1、09FTM13AGMA Technical PaperBending Fatigue,Impact and PittingResistance of AusformFinished P/M Gearsby N. Sonti and S.B. Rao,Pennsylvania State University andG. Anderson, Keystone PowderedMetal CompanyBending Fatigue, Impact and Pitting Resistance of AusformFinished P/M GearsNageshSonti andSurenB. R
2、ao, PennsylvaniaStateUniversityandGaryAnderson,Keystone Powdered Metal CompanyThe 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.AbstractPowder metal (P/M) process is making i
3、nroads in automotive transmission applications because ofsubstantially lower cost of P/M steel components for high volume production as compared to wrought orforgedsteelparts. AlthoughP/Mgearsareincreasinglybeingusedinpoweredhandtools,gearpumps,andasaccessorycomponentsinautomotivetransmissions,P/Mst
4、eelgearsarecurrentlyinlimiteduseinvehicletransmissionapplications. TheprimaryobjectiveofthisprojectwastodevelophighstrengthP/Msteelgearswith bending fatigue, impact, and pitting fatigue performance equivalent to current wrought steel gears.Ausformfinishingtoolingandprocesswasdevelopedandappliedtopow
5、derforged(P/F)steelgearsinorderto enhance the strength and durability characteristics of P/M gears, while maintaining the substantive costadvantage for vehicle transmission applications.BendingfatigueandimpactstrengthofausformfinishedP/Fsteelgearsweredemonstratedtobecomparableto conventional wrought
6、steel gears. Pitting fatigue life of ausformfinished P/Fgears was about 85%higherthan wrought steel gears produced by current conventional processing techniques. Scoring and wearresistance of ausform finished P/F steel gears were also shown to be substantially superior to conventionalwrought steel g
7、ears. This paper presents the processing techniques used to produce ausform finished P/Fsteel gears, and comparative bending fatigue, impact and surface durability performance characteristics ofausform finished P/F steel gears as well as conventional wrought steel gears.Copyright 2009American Gear M
8、anufacturers Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314September 2009ISBN: 978-1-55589-966-03Bending Fatigue, Impact and Pitting Resistance of Ausform Finished P/M GearsNagesh Sonti and Suren B. Rao, Pennsylvania State University and Gary Anderson,Keystone Powdered Metal
9、CompanyIntroductionPowder metal (P/M) process has the potential forsubstantial cost advantage for high volume vehicletransmission applications as compared to currentwrought or forged steel parts. 1, 2 Recentadvances in P/M technology such as doublepressing/sintering, high temperature sintering,surfa
10、cedensificationandpowder forging havesub-stantially enhanced the tooth bending fatigue androlling contact fatigue performance of P/M steelgears to be comparable to current automotivewrought steelgears. 3-16 Useof P/M steelgearsfor power transmission applications, however, iscurrently limited to mini
11、mally loaded componentsdue to surface durability performance constraints.Ausform gear finishing process developed at PennState University has the potential to induceimproved accuracy and surface finish in a costeffective manner and thereby enhance the perfor-mance characteristics of P/M steel gears
12、to beequivalent tocurrent wrought or forged steelgears.17-20 Powder forged (P/F) and case hardenedP/M steel gears were ausform finished in order toachievebendingfatiguestrength,impactresistanceand surface durability performance at least equiva-lent to current wrought steel gears. Test resultsfrom a
13、comparative performance testing programare presented comparing ausform finished P/Fsteel gears, and baseline wrought steel gears.Materials and processing detailsP/M standard gear design (24 teeth, 8 diametralpitch, 0.5” face width), which had been previouslyusedinaCPMTsponsoredGearResearchInstitutep
14、rogram to compare sixteen different P/Mformulations and/or processing techniques, wasselectedasthecandidatetestgear,withsomealter-ations. Table1summarizes thetest gear geometrydetails for baseline wrought steel gears andausform finished P/F steel gears. Table 1 also de-scribes the mate gears used fo
15、r surface durabilitytesting including pitting fatigue, scoring and wearresistance tests.Table 1. Gear dimensionsTest gears Mate gear# of teeth 24 40Diametral pitch 8 8Pressure angle 18.65 18.65Tooth thickness 0.235-0.237 0.146-0.148Base diameter 2.842469 4.737449Root diameter 2.744-2.754 4.543-4.555
16、SAP diameter 2.8944 4.77Pitch diameter 3 5EAP diameter 3.357 5.147Outside diameter 3.36 5.152Fillet radius 0.0536 0.0507P/FsteeltestgearsweremadefromprealloyedP/M4620 steel composition, a carburizing grade P/Mformulation that is typically used to produce auto-motive P/M parts. Baseline gears were pr
17、oducedfrom wrought 4023steel, andmategears (40teeth,8 diametral pitch, 1.0” face width) used for powercirculating (PC) surface durability tests wereproduced from wrought 8620 steel. Table 2 showsthe chemical composition of the respective steels.Case hardening specifications for all test and mategear
18、s required a surface hardness of 58-63 HRC,effectivecasedepthto50 HRCof 0.030 0.040” atmid tooth height and 0.15” minimum at root fillet,and core hardness of 28-34 HRC. Processsequence used for producing P/F steel test gearsinvolved pressing, sintering, powder forging, casehardening heat treatment,
19、followed by ausformfinishing. In comparison, process sequence usedfor producing baseline and mate gears involvedblank machining, forging, hobbing, shaving, casehardeningheat treatment, andshot peening. Toothprofile design of mate gears incorporated a pro-nouncedtipandrootrelief inorder toaccommodate
20、tooth bending effects in the test gear teeth duringsurface durability testing. Table 3 compares the4processing steps used for the manufacture ofausform finished P/F steel test gears and wroughtsteel baseline and mate gears.Table 2. Chemical compositions4620 4023 8620C 0.17-0.22 0.2-0.25 0.18-0.23Mn
21、0.45-0.65 0.7-0.9 0.7-0.9Si 0.15-0.35 0.15-0.35 0.15-0.35Ni 1.65-2.0 0.4-0.7Cr 0.4-0.6Mo 0.2-0.3 0.2-0.3 0.15-0.25P (max) 0.035 0.035 0.035S (max 0.04 0.04 0.04Table 3. Process sequence comparisonP/F test gears Baseline/mate gearsPress Barstock cutSinter ForgePowder forge NormalizeMachine MachineHea
22、t treat HobHone bore ShaveAusform finish Heat treatTemper Hone boreDeburr Finish facesFinish faces Shot peenAusform gear finishingAusforming is a modified heat treatment processapplicable to medium to high carbon low alloyedsteels, wherein the steel is first austenitizedfollowed by interrupted quenc
23、hing to above the MStemperature to metastable austenitic state. Thepart is then plastically deformed in the metastableausteniticconditionandfinallycooledtomartensite.Figure 1 shows a schematic time-temperature-transformation diagram that describes theausforming process. Research has shown thatausfor
24、med martensite resulting from deformedaustenite possess substantially higher strength ascompared to conventional martensite transformedfrom undeformed austenite. Up to 50% increaseintensile and yield strength was reported in varioussteels depending on amount of deformationinduced during ausforming.
25、21-25 Over 600%increase in rolling contact fatigue of ausformedcylindricalM50steelspecimenswasdemonstrated,and degree of B10fatigue life improvementincreasing with the amount of deformation. 26Bambergerreportednine-foldincreaseinB10lifeofausformed M50 steel bearings over conventionallyheat-treated b
26、earings. 27Figure 1. Ausform process schematicPennStatestechniqueappliesausformingtolocal-izedsurfacelayers ofcontactingmachineelementssuchasgears.28-36 Figure2showsaschematicdescription of the ausforming process as a geartoothfinishingoperationforatypicalcasehardenedlow alloy steel gear, and involv
27、es induction auste-nization,marquenching,rollfinishingandthencool-ingtomartensite. Ausformfinishingofspur andhe-licalgears results invery finesurfacefinishof 4 to8in Ra in both the radial and tangential tooth profiledirections. Furthermore,fullyoptimizedausformingtoolingand process have beendemonstr
28、ated tore-sult in finished gear tooth accuracy of less than0.0002” in both the profile and lead inspections.Fine surface finish and gear teeth accuracy havebeen shown to contribute significantly to improvedsurface fatigue performance. Cycle times involvedin ausform gear finishing are of the order of
29、 severalseconds per gear as compared to several minutesforgeargrinding,andthereforetheprocessiscapa-ble of integration in large production applications.5Figure 2. Ausforming process stepsAusform finishing of P/F steel gearsDoubledieausform gear finishingmachine at PennState was used to develop the t
30、ooling and processvariables to process P/F steel gears. The first stepwas to establish the dual frequency induction heat-ing process to be used to austenitize the case priortorollfinishingofgearteethinmetastableausteniticcondition to final dimensions. The requiredinduction heating process parameters
31、 wereestablishedfor theP/F gears by experimentalitera-tions. X-ray diffraction measurements showed acompressive residual stress at tooth surface in theroot fillet region of about 31 ksi for the ausformedP/F steel gear as compared to about 22 ksi mea-sured for preausformed P/M steel gears.The next st
32、ep to ausform finish P/F steel gear wasto establish gear roll finishing tooling and process.Ausform finishing resulted in reduction of chordaltooth thickness of about 0.003 to 0.004”. Rollfinishingoperationofausformingrequiresoptimiza-tionof therollingdietoothprofileinorder toachievethedesiredfinish
33、edgeartoothaccuracy. Therollingdietoothprofilesmustbemodifiedaway fromnomi-nally involutetoothshape,anddietoothprofileopti-mization typically requires several experimental it-erations. For thecurrent program to ausform finishP/F steel gears, the initial die tooth geometry wasestimated based on numer
34、ical FEA based processmodelingaswellaspriorresultsfromsimilarlysizedtest gears. Figure 3 shows the rolling die toothprofileusedonthedriveandcoastsidesforausformfinishingofP/Fsteelgears. Figure4and5comparethe profile and lead charts of P/F steel test gearsbefore and after ausforming, and show the en-
35、hancements in both accuracy and surface finish ofausform finishedgear teeth. Profilecharts infigure4 show the profile accuracy on right flank of gearteeth to be less than 0.0001” achieved across thecontact region of the teeth. Furthermore, ausformfinished tooth profiles in figure 4 also show thedesi
36、redtipreliefofabout0.001”implementedbytherolling dies. Die tooth profile require further devel-opment to optimize the profile accuracy on the leftflanks which shows a localized hollow of 0.0003”.Figure 3. Ausform finishing rolling die tooth profiles6Figure 4. P/F gear tooth profile charts: preausfor
37、m (left) and ausform finished (right)Leadcharts shown infigure 5also demonstratetheability of ausforming process to produce uniformtooth surfaces with very fine surface finish.Ausform finishedP/F steelgearsshowacharacter-isticleadcrown- astraightregioninthemiddlewithedges rounding that falls off by
38、over 0.001” - that isinherently produced as a result of the inductionheating and roll finishing process characteristics.As a result, the effective contact of test gears wasreduced to an effective face width of about 0.37”,requiring the power circulating test conditions to beadjusted as described lat
39、er to achieve equivalentcontact stresses for ausform finished P/F steelgears as compared to the baseline wrought steelgears.Ausform finishing of P/F steel gears resulted inadditional densification in surface layers.Densification effect due to ausforming of P/F steelgears was determined by weight mea
40、surements inair and water of sections cut from preausform andausform P/F test gears. Preausform P/F gearswhich were already nearly fully dense weremeasured to have an average density of 7.81 g/cc,and subsequent ausform finishing of P/F steelgears resulted in an average density (of a sector ofgear)of
41、7.83g/cc. Thedensificationduetoausformfinishing is localizedin thesurface layers of theP/Fsteel gear teeth.After ausform finishing, post processingoperationsincluded final tempering operation, deburring andmachiningofboreandendfacestofacilitatesurfacedurability testing.Figure 5. P/F gear lead charts
42、: preausform (left) and ausform finished (right)7Baseline and mate gearsMate wrought steel gears required for powercirculating surface durability tests as well asbaseline wrought steel test gears required for com-parative performance evaluationwere producedbyNew Process Gear, Syracuse New York, usin
43、g aprocess sequence as described in table 3, and toheat treatment specifications as described above.Figures 6 and 7 show profile and lead charts forbaseline and mate gears.Gear performance testingGears can fail due to bending fatigue, surfacedistress due to subsurface shear induced pittingfatigue, w
44、ear, surface and or subsurface pittingfatigue due to and initiating at intermetallicinclusions, scoring of tooth surfaces due to break-down of lubrication film, and fracture due to impactloadingconditions.37 Test resultstoestablishthecomparative performance of ausform finished P/Fsteel gears are pre
45、sented in the following sections.Tooth bending fatigue testingSingle tooth fatigue (STF) testing of individual gearteethhasbeenusedwidelytogenerateacceleratedbending fatigue data at comparatively high cycleswithout risk of losingtests toother modes of failure.STFtestingof ausformfinishedP/Fsteelgear
46、sandbaselinewrought steelgears was carriedout ona5kip servo-hydraulic universal fatigue testing ma-chine utilizing a specially designed test fixture toholdthetestgearandtofacilitatefatigueloadingviaa flexure arm. STF test fixture is shown in Figure8Figure 6. Profile charts: baseline gear (left) and
47、mate gear (right)Figure 7. Lead charts: baseline gear (left) and mate gear (right)8Figure 8. STF test fixtureFigure 9. STF test loading detailsFigure 9 shows the loading geometry for the testgears, showing the location of cyclic test load at1.668” radius (35.2 roll angle), as wellas thewidthand heig
48、ht of the critical section in the root filletregion determined by the Lewis parabola. Twoteethweretestedsimultaneously by applyingcyclicload at the above radius as shown in figure 10 untilrun out (defined as no failure after seven millioncycles) or when one of the teeth failed called instatisticalan
49、alysis as asudden deathtest. Suddendeath fatigue tests are widely used in the bearingindustry whereinseveral bearings are testedsimul-taneously until one bearing fails, then all bearingsare replaced. Statistical analysis of sudden deathfatigue tests takes advantage of improved statisti-cal confidence due to the other parts not havingfailed yet. In this case of one of two teeth failing,rankingtheorydeterminedthatthelowestvalue,inaset of two, clustered about a median location ofB29.29point of the population. Also, as both testte
