1、11FTM14AGMA Technical PaperAGMA 925-A03Predicted Scuffing Riskto Spur and HelicalGears in CommercialVehicle TransmissionsBy C.H. Wink, Eaton CorporationAGMA 925-A03 Predicted Scuffing Risk to Spur and HelicalGears in Commercial Vehicle TransmissionsCarlos H. Wink, Eaton Corporation Vehicle GroupThe
2、statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Association.AbstractThis paper presents the AGMA925-A03 scuffing risk predictions for a series of spur and helical gear sets oftransmissions
3、 that are used in commercial vehicles ranging from SAE class 3 through class 8. Contacttemperaturewascalculatedfor50spurandhelicalgearsets. Limitingscuffingtemperaturesusing2mineraland 3 synthetic gear lubricants were determined from FZG scuffing tests. The risk of scuffing was thenassessedcomparing
4、thecontacttemperaturewithlimitingscuffingtemperature. Theresultswerecomparedwith field and warranty data, and dynamometer test results. A good agreement was obtained for all cases.The agreement between prediction, test results and actual usage provides confidence in the predictor ofscuffing risk for
5、 gears in commercial vehicle transmissions.Copyright 2011American Gear Manufacturers Association1001 N. Fairfax Street, 5thFloorAlexandria, Virginia 22314October 2011ISBN: 978-1-61481-013-13 11FTM14AGMA 925-A03 Predicted Scuffing Risk to Spur and Helical Gears in CommercialVehicle TransmissionsCarlo
6、s H. Wink, Eaton Corporation Vehicle GroupIntroductionThe risk of gear tooth scuffing in commercial vehicle transmissions has gained more attention because ofincreasing demand for fuel-efficient powertrain systems in which diesel engines run at lower speeds, powerdensity is higher, andlubricants are
7、modifiedtoimproveefficiency andcompatibility withcomponents of newtechnologies, such as dual clutch transmissions.Thus, predicting scuffing risk during the design phase is vital for the successful development of commercialvehicletransmissions. AGMA925-A031isacomprehensivemethodtopredicttheprobabilit
8、yofgearscuff-ing. Therefore,thispaperpresentstheAGMA925-A031scuffingriskpredictionsforaseriesof50spurandhelical gear sets in transmissions that are used in commercial vehicles ranging from SAE class 3 throughclass8. Limitingscuffingtemperaturesusing2mineraland3syntheticgear lubricantsweredeterminedf
9、romFZG scuffing tests. The risk of scuffing was determined for each gear set according to AGMA 925-A03 1.The predictions were compared with field and warranty data, and dynamometer test results. The predictorwas correct for allcases. Highscuffingrisk was predictedfor gears knowntoscuff, andlowscuffi
10、ngrisk waspredictedfor allother cases that didnot haveahistory of scuffing. The goodagreement betweenprediction,test results and actual usage provides confidence in the predictor of scuffing risk for gears in commercialvehicle transmissions.BackgroundScuffing failure causes localized damage to the g
11、ear teeth resulting in matte and rough finishes of thecontactingsurfaces, andtoothform changes as well. This typeof damagegenerally occurs inthe toothcon-tact zonewherecontact pressureandslidingvelocity arehigh, far from thepitchline. Thetooth damagecanincrease vibration and noise, and be detrimenta
12、l to gear load capacity, ultimately leading to catastrophicfailures2. Asasevereadhesivewearphenomenon,scuffingoccurswhentheoilfilmthicknessbetweenthetooth contacting surfaces is not enough to prevent metal-to-metal contact, which causes local welding andsubsequent tearing. Metalparticles aretransfer
13、redbetweenthetwosurfaces or lost from them, andscratchtoothflanksintheslidingdirection. Scuffingis notafatiguephenomenonandit mayoccur atthebeginningofthe operation 3, 4.Thereareseveralanalyticalmethodstopredictscuffingrisk;however,thethresholdtodetermineifagearsetwill scuff is still mostly dependen
14、t on empirical results.ThemethodofevaluatingscuffingriskinAGMA925-A031isafunctionofoilviscosityandadditives,operat-ing bulk temperature of the gear, sliding velocity, surface roughness of gear teeth, gear materials and heattreatments,andsurfacepressure. Theriskofscuffingisdefinedbycomparingthetoothc
15、ontacttemperature,whichis calculated, withthelimitingscuffingtemperature, whichcanbeobtainedfromagearscuffingtestforeach gear lubricant.Contact temperature calculationFlash temperature across the tooth contact path is calculated by Bloks equation 1.(1)fli= 31.62 K mmiXiwnbHi0.5vr1i vr2iBM1vr1i0.5+ B
16、M2vr2i0.54 11FTM14wherefliis flash temperature, C;K is 0.80, a numerical factor for the Hertzian distribution of frictional heat over the instantaneouscontact band width;mmiis mean coefficient of friction;Xiis load sharing factor;wnis normal unit load, N/mm;vr1i,vr2iare rolling tangential velocities
17、 in m/s of the pinion and gear respectively, m/sBM1, BM2are 13.6 N/mm2K, thermal contact coefficient of steel;bHiis the semi-width of the Hertzian contact band, mm;i is a subscript of line-of-action points.Contact temperature at each line-of-action point is given by equation 2.(2)Bi= M+ fliwhereBiis
18、 contact temperature, C;Misthetoothtemperature,inC,whichisthetemperatureofthetoothsurfacebeforeit entersthecontact zone. The tooth temperature can be estimated by calculation, testing or experience.In terms of calculation of the tooth temperature, it can be estimated by equation 3.(3)M= ksumpoil+ 0.
19、56 fl maxwhereksumpis 1 for splash lube and 1.2 for spray lube;oilis oil supply or sump temperature in C;fl maxis the maximum flash temperature found over all line-of-contact points i (see equation 1).The maximum contact temperature is obtained by equation 4.(4)Bmax= M+ fl maxwhereBmaxis maximum con
20、tact temperature, C.Whenthemaximumcontacttemperatureis closertoorhigher thanthelimitingscuffingtemperature,scuffingmay occur.Limiting scuffing temperatureThe limiting scuffing temperature is the tooth contact temperature at which scuffing is likely tooccur, givenacombinationoflubricantandgearmateria
21、l1. Itcanbeobtainedfromgearscuffingtests,suchasFZGtests,whichis anindustry standardtest usedworld-wide torate different lubricants for resistance toscuffing. TheFZGtestmethodperCECL-84-026wasdevelopedbytheCEC,andisextensivelyusedbytheautomotiveand petroleum industries in Europe and throughout the wo
22、rld.5 11FTM14TheFZGgeartestmachineisoperatedatconstantspeedforafixedperiodoftimeatsuccessivelyincreasingloadsuntilthefailurecriteriaisreached. Thetestgearsareexaminedinitiallyandaftertheprescribeddurationateachloadstageforscuffingdamagetothegeartoothflanks. Thetestgearfailswhenthesumofthescuffedarea
23、 widths on all teeth exceeds the gear face width 3.Back in2004EatonCorporation - VehicleGroup developedan extensivework toinvestigate industry stand-ard scuffing test methods, and correlate their results with different gear lubricants in transmission tests indynamometers. Theinvestigationwasmotivate
24、dbylongeroildrainintervalsandnewlubricantsbeingusedinEaton transmissions. Under these conditions the former ASTM D-5182-9 3 method did not correlate toknown scuffing occurrences.TheFZG method definedby CECL-84-02 6 and designatedas A10/16.6R/120showed consistent resultsandgoodcorrelationwithdynamome
25、ter tests. Sincethen, EatonCorp. has adoptedthis testmethodaspartof its internal procedure for gear lubricant qualification and approval.TheA10/16.6R120stands for A-typegear geometry, 16.6m/s pitchlinevelocity, andasumptemperatureof120C. TheA-typegearisdesignedwithlongeraddendum geometrytogeneratehi
26、ghslidingvelocity, andismanufactured with only a 10mm face width to increase the contact stress so that the scuffing could be moreeasily producedthanwiththe 20mmfacewidth. The10mmfacewidthgear testis commonlyreferredtoasthe half-tooth, double speed test. Figure 1 shows a FZG A-type gear after a test
27、.Scuffing risk predictionWhen the maximum contact temperature is close to or above the limiting scuffing temperature for thecombination of lubricant and gear material scuffing is likely to occur. The probability of scuffing can be pre-dicted by comparing the contact temperature with the limiting scu
28、ffing temperature. The AGMA 925-A03probabilityofscuffingisobtainedfromaGaussiandistributionofscuffingtemperatureassuminga0.15coeffi-cientofvariation. Thecontacttemperatureisarandomvariable,thelimitingscuffingtemperatureisthemeanvalueof therandom variable, andthestandarddeviationof therandom variable
29、is assumedto be0.15 of thescuffing temperature. Then AGMA 925-A03 1 proposes a criterion to assess the risk of scuffing based onthe calculated probability of scuffing as shown in Table 1.Figure 1. FZG A-type gear after testTable 1. Scuffing risk per AGMA 925-A03Probability of scuffing Scuffing risk3
30、0% High6 11FTM14Application examplesIn this study 50 gear sets were evaluated usingthe AGMA 925-A03 scuffingrisk method1; that is, 24spurgearsetsand26helicalgearsets,whosehelixangleswerebetween12and33degrees,wereevaluated. Thegearsetsareusedin9differentmodelsofcommercialvehicletransmissionswith5to18
31、forwardspeeds,andtransmission torque capacity between 420Nm to 2500Nm. Their applications include light duty pick-uptrucks, delivery trucks, buses, and vocational and heavy haul trucks.Figure 2 and Figure 3 show basic geometric and operating parameters of the gear sets, such as module,pressure angle
32、, contact ratio, pitch line velocity, and sliding velocity.Figure 2. Characteristics of the studied gear setsFigure 3. Sliding velocity and transverse contact ratio of the gear sets7 11FTM14Allgearswerecasehardenedsteelwithasurfacehardnessof 58-63HRc. Amongthegearsets threetooth-flank finish process
33、es were used - shaving (surface roughness Ra=1.25mm), honing (Ra=0.40mm), andCBN grinding (Ra=0.63mm).The flash temperature was calculated for each gear set using equation 1 for 25 line-of-action points, asrecommended by AGMA 925-A03 1.The coefficient of friction was calculated withthe formulaein AG
34、MA 925-A03 1 for anapproximation of thecoefficient of frictionby aconstant. Thecoefficient of frictionwas calculatedwiththeaveragesurfacerough-nessofthepinionandgearteeth,andwasassumedconstantalongtheline-of-action. Themaximumcoeffi-cient of frictionvaluewas limitedto0.11. Theeffect ofthecoefficient
35、of frictiononthemaximum contacttem-peraturepredictedbytheanalyticalmodelwasinvestigatedforoneofthegearsets. Thecoefficientoffrictionwas arbitrarily changed from 0.03 to 0.11, and the contact temperature calculated.Theresults showedthat smallvariations oncoefficient of frictionmay not significantly a
36、ffect the contact tem-perature results; for example, a 10% variation of a 0.10 coefficient of friction caused the maximum contacttemperaturetovaryonlyabout6%(seeFigure 4). However,morepronouncedcoefficientoffrictionchanges,e.g., when changing finish processes and obtaining much smoother surfaces - c
37、an significantly affect themaximum contact temperature.Another uncertainty whencalculatingthecontact temperatureof gear setsusedincommercialvehicletrans-missions was the load condition for the calculation, i.e., torque and speed, or power. Vehicles powered withcombustionengines rununder acertainpowe
38、r regime, whichis a functionof thecharacteristic enginepowercurve. In typical engine power curves, the point of maximum power does not match the point of maximumtorque. Typically themaximum torqueis at alower enginespeedthanthemaximum power point. Inordertofindthecriticalloadconditionforscuffing,the
39、maximumcontacttemperaturewascalculatedto10pointsalongthe engine power curve. The results are shown in Figure 5.The maximum contact temperature followed the engine power curve; i.e., as power increases the contacttemperature also increases. The maximum value of contact temperature was obtained at max
40、imum enginepower. Based on these results all calculations were made at maximum engine power instead of at peaktorque.Figure 4. Effect of coefficient of friction on contact temperature results8 11FTM14Figure 5. Effect of engine power on contact temperature resultsFivedifferent gear lubricants wereuse
41、dinthis study, 2mineraland3synthetic lubricants. Table 2showsthebasic characteristic of the gear lubricants and their load stage (the torque at that stage of the test) at whichscuffing occurred.ThelimitingscuffingtemperaturewasthencalculatedfortheFZGA-typegear(A10/16.6R120)atthescuffingfailure load
42、using the AGMA 925-A03 method, which was implemented into a Microsoft Excel spreadsheet.Oncethescuffingtemperaturewasdefinedforeachoneofthe5gearlubricantsofTable 2,andthemaximumcontact temperaturecalculatedfor eachof the50 gear sets, thescuffing risk was obtainedby comparingthecontacttemperatureands
43、cuffingtemperature. Twoapproachesweretakenforthis,firstlytheratioofcontacttemperature and scuffing temperature was calculated, and named scuffing ratio. Ratios close or above 1meansscuffingislikelytooccur. Secondlytheriskofscuffingwasassessedaslow,moderateandhighbasedon the probability of scuffing d
44、efined in AGMA 925-A03 1.Amongthose50gearsetsthatwerestudied,4ofthemshowedscuffingfailures,indynamometertests andinthe field. A picture of one of the scuffing failures is shown in Figure 6.Table 2. Gear lubricantsGear lubricants 1 2 3 4 5Oil base Synthetic Mineral Synthetic Mineral SyntheticViscosit
45、y index (ASTM D227) 151 194 146 164 150Kinematic viscosity 40C (mm2/s) 135.0 33.8 132.0 51.0 40.5Kinematic viscosity 100C (mm2/s) 18.2 7.6 17.5 9.1 7.4Scuffing failure load stage of FZG test* 3 6 9 7 6* FZG A10-16.6R1209 11FTM14Figure 6. Scuffed gear set from a dynamometer testFigures7andFigure8show
46、thescuffingratioandprobabilityofscuffing,respectively,forall50gearsets. The4 scuffed gear sets are identified in red and with an asterisk in front of the results in Figure 7 and Figure 8.Scuffing ratios closer to and above 1 were predicted for all 4 gear sets at which scuffingfailures wereknown(see
47、Figure 7). However, looking at the scuffing ratio only can lead one to think that scuffing could occur ongear set number 5, which has a scuffing ratio around 0.9.Figure 7. Scuffing ratio results (scuffed gear set)10 11FTM14Figure 8. Probability of scuffing, 30% means high risk, (*scuffed gear set)Th
48、ispointedouttheadvantageofusingAGMA925-A03probabilityof scuffing. Figure 8shows theprobabil-ity of scuffing results. The results clearly show high risk of scuffing (greater than 30%) to all 4 gear sets atwhich scuffing was observed.When looking at gear set number 5, its probability of scuffing is ar
49、ound 20% which is classified as moderaterisk in AGMA 925-A03 1. Thus, aclearer assessment can bedone usingthe probability of scuffing, as it isrecommended in the AGMA standard.DiscussionsTheAGMA925-A03standardwasusedtopredict scuffingrisk of50spurandhelicalgear setsof 9transmis-sion models, which are applied to commercial vehicles ranging from SAE class 3 through class 8. Scuffingtemperature of 2 mineral and 3 synthetic gear
