《AGMA-93FTM11-1993.pdf》由会员分享，可在线阅读，更多相关《AGMA-93FTM11-1993.pdf（17页珍藏版）》请在三一文库上搜索。
1、93FTM11 The RelativeNoise Levels of ParallelAxis Gear Sets with Various ContactRatiosand Gear Tooth Forms by: R. J. Drago, J. W. Lenski and R. H. Spencer, Boeing Defense and Space Group; M. Valco, Lewis ResearchCenter, U.S. Army; E Oswald, Lewis research Center, NASA AmericanGearManufacturersAssocia
2、tion II TECHNICALPAPER Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or networking permitted without license from IHS -,-,- he Relative Noise Levels
3、 of Parallel Axis Gears Sets with Various Contact ratios and Gear Tooth Forms R. J. Drago, J. W. Lenski and R. H. Spencer, Boeing Defense and Space Group; M. Valeo, Lewis Research Center, U.S. Army; and F. Oswald, Lewis Research Center, NASA The statements andopinions contained herein arethose ofthe
4、 authorandshould notbe construed as anofficial action or opinion of the American Gear Manufacturers Association. ABSTRACT: The real noise reduction benefits which may be obtained through the useof one gear tooth form as compared to another is an important design parameter for any geared system, espe
5、cially for helicopters in which both weight and reliability are very important factors. This paper describes the design and testing of nine sets of gears which are as identical as possible except for their basic tooth geometry. Noise measurements were made atvarious combinations of load and speed fo
6、r each gear set so that direct comparisons could be made. The resultant datawas analyzed so that valid conclusions could be drawn and interpreted for design use. Copyright1993 American Gear ManufacturersAssociation 1500 King Street, Suite 201 Alexandria,Vtrginia, 22314 October, 1993 ISBN:1-55589-623
7、-5 Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or networking permitted without license from IHS -,-,- The RelativeNoiseLevelsof ParallelAxisGear S
8、ets WithVariousContactRatiosandGear ToothForms By Raymond ,I.Drago,PE AssociateTechnicalFellow BoeingHelicopters RobertH.SpencerJosephW.Lenski,Jr. StaffEngineerAssociateTechnicalFellow BoeingHelicoptersBoeingHelicopters MarkJ. ValcoFredB. Oswald AerospaceEngineerResearch Engineer LewisResearchCenter
9、/U. S.ArmyLewisResearchCenterNASA I INTRODUCTIONmodifiedcontactratiosandthe geartoothform, separatelyand in combination,for spur and helical The problem of gear noise in helicopter transmissionsgears,on the noiselevelsproducedby otherwise is ever present.The main excitingforces whichidenticalspurand
10、helicalgears.Inorderto produce this noise are the meshingforces of the gearaccomplishthis objective, a program was defined to teeth in the transmission. While this is certainly andesignappropriategears(TableI),fabricatea oversimplification,sincemanyfactorsinfluencesufficient number of test specimens
11、,and conduct the transmissionnoise aside from the gear mesh forces,testing required. the simplefactrernain._ that if the basicexciting forcesare reducedand no amplifyingfactorsareWhile a wide range of specimens is shown, they were present, the overall noise level of the system will beall configured
12、as nearly alike as practical,within the reduced,limitations imposed by manufacturingconsiderations and the test stand. Testing was conducted in a single Amongthe severalwaysin whichthe gear toothmeshgear box undercontrolledconditionswhich meshingforcesmay be reduced,two of the mostweremaintainedasne
13、arlyidenticalaspossible. directly applicableto helicopter transmissionsare theAcoustic intensity measurementswere taken with the form of the teeth and the overall contact ratio. Bothaid of a robot to insure repeatabilityof measurement approachesarequiteattractiveforanaerospacebetween gear sets and t
14、o minimize human technique applicationsince, unlikeother “treatment“methods,influence. whichare appliedwithpenaltiesto eithersystem weightor performance,theseapproacheshavethe potentialfor reducingnoise withoutcausinganyTESTGEAR DESIGN increaseinoverallsystemweightorreducing performance. In fact, bo
15、th approaches also offer theEight (8) sets of gears, four (4) spur and four (4) possibilityof.actuallyprovidingimprovedgearhelicalas listedin TableI, compatiblewith the performancein termsof longerlife,higherloadNASA Lewis gear noise test rig, were designed.Of capacity,improvedreliability,and reduce
16、dweightthe four sets of spur gears, two sets have an involute while simultaneouslyreducingnoise levels,tooth form and two utilizea nouinvolute,constant radiusofcurvaturetoothform.Allgearswere Theobjectiveof thisprogramwas to define,bydesignedinaccordancewithnormalBoeing controlled testing and actual
17、 noise measurements,theHelicopterspractice so that, except for size, they are effect of changesin the profile, face, andrepresentativeof typical helicoptergears. Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Be
18、rnie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or networking permitted without license from IHS -,-,- TableI Gear NoiseTest Matrix Contact Ratios ConfigurationToothFormTvlae ProfileFaceModifie.d 1.ConventionalInvoluteSpur1.250.001.25 Spur Baseline 2.HCR-INVInvoluteSpur2.150.002.15 3.Con
19、ventionalInvoluteHelical1.251.251.77 SingleHelicalBaseline 4.DoubleHelicalInvoluteHelical1.251.251.77 5.HCR-INVInvoluteHelical1.251.752.15 6.HCR-INVInvoluteHelical2.152.253.11 7.NIF BaselineNonInvoluteSpur1.250.01.25 8.NIF-HCRNonInvoluteSpur2.150.02.15 Figure 1 - Test Gears Copyright American Gear M
20、anufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or networking permitted without license from IHS -,-,- Table II - Basic Test Gear Configuration II PinionGear Number of Teeth25
21、31 DiametralPitch, Transverse8. Center Distance3.50 Pressure Angle, Transverse25 (Std Profile Contact Ratio) 20 (High ProfileContact Ratio) Face Width (Spur recorded acoustic intensity spectra in theanalyzerfor eachnodeof thegrid;and transmitted the spectra to the computerfor storage on _disk. The a
22、constieintensityprobeconsistsof a pair of aEAr_ NmSE at_phase-matched6 mm microphonesmounted face-to- face with a 6 mm spacer. The probe has a frequency range (1 dB) of 300-10,000I-Iz.Measurementswere Figure 2 - Nasa Gear Test Rigmade at a distance of 60 mm betweenthe acoustic center of the micropho
23、nesand the gearbox top. A 20 node measurementgrid was drawn on the topThe 20 intensityspectra collectedat each operating cover of the gear box and used to insure repeatabilityconditionwereaveraged,thenmultipliedby the of the noise measurementsand to aid in avoidingradiationarea to compute an 800 lin
24、e sound power operator induced errors. The grid covers an area 228spectrum. The radiationarea was assumed to be the x 304 mm (9 x 12 in) centered on the 286 x 362 mmarea of the grid plus one additionalrow and column (11.25 x 14.25 in) top. A cutaway section of the testof elements or 0.0910 m2.The ae
25、tnal area of the top gear box is shown in Figure 3. All data was collectedis 0.1034 mz. The measurementgrid did not extend Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:4
26、1 MDTNo reproduction or networking permitted without license from IHS -,-,- completelyto the edges of the gearbox top because the edge of the top was boredto a stiff mounting flange which wouldnot allow much movementand measurementstakencloseto the edge of the top would be affectedby noise radiated
27、from the sides of the box. Noise measurementsfrom the gearboxsides were not attemptedfor the followingreasons: (1) the top is not as stiff as the sides; thus, noise radiationfrom the top dominates (2) the numberof measurementlocationswere kept reasonable;and (3)shaftingandotherprojectionsmadesuch me
28、asurementsdifficult. Sound power measurementswere made over a matrix 9_ . )_;II,_ _“:. of nine test conditions:3 speeds (3000, 4000, 5000 rpm) and at 3 torque levels (60, 80 and 100 percentFigure 4 - Robot Noise MeasurementSystem of the referencetorque 256 N-m (2269 in-lb). During each intensity sca
29、n, the speed was hetd to within 5 rpm and torque to _-2N-m. At least five complete setsPROCESSINGSOUND POWERDATA of scans were performedon each gear set. The sound powerdata as capturedby the method Acousticintensitydatawererecordedovertheoutlined above consists ofmanydata files of 800 line bandwidt
30、h896-7296Hz. On the 800 line analyzer,sound power spectra.A typical spectrum is shown in this produceda line spacing of 8 Hz. We chose thisFigure 5. This trace(takenat 5000 rpmand100 frequencyrange becauseit includes the first threepercent torque)includes the first three harmoniesof harmoniesof gear
31、 meshingfrequency for the speedgear mesh frequency. Each harmonic is surroundedby range (3000-5000rpm). In addition to the intensitya numberof sidebands. data,signalsfromtwomicrophonesandtwo aecelerometerswere recordedon four-channeltape. Ba_I,n_ S_e“fieor_No4/8 I_1_ i f_ Frequeem:9,kHz Figure 5 - B
32、aseline Spur Spectrum To characterizegear noisedata, it was decidedto reducethe 800 line sound power spectrato a single numberthatwouldrepresenteachgearmesh harmonic.For the subjectreport, this is referredto as Figure3 - Test Gear Box CutawaySectionthe harmonicsoundpowerlevel.Five alternatives were
33、consideredfor reportingof each harmoniclevel: Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or networking permitted without license from IHS -,-,- (
34、1) The amplitudeat gear mesh frequencyonly (noBo_,_p_r G No 4/8 sidebands) (2) The value oftbelargest amplitudemesh frequencyst harmonic or sideband, whicheveris highest_i_fq (3) The log sum of the sound intensity amplitudesin a fixed-width frequency band centered on the mesh_ frequency. (4) A value
35、similarto (3) exceptthe size of the8 _,_,_,9_2_2, _5 frequency band varied with speed. The total numberF_,=_.kH_ of values added is not constant. (5) Sum of gear mesh and fixed number of sidebands.Figure6 - Enlargementof Figure5 (Around First Harmonic) Alternative(5)waschosenforcomputingthe harmonic
36、sound power level. We used three pairs of sidebands plns the harmonies(i.e., seven peaks) in the calculation.Sound power levels were convertedtoCl=t(6/v/-n)( 1 ) Watts prior to calculatingsums. where: In the analysisof the intensitydata, each harmonicof gear mesh frequency was defined by several dig
37、italC_= confidencelimit, dB lines of the frequency analyzer.In order to capture the total effective magnitude at each harmonic, whilet= probabilitydistribution(“Student accountingfor speed drift, ete, the peak value and twot“ distribution) frequencylineson eithersideof the peakwere summed. These val
38、ues were converted to dB (re 10_28= standard deviationof data, dB W) to definea mesh harmoniclevel.Since seven peaks were used, 35 values (5x7) were summed ton= number of samples (typically5) produce the mesh harmonic sound power level Figure 6 illustratesthe data (marked with the symbol“*“)The valu
39、es for the “t“ distributionare found in any used to produce the harmonic sound power level. Thisstandardstatisticstext.A confidencelevelof95 is a portion of the spectrum of Figure 5 showing thepercent correspondsto a 5% probability.The number first harmonic(at 2083 Hz). The sidebandspacingof degrees
40、of freedomin the “t“ distributionis the (for 5000 rpm) is 83 Hz, thus there are about10numberof samples minus 1 (typically 4). analyzer lines per sideband. At lower speeds, there are fewer analyzer lines per sideband.Toestimatetheeffectduetosample-to-sample variation,two setsof gearsfor eachdesign w
41、ere fabricatedand tested.Each gearwas inspectedin DATA SAMPLINGdetail in accordance with typical production helicopter standards.Theoverallaccuracyof the gearswas In order to be assured that data measuredon eachfoundto beconsistentwithwhatweexpectof gear set could be reliablycomparedwith data frompr
42、oductionhelicoptergearsofsimilarsizeand other gears, it was desired to have sufficient recordseonfignration.Based on our evaluationof the gear to establish a 95% confidence level of 1 dB. Thistooth inspection data, the variation betweenthe two level is well beyond the practicaldifference(i.e., asets
43、ofgearsisreasonablytypicalofnormal change of about3 dB) whichmostpersonswithproductionfor gears in the same manufacturinglot. normal hearingcan detect.Lot to lot variationsmay be and differencesbetween differentmanufacturersof the same parts certainly Based on these considerations,the confidencelimi
44、twill be higherbut the overalltrend of the effect is given byEquation1:should be about the same. 6 Copyright American Gear Manufacturers Association Provided by IHS under license with AGMA Licensee=IHS Employees/1111111001, User=Wing, Bernie Not for Resale, 04/18/2007 11:20:41 MDTNo reproduction or
45、networking permitted without license from IHS -,-,- We have also noted that a large differencein noiseBetter load capacity, due to lower messes,is another level is sometimes observed on large productiongearfactor but will be ignored for our purposes. boxes simply as a result of rebuildingthem after
46、they weredisassembledfora visualinspection,even though no parts were chon_ged.Considering this effect,z_.000 in addition to the manufacturingvariability checks, weBASEUNE SPUR also checkedfor variabilitydue to disassemblyand1so.000 reassembly._ We accomplishedthis by testing three “builds“ of the_ 1
47、2o,ooo first gear set. Each build used exactly the same parts and each was accomplishedby the same technician_ _“ eo.ooo usingthe sametools,andmiscellaneous parts._ 0HeRi: HELICAL 40,000 TESTGEAR LOADING 0,I,I, 1.50020002,500 Theloadsappliedtothetestgearsduringthis1.000 programpresenteda problemin t
48、he designof theTORQUE(INCH-LBS) experiment.Obviously, if the overall gear geometry is kept constant, the stress levels under identical torqueFigure 8 - Contact Stress v Torque In order to providean overviewof the stress levels to 3o,ooorwhichthesegearsweresubjectedduringtesting, 8ASEUNESPURFigures 7, 8, Id136151/2269 ,_31d18164W13614kr22695k/18t6 Speed(rpm) t Torque (in*lb) Baseline Spur HCRSpur NIFSpur NIFHCRSpur Figure 19 - Modified Contact Ratio Effect ToothForm - In general, the noninvohite tooth form,Figure 21 - Tooth Form Effect wheth