NASA/TM--1999-208901 AIAA-99-0370

NASA/FAA Tailplane Icing Program Overview

Thomas P. Ratvasky Lewis Research Center, Cleveland, Ohio

Judith Foss Van Zante Dynacs Engineering Co., Inc., Brook Park, Ohio

James T. Riley FAA Technical Center, Atlantic City Airport, New Jersey

Prepared for the 37th Aerospace Sciences Meeting & Exhibit sponsored by the American Institute of Aeronautics and Astronautics Reno, Nevada, January 11-14, 1999

National Aeronautics and Space Administration

Lewis Research Center

lanuarv 1999 Acknowledgments

The authors would like to thank the technical staffs from the NASA Icing Research Tunnel, OSU Low Speed Wind Tunnel, and the NASA Twin Otter. Special recognition goes to Mr. Richard Ranaudo for both his superb skills as a test pilot and his keen research insight, and to Dr. Dale Hiltn_r for all of his effort in developing TAILSIM. We thank Mr. John P. Dow, Sr., of the FAA Small Directorate, for his continued support and active promotion of the TIP. We also would like to express our appreciation to our sponsors, NASA's Aviation Operations Systems Base program and the FAA Technical Center.

Trade names or manufacturers' names ire used in this report for identification only. This usage does Pot constitute an official endorsement, either expressed or irr plied, by the National Aeronautics and Space Ad ninistration.

Available from

NASA Center for Aerospace Information National Technical Information Service 800 Elkridge Landing Road 5287 Port Royal Road Linthicum Heights, MD 21090-2934 Springfield, VA 22100 Price Code: A03 Price Code: A03 NASA/FAA TAILPLANE ICING PROGRAM OVERVIEW

Thomas P. Ratvasky Aerospace Engineer NASA Lewis Research Center Cleveland, OH 44135

Judith Foss Van Zante Member, AIAA Senior Engineer Dynacs Engineering Co., Inc. Brook Park. OH 44142

James T. Riley FAA Technical Center Atlantic City Airport, NJ 08405

Abstract The effects of tailplane icing were investigated in a four-year NASA/FAA Tailplane Icing Program (TIP). This research program was developed to improve the understanding of iced tailplane aeroperformance and aerodynamics, and to develop design and training aides to help reduce the number of incidents and accidents caused by tailplane icing. To do this, the TIP was constructed with elements that included icing wind tunnel testing, dry-air aerodynamic wind tunnel testing, tests, and analytical ccx:le development. This paper provides an overview of the entire program demonstrating the interconnectivity of the program elements and reports on current accomplishments.

List of Symbols and Abbreviations Introduction Ice impedes the productivity and sali_ utililization of all S&C stability and control aircraft. As a result, substantial efforts have been under- Ca drag coefficient taken to reduce the safety risks associated with aircraft C_ lift coefficient icing. Although considerable progress has been made in Cm pitching-moment coefficient icing research and engineering, some aircraft designs Cr_ hinge-moment coefficient are still susceptible to certain ice-related problems, one CT thrust coefficient being ice contaminated tailplane stall (ICTS). G acceleration due to gravity Tailplane stall due to icing is not a new problem. cg aircraft center of gravity Aircraft incidents and accidents have occurred V_ extension speed sporadically since the late 1950's. At that time, the Vs stall speed cause of these incidents and accidents was unknown. But now, it is clear that these events were related Greek: through loss of pitch stability and control probably due angle-of-attack, deg to ice on the horizontal tail. Aircraft accident analyses

Oqail tail angle-of-attack deg have revealed ice contamination on horizontal tailplanes

13.a/c aircraft angle-of-attack, deg as the primary cause of over 16 accidents resulting in 139 fatalities. _ 13 angle-of-sideslip, deg _5c,dE elevator deflection, deg The ICTS events usually occurred on final ¢3F.dF flap deflection angle, deg approach, when flaps were extended. Ice on the hori- zontal tail caused premature flow separation. The separated flow could not attain pressure recovery over the elevator and resulted in stick force reversal (control column pulled away from the pilot). The aircraft would Ct_pyrighl _;' It19_ h_, the AIAA. Inc. N_ c_pyrighl is as_rted in the trniled Stalt.s under Title 17. U.S. Code. The2 [[.S. (io_ernmenl has a itL',.alty-Ii ct' lit2ell,_, ltt CXelCi_a.' an i-i_2111s pitch nose down and rapidly lose altitude (Figure I). under Ihc copyright tiara'ted herein [_tr Ch_',crnnT,..'nt purp_'s. All oIbcr rights are rc_'rvcd by the copyright owner

American Institute of Aeronautics and Astronautics Generally,theaircraftcouldbe recoveredonlyby requestbydevelopingtheNASA/FAATailplaneIcing retractingtheflapsandby thesheerstrengthof the Prograni(TIP).TheTIPwasco-sponsoredbyNASA pilotspullingbackonthecontrolcolumn. LeRCand the FAA TechnicalCenter through an Althoughtailplanestallduetoicingcanoccuron Interagency Agreement. anyclassof airplane,theproblemhashadthehighest The purpose of this paper is to provide an overview rateof occurrenceonthecommuterandlighttransport of the entire Taiiplane Icing Program and illustrate the .Variousreasonsareofferedto explainthis. interconnectivity of the program elements. It also (1) Commutersoperatetbr greaterperiodsof time summarizes the current results from each element. withinpotentialice zonealtitudesthando thelarge transportsandthereforehavea greaterlikelihoodof NASA/FAA Tailplane Icing Program History encounteringicing. (2) Ice protectionsystemson commutersaretypicallyde-icers,whichmayleadto In early 1994, representatives from NASA LeRC, the reducedairfoilperformanceduetoresidualiceandice FAA Technical Center, FAA Certification Service, and buildupbetweendeicingcycles. the Ohio State University met to discuss potential Previousresearcheffortsto understandtailplane program activities described in a NASA-developed icingwereconductedby a Swedish-Sovietworking work plan 6. The primary results from these planning groupduringthe1970'sto 1980's234.Thefirstreport meetings were: describedan experimentalstudyof icing on the I. the confirmation of the NASA/FAA Tailplane Icing aerodynamicsof high-lift, swept-wingsections. Program with funding from both NASA LeRC and Experimentalmethodsweredevelopedto estimate the FAA Technical Center, "critical"iceshapes,tosimulateicingconditions,andto fabricateice"imitators".Thesecondandthirdreports 2. the establishment of a cooperative agreement focusedontailplaneicinganditseffectsonlongitudinal between NASA and the Ohio State University, and stabilityandcontrol.Windtunnelandflighttestswere 3. concurrence between the various parties on the conductedtostudythetaitplanestallphenomenon.This tasks to be accomplished in the program. researcheffortprovidedexcellentinsightsintosomeof theaspectsoficecontaminatedtailplanestall. In _)rder to expand the understanding of iced TheicingresearchprogramatNASALeRCalso tailplane aeroperformance, icing wind tunnel testing and studiedthestabilityandcontrolchangesdueto tailice aerodym_mic (dry air) wind tunnel testing of a tailplane ona DHC-6TwinOtteraircraft5.Resultsshowedthe model were determined necessary. Flight testing was longitudinalstabilitydecreasedsignificantlywith also nect:ssary to verify wind tunnel results and evaluate artificialiceonthehorizontaltail,andthatthestability wasfurtherdecreasedwiththeflapsdeflectedto 10°. the maneuvers of interest to the FAA and other parties. Highthrustcoefficientandlowaircraft-angle-of-attackFinally, it was desired to have an analytical method for were also significantcontributorsto the reduced discriminating tailptane sensitivity that would use stability.Inaddition,elevatorcontroleffectivenesswas results from the wind tunnel and flight tests for significantlyreducedwiththeartificialice. developilent and evaluation. These ideas formed the To promoteawarenessof the tailplaneicing foundati_ m of the program. problem,theFederalAviationAdministration(FAA) The TIP became a four-year research program that sponsoredthree InternationalTailplane Icing utilized _:combination of icing experts and test facilities Workshopsin November1991,April 1993,and that included NASA Lewis' Icing Research Tunnel September1994.Theseworkshopsgeneratedapproxi- (IRT), The Ohio State University (OSU) Low Speed mately30recommendationsaddressingissueson the icingenvironment,aerodynamics,ice detectionand Wind Tunnel (LWST), and NASA Lewis' DeHavilland protectionsystems,flightoperations,andmaintenance.DHC-6 l"win Otter Icing Research Aircraft. These In responseto someof theserecommendations,the rcsource_ were used to accomplish the following FAA requestedthatNASAconductresearchintothe program goals: characteristicsof ICTSandto developtechniquesand I. lmpJ ove understanding of iced tailplane methodologiesto minimizethe hazard.A specific aeroi)erformance and aircraft aerodynamics, requestwasmadeto improvetheunderstandingof the 2. Develop analytical tools to help assess tailplane dynamicandaerodynamiccharacteristicsof theso- sensitivity to icing, and calledpushoverandsideslipflightmaneuvers,andto 3. Develop training aides to expand the awareness of developa bodyof knowledgeandtheorybehindthe the ICTS aviation hazard criticaldegradationoflongitudinalstabilityandcontrol causedbytailplaneicing.NASArespondedtotheFAA

2 American Institute of Aeronautics and Astronautics Model Selection labeled S&C Ice (Figure 4). The shape was determined The first step in the program was to select an using a method described in Reference 7, and the icing appropriate model to test. A DHC-6 Twin Otter tlight conditions and photographs from one of NASA LeRC's 1984 icing . The second shape tested was tailplane was chosen as the subject of the study because: predicted by LEWICE version 1.3 (Figure 5). The • The DHC-6 had a known susceptibility to ICTS conditions used to derive the LEWICE shape were: • NASA-LeRC owned a DHC-6with a full • V= 120 knots, _=0 ° compliment of flight dynamic and cloud physics • LWC=0.5g/m 3, MVD=201am instrumentation • T0=-4 ° C, time=45 minutes • NASA LeRC had 12 years of experience with the Both the S&C and LEWlCE shapes were fabricated DHC-6 flight characteristics in icing, reflected in for wind tunnel testing using Ren Shape ® and were extensive documentation and a databank of strictly 2D. After machining, the fabricated shapes were performance, stability and control derivatives with sanded smooth and painted. No attempt was made to and without icing on the aircraft vary the shape in span or simulate ice roughness with grit material. Since wind tunnel testing was to be conducted at the NASA IRT that has a 6'x9' test section and the Two other ice shapes were selected for aero- OSU LSWT that has a 7'x10' test section, it was opted perlormance testing. These shapes were the result of the to obtain two models, one tbr each tunnel. The IRT test. (More details on the IRT test are found in the characteristics of each model are described in the next section). One shape represented ice accretion following paragraphs. remaining on the tail between the pneumatic de-ice boot The model used in the IRT was made from an operation - Inter-cycle Ice (Figure 6). Another shape actual Twin Otter tailplane (Figure 2). The flight represented the ice accretion during a failed de-ice boot hardware was cut to provide a 6-foot span, 2D model condition - Failed Boot Ice (Figure 7). The icing for IRT testing. A new BFGoodrich pneumatic boot conditions used to form these ice shapes were the same with standard coverage for a Twin Otter was installed with the exception of the time. These conditions were: on the to determine inter-cycle ice accretions. Load cells were incorporated with the • V=135 knots, alpha=-2.9 ° elevator hinge brackets to be able to measure the • LWC=0.5g/m 3, MVD=20_tm elevator hinge moment throughout the testing. • T0=-4 ° C, The model used in the OSU LSWT was made from • time=22 minutes for Failed Boot Ice a material called Ren Shape ® 450 (machineable plastic). • time=15 minutes, with boot cycle every 3 minutes The 2D model was made to be full-scale 4.75-foot up to 12 minutes for Inter-cycle Ice , and 7-foot span (Figure 3). The model was designed to replicate the two-section geometry of the After the ice accretion was formed on the model, Twin Otter and had an elevator that could be set to molds were made of approximately a 15-inch span of discrete angles for testing elevator effects. The model the ice on the model. New methods were employed to incorporated approximately 90 chordwise pressure taps extract multiple polyurethane castings from the molds to in a mid-span location for acquiring data on the have full-span ice accretion in the aeroperformance aeroperlormance. A pressure belt, similar to the one wind tunnel test and flight test (described in later used in flight, was also used to correlate with the sections). pressure tap data. Likewise, a 5-hole probe, similar to those used in the flight tests was mounted for Icing Wind Tunnel Test comparison purposes. The icing wind tunnel test was conducted in February 1996 at the NASA LeRC Icing Research Tunnel. The Ice Shape Selection IRT is a closed-loop, refrigerated wind tunnel (Figure The second step was to develop ice shapes for testing. 8). The temperature is controllable from 40°F to -40°F. Developing accurate ice shapes for this program was A water spray system has been calibrated to simulate very important, so an IRT entry was scheduled. Due to icing clouds with droplet MVD of 14-40 _tm, and liquid scheduler constraints at the IRT, testing was not water content (LWC) of 0.2 to 3.4 g/m _. The test possible in the early phasc of the program, so it was decided to perlbrm initial aeroperformance tests at OSU section is 6-feet high by 9-feet wide, and a 5,0(0) hp on two other ice shapes. The first ice shape was used in motor drives a fan to provide test section speeds up to previous stability and control projects and was therefore 400 mph (empty test section). A turntable in the tloor

3 American Institute of Aeronautics and Astronautics providesangleof attackchangesto modelsmounted having :he most degradation in C_,_, and reduction in vertically. stalling . The reductions in Cim_xand stalling TheTwin Ottertailplanemodelwasmounted angle of attack for each shape are listed in Table I. verticallyandattachedtotheIRTforce-balancesystem ontheturntableandtunnelceiling.Approximately30 Table t Reductions in Clm_ & o_t_H testpointsweremadevaryingicingcloudconditions Ice Shape A Ci.x (%) A a_t_l (°) andtemperatures,deicerboot-cycletime,spraytime, Inter-cycle 30 2.3 airspeed,angleof attackandelevatordeflectionto Failed Boot 41 7.3 simulatecruise,holding,andapproachphasesofflight. S&C 50 9.5 Altereachicingcondition,themodelwasrotatedfrom LEWICE 50 9.5 +4° to -22° AOA while recording force-balance and hinge moment data with the ice accretion. Afterwards, Similarly, there was an increase in drag coefficient ice tracings and photo documentation were made of the with each ice shape. The LEWICE shape had the greatest ice shape. At the end of each night, a mold was made of increase in drag overall. The coefficient the final ice accretion. The primary results from the IRT followed a similar trend to the lift coefficient. All shapes entry were the ice castings described in the previous caused an early departure to a positive pitching moment, section and viewed in Figure 6 and Figure 7. These but the LEWICE and S&C shapes caused this to happen castings were utilized in both the aeropertormance wind at the least negative alpha. The hinge moment coefficient tunnel and flight test studies. had sortie unusual characteristics with two of the ice shapes. With the S&C and LEWICE shapes, the CH¢ Aeroperlbrmance Wind Tunnel Tests behavior was as expected. Premature separation caused a Initial aeroperformance testing was conducted during rapid increase in the CH_, which is typically attributed to September 1994 at the OSU 7"x10" LSWT (Figure 9). the " snatch" experienced by pilots during a tailplane The LSWT is a closed loop tunnel with a 2,000 hp stall. But for both ice castings, the CH_ was initially electric motor to provide dynamic pressures up to 65 reduced below the baseline and then followed the psf in the 7'x10' section. A turntable in the floor allows expected trend of a rapid increase in CH_ as the angle of vertically mounted airtbil models to be tested between attack became more negative. One possible explanation -90 ° to 270 ° angle of attack. The Twin Otter tailplane for this behavior may be the very rough texture and 3D model was tested with a no-ice (baseline) and with the nature oi' the ice castings. These irregular protrusions may S&C and LEWICE fabricated ice configurations. The have acted as vortex generators, which energized the test matrix included 2 airspeeds, 6 elevator angles, and boundary layer and reduced the hinge moment tbr the I0-13 angles of attack for 2 section geometries. lower angles of attack. However, as the angle of attack Surface pressure tap and drag wake survey data was further increased, the separation bubble extended far were acquired to calculate the CI, Ca, Cm and CH_ lor enough down the chord to cause the hinge moment shift each configuration and test condition. The results from as was seen in the other two ice shape cases. This this test are reported in Ref 8. phenomt.,non may be airfoil specific and was not explored A second phase of testing was conducted during August further because it was not central to the program tasks. 1996 at the OSU LSWT using the ice castings developed Similar data was collected for each elevator setting, fi'om the IRT test. Similar to the 1994 test, measurements of airspeed, and airfoil section. These results were directly the surface pressures and wake drag survey were used to used in the analytical code development and flight test calculate the aeroperlbrmance coefficients. The results from planning tasks. This work was fundamental to the this test are relx_ned in Ref 9. continuation of the prqiect and provided valuable A sample of the aero wind tunnel results lot one informal ton to reduce risk during the flight tests. elevator deflection (te=0 °) can be seen in Figure 10-Figure 13. All aeroperlormance coefficients (Ci, Cd, Cm and Cn¢) Flight Tests are affected by the ice contamination to some de_ee. Flight t,_sts were conducted with the NASA Icing Ranking the ice shapes in terms of lift coefficient, the least Research Aircraft. The aircraft is a modified DHC-6 affcctcd was the inter-cycle ice, lbllowed by the thiled boot Twin O_ter that has been used extensively at NASA ice, and ending with both the S&C and LEWICE ice shapes LeRC f(r over 16 years to acquire natural icing flight

4 American Institute of Aeronautics aJ4dAstronautics dataaswellasexploretheicingeffectson aircraft The steady wings level and steady heading sideslip performance,stabilityandcontrol.FortheTIP, the test points were I-G unaccelerated steady test points aircraftwasequippedwithinstrumentationtosenseand and are self-explanitory. The pushover was a long- recordthefollowing: itudinal-axis maneuver to reduce normal acceleration to • inertialparameters(linearacceleration,angular zero-G by the pilot performing a series of nose-up and ratesandposition) nose-down control inputs. Figure 15 provides a • airdata(airspeed,angleof attack,sideslip,outside graphical representation of a pushover to zero-G. The airtemperature,pressurealtitude), trim point recovery was similar to the pushover, but the • enginedata (propellerRPM, enginetorque nose-down push on the control column was limited to pressure,fuelflow) the column position where trim was established. Stall • controlsurfacedeflections(elevator,, recovery maneuvers were initiated from a near-stall ,flap) speed where a nose-down control column was • pilotforces(elevator,,rudder) implemented to recover from the stall. Wind-up turns • videoof pilot action & horizon were performed through steep-turns to reach 2-G Due to the nature of this testing, new instrumentation normal acceleration. was added to record tailplane specific data: The results from this initial effort were used directly • tailplane flow field & surface pressures. in the IRT test and analytical c_e development, and • tailplane flow visualization provided critical information needed to proceed to the Three 5-hole probes were mounted along the span of the next phase of flight test with the ice shapes. left horizontal taws leading edge to record the flow field at the tailplane (Figure 14). A pressure belt was Phase 2 htter-cvcle Ice Tests wrapped around the and elevator to allow The second phase of flight tests was conducted on calculation of the tailplane lift performance. A video the Twin Otter between July and October 1997. After camera was mounted below the tail to monitor and repeating selected baseline tests, the inter-cycle ice record tuft activity on the suction surface. castings were mounted to the horizontal tail (Figure 16- Figure 17). Flight tests points included I-G steady Phase I Baseline Tests wings level, steady heading sideslips, thrust transitions, Initial baseline (clean tail) flight tests were pushovers, and elevator doublets. These maneuvers conducted September to October 1995. The purpose of were conducted for the full range of flap settings [0, I0, these flights were to: 20, 30. & 40] and airspeeds [Vs to Vf_]. Note: The • confirm DHC-6 hydraulic system functionality DHC-6 is placarded to warn pilots not to extend the during low-G pushovers flaps greater than 10° in icing conditions. This flight test • determine tail flow conditions and elevator settings was executed in a highly structured research for cruise, hold, and approach flight phases which environment in order to capture a full compliment of were used in the IRT tests aircraft configurations. NASA LeRC subscribes to the • determine range of tailplane inflow angles for all 10° flap limitation in routine icing operations and does flap, speed, thrust and normal accelerations to be not recommend exceeding the manufacturers stated tested with the ice shapes limitations. Table 3 provides a summary of the In this first phase+ 17 flights were made to document maneuvers flown with the inter-cycle ice shape. baseline tailplane aerodynamics in steady and maneuvering flight. Control variables were thrust Table 3 later-cycle lee Flight Test Maneuvers coefficient (CT), flap deflection (+F), airspeed (V), Maneuver Number of test points angle of attack (_), angle of sideslip (13), pitch rate (q) I-G Steady win_,s level 67 and normal acceleration (N2). Data was recorded for all Steady headin_ sideslip 15 maneuvers listed in Table 2. Pushover 32 Table 2 Baseline Flight Test Maneuvers (Phase 1) Thrust Transition 29 Elevator Doublet 16 Maneuver Number of test points Take off & Landing, 20 Flow unsteadiness on the lower side of the tail was 377 I-G Steady win_s level detected through minor yarn tuft activity in the I-G Steady headin_ sideslip 233 steady wings level points when flaps were extended to Pushover 419 8F=30 & 40 °. During the pushover and elevator Trim point recovery 2O doublets, flow separation occurred when flaps were [Stallrecovery 24 extended to _iF=20 ° or more. It was during the inter- Wind-up turns cycle ice configuration that the thrust transition

5 American Institute of Aeronautics and Astronautics maneuverwasdeveloped.It wasfoundthatbyholding the worst and nearly identical to the LEWICE shape. velocityconstantandslowlyincreasingthethrust,that For thi_ reason the LEWlCE shape was not flight at somelevelof thrust,flow separationandtail stall tested. Table 5 lists the details of this limited test couldbeinitiated.Moreover,theelevatorwasdeflected matrix. trailingedgeup(TEU),whichwasacommontailplane configurationtbr landing.(Notethatforthepushover, Table 5 S&C Ice Flight Test Maneuvers the elevatorwasdeflectedtrailingedgedownat minimum_). Thrusttransitionswereincludedin the Maneuver Number of test points I-G Steady wings level 62 testmatrixfor theremainingcases.Elevatordoublets, 12 typicallyusedfor systemidentificationandparameter Stead_' heading, sideslip estimationflight tests,werealsoaddedto thetest Pusho"er 27 matrix.Theelevatordoubletconsistedof fourstep Thrust Transition 26 inputsto theelevatorinitiatedfromstraightandlevel Elevat,_)r Doublet 24 flight.Parameter estimation analysis and examination of the damping characteristics was possible with the Dut_ to stability and control problems with the S&C elevator doublets. ice, maximum flap deflection was limited to 30 degrees for the steady points, the elevator doublets and power Phase 2 Failed Boot Ice Tests transition points. The maximum flap deflection for the pushover was limited to 20 degrees. Even with these Alter fully testing the inter-cycle ice shape, the limitations, flow separation and control force reversals failed boot ice shape was mounted on the tailplane were experienced many times during these tests. (Figure 18) and flight tests proceeded to explore the boundaries of tail stall with this level of ice Flight Test Result Summary contamination. Maneuvers similar to the inter-cycle ice werc flown, and are listed in Table 4 Results from these flight tests are being published in References 10, I I & 12. Three of the key flight test Table 4 Failed Boot lee Flight Test Maneuvers accomplishments were (1) the identification of the dominart drivers that led to ICTS (termed "Paths to Maneuver Number of test points Tailplane Stall"), (2) the recognition of tactile cues of I-G Steady winLzs level 64 an impending tail stall, and (3) the development and Steady heading sideslip 12 demonstration of an effective tail stall recovery Pushover 29 procedure. These accomplishments are outlined below. Thrust Transition 23 The paths to tailplane stall on the test aircraft were: Elevator Doublet 17 • Increased Ice Shape Severity Wind up turns • Increased Flap Deflection Flap transitions • Increased Speed * Increased Thrust (may be airplane specific ) As expected from the wind tunnel results, there was The increase in ice shape severity (based on wind greater aeroperformance degradation with the failed tunnel tests) reduced the stalling angle of attack (_i_) boot ice shape than the inter-cycle shape. The lull test and C_, of tailplane, so that ICTS was encountered at matrix was not completed because the boundary of premature test conditions (_iF, N_, _/_). As the flaps tailplane stall with this ice shape was found during the were deflected, the _it was made more negative due to thrust transition maneuver with _iF=40 °. During this an increase in wing downwash and a decrease in _ to maneuver, a full tail stall occurred when the thrust maintair aircraft lift coefficient. As the speed increased, reached approximately CT=0.13 (nominal cruise power) the oq_i_ was made more negative due to a decrease in at an indicated airspeed of VIAS=85 knots. The stall _u_. Fir ally, as thrust was increased, a greater tail caused the aircraft to pitch nose-down for 3-sec to 0 =- downloz:i was required to counteract the nose-down 37 ° and lose 300 feet of altitude even though recovery pitching moment because the thrust line was above the procedures were employed within 1,4second of the stall. cg. For some configurations, the tail CI _m,.,_a may Alter this test point, the pushovers were limited to exceed tile Ci _.,il_J_ and result in a ICTS event. 8F=20 ° and elevator doublets were limited to a Tac:ile cues that preceded the tail stall events were maximum _iF=3()". an inability or difficulty to trim, pitch excursions, onset of pilot :nduced oscillations, buffeting in the controls - Phase 2 S&C Ice Tests not the axframe. The final ice shape tested was the S&C shape When the full tail stall was experienced during the (Figure 19). As the wind tunnel tests indicated, the power transition, the stall recovery procedure was: aeroperformance characteristics for this ice shape were • Reduce thrust (may be airplane specific )

6 American Institute of Aeronautics ar d Astronautics • Pullbackonyoke/ increase Analytical Tool • Raise flaps The third element of the TIP was the development of an Reducing thrust was the first part of the procedure analytical tool to help assess an aircraft susceptibility to because it was increasing thrust that led to the stall ice contaminated tailplane stall. The tool was a flight event that was encountered. Pulling back on the yoke path simulation code based on a 6-degree of freedom, increased the camber of the tailplane, which provided non-linear model. The computer code, TAILSIM, was enough tail download to counteract the nose-down constructed using a database of stability and control pitching moment and increase the at,_il. Raising the flaps derivatives obtained from flight tests [Ref 5], wind was initiated by the copilot immediately, but the flaps tunnel results [Ref 8, 9], and standard methods for are hydraulically actuated and movement is rather slow estimating aircraft/tailplane parameters, downwash (- l°/sec). The major lesson learned to recover from a angles, and tailplane dynamic pressure ratios. The result tail stall was to undo what was just done to cause the was a new database that combined the aeroperlbrmance event. effects of the ice contamination measured in the OSU It was noted that this tail stall recovery procedure is tunnel with the flight dynamics of the Twin Otter. The opposite of the recovery from a wing stall. The reason simulation estimated the flow field at the tail, for the difference is the location of the flow separation. determined the lift, drag, pitching moment, elevator In a wing stall, the flow separates from the upper hinge moment, and the resulting motion of the aircraft. surface of the wing, therefore reattachment is made by Test maneuvers were "flown" and the response decreasing the wing or. In a tail stall event, the flow characteristics were noted in terms of tailplane angle of separates from the lower surface of the tail and requires attack, pitch stability, and control-force reversal. a positive increase in tail _to reattach the flow. Comparisons to flight test data were made with Because of these differences in the stalling mechanisms acceptable results (Figure 20). This work was and recovery procedures, it was determined that pilots accomplished as part of a Ph.D. dissertation and is fully should be made aware of the cues that may occur prior explained in Ref 13. to a tailplane stall. Eftorts to increase pilot awareness on this topic are described in the following section. Conclusions The NASA/FAA Tailplane Icing Program was a four- Guest Pilot Workshop & Tailplane Icing Video year research program that utilized a combination of At the conclusion of the flight tests, it was felt the TIP icing experts and test facilities that included NASA gathered enough new intormation that it needed to be Lewis' Icing Research Tunnel (IRT), The Ohio State quickly shared. A Guest Pilot Workshop was developed University (OSU) Low Speed Wind Tunnel, and NASA to demonstrate the unique flying qualities of an aircraft Lewis' DeHavilland DHC-6 Twin Otter Icing Research with an ice-contaminated tailplane. An international Aircraft. These resources were used in combination to group representing various facets of the aviation accomplish the following program goals: industry aviation regulatory agencies, aircraft 1. improve understanding of iced tailplane manufacturers and aviation media pilots/reporters aeroperformance and aircraft aerodynamics, were invited. In total, 15 guest pilots and engineers had 2. develop analytical tools to help discriminate the opportunity to fly the NASA Twin Otter. This tailplane sensitivity to icing, and demonstration program provided a mutually beneficial 3. develop training aides to expand the awareness of forum for the exchange of information between NASA the ICTS aviation hazard and the user community. Feedback from the guest pilots was very positive. Each indicated a greater appreciation The interconnectivity of the program elements led of ICTS as a result of this guest pilot workshop. Media to the safe and successful outcome of reaching the articles were written in the winter of 1997-98 to provide program goals. Insights gained through this program a rapid dissemination of some of the key lessons learned were the paths to stall, recognition of tactile cues prior to the pilot community. to an impending tail stall, and a demonstrated recovery Further dissemination of vital information was procedure. The rapid dissemination of key information accomplished through the production and distribution to pilots was accomplished through the guest pilot of a 23-minute educational video entitled, "Tailplane workshop, the media articles, and the "Tailplane Icing" Icing". The target audience is primarily pilots who video. All of these serve as good examples of NASA's might encounter in-flight icing. The video has been response to flight safety initiatives and the quick release enthusiastically received, and is available upon request of information to the user community. through the authors.

7 American Institute of Aeronautics and Astronautics Acknowledgments The authors would like to thank the technical staffs 10 Rat_,asky, T.P., Van Zante, J. Foss & Sim, A.G., from the NASA Icing Research Tunnel, OSU Low "NASA/FAA Tailplane Icing Program: Flight Test Speed Wind Tunnel. and the NASA Twin Otter. Special Rep_rt," being reviewed for NASA/FAA TP recognition goes to Mr. Richard Ranaudo tor both his II superb skills as a test pilot and his keen research insight, Ratvasky, T.P. & Van Zante, J. Foss, "In-flight and to Dr. Dale Hiltner for all of his effort in Aerodynamic Measurement of an Iced Tailplane," AIAA 99-0638, 1999 developing TAILSIM. We thank Mr. John P. Dow, Sr., of the FAA Small Airplane Directorate, for his 12 Van Zante, J. Foss & Ratvasky, T.P., "'Investigation continued support and active promotion of the TIP. We of Dynamic Flight Maneuvers with an Iced also would like to express our appreciation to our Tailplane," AIAA 99-0371, 1999 sponsors, NASA's Aviation Operations Systems Base program and the FAA Technical Center. 13 Hilmer, D. W., "An Investigation of the Tailplane Aerodynamics and Aircraft Dynamics of the Ice Contaminated Tailplane Stall Phenomenon," Ph.D. References Dissertation, Department of Aerospace, Aviation, and Applied Mechanics, The Ohio State University, Dow, J.P. Sr., FAA Small Airplane Directorate, Columbus, OH, 1998. private communication

2 Ingelman-Sundberg, M. & Trunov, O.K., "Methods For Prediction Of The Influence Of Ice On Aircraft Flying Characteristics," Swedish-Soviet Working Group on Flight Safety Report No. JR-I. 1977

3 Ingclman-Sundberg, M. & Trunov, O.K., "Wind Tunnel Investigation Of The Hazardous Tail Stall Due To Icing," Swedish-Soviet Working Group on Flight Safety Report No. JR-2. 1979

4 Trunov, O.K. & Ingelman-Sundberg, M., "On The Problem Of Horizontal Tail Stall Due To Ice," Swedish-Soviet Working Group on Flight Safety Report No. JR-3. 1985

5 Ratvasky, T.P. & Ranaudo, R.J., "'Icing Effects on Aircraft Stability and Control Determined from Flight Data, Preliminary Results," AIAA-93-0398, NASA TM 105977, 1993

6 Ratvasky. T. P., "NASA/FAA Tailplane Icing Program Work Plan," April 1994 7 Bowen, D.T., Gcnsemer, A.E., Skeen, C.A., "Engineering Summary of Icing Technical Data." General Dynamics/Convair, San Diego, CA, Technical Report ADS-4, 1964

8 Hiltner, D.W.. McKee, M., La No6, K.B., "DHC-6 Twin Otter Tailplane Airfoil Section Testing in The Ohio State University 7x 10 Wind Tunnel." 1995, being reviewed for publication

9 Gregorek, G.M., Dreese, J.J., La No6, K.B., "'Additional Testing of the DHC-6 Twin Otter Iced Airfoil Section at The Ohio State University 7'xl(l' Wind Tunnel." 1996. being reviewed for publication

8 American Institute of Aeronautics a_d Astronautics 50 L. Flap 32 °

Elevatorup

evator down 2o ¢1

100 200 300 400 500 HorizontalDistance (meters)

Figure 1 Trajectory of Vickers Viscount accident in Stockholm, Sweden 1977

Figure 3 DHC-6 Tailplane model in OSU LSWT

_e Shape _ derived from in-flight photos and _?{ . AOS-4 _ used in previous stability & '_:'_ control flight tests

Figure 4 S&C Ice Shape

jJ . Ew,cS.ape ) _ v=,20_s,,,Iph,,=oo i,J _ • LWC=O.5g/m 3, MVD=-2011m • T0=-4° C, time=45 rain

Figure 5 LEWICE Ice Shape Figure 2 DHC-6 model in NASA IRT

9 American Institute of Aeronautics and Astronautics Ice Shape Effects On Lift Coefflciem Section 1, V=6Okts, Be=0.0* -,Inter-cycle IRT Shape

v=135 kts, alpha=-2.9 o • LWC---0.5g/m3, MVD=20_m • T0 oC time=15 min, with boot I cycle every 3 minutes

Figure 6 Inter-cycle IRT Ice I

Failed Boot IRT ShaDe Figure 10 Ci Summary for all ice shapes • V=135 kts, alpha=-2..9 ° !_ • LWC=0.5g/m 3, MVD--201un Ice Shape Effects On Drag Section 1, V--6Okts, 6e=0.O o T " "__0----4 ° C, time=22 rain '" / q

• q [ Figure 7 Failed Boot IRT Ice

-o-sAc _, s_l_,,

Figure 11 Cd data summary for all ice shapes

Ice Shape Effects On Pttching Moment Section 1, V--60kts, Be=-0.0o

, I I t _ Figure 8 NASA Icing Research Tunnel

-it--c_

I O BLADE FAN - -90 FI. DIA -7 I IN___ TI ,% |1 _ I r- ...... 7 4o Frl I / I I // , .,q_==_! Figure _2 Cm Summary for all ice shapes _ITL lo_Lr I A _g,l_i.'t I'res'LsEt,non I qCREEN / I TES'I SECI'ION I _ I ._,.__ BU1LDING WALLS

ISFT 1.'_.1F]" - 0 IN

Figure 90SU 7'x10' Low Speed Wind Tunnel

10 American Institute of Aeronautics and Astronautics Ice Shape Effects On Hinge Mome_ Section 1, V=6Okls, 6e=0.0 _ Altitude (ft)

6000 ......

5260 J: .... \- ' _IL.,I., 5ooo ! ; ; , o 5 lO 15 20

Figure 13 C=,_Summary for all ice shapes G - Load " i o 0 5 10 15 20 time (sec)

Figure 15 Zero-G Pushover Sample, 5F--0, V=I.4Vs

..... t

| ! i [....

Figure 14 NASA Icing Research Aircraft with Tail Flow Probes Figure 16 Installation of ice cast on tail

II American Institute of Aeronautics and Astronautics --THS_

_U_ ...... Jl

o _o ee 3e 40 so _o 70 _°i

Figure 17 Inter-cycle Ice on Tailplane I

_T ......

Figure 18 Failed Boot Ice on Tailplane

=

Figure 20 Comparison of TAILSIM to Flight Test Figure 19 S&C Ice on Tailplane for Pushover (Ice=S&C, 8F=20, CT--'0.10)

12 American Institute of Aeronautics aad Astronautics

REPORT DOCUMENTATION PAGE Fo_ Approved OMBNO.0704-0188

Pubic report ng burden or this collection of information is estimated to average 1 hour per response, inct iding the time for reviewing instructions, searching existing data sources. ! gathering and maintaining the data needed, and completing and reviewing the collection of information, c end comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Sen,ces, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204. Arlington. VA 22202-4302, and to the Office of Management and Budget, Pa)erwork Reduction Project (0704-0188), Washington, DC 20503. 1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

January 1999 Technical Memorandum 4. TITLE AND SUBTITLE 15. FUNDING NUMBERS

NASA/FAA Tailplane Icing Program Overview

WU-548-20-23q30 6. AUTHOR(S)

Thomas E Ratvasky, Judith Foss Van Zante, and James T. Riley

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORG=ANIZATION REPORT NUMBER National Aeronautics and Space Administration Lewis Research Center E-I1502 Cleveland, Ohio 44135-3191

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING AGENCY REPORT NUMBER

National Aeronautics and Space Administration NASA TM--1999-208901 Washington, DC 20546-0001 AIAA-99-0370

11. SUPPLEMENTARY NOTES

Prepared for the 37th Aerospace Sciences Meeting & Exhibit sponso: ed by the American Institute of Aeronautics and Astronautics, Reno, Nevada, January I 1-14, 1999. Thomas P. Ratvas

NAS3-98008); James T. Riley, FAA Technical Center, Atlantic City Airport, New Jersey 08405. Responsible person, Thomas P. Ratvasky, organization code 5840, (216) 433-3905. 12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Unclassified - Unlimited

Subject Categories: 08, 03. and 05 Distribution: N)nstandard

This publication is available from the NASA Center for AeroSpace Information, q301 ) 621-0390. 13. ABSTRACT (Maximum200 words)

The effects of tailplane icing were investigated in a four-year NASA,'FAA Tailplane Icing Program (TIP). This research

program was developed to improve the understanding of iced tailplane aeroperformance and aircraft aerodynamics, and to develop design and training aides to help reduce the number of incidents and accidents caused by tailplane icing. To do this, the TIP was constructed with elements that included icing wind tunnel testing, dry-air aerodynamic wind tunnel testing, flight tests, and analytical code development. This paper pro/ides an overview of the entire program demonstrat- ing the interconnectivity of the program elements and reports on cur 'ent accomplishments.

14. SUBJECT TERMS 15. NUMBER OF PAGES 18 Aircraft icing" Stability and control; Tailplane icing: Tailplane perforl mnce 16. PRICE CODE

A03 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECU;_IITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF REPORT OF THIS PAGE OF AE STRACT

Unclassified Unclassified Unclassified

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. Z39-18 298-102