1、90 FTM 5POLISHING WEARby: A. Milburn, Milburn Engineering;R. Errichello,.Geartech; andD. Godfrey, Wear AnalysisiAmerican Gear Manufacturers AssociationTECHNICAL PAPERPOLISHING WEARA. Milburn, Milburn Engineering; R. Errichello, Geartech; andD. Godfrey, Wear AnalysisThe Statements and opinions contai
2、ned herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Association.ABSTRACT:Polishing wear has been known to occur on piston rings, cylinder bores, valve lifters, hydrauliccomponents, rolling-element bearings and gear teeth.
3、The bright, polish surfaces may look good, butpolishing wear is detrimental because it is a high wear phenomenon which reduces the geometric accuracyof the components. A case history is presented of a gearbox which suffered extensive polishing wear ofthe gear teeth and rolling-element bearings. The
4、results of research into the basic mechanism of polishingwear, and laboratory analyses of materials and lubricants are presented. Polishing wear is shown to be dueto fine-scale abrasion. It is promoted by a combination of a fine abrasive and a gear oil withchemically-active additives.Copyright 1990A
5、merican Gear Manufacturers Association1500 King Street, Suite 201Alexandria, Virginia, 22314October, 1990ISBN: 1-55589-557-3Polishing WearAndrew MilburnMilburn EngineeringRobert ErrichelloGEARTECHDouglas GodfreyWear AnalysisINTRODUCTIONThe first mode of failure listed in the precursor to scuffing. B
6、ore polishing isAmerican Gear Manufacturers Association believed to be caused by fuels with high(AGMA) Standard i on gear tooth failure sulfur content in combination withmodes is polishing. The Standard gives crankcase oils with low alkalinity 6.examples of polishing wear on hypoid and Oils with low
7、 alkalinity fail to neutralizeherringbone pinions and it states that the sulfuric acid from the fuel.“extreme-pressure additives in thelubricant may have promoted polishing on References 7 and 8, list polishing wearthe tooth surfaces“. References i and among lubrication wear problems, and define2 im
8、ply that polishing is a normal, low- it as “continuous removal of surface filmswear-rate mechanism that may promote by very fine abrasives“.geometric conformity and is not necessarilyharmful. While this may be true in some Reference 9 has shown that silica,cases, this paper will show that there are
9、alumina and soot are polishing wear agents.instances where polishing wear can bedetrimental. The fact that the AGMA CASE HISTORYStandard describes polishing as a failuremode, yet implies that it is not harmful, The following is a case history of ais evidence that polishing wear is not well gearbox t
10、hat suffered extensive polishingunderstood. There is controversy over how wear of the gear teeth and rolling-elementmuch wear constitutes a failure. One bearings.observer may look at a polished gear toothand pronounce it “well run-in“, while Gearbox descriptionanother observer will describe the same
11、gear tooth as failed. This paper will show The gearbox is a two-stage, planetary speedthat polishing wear can be a true failure reducer. Each stage consists of a spur sunmode. gear, three planet gears and a fixedinternal gear. Initially, each planet gearA sulfur-phosphorus gear oil additive has was
12、supported on two spherical-rollerbeen associated with the polishing of bearings. Later, these bearings weredifferential gears, spider gear thrust replaced with cylindrical-roller bearings.washers and side gear thrust washers 3,4. The bearings have machined bronze, roller-No polishing occurred with a
13、 non-chemically riding cages. The gear ratio of the low-active, borate additive. Reference 5 speed stage is approximately 5:1 and thepresents data showing the loss of 76 _m of gear ratio of the high-speed stage is 4:1;metal from a polished gear tooth, giving an overall gear ratio ofapproximately 20:
14、1. The reducer is drivenCylinder bore polishing in diesel engines by a 150 kW, 1800 rpm electric motor.increases oil consumption and is a1The sun and planet gears are made from AISI The teeth of the low-speed planet gears4820 steel and carburized and hardened to also showed polishing wear (see Figur
15、e 2).58 HRC minimum. The teeth of the sun and The flanks of the teeth that mated with theplanet gears are ground to AGMA quality ii. internal gear teeth were highly polished toa mirror finish as shown by the reflectionThe internal gears are made from AISI 4340 of the graph paper grid in Figure 2. Th
16、esteel and through-hardened to 321-363 HB. opposite flanks of the planet gear teeth,The teeth of the internal gears are shaped which contacted the sun gear teeth, showedto AGMA quality i0. no unusual wear and the original grindingmarks were still visible.The gears and bearings are splashlubricated w
17、ith a mineral gear oil whichcontains sulfur-phosphorus, anti-scuffadditives. Initially, a viscosity of ISO320 (AGMA 6EP) was used. Later, the _ 2_ _ _ _ _viscosity was changed to ISO 460 (AGMA _ _ _,_i._ _ _ _“ _r_7EP). The static oil level is 50 mm below _/_ #_%1_ _ _ J_ _ ,_II_the centerline of th
18、e gearbox. _ ._ _Service history _ _P_ The first sign of a problem occurred after _approximately 300 hours of operation under _ _50-60% of full load. Analyses of the _ _lubricant from several gearboxes indicated _9 =iron contents ranging from 200 to 2400 ppm. _#_#_ _i:_ _:_The lubricant was drained
19、and replaced with _ _ _ _fresh oil and the lubricant analyses werecontinued using emission spectroscopy, ,_II_4_ _,_particle counting and analyticalferrography. The analyses confirmed that _unusual wear was occurring. A gearbox was _ :removed from service and disassembled. _ _ _Inspection of the gea
20、rbox componentsrevealed unusual wear in two areas. The _rollers from the high-speed planet bearings ,were worn in an area where they contactedthe bearingcages (seeFigurei). _5ramFigure 2. Low-speed planet tooth showing._! polishing wear.The bearings, sun gears and planet gearsfrom this gearbox were
21、replaced and theworn parts were taken for study. Some ofthe bearings were sent to the bearingmanufacturer for their review. They_m reported that the centrifugal loads on thehigh-speed planet bearings were higher thanrecommended. They suggested that theplanet bearing be changed to a bearing withFigur
22、e i. High-speed planet bearing a cage that is guided by the outer race toshowing polishing wear on the rollers, reduce the inertial loads of the cage onthe rollers. The gear teeth were inspectedA distinct step was visible at the boundary on a Maag PH 60 inspection machine. Nobetween the worn and unw
23、orn areas. The unusual wear was found on the teeth of anyworn surface of the rollers had a highly- of the gears except the low-speed planetpolished, mirror finish. Measurements of gear. This gear had excessive wear on thethe wear step indicated that the polishing flanks of the teeth that contacted t
24、hewear had removed up to 20 _m of metal from internal gear. Figure 3 shows profilethe rollers, charts from a low-speed planet tooth.2Hardness measurements on the high-speedplanet bearing rollers and the low-speedplanet gear teeth indicated that the- -h hardnesses were greater than 58 HRC., TIP TIP,
25、A borate gear lubricant was tried in someof the gearboxes in the field. Analyses ofthe new lubricant showed lower ironcontents compared to oil from gearboxeswith sulfur-phosphorus lubricant, but theROOT ROOT_ wear had obviously not stopped.I , , Full load tests were conducted on new25_m gearboxes at
26、 the gear manufacturers shop.Initial testing with sulfur-phosphorus gear3a 3b oil of ISO 320 viscosity produced the samepolishing wear as was observed on thefield-tested gearboxes. These resultssuggested that the polishing wear was notFigure 3. Profile charts of low-speed caused by environmental abr
27、asives. Aplanet tooth. 3a-unpolished flank. 3b- borate lubricant of ISO 320 viscosity and apolished flank, sulfur-phosphorus lubricant of ISO 460viscosity were tried. Both theseFigure 3a shows that the flank that ran lubricants helped to reduce the degree ofwith the sun gear is essentially unworn, p
28、olishing, but it was not completelyFigure 3b shows that the polishing wear eliminated. The high-speed planet bearingsthat occurred on the flank that ran withwere replaced with cylindrical-rollerthe internal gear had removed approximately bearings with roller-riding, machined30 _m of metal, decreasin
29、g its accuracy bronze cages. Bearings with outer racefrom AGMA quality ii to AGMA quality 8. riding cages were preferred, but were notFigure 4 shows helix charts from a low- available in the size needed. After a i00speed planet tooth, hour test with sulfur-phosphorous lubricantof ISO 460 viscosity,
30、inspection of thegears and bearings showed no unusual wear./ _ The larger cage area in contact with thestraight roller and the higher viscosityoil may have resulted in a thicker oil film and less sensitivity to abrasives. Also,with contact along the entire length ofORIGINAL- I rollers, no wear steps
31、 were formed.Reducing the wear debris from the bearingsmay have suppressed the gear toothPOLISHED-“-. I polishing by reducing the amount of ironI oxide abrasives reaching the gear teeth.i/JJ!/ INSPECTION OF GEARBOX COMPONENTS BY LIGHTj MICROSCOPY, , Apparatus25_mThe gearbox components were examined
32、with a4a 4b stereo microscope with magnification up to250X. Photographs were taken through onelens of the microscope using incident flashFigure 4. Helix charts of low-speed planet illumination.tooth. 4a-unpolished flank. 4b-polishedflank. SpecimensFigure 4a shows the helix chart for an A tooth was r
33、emoved from a low-speed planetunpolished, sun-gear flank. Figure 4b gear (see Figure 2). A roller and ashows the chart for the polished, internal- section of the cage were removed from thegear flank of the same tooth. Comparing high-speed planet bearing shown in FigureFigures 4a and 4b, it can be se
34、en that the I. The roller and cage were examined incrown of the tooth which originally the condition they were received. Theymeasured 30 _m has been completely worn were also examined after the cage was cutaway. These measurements show the high and the parts cleaned in a series ofwear caused by poli
35、shing, hydrocarbon solvents.3Results and discussionGear ToothTo the unaided eye, the polished surface ofthe gear tooth had a bright, mirror finish.When viewed at an oblique angle, thesurface appeared to be slightly wavy with afew, local bumps. Some relatively short,smooth furrows were evident in the
36、 addendumof the tooth extending radially upward fromthe pitch line toward the tip of the tooth.One small area of the addendum at the endof the tooth had traces of the originalgrinding marks.On the opposite, unpolished flank of the 6agear tooth the central portion was smooth,with a matte texture. The
37、 ends of thetooth were rougher with the originalgrinding marks clearly visible.Figure 5 shows micrographs of the polishedflank of the gear tooth, above and belowthe pitch line. The principle feature inthe Figure 5 micrographs is the short, finescratches in the direction of sliding.Intense, incident
38、light was used to revealthe scratches.t J6b 200pmFigure 6. Micrographs of low-speed planettooth on unpolished flank. 6a-addendum.6b-dedendum.i The principle feature in the Figure 6_I micrographs is abrasion marks in the1 direction of sliding that are wider andlonger than those on the polished flank.
39、5a Figure 6a also shows some original grindingmarks approximately perpendicular to theabrasion marks. There were no otherfailure modes such as scuffing, pitting orcorrosion on either flank of the geartooth.The main difference between the polishedand unpolished flanks of the gear tooth wasthe charact
40、er of the abrasion marks whichtended to be shorter and finer on thepolished flank of the tooth. The short,fine scratches are consistent with 2-bodyabrasion caused by embedment of fineabrasive particles in the internal gearteeth. With the particles fixed on theinternal gear teeth, they would be expec
41、ted, , to cut relatively short abrasion furrows onSb 200pm the planet gear teeth. The scratch lengthsincreased with distance from the operatingFigure 5. Micrographs of low-speed planet pitch line; consistent with greater slidingtooth on polished flank. 5a-addendum. 5b- motion. The longer, coarser sc
42、ratches ondedendum, the unpolished flank of the tooth areconsistent with 3-body abrasion caused byFigure 6 shows micrographs of the hard particles entrained in the oil. Aunpolished flank of the gear tooth, above loose particle would be expected to abradeand below the pitch line. a relatively long fu
43、rrow.4Roller Bearingvisual inspection of the raceways of thespherical-roller bearing showed that theywere damaged by indentations and scratchescaused by hard debris in the oil. Theoriginal grinding marks were completelyobliterated. The raceways were notpolished.Composite photographs of a roller in a
44、segment of the cage are shown in Figure 7.Each roller was polished on the outsidediameter for 2/3 of its length and at oneend. These areas corresponded to siteswhere sliding occurred between the rollerand the guiding surfaces of the cage. Thepolished end of the roller had a small,unpolished circle i
45、n the center whereoriginal grinding marks where visible.This area matched a small depression in the 8acage pocket, but was smaller in diameter,indicating that there had been small,relative translation between the cage androller.7a8b I !600pmFigure 8. Macrographs of roller andi_%_ segment of cage fro
46、m high-speed planet_ _ ._i._ bearing. 8a-cage pocket. 8b-roller end./;C/I_ :_ Figure 9 shows macrographs of a cage rib_ _ and the outside diameter of a roller.Figure 9a shows discolored areas and7b 7c abrasion marks on the surface of the cagerib. The left side of Figure 9b shows thatthe polished por
47、tion of the roller was covered with fine abrasion marks which5mm formed continuous scratches around theperiphery of the roller. The right side ofFigure 7. Photos of roller and segment of Figure 9b shows the unpolished portion ofcage from high-speed planet bearing, the roller that did not contact the
48、 cage.The fine, continuous scratches on theFigure 8 shows macrographs of the end of a roller are consistent with 2-body abrasioncage pocket and the end of a roller, caused by embedment of fine, abrasiveFigure 8a shows the unworn depression in particles in the surfaces of the cage ribthe center of th
49、e cage pocket which was and cage pocket.surrounded by a discolored area whichexhibited abrasion marks. The abrasionmarks were not continuous circles aroundthe central depression. Figure 8b showsthe polished end of the roller with itscentral, unpolished circle surrounded bycontinuous circles of fine abrasionfurrows.59alOa9b _600gmFigure 9. Macrographs of roller andsegment of cage from high-speed planetbearing. 9a-cage rib. 9b-roller O.DINSPECTION OF GEARBOX COMPONENTS BY lobSCANNING ELECTRON MICROSCOPYApparatusThe gearbox components were e
