890092 TestingofAutomotive HubUnits L.Berryman SKFEngineering &Research CentreB.V. ABSTRACT Methods aredescribed forthetest­ ingofautomotive hubunitsunderquasi­ static, impact, anddynamic conditions. Itisshownthattheoverall timerequir­ edtoreliably determine therolling con­ tactfatigue lifeofsuchunitscanbe reduced bytheselection ofasuitable teststrategy. Astrategy isdescribed in whichfifteen 'sudden death' groups of fourunitsaretested simultaneously. Whenanyone unitinagroupfails, then thecomplete groupisconsidered tohave failed. Weibull analysis isthenapplied totheseresults toestimate theL10life oftheunit.This'sudden death'strategy hasabetter discriminating powerthana' conventional 'fivefailures fromtentest­ edunits' strategy, andcanbecompleted inamuchshorter testtime. THISPAPER descri~es arangeoftestsde­ signedtoinvestigate allaspects ofhub unitdesign: Inparticular, anapplicat­ ion-related testfordetermining therol­ lingcontact fatigue lifeofsuchunits ispresented. Itwillbedemonstrated thatbyselecting asuitable teststrat­ egy,theoverall testtimerequired to determine areliable fatigue lifeestima­ tioncanbereduced. AUTOMOTIVE HUBUNITS Theuseofautomotive hubunitsfor boththefrontandrearwheelsofpassen­ gercarshasbecome acommon practice throughout theautomotive industry (1)*. Automotive hubunits(Fig.1) aredouble rowangular contact bearings whichare pre-assembled, greased forlifeand .*Numbers inparentheses designate refer­ encesatendofpaper. 141sealed bythemanufacturer. Hubunits offerdistinct advantages overconVen­ tional twobearing arrangements forwheel assemblies, namely: compactness, easeof assembly, andreliability. Hubunitsmust satisfy threebasicrequirements: theymustallowwheelrotation under allservice conditions withminimum friction losses, minimum noise, and satisfactory service life. inthecaseofflanged hubunits, they mustmaintain thestructural integrity ofthewheelassembly. andinsomecases(i.e.driven wheel arrangements), theymusttransmit tor­ quefromthedriveshafttothewheel flange. Hubunitsareamongst themosthigh­ lyloadedballbearings usedtoday. The Hertzian contact pressures between the ballsandtheinnerringraceway canbe theoretically ashighas5000N/mm', whichisintheorderof50000atmos­ pheres. Theballs andraceways aresepar­ atedduring running byanoilfilmless than0.5urnthick. Whenpassing through thiscontact theoilundergoes achange fromliquidtoasemi-solid andbackto liquid inafraction ofasecond. Thedevelopment ofautomotive hub unitsforsucharduous conditions hasre­ quired aparallel development oftesting methods toinvestigate allaspects ofhub unitdesign andtoprovetheirreliabil­ ity. HUBUNITLIFETESTING Functional testing ofautomotive hub unitsmust: beperformed underrealistic testcon­ ditionssuch.thatthefailure modere­ sulting fromthetestissimilar to thatexperienced innormal service . bestatistically validintermsofDownloaded from SAE International by University of New South Wales, Wednesday, September 19, 2018confidence inthedatagenerated. satisfy somepractical testing re­ qUirements (i.e.shorttesttime,ease ofapplication ofthetest,etc.), whilemeeting thetestobjectives. investigate allthreeoftheabove­ mentioned basicrequirements ofthe hubunit. Hubunitservice lifeisdetermined byrolling contact fatigue (asub-surface initiated fatigue phenomena associated withrolling element bearings), thelub­ ricant (grease) life,theseallife,and thefatigue lifeoftheflanges. Anyin­ vestigation ofthelifeorfunctional performance ofahubunitforapa~tic­ ularvehicle application mustconslder alloftheseaspects ofservice life. TESTCONDITIONS -Thefunctional testing ofautomotive hubunitsmustbe performed undertestconditions which result inthesamefailure modeasthat experienced innormal service. Thedeter­ mination ofsuchtestconditions requires detailed knowledge oftheservice envir­ onment towhich hubunitsaresubjected. Suchknowledge canonlybegained byroad loaddatacollection andanalysis. Asa sub-supplier totheautomotive industry, SKFdoesnotoperate itsownproving ground andtest