NASA/Industry Advanced Turboprop Technology Program

NASA Technical Memorandum 100929

Nd6-24t4 1 I_AS_-T_- I£C_9) NAS A/I NLLS_I_/ _EVANC££ I[_EC_bC£ 21_(hI_(ICG_ £FCG_A_ II_A2A) 26 F CSC£ 21_

G3,/07

Joseph A. Ziemianski and John B. Whitlow, Jr. Lewis Research Center Cleveland, Ohio

Prepared for the 16th Congress of the International Council of Aeronautical Sciences--ICAS '88 Jerusalem, Israel, August 28--September 2, 1988

NASA/INDUSTRY ADVANCED TURBOPROP TECHNOLOGY PROGRAM

Joseph A, Ziemianskl and John B. Whitlow, Jr. Natlonal Aeronautics and Space Admlnlstratlon Lewis Research Center Cleveland, Ohlo 44135

Abstract shown Is an artlst's rendltlon of the new advanced turboprop, or "propfan" concept and, for compari- Experimental and analytlcal effort shows that son, the old four-bladed prop typical of those use of advanced turboprop (propfan) propulsion used on the Lockheed Electra. Although the old Instead of conventlonal turbofans In the older turboprops were fuel efficient up to airspeeds narrow-body alrllne fleet could reduce fuel con- of slightly over Mach 0.6, they experience a rapid sumption for thls type of alrcraft by up to increase In compressibility losses beyond these 50 percent. The NASA Advanced Turboprop (ATP) speeds due to thelr thick, unswept, large-diameter program was formulated to address the key technol- blades. Their propulsive efficiency is much higher ogles required for these new thin, swept-blade than that of hlgh bypass turbofans at speeds up propeller concepts. A NASA, Industry, and univer- to Mach 0.6+ because In generating thrust a prop sity team was assembled to develop and valldate Imparts only a small Increase In axial velocity to applicable new design codes and prove by ground a large mass flow of alr, thereby reduclng Kinetic and flight test the vlablllty of these new propel- energy losses In the discharge flow. The advanced ler concepts. Thls paper presents some of the turboprop uses very thin, highly swept blades to ' history of the ATP project, an overvlew of some reduce both compressibillty losses and propeller of the Issues and summarizes the technology deve- noise during hl_h-speed cruise. High dlsk power loped to make advanced propellers viable In the loadlngs (SHP/D Z) at least double that of the hlgh-subsonIc cruise speed application. The ATP Lockheed Electra are required for high-speed cruise program was awarded the prestlglous Robert O. and are achieved by Increasing the number of Colller Trophy for the greatest achievement In blades and lengthening blade chord. Counterrotat- aeronautics and astronautlcs In America In 1987. Ing blade designs can be used to provide still higher disk power loadlngs and eliminate some of the exit "swirl" losses assoclated with single- I. Introductlon rotation propellers. Installed propulsive efflclencles roughly equivalent to those achieved with Until the advent of the ATP program in 1978, the old Electra technology can be extended to the propeller technology development had stopped In Math O.B regime with advanced propellers. The the mld-1950's. Thls 20-year gap in development advanced turboprop produces fuel savings conserva- occurred because we did not know how to build prop tively estimated at about 30 percent due to the blades wlth the comblned structural and aero- Improved propulslve efficiency of the propfan and dynamic chaFacterlstlcs to operate reliably and 50 percent better overall with core engine effIclently at the hlgh subsonic cruise speeds improvements Included. (2) To illustrate the obtained by the turbojets then being Introduced impact of such a savings on today's U.S. alrllne Into commerclal servlce. fleet, if advanced turboprops were substituted for the turbofans used in only the narrow-body Although propellers were more fuel efficlent 727's, 737's, DC9's, and MD80's, about 2.5 billion than turbojets and turbofans, propulsion develop- gallons of fuel would be saved per year. (3) ment was largely concentrated on Improvements to the turbojet durlng thls era of cheap fuel. II. ATP Programmatic ObJectlves and Plans Perspectives changed beginning In 1973, as a result of the Mlddle East oll embargo. Fuel costs NASA formally began the Advanced Turboprop escalated and by 1980 had gone from about one- (ATP) project In 1978 with the overall objective quarter to more than half of direct operating of valldatlng key technologies required for both costs, as Illustrated In Fig. 1. slngle- and counterrotatlng propfans in Mach 0.65 to 0.85 appllcatlons. Project goals were to verlfy In January 1975, the U.S. Senate Commlttee on projected propfan performance and fuel savings Aeronautical and Space Sclence requested that NASA beneflts; to verify the structural integrity of develop a program to address the fuel crlsls. (1,2) these radically different blade deslgns under In response, NASA formed an Inter-agency task force actual operatlng conditions; to establlsh passenger that consldered many potentlal fuel-savlng con- comfort levels (i.e., cabin noise and vibratlon) cepts. They proposed the Alrcraft Energy Effl- approaching those in modern turbofan-powered alr- clency (ACEE) program, which Included three liners; and to verify that propfan-powered alr- propulslon projects managed by NASA Lewis. One craft could meet the airport and community noise of these, strongly advocated by NASA Lewls and standards speclfled by U.S. Federal Alr Regula- Hamllton Standard Dlvlslon of United Technologles, tlons (FAR-36). The project plan projected system was to develop advanced turboprops whlch could technology readiness by the late 1980's. overcome the hlgh-speed compresslblllty losses of conventlona} propeller designs. The NASA ATP project was organized in such a way that technlcal Issues were first resolved NASA conducted several system studies which through wlnd tunnel testing of sma11-scale models showed that advanced propellers could have propul- before more costly large-scale ground and flight sive efflclencles about 1.3 tlmes as hlgh as those testlng. The early years of ATP (prior to 1980) of equivalent turbofans at cruise speeds of Mach 0.8. These results are illustrated In Flg. 2. Also providedtheenabllngtechnologyvla sma11-scale tunnel tests leading to design valldation and testing anddeslgncodedevelopmentto establlsh verification of aerodynamlc, acoustic, and struc- the feaslb111tyof thepropfan. Afundamental tural codes. This on-golng research program will databaseof design,analysis,andtesting tech- make extensive use of the previously acquired ATP niqueswasdeveloped.A large-scaletechnology database. Advanced concepts of a slngle-rotatlon }ntegratlonphasethendrewfromthls knowledge propfan with stator vane swirl recovery and a to deslgn,fabricate, andgroundtest large-scale hlgh-bypass-ratlo ducted-fan configuration are propfansystems.Thelarge-scaleeffort was Illustrated in the two bottom drawings. Hhlle neededto e11mlnateuncertalntlesconcernlngthe future ducted props will not have the efficiency scale-upof structural andacousticdataobtalned of unducted propfans they may be more sultable wlth windtunnelmodels.(1) F11ghtresearch for "packaging" on large alrcraft such as the testlngof large-scalepropfanpropulslonsystems Boeing 747, wasInitiated in 1986to verlfy bladestructural Integrity, to determlnecabincomfortandground noiselevels, to provldescalingcomparlsonswlth Ill. Slngle-Rotation Systems modeltunneldata,andto validatecomputeranal- yses. Asthesetests werecompleted,anexten- The first 2-ft-dlameter propfan single- slve analyslseffort wasImplementedto assess rotation model, deslgnated SR-I and Incorporat- thedataacquired. ing eight thln, swept blades was developed by NASA and Hamllton Standard i_ 1976. (2) Fromthebeglnnlngof the ATP project, a systems approach which consldered the entire aircraft When tested In a wind tunnel, the SR-I was used In deslgnlng the propulslon system, as achieved an efficiency of 77 percent at shown in Fig. 3. This Included elements such as Mach 0.8. The model blades were stable even the propeller and the nacelle, the drive system, when an attempt was made to force flutter. Installatlon aerodynamics, and the aircraft Inte- Encouraged by thls but st111 needing to fully rior and communlty environments and the effect of understand the efflclency and noise potential of these elements on meeting the goals of reduced the propfan, several more models were designed fuel consumptlon, low operating costs, and pas- and tested. senger acceptance. Thls approach followed the 1oglc path illustrated In Fig. 4. The sequence One was an Improved verslon, the SR-IM, wlth started with analyses and systems studles and a modlfled spanwlse twist to better dlstrlbute proceeded to design code development based on the blade loading, which resulted in a l-polnt scale-model wlnd tunnel tests or component tests. gain In overall efficiency, Another model, the Finally, large-scale systems were designed, stralght-bladed SR-2, was designed to provide a built, and tested both on the ground and in basellne for comparlson. Its efficiency was fllght as proof of the concept. This approach sllghtly less than 76 percent at Mach 0.8. The was used In the three propfan conflguratlonal subsequent SR-3 model, whlch incorporated 45° of areas shown: slngle rotation, gearless counter- sweep for both aerodynamic and acoustic purposes, rotation, and geared counterrotatlon. achleved an efflclency of nearly 79 percent. Some of the early blade models that were wind- The technical expertise of all three NASA tunnel tested are shown in Fig. 7. aeronautical research centers (Lewis, Langley, and Ames), more than 40 contracts distrlbuted The performance gains (i.e., Incremental over the maJorlty of the U.S. aircraft Industry, gains in propeller efficiency) and noise reduc- and over 15 universlty grants were required to tions due to increased blade sweep for the SR-2, complete the project. Major contracts were SR-IM and SR-3 configurations are summarized in awarded to General Electric on the Unducted Fan the plot of measured data In Fig. 8.(2) All (UDF), Hamllton Standard on the Large-Scale three blades were very thin but varied In amount Advanced Propfan (LAP), and Lockheed-Georgla on of tip sweep from 0° to 45°. All three of these the Propfan Test Assessment (PTA). In addition configurations are compared at their design polnt to NASA-sponsored research, a significant inde- condition _f Math 0.8, disk power loading of pendent Industrlal research and development 37.5 SHP/D L, and tip speed of 800 ft/sec. It Is effort was applled to develop these new concepts. clear that both performance and acoustics bene- Beglnnlng in August 1986, the advanced turboprop fit as blade sweep is Increased. These model propulsion concept was proven by three flight tests eventually led to the wind tunnel test of programs uslng large-scale hardware (Fig. 5). the SR-7A model, which Is an aeroelastically The NASA-General Electrlc-Boelng fllght test and scaled 2-ft model of the 9-ft proDfan flight the General Electric-McDonnell Douglas flight tested later In the PTA program. (4) test used the Unducted Fan as a proof-of-concept demonstrator for the gearless counterrotatlng Net efflclencles of the propeller models are concept. The NASA/LocKheed-Georgla Propfan Test shown as a function of Math number In Fig. 9. (5) Assessment verlfled the structural Integrlty and At Math 0.80 the SR-7A propfan has the hlghest acoustic characteristics of the slngle-rotating measured propeller efficiency - 79.3 percent. LAP propfan bullt by Hamilton Standard. On the The performance of the SR-2 propeller Is lower basis of the success of these tests and previous than that of the others because of its unswept scale-model work, Pratt & Whttney-Alllson built blade design. The number of blades and amount of a geared propulslon system with Hamilton Standard tlp sweep are tabulated below for each of these counter-rotatlng propellers that they plan to models. Fly on the MD-80 in late 1988.

The three top pictures of Fig. 6 illustrate the post-fllght test NASA generic propeller research program of analysis and scale model wlnd the upward and downward blade rotational veloclty Design Numberof Sweep vectors. Pressure contours for the blades at the blades angle, top and bottom of the rotation are relatively deg unaffected by the tilt of the rotatlona] axls SR-7A 8 41 because at these positions the rotatlonal velocity SR-6 I0 40 component Is perpendicular to the upward free- SR-3 8 45 stream component. SR-IM 8 30 Cabin noise and vlbratlon levels with past SR-2 8 0 turboprops has been less favorable than wlth turbofans. To achieve a cabin environment with propfans that Is comparable to current turbofan transports, Near-fleldacousticresults wererecently a reduction of 25 to 30 dB beyond the capacity of obtalnedwith theSR-TAmodelat hlgh-speedcruise a bare-wall untreated cabln is likely to be conditionsin theNASALewis8x6ft tunnel,as Fequlred for a wing-mount installation. NASA shownin Flg. IO. Peakfundamentaltonelevels Langley Research Center has been involved in sev- areplotted againsthelical tlp Machnumber(I.e., eral ATP noise reduction activities, includlng the thebladerelative Mathnumber)for threeloading evaluation of advanced cabin sldewa]] concepts. A levels.(6) Thedatashowthat fundamentaltone Langley/Lockheed-Callfornla effort has led to the levelsmaypeak,level off, OFdecreasebeyonda development of an advanced cabin wall acoustic hellca] tlp speedof Machl.l, dependlngon treatment utlllzlng Helmholtz resonators tuned to oadlng. the fundamental blade passing frequency. Thls concept was flight tested in the PTA program and is dls- Althoughtheeffect of increaslngbladesweep cussed later In con_ectlon with those fllght tests. s positivein termsof Improvementsto propulsive efficiencyandacousticsat highflight speeds, Under NASA sponsorship, Hamilton Standard the structural andaeroelastlcdeslgnbecomesmore initiated the design of a Large-Scale Advanced difficult. Onestructuralconcernrelatesto Propeller (LAP) In 1982. The resultlng 9-ft steadystate stresslevelsdueto centrlfugaland dlameter SR-7 propfan deslgn Incorporates the steadyaerodynamicloads. Anotherrelatesto an spar-shell type of blade construction Illustrated aeroelastlcinstablllty phenomenoncalled flutter in Fig, 13. (1) All new Hamilton Standard straight- which can occur In response to forced excitations bladed commuter aircraft propellers use a similar caused by unsteady, unsymmetrical airflows pro- spar-shell type of construction which has proven duced by gusts, upwash from the wing, and airframe- to be very safe, reilable, and lightweight. The induced flow fleld distortions. The presently FOD problems inherent in earlier solid aluminum lll-deflned boundary for hlgh-speed classical blades are avoided by protecting the single Ioad- flutter will occur at increasingly reduced flight bearing spar with an aerodynamlcally-shaped flber- speeds as blade tlp sweep is increased, all other glass shell. This construction technique, however, things (e.g., materials, constructlon) remalnlng was unproven for the thin, swept, LAP blade deslgn equal. Avoidance of this flutter boundary is one which is subjected to complex nonlinear deflec- reason why the sweep of the SR-7 was llmited to tions under load. The possibility of hlgh-speed 41 ° at the tlp. classical flutter and the need to verify the design codes that were used reinforced the need for An experlmental and analytlcal research pro- large-scale fabrication and fllght testlng. gram Is belng conducted to better understand the flutter and forced response characteristics of The 9-ft dlameter size for the LAP rotor advanced hlgh-speed propellers. A comparlson of assembly was selected as the minimum size that measured and calculated flutter boundarles for a would permlt a reallstlcally-scaled blade cross propfan model deslgned to flutter is shown in sectlon wlth the mlnlmum allowable shell gauge Flg. 11. (7,8) The theoretlcal results, from the thlckness. Exlsting drive system capabIllty was NASA Lewls-developed ASTROP3 analysls, _nclude the limited to about 3000 SHP at the Mach 0.8/35 000 ft effects of centrlfugal loads and steady-state, design polnt and also dlctated a diameter of about three-dlmenslonal alr loads. The analysis does 9 ft If disk power load_ngs (SHP/D 2) in the desired reasonably well in predicting the flutter speeds deslred 30 to 40 SHP/ft L range were to be obtained. and slopes of the boundaries. However, the dif- ference between the calculated and measured flut- A LAP static rotor test was completed In late ter Mach numbers Is greater for four blades than 1985 at a Wrlght-Patterson Air Force Base facll- for eight blades. This Implles that the theory Ity, as shown In Flg 14, using a facility elec- Is overcorrecting for the decrease In the aerody- trlc drive motor. (IUi Forty-slx strain gages, namic cascade effect wlth four blades. some of which can be seen In thls photo, were installed on the LAP durlng manufacture. Of this EuleF code so]utlons have recently been deve- total, 30 strain gages were connected through sllp loped to descrlbe the unsteady, three-dlmenslonal rlngs to a data system and continuously recorded flow fleld generated with advanced propeller during propfan operation while the remainder were deslgns. (9) A graphlca] dlsplay of the blade pres- spares which could be used in the event of a mal- sure contours from this analysls is shown In functlon of one of the primary gages. This %nstru- Flg. 12 for an SR-3 propfan in regions where the mentatlon and data system were retained throughout flow is supersonic when the axis of rotation Is the LAP and follow-on PTA testing. In early 1986, tilted upward 4° from the Math 0.8 free-stream the LAP was Installed In France's Modane wlnd tun- flow. The downward moving blades (on the right nel to verify blade structural integrity at speeds _n the flgure) experience the hlghest ]oadlngs, up to Mach 0.83. A second Modane tunnel entry whereas the upward movlng blades experience some occurred In early 1987 to acquire blade steady and unloading due to the alternating alignment of the unsteady pressure data for verifying and improving upward component of the free-stream veloclty with aerodynamlc predlctlon codes. Figure 15 depicts the second LAP Modane entry, blade pressure Instru- and predictions obtalned with an analytical code. (15) mentatlon locatlon schematics for two blades spe- Comparatlve maximum noise data along the fuselage cially instrumented for this particular test, and exterlor are presented at axial locatlons fore and also one of the two completed propfans delivered aft of the plane of rotation In Flg. 18. A maxi- to the PTA project. mum sound pressure level of 147 dB was measured at the fundamental blade passlng tone of 225 Hz. The The two-bladed verslon of the elght-blade measured local noise reduction at an adjacent propfan shown In F_g. 15 was used In some of the location Inside the bare-wall cabin was 25 dB. Modane testing because of the llmlted fac111ty power available to drive the propeller. In th_s Subsequent to the basic bare-wall cabin fllght way the propeller could be operated at a reason- test effort, a lO-ft sectlon of the PTA cabin was able power 1oadlng per blade. The large size of cleared for acquiring data wlth an advanced cabln this propeller allowed much more detailed blade acoustic treatment In early IgBB. The treated pressure measurements than could be obtalned on enclosure, located fore and aft of the propfan the 2-ft diameter models tested prevlously. plane of rotation, consisted of tuned Helmholtz resonator wall panels attached to a Framework Prior to f11ght test, several scale-model mounted to the cabin floor through vibration iso- tests of the PTA alrplane were conducted In NASA lators. Preliminary results obtained from the 31 wlnd tunnels In the Interest of fllght safety and cabin Interior microphones Indlcate that noise to obtain data for valldatlng aerodynamic predIc- levels 25 to 30 dB below that of the bare-wall tlon codes. Both a 11g-scale airplane aeroelas- cabin were obtalned. This is the approximate tic model 11 and a !_-scale aerodynamlclstab111ty level required for comparability with existing and control model _m_j were built and tested. The turbofan-powered airliners. results From both models showed excellent agreement with analytlcal predlctlons. A rake survey The PTA flight test effort was concluded in of the flowfleld at the propfan plane was also March 1988. Although some preliminary results are conducted wlth a modified version of the PTA aero- available, because of the masslve quantity of data dynamic model to verlfy the flowfleld predIctlon to be analyzed the final results will not be avall- code. (13) The predicted ?lowfleld at the varlous able until October 1988. It is clear, however, that f11ght test conditions was used as a correlatlon these fllghts, as intended, verified propfan struc- parameter for measured propfan stress. tural Integrlty. There was no evidence of flutter anywhere In the fllght regime and blade stressing A ground static test of the entire PTA pro- was In good agreement with predictions. Measured pulsion system - Including the straln-gage- blade stresses were wlthln limits established by Instrumented propfan, englne/gearbox, and forward Hamilton Standard for Inflnlte llfe. Prellmlnary nacelle - was conducted In the sprlnq of 1986 at acoustic data analysis Indicates that magnltudes an outdoor thrust stand (Fig. 16). (14) Over 50 hr and trends are generally as predicted wlth prop- of extenslve testing was accomplished, with essen- fan nolse sllghtly lower than predicted. It tially flawless system operation and with blade appears that advanced cabin acoustic treatments stresses at a level somewhat below those seen at can reduce interior noise to acceptable levels and the previous static rotor test. In neither of that FAR36 (stage 3) airport communlty standards these static tests was there any evidence of blade can be met when data are extrapolated to a product flutter and stresses were low except at very hlgh design. blade angles where buffeting characteristic of separated flow was sometimes Indlcated. IV. Gearless CounterrotatIon Systems After G-II alrcraft modlflcatlons to Install the propfan propulsion system on the left wing and CounterrotatIon propeller systems are of Interest because of their potential to further the subsequent ground checkout testlog_)PTA_I flight testing began In the spring of 1987. Some enhance propulsive efflclency by reducing or ellm- photos of the PTA fllght testing are shown In Inatlng the swirl component of the discharge Fig. 17. One unique feature of the PTA design was velocity. In order to generate propulsive thrust, the varlable tilt nacelle whlch allowed the for- a propeller must take essentially axial flow and ward portlon of the nacelle to be tilted up or turn it to do work, in much the same way that an down so that blade stresses could be assessed as a airplane wing must turn the flow slightly downward functlon of Inflow angle. The PTA flight test In order to generate upward llft. The result with program was performed to verlfy LAP structural a single-rotatlon propeller is that the discharge Integrlty and characterize LAP acoustics both out- flow must have a nonax1al rotational component of side and Inside the cabin as well as on the perhaps several degrees, depending on disk loading ground. Data were obtained over a Flight envelope and tlp speed. A decrease In net thrust (and, extendlng from Just above low-speed stall to hence, propulsive efficiency) results from thls Mach 0.89 and at altltudes from 8OO to 40 OOO ft. nonaxlal, or "swlrl," velocity component slnce the total change In momentum is not In the axial dlrec- A total of up to 613 acoustlc, g-loadlng, tlon, as It Ideally should be for the production of pressure, strain, temperature, and miscellaneous thrust. The swlrl losses for an Isolated single- operatlng parameters were recorded onboard the rotatlon propfan at Math 0.8 design point operat- aircraft at each of almost 900 completed test runs Ing condltlons are typically equivalent to about durlng the PTA research fllght tests. The PTA elght points In efficiency. These losses can be flight test program Involved more than 133 hr of reduced or ellmlnated by usln9 counter-rotatlon flight tlme over a total of 73 flights. as a swirl recovery technique. Wlth counterrotation, the second, or aft stage, propeller Inltlal fllght nolse test data agree favor- rotating in the opposite dlrectlon returns the ably wlth both scaled-up NASA wlnd tunnel data flow to the axial direction as It performs Its work. By 1983 General Electric became convinced Both steady and unsteady flow prediction that a gearless counterrotattng Unducted Fan (UDF) codes have been developed for counterrotation pro- engine would be a vlable fuel-saving alternative pellers. Flgure 24 shows the three-dimensional to the turbofan. Rather than venture into the Image of the UDF pressure distrlbution generated uncertain area of gearbox design for a 20 000 hp_ with a NASA Lewis-developed Euler predIctlon class engtne, GE chose to ellmlnate the gearbox by code. (18) The coupling between rows is done in a using a counterrotatlng power turbine to dlrectly clrcumferentially-averaged sense which does not drive the props. (2,16) A cutaway of thls concept conslder blade-wake interactlon effects. The pres- ls shown in Fig. 19. The gas generator ahead of sure dlstributlon is shown on the nacelle, blade the power turbine tn th_s concept demonstrator is surfaces, and on a flow cross section downstream not mechanically linked to the power turblne, of the aft rotor. An unsteady Euler code solutlon which is drtven solely by hot exhaust gas. has also been applled to the FT-A7 UDF configuration to obtain the full unsteady three-dimensional The UDF prop blades were designed for an over- solutlon for the flow field. _> Pressure contours all disk power loading almost twice as great as at a particular instant in time are shown in that of the slngle-rotatlon designs. Hub-to-tlp F_g. 25 at a plane just downstream of the aft radlus ratio of these blades _s about 75'percent blade row. The low pressure islands shown are the hlgher than that typical of geared deslgns In result of tlp vortex shedding. order to accommodate the large-dlameter power turbine. The large turbine dlameter Is required for The UDF demonstrator engine is shown during power generation and compensates for the low rpm ground static testlng at GE's Peeb[es, Ohlo out- restraint imposed by the prop tlp speed IImlt. door test faclllty In the top photo of Fig. 26. Except for a set of Inlet and outlet guide vanes, After completlng thls ground testing, the UDF this 12--stage power turbine Is unlque in that It demonstrator was installed on a Boelng 727 air- has no stator vanes between the alternating plane for flight testlng In August 1986. The opposlte-rotatlon blade rows. engine was later Installed on an MD-80 In May 1987 for a Douglas f11ght test program. These flight The UDF blade structural deslgn chosen by GE test conflguratlons are shown in the two bottom Is somewhat dlfferent from that used by Hamilton photos of Flg. 26. The UDF was substltuted for Standard for the LAP blades. The LAP blade con- the rlght-hand JTSD powerplant on the 727 and for sisted of a full-length structural spar and a the left-hand JT8D on the MD-80 alrplane. F_berglass she11; the shell of the UDF blade, In contrast, Is the structural element and the spar Fundamental tone dlrectivities for the F7-A7 iS used for attachment. (2) The blades have a blade comblnatlon, the proof-of-concept UDF con- half-span tltanium spar covered wlth an aerody- flguratlon, are shown in Fig. 27 for scaled wind namlc shell of epoxy-bonded carbon-flber/ tunnel model data and fu11-scale fllght data flberglass plies, as shown In Fig. 20. The plles obtalned by the Instrumented NASA Lewls Leafier are orlented In such a way as to tune the dlrec- in formation flights with the UDF-powered 727. (5) tlonal stlffness for blade shape control, There Is excellent agreement among the model wind- strength, and aeromechan_cal stablllty. The blade tunnel measurements, fu11-scale f11ght data, and deslgn also Includes a nickel leadlng-edge sheath prediction at most sldeline angles, although the and polyurethane film bonded to the outer shell to wind tunnel data appear to be somewhat high at provide addltlonal protectlon. Blade design and forward angles. constructlon Is baslcally the same for both forward and aft rows. V. Geared Counterrotatlon Systems The NASA Lewls counterrotatlon pusher propel- ]eF test rlg shown wlth a UDF blade conflguratlon NASA has also sponsored research leadlng to in the 8- by 6-Ft Wlnd Tunnel In Flg. 21 was one the development of a more conventlonal geared of three bullt by GE for model blade testlng. (17) counterrotating propfan system uslng blade tech- The other two were used at Boelng and GE. Per- nology which _s basically an extenslon of that formance, flowfleld, and acoustic measurements pioneered in the LAP slngle-rotation propfan. As were made during thls testlng. The UDF model part of this effort, A11ison Gas Turblne Dlvlsion blade configuratlons tested at NASA Lewis are of General Motors designed, fabricated, and rig- shown In Fig. 22. The designs dlffered in tip tested a 13 OOO shp-class advanced In-llne differ- sweep, planform shape, alrfoll camber, and entlal planetary counterrotatlon gearbox. (15) Included one case with a s_gnlf_cantly shortened Ease of ma_ntalnabillty, hlgh (over 99 percent) aft rotor. The planform shapes for most forward mechanical efficiency, and a high durability were and aft rotors were very slmilar. These blades of paramount Importance In thls gearbox design. were deslgned and bullt by General Electrlc. Net A111son used the results of these tests in deve- efflclences for the F7-A7 blade configuration are 1oplng the technlcally slmllar flightwelght gear- shown In Flg. 23 as a functlon of crulse speed for box for the Pratt & Whltney-A111son/Douglas three power 1oadlngs. (5) At Mach O.72 design con- 578DX/MD-80 flight tes_ program. ditions, efflclency Is strongly affected by 1oad- Ing, but as Math number Increases, compresslbillty The Hamilton Standard CRP-XI propeller model losses domlnate and efflclencles fall off essen- (Flg. 28) was evaluated in wlnd tunnels for per- tlally Independent of loading. formance and flow fleld data, blade stresses, and acoustics.(19, 20) At the Math O.8 deslgn power General Electrlc used NASA design codes for 1oadlng, a net effIclency improvement of ~6 percent propeller ply deslgn, flutter analysls, aerody- was obtalned over the analogous SR-7A single- namic design, and nolse predlctlon. Model rlg rotatlon model. The counterrotatlon propfan data was also used by GE to modify, Improve, and deslgn used In the MD-80/578DX flight test is verify their own Inhouse codes. based on the technology of the CRP-XI model and the earller SR-7.

ORIGINAL PAGE IS OF pOOR QUALIT_ Flowcharacteristics of these types of coun- counter-rotatlng concept. The NASA/Lockheed- terrotatlon propellers have also been predicted Georgia Propfan Test Assessment fllght test used with analytical codes. One such example, shown in a propfan built by Hamilton Standard to prove the Fig. 29, Is the analytically slmulated flow vlsu- geared slngle-rotation concept and to compare a11zatlon of an off-deslgn vortex shedding phe- large-scale fllght data with an extensive wind nomenon with the CRP-XI nx)del design. If not tunnel model data base. On the basls of the suc- accounted for In analytical models, errors In pre- cess of these tests and previous scale-model work, dicted performance and noise wlll occur. The Pratt & Whitney - Allison built a geared counter- results shown are from an Euler code developed at rotatlng propulslon system that they plan to fly NASA Lewis and ruff to slmulate takeoff with a thls year on the MD-80 aircraft. It is clear From high blade Incidence angle. (18) these efforts that much of the predicted fuel sav- Ings potential can be realized with designs which Levels of the first flve harmonics of slngle- malntaln their structural Integrlty In flight. rotatlon and counterrotatlon propeller nolse, Prellmlnary Indicatlons are that propfan noise based on Hamilton Standard model test data, are trends are also generally as predlcted and that shown in FIq. 30 at three axial locations in the acoustic treatments can be deslgned to satlsfacto- far field. (20) The single rotatlon tone levels rlly attenuate cabin noise. have been adjusted upward 3 dB to compare the equivalent of two independent propellers wlth the Because of the fuel savings potential for CRP-XI counterrotation conflguration. Single and thls concept, whlch can be implemented without any counterrotatlon fundamental tones are then roughly sacrlflce In the speed and comfort we have grown equal, but the counterrotation higher harmonics to expect, it Is anticipated that this technology are dramatlcally higher at all locations due to will lead to a whole new generation of propfan- the unsteady aerodynamlc Interactions between powered alrcraft - both clvll and mllltary. blade rows. The hlgh fore and aft harmonlc levels must be dealt wlth to achieve acceptable counter- Although more work is yet to be done in ATP rotatlon community nolse levels. data acqulsltlon and analysls, the rapidly expand- Ing database already In existence w111 permit technIcally sound marketing decisions to be made VI. Future Research by industry in the deployment of prlvate invest- ment capltal. Early candldate production aircraft In future work, NASA w111 attempt to achleve are now on the drawing boards. The Lewls Research some of the swirl recovery benefit of counterrota- Center and the entire NASA/Industry Advanced Tur- tlon without the addltlonal complexity and noise, boprop Team were cited In the Collier Trophy award by uslng swirl recovery vanes wlth a single- (Fig. 31) presented in May 1988, by the National rotatlon propfan model (Flg. 6, bottom left). Aeronautic Assoclation for this work which they conslder as the greatest achievement in aeronau- Future NASA research effort will also be tics or astronautics In America demonstrated by directed toward the ducted propeller, whlch was actual use in the previous year. dlscussed brlefly in connection wlth Fig. 6. Ducted props are more easily integrated Into the design of large long-range aircraft than unducted Vlll. References props because they are more compact than unducted propfans. Underwlng %nstallatlons of propfans on Whltlow, J.B., Jr., and Slevers, G.K., "Fuel large heavy aircraft are unllkely because the Savlngs Potential of the NASA Advanced large tlp diameter requirements will cause ground Turboprop Program," NASA TM-B3736, 1984. clearance problems. There are several technical Issues which must be addressed with regard to Hager, R.D., and Vrabel, D., "Advanced ducted props. (5) At crulse, the drag of the Turboprop Project," NASA SP-495, 1988. large-diameter thln cowl must be Kept low while maintaining acceptable near-field noise levels. Whitlow, J.B., Jr., and Slevers, G.K., "NASA Tradeoffs between propeller and fan aerodynamic Advanced Turboprop Research and Concept deslgn methods are required to arrive at the opti- Valldatlon Program," NASA TM-lO0891, 1988. mum combination of ducted prop design parameters. At low-speed condltlons, far-field noise tn the Stefko, G.L., Rose, G.E., and Podboy, G.G., community; tlp flow separation and blade stresses "Wind Tunnel Performance Results of an wlth a short, thtn cowl at high angles of attack; Aeroelastlcally Scaled 219 Model of the PTA and reverse thrust operation are technlcal lssues Flight Test Prop-Fan," AIAA Paper 87-1893, requiring further Investigation. June 1987. (NASA TM-89917).

5. Groeneweg, J.F., and Bober, L.J., "Advanced VII. Concludln 9 Remarks Propeller Research," Aeropropulslon '87, Sesslon 5, Subsonic Propulsion Technology, Propfan technology In llttle more than a dec- NASA CP-IOOO3-SESS-5, 1987, pp. 5-123 to ade has evolved from a fuel savlng idea, through 5-152. the test of small-scale models, to the current fllght test of large-scale complete propulslon 6, Dlttmar, J.H., and Stang, D.B., "Cruise Nolse systems. In 1987 the advanced turboprop propul- of the 2/9 Scale Model of the Large-Scale slon concept was proven by three fllght programs Advanced Propfan (LAP) Propeller, SR-TA," uslng large-scale hardware. The NASA/General AIAA Paper B7-2717, Oct. 1987. (NASA ElectrIc/Boelng fllght test and the General Elec- TM-I00175). trlc/McDonnell Douglas flight test used the Unducted Fan demonstrator to prove the gearless 7. Kaza,K.R.V.,Mehmed,0., Narayanan,G.V., 14. Withers, C.C., Bartel, H.W., Turnberg, i.E., andMurthy,D.V.,"AnalyticalFlutter and Graber, E.J., "Static Tests of the Investigatlonof a CompositePropfanModel," Propulsion System - Propfan Test Assessment 28thStructures,StructuralDynamicsand Program," AIAA Paper 87-1728, June 1987. MateFlalsConference,Part2A,AIAA, NewYork,1987,pp.84-97. (NASATM-88944). 15. Graber, E.J., "Overvlew of NASA PTA Propfan Flight Test Program," Aeropropulsion '87, 8. Ernst,M.A.,andK1raly,L.J., "Determlnlng Session 5, Subsonic Propulsion Technology, StructuralPerformance,"Aeropropu]sion'87, NASA CP-IOOO3-SESS-5, 1987, pp. 5-97 to 5-122. Session2, AeropropuIslonStructuresResearch NASACP-IOOO3-SESS-2,1987,pp.2-11to 2-24. 16. Adamson, A., and Stuart, A.R., "Propulsion for Advanced Commercial Transports," AIAA 9. Nh_tfleld,D.L., Swafford,T.N.,Janus,J.M., Paper 85-3061, Oct. 1985. Mulac,R.A.,andBelk,D.M.,"Three-Dimensional UnsteadyEulerSolutlonsfor Propfansand 17. Delaney, B.R., Balan, C., Nest, H., Humenik, Counter-RotatingPropfansIn TransonicFlow," F.M., and Craig, G., "A Model Propulsion AIAAPaper87-1197,June1987. Slmulator for Evaluating Counter Rotating Blade Characteristics," SAE Paper 861715, 10.DeGeorge,C.L., "Large-ScaleAdvancedProp-Fan Oct. 1986. (LAP),"NASACR-182112,1988. 18. Celestlna, M.L., Mulac, R.A., and Adamczyk, ll. Jenness, C.M.J., "Propfan Test Assessment 3.J., "A Numerlcal SImulatlon of the Invlscid Testbed Alrcraft Flutter Model Test Report," Flow Through a Counterrotatlng Propeller," LG85ER0187, LocKheed-Georgla Co., Journal of Turbomachlnery, VoI. I08, No. 4, Marletta, GA, June 1986, NASA CR-179458. Oct. 1986, pp. 187-193.

12. A1Jabrl, A.S., and Little, B.H., Jr., "Hlgh 19. Natnauskl, H.S., and Vaczy, C.M., "Aero- Speed Wind Tunnel Tests of the PTA Aircraft," dynamic Performance of a Counter Rotating SAE Paper 861744, Oct. 1986. Prop-Fan," AIAA Paper 86-1550, June 1986.

13. A1Jabrl, A.S., "Measurement and Predlctlon of 20. Magllozzl, B., "Noise Characterlstics of Model Propeller Flow Field on the PTA Alrcraft at Counter-Rotatlng Prop-Fans," AIAA Paper Speeds of up to Mach 0.85," AIAA Paper 8?-2656, Oct. 1987. 88-0667, Jan. 1988. 100

CREW

80-- b"///, r ,FUEL & OIL 60 m J,/ (24%) / DIRECT OPERATING Y/Y/, COST, %0

40 m MAINTENANCE \

20 m DEPRECIATION, \ INSURANCE N & OTHER

YEAR YEAR 1973 1980

FIGURE 1. - FUEL COST ESCALATION AS PERCENTAGE OF DIRECT OPERATING COST. U.S. DOMESTIC TRUNK OPERATIONS. (CAB DATA) ORIGINAL P_._,.,., IS OF POOR QUALITy

90 -- /-Advanced high- / / speed turboprop E 0 o 80 -- O.

0 r- Electra -/11 '__ "_ _ 70 _E 0 _.o 60

r- I I I I I 5O I .3 .4 .5 .6 .7 .8 .9 Cruise Mach number

FIGURE 2. - EFFICIENCY TRENDS FOR TURBOPROP AND TURBOFAN ENGINES.

PROPELLER/ NACELLE • AERODYNAMICS DRIVE SYSTEM • ACOUSTICS "PROPPITCHCHANGE • STRUCTURES ,,GEARBOXES • INLETS AIRCRAFT • PUSHERENGINES TRADEOFFS

GOALS • LOWFUELCONSUMPTION INSTALLATION •LOW0PERATM6COST NOISE AND AERODYNAMICS • PASSEN6ERACCEPTANCE VIBRATION • DRAG • PASSENGERNOISE • STABILITYCONTROL • COMMUNITYNOISE

FIGURE 3, - ELEMENTS NEEDED TO DEVELOP ADVANCED TURBOPROP AIR- CRAFT. ORIGINAL PAGE I3 OF POOR QUALITY

FIGUREq. - NASA/INDUSTRYADVANCEDTURBOPROP(ATP) PROGRAR.

PTA/GULFSTREAMGII

UDFIBOEING727 UDFIMD-80 AND 578DX/MD-80

FIGURE5. - FLIGHTTESTINGOF ADVANCEDTURBOPROPS.

10 FIGURE 6. - POST-FLIGHT-TEST AREASOF ON-GOINGPROPELLERRESEARCH AT NASA LEWIS RESEARCHCENTER.

FIGURE 7. - ADVANCED PROPELLER BLADE WIND TUNNEL MODELS.

II O( 1

SR-2 u_ ..Q Oil "13 SR-1M

,c- -2 O

0 -.j

0 e'- E -4 m S -@ (b

SR-3 -6 I I 0 1 2 Efficiency gain, percent

FIGURE 8. - NOISE AND PERFORMANCE CHARACTERISTICS OF PROPELLER MODELS AT MACH 0.8, 37.5 SHP/FT 2 DISK POWER LOADING, AND 800 FT/SEC TIP SPEED.

85F

80 "------"...... _ /- $R-7A

NET EFFICIENCY, qNET 75 "_',.\ _ SR-1M -,,%-SR-6 \_- SR-2

70 J I J I i I .60 .65 .70 .75 .80 .85 .90 FREE-STREAMMACH NUMBER,MO

FIGURE 9. - SINGLE-ROTATION PROPELLER PERFORMANCE.

12 O_IG!NA!. PAGE _'1,_ OF POOi_ (_L;AI.[TY ,,oF

BLADE 160 -- SETTING MAXIMUM ANGLE, SOUND deg PRESSURE O 60.1 (DESIGN) LEVEL, [] 57.7 dB 150 -- A 63.3 / d I [ I J ,I 140 .8 •9 1.0 1.1 1.2 1.3 HELICALTIP MACH NUMBER

FIGURE 10. - SR-7 PEAK BLADEPASSINGTONE VARIATIONWITH HELICAL TIP I'tACHNUMBER.

4 I_ADES B BLA_S

k _ I" THEORY \ .%

ROIATS_REP[MD"IO_A( [:COO '_ UNSIAI_L [ U_SIAPa [

,,> .& , ,! . 5 6 .7

FRL_ SIR_ FIA(H HUMOR

FIGURE 11. - COMPARISON OF MEASURED AND CALCULATED FLUTTER BOUND- ARIES WITH SR3C-X2 PROPFAN MODEL.

FIGURE 12. - UNSTEADY THREE-DIMENSIONAL EULER CODE SOLUTION AT MACH 0.8 WITH SR-3 PROPFAN AXIS AT 40 ANGLE OF ATTACK.

13 Fill--_ /-- Spar /--- Shell / /

Ice protection--_ \ /-- Erosion

/11// protection

( /

FIGURE 13. - SCHERATIE OF THE SPAR-SHELL BLADE CONSTRUCTION CONCEPT USED FOR THE LAP.

]4 ORiC_NAL I'I'_GE I$ OF P()O_ (_U,,\[_ITY

FIGURE 14. - LAP STATIC ROTOR TEST AT WRIGHT-PATTERSON AIR FORCE BASE FACILITY.

TRANSOUCER LOCATIONS

TWO-BLADE / , TWO SR-7 / Y SR-71N MODANE, ASSEMBLIES FRANCE,WIND DELIVERED TUNNEL TO PTA ", ()

STEADYPRESSURE UNSTEADYPRESSURE INSTRUMENTATIONFOR MODANETESTING

FIGURE 15. - NASA LARGE-SCALE ADVANCED PROPELLER (LAP) PROJECT HIGHLIGHTS.

15 ORIGINAL PAGE I$ OF POOR QUALITY,

___J

FIGURE 16. - ROHR PTA PROPULSION SYSTEM STATIC TEST.

16 ORIGINAL PAGE IS OF POOR (_UALITY

FIGURE 17. - PROPFANTEST ASSESSMENT(PTA) FLIGHT TESTING.

--- PREDICTED <> WIND TUNNEL(LEWIS8x6) O PTAFLIGHT(MACH0.8; 35 000 ft) 150

140 BLADE PASSING TONE, dB

130

/ / PROPFANPLANE \ / UPSTREAM= I _ DOWNSTREAM \ ¢ I \ 120_ I I I Y 1.0 .5 0 -.5 -1.00 FORE AND AFT LOCATIONS IN PROPELLER DIAMETERS

FIGURE 18. - COMPARISON OF PTA FUSELAGE SURFACE NOISE WITH PRE- DICTION AND WIND TUNNEL DATA.

17 FIGURE19. - UNDUCTED FAN ENGINE (UDF).

POLYURETHANEFILM SURFACEPROTECTION %

NICKEL LEADING EDGE PROTECTION-,,

CARBON-FIBER/2Oe/o FIBERGLASSAIRFOIL

_- TITANIUM SPAR

FIGURE 20. - GE UDF BLADE CONSTRUCTION,

017, POOR QUALITY. FIGURE 21. - UDF COUNTERROTATION PROPELLER MODEL IN NASA LEWIS 8- BY 6-FOOT WIND TUNNEL.

FIGURE 22. - WIND TUNNEL MODELS OF UDF COUNTERROTATION BLADE CON- FIGURATIONS.

19 90--

PERCENT OF DESIGN LOADING 85-- 80

NET 80-- EFFICIENCY, PERCENT 120 -_"

75-- "N \

?o I I I I I I .60 .65 .70 .75 .80 .85 .90 MACH NUMBER

FIGURE 23. - FT-A7 UDF RODEL BLADE PERFORHANCE RESULTS. 8 + 8 BLADE CONFIGURATION, 780 FT/SEC TIP SPEED.

FIGURE 2q. - THREE-DIRENSIONAL EULER ANALYSIS OF UOF PRESSURE DISTRIBUT[ON.

Oh. .... '_',AL PACE IS OF POOR QUALITY 2O FIGURE 25. - UNSTEADY THREE-DIMENSIONAL EULER SOLUTION OF PRES- SURE FIELD JUST DOWNSTREAM OF UDF AT ONE INSTANT IN TIME.

GE STATIC TEST AT PEEBLES, OHIO

BOEING 727 FLIGHT DOUGLAS MD-80 TEST FLIGHT TEST

FIGURE 26. - NASA/GENERAL ELECTRIC UNDUCTED FAN ENGINE GROUND AND FLIGHT TESTING.

21 130 --

120-- BLADE PASSAGE TONE URCE SPL 0 /" deg PERCENT 110 -- " MXCL AT / 155-FT / 0 58.5155.7 100 8×6 WIND TUNNEL SIDELINE, / dB ,_ 59.0153.3 76 100 -- " 59.2/52.8 77 LEARJET O 59.3/52.9 87 { PREDICTED I ------I 58.7157.7I 1pO I I 9020 40 60 80 100 120 140 SIDELINE ANGLE, deg

FIGURE 27. - UDF FUNDAMENTAL TONE DIRECTIVITY MEASURED IN FLIGHT, COMPARED WITH SCALED MODEL DATA AND PREDICTION. MACH 0.72 DE- SIGN SPEED.

FIGURE 28. - HAMILTON STANDARD CRP-XI COUNTERROTATION PROPFAN MODEL IN UTRC WIND TUNNEL.

ORIGINAL PACE IS 22 OF POOR QUALITY FIGURE 29. - NASA LEWIS PREDICTION CODE SIMULAIES LEADING EDGE AND TIP VORTEX SHEDDING AT OFF-DESIGN MACH 0.2 CONDITIONS FOR CRP-XI PROPFAN.

--_ CRP ---C)--- SRP+3 dB INTERACTION NOISE

120 --FORWARD --r'-I PLANE OF AFT

PI1ESSURE 100 LEVEL, SOUNDdB 110 i

8o I I I I I 0 1 2 3 4 5 1 2 3 4 5 0 1 2 3 4 5 HARMONIC OF BLADE-PASSAGE FREQUENCY

FIGURE 30. - CRP-XI MODEL NOISE TEST DATA COMPARED WITH ADJUSTED SINGLE-ROTATION PROPFAN DATA.

23 OF POOR QUALITY

FIGURE 31. - 1987 COLLIER TROPHY AWARDED TO LEWIS RESEARCH CENTER AND THE NASMINDUSTRY ADVANCED TURBOPROP TEAM.

Report Documentation Page Nallorbal Ae=_)f_auh( _ and Space Adm.,_shahun

1. Report No. 2. Government Accession No, 3, Recipient's Catalog No. NASA TM-100929

4. Title and Subtitle 5. Report Date

NASA/Industry Advanced Turboprop Technology Program 6. Performing Organization Code

7. Author(s) 8. Performing Organization Report No.

Joseph A. Ziemlanskl and John B. Whltlow, Jr. E-4198

10. Work Unit No.

535-03-01 9. Performing Organization Name and Address 11, Contract or Grant No. National Aeronautics and Space Admlnistr&tlon Lewis Research Center

Cleveland, Ohlo 44135-3191 13. Type of Report and Period Covered

12. Sponsoring Agency Name and Address Technical Memorandum

National Aeronautics and Space Administration 14. Sponsoring Agency Code Washlngton, D.C. 20546-000]

15. Supplemenla_ Notes Prepared for the 16th Congress of the International Council of Aeronautlcal Sciences, Jerusalem, Israel, August 28 - September 2, 1988.

16. Abstract

Experlmental and analytlcai effort shows that use of advanced turboprop (propfan) propulslon instead of conventional turbofans in the older narrow-body airline fleet could reduce fuel consumption for thls type of aircraft by up to 50 percent. The NASA Advanced Turboprop (ATP) program was formulated to address the key technologies requ|red for these new thin, swept-blade propeller concepts. A NASA, industry, and unlverslty team was assembled to develop and valldate applicable new design codes and prove by ground and f11ght test the viability of these new propeller concepts. This paper presents some of the history of the ATP project, an overview of some of the issues and summarizes the technology developed to make advanced propellers viable In the hlgh-subsonic cruise speed application. The ATP program was awarded the presttglous Robert J. Collier Trophy for the greatest achievement in aeronautics and astronautics In America in 1987.

17 Key Words (Suggested by Author(s)) 18. Distribution Statement

Turboprop Unclasslfled - Unllmtted Propeller Subject Category 07 Commercial aircraft Fuel conservation

19. Security Classif (of this report) 20. Security Cla_if. (of this page) 21. No of pages 122. Price" Unclassified Unclasslfled 26 A03

NASA FORM 1626 OCT 86 "For sale by the National Technical Inlormation Service, Springfield, Virginia 22161

National Aeronautics and SECOND CLASS MAIL Space Administration IIIIII Lewis Research Center ADDRESS CORRECTION REQUESTED Cleveland. Ohio 44135 om:_ seeing8 Pqmlly kx I_va_ _ 11300 PoMIIge Ind Fees Pldd Nalk3nalA_ronautt_ Sio_e .,_rninistrm_ N.'_.Sk461