1、06FTM12A Crane Gear Failure Analysis - Case Study,Observations, Lessons Learned,Recommendationsby: R.J. Drago, Drive Systems Technology, Inc.TECHNICAL PAPERAmerican Gear Manufacturers AssociationA Crane Gear Failure Analysis - Case Study,Observations, Lessons Learned, RecommendationsRaymond J. Drago
2、, Drive Systems Technology, Inc.The 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.AbstractCranes, large and small in a very wide variety of types, are ubiquitous in just abou
3、t every industrialenvironment.Manyhavebeeninservicefor20,30,and40yearsormore!Thisisespeciallytrueforverylargecranes such as those used in heavy industrial environments including metals processing, automotivemanufacturing, mining, etc. The basic technology of crane design has advanced in many areas i
4、ncludingmotors,controls,humaninterface,andwireropedesign,amongothers.Oneareathathasnotchangedverymuch over the years, however, is the design and manufacture of the gearboxes that form major links in theoverall crane system. In general, the gearboxes used in cranes have proven themselves to be reliab
5、le workhorses capable of delivering years of service with minimal maintenance. Recent crane gear failures (one ofwhich, most unfortunately, resulted in a fatality), however, gave rise to a reevaluation of the design,configuration, and manufacture of the gearboxes in large cranes.In addition, as part
6、 of a recent failure investigation, it became obvious that while the cranes themselves aresubjected to routine, thorough safety inspections, the internal gearbox components are not generallyevaluatedperiodicallybyskilledgeartechnologiststrainedtorecognizepotentialgearsystemproblems.Forexample,inarec
7、ent,specialseriesofcranegearboxinternalinspectionsgeardefectsofapotentiallyseriousnature were identified in more than 37% of the gearboxes examined!Since crane gearboxes do not operate either for long periods of time or “continuously,” as do most othergeared systems, gear system fatigue characterist
8、ics have not been in the forefront of crane gear systemoperation.Recentstudieshave,however,indicatedthatinmanycasesusageratesandloadingand,inmanycases both have increased dramatically. In some applications, “production” (loaded crane usage) hasincreasedbyfactorsoftwoorthreeorevenmorewhileunitloading
9、hassimilarlyincreased.Thismuchhigherusage makes the cumulative effects of fatigue much more important in these typically intermittent usedevices.This paper presents a case study of one particular crane gear failure, including both the failure analysis andresultant remedial actions, along with a disc
10、ussion of the results of and implications from extensive gearboxinspectionsthatwereconductedasaresultoftheinitialfailure.Basedonthisexperience,additionalanalyseswereconductedonhistoricalinspectiondatathatprovidedthebasisforacriticalanalysisofcranegearusage,failure incidence, and design practices. In
11、 addition, recommendations for specific actions required to insurecrane gearbox safety and improve reliability in todays much more challenging environment are presentedand discussed.Copyright 2006American Gear Manufacturers Association500 Montgomery Street, Suite 350Alexandria, Virginia, 22314Octobe
12、r, 2006ISBN: 1-55589-894-71A Crane Gear Failure Analysis Case Study, Observations,Lessons Learned, RecommendationsRaymond J. Drago, Drive Systems Technology, Inc.The Initial IncidentInthespringof2004,afailureoccurred ina largein-dustrial crane system. As the crane was lifting aheavy load, the load b
13、ecame “hung.” That is, it wasneither possible to raise the load further nor lower ittothefloor.Sincesuchproblemsareoftenrelatedtoelectricalissuesincludingmotorfailureandcontrolsissues, an electrician was dispatched to troubleshoot the problem. The electrician climbed aboardthecranetrolleyandproceede
14、dtoworkthecontrolswhile observing the motor shaft. He observed thatthe shaft turned in the appropriate direction whenthe controls were activated. A mechanic was thendispatched to examine the mechanical portion ofthe crane, including the gearbox.The mechanic examined the couplings and othervisible po
15、rtions of the drivesystem butfound noap-parent problems. The sheet metal cover, similar tothat show in Figure 1,of thefinal drivespur gearsetwasthenremovedsothatthegearscouldbevisual-ly examined.Figure 1. Primary Gearbox, Final Drive SpurGear Set Case and Wire Rope DrumAs he was examining the final
16、drive spur gear set,the mechanic observed an object that appeared tobe wedged in the gear mesh. He reached into thegearbox with a tool in an attempt to free the objectfrom the mesh. Unfortunately, when he did so, theload dropped immediately and precipitously. As theload crashed to the floor, the wir
17、e rope spun fromthe drum in an over speed unwrap. The mechanicwas ensnared by the whipping rope and, tragically,injured fatally. The electrician who was watching ashort distance away on the crane trolley was not in-jured but was seriously shaken by the experience.Thecranesystemwasimmediatelytakenout
18、ofser-viceandsimilarcranesinusewerealsotakenoutofservice while an investigation into the cause of thefailure was mounted.The FailureSubsequent to securing the site, as the gearboxwas being examined, it was discovered that the ob-ject that the mechanic saw in the gear mesh was apieceofkeystock.Examin
19、ationofthegearboxcon-tinuedinanefforttodeterminethesourceofthekey.Thespurpinionwasfoundtobeashelldesign,simi-lartothatshowninFigure2,whichispressfitontheoutput shaft of the primary drive gearbox. A key-way in the bore of the pinion accepts a key that pro-videstorquetransmissionfromtheshafttothegear.
20、This is the key that became entrapped in the gearmesh.Figure 2 - Shell Pinion With Keyway2The pinion that was installed in the failed gearbox,Figure 3, experienced a tooth fracture failure. Ex-aminationofthispinionshowsclearlythatitcrackedthrough the rim, at the keyway.Figure 3 - Failed Final Drive
21、Spur PinionWhile the failure itself is due to fatigue, it is very im-portant to note that the mode of failure, crackingthrough the rim rather than a tooth fracture, is theprimarycauseofthelossoftorquetransmissionca-pability of the pinion when the crack progressed.This is a clear instance of a thin r
22、immed gearfailureby cracking through the rim. In addition, careful ex-amination of the keyway shows that the corners ofthe bottom of the keyway are quite sharp with virtu-ally no significant radius present. This condition ex-acerbates the thin rimmed configuration and pro-vided a very definite crack
23、 progression path.As Figure 4 shows, when a gear blank has a thickrim, if a crack initiates in the tooth root, the progres-sion path is across the base do the tooth, ultimatelyliberating either a piece of the tooth or, if the face isnarrow, the entire tooth.Figure 4 - Comparison of Thick and Thin Ri
24、mFailure ModesWhen the gear rim is thin, it “participates” in thetooth bending and the stresses at the tooth rootchange character dramatically, as Figure 5 shows.Figure 5 - Comparison of Fillet Stress WaveForms for Thick and Thin Rimmed GearsCarefulexaminationofthestresswaveformsinFig-ure 5 shows th
25、at both the magnitude and the posi-tionofthemaximumrangeofstresschanges astherimgetsthinner.Theresultsofthesechangesresultin a change in failure mode (i.e. crack propagatesthrough the rim rather than through the tooth base)and an increase in stress level. Considering thisscenario, it is easy to unde
26、rstand the how the shellpinion failure shown in Figure 3 occurred.This phenomenon was first described in detail andbrought to the forefront when a new factor, the RimThickness Factor, Kb(Figure 6), was added toAGMA Standard 218 based on the work presentedin Reference 1. This factor was carried forwa
27、rdwhen AGMA 218 was replaced with AGMA Stan-3dard 2001 “Fundamental Rating Factors and Cal-culationMethodsforInvoluteSpurandHelicalGearTeeth” in 1995. Prior to the development of the RimThickness Factor, the thickness of the rim support-ingthegearteethwasnotconsideredintheanalysisof the bending stre
28、ngth of the teeth.Figure 6 - Rim Thickness Factor from AGMAStandard 2001The backup ratio, mB, of the shell pinion shown inFigure3,awayfromthekeywayisabout1.33thusitisjust intothethick rimregion (i.e., = 1.2). In there-gion of the keyway, however, the backup ratio isabout 0.66 thus the rim thickness
29、factor in this re-gion is 1.94. This means that the effective bendingstress at the tooth roots over the keyway is 194%greater than the bending stress that would exist atthe roots of the teeth if the rim were thick. This veryhigh stress condition is further aggravated by thesharp corners of the keywa
30、y which are located justbelowthehighlystressedtoothroots.Inlight ofthis,it is not surprising that this shell pinion failed in themanner that it did. It is a classic example of a thinrimmed gear failure mode.Crane Gear System InspectionsNot surprisingly, this failure prompted a program toinspect the
31、gear systems of similar crane gear sys-tems. Ina typicalcrane system inspection, theindi-vidual component parts of the crane - brakes, wireropes,motors,controls,etc.- areinspectedingreatdetail and their condition is documented carefully.Appropriate actions are taken to repair or replacecomponentstha
32、tarefoundtobedamaged,worn,orthat show other signs of distress. In general Cunningham, Roy PA.5. Drago, Raymond J.; Cunningham, Roy & Cym-bala, Steve, “The Application Of Very Large,WeldFabricated, Carburized, Hardened & HardFinished Advanced Technology Gears In SteelMillGearDrives”American GearManufacturersAssociation Fall technical Meeting, TechnicalPaper No. 05FTM19, Detroit, MI.
