14
NASA AVSCOM Technical Memorandum105665 Technical Report 92-C-011 AIAA-92-3365 AD-A252 379 Advanced Rotorcraft Transmission (ART) Program Summary DTIC ELECTE TL.KS JUL 0 2 1992 Propulsion Directorate A U.S. Army Aviation Systems Command Lewis Research Center Cleveland, Ohio and J.G. Kish Sikorsky Aircraft Stratford, Connecticut IThi document has been approved for public elease and sale; its disthibution is urtlijimited. Prepared for the 28th Joint Propulsion Conference and Exhibit cosponsored by the AlAA, SAE, ASME, and ASEE Nashville, Tennessee, July 6-8, 1992 92-17102 US ARMY AVIATION PJASASYSTEMS COMMAND

Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

NASA AVSCOMTechnical Memorandum105665 Technical Report 92-C-011AIAA-92-3365

AD-A252 379

Advanced Rotorcraft Transmission(ART) Program Summary

DTICELECTE

TL.KS JUL 0 2 1992

Propulsion Directorate AU.S. Army Aviation Systems CommandLewis Research CenterCleveland, Ohio

and

J.G. KishSikorsky AircraftStratford, Connecticut

IThi document has been approved

for public elease and sale; itsdisthibution is urtlijimited.

Prepared for the28th Joint Propulsion Conference and Exhibitcosponsored by the AlAA, SAE, ASME, and ASEENashville, Tennessee, July 6-8, 1992

92-17102 US ARMYAVIATION

PJASASYSTEMS COMMAND

Page 2: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

ADVANCED ROTORCRAFT TRANSMISSION (ART) PROGRAM SUMMARY

T.L. KrantzPropulsion Directorate

U.S. Army Aviation Systems CommandLewis Research Center

Cleveland, Ohio 44135-3191

and

J.G. KishSikorsky Aircraft

Stratford, Connecticut 06601-1381

Abstract gram managed jointly by the Army and NASA.The goal of the ART Program is to develop and

The Advanced Rotorcraft Transmission demonstrate the technologies needed for rotorcraft(ART) Program was initiated to advance the transmissions that will be 25 percent lighter andstate-of-the-art for rotorcraft transmissions. The 10 dB quieter than the state-of-the-art while in-goal of the ART Program was to develop and creasing the mean-time-between-removal to a mini-demonstrate the technologies needed to reduce mum of 5000 hr. The ART program consists oftransmission weight by 25 percent and reduce four major contracts. Project status reviews fornoise by 10 dB while obtaining a 5000-hr mean- all four contracts are documented in Refs. 1 to 8.time-between-removal. This'paper summarizes the This article will overview the experiments andresearch done at Sikorsky Aircraft under the ART analysis phase of the program conducted byProgram. A split path design was selected as best Sikorsky Aircraft.able to meet the program goals. Key componenttechnologies needed for this design were identified, Technology Needs and Research Planstudied, and developed. Two of these technologiesare discussed in detail: the load sharing of split The vehicle that defined the transmissionpath designs including the use of a compliant elas- requirements was an Advanced Cargo Aircrafttomeric torque splitter and the application of a (ACA) heavy lift helicopter with a projectedhigh ratio, low pitch line velocity gear mesh. 80 000-lb gross weight, 25 000-lb payload, andDevelopment of an angular contact spherical roller 500-kn mission radius. Many drive system config-bearing, transmission error analysis, and fretting urations were identified and evaluated against thefatigue testing are discussed. The technologies for ART program goals and the requirements of thea lightweight, quiet, and reliable rotorcraft trans- selected ACA helicopter. The configuration selectedmission have been demonstrated. as best able to meet the program goals was the split

path transmission illustrated in Fig. 1. TheIntroduction important design parameters and resulting charac-

teristics for this transmission are summarized inThe next generation of rotorcraft vehicles will Table 1. An evaluation of this design predicted

require drive systems that are lighter, quieter, that it would meet the goal of a 10-dB noiseand more reliable. These improvements are needed reduction, be 23 percent lighter than the baseline,to increase the vehicle's payload and performance, and have a 3890-hr mean-time-between-removal.improve passenger comfort and safety, lower oper- Although the reliability for this design is less than ..........ating costs, and improve readiness. The Advanced the 5000-hr ART program goal, the 3890-hr figureRotorcraft Transmission (ART) Program was ini- is approximately four times greater than the base-tiated to address these needs and advance the line and was a difficult goal to achieve for a heavystate-of-the-art for rotorcraft transmissions. The lift vehicle. The impact of the advanced transmis-ART program is an Army funded research pro- sion on vehicle perf aance and cost was studied.

Cop~rnghl 1992 by the Amcrican Institute of Aeronjutic,;.ind A~ir.i.uiwc. Inc No copyright is as.ened in the

United Siate under Title 17. U.S CodIe. The U.S. Go~ern. -

nient hi% a royalty-free Iicc,.. ti) eierckc~ all nghts underihe t ,'t ri .lt c m . d herein tfor .ternilntil purpseA

Page 3: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

One of the results of that study was that this deflections as the full scale design. Also, componenttechnology would reduce the 35-year life cycle cost life and reliability are the same notwithstandingof a 600-vehicle fleet by $1.7 billion dollars. A material allowables based on size effects. How-more detailed discussion of the design and how it ever, the sliding velocities of the half-scale gearwas selected can be found in Refs. 1, 2, and 10. meshes are half of the full scale components.

The significant characteristics of the selected The test program for the half-scale gearboxtransmission design are the split path configuration included the following testing:and several key component technologies including:

(1) No load lubrication survey

(1) Composite gearbox housing (2) Gear pattern development and verification(2) Composite drive shafts (3) Static and dynamic surveys(3) High speed spring clutch (4) 200-hr endurance test(4) High hot hardness steel (5) 200-hr, 120-percent overtorque test(5) Angular contact spherical roller bearings(6) High reduction ratio gear mesh at the The test facility uses two identical gearboxesoutput stage in a back-to-back, closed loop type arrangement.

(7) Wide face width, high contact ratio double More detailed descriptions of the developmenthelical mesh plan, analytical studies, and test facilities have

(8) Split path load sharing methods been documented previously. 1 ,2 '9 "11

(9) Topologically ground tooth profilesResearch Results

Items 4 to 9 were selected for study, devel-opment, and demonstration in the ART program. The experiments and analysis to study andA component technology development plan was develop the key component technologies have beenestablished that consisted of research areas that completed. The technologie3 for a lightweight,were key to the successful advanced transmission quiet, reliable rotorcraft transmission have beendevelopment, demonstrated. Some results of the research, fully

documented in Ref. 10, are summarized here.Experiments were used to study the fretting

characteristics of a high hot hardness steel. A Angular Contact Spherical Roller Bearingcode was developed for analysis and design ofangular contact spherical roller bearings with A computer code, SASHBEAN, was developedceramic rollers, and bearings were built and tested to assist in the design and analysis of angular con-in a special test rig. An analytical study investi- tact, spherical roller bearings. For a helicoptergated how to optimize the gear mesh parameters to application, the angular contact spherical rollerminimize transmission error. A half-scale gearbox bearing must operate at significantly higher speedand test facility were built and tests conducted to than in traditional applications. Used on the heli-study the high contact ratio double helical gear copter gearbox input section, it replaces the currentmesh, split path load sharing, transmission error, state-of-the-art ball and roller bearing combination.topological tooth profiling, and design durability.The half-scale gearbox, shown in Fig. 2, duplicates The SASHBEAN code was used to design athe final two stages of the split path gearbox for bearing that would support the spiral bevel pinionone engine path but at one-half geometric scale. of the ART transmission. Test bearings wereOne-half scale was chosen to reduce fabrication manufactured with two types of rollers; steel andcosts of the hardware and test facility. By using ceramic. Bearings were tested both at nominalthe same speed as the full scale gearbox, red-acing operating conditions and at loss of lubricant condi-the power by one-eighth, and scaling thi- compo- tions. The results of endurance testing done atnents by one-half, the tested components experi- nominal operating conditions showed that theence the same bending stresses, Hertz stresses, and design was durable. Figure 3 illustrates the results

2

Page 4: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

of loss of lubricant testing for the ceramic bearing An analysis was conducted to study the toothcompared to an analytical prediction of the maxi- load distribution and transmission error of bothmum bearing temperature. The actual maximum the second stage high contact ratio spur mesh andtemperature of the tested bearing was somewhat the final stage double helical mesh (Fig. 4). Thegreater than the experimentally measured outer analysis indicated that changes could be made torace temperature. The slopes of the plots of ana- both meshes that would improve the load distribu-lytical data and experimental data are similar. tion. Also, by changing the double helical meshThe cage of the bearing failed 21 min after loss of from the initial herringbone design with alignedlubricant. Further design optimization is required left and right hand helixes to one in which the twoto increase the loss of lubricant survivability time halves of the double mesh are staggered, the pre-to achieve the desired target of 60 min. dicted transmission error could be reduced by

about 50 percent.Fretting Fatigue

Split Path Load SharingFretting fatigue experiments were conducted

for a high hot hardness steel, Pyrowear 53, that is A split path design at the final stage of thea candidate material for gears. Two types of spec- gearbox is desired to obtain the advantages ofimens were tested, one plated with thin dense sharing the torque among multiple pinions, as ischrome and one without plating. It was found done in a conventional planetary stage, while alsothat the plating did not significantly change the obtaining a larger reduction ratio than is possiblefretting fatigue life, and the Pyrowear 53 material for a planetary design. A split path design with ahas a fretting fatigue life similar to 9310 steel, large reduction ratio at the final gear stage weighs

less than a planetary design. However, the splitGear Mesh Technology path design must have good load sharing between

the two power paths. If the torsional stiffnessThe tooth profiles of the half-scale gearboxes between the torque splitting gear and combining

were modified to account for deflection under load. pinion is high compared to the load carried, andThe types of modifications used have become feasible manufacturing errors are considered, one of theonly recently with the introduction of computer two power paths will carry more than half of thenumerically controlled gear grinding equipment total power. One method to reduce the load shar-capable of lead and root modification. A finite ing error is to properly index the torque splittingelement analysis of the gearbox housing, shafts, gear and pinion while maintaining precise manu-and gears was conducted to determine tooth deflec- facturing tolerances for machining and assembly.tions. The teeth of the gears were coated or plated This is feasible using today's manufacturing capa-with silver, copper, gold, black oxide, and blueing bilities but may not be the optimal solution. Otherto aid in visualization of the tooth contact pat- methods have been proposed, including an axiallyterns. The silver and copper plate were most effec- floating quill shaft, torsionally compliant shafts,tive for the tooth development process. Strain balancing mechanisms, and laterally compliant bear-gage measurements across the face of the teeth ing supports. These methods may significantly alterwere also used to evaluate tie load distributions, the vibration properties of the gearbox. The twoThe gear teeth patterns and strain measurements methods tested in this program were the use ofwere analyzed after running under load. The teeth precise tolerances and the use of a special device towere contacting across the full face width and the provide torsional compliance.strain readings showed good load distribution suchthat further grinding for pattern development was The device used to provide torsional compli-not required. The capability to obtain proper ance between the torque splitting gear and pinion,tooth load distributions without experimental trial an elastomeric load sharing device, is shown inand error was demonstrated. Fig. 5. Alternate thin layers of nitrile rubber and

3

Page 5: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

steel are stiff in the direction perpendicular to the consistent with the lower transmission errors thatlaminates but allow large deflections in the parallel were measured. The elastomers provide a highdirection. The laminates are located at an angle degree of damping not normally found in helicop-to the axis of rotation on two halves that are bolted ter transmission components. More research andtogether. The laminates are compressed as the two study is needed to state conclusively whether thehalves are drawn together during assembly. The reduction in transmission error and noise was themagnitude of the compressive preload force is result of added compliance, added damping, orimportant because it controls the magnitude of the both.friction force between the spur gear and isolatorhalves. The torque on the gear is transmitted by The transmission error measurements andthe frictional forces between these surfaces. Larger observed noise difference makes the torsional isola-preloads create a larger torque capacity, but if the tor an attractive option. The isolator also wouldpreload is too large the material will yield and fail. have the advantage of relaxing costly, precise manu-The elastomeric torsional isolator shown is about facturing tolerances. However, an unexpected andfive times more compliant in torsion than an all undesirable characteristic of the present design wassteel assembly of the same dimensions. The all discovered during endurance testing. The torquesteel versions were precisely assembled to properly split between the two power paths changed slightlyindex the two gears on the common shaft. Both during the testing. It was later verified that thedesigns were tested. spur gear and isolator halves that make up the iso-

lator assembly had slipped with respect to oneBoth methods for torque splitting proved to another, which affected the torque split. The

be feasible. The mean torque carried by the dual nitrile rubber and steel have different thermalpaths was equal within 5 percent for both methods. expansion rates. For the isolator geometry used,The relationship between the torsional compliance of as the temperature of the assembly increased, thethe dual paths and the precision required in assem- load in the laminated area increased. A test wasbly was verified by the experiments. Although conducted by the Lord Corporation, the manufac-both methods were feasible, the behavior observed turer of the load sharing device, where a steel andduring the two tests was different. nitrile rubber laminate was compressed by a prede-

termined amount and held at that dimension. TheThe transmission error measured at the heli- test specimen was subjected to temperature cycles

cal pinion for each of the two tests at identical while being held at a constant compressed dimen-operating conditions is shown in Fig. 6. One sig- sion, and the compressive load in the laminate wasnificant difference is that the magnitude of the measured. Figure 6 is a plot of the test results.maximum peak-to-peak transmission error is more Note that during the first cycle over a time scalethan 50 percent less when using the elastomeric of 600 min, with the temperature and deflectionisolator. Also, the speed at which the maximum held constant, the compressive load decreased bytransmission error occurred is different. One about 30 percent. Using the results of these tests,would expect that the large compliance of the the effect of the temperature cycles experienced byisolator would shift the resonance condition to a the elastomeric torque splitter during testing inlower frequency. The maximum response, how- the half-scale gearbox was calculated. It wasever, occurred at a higher speed with the elasto- found that under repeated heating and cooling, themeric isolator compared to the steel assembly. It preload had decreased significantly enough tois likely that the two-peak responses are two differ- reduce the torque capacity below the designent mode shapes being excited within the speed torque, and slippage occurred during testing.range shown. Furthermore, the assembled preload at room tem-

perature could not be adjusted to compensate forAnother observed difference in behavior was the desired operating temperature range without

that the audible noise produced while testing the either reaching a yield stress at the highest tem-elastomeric isolator was significantly less compared perature or losing needed torque capacity at theto that while using the all steel assemblies. This is lowest temperature. The elastomeric torque

4

Page 6: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

splitter, in its current configuration, does not have The operating speed and load of the test andthe needed torque capacity over the entire range of dummy gearboxes are identical since they are con-temperatures required for a fielded system. How- nected in a back-to-back arrangement. The direc-ever, an alternate temperature compensating design tion of rotation of the dummy gearbox is oppositethat matches the compliance and damping of the to that of the test gearbox since the dummy gear-present design may be possible. box is a driven compared to driving for the test

gearbox. However, for the two gearboxes theHigh Reduction Ratio Double Helical Gear Mesh loaded side of the tooth, load magnitudes, and

load directions are identical. The only essentialDuring a scheduled 150-hr visual inspection of difference between the operating conditions of the

the 200-hr endurance test of the half-scale gearbox, test and dummy gearboxes was the separate lubri-surface distress was discovered on a gear tooth. cation systems and operating temperatures. TheExamination through the inspection port on the test gearbox inlet oil was preheated to obtain antest gearbox revealed a distressed surface condition average oil outlet temperature of approximatelyon one of the two double helical pinions. The dis- 190 *F whereas the dummy gearbox had no facilitytressed area was typical of a spalling type of sur- oil heating system. The dummy gearbox oil tem-face distress. A decision was made to continue perature ran at approximately 120 *F oil out.testing and monitor the spall to determine propa- With the DOD-L-85734 oil used in both gearboxes,gation rates. An advantage of a double helical the viscosity of the oil in the test gearbox wasmesh with such a high contact ratio as in the ART estimated to be 5.8 cS and in the dummy gearboxgearbox (Fig. 4) is that in the event of a surface to be 16.5 cS. The difference in viscosity isdistress such as that experienced, adjacent teeth because of the temperature difference.will carry the load for a considerable period oftime before the tooth becomes totally deteriorated. The evidence of a spalled tooth and polishedThe 200-hr test was completed and the gearbox surfaces, indicating a high degree of working in thedisassembled for a closer look at the spall. test gearbox, and no spalling with much less pol-

ishing and wear in the dummy gearbox, along withThe original spall had increased in size from the factor of differences in lubrication led to an

the estimated 0.09 to 0.23 by 0.15 in: in diameter, investigation of poor lubricant film thickness orIn addition, another tooth on the same pinion had elastohydrodynamic (EHD) film thickness effectsa smaller spall measuring approximately 0.23 by as the cause of the surface distress. Most of the0.05 in. The general locations and conditions of work in this area has been related to bearing oper-the spalled areas are illustrated in Fig. 8. ation. The theory has been extended to gears by

several researchers. An investigation of the litera-There was no evidence of other surface dis- ture showed that two basic methods of calculation

tress conditions on the gears of the test or dummy exist for EHD film thickness in gears. The Cheng,gearboxes. There was a considerable difference in modified Gruben analysis 12 is easier to use becauseappearance of the teeth of the double helical pin- it makes use of a single lubricant property, theions of the test gearbox compared to the pinions of lubricant parameter. The more popular Dowsonthe dummy gearbox. The teeth of the test gear- Higginson " procedure requires the use of thebox double helical pinions showed a high degree of lubricant absolute viscosity in microreyns and thepolish whereas the dummy gearbox showed only a pressure viscosity coefficient in inches squared persmall degree of polish and had a surface finish pound.close to the manufactured condition. The surfacefinish of the teeth were measured using a Taylor After the film thickness is calculated, theHobson surface finish measuring machine with a ratio of film thickness to the composite surface0.030-in. cutoff. The surface finish of the test roughness, A, is found. When A is equal to 1, thepinion was measured 10 to 17 Ain. rms while the film thickness is equal to the surface roughness.surface finish of the dummy pinion was measured Thus the surface is just immersed in oil and wear isas 21 to 27 pin. rms. minimal. As A increases over 1, the film thickness

5

Page 7: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

is larger than the surface asperities and the chance flight hours with calculated A ratios in the rangeof surface distress becomes lower with higher A. On of 0.36 to 0.46. The allowable A ratio was inves-the other hand, when A falls below 1, the peaks of tigated by Wellauer and Holloway"8 in the 1970's.the surface finish will be sticking out of the oil Unfortunately, little work has been accomplishedlocated in the valleys. When A gets very low, as since. Moreover, the Wellauer and Holloway workin the ART half-scale testing which was done with was based substantially on test data for througha calculated A in the range of 0.2, the gears are hardened gears along with a limited amount of dataoperating in the boundary lubrication regime, and for carburized gears. The allowable A ratio versusthe chance of surface distress increases, probability of distress curves that are shown in

many references are taken from the original WellauerThe film thickness is a function of load, speed, and Holloway test data. Carburized gears have a

and lubricant properties. The most influential much lower probability of EHD related surface dis-parameter in calculated film thickness is speed, or tress than through hardened gears, as evidenced bymore precisely pitch line velocity. In the ART the limited testing done by Wellauer and Hollowayprogram, the half-scale gearbox operates at one- and also by comparing data for successful produc-eighth power and the same speed as the full scale tion helicopter gear output stages to the allowablegearbox, which duplicates the tooth fatigue bend- curve. Figure 10, taken from AGMA Ref. 17,ing stress, tooth compressive (Hertz) stress, gear shows a high probability of distress at A = 0.4tooth deflections, shaft fatigue bending stress, shaft and 2000 ft/min, which is typical for a productiondeflection, housing stress, and housing deflections helicopter output stage which operates successfullyuf the full scale gearbox. However, although most without surface distress. There is a need for moreimportant parameters match, the half-scale gearbox work to be done in this area for carburized gears.pitch line velocity is one-half of the full scale gear- The fact that no allowables exist for carburizedbox. In the half-scale ART test, the pitch line gears is perhaps a reason why aerospace designersvelocity of the double helical pinion was 817 ft/min have been reluctant to use EHD analysis as aat 100 percent speed. In comparison, the full scale design tool.ACA ART transmission would have twice thepitch line velocity, or 1634 ft/min. Production Split path designs have pitch line velocitiesSikorsky helicopter output stage pitch line veloci- which are on the low side of current helicopterties are in the range from 1300 to 1900 ft/min. experience. This is inherent in the design if full

advantage is taken of the weight benefit from max-From the above discussion, it is seen that the imum reduction ratio at the final output stage. If

half-scale ART test gearbox has an artificially pitting resulting from low EHD oil film becomes ainduced low pitch line velocity compared to what design driver or a source of gearbox removal, thereit would be in the full-scale gearbox or compared is an easy solution available. The solution is toto other production helicopter output stages. use an oil having a higher viscosity than currentlySince velocity is the most influential parameter in used in helicopter transmissions. Oil choice has beenthe calculation of EHD film thickness, the ART largely driven by military requirements to use ahalf-scale test gearbox has an artificially low EHD common oil in the turbine engine and transmission.film thickness. Figure 9 is a plot of calculated If this thinking can be reversed, low EHD filmsEHD film thickness for the half-scale gearbox as a will not be a problem for split path transmissions.function of temperature. As seen, the film thick-ness at the 120 *F temperature of the dummy Conclusionsgearbox was approximately twice that of the testgearbox operating at 190 *F. The technologies needed for the next genera-

tion rotorcraft transmission have been identifiedGears are capable of operating in the bound- ahd research conducted in the Advanced Rotorcraft

ary lubrication regime for many cycles, as evidenced Transmission program. Research has been con-by typical helicopter output stages which have ducted to develop and demonstrate the technologiesoperated without surface distress for millions of required for a transmission that has a 5000-hr

6

Page 8: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

mean-time-between removal and also is 10 dB conditions where EHD related surface failures willquieter and 25 percent lighter than state-of-the art occur are not precisely established for carburizedtransmission. The following results were obtained gears.in the ART program research.

References1. A split path design was developed that met

the program goals for less noise and less weight 1. Kish, J., "Advanced Rotorcraft Transmissionwhile obtaining a 3890-hr mean-time-between (ART)-Program Status,' AIAA Paperremoval. Two of the new, significant technologies 91-1909, June 1991.needed for this design are split path load sharingand a high reduction ratio, low pitch line velocity 2. Kish, J., "Advanced Rotorcraft Transmissionmesh at the final output stage. (ART) Program Review," Rotary Wing Pro-

pulsion Specialists' Meeting; Proceedings,2. An angular contact spherical roller bearing American Helicopter Society, Alexandria, VA,

was developed and demonstrated. This bearing, 1990, 11 p.with ceramic rollers, ran for 21 min without oilbefore failure, about one-third of the desired 3. Bossier, R.B., and Heath, G.F., "Advanced60-min target. Rotorcraft Transmission (ART) Program

Status," AIAA Paper 91-1906, June 1991.3. Thin dense chrome plating does not signifi-

cantly change the fretting fatigue life of the high 4. Bossler, R.B., and Heath, G.F.: "Advancedhot hardness gear steel Pyrowear 53. The fretting Rotorcraft Transmission (ART) Programlife of the material without plating is similar to Status," Rotary Wing Propulsion Specialists'9310 steel. Meeting; Proceedings, American Helicopter

Society, Alexandria, VA, 1990, 14 p.4. Two methods for achieving an equal torque

split in a split path gearbox were tested: precision 5. Henry, Z.S., "Preliminary Design and Analysistolerances and indexing at assembly and a compli- of an Advanced Rotorcraft Transmission,"ant elastomeric torque splitter. Both methods AIAA Paper 91-2018, June 1991.were demonstrated and are feasible.

6. Henry, Z.S., "Preliminary Design and Analysis5. The transmission error and noise from the of an Advanced Rotorcraft Transmission,"

gearbox were reduced by using the elastomeric Rotary Wing Specialists' Meeting; Proceed-torque splitter. ings, American Helicopter Society, Alexan-

dria, VA, 1990, 14 p.6. The elastomeric torque splitter in its current

configuration does not have the needed torque capac- 7. Lenski, J.W., and Valco, M.J., "Advancedity for the entire range of temperatures required for Rotorcraft Transmission (ART) Program-a fielded system. Boeing Helicopters Status Report," NASA

TM-104474, 1991.7. A surface distress failure occurred that is

related to the low EHD film thickness of the low 8. Lenski, J.W., "Boeing Helicopters Advancedpitch line velocity at the final stage mesh of the Rotorcraft Transmission (ART) Programhalf-scale gearbox. Surface distress is not likely in Status," Rotary Wing Propulsion Specialists'the full scale gearbox because the pitch line veloc- Meeting; Proceedings, American Helicopterity is double that tested. Society, Alexandria, VA, 1990, 15 p.

8. The calculation techniques for low EHD filmthickness have been developed, but the operating

7

Page 9: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

9. Mitchell, Jr., G.D., "Manufacturing 14. Errichello, R.E., "The Lubrication of Gears,Techniques--Split Torque Path Helicopter Part 2," Gear Technology, Vol. 8, May/JuneTransmission," AHS Paper A-91-47130-5000, 1991, pp. 18-22.1991, Presented at American Helicopter Soci-ety 47th Annual Forum, Phoenix, AZ, May 15. Errichello, R.E., "The Lubrication of Gears,6-8, 1991, NASA Contract NAS3-25423. Part 3," Gear Technology, Vol. 8, July/Aug.

1991, pp. 14-22.10. Kish, J., "Sikorsky Helicopter Advanced Rotor-

craft Transmission Project. Final Report," 16. Errichello, R.E., "The Lubrication of Gears,To be published as NASA CR- , 1992. Part 4," Gear Technology, Vol. 8, Sept./Oct.

1991, pp. 18-25.11. Hochmann, D., Smith, D., Thomas, J., and

Houser, D.R., "Transmission Error and Load 17. American National Standard Institute/AmericanDistribution Analysis of Spur and Double GearManufacturers Association, "FundamentalHelical Gear Pair Used in a Split Path Heli- Rating Factors and Calculation Methods forcopter Transmission Design," Presented at Involute Spur and Helical Gear Teeth,"AHS Technical Specialists Meeting, Philadel- ANSI/AGMA 2001-B88, Sept. 1988.phia, PA, Oct. 15-16, 1991.

18. Wellauer, E.J., and Holloway, G.A., "Applica-12. Mobil EHL Guidebook, Third Ed., Mobil Oil tion of EHD Oil Film Theory to Industrial

Corp., Fairfax, VA, 1981. Gear Drives," Journal of Engineering forIndustry, Vol. 98B, No. 2, May 1976,

13. Errichello, R.E., "The Lubrication of Gears, pp. 626-634.Part 1," Gear Technology, Vol. 8, Mar./Apr.1991, pp. 18-26.

TABLE I.-TRANSMISSION DESIGN PARAMETERS AND SUMMARY

Engine, rpm ......................................... 15 000Main rotor, rpm ........................................ 130Overall MGB, reduction ratio ............................. 115.4Main gearbox input, hp ................................ 16 030Main gearbox input, hp (limit) . ......................... 32 060Engine, hp (OEI) ...................................... 6209Main rotor, hp ....................................... 14 825Main rotor, hp (limit) ................................. 29 650Tail rotor, rpm ......................................... 580Tail rotor, hp (gears) .. ................................. 2400Tail rotor, hp (limit .................................... 4328Drive system weight, lb ................................ 7870

8

Page 10: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

(a) Isometric view.

#1 engine #3 engine

1270 rpm

15 000 rpm in

C. L.(b) Schematic plan.

Figure 1 .- Advanced cargo aircraft split path transmission.

Page 11: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

650 Hp output/atl130rpm

Input stage torque splitter-high contactratio spur mesh

L6 5 0 Hp input L Final sage UTEHOOGE

at 4933 rpm double SIKORSKY

helical mesh A AIR CRAPT

Figure 2.-ART 1/2 scale split torque gearbox.

800

700

~600Failure of

Ei-400

300

2000 10 20 30 40

Time following loss of lubicant, min

Figure 3.-Loss of lubicant survivability, angular contact, Figure 4.-High reduction ratio double helical gear mesh.ceramic spherical roller bearing.

10

Page 12: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

00

Isolator(a) Steel.half (TYP) E

~ 800

L60-Spline with 4 -

every ther0. .4801-

tooth missing & .32

-18 3/8 dia bolts -. 6w...i

friction drive 2400 2920 3440 3460 4480 5000

Shim Shaft speed, rpm

(b) Elastomer.Figure 6.-Transmission error vs rpm for upper helical mesh with

and without elastomeric load sharing device installed.

(a) Cross section.

5X1 03

1 250 IF Constant temperature

- - - Cooling/heating cycle

4

250 IF

3 * 250 IFS 85 IF '

(b)~~~~ Spure gerad1C e2

AI 2 ,Cce ' Cyl 2

I * ma Cycle 3()Sugeradisolator component.1

Figure 5.-Elastomeric, torsionally compliant load sharing device. 85 IF89 IF0I~ 1 870 I

0 1000 2000 3000 4000 5000Time, min

Figure 7.-Measured laminate compressive load vs time withtemperature cycling and constant deflection.

....1..

Page 13: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

(a) Location cf distress.

(b) Firs' distress area. (c) Second distress area.Figure 8.-Surface distress faiiure of double nelical pinion.

2.0- Probability of

- 2 Dowson Higginson analysis wear related10 distress,

c .6 -- percent

Z .2

o ART dummy(No pall) .04

ART test 0 .04

4, (Spall) U) .02- ~~~~~~~~11 1 10 Iii 11fIiIll l

0 .01o 60 100 140 180 220 50 100 5001000 5000 10000 50000

Temperature IF Pitch line velocity, FPM

Figure 9.-Calculated specific film thickness, \, vs temperature Figure 10.-Probability of Wear distress, percent.for ART 1/2 scale gearbox at full power.

12

Page 14: Advanced Rotorcraft Transmission (ART) Program Summary DTIC · helical mesh plan, analytical studies, and test facilities have (8) Split path load sharing methods been documented

Form ApprovedREPORT DOCUMENTATION PAGE OMB No. 0704-0188

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources.gathering and maintaining the data needed, and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of thiscollection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for information Operations and Repots, 1215 JeffersonDavis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188). Washington. DC 20503

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

1992 Technical Memorandum4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Advanced Rotorcraft Transmission (ART) Program Summary

WU-505-63-366. AUTHOR(S) 1L162211A47A

T.L. Krantz and J.G. Kish

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATIONNASA Lewis Research Center REPORT NUMBERCleveland, Ohio 44135-3191and E-7027Propulsion DirectorateU.S. Army Aviation Systems CommandCleveland, Ohio 44135-3191

9. SPONSORING/MONITORING AGENCY NAMES(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING

National Aeronautics and Space Administration AGENCY REPORT NUMBER

Washington, D.C. 20546-4141f NASA TM-105665andU.S. Army Aviation Systems Command AVSCOM-TR-92-C-01 ISt. Louis, Mo. 6312G-1798 AIAA-92-3365

11. SUPPLEMENTARY NOTES

Prepared for the 28th Joint Propulsion Conference cosponsored by the AIAA, SAE, ASME, and ASEE, Nashville,Tennessee, July 6-8, 1992. T.L. Krantz, Propulsion Directorate, U.S. Army Aviation Systems Command. J.G. Kish,Sikorsky Aircraft, Stratford, Connecticut 06601-138 1. Responsible person, T.L. Krantz, (216) 433-3580.

12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Unclassified - UnlimitedSubject Category 37

13. ABSTRACT (Maximum 200 words)

The Advanced Rotorcraft Transmission (ART) Program was initiated to advance the state-of-the-art for rotorcrafttransmissions. The goal of the ART Program was to develop and demonstrate the technologies needed to reducetransmission weight by 25 percent and reduce noise by 10dB while obtaining a 5000 hr mean- time-between-removal. Thispaper summarizes the research done at Sikorsky Aircraft under the ART Program. A split path design was selected as bestable to meet the program goals. Key component technologies needed for this design were identified, studied, anddeveloped. Two of these technologies are discussed in detail: the load sharing of split path designs including the use ofa compliant elastomeric torque splitter and the application of a high ratio, low pitch line velocity gear mesh. Developmentof an angular contact spherical roller bearing, transmission error analysis, and fretting fatigue testing are discussed. Thetechnologies for a lightweight, quiet, and reliable rotorcraft transmission have been demonstrated.

14. SUBJECT TERMS 15. NUMBER OF PAGESHelicopter transmission; Split torque; Gears; Gear lubrication 14

16. PRICE CODE

A0317. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURmI CLASSIFICATION 20. LIMITATION OF ABSTRACT

OF REPORT OF THIS PAGE OF ABSTRACT

Unclassified Unclassified Unclassified

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)Prescribed by ANSI Sid Z39 18298-102