濕式離合器設(shè)計(jì),離合器,設(shè)計(jì) Wear 288 (2012) 54–61 ContentslistsavailableatSciVerseScienceDirect Wear journal homepage: www.elsevier.com/locate/wear Apparatusforcontinuouswearmeasurementsduringwetclutchdurabilitytests NiklasLingestena,P?rMarklunda,?,ErikH?glunda,MartinLunda,JoakimLundinb,RikardM?kib aDivisionofMachineElements,Lule?UniversityofTechnology,97187Lule?,Sweden bVolvoConstructionEquipment,Eskilstuna,Sweden a r t i c l e i n f o a b s t r a c t Articlehistory: Received23May2011 Receivedinrevisedform14February2012 Accepted21February2012 Available online 1 March 2012 Wet clutches are used in many applications today such as automatic transmissions and limited slip differentials in cars as well as in heavy duty equipment such as wheel loaders. The present study is concerned with the wear and engagement behavior ofwet clutches inthe latter type ofapplication. A test rig isdeveloped inwhich the wet clutch engagement ismonitored during an arbitrary number of testcycles. This rig has many similarities with the SAE #2 test rig in that they are both inertia type test rigs. However,thetestrigpresentedherehasseveraloriginalpartsfromheavydutyequipmentinproduction incorporated intoit.Thedatacollectionincludesacontinuous measurement ofthepositionofthepiston usedtoapplyforceontheclutchpackinadditiontotheseparatordisctemperatures, hydraulicactuating pressureandtorquetransfercharacteristics. Themeasurements ofthepistonpositioncanthenberelated totheclutchwearduringalongtestseries. Keywords: Wetclutch Wear Testbench Tribology Temperaturemeasurement ? 2012 Published by Elsevier B.V. 1. Introduction etal.[6]intestrigsdesignedspeci?callyforthisapplication.The temperatureintheclutchseparatordiscswasmeasuredthrough Wetclutcheshavebeencriticalcomponentsofautomatictrans- missionsandlimitedslipdifferentialsforseveraldecades.However, phenomenaintheclutchsuchasshudderandtorquetransferfail- urehaspromptedinvestigationsintowetclutchbehavior.Thetwo mainmethodsusedarecomputersimulationsoftheclutchbehav- iorandexperimentalinvestigationsusingwetclutchtestrigs. Oneofthemostcommonlyreportedmethodsusedinwetclutch studiesistheSAE#2testrigwhichmainlyfunctionsasaniner- tia dynamometer where a ?ywheel is accelerated to accumulate energywhereuponaclutchisengagedtobrakethe?ywheel.This test setup allows for measurement of torque transfer properties andfrictioncharacteristicsforawidevarietyofoperatingcondi- tionsandhasbeenusedextensivelyinresearch[1–3]andisalso frequentlyusedinindustry.AsimilartestrigwasdevelopedbyHol- gerson[4]withtheadditionofallowingacontributionbyadriving torquetotheenergyinputtotheclutch.InHolgerson’stestrigthe clutch temperature was measured using an infrared thermome- ter,allowingforinvestigationsonthein?uenceoftemperatureon clutchbehavior. the use of thermocouples which allowed for investigation of the dependenceoffrictioncharacteristicsontemperature. Marklundetal.[7]usedpin-on-discmethodtodevelopafriction model for purposes of simulating wet clutch behavior. However, thisapproachisnotsuitableforstudyingclutchwearinparticular asdemonstratedbyOstetal.[8]norisitsuitedforinvestigationsof clutchagingingeneral.Thisisinpartbecauseofthelargedifference incontactareatosystemsizeratiointhissetupascomparedtothe clutch. Deterioration of clutch performance during service life is an important issue, limiting the development towards smaller and moreef?cientclutches.Theclutchagingcanbedividedintotwo categories.OneislubricantagingasinvestigatedbyBerglundetal. [9]forthelimitedslipclutch.Thesecondisfrictionmaterialaging, includingforexamplewear.GlazingasdescribedbyNewcombetal. [10] could be described as a combination of the two where the lubricant degrades due to the high temperature in the interface formingsmoothlayeronthefrictionmaterialsurface.Wearinpar- ticularwillseverelyaffecttheclutchcontrolsincethethicknessof clutchdiscsshouldbeconsideredwhenconstructionclutchcontrol algorithmsandmightaffectthefrictioncharacteristics. For the limited slip differential application, however, the SAE #2testrigisnotsuitableduetothecontinuousslipconditionsof theclutch.Investigationsintotheclutchcharacteristicsoflimited slipdifferentialshasbeenperformedbyM?kietal.[5]andIvanovic Wearofwetclutchfrictiondiscshasbeeninvestigatedbefore usingafewdifferenttechniques.Theearlystagesofwearhasbeen studiedbymanyauthors[3,11,12]intermsoftherealcontactarea betweenfrictionmaterialandsteeldisc.Forpurposesoflongterm durabilityofaclutch,thewearofthefullclutchpackismoreinter- estingasithasmoreofaneffectonhowtheengagementmustbe ? Correspondingauthor.Tel.:+46920492415;fax:+46920491399. E-mailaddress:par.marklund@ltu.se(P.Marklund). 0043-1648/$–seefrontmatter ? 2012 Published by Elsevier B.V. doi:10.1016/j.wear.2012.02.014 N.Lingestenetal./Wear 288 (2012) 54–61 55 Fig.1. Thetestriglayout:(1)driveshaft,(2)external?ywheel,(3)gables,(4and 6)gears,(5)sealingdoor,(7)clutchaxlewithclutchdrumand(8)torquearmwith loadcell. Fig.2. Theassemblyoftheclutchinthetestrig:piston(A),springseat(B),balance piston(C),pressureplate(D),separatorandfrictiondiscs(EandF),washer(G)and endplate(D). controlled.Thewearoffullclutchpackshasbeeninvestigatedusing theSAE#2testrig[1,3]orcustomtestrigssuchastheoneusedby Saitoetal.[13].Themethodrequiresthetestrigtobedisassembled regularlyinordertomeasurethewearoftheclutch.However,itis possibletocontinuouslymonitorthewearofaclutchpackusing thesensortechnologiesavailabletoday. Developmentofmoreef?cientclutchesrequiresreliabledesign criteria with regards to the different failure modes of the clutch. Inheavydutyequipment,theenergiesandpowerreleasedinthe engagementsaregenerallylargeandplacesanimportantlimitation ontheclutchdimensions.Thisworkfocusesonthedevelopment ofawetclutchtestrigwhichhastheabilitytocontinuouslymon- itor the wear in the clutch in addition to the clutch engagement characteristics.Theperformanceofdifferentclutchdesignscanbe evaluatedbystudyingthewearandfrictiondatawhichiscollected. Tobeabletostoreenergyinthesystem,thedriveshaft(1)has anintegrated?ywheel.Theshaftwasmanufacturedfromasingle pieceofsteeltoobtainawellbalancedshaftwhichwouldminimize theriskofvibrations.Rotationistransferredfromthedriveshaft totheclutchbyahelicalgearpair(4and6)withthegearratio1. Thegearontheclutchaxle(6)hassplinesmachinedonthehubfor mountingofthefrictiondiscs.Thedriveshaftisdrivenbyanelectric motorwhichhasamaximumrotationalspeedof3000rpm.Intotal, thetestrighasamomentofinertiaof0.6318kgm2.Anexternal mass(2)canbeaddedtothedriveshaft.Ifthisisdonethetotal momentofinertiawillincreaseto0.7375kgm2. The clutch axle (7) is a slightly modi?ed original part from a wheelloader.Channelsforoilrunthroughtheshaftforthepur- poseofclutchlubricationandhydraulicclutchcontrol.Theclutch ismountedintheclutchdrum.Twosensorsformeasurementof engagementcharacteristicsaremountedthroughholesintherear walloftheclutchdrum.Duringclutchengagementthetorquearm (8)keepstheclutchaxlestationarywhilepressingagainstaload cellwhichregisterstorquetransfer. Toallowforassemblyanddisassemblyoftheclutchpackahole wasmadeinthegablenearthegears.Throughthisholethegear(6) andclutchassemblycanberemovedandreplaced.Duringexper- iments the hole is sealed by a steel door (5) which supports the clutchaxle. 2. Methods To carry out wear studies on wet clutches a new test rig was designed.Themainfocusoftheauthorsistheapplicationofwet clutchesinheavydutyequipmentandthustherequirementsare basedonclutchusageissuchmachines.Throughconditionsiden- ti?edinsuchvehiclesalistofrequirementswascompiled.Thetest rigmustbeableto: TheclutchsetupinsidetheclutchdrumisshowninFig.2.Two frictiondiscswithinwardsplinesareused,squeezedbetweenthree separatordiscswithoutwardsplines(EandF).Theinnersepara- tordiscisloadedbyapressureplate(D)whichisindirectcontact with a hydraulic piston (A) applying the axial load on the clutch pack.Theouterseparatordiscisincontactwithasteelwasher(G). Thewasherallowsfortheassemblyofthetestspecimensinthe standardclutchdrumontheclutchshaft.Theendplate(H)isposi- tionedoutsideofthewasherandheldinplacewithalockingspring. Betweenthehydraulicpistonandthepressureplate,thereisabal- ancepistonandaspringseat(BandC).Betweenthesecomponents aspringismountedinordertoreturnthepistontotheoriginalposi- tionwhenthehydraulicpressureisreleased.Thestandardsprings ? Measurethecontinuouswearoftheclutchpack,thetorquetrans- fer,theclutchtemperatureandclutchrotationalspeed. ? Use standard size clutch discs from construction equipment in production. ? Achieveengagementenergiesabove260kJ/m2. ? Finishtheengagementwithin1swithgoodreproducibility. ? Allowforfeedbackcontrolwithregardstoanyoutputparameter. ? Allowforeasyandaccurateassembly/disassemblyoftheclutch pack 2.1. Mechanics AnillustrationofthecorepartofthetestrigisshowninFig.1. Thestructuralstabilityisprovidedbygables(3)andabottomplate insteel(AISI1148).Theclutchaxle(7)anddriveshaft(1)aresup- portedbytaperedrollerbearingsinthegablesandeverythingis mountedonasteelfoundation. has been replaced by a less stiff wave spring with a linear force displacementbehaviorandthespringconstant48×103N/m. There are three separate oil circulation systems active in the testrig,allusingthesametypeoflubricant.Twoforclutchlubri- cationandcooling,sharingoneoiltank,andonehydraulicsystem 56 N.Lingestenetal./Wear 288 (2012) 54–61 Table1 Testrigdimensions. Componentdimension Size Gablethickness Gableheight Gablewidth Flywheeldiameter Flywheelwidth Driveshaftlength Torquearmlength 41mm 350mm 540mm 275mm 117mm 527mm 150mm forclutchengagementwithaseparateoilsupply.Onelubrication system and the hydraulic system pump oil into the clutch shaft channelsthroughanoildividermountedonthepreviouslymen- tioneddoor.Anoil?lterforthelubricationoilisalsomountedon the door as well as the solenoid servo valve which controls the engagementpressure. Thelubricationoiltankhasavolumeof20l.Theoilispumped by?xeddisplacementgearpumpsthroughthetwocirculationsys- tems, both of which has a maximum ?ow rate of 10l/min. The primarycirculationsystempumpthelubricantthroughtheclutch axleintothefrictioninterface,whileaverysmallfractionofthe oil is directed to lubricate the bearings of the drive shaft. In the secondarycirculationsystem,oilispumpedthrougha?lteranda heatexchangerwheretheoiltemperatureiscontrolledbymeans ofwatercooling.Thereisalsoapossibilitytoheattheoilinthetank Fig.3. Schematicofpositionsensormeasurementsetup. testserieswillyieldinformationaboutthethicknesschangeofthe clutchpack,thusyieldinginformationaboutwear. The hydraulic actuating pressure which controls the clutch engagementiscontinuouslymeasuredbyapressuresensorwhich isalsomountedatthebackoftheclutchdrum.Thesensormeas- uresrelativepressureincreasewitharesponsetimelessthan1ms. Thepressuremeasurementisdirectlyrelatedtotheaxialforceand clutchdiscfacepressurethroughthepistonarea. Therotationalspeedofthesystemismeasuredbyamechanical counter built into the electric motor that accelerates the rig. For onerevolutionoftheshaft,4096pulsesaresentfromthemotor.A settimeintervalisspeci?edandthenumberofpulsesarecounted withinthisinterval,givingtherotationalspeed.Inthisinvestiga- tion,thenumberofpulsesweresampledatarateof100Hz. usingasuspensionheaterallowingforcontrolofthetesttemper- aturewithavariationlessthan2?C.Thedimensionsofthetestrig issummarizedinTable1. 2.2. Sensors A number of different sensors are installed in the rig to mea- suretheclutchcharacteristics.Thesensorworkingrangesandtheir resolutionissummarizedinTable2. 2.3. Dataretrievalandsystemcontrol Theverticaltorquearmisindirectcontactwithaloadsensor. Thesensorismountedinaholderattachedtothebottomplate.The loadcelloutputvoltageisconvertedtoloadinthecontrolsoftware. Thetemperatureismeasuredintheseparatordiscsbytheuseof thermocouples(typeK).Thesethermocouplesaremetalsheathed withadiameterof0.5mmwithaspeci?edresponsetimeof14ms. Theshortresponsetimeallowsforthecaptureofthefastchangesin temperatureduringclutchengagement.Onethermocoupleisalso usedtomonitortheoiltanktemperature. Onethermocoupleisinstalledineachseparatordisc.Thether- mocouplesareinsertedintoholesofdiameter0.8mmdrilledfrom theseparatordiscedges.Threedifferentmaterialsweretestedin theinterfacebetweenthethermocoupleandtheholewallinterms oftemperatureresponseperformance.Materialoneissimplythe ATF which will ?ll the gap between thermocouple and disc wall duringtestrigrunning.Materialtwoisaregulartinbasedsolder. ThetestrigiscontrolledbyaCompactRiocomputerthrougha LabViewinterfacerunningonaPC.Thecomputerinturncontrols severalsubsystems;electricmotorcontroller,thehydraulicpump andvalves,thepumpsforthecoolingandlubricationoftheclutch. Thehydraulicpumpandvalvecontrollingclutchengagementis switchedonandoffthroughsimpledigitalswitchesinthecontrol program. In the current setup, the hydraulic pump yields a sys- tem pressure of 5 MPa. The servo solenoid valve controlling the actuatingpressureiscontrolledthroughafeedbackloopincluding acontrolchannel.Anysensorreadingcanbeusedasthecontrol channel, but in the present cases the actuating pressure reading governsthevalveaction.Thepressuresensorreadingiscompared tothedesiredpressureandcombinedintoacontrolvoltagefedto thevalve.Theresponsetimeforthevalveislessthan5ms,allowing forprecisecontroloftheactuatingpressure. Material three is a gallium alloy which is in liquid state at room temperature with a boiling point higher than 1350?C. This alloy is commercially available under the name Coollaboratory Liquid Pro with application as thermal interface material for computer processors. The position of the piston (Fig. 2H) as the clutch engages is measuredbyacontactDVRT(differentialvariablereluctancetrans- ducer)sensor(Fig.3)mountedatthebackoftheclutchdrum.The sensorpinisspringloadedandfollowsthepistonasitmoves.The maximumlengthofmovementthatcanbemeasuredbythesen- sor is 6mm and it gives resolution down to single micrometers. Theoutputisavoltagethatiscalibratedagainstatestcurveand interpretedinthepostprocessingofexperimentaldata.Thechange in distance traveled by the piston during an engagement over a Thetestrigcanrunintwomodes.Onewhereeverythingiscon- trolledmanuallyandonewhichisasequencebasedrunningmode. Inthesecondmode,atestsequencecanbespeci?edandsettorun a number of clutch engagements. This mode is ideal for running extendedweartestsonaclutchpack.Themanualcontrolismainly usefulfortestingsensorsandtestrigfunctionality. Thecomputeralsohandlesthedataretrieval,savingthedataon anexternalharddriveinbinaryform.Dataisonlystoredduringthe actualclutchengagement,whichlastsforapproximately1s.The systemsamplesthedataatarateof2000Hz,enoughtoresolveall thephenomenapresentduringtheengagement. Parallel to the control system, there is a security system including relays to switch off power if the temperature or actu- ating pressure exceeds a preset value. There are also relays for