1、08FTM08AGMA Technical PaperPM Materials for GearApplicationsBy S. Dizdar, P. Johansson andU. Engstrm, Hgans AB, andI. Howe and D. Milligan, NorthAmerican HgansPM Materials for Gear ApplicationsSenadDizdar,PernillaJohanssonandUlfEngstrm,HgansAB,andIanHoweandDavid Milligan, North American HgansThe sta
2、tements 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.AbstractThispaperreviewsrecentachievementsingeartoothbendingstrengthandrollingcontactfatigueresistanceof powder metal materials in
3、comparison to wrought steel materials for automotive applications. Prototypepowder metal gears and rolling contact fatigue (RCF) rollers were pressed from commercially availablelow-alloyedironpowdermixes,sintered,surfacedensified,casehardenedandoptionallyhardfinishedusingcommercially available equip
4、ment for all manufacturing steps. The reference gears and rollers weremachined from common wrought steels. The results evaluated on the prototypes show that powder metalmaterials meet wrought machined materials gear tooth bending strength and RCF-resistance.Copyright 2008American Gear Manufacturers
5、Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314October, 2008ISBN: 978-1-55589-938-73PM Materials for Gear ApplicationsSenad Dizdar, Pernilla Johansson and Ulf Engstrm, Hgans ABIan Howe and David Milligan, North American HgansIntroductionDevelopments in the automotive industry
6、towardrelatively inexpensive, high performance vehicleswith low fuel consumption and low environmentalimpact put high demands on materials for automo-tive gear applications. Such developments requirecompact size but high strength automotive gear-boxes. Thegearsthatmeetautomotiveindustryde-mandsarema
7、deoflowalloyedsteelsforcasehard-ening such as AISI 5115 (DIN EN 16MnCr5) or8620(21NiCrMo2). Thegearsaremanufacturedbya process route consisting of three principal opera-tionsteps:softmachining, casehardeningandhardfinishing. These gears achieve higher than1000 MPa in gear tooth bending strength FEan
8、d1500 MPa in gear pitting resistance HlimperISO 6336.Conversely, powder metalmaterials have, untiloneor two decades ago, been associated with low costand low performance gear applications, such aspumps,hobbyandhouseholdapplications. Howev-er, the introduction of high density technologies im-provedge
9、ardensitylevelsandbythistheirmechan-ical strength. Warm compaction 1 improveddensitylevelsofpartssuchaspowertoolsgearsupto 7.27.3 g/cm3, and offered a lower cost alterna-tive to double press double sinter (DPDS) route.High velocity compaction 2 offered the possibilityto cost effectively single press
10、 large single levelpowder metal parts such as parking gears up to7.2-7.3 g/cm3. Finally, gear surface densificationtechniques 3 open possibilities to fully densify thetooth surface to a depth that Hertzian contactstresses inthe gear flank andbending stress gradi-ents in the gear tooth root are withi
11、n it.This paper has been aimed to review recentachievements in high performance metal powdergears manufactured by powder metallurgy routesconsisting of pressing, sintering, surface densifica-tion and case hardening.Experiment and materialPerformance levels that powder metal gearsachieve were evaluat
12、ed on simple spur gears fromanautomotiveapplicationandrollersfrom RCFtestby ZF Friedrichshafen. Figure 1 shows theprototypegear,basicfactsaboutthegeartoothrootbending testing and a brief schedule of gearbending stress calculations accordingto ISO 6336.Figure 2 illustrates the ZF-RCF test and lists s
13、omefacts bout the RCF-testing.Gear dataTeeth number z =18mmModule mn= 1.5875 mmPressure angle n=20Face width b =10mmFatigue testingElectromagnetic resonance machine = VibrophoreTest frequency f = 80 - 120 HzStress ratio R = F0 min/F0 max=0.1Test stop criteria:- 3 106load cycles (runout) or 5 Hz freq
14、uency dropTip contact ISO 6336sFn=3.03mmhFa=2.95mmF=1.0mmen= 29.6Tooth root bending stress ISO 6336FO=FpmnbYFaYSaYFa=6hFamncosensFnmn2cosn= 2.83YSa= 1.2 + 0.13sFnhFaq1.2+ 2.3sFnhFa 1s = 1.50qs=sFn2F= 1.52Figure 1. The prototype gear, its gear toothroot bending testing and a brief schedule ofgear too
15、th bending stress calculations4ZF-RCF test variant used in brief- R1/R2= 30/70 mm- Line contact between cylinders- Rotational velocity 3000 RPM- Full fluid film lubrication- Relative sliding 24%- Test lubricants:S Gear oil SAE 80WS ATF Dextron III- Lubricant temperature 80C- Test stop criteriaS Run-
16、out at 50 106load cyclesS Width-through crater on contact surface- RCF Hertzian resistance is highest Hertzianstress at which test rollers survive at least 50 106load cycles. This is estimated by tests at loadlevels with limited fatigue life.Figure 2. ZF-RCF test in briefThePM materials usedinthis s
17、tudy weretheatom-ized prealloyed powder grades Astaloy 85Mo andAstaloy CrL. Astaloy 85Mo is prealloyed with0.85%MowhileAstaloyCrLisprealloyedwith1.5%Cr and 0.2% Mo.The manufacturing process route for each proto-type gear and roller are described in Table 1 andTable 2. The reference gears and rollers
18、 weremachined from wrought round bars and case hard-ened at a heat treat vendor. The case hardening(CQT) was performed according to commonpracticefor components madeof lowalloyedsteelsaimedfor casehardeningi.e. they were gas carbu-rizedfor 30and150minutes at 920C, quenchedin60C oil bath and air temp
19、ered at 200C for 60minutes. ThecommonDIN 550 HV casehardeneddepth (Eht5550HV) was used. The reference gearswere finish ground to achieve gear quality DIN 7.The PM prototype gears and rollers were pressedas D34/d14 x h10 mm rings and D40 x 28 mmcylinders and sintered for 30 minutes at 1120Cin90N2/10H
20、2atmospheretoachievethenominalcoredensity. To achieve core densities greater than 7.1g/cm3, the prototypes were double pressed doublesintered(DPDS). The first press was to 7.0g/cm3,sintered for 30 minutes at 800C in 90N2/10H2at-mosphere, then re-pressed to 7.4 and 7.6 g/cm3densities and then sintere
21、d for 30 minutes at1120C in 90N2/10H2.Therings/cylinderswerethenmachinedtogearandroller blanks with rollingreadygeometry. Thismeans that the gear blanks were hobbed using agear hob with intentionally modified profile to a ge-ometry that includes allowance for flank surfacematerial to be displaced by
22、 the gear rolling die. Atypicaloversizefor testedgears was 0.3 mm intheover ball measure (OBD).Table 1. Material, manufacturing routes and achieved surface densification and hardness of thetest gearsMaterial and base(core) densityManufacturing route SDD0.98/Eht550HV(mm/mm)1 AISI 5115 (DIN EN 16MnCr5
23、)Wrought steelMachining, CQT, geargrinding- - /0.192 Astaloy CrL + 0.2 C (Fe-1.5Cr-0.2Mo+0.2C)7.1 g/cm3DPDS, machining, gearrolling LPC-GQ-T0.20/0.253 Astaloy CrL + 0.2C (Fe-1.5Cr-0.2Mo+0.2C)7.4 g/cm3DPDS, machining,LPC-GQ-T0.20/0.254 Astaloy 85Mo + 0.2C (Fe-0.85Mo+0.2C)7.1 g/cm3Press and sintermach
24、ining, gear rolling,CQT0.20/0.305 Astaloy 85Mo + 0.2C (Fe-0.85Mo+0.2C)7.4 g/cm3DPDS, machining, gearrolling, CQT0.20/0.326 Astaloy 85Mo + 0.2C (Fe-0.85Mo+0.2C)7.6 g/cm3DPDS, Machining, CQT - - /0.255Table 2. Material, manufacturing routes and achieved surface densification and hardness of thetest ro
25、llersMaterial and base(core density)Process route SDD0.98/Eht550HV(mm/mm)1 AISI 8620 (DIN EN 21NiCrMo2) Wrought steel Machining, CQT - - /0.82 Astaloy CrL+0.2C (Fe-1.5Cr-0.2Mo+0.2C)7.1 g/m2Press and sintermachining, rolling, CQT0.7/0.73 Astaloy CrL+0.2C (Fe-1.5Cr-0.2Mo+0.2C)7.6 g/m2Press and sinterm
26、achining, rolling, CQT1.4/0.94 Astaloy 85Mo+0.3C (Fe-0.85Mo+0.3C)7.0 g/cm3Press and sintermachining, rolling, CQT1.0/1.05 Astaloy 85Mo+0.3C (Fe-0.85Mo+0.3C)7.0 g/cm3Press and sintermachining, CQT- - /1.0The gear blanks were then rolled using a commer-cialCNCradial(transverse)rollingmachine(Figure3)
27、and achieved 0.3 mm surface densified layerdepth the depth in millimeters at which poresoccupy 98% of the relative full density (SDD0.98)(Figure 4). The PM gear rolling process is analo-gous to soft finishing of wrought gear 4, aprocessused by some auto manufacturers. The PM rollerblanks had an over
28、size diameter of about 0.5 mmfor rolling allowance, and were rolled using thesame machine as for the PM gears. Reference 5includes an overview of PM rolling technology. AllPM prototypes except those made of Astaloy CrLwere case hardened at the same heat treat vendorusingthesamefurnaceandthesameparam
29、etersinorder to achieve compressive residual stresses bycreatingahard martensitic surface witha soft core.Figure 3. A photograph of a P/M gear rollingin a CNC radial rolling machineThe gears made from Astaloy CrL were low pres-sure carburized, gas quenched and tempered(LPCGQT) by a heat treat vendor
30、. Carburizingwas done at 960C using acetylene gas with flow700 nl/hatpressureof3mbarunderboostcyclesof4, 2 and 2 minutes separated by diffusion cycles ofrespective 6, 11 and 40 minutes. The gears werethengasquenchedwithnitrogenunder10barpres-sure. However, this process was not optimized forthe gears
31、.Figure 4. Pore structure of a surfacedensified gear tooth made of AstaloyCrL+0.2C, with 7.4 g/cm3 core density anddensified depth of 0.3 mmMicrostructureincasehardenedAstaloy 85Moma-terials (Figure 5) consists of plate martensite withhighcarboncontent at the surfaceand lathmarten-site in the core w
32、ith low carbon content. Betweenthesurfaceand thecore therewill bea carboncon-tent gradient where the martensite gradually willhavelowercarboncontent. CasehardenedAstaloyCrL material (Figure 6) is characterized by platemartensite with high carbon content at the surfaceand deeper in the material the c
33、arbon content willdecreaseandtheplatemartensitewillbemixedwithlath martensite. Eventually the lath martensite willbe mixed with more and more bainite so in thecorethe microstructure will be bainite mixed with lathmartensite.6Figure 5. Etched structure of the tooth flanksurface of an Astaloy 85Mo+0.2
34、C, surfacedensified case hardened gear toothFigure 6. Etched structure of the tooth flanksurface of an Astaloy CrL+0.2C, surfacedensified case hardened gear toothMicrohardnessprofilesoftestedgearsareshowninFigure 7. Comparing to the reference AISI 5115gears, surface densified Astaloy CrL gears haveh
35、igher surface hardness but lower core hardnesswhile Astaloy 85Mo gears have similar surfacehardnessbutlowercorehardness. Thereasonsforsuch microhardness picture are likely core porosityand alloying content.Microhardness profiles of the tested RCF-rollersareshowninFigure8. Alltherollers,eventhosemade
36、 of Astaloy CrL, were gas carburized, oilquenchedandair tempered. Microhardness profileof the surface densified rollers made of Asta-loy CrL+0.2C, 7.1 and 7.6 g/cm3density, are closeto the microhardness profile of the referenceAISI 8620 rollers. Microhardness profile of Asta-loy 85Mo+0.3C,7.0 g/cm3d
37、ensityrollers showedaconsiderable difference from the others. The sur-face densified gears achieved a very high surfacehardness while the non-surface densified gearsachieved a micro-hardness profile with a surfacehardness of slightly over 600 HV0.1. A generalcomment for the rollers case hardening is
38、 that sur-face densified powder metal rollers behave thesame as full dense wrought material rollers, whilenon-surface densified rollers powder metal withdensity level of approximately 7.0 g/cm3, and inter-connected pores, more difficult to case hardening.Figure 7. Microhardness profile of testedgear
39、sResults ofthegeartoothbendingtestingareshowninFigures 9and10. Casehardened surfacedensi-fied gears made of Astaloy 85Mo + 0.2% C show ahigh gear tooth bending performance. Those withcore density of 7.4 g/cm3have even higher perfor-mance than case hardened wrought machinedAISI 5115 gears. The Astalo
40、y CrL + 0.2% C gearshaveslightlylowerperformancebutcasehardeningfor this materialwas not optimizedat thetimeofpu-blication. The effect of surface densification on thegear tooth bending performance appears to be again of 100 MPa in gear tooth bending strengthwhen core density increases from 7.1 to 7.
41、4 g/cm3.7However, absence surface densification lowers thegear tooth bendingstrength downto 850 MPaleveldespite a high core density of 7.6 g/cm3.Whencomparinggeartoothbendingstrengthofdif-ferentgearsizesi.e. modules,thesizeeffecthastobeconsidered. ISO6336includesgear toothbend-ing strength levels ev
42、aluated on larger wroughtgearswithmodule3-5mmdependingoncorehard-ness level and Ni content according to ISO 6336.The test gears in this investigation had a relativelysmall module of 1.5875 mm (DP 16). Their geartooth bending strength exceeded ISO 6336 levelsfor 3-5mm modulegears. Thesamephenomenonwa
43、s previously reported by e.g. Jeong in 1992 6,whofoundadifferenceincasehardenedgeartoothbending strength of 22% between modules of 1.5and 5 mm. Dividing 1350 MPa in gear tooth bend-ingstrengthofcasehardenedAISI 1515gearswiththe core hardness of 440 HV0.1 by 1100 MPa forcasehardenedwroughtgearswithco
44、rehardnessof40 HRC (ca. 400 HV), yields a difference of 23%.This is additional evidence of validity of the resultsshown here.Figure 8. Microhardness profile of testedRCF-rollersFigure 9. Gear tooth bending strength FEasa function of gear tooth root surface hardnessFigure 10. Whler curves for test ge
45、ars8RCF-resistanceofthecasehardenedsurfaceden-sifiedpowder metalrollersreached2150 MPa(Fig-ure 11). The best powder metal rollers were madeof Astaloy CrL + 0.2% C with core density of 7.1g/cm3 and surface densification and case depth of0.7 mm reached 2100 MPa level. At nearly thesame level of 1950 M
46、Pa, were Astaloy CrL+0.2Crollers with core density of 7.6 g/cm3, and surfacedensification/casedepthof1.4/0.9mm. Bothrollershad very similar profiles of residual compressivestresses down to 0.5 mm depth. The reasons forthis 150 MPadifference inthe RCF-resistancecanbefoundinotheraspects ofcasehardenin
47、gquality,uniformity of surface densified depth, ratio of casehardened and surface densified depth and coredensity/hardness differences.Astaloy 85Mo+0.3C rollers with density of7.0 g/cm3, casehardenedbut not surfacedensifiedreached only about 1000 MPa. Similarly manufac-tured rollers surface densifie
48、d, reached 1800 MPaillustrating how surface densification increases theRCF-resistance. Benefits of the surfacedensifica-tion are also illustrated in the slope of the WhlercurvesinthelimitedRFC-resistanceregionfor50%survival life according to lognormal probability (Fig-ure 12). Surface densified Asta
49、loy CrL+0.2C, 7.1g/cm3rollerscurveoverliethereferenceAISI 8620curve while the non-surface densified rollers madeofAstaloy 85Mo+0.3Cwithdensityof7.0g/cm3aresignificantly lower.ConclusionsPowder metal materials for gear applications havebeen reviewed based on results of investigationsinto gear tooth bending strength and rolling contactfatigue resistance. Following conclusions werereached:1) Surfacedensifiedcasehardenedpowdermet-al gears reached ge
