ly . The computed balancing mioments were obtained hj the following equation, which is based on the geometric dim.ensions of the balance arrangemient for a ^jnlt soecn ( two-dim-ens i onal } ' Ac h. Pr '■ '^ (1 + ,''t/2 - (^ 2I 1 n I Fr •- 2 I = -f L(o.7l+Ji}'-(i + mg) - (0.1G75) J Pr- "5" 0.555a + ^3) - 0.0552 FACA ARR No. L5F50a IJ Values of Ft for use in this equation viere obtained from figure kO. The width of the seal used in the investigation was not measured but was believed to have been arfprozirr.ately 0.[(.. Values of m for computing the exact balancing morr.ent over the deflection range for a seal width of O.lj. and a sealed-gap width of O.O336 v;ere interpolated from the data of figure 22 and fig- ures ik and 1^. (The characteristics of the seals with vertical and circular beckplates are sornev/hat siinilar at this seal width and sealed-gap width.) The balanced-aileron hinge-moment coefficients wrere obtained by the equation ^h "' ^h< "*" ^'^h balanced aileron ^plain aileron " ^ and figure Lo shows fair agreement between experimental and computed results. The discrepancy between the balanced-aileron characteristics obtained experimentally and those computed is probably caused by the difference in the size of the vent gap (0.005c for plain aileron, 0.010c icr balanced aileron) and by possible small dif- ferences in model configuration such as seal width, chord of overhang, and thickness of overhang. Gom.parison of the comouted balanced-f lao hinge- momient coefficients with and wdthout bals'nce-chamber restrictions . - A comparison of the com.puted balanced- flap hinge-rfjomient coefficients, with curves showing the effect of the vertical cham^ber restriction either neg- lected or considered, is presented in figure kl to illustrate the type of balanced-flap hinge -mom.ent coef- ficient that v;ould be obtained if a very wide seal v/ere used in a configuration of an internally balanced aileron. For these com.putations ^ an overhang chord of O.^SCg^, a seal width of 0.6, 5|3 = 21°, and a sealed-gap width of 0.1 (values of mg obtained from figure 22(c)) were assumed installed on the plain sealed aileron of figure 3} a:nd the other data, supplied in figures 39 and kO, were assumed to rem.aln the sam.e. (See fig. kl.) The miethod of com.puta- tlon was the same as that previously outlined. Neglecting the effect of the vertical balance- cham.be r restriction (that is, wdth no limiting value for 6-j_) results in lL\. KACA ARR wo. L^FJOa considerable error at lt.rge deflections, when the seal contacts the balance-chamber ceiling. (See fig. ij.1 . ) The optimum seal width would be such that the seal would barely touch the chamber ceiling at the maximum overhang deflection fsee section entitled "Effect of vertical restriction") and, for this aileron configuration, fig- ure 22 indicates that a seal vidth of about O.5 would be optimiar.1. Up to deflections of il6°, the curve for aileron-hinge-m.oment coefficients computed for s = O.5 and 5p = 21*^ is approximately the same 'as that shown in figure III for s = 0.6 and no limiting value for 5-„; for deflections beyond tlS"^, this cu-Pve is almost parallel to the curve shO'.'.Ti in figure L.l for s = 0.6 and 5^ = 21°. The error involved in the "os e of the approximate method is also illustrated ixi figure kl; the approximiate m.ethod indicates that the seal contributes more balancing moment thazi that actually produced over alraost the entire deflection ranee. COKOITSIONS A.S an experim.ental verification and an extension of a previous analytical investigation, tests -^^^ere made to deterrrd.ne the hinge momients produced in an internal- bs.lance arrangement by fabric seals of various widths that seal flap-nose gaps of various widths in the presence of a thin-plate overhang. The tests were con- ducted v;ith horizontal, vertical, and circular types of balance-chamber wall forward of the balance and \s/ith various heights of the balance chamber. The investiga- tion indicated the following conclusions s o 1. The moment of the seal m.ay be a balancing or an unbalancing m'Omant and may be an appreciable part of the total balancing moment of an internally balanced flap, depending on the overhang deflection and the configura- tion of the internal balance. 2. Variation of the width of the seal, the sealed gap, or the location of the seal attachment to the wing structure affected the seal mioments through most of the overhang deflection range. NACA ARR No, L5F.30a T C 5. The shape and size of the balance chamber affected the seal-r.iomenb characteristics in the deflection range at which the seals contacted and were constrained by the chamber walls; the values of the seal moraents were usually reduced when the seals were constrained. k. An optimum balance configuration empD.oys a seal width such that the seal barely touches the chamber ceiling when maximum overhang deflection is attained. Langley Memorial Aeronautical Laboratory National Advisory Committee for Aeronautics Langley Field, Vu. REPLRSNGJilS 1. Rogallo, P. M., and Lowry, John G,: Resume of Data for Internally Balanced Ailerons. NACA RB, March 1943. 2. Murray, Harry E,, and Ji^rwin, i-'ary A,; Hinge i.Ioments of oealed-Internal-Balaiice jT.rranj|-era.ents for Control Surfaces, I - Theoretical Investigatioxi. MCA ■' ARR iJo. LSP30, 19^5. NACA ARR No. L5F30a 16 TABLE I.- BALANCE CONFIGURATIONS TESTED Location of seal attachment (fraction of c^j) Width of sealed gap (fraction of c^) width of seal (fraction of c^) (deg) Fl gure Vertical backplate ( "■h None, 15 li4.(a) None, 16, 21, 26 ll|(b) .1 .i None, 16 15(a) .1 None, 16, 21, 26 15(b) .2 , 5 None, 16 16(a) C9nt«r ; .2 \ .2 . 6 None, 16, 21 16(b) of .8 None, 16, 21, 26, 30 16(c) haokplate .3 .6 None, 16 17(a) ■\ .8 None, 16, 21, 26 17(b) 16(a) .7 None, 16, 21 ■ h .9 None, 16, ZZ, 26, 30 18(b) •5 .7 None , 16 19(a) ^ -5 • 9 None, 16, 21, 26 19(b) O.3I4. above and c below center J 'I .8 20 20 of backplate \ -5 • 9 20, 25 21 c Iroular backplate f 0.1 O.i; None, 16 22(a) .1 . 5 None, 16, 21 22(b) .1 . 6 None, 16, 21, 2S 22(c) Center .3 .6 None, 16, 20 23(a) of ^ -5 •7 None, 16, 21, 26 23(b) backplate .3 .8 None, 16, 21, 25 23(c) 2l;(a) .5 ,7 None, 16 •5 .8 None, 16, 21 2lf(b) ^ .5 .9 None, 16, 21, 25 2U(o) O.5I1. above and f _ .8 below center / 'I 20 25 of backplate I -5- .9 20, 25 26 Ho rlzontal backplate C °:i None 27(a) None, 16 27(b) .8 None, 15, 20, 25 27(c) 28(a) .1 .u None .1 .5 None, 16 28(b) .1 .7 None, 15, 20, 25 2B(c) .1 .9 None, 25, 30 28(d) .2 .5 None, 16 29(a) Center .2 .7 Fine, 15, 20, 25 29(b) of C .2 • 9 None, 26, 31 29(c) backplate .3 .6 None, 16 30(a) .3 .7 None, 15, 20 30(b) •t •9 None, 25, 31 30(c) .6 None, 16 31(a) •^ •7 None, 15, 20 31(b) .1+ .9 None, 16, 21, 25 31(c) .5 • 7 None, 15 32(a) ^ -5 •9 None, 16, 20, 26 52(b) O.3I4. above and below center of backplate ( -^ • 7 .9 20 20,26 15 NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS NACA ARR No. L5F30a Fig. la, b (a) Plan form of a semis pan wing. Backplafe 'Co^er plafe 'Flexible seal 'Overhang (h) Section A-A. NATIONAL ADVISORY COHHITTEE FOR AERONAUTICS Figure I . - Schematic diagram of a typical internal-balance arrangement for an aileron on a tapered yving. Fig. 2a NACA ARR No. L5F30a I/I u > K o S -.8 < "- 2[ o 4) 3 I «3 J) ^ I I ■^15 "§■ NACA ARR No. L5F30a Fie. 2b /-//nge - moment equ/pmen / ^/cromanorneter Figure 2(b).- Three-quarter rear view of testing apparatus. NACA ARR No. L5F30a Fig. 5a-d o o tJ ol •-{ Ck M o cd ^ o C4 X3 t<3 3 Ol Ol a) o ^H o a) tH O) cn II J3 '. OO to , 4J m 3 4J O) ^1 od A A a) 4-3 m (D 4-> Od rH u CO Cd "^ U 0} m f> a) CQ to ft ft • cd • b .. ID ID O 4J 4J 4J O m cd C "-I ft o n .-t o 3 ^ ol Hi O CO • ID •H CJ o «J ■* a> ^ w u ^ IB • P > o to bD a 4J • «-l o s ,» o ID (M m • p O II 4-> Cd II __ P. ^ Q) m CO ft ft Cd • • a. o> o 4-> 4J o m a) C»J a) Id CO OJ X3 cn i-H ,» 1 cd vH • o • in 4J o a> t^ H (-. 60 60 ^ 43 4J O • ^^ o) .» o s 4-> CJ) in *J • CD CO at o P u 4J r-H n OJ a) -^ ^ ID n £> od CQ CO ft ft • cd • ^ • » .-1 -H o ID 43 m cd > CO -* 0) t^ to 11 (h ID • P > o to bO NACA ARR No. L5F30a Fig. 9a-d' o o o 00 J3 CO o a 60 O a m Q. ^. o • » ■* m ai o ;c o Si x: 11 • o. Ck m o a) e »-i x> ^. •H f- o x> u w t- aj o o, 3 o • o ^ CJ C 0} ;« m (-. T3 ,-1 tn Ck oS o (-> *-H OS c: a) a O V *H x> u e-H tS a fi u f-l « o 4J ■ ■ ■1 «H I E pi » 1 ^K t • o 00 o 0] at . m Ck a) 4-) m 4J a) Ol i-H a >{ o a X3 a . oo C4 o »H 01 ^ o II ^ « ^ ■»J oo •rH 3 to 3 *J a) t^ ad • O. o n o ol ^ 4-> H OO .-H ,^-s at a) in o P fH • o 4J n) u T-i a) £> O, * o oo as • »-J i4 43 o i-H cd J3 n Ol X3 O 0) •* bO CO r-l to • Cd • o 1 4J o .. lO • (=! a> fj o II ~ U O a) ^ P ^ 4-J a> ^i CO 4J u cd P! •Ol IB ^ O. • O «3 a) (D JJ a> »j Od > ,» to rH o ID ID 13. J3 • 4J hi Od O .-1 od fl II oS ^ o -* 00 CO .-H to • od • o 1 N N o •o • *-t tsi (D o u •»H Xi