1、92 FTM6Comparison of Carburized GearMaterials in Pittingby: Louis Faure, C.M.D., FranceJ.L.Vasseur and C. Lefleche, CETIM, FranceAmerican Gear Manufacturers AssociationTECHNICAL PAPERComparison of Carburized Gear Materials in PittingLouis Faure, Technical Director, C.M.D., FranceJ.L. Vasseur and C.
2、Lefleche, CETIM Senlis, FranceThestatementsandopinionscontainedhereinarethoseof theauthorandshouldnotbe construedasanofficial actionoropinion of the AmericanGear ManufacturersAssociation.Copyright 1992American Gear Manufacturers Association1500 King Street, Suite 201Alexandria, Virginia, 22314Octobe
3、r, 1992ISBN: 1-55589=586-7iCOMPARISON OF CARBURIZED GEAR MATERIALS IN PITTINGLouis FAURE - Technical Director - C.M.D. - FRANCE3.L. VASSEUR - C. LEFLECHE - CETIM Senlis - FRANCEABSTRACT:This paper compares the pitting resistance of five different steels commonly used for casecarburized gears. The co
4、mparison is based on the test results obtained on the CETIM gear benches.A presentation of the test bench capabilities and of the first results obtained was given in 1988 inNew-Orleans at the Fall Technical Meeting. Since then new tests have been performed on severalcarburized steels and numerous re
5、sults are now available.The paper first describes how the tests were set up and how the test results were interpreted foreach steel, representative curves of pitting performances were drawn. The comparison is based onthe appearance of these curves and all the deviations encountered are discussed in
6、details.Introduction The torque measurement is made by strain gaugesFor more than ten years CETIM has been conduc- fitted on the “U“ joint tube. The power supply of theting experimental studies to determine the pitting gauge bridge and the torque signal are providedresistance of case carburized gear
7、s. First there is a through a rotating transmitter fitted on the free shaftshort discussion about the test benches and gears used end of the loop goarbox. A schematic view of thein these studies. Then details are given on the method bench is given on figure I and the main characteristicsused to iden
8、tify the typical surface aspects which are in the table 1.taken as references for pitting resistance evaluation.Next, the results of the test on the teeth of gearsfrom five different materials are compared. The mate-rials are :- 20 MnCr5 (20MC5)- XC 18- 16 NC 6 Torque application- 17 CrNiMo6 Gear to
9、 be rmotor- 8NCD6 tested “ _ FLoop gearboxThe Iirst one is a chromium manganese steel common- /ly used in Germany for case hardening. The fourth one _ - DO motoris a chromium nickel and Molybdenumsteel which is _ -_ _also very common in Germany for gear carburizing. -EThe secnd ne is a carbn steel n
10、t aliyed and it !st_re d -_r_r_im_RrOtnastmi_tgte _can be carburized. The two others are steels for case -Ehardening often used in France. The chemical compo-sition of these steels is given in appendix I.ucer1. Test benchesTo perform tests for pitting endurance, CETIMbuilt g test benches. These test
11、 benches have closedpower circulation and the load application is eilber t_v _ “Ujoint shaftmechanical or hydraulic means. A static torque isintroduced in either direction on one of the two shaftsplacing the gears under load. The IDEFIX #00 testbench described below is made-up of two industrialgearb
12、oxes. One is located near the drive motor, is Fig. 1 : IDEFIX t_00 Test benchlargely dimensioned and is used to close the loop. Theother one is at the other end and houses the gear tobe tested.The two gearboxes are connected together on one sideby a “U“ joint and on the other side by a torqueintrodu
13、ction device (rotating jack or torsional bar).tThe point obtained on the endurance curve for thecorresponding pressure is an industrial use limitIDEFIX 401 TEST BENCH for the gear when we can tolerate the presence ofCHARACTERISTICS micro pitting on the teeth flanks. We stopped the testwhen we ascert
14、ained there was general spalling on theSPEED(R.P.M.) VARIABLEFROM0 TO 3000 whole width of one or more teeth.By comparing the number of cycles obtained with theSTATICTORQIJE VARIABLEBYROTATIVE number of cycles recorded at the end of the firstCYLINDER(INROTATION) two phases defined previously, we can
15、have a goodFROM0 TO400 m.dan idea of how rapidly the surface degrades for a givenCEtCrERDISTANCE 200 mm level of pressure. Figure 2 below provides a generaldescription of the gear teeth flanks after microTYPEOF BEARING SPHERICALANDCYLINDRICAL pitting appears.ROLLERSTYPELUBRICATION BY INJECTIONONEACH
16、SIDEOFTHEGEAROILTEMPERATIlllE IN SERVICEREGULATIONBETWEEN60 AND900 C Grmd,ng lchet_an I verttcal ridgesighly cold* rolledTable 1 h tal ridges )Pitch dlaeened2) Test gears 14toSmmlI Typical mlcroplttlng Choice of the module : B.,e d,_The aim of the tests is to determine the onset ofsurface degradatio
17、ns. The value of the module hasbeen chosen to avoid tooth breakage (a safety factor W_EEL0,for bending resistance of more than 1.5 is used).Face width :The face width was chosen to withstand a toothload, which can reach 3000to 3500N/ram2 at themaximum torque of the test bench.Case depth :The case de
18、pth has been chosen to avoid case Fig. 2 : condition of wheel on secondary shaft aftercrushing in operation (_, 1 mm for m=10 mm, 1,22.10 cycles(torque mdaN)1.5 mm for m=12 ram).We record 3 distinctive zones :20MC5 XC18 16NC6 I7CrNiMoS the upper part where we see grinding traces and16NC6 m = 12 18NC
19、D6(m=_0) fhat corresponds to the zone where we performedNumbersof ZI = 20 Zl = ZZ Zl = m Zi _ ZZ a tip relief. The contact pattern of the flank isteeth Z2 = 21 = 16 = 16 =20 decreasing when we go towards the tip of thetooth where it is minimum,Addendum xl = x2 xl = x2 xl = x2 xl _ x2modification = 0
20、 = 0.333 = 0.333 = 0coefhclent the part situated near the pitch line, approximate-ly at mid height of the flank which is the zoneFace width 15 16 17 15 cold-rolled under superficial pressure,b minimum(ram)the lower part which has an unpolished appea-Modulem 10 12 12 10 rance contains micro pitting t
21、o a height of appro-mately # to 5 ram.Center distance 205 200 200 200aCase depth 1 2.5 1.5 1 #) Determination of pressure limit values for_mm_ endurance of case hardened steels : _6_L.This value is the pressure under which there isTable 2 : Characteristics of test wheels used practically no risk of
22、tooth degradation. According toISO Standard 6336, this value is 1500 N/mm2for casecarburized steel. We used this value.3) Criteria and method used for the tests follow-up ISO 6336 standard has also limited the maximum pres-During the daily tooth examinations, we noted the sure at 1.6x1500 N/ram2 = 2
23、#00 N/ram2. We also usednumber of cycles at which we saw the first appea- this value. The ISO Standard has defined two differentrance of micro pitting. For each specific load level, curves:the point obtained will be correlated with the corres- - the lowest corresponds to a lower level of pressurepon
24、ding value of the endurance curve to determine which garantees the operation of the gear withoutthe limit pressure for a gear without micro pitting.“ apparent pitting on the active flank. This curveWhile continuing the test, we recorded the number of reaches the value of 0_m_ for 5x10 V“ cycles,cycl
25、es which corresponds to the appearance of thefirst pit.2the other corresponds to a greater admissible pres- d) for each graph the lowest limit will be obtainedsure for the same number of cycles because it by shifting a line, parallel to the others, to the leftaccepts a certain degree of “pitting“. T
26、he maxi- side, until it reaches the last experimental pointmum pressure is reached for 6x10 _ cycles and on that side of the graph. We have to verify thatis obtained for 10 _ cycles. These two curves will this last point is not an abnormal value. If such isbe taken as a basis for the comparison of a
27、ll the the case, the space between the limit curvesresults for the steels tested. Each graph has been concerning the first pit criteria and the microdrawn to show this comparison, pitting criteria, and in the same manner betweenthe general spalling criteria and the first pit, will#.I Analysis of tes
28、t results be reduced. In extrem conditions the limit curvesFig. 3 presents an example of typical results of can cross over, that is not logical. For exampletests for 18NCD6 steel. These results concern the the point No. 22 of Fig. 3 is an aberrant point.general spalling appearance. Twenty one differ
29、enttests have been performed on 6 different load levels #.2 Presentation oI the test resultswhich correspond to the application of 6 different The graphs of Fig. # to 9 give the results obtainedtorque levels on the test bench. To construct a curve for the five steels tested.for this family of result
30、s which can be considered asthe lowest limit which garantees the no-risk of failure, z, CH_kl I IIIIIIIJ/rlllA I IINIIIIx_IIINII “_.111 Jof the results are : 16I.,=,0,T.,0,., I lllrIJll “lXU-flllll I FLIa) with two logarithmic scales along the axis, asdefined on the graphs, the three curves limiting
31、 I ! lllll “,I llllllft l _lllll l IIlI_WIthe apparition of micro pitting, of the first pit I 5, II IIIIIIII I II/IIIII II IIIIIII_and of the general spalling are three straight lines, ,bs _ l_e _ ,_7 _ ,_ _ ,_-b) all the results obtained must take place on a Fig. 5 : XCI8 Test resultsfamily of para
32、llel straight lines which normaly in-clude all the points of the graph concerned,c) the single slope will be obtained at the beginningby taking the slope of the lines which go througha maximum number of experimental points (no lessthan 3 or #). These lines generally have a distribu-tion of points wh
33、ich is partially similar when weconsider the three criteria defined above (micropitting, first pit, spalling),3Z,I O“HI N/m rnnl ZN OH :N/m n_I 16N_ 6 m - IO - TESTS_gJ-TS 18NQ) 6 - TES3SRESLETSt I 11111# I I Pill 11111/ I I IIIIrl i I Ililll I I IF| I r I IIIIll i i ii1111 i = t t 11111,o24OQilI “_
34、“,_,I;T,_U_II I NJIIl%l I II LHIk “I-. / I1160C IIIIIII 1111111147_lJflll Ntl% Ill 16NC6 m=12 8.6 8.7 12.5I IIIIIII I IIIIIIF“., 13!LIIIIII I“l-IIli.IJI,. ,o0 IIIIIII I IIIIIIII I IIIIIIII I IIIIII11_ t7CrNiMo6 9.5 9.0 13.8liR It l_lll $ I07 _i 1(Is t 18 _ 8NCD6 t 3.7 13.2 15.7Fig. 8 = 17CrNiMo6 Tes
35、t resultsTable 3-Slope values for different limit lines + steels#IWhen all the families of curves corresponding to these a higher level : _ = 2300 N/mm2five steels are considered, the following conditionscan be drawn : For each level we have defined the three steps cor-responding to micro pitting, f
36、irst pit and spalling bya) the ISO Standard “limit for pitting allowed“ corres- three horizontal lines, and we have pointed out forponds to a generalization of spaUing on one or each steel the number of cycles necessary for eachseveral teeth of the gear. In fact, when the gear load level to reach th
37、e three reference steps. Fig. 10is in such a condition, it is very near to tooth represents the graph obtained for low load and fig. 11breakage by crack development from the surface, for high load. It can be seen from fig. 10 thatTherefore, it can be considered that this limit cor- lgNCD6 steel seem
38、s the best for low load and fromresponds to the end of the life of the gear. When fig. 11 that 20MC5 steel is the best for high load.the slope of the spalling curves are compared tothe ISO curve, it can be concluded that, for only 5.3 Influence of the moduleone material (18NCD6), we are in complete
39、agree- For the same steel (16NC6) we have two sets ofment. For two others (17CrNiMo6 and 16NC6) the test results :results otained are not too different (12,5 to 13,8 the first one with m = 10 mm,against 15,8). It is necessary to apply some cor= the second one with m = 12 mmrection if a very long dur
40、ation in operation is t_L_ _H:l?OONimm,needed. For the last two materials (20MnCr5 andmet,o II IlLdirectly for a long duration without taking these SPALU,G results into account. However, it should be pointed ._ 7out that there were good performances from _;_/“_1_20MC5 steel for a low number of cycle
41、s. This y /means that 20MC5 can be used for low speedgears with limited life, for example in automobilegears, .,s, _ IIl_?%)_J_ I I I Ib) the line of ISO Standard called “without pitting“corresponds to the appearance of micro pitting onthe teeth. It can be se.en that material (18NCD6) ,0 _!_- ,_is t
42、he only good correlation we have with ISO ; “_“for XCI8 steel, the slope obtained for the micro MICR(PITTINGpitting lineis very different from ISO ($,9 against II I _ _13,2) and it is clear that this material is verysensitive to surface micro damages. For the other T._,_b8 _b_steels, the slope for t
43、he micro pitting line is rela-tively the same but lower than ISO slope (8 to 10against 13,2). For these steels it is necessary to Fig. 10- Performance comparisonbe careful if micro pitting is not allowed, with low loadHowever, we have also to consider that the deve- H_u_ -,-0“ =2300“Nlm_lopment of m
44、icro pitting is in connection with theoil film thickness, which depends into a large IIdegree on the choice of the oil viscosity, SPALLING .lc) if one considers the slopes of the first pit lines, itcan be observed that they are all lower than theslopes of spalling curves. This can be explained bythe
45、 fact that the crack propagation velocity ismore important when the load is high. Therefore,for high loads the distance between first pit line FIRSTPITand spalling line will decrease. For 18NCD6 steelthis distance is relatively stable and it seems toreact well under high loads. 16NC6 and 17CrNiMo6st
46、eels seem to be more prone to pitting underhigh loads, but the propagation of pits will beslower under low load. Industrially, with these MICROPPITTINgwithout great danger because the crack propaga-tion is slow for low load. Howerver, 16NC6 and17CrNiMo6 will not perform well under heavy 1 7load. For
47、 example if a crack or a pit appears Fig. 11- Performance comparisonafter an overload application, it can be observed with high loadafter a quick evaluation of the damage thatspalling can be dangerous to the life of the gear.20MC5 steel is more suitable for overload condi- Referring to fig. 6 and 7
48、to the table 3 and to fig. 10tions and for limited duration (see Fig. $). and 11, there seems to be no important difference ofperformance between these two modules except for5.2 Influence of the load level high load level.To summarize this comparison we have considered In this case, the best results
49、 seem to be obtained bytwo levels of load : the largest module (12). This comparison has not beena low level : _ = 1700 N/ram2 confirmed by other tests at this time.58) ConclusionThe test performed on industrial gears by CETIMduring more than ten years have provided a betterknowledge of the behavior of case carburized steelsfor gears. The data obtained in CETIM test have beenused in the ISO method in order to have a good baseof comparison. It has been established that the beha-viors of the steels tested are equivalent on somepo
