ARR No. L5C08 "r NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WARTIME REPORT ORIGINALLY ISSUED March ISWi) as Advance Restricted Report L5C08 ERELBCOTARY FLIGHT RESEARCH On M ALL-MOVABLE HORIZONTAL TAIL AS A LOHGITDDIITAL COKTROL FOR FLIGHT AT HIGH MACH NUMBERS By Harold F. KLeckner Langley Memorial Aeronautical Laboratory Langley Field, Va. 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. 89 DOCUMENTS DEPAPTMcNT Digitized by tlie 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/preliminaryflighOOIang MAC A AR^ No. LSCOS NATIONAL ADVISORY COmiTTEE FOR AERONAUTICS ADVANCE RESTRICTED REPORT PRELIMINARY FLIGHr RESEARCH ON AN ALL-MOVABLE HORIZONTAL TAIL AS A LONGITUDINAL CONTROL FOR PLIGHT AT HIGH WACH NWiBERS By Harold F. Kleckner SUfJMARY The NACA is conducting flight tests of an all- movable horizontal tail installed on a Curtiss XP-I(-2 air- plane because of its oossible advantages as a longitudinal control for flight at high Mach numbers. The results are uresented for some iDreliniinary tests in the low-speed range for which the tall was very closely balanced aero- dynamically and a bobweight was used to obtain stable stick-force variations with speed and acceleration. For these tests, the tail was hinged at 0.2l\. chord and was tried with two arrangements of servotab control. The elevator control was found to be ijnsatisfactory with the control arrangements tested. Although there were sufficient variation of stick force with accelera- tion in steady turns and a stable stick-force variation with speed, the near-zero variation of stick force with stick deflection resulted in an extremely sensitive control that required continuous attention in order to avoid motions of the airplane due to inadvertent move- ments of the control stick. For subsequent tests, tne servotabs are being connected as geared unbalancing tabs in order that more conventional elevator hinge- moment characteristics may be obtained. The expected advantages of the all-movable tail with a control system utilizing tabs would of covirse be limited to flight at Mach numbers below those for v»hich severe compressibility effects are encountered on the tail itself. For higher Mach numbers, the all-movable tail would require an irreversible power-boost control in order to handle the large hinge-moment increases that are exr^ected. KAGA ARR No. L5C08 INTRODUCTION The possible advantages of an all -movable horizontal tall for longitudinal control have been considered for somo time. This type of tail recently has been suggested for flight at high Mach numbers since it offers a means of eliminating the pitching oioments that result from the downwash changes which accomoany compressibility effects on the wing when conventional fixed stabilizers are used, A3 a consequence, an all-movable horizontal tail was designed for the Gurtiss XP-l|.2 airplane (a modified P-56 airplane), and flight tests were made of the air- plane with this tall installed. The XP-I4-2 airolane can- nob be f].own at Mach numbers at which control difficulties ordinarily arise; however, srieeds and accelerations that needed to be covered in a preliminary research iDrogram could be obtained. With the all-movable horizontal tail installed on the XP-'42 airplane, a series of ground handling tests and two flights have been made. In the two flights, the maneuvers were limited to low-acceleration turns and the speed was limited to 200 miles per hour. The present report su^Tiinarizes the data obtained. TAIL GHARACTERI3TICS The all-movable horizontal tail that was designed for the XP-I|.2 airplane incorporates three distinguishing features : (1) An all-movable tail plane (elevator) hinged at its aerodynamic center (2) Servotab control (3) A bobweight in the control syste m A tail arrangement incorporating these features offers several advantages. '.Vhen the tail plane is hinged at its aerodynamic center, the variations of hinge-moment coefficient with elevator angle C]^= and with angle of attack of the tail C^ ^^^ very near zero. The stick force required to produce a speed or acceleration cbange then depends only on the bobweight I7ACA A HP. No. L5C08 5 e'^fect; is independent of the tail load and the flow direction at the tail; and is therefore inde-oe^nder^t of airplane center-of-grav" tv position, power and flap effects (except for changes in dynamic pressure at the tail), alti tilde, and downwash chai-iges that accompany Mach number effects on the wing. In addition, an all- ^novable horizontal tail provides a greater down load in the landing attitude than a conventional stabilizer and elevator. This characteristic penrits an airplane to meet landing requirements for a greater center-of-gravity range or, for the same range, pennits a reduction in tail area if more-forward center-of-gravity positions are used. It was realized that compromises would probably be required as the tests progressed. From the beginning it was apparent that longitudinal oscillations might exist and that the pilot m.ight move the tail to a position at which the tail load might be excessive before the airplane acceleration, and therefore the force from the bobweight, vi/ould be experienced. Pertinent characteristics of the tail and its installation are shown in figures 1 to 5 • The tail area was not reduced in comnarison with the original tail area because moving the center of gravity of the airplane forward was not feasible. The aspect ratio was Increased in com.parison with the original tail to compensate for the shorter- tail length that was required for installation purposes. Ihe elevator was mass-balanced about its hinge line, and the servotabs were m.ass-overbalanced about their hinge lines to give d^^mamic balance for rotation of the elevator. The bobweig^t in the control system, which was located "J .0 feet behind the center of gravity, gave a ^orce of 6 pounds at the pilot's grip, T'J'io arrangem.ents of the servotab control system have heen used to date. Schematic drawings of these systems are given in figure [(.. In arrangement 1, a spring was incorporated on the servotabs to keep them from banging against their stops in ground handling tests. The spring did not alter essentially the condi- tion of Ciy - and Ch^ - for this arrangement when the aerodynamic center was at the hinge line. After a fevif tests, the control system -was changed to arrangement 2. In this arrangement, the point at v/hich NAG A ARK No. L5CO0 the additional link was attached was so chosen that the tabs would deflect in tho same direction as the elevator (unbalancing) in a ratio of tab angle to elevator angle 5t — = 1 (no soring deflection). This arrangement made 6e Che- negative but Thermit ted G^c: to approach zero as the speed increased, since the action of the spring 64- reduces ^— with increase in speed. This arrangement did not alter Gh and retained the servotab action of a-"rangeraent 1. In arrangement 2 the trim tabs v;ere locked vn the neutral oosition; trim changes were made by changing the servotab position. Arrangement 2, with spring added, is the geared unbalancing tab arrangement used to obtain a stable variation of rudder force with rudder deflection on two all-raovable vertical tails previously tested at the Langley Laboratory (references 1 and 2 ) . TESTS AND INSULTS With the elevator control system connected as in figure k-isi), the control felt Tincertain to the pilot in taxi runs and ground flights (take-off, flight along runway, and landing). In an attempt to isolate the trouble, the servotabs were locked and taxi runs with the tail down were made at about [|.5 miles per hour with the elevator moved slowly through its deflection range. The data of figure 5(a) were obtained in a run of this type. The slone of the curves in figure 5(a) indicates that the aerodynamic center of the tail was between 3 and U- percent of the mean aerodynamic chord ahead of the hinge I'ne, an indication that the tail v>/as overbalanced and that Oh„ was nositive and not zero. The breaks in the curves at large down elevator deflections are the result of stalling of the tail. Since wind-tunnel tests have shown that strios on the trailing edge of an airfoil move the aerodynamic center rearward, this convenient method was used to bring the aerodimamic center to the elevator hinge line. Strips of different sizes v/ere tried until the aerodynamic center was moved back to the hinge line (fig. 5(t>)) by 0.28-lnch strips attached outboard of the servotabs. No strips were attached to the trailing edges of the servotabs because the strips NACA ARR No. L5C08 5 would make the variation of servotab hinge-moment coef- ficient with angle of attack of the tail negative; this effect is similar to moving the aerodynamic center of the elevator forward. With Ch and Ch^- zero, the control still felt •uncertain to the pilot and was unsatisfactory in ground flights. The uncertain feel of the controls probably resulted from the absence of stick forces associated with stick movement. The stick forces from accelerations (due to the bobweight) did not give significant feel to the pilot for these ground flights because the normal accelerations v;ere small and lagged behind the stick movements too much at these low speeds. It was con- cluded that, in take-offs and landings in which rather large and rapid movements of the control stick are made, variation of stick force with stick deflection must be provided in order to give the control the feel necessary for the pilot to fly the airplane with assurance. The control system was therefore changed to arrangement 2; and, after satisfactory ground tests, two flights were made viJith this arrangement. For these flights, the air- plane weight was 61OO pounds and the center of gravity was at 28.1 percent of the mean aerodynamic chord with wheels up. The longitudinal characteristics of the air- plane were recorded in abrupt pull-ups, steady turns, and steady flight through the speed range. Records of stick-free oscillations were also obtained. The data of these flights are given in figures 6 to 3 and indicate that (1) 'The airplane exhibited stick-free and stick- fixed static longitxidinal stability (2) The airplane would trim throughout the speed range tested (5) There v;as a stick-force gradient in steady turns of about 8 po\mds per g (1|) Stick-free oscillations at II5 and 157 miles per hour damped satisfactorily Despite these satisfactory characteristics, the pilot considered the control sensitive and uncertain and therefore unsatisfactory. Continuous attention to the control was necessary in rough air in order to avoid motions of the airplane due to inadvertent move- ments of the control stick. In addition, the control NAG A ARH No. L5GO8 was considered sensitive because In abrupt maneuvers the reactions of the airplane were not proportio::al to the forces exerted by the pilot, A comparison is made in figure 9 of data obtained during abrupt null-ups made with the XP-I|-2 airplane and a Gurtiss P-i|0 airr^lane, for which the stick-force gradient in steady turns was also about 8 oounds ner g. Itie curves of figure Q show that the force required for the initial deflection of the all-movable horizontal tail was about 5 0^ 10 percent of the force required for the deflection of the P-.^O elevator for approximately the same resulting normal acceleration. Ihis com.parison Indicates that insufficient variation of stick force with stick deflection is the reason for the pilot's dissatisfaction with the control. It is interesting to note here that, since these tests were made, the NAGA pilots have expressed dis- satisfaction with closely balanced elevators of con- ventional design on other airplanes because of the sensitivity of the control. In this connection, a theoretical investigation (reference J) has been made of the effect of various hinge-moment -oarameters on elevator stick forces in rapid maneuvers. SUBSE^'JimiT 'T3STS ?or subsequent flight tests, the control system is being modified to incorporate preload in the servotab s-oring, With this arrangement conventional stick- force characteristics can be obtained, but the advantages that accrue from Che = of course cannot be realized. 'Tie -oossibility still exists of making the all-movable tail a satisfactorv longitudinal control with Ght - 0« In this regard, provision has been made to incor-oorate a damner on the servotab spring. The dam-oer would in effect eliminate the spring action from abrupt control deflections. Preload in the servotab spring will not destroy the expected advantage of the all-movable tail for flight at high Mach numbers, because the servotab will come into action when the stick force is sufficient to overcome the preload in the spring. NAG A ARH :To . L5G08 The use of an all-movable taf.l that defends on tab action Is necessarily limited to flight at Mach numbers below those for which severe compressibility effects are encountered on the tail itself. For higher Mach numbers, the all-movable tail will require an irreversible power- boost control in order to handle the large hinge-moment increases that are exnected. It ap'oears that the devel- onment of such a power-boost control warrants consideration . CONCLUDING REMARKS In preliminary flight tests of an XP-J-i-2 airplane with an all-movable horizontal tail chat incorporated very close aerodynamic balance and a bobweight, the elevator control was found to be unsatisfactory. There were sufficient variation of stick force with accelera- tion in steady turns and a stable stick-force variation with speed, but the control was sensitive and required continuous attention in order to avoid motions of the airplane due to inadvertent movements of the control stick. The vmsatisfactory qualities of the control were attributed to insiif f icient variation of stick force with stick deflection, which resulted from the very close aerodynamic balance. Langley Memorial Aeronautical Laboratory National Advisory Conmittee for Aeronautics Langley Field, Va. 8 NACA AR.R No. L5CO0 REPER^TCES Jones, Robert T. , and Klecloier, and Preliminary Plight Tests Vertical Tail Surface. NACA Flarold P. : Theory of an All-Movable ARR, Jan. 19i4-5. 2. Kleckner, Harold P.: Plight Vertical Tail on the Pairchild NACA ACR No. 3P26, 191+5. Tests of an XR2K-1 All-I.Iovable Airplane . 3. Jones, Robert T., and Greenberg, Harry: Effect of Hinge -Iv'oment Parameters on Elevator Stick Porces in Rapid Maneuvers. NACA ARR No. Li+J12, I9IJJ4.. NACA ARR NO. L5C08 Fig. > o B I 03 C «3 I— I D. U • (d 3 O -H o c o S N 0) -H > o u CO 0) u u 0) ♦J (d 3 cr I 0) 0) Li x: Li bo NACA ARR NO. L5C08 Fig. 2 CO CO u a o «3 ♦J CO c d C 0) O G — t .-( »-■ o. o u (U ■H CM > a, o X E I o a. u O < f I a, X x: -^ ^ x: S H ■a X W O 3 J3 CO • E 3 0) o u »s S O o o x: u ID CO E 0) x; w (1) 3 bo NACA ARR NO. L5C08 Fi?. 5 (a) Original trailing edge, no strips, 10 :a. -^10 p, — — o- ■r^ -o ^ ^ -Q D-^ -^ 1 / 7 C^^"'-' d COJ NAIIUNi MinEE IL AUViaUKY •OR AERONAUTICS 8 Dotvn 4 Elevator 4 angle 8 12 , deg Up (b) Trailing edge equipped with 0.28-lnch strips outboard of servotabs . Plgiare ^.- Variation of stick force with elevator angle for steady elevator movements In taxi runs, three-point attitude, Curtlss XP-I4.2 airplane with all-movable horizontal tall. Readings taken every ^ second; arrows Indicate direction of elevator motion. Figs. 6,7 NACA ARR No. L5C08 20 ^ 10 15 0/234 Norma/ acce/eraf/on, g Figure 6.- Variation of stick force with normal acceleration in steady turns at an altitude of 5OOO feet. Curtlss XP-lj.2 airplane with all-movable horizontal tail; 6-po\ind bobwei^t in control system. ^ ^ ^ Speed (mph ) /SO Q 200 / Q.5 |2 /Power i in. Hg j ( rpm ) o 25 £200 Q /3 /SOD , fOO /20 /¥0 160 Indicated airspeed, mph Figure 7'- Variation of stick force and elevator angle with Indicated airspeed. Curtlss XP-i|.2 airplane with all-movable horizontal tail. NACA ARR NO. L5C08 Figs. 8,9 I 0/20/2 T/me f ^ec Time j sec (a) Speed, 115 miles per hour, (b) Speed, 157 miles per hour, Flgxire 8.- Records of longitudinal oscillations made by abruptly moving and then releasing the stick. Curtlss XP-1|2 airplane with all-movable horizontal tail; no stick force available. y^ — ~ r ^ \ Servofab E/ei/a/or ,'~^^. A . /^ > ^^ - 80 h$^^^5 V%io ^1" u2 5: -^ / Time J / sec NATIONAL ADVISORY efiMMITTEE FOR AEfiONAUTICS . r- orce 1 / \ C le vaf-or C. / 4 1 sl Accelerafion 1 1 11 jl 1 V A jl A / \ fl ' 1 ' \ jl jl ' u A ^->> V V T/me / , sec (a) P-I4.0 airplane; speed, (b) X?-U2 airplane; speed, 188 miles per hour. 205 miles per hour. Figure 9«- Comparison of data obtained during abrupt pull-ups of Curtlss XP-q.2 airplane with all-movable horizontal tail and of Cta:*ti3a P-I4.O airplane. I » UNIVERSfTY OF FLORIDA 3 1262 07749 250 1 UNIVERSITY OF FLORIDA DOCUMENTS DEPARTMENT 1 20 MARSTON SCIENCE UBRAFIY P.O. BOX 117011 GAINESVILLE, FL 32611-7011 USA