1、06FTM09Opportunities to Replace Wrought Gears withHigh Performance PM Gears in AutomotiveApplicationsby: U. Engstrm and D. Milligan, North American Hgans, USAand P. Johansson and S. Dizdar, Hgans AB, SwedenTECHNICAL PAPERAmerican Gear Manufacturers AssociationOpportunities to Replace Wrought Gears w
2、ith HighPerformance PM Gears in Automotive ApplicationsUlf Engstrm and David Milligan, North American Hgans, USAand Pernilla Johansson and Senad Dizdar, Hgans AB, SwedenThe statements and opinions contained herein are those of the author and should not be construed as anofficial action or opinion of
3、 the American Gear Manufacturers Association.AbstractPowder metallurgy (PM) enables cost effective production of components with complex geometries such asgears. During the last decades the use of PM components in automotive applications has show continuousgrowth. In order to continue this growth it
4、 is important to maintain the strong points of PM, i.e. the net shapecapabilitywhileperformanceisfurtherimprovedtofulfillmoreandmoredemandingrequirements.Gearsforautomotive applications are complex in shape and require both very high geometrical accuracy in terms ofgear quality and very high mechani
5、cal performance in terms of durability of tooth flank and root. By utilizingselective densification of the teeth the accuracy and mechanical performance requirements can be met at avery low added cost.In this paper a PM process route consisting of compaction, sintering, surface densification and fin
6、ally heattreatment has been studied to assess the feasibility of producing high performance gears by powdermetallurgy.Helicalandspurgearswereusedinthestudywherethedensificationaswellastheresultinggearquality and durability were tested.Copyright 2006American Gear Manufacturers Association500 Montgome
7、ry Street, Suite 350Alexandria, Virginia, 22314October, 2006ISBN: 1-55589-891-21Opportunities to Replace Wrought Gears with High PerformancePM Gears in Automotive ApplicationsUlf Engstrm and David Milligan, North American Hgans, USAand Pernilla Johansson, Senad Dizdar, Hgans AB, SwedenIntroductionPo
8、wder metallurgy enables cost effective produc-tion of parts with complex geometries such asgears. The amount of waste material is often negli-gible as a consequence of the minimal necessity tomachinethepartsandthe useof ecologicallyharm-ful cutting fluids is avoided.The growthof thePM markethas alon
9、g historyandwe have not seen significant downturns in thisgrowth. In order to continue this growth it is impor-tant to maintain the PM strong points, i.e. the netshape capability while performance is improved. Inorder to expand the PM market new applicationscouldbewonbycombiningtheinherentadvantages
10、of PM while performance is increased by surfacedensification.Gear rolling is a well-known technology for improv-ing the shape and surface finish of solid steel gears1,2,3,4. When applied on PM gears, rolling alsoenhances the fatigue properties since the densityandhencethestrengthinthehighly loadedre
11、gions,i.e. the surface regions, increases to full density 2,3, 5, 6, 7. This technique was first developed morethan 20 years ago, but has not reachedwidespreaduse due to several reasons. The difficulty in obtain-ing good densification and good gear quality simul-taneously has limited the applicabili
12、ty and cost ef-fectivenessoftheprocess.Alsothelongdevelopingtimes have hindered the utilization of the surfacedensified gears. Improvements in rolling machinesand increase in knowledge and experience havesignificantlychangedthesituation7.Severalpartsproducers have after significant and committed de-
13、velopment work started to offer surface densifiedcomponents to the market 5, 8, 9. Another factorlimiting the use of surface densification technologyhas been the continuous market growth. In fact aslong as the requirements fromthe moredemandingapplications could be met using existingtechnologies the
14、 driving force for developing sur-face densification technology was limited.This paper aims atpointing outsome possibilitiesinusing the combination of powder metallurgy andsurface densification. It will be demonstrated thatthe performance oftwo surfacedensified gearscanbe adequate for highly loaded
15、applications. Thegear quality is also presented and discussed.Gears studiesThe gears examined in this study are the fifth gearused in a manual gearbox in a passenger car andthe planetary gear in the reduction unit in a heavytruckgearbox.ThetwogearsareshowninFigure1.Figure 1a. Fifth gear of a manualt
16、ransmission in a Volvo passenger car.Figure 1b. Planetary gear of a reduction unitin a Scania heavy truck gearbox.Typicaldataofthegearsarepresentedintable1. Inthis study both gears were manufactured by com-bining the PM process route with gear rolling. Thehelical gear was redesigned in order to faci
17、litatetesting in a pitting test rig. The clutch gear was re-moved,thediameteroftheborewasenlargedandakeyway was introduced.2Table 1. Gear data.Spur gear HelicalgearNumber of teethZ20 28Normal modulemn(mm) 3.650 2Pressure anglen22.5_ 15_Helix angle 0_ 32_Addendummodification co-efficient x0.471 0.136
18、Over ball diam-eter (mm) 89.350 74.46ExperimentalThe investigated gears were pressed and sinteredto cylindrical blanks. The cylinders were machinedto larger dimensions than on the drawing to havesome stock material to compress during surfacedensification. The work piece was rolled betweentwo dies in
19、 order to densify the surface of the toothflank and root regions. The surface of the compo-nentwascompressedandthedensityinthesurfaceincreased to full density. The tooth-root-fatigueproperties of the gears were tested in an electro-magnetic pulsator.MaterialsReference gears of wrought steels were ma
20、nufac-tured by turning, hobbing, shaving and heat treat-ment.Thematerialsusedforthesegearsareofsim-ilar type (designation) as the material for the gearsused in production. The planetary gear is manufac-tured using the Swedish standard steel SS92506which corresponds to DIN 20NiCrMo2 and AISI8620. The
21、 helical gear is made in DIN 16MnCr5,which corresponds to AISI 5115 and is very similarto AISI 5120.The powder mix selected for this study was Astaloy85 Mo (Fe-0.85Mo, FL4400) with low graphite con-tent. The compositions in the sintered condition aregivenintable2.Table 2. Chemical composition of sin
22、teredand solid steel.Gear MaterialC(%)Mo(%)Cr(%)Mn(%)Ni(%)Spur gearReference SS 92506DIN 20NiCrMo2AISI 86200.20 0.20 0.55 0.50 0.55PM gear(SG)Astaloy 85MoFe-0.85MoFL44000.28 0.85 - - -Helical gearReference DIN 16MnCr5AISI 51150.16 - 0.95 1.15 -PM gear(HG)Astaloy 85MoFe-0.85MoFL44000.20 0.85 - - -Man
23、ufacture of PM gear blanksPM gear blanks were manufactured by compactionofcylindricalblanksfollowedbysinteringat1120_Cfor 30 min in a mixture of 90% N2and 10% H2gasatmosphere. The blank densities were 7.15 g/cm3forthespurgearand7.20and7.10g/cm3forthehe-licalgear.Thesinteredblankswerethenturnedandhob
24、bed to the desired preform shape. The amountof stock material is estimated using simple calcula-tions of material needed to achieve the requireddensification.Surface densification by rollingIn the present work a HC20CN rolling machine(Escofier Technologie SA. Chalon. France) wasusedforthedensificati
25、ontrialsasshowninFigure2.Figure 2. Surface densification of a helicaltransmission gearTherollingcyclecanbedividedintothreeparts;theinitiation where the tool and the gear mesh andmakecontact.Thesecondphaseiswhentheloadisapplied and the tools deform the gear until a presetcenter distance has been reac
26、hed. The third phaseisthetoolwithdrawalwhere thetools moveoutwardsothatthegearcanbeunloaded.Inaserialproduc-3tion set-up a cycle time including loading and un-loading would be 8-15 seconds.Heat treatmentThe heat treatments selected for the two differentgearswerenotthesame.The spurgear wascarbu-rized
27、 at940oCwith acarbon potentialof 0.8%Cfol-lowed by quenching in oil. A stress relieving opera-tion was performed at 160 _C for 60 minutes.The helical gear was gas carburized at 920_Cfol-lowed by a quench in oil at 80 _C. The stress reliev-ing was performed at 170 _C for 60 minutes in air.The side fa
28、ces of the gear were masked using acommercialmaskingpasteinordertolimitthepene-tration of carbon through the low-density sidefaces.Ineachcaseisthecarburizing timeselected sothatthe required case depth should be reached for thedifferent gears. The case depth is defined as thedepth below the surface w
29、here the hardness hasdropped to 550 HV0.1.Gear qualityThe geometrical quality of the gears was measuredby a Klingelnberg P26 3D measuring center. In allcases were the reference axis defined by the gearteeth themselves. The target gear quality for all thegears was DIN quality 8. Some limited trials w
30、ereperformed in order to assess the consistency of theprocess. The diameter over balls was measured onthe blanks and on rolled gears. The densification ofthegearswasassessedbyusingimageanalysisonpolished samples. The response to the heat treat-ment was examined using etched microstructuresof tooth f
31、lank and root regions. The heat treatmentwas further examined using micro hardness mea-surements. Adensification depthwas arbitrarilyde-finedasthedistancefromthesurfacewherethepo-rosity has increased to 2%.TestingThe spur gear was tested in a servo-hydraulic test-ing machine. The load was applied on
32、 the flank ofthetooth.The loadratio was0.1. Samplesreaching2 million cycles were considered as run outs. In all15 tests were performed. The helical HG-Mo gearwas tested in an electro-magnetic pulsator. Thetestrigwasfittedwithaspecialclampingdevicethatappliedtheloadatthetopofthegearteeth.Further-more
33、, a fixture ensured that the gear could not turnor twist during loading. The load ratio was also herekept at 0.1 in order to ensure contact between theclamping device and the gear teeth at all times. Asthe load is applied at the tooth top and not on thetoothflankandthedifferenceingeometryintherootre
34、gion it was not attempted to calculate the maxi-mum bending stresses in the root region. In thiscase the specimens are considered run-outs whenthey reach 30 million cycles.ResultsDensificationThe result of the densification of the spur gear canbe seen in Figure 3. The densification depth isapproxima
35、tely 0.5 mm on the flank and slightly low-er in the root region. For the helical gear the situa-tion is the reverse as can be seen in Figures 4. Thedeepest densification is in the root region reaching0.4-0.5 mm. On the flank the densification depth isapproximately 0.2 mm.2mmFigure 3. Metallographic
36、cross section of thesurface densified spur gear.Figure 4. Metallographic cross section of thesurface densified helical gear.Gear qualityThe gear quality obtained for the two gears areshown in table 3. The quality of the spur gear afterrollingisestimatedtoDINquality8.Thereisonepa-rameter that falls o
37、utside DIN 8. The total profiledeviation reaches 26 mm and the requirement forDIN 8 is 25 mm. DIN 9 allows 36 mm of total profiledeviation. The gear quality ofthe helicalgear isDIN8 or better.4Table 3. Measured gear quality after rolling.Spur PMgearHelicalPM gearProfileForm deviation ff?88Angular de
38、viation FH?68Total deviation F?9* 7HelixForm deviation ff?78Profile deviation fH?77Total deviation F?87Single normal pitchdeviationfu6 6Difference betweenadjacent pitchesfp74Total cumulative pitchdeviationFp55Cum. circular pitcherror over z/8 pitchesFpz/855Tooth thicknessvariationRs7 4Runout toleran
39、ce * Fr62* 1 mm from gear class 8* The gear bore is not used as reference. The rotationaxis is calculated by measurements on the tooth flanksConsistency of gear qualityMeasuring the diameter over balls before and afterrolling makes it possible to assess the consistencyof the rolling process. After h
40、obbing of the helicalgear the scatter in diameter over balls was approxi-mately 0.0225 mm as can be seen in Figure 5. Afterrolling the standard deviation dropped to 0.0031mm. The helical gear was also used to assess theconsistency in gear profile. Four teeth were mea-suredforeachgearandtheresultsare
41、plottedinthescatter diagram in Figure 6.Case hardeningThe hardness profiles for the tested gears areshown in Figures 7 and 8. The target case depth ontheflankwas0.7-1.2mmforthespurgearwhiletheprofiles for the helical gear had a target of 0.3 0.4mm. The case depth for the spur gear is 1.0 mmwhich is
42、well within the target range. The helicalgear had a higher surface hardness compared tothe reference gear. On the other hand it exhibits amore shallow case depth of approximately 0.5 mmcompared to 0.70 mm for the reference gear.74,3574,4074,4574,5074,5574,6074,650 1020304050Gear #Overballdiameter(mm
43、)before rollingafter rollingFigure 5. Scatter in diameter over ballsmeasurement before (diamonds) and afterrolling (circles). Scatterband3 m is indicatedby shadowing. Upper and lower tolerancesare shown as solid lines.0246810121416180 5 10 15 20 25Gear #DIN 9DIN 8DIN 7Profileformdeviation(ff)3 scatt
44、erbandFigure 6. Scatter in profile form deviation.Four teeth are measured on each gear.020040060080010000,0 1,0 2,0 3,0 4,0Distance from surface (mm)Hardness(HV0.1)PM spur gearReference spur gearFigure 7. Micro hardness profile for the spurgear.5020040060080010000 0,5 1 1,5 2 2,5Distance from surfac
45、e (mm)Hardness(HV0.1)Helical PM gearHelical reference gearFigure 8. Micro hardness profile for the helicalgear.Distortion after heat treatmentThe profile and lead traces of the helical HG-Crgear is shown in Figure 9. The hobbed gear has aquality of DIN 10 regarding the profile form devi-ation,whilee
46、.g.theprofileangulardeviation isquitegood at DIN 7. The lead errors are better betweenDIN 7 and 9. After rolling some improvements canbe seen. The profile form deviation has decreasedto DIN 8 near DIN 7, while the profile angular devi-ation is a DIN 8 near DIN 9. In the lead trace thechamfering can
47、be clearly seen. After heat treat-ment it can be seen that the profileform deviationisvirtually unaffected, while the profile angular devi-ation and the profile total deviation changes signifi-cantly. The lead trace shows a similar behavior inthat the helix form deviation is unaffected while theheli
48、x total deviation and helix angular deviationchanges slightly.DiscussionDensificationIdeally a larger densification depth would be pre-ferred. However, especially in the flank region it isdifficulttoreachdeeperdensificationdepthswithouttoo severe deformations. If too much densificationis attempted,
49、the work piece may break due to frac-tureofteethnearthetop.Thismayleadtobreakageof the rolling dies. In the present investigation wehave reached a densification depth of 0.5 mm forthe spur gear and 0.2 mm for the helical gear. How-ever, the densification depth isarbitrarily definedasthe depth where 2% porosity is reached. From acasehardeningpointofviewitissufficienttoreachastateofclosedporosity(around7.2g/cm3)werethematerialstartstobehavelikesolidsteel.Thisregionis significantly thicker in fact it is in the same rangeasthetarget
