1、07FTM15Experience with a Disc Rig Micropitting Testby: Dr. M.G. Talks, QinetiQ Ltd., and W. Bennett, DefenceEquipment and Support MoDTECHNICAL PAPERAmerican Gear Manufacturers AssociationExperience with a Disc Rig Micropitting TestDr. Miles G. Talks, QinetiQ, Ltd., and W. Bennett, Defence Equipment
2、andSupport, MoDThe 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.AbstractThe experimental work carried out was aimed at developing a test method that was able to consistently
3、produce micropitting damage and could discriminate between a good oil (i.e., one that rarely producesmicropitting in service) and a poor oil (i.e., one that does produce micropitting in service). The small-scale3-Disc test rig that was used for this work employs 3 discs to apply the test load to a 1
4、2mm-diameter testroller.Thistestgeometryallowsalargenumberofstresscycles(typically600,000to800,000cycles/h)tobegenerated at the contact track on the roller.Thedisc rigcontrolsystemallows testparameters suchas entrainmentvelocity,contactstress andslide/rollratio at the disc/roller contacts to be accu
5、rately and independently controlled. This enables the effect of keyparameters to be studied in isolation, which is something that cannot be easily achieved using conventionalgear test rigs.Theearlyworkcarriedoutusingthediscrigwasaimedatproducingmicropittingdamagebyoperatingtherigat contact condition
6、s similar to those used in the FZG micropitting gear test method. These early testsconfirmedthatthedamageproducedtotherollertrackexhibitscharacteristicsthataretypicalofmicropittingdamage,andshowedthattheseverityofthemicropittingproducedwasaffectedbytheamountofrunning-incarried out on the roller prio
7、r to applying the full test load.A test procedure has been developed which provides a good level of repeatability and which allowsdiscriminationbetweenoilswhichproducemicropittinginserviceandthosewhichdonot.Inaddition,astudyof the effect of slide/roll ratio (SRR) has shown that the severity of micro
8、pitting damage increases as SRRincreased,whereasat0%SRRnomicropittingoccurredand,atnegativeSRRs,microcrackingoccurredbutnot micropitting. This is the way that SRR seems to affect micropitting in gears.Copyright 2007American Gear Manufacturers Association500 Montgomery Street, Suite 350Alexandria, Vi
9、rginia, 22314October, 2007ISBN: 978-1-55589-919-61Experience with a Disc Rig Micropitting TestAuthors: Dr. Miles G. Talks (QinetiQ Ltd, UK) and Mr. W. Bennett (Defence Equipment60%SRR; 1.7GPa contact pressure; 0.45 mm Ra discs;and 90C oil temperature. In all of the five tests, asmallamountofmicropit
10、tingwasproducedafterthefirsthourof testing,about halfof theroller trackwas7micropitted after 10 hours and virtually all of thetrack was micropitted after 20 hours and 30 hours.The mean track profile losses obtained during thetests areshowninFigure6. Infour ofthefivetests,theprofilelosswasbetween8.6a
11、nd10.5mm. Inthefifth test, the profile loss was a little higher (14 mm)after the same test period.Figure 6. Profile losses for the fiverepeatability tests carried out using Oil ADiscussionAchievement of the initial objectivesEstablishmicropittingasthedominantdamagetypeFor allof thetests carriedout u
12、singa positivevalueof SRR, where roller track damage occurred it tooktheformofaxialmicrocracks(i.e.,cracksorientatedacross the track). These were typically up toabout50 mm long in the early stages of thetest, but thesesubsequently developed and joined up to form mi-cropits that were typically about
13、50 mm wide and300 mm long. Examination of the sectioned rollertrack using optical microscopy (after coating thetrack with electroless nickel to stabilize it and pre-ventthelossofloosemicrocrackedsurfacematerialduring sectioning and polishing) revealed micro-crackswhichpropagatedatashallowangleof(typ
14、i-cally20to30degrees)tothesurfacebeforebranch-ing towards the surface (Figure 7). Thesecharacteristics are typical of the micropitting pro-cess(7), (8)and indicate that the damage that wasproduced during this study is, indeed, micropitting.In addition, the appearance of the micropitted sur-face (e.g
15、., as shown in Figure 5) is very similar totypical micropitting damage produced in the FVAtestandothergear-basedmicropittingtests.7,8Figure 7. A typical section of the micropittingdamage to the roller track(NB: The coloured area is the electroless nickelcoating which was applied to stabilize the sur
16、face)Achieve an acceptable level of repeatabilityThe results of the five repeatability tests showedthat the level of the micropitting damage producedafter 30hours of testingwas consistent andrepeat-ableintermsofappearanceandprofileloss. Infourof the five tests, the profile loss results were verycons
17、istent. The mean profile losses after 30 hoursof testing for each of these tests was within 10% ofthe average mean profile loss for the four tests. Inthe fifth test, the mean profile loss after 30 hourswas 40% higher than the average for the other fourtests. Theseresults indicatethat thetest methodi
18、scapableof generatingmicropittingresults that havean acceptable level of repeatability.Achieve adequate discrimination between the tworeference oilsTheresultsofthisstudyhaveshownthattheseveri-ty of the micropitting produced in the disc rig test isaffected by many factors. These include contactstress
19、, SRR, surface roughness, oil type and disc/roller running-in conditions. These are factorswhich have been found to affect micropitting ingears. Itisreasonabletoexpectthatiftheoperatingconditions used in the disc rig are too aggressive,severe micropitting damage will be produced withboth of the refe
20、rence oils and discriminationbetweenthemwillbepoor. Conversely,ifthecondi-tionsarenotsevereenough,littlemicropittingwillbeproduced with either oil and, again, the discrimina-tion will be poor. Consequently, if adequatediscrimination between oils is to be achieved, it isimportant to carry out the tes
21、ts at an appropriate8level of severity. For the two reference oils used inthis study, andusing0.30mmRadiscs, thebest dis-crimination was obtained at 60% SRR at the1.7GPa and 90C test condition. NB: In gears,nearly all of the micropitting damage that developsdoessoinregionswhereSRRsofgreaterthan40%oc
22、cur. It is possiblethat further development of thetest method could result in further improvements inthe discrimination but the important feature of theresults obtained so far is that the test can discrimi-natebetweentwogearoilswhichareknowntohavedifferent micropitting performances in serviceapplica
23、tions.Comparison of the disc rig results with FZGgear test resultsFigure8shows thepositionofthemicropittingbandafter a typical gear-based micropitting test usingthe FZG C-type gears. The band of micropitting(whichhas beencolouredtohighlightits position)isbelowthepitchline,intheregionwherehighcontact
24、stresses and high positive values of SRR occur. Agraphof profileloss inamicropittingtest carriedouton Oil A, using theStrama gear test rig, is showninFigure 9, together with a summary of the SRR val-ues at various positions on the gear flank. Thisshows that most of themicropitting damageoccursinthed
25、edendumofthetestgearinaregionwhereacombination of high contact stresses (about1.6 GPa) and high SRRs (typically up to about+80%) occurs. The results from the disc rig exhib-ited similar behavior, with the highest levels ofmicropitting being produced using a high positiveSRR (+60%) and contact stress
26、es of 1.7 GPa and1.85 GPa.Figure 8. Micropitting damage (the colouredband) to an FZG C-type gear after a typicalmicropitting testFigure 9. Profile loss of an FZG C-type micropitting gear after testing with Oil A9ConclusionsA series of tests carried out using the miniaturethree-disc test rig has show
27、n that it is able to pro-duce micropitting damage and to achieve an ac-ceptable level of repeatability. In addition, once asuitable set of test conditions has been identified, itisabletodiscriminatebetweengearoilswhichhavedifferent micropitting performances.Aswithgear-basedmicropittingtests(e.g.,the
28、FVAtest), adequate control of the surface roughness ofthe test specimens is important if acceptable re-peatability and adequate discrimination are to beobtained in the disc rig test. It is possible that theeffect of disc roughness is less pronouncedathigh-erSRRs(e.g.,60%),however,duetotheoverridinge
29、ffect of the sliding conditions.Theseverityofthemicropittingdamageproducedinthediscrigvariesinasimilarmanner withoperatingparameters such as surface roughness, SRR andsliding direction, as it does in gears. Micropitting isnot producedunder purerollingconditions andverylittle is produced at a negativ
30、e SRR, which is thesame pattern as that observed in gear-based mi-cropitting tests.Micropittingcanbeproducedusingthediscrigwith-inabout 25 millionstress cycles, which is compara-blewiththecycles involvedingear-basedmicropit-ting tests. The disc rig test contact conditions arethus similar in severity
31、 to those occurring in testgears. Asthetestrollerissubjectedtoahighstresscyclerate(typically810,000cycles/h),however,thedisc rig test is potentially a much quicker test thangear-based micropitting tests.Becauseofthesmallvolumeoftestoilusedandtherelatively simple specimen design, the miniaturedisc ri
32、g micropitting test should be a useful tech-nique for future assessments of gear oils and addi-tives, and gear materials, as well as for studies ofcoatings and the effects of surface finishing tech-niques. In addition, the disc rig allows more funda-mental studies of micropitting to be carried out a
33、tspecific combinations of gear parameters (such ascontact stress and SRR). This type of study couldnot be performed using conventional gear tests.References1 Shotter, B.A., Micropitting: Its Characteristicsand Implications on the Test Requirements ofGearOils, ProceedingsoftheInstituteofPetro-leum, N
34、o.1, pp 91-103, London, 1981.2 Ariura,Y.,AnInvestigationofSurfaceFailureofSurfaceHardenedGearsbyScanningElectronMicroscopy Observations, Wear, 87, pp 305 -316, 1983.3 Benyajati, C., An Experimental Study of Micro-pitting Using a New Miniature Test Rig, Pro-ceedings 30thLeeds-Lyon Symposium onTribolo
35、gy, Lyon, France, pp135-143, Sept2003.4 Webster, M.N. and Norbart, C.J.J., An Experi-mental Investigation of Micropitting Using aRoller Disc Machine, Proceedings 50th STLEAnnual Meeting, Chicago, Illinois, May 1995.5 Olver,A.V.,Wearof HardSteelinRollingLubri-cated Contacts, PhD Thesis, Imperial Coll
36、ege,London, UK, 1986.6 FVAInformationSheetNo. 54/I-IV,FVAMicro-pittingTest Method, ForchungsvereinigungAn-treibstechnik E.V., Lyoner Strasse 18, 60528Frankfurt/Main, Germany.7 Shaw, B.A. and Evans, J.T., Micro-pits Underthe Microscope, Session 3, Paper 6, BGADrives and Controls Conference, Telford,S
37、alop, UK, March 1997.8 Brimble,K.,AComparisonofMicro-pittingPer-formanceof IdenticalOilsUsing StandardFZGTest Gears and Helical Test Gears, Session 8,Paper 8, Proceedings 2001 Drives andControls, and Power Electronics Conference,London, UK, March 2001.AcknowledgementsTheauthors wouldliketoacknowledgetheUKMin-istry of Defencefor financingthis work. Theywouldalsoliketothank theMoD, QinetiQ LtdandtheBrit-ishGearAssociation(Project6Consortium)forper-mission to publish this paper.
