1、05FTM17Influences of Bearing LifeConsiderations on Gear Drive Designby: F.C. Uherek, Rexnord Geared ProductsTECHNICAL PAPERAmerican Gear Manufacturers AssociationInfluences of Bearing Life Considerations on GearDrive DesignFrank C. Uherek, Rexnord Geared ProductsThe statements and opinions contained
2、 herein are those of the author and should not be construed as anofficial action or opinion of the American Gear Manufacturers Association.AbstractHistorically, catalog gear drives have been designed with 5000 hours of L10 bearing life at service factor 1.0power. Advances in bearing analysis methods
3、 have brought new considerations to the design and selectionprocess. The impact of new modeling techniques, additional considerations, and various extensions to thetraditional bearing fatigue calculations are explored. The modeling of these various additions to a traditionalcatalog L10 calculation i
4、s illustrated by bearing selections for cases of single, double, and triple reduction geardrives. A roadmap is presented listing critical considerations when applying various bearing manufacturerrecommendations.Copyright 2005American Gear Manufacturers Association500 Montgomery Street, Suite 350Alex
5、andria, Virginia, 22314October, 2005ISBN: 1-55589-865-31Influences of Bearing Life Considerations on Gear Drive DesignFrank C Uherek, Rexnord Geared ProductsIntroductionHistorically catalog gear drives have been designedwith 5000 hours of L10 bearing life at service factor1.0 power. Advances in bear
6、ing analysis methodshave brought new considerations into the designand selection process. Reviewing this history usingthree sample gear drives will illustrate past and cur-rent thinking on this subject and the implications ofthe various assumptions each of the methodsmake.Target Design PointBy its n
7、ature, a catalog design gear drive is not de-veloped with a known set of load conditions. Thedesigner does not have the luxury of the certainknowledge that it will be operated only at moderatespeeds occasionally on Sundays during monthsthat do not have a letter R in them. Catalog speedranges run fro
8、m 1800 to 100 rpm. Application or ser-vice factors per AGMA standards range from 1.0 to2.0 for various applications. Customers may specifyadditional requirements such as increased bearinglife or limited cooling options that also influence boxsize. The unit design must be capable of operatingat any o
9、f these conditions.For the purposes of this paper, we will make the as-sumption that the gearing is the limiting member inthe drive. Bearings, shafting, and other compo-nents, keys, fasteners, etc. will be selected andsized to exceed the rating of the gears. Althoughoptimal, this gearing only limita
10、tion is sometimes dif-ficult to achieve in practice when one is balancingthe competing options of having a wide variety of ra-tios and maximum flexibility combined with mini-mum variation in part dimensions and inventory ofparts. For this study, we will assume that we canpick the optimum bearing siz
11、e regardless of sizevariation across ratios and multiple reductiondrives.The gear drive will be operated for 60 Hz service at1750 rpm input speed. Ambient temperature rangeis 17 Cto+38C. Altitude is assumed to be lessthan 700 meters. The effect of sump temperature isreviewed by looking at both a 28
12、C rise and a 55 Crise in sump temperature above the ambient of 38C. The control in temperature rise can take theform of forced air cooling, water cooling coils, ordedicated pump and oil-to-water heat exchangers.Gear Drive DesignsIn 1997 AGMA was involved in a worldwide study tocompare and contrast t
13、he various methods used todetermine the load distribution factor KH.Tofur-ther efforts in this area, three parallel shaft geardrives were developed to test various assumptionsin the calculation models. These gear drives wereutilized as the design in the analysis.The single reduction gear unit has a
14、space envelopeof 883 mm wide by 635 mm tall by 1021 mm deepshaft-to-shaft with the following key dimensionslistedintable1.Based on the above geometry, gear ratings werecalculated to AGMA 2001-D04 using the AGMAGear Rating Suite Version 2.2. Setting requiredgear life to 10000 hours as recommended in
15、AGMA6010-F97, and setting service factor equal to unity,we calculate a unit rating of 990 kW at 1750 rpm in-put speed per AGMA 2001-D04.The double reduction unit has a space envelope of1048 mm wide by 543 mm tall by 870 mm deepshaft-to-shaft with the following key dimensionslistedintable2.Based on t
16、he above geometry, gear ratings werecalculated to AGMA 2001-D04. Setting requiredgear life to 10000 hours as recommended in AGMA6010, and setting service factor equal to unity, wecalculate a unit rating of 308 kW at 1750 rpm inputspeed per AGMA 2001-D04.The triple reduction unit has a space envelope
17、 of1676 mm wide by 889 mm tall by 1155 mm deepshaft-to-shaft with the following key dimensionslistedintable3.2Table 1. Single Reduction DriveItem Pinion Gear Both/Units Pinion Gear UnitsOverall Ratio 4.214Tooth Combination 14/59Normal Module 8.00 mmNormal Diametral Pitch 3.175 in- 1Normal Pressure A
18、ngle 20 degHelix Angle 11 degCentre Distance 304.8 mmCenter Distance 12.00 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 137.97 503.63 mm 5.432 19.828 inOperating Pitch Diameter 114.10 480.82 mm 4.492 18.930 inRoot Diameter 100.20 465.81 mm 3.945
19、18.339 inActive Face width 127 127 mm 5.00 5.00 inTable 2. Double Reduction DriveItem Pinion Gear Both / Units Pinion Gear UnitsOverall Ratio 18.264High Speed SetTooth Combination 15 / 63Normal Module 4.5 mmNormal Diametral Pitch 5.644 in- 1Normal Pressure Angle 20 degHelix Angle 11 degCentre Distan
20、ce 182.88 mmCenter Distance 7.2 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 82.19 301.55 mm 3.235 11.872 inOperating Pitch Diameter 70.33 295.42 mm 2.769 11.630 inRoot Diameter 61.30 280.43 mm 2.413 11.041 inActive Face width 76.2 76.2 mm 3.00 3
21、.00 inLow Speed SetTooth Combination 17 / 69Normal Module 6.00 mmNormal Diametral Pitch 4.233 in- 1Normal Pressure Angle 20 degHelix Angle 9 degCentre Distance 266.7 mmCenter Distance 10.50 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 121.03 436.
22、35 mm 4.765 17.179 inOperating Pitch Diameter 105.44 427.96 mm 4.151 16.849 inRoot Diameter 93.49 408.69 mm 3.681 16.090 inActive Face width 120.65 120.65 mm 4.75 4.75 in3Table 3 Triple Reduction UnitItem Pinion Gear Both / Units Pinion Gear UnitsOverall Ratio 30.8High Speed SetTooth Combination 20
23、/ 44Normal Module 7.00 mmNormal Diametral Pitch 3.6286 in- 1Normal Pressure Angle 20 degHelix Angle 11 degCentre Distance 236.22 mmCenter Distance 9.3 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 163.77 336.702 mm 6.448 13.256 inOperating Pitch D
24、iameter 147.64 324.80 mm 5.812 12.788 inRoot Diameter 131.70 304.62 mm 5.185 11.993 inActive Face width 101.6 101.6 mm 4.00 4.00 inIntermediate Speed SetTooth Combination 15 / 50Normal Module 9.00 mmNormal Diametral Pitch 2.822 in- 1Normal Pressure Angle 20 degHelix Angle 12 degCentre Distance 304.8
25、 mmCenter Distance 12.00 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 162.99 482.57 mm 6.417 18.999 inOperating Pitch Diameter 140.68 468.92 mm 5.538 18.462 inRoot Diameter 120.03 439.65 mm 4.726 17.309 inActive Face width 127 127 mm 5.00 5.00 in
26、Low Speed SetTooth Combination 15 / 63Normal Module 11.00 mmNormal Diametral Pitch 2.309 in- 1Normal Pressure Angle 20 degHelix Angle 12 degCentre Distance 457.2 mmCenter Distance 18.00 inAGMA Quality Number A6 - Q11Heat treatment CarburizedManufacturing Method GroundOutside Diameter 203.43 754.96 m
27、m 8.009 29.723 inOperating Pitch Diameter 175.84 738.55 mm 6.923 29.076 inRoot Diameter 152.46 703.97 mm 6.002 27.715 inActive Face width 177.8 177.8 mm 7.00 7.00 in4Based on the above geometry, gear ratings werecalculated to AGMA 2001-D04. Setting requiredgear life to 10000 hours as recommended in
28、AGMA6010, and setting service factor equal to unity, wecalculate an unit rating of 712 kW at 1750 rpm inputspeed per AGMA 2001-D04.Bearing Selection ParametersBased on the size range of the above units, bearingselections were limited to single row taper rollerbearings. The initial criteria are to ha
29、ve the bearingbore below the root diameter of the pinion and largerthan any required shaft extension based on 72 N/mm2allowable stress. Based on the class one pow-ers of the gear drives we set the following sizelimitations:Single Reduction UnitShaftBearing Bore Range(mm)Input Shaft 9095mmOutput Shaf
30、t 120 130Double Reduction UnitShaftBearing Bore Range(mm)Input Shaft 55 - 60Low Speed PinionShaft90Output Shaft 130Triple Reduction UnitShaftBearing Bore Range(mm)Input Shaft 70Intermediate PinionShaft120Low Speed PinionShaft170Output Shaft 220Selected bearings met the 5000 hours life require-ment e
31、xcept for the double and triple reduction lowspeed pinion shafts that had 4100 and 4700 hoursat gear mesh. The root diameter requirement wasmet for all shafts other than the low speed pinion forthe triple reduction unit.Catalog Selection MethodsThe process of selecting bearings is well docu-mented i
32、n manufacturers catalogs and computerprograms. Based on the required bore size, all stan-dard ISO taper bearings available for that rangewere audited. In these sizes, taper roller bearingsprovide a good balance of performance versus costand availability. A constant bearing span for eachshaft was ass
33、umed. The same manufacturers costbasis was assumed for all selections.Three MethodsIn addition to selecting bearings to an unadjustedcatalog rating, three methods were used to selectbearings. These methods are described as follows.Timken fTfVmethodIn 1986 the Timken Company published a life modi-fic
34、ation method for bearing selection. This methodaddressed the effects of speed, operating tempera-ture, and oil viscosity at fixed value. A catalog bear-ing life was first calculated and then two adjustmentfactors were applied. The first factor covered speedand temperature, and the second addressed t
35、he im-pact of viscosity at a fixed temperature. To calculatethese factors, the following formulas were used.For shaft speedsS 15000dfT= 600 (1.89 + 32) 1.8 S0.38wherefTis the temperature factor is the bearing operating temperature, CS is the shaft speed, rpmd is the bearing bore diameter, mmWhenS 4.
36、0a23= 2.5wherea23is the lubrication life adjustment factor.The resultant L10 life is calculated byL10a= a23 L10This method has a maximum increase of 2.5 timescalculated L10 life and a potential decrease to 0.10times calculated life. The inclusion of EP additivesassists this method when modeling low
37、speed bear-ing performance.DIN ISO 281 Amendment 4 aDINmethodIt has been noted that it is rare for bearings to fail in aclassical fatigue mode predicted by the formulas inbearing catalogs. Bearings can fail due to a wide va-riety of conditions such as overload, improper as-sembly, lubricant difficul
38、ties, as well as fatigue. Tobetter correlate the classical formulas with field ex-perience, a new adjustment factor was needed. In2003 work begun on an update to ISO 281 RollingBearings Dynamic load ratings and rating life.Amendment 4 presented a new life adjustment fac-tor aDIN.This factor consider
39、s not only oil viscosityat the operating temperature of the bearing but alsothe contamination level in the lubricant. Contamina-tion in the lubricant will have a major effect on bear-ing life. Particles can cause local stress risers asthey flow through the bearing damaging the rollerand raceway. Hig
40、h film thickness or soft particlescan reduce the effect of these problems. Since geardrives rarely operate in laboratory clean rooms,filtration is required to remove the particles gener-ated either internally from the bearings and gearmeshes or externally from the processessurrounding the drive.The
41、critical parameter in the aDINanalysis is an esti-mate of lubricant contamination. This is very difficultto estimate yet its importance cannot be under-stated. An incorrect assessment of this value willlead to dramatic misestimates in L10 adjusted life.cis the factor that considers the contamination
42、 lev-el of the lubricant in the bearing life calculation. Theinfluence of contamination on bearing fatigue de-pends on a number of parameters including bearingsize, relative lubricant film thickness, size and dis-tribution of solid contaminant particles, and types ofcontamination, (soft, hard, etc.)
43、A method of defining cis based on the level of con-tamination as defined in ISO 4406. This methodcounts the number of solid particles of given sizes ina fixed volume of oil. The particles are sorted intogroups by size. Very clean gear drive oil would begiven a rating of -/15/12 meaning that between
44、160to 320 particles of a size greater than or equal to 6mm and 20 to 40 particles of a size greater than orequal to 15 mm are found per milliliter of oil. Newerparticle counters return a three-digit code reportingthe number of particles greater than or equal to 4mm, 6 mm, and 14 mm in the same sampl
45、e size.When comparing between methods of measure-ment, it is customary to drop the first group of num-bers reported in a three-digit measurement.To achieve these levels of oil cleanliness, a betarated filter is required. A filter rating is an indicationof filter efficiency. The efficiency of filters
46、 is definedas the filter rating or reduction factor , which is re-7lated to a given particle size. The higher the value,the more efficient the filter is for the specified par-ticle size. Therefore both the value and the speci-fied particle size have to be considered. The filterrating is expressed as
47、 the relationship betweenthe number of specified particles before and after fil-tering.Thiscanbecalculatedbyx=n1n2wherexfilter rating related to a specified particle sizex;x particle size, mmn1number of particles per volume unit (100 ml)larger than x mm upstream of the filtern2number of particles pe
48、r volume unit (100 ml)larger than x mm downstream of the filterNote: The filter rating only relates to one particlesize in mm, which is shown as the index e.g. 3,6,12,etc. For example, a complete rating of 6=75means that only 1 of 75 particles of 6 mm or largerwill pass through the filter. Therefore
49、 both the val-ue and the specified particle size have to be consid-ered.An initial estimate for cis 0.1. This is typical forsplash lubricated drives in contaminated environ-ments or for drives using pressure lubrication withfilters coarser than filtration ratio 25=75.If actual contamination levels measured to ISO4406 are available and the specified filtration size isused, i.e. 12=200, an estimate for ccan be foundfrom the following curve fits listed in table 4 derivedfrom the graphs in DIN ISO 281 amendment 4
