1、90 FTM 10v The Mechanism of Failure With and WithoutTitatanium Nitride Coating in Roller Testsby: Joge Vigintin, University LjubljanaAmerican Gear Manufacturers AssociationI III IIII ITECHNICAL PAPERThe Mechanism of Failure With and Without Titatanium Nitride Coating in Roller ATestsJoKe Viintin,Uni
2、versity LjubljanaThe 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.iABSTRACT:To clarify the effect of Titanium Nitride (TIN) coatings on the failure resistance and frictional
3、characteristic and compare this effect with that produced by the heat treated coatings, two roller tests havebeen made and the stress resulting from the combination of the Hertzian stress field and frictional forcefield on and below zhe contacting surface as well as the flash temperature rise were c
4、alculated. The failureresistance of the TiN-coated roller pair was greater than that of the heat treated roller pair. Themechanism of failure resistance can be explained by the shearing stress (Hertzian stress + frictional force)acting on the contact surface. This stress modifies the structure in th
5、e vicinity below the TiN layer whichis then sheared in the wear track direction.Copyright 1990American Gear Manufacturers Association1500 King Street, Suite 201Alexandria, Virginia, 22314October, 1990ISBN: 1-55589-562-XTHE MECHANI._g_ OF FAILUREWITH AND WITHOUTTITANIUM NITRIDE COATING IN ROT.I.ER TE
6、STSJo_.e Vi_.intin, ProfessorDepartment of Tribology, Faculty for Mechanical Engineering,University Ljubljana, 61000 Ljubljana, Yugoslaviamade of X155 Cr VMo 12-1 (_ 4850) and $6-5-21 INTRODUCTION (_ 7680 tool steels. The comb,nat,on of roller pa,rs, heattreatment and coating treatment are shown in
7、Table 1To increase wear resistance of cutt,ng tools hard The TiN coating of a thmkness of 2 to 4 :m wascoatings such as tltamum carbide, titanium nltrlde and produced by the physical vapour deposition (PVD) process.alum,mum oxrde are apphed to the surface of tools These The thermal properties and ha
8、rdness of the matenat ofcoatmgs are also wldety applied to wear resistant parts the rollers and the coatings are shown in Table 2and mechanical components There are a great number of 3mechanism of scoring was investigated using a two-roRer 4i !machine Scoring resTstance of TIC- and TiN coated gearsw
9、as investigated by Terauchl et I. /1/ who found outthat the scuffing resistance of the coated roller pmrs washegher than that of the uncoated roller pair. Themechamsm of smzure in a two-roller test was investigated 3,01 -_L.-by Nadano et . /2/ He has found that a part of the ,4. B.Figure 1. The prof
10、ile and dimensions of the rollerswear track on the upper roller was fractured due toMateriel Roller Upper roller Lower rollershearing stress at the posmtlon where the hardness of the DIN,JUS pairFoiler was minimum, and adhered to the contacting surface AiSI Heettreatment coating Heattreatment coatin
11、g980“C,oil,12rnln 980“C,oI.12rainof the lower roller C.4.850 Type A ._O0“C,air,2h / 500“C,air,2h /X155CrVM( 500C,air,2n %Ou“C,air,2h- 1 Type B / “IIN / “_INIn this paper roller paws were examined m the 12D21235 “C,ail,2min 1_1.5 “C.ail,2minrecipient stage of failure Observatmns of microhardness of (
12、_.7680Type C 600“C.a|r,lh / 600C.air,lh /the structure at the fad ed portion of the heat treated-, $6-5-2 600“C_Tr,lh 6130“C_lr,lhM2 Type D / “IIN / “IINand TiN- layer ,n two-roller tests, and on the basis ofTable 1 Combination of roller pmrs, heat treatment andcalculation of the stress on and below
13、 the surface of theroller, the mechanism of the failure of the TiN-layer in coating treatmentcomparison w_th the fracture of the heat treatedThe surface roughness of the rollers in the axial2. TEST SPECIMENS AND EXPERIMENTAL directmn Is shown in Table 3 The surface roughness cfMETHOD the coated roll
14、ers was slmdar to that before the coat,ngwas applied The peripheral velomtles the spemflc slidm CThe drawings of the cylindrical rollers used in the and other test conditions of the two-roller test are showntwo-roller test are shown In Figure 1. The rollers were ,n Table 42Material while the tempera
15、ture rise of the roller pairs Type BX155CrVMo12-1(_.4850) S6-5-2 (e.Teeo) and Type D lubricated was approximately by 70-95 KProperties Treated(Untreeted)+_NTreated (Untreeted)+TiNTypeA Typee TypeC Type0 higher than that Tn Type A and Type C The mm_mum oil_1ckers microhardness(HV) 683 (230) 2000 657
16、(250) 2000 film thickness calculated by Dowsons equation /4/ in aThermal condictivityK(W/mk) 28 19 24. 19 line contact on the roller pairs was approximately 0.7 XThermal diffueivityk(m/e) 662x10_ 5Lx10“a 662xlOe 5.3x10“a 10-4 mm As he value defined by Wellauer /5/, TheSpecific heaL,c(kd/kqK) 0.4-60
17、0.700 0.460 0.700 raho of the him thickness to combined surface roughnessDe.si_,(k /m_) 7750 5440 7750 5440 as defined by Wellauer /5/ was from 0,095 to 0,248T he tests were run at very low lambda value perhapsTable 2 Hardness and thermal properties of rafter andalmost metalhc contact.coatingSu_f_=e
18、 Ro_ p,_ 3.1.1. Lubricated frictioncoucJhess ,um Type A Type B Type C Type DBaler Ra O.S 1.s2 1.o7 Z.73 The relationship between the coeffiment of fr_ctloncoating Rmax 4.97 7.81 _.2_ 7.?z and the wear length is shown _n Flgu r e 3 Themeasurement of the coeffiment of frlcUon was carried outAftec Ra 0
19、.24 O.t4 0.46 0.28Unlub_*cated twice for each roller pair Since the two values ,wereTe_t Rmax 1.64 2.55 3.38 4.53similar, their mean value is shownAfte c Ra 0.20 0.14 0.36 0 345obclcatedTe_t Rmax 2.02 3.55 2.92 6.3 1000 I 1000_L “_ Bulk temperature : % =300-350 KTable 3. Surface roughness before and
20、 after the roller 900 -900test800- - -800Conditlons Upper Lower _ 700 - v“roller“ roller -700 _“Peripheral veloal%y, W (m/s) %42 0r38S_ecifRadlus,%cRSl(mm)idin_r 020095 -020105 _ 600- - 600WldLh, b (ram) 3 i0Normal Load. F _N) 500 500Initial oil %empera%ure IK) 293“-2 29322 _ 500- -500Table 4 Specif
21、leatmn of the test conditions _z.O0- B -400 _The test condltlons were kept constant during the _300- -300-test The test was carried out using a two - roller _. “_machine The rollers were lubricated with describe 200- -200LT.additives EPOL SP 150 They prevent scuffing a strmght O0-mineral off with ad
22、ditives, with kinematic viscosities of -100214 10-a m:/s at 313 IK and 17 2 10-e m2/s at 373 KThe roller tests were carried out with unforced Type A TypeB TypeC Typedlubrication as shown In Figure 1 The oil temperaturewas controlled to +/- 4 K by a thermostat Tf Hertzian pressure max Prnax Tf - lubr
23、icated Tf - start stage - unlubrleated3. RESULTS AND DISCUSION Tf - Inclplent stage of failure unlubrlcated3.1. Failure resistanceFigure 2 Calculated s urface temperature rise andmaximum Hertzian stress at the inmpentFigure 2 shows the calculated values of the flashtemperature rise and the maximum H
24、ertzlan stress Pmax stage of failure and at the imtml stagea_ the start and at the recipient stage of failure, e=5unlubr_cated and recipient stage when _ubrlcated of the _ _roller pairs tested The flash temperature rise was _ m-. QO8calculated by the equation given in reference /3/ and _, _oo _ “_,t
25、he vaTue of the coefficient of friction that was measured _ eo_. _.ujust before failureoccurred and at the beginning of the o_.(103“roller pair tests The Hertzian stress was constant _o_.during the tests The flash temperature rise of the roller _“pairs Type B and Type D unlubncated was approximately
26、 o o_ o_ ob o_ o:s Q_ o;_ o_ q_ _ ,:_ _.“L WEAR LENGTH /lOrn/370 to 430 K higher than that Type A and Type C, Figure 3 Relatmnshtp between coefficient of friction andwear length - lubricated conditionsFor all roller pairs, the coefficient of friction just 4 OBSERVATION OF FAILED AREASafter the begin
27、ning of the test had a maximum valueAfter that, Jt decreased to a constant value with 4.1. Lubricated testsincreasing wear length The coefficient of friction of theFigures (5 A B show a micrograph of the wearroller pairs Type A and C was 0,0(55. that of the rollertrack of the rollers Type A and C af
28、ter a 1.2 10+6 mpair Type D was 0078 and that of the roller pair TypeB was highest 0.09pe A _0,5 I6“ I-Typ_0 i9“/, /o._ _ 5,* /uu L. !/ /803 _ /, SS/ /3- /_/ / /L_J/_ /al 1 / /TypeA Typee TypeC Type0ROLLER PAIRS0 ,_I150II1._0 14_2 Frgure 5 Time to fadure of unlubrlcated roller pairsWEAR LENGTH L mFi
29、gure 4. Relabonshtp between coeffiment of fr=ctmn and wear length The surface layer of the roller Type C wasunlubmcated condibons for four roller pairs shghtly more sheared in the direction of fr_ctional forceAlong the contact surface and m the wcm_ty of theThe coefficient of friction vaned with the
30、 surface there are distinct modifications of mmrostructuredifference rn the base material and coahng on the roller induced by shear stresses and temperature This can beFor all the roller pmrs tt appeared that the surface was confirmed by the fact that the hardness rs conmderabtyuniformly torn out du
31、e to frictional forces and wear: higher than that in the subsurfaces Figure (5 B showshowever, the contacting surfaces became smoother as the change in Vickers mlcrohardness wTth depthsshown m Table 3 measured along the hne A - A. probably caused bywork-hardening due to the action of compresswe stre
32、ssfr=ctmnal force and cooling wtth lubricating ml Figure 7A3.1.2. Unlubrlcated teats shows a mlcrograph of the wear track of the rollerType B after failure at about 12 10- 6 m AnAs the failure resistance of the coated and undulation was observed in the cfrcumferenhal direcbonuncoated rollers could n
33、ot be estimated from the test of the roller and the TIN surface layer wasresults obtained under lubrlcatton, the failure test was fahgue-damaged m the direchon of the frtctlonal forcecarried out unlubrlcated The testing was stopped when, Figure 7B shows a mfcrograph of the section plane cutfmlure wa
34、s observed. The initial fr_cbon coefflments were in axial dlrectmn of the roller and the distnbutmn ofabout 012 to 0,1(5 respectively The relationship between Vickers mlcrohardness measured along the hne A-Athe coeffrment of fricbon and the wear length fs shown From this mlcrograph _t can be seen th
35、at the metal _nin Figure 4. the wcm_ty below the TiN layer was distinctly deformedThe hardness in the wcmtty of the boundary of the T_NFigure 5 shows the running time until failure layer was the same as that before the tests The T_Noccurred. The test was carried out twice for each roller layer had a
36、 larger elasbc_ty modulus than the basepair, and the mean value out of two measured values _s matertal, and that the compressive, shear and thermalshown The life of the coatings of the roller pairs Type stresses cause a much larger deformation _n the baseB and D was longer than that of roller pairs
37、Type A mater_al =n th=, vlcmty of the TIN layer than that m theand C. T_N layer On the other hand the surface temperaturewas lower than that used for heat treatmentu_ _ Vickers m,crohordness Lood:0,3/10N600 800 1000 1200_ o,ol_ oi-t=_u_ 0,02 _._d_rectlon of the roller Type C, and the surface layer o
38、f _i . ,- i_!_ o,othe roller was sheared in the direction of the frlcbonal 500 - TypeA-un_ubr_atedforce. Along the tested surface there is an approx 1 _.m :!_y_/_00()_:,_,._. x_thick layer of austenite.Th_s layer was produced in thesame way as that on roller pairs type A.(B) SECTIONAL PLANE CHT INAX
39、iaL _P_CTIO_o_ _OLLE_Figure 9, A,B shows a micrograph of the weartrack and the sectional plane cut in the axml dmectton of FigJre 9 Micrograph of failure portion of the rollerthe roller Type 8 after failure.Comparing the roller Type Type B w_th T_N coabngD w_th the roller Type B, the degree of surfa
40、ce fabguedamage by failure was htgher than that found on the In Ftgures 10 and 11 the ordinate and absmssarollers pairs type D. Figure 9 B shows the distribubon of indicate the posit_on of the evaluated stress m terms ofVickers m_crohardness which _s the same as that before the parameters y and zthe
41、 tests. The mechanism of the failure of T_N-layer wasslightly different appearance than that of the lubricated The patterns show that maximum value of theThe elastic deformation of the base materm was largerstresses “Ok. 5 45c and 5zy occurs below the ccntacbngthan elastic deformation of the TiN-lay
42、er and larger thansurface, and the profile of distrtbutlon of the stresses _sthat of the heat-treated base material of the roller _nthe lubricated conditions approximately symmetrical to the orig_n for the rollerparos whose value of the coefficient of fHcbon wasunder 0,1 On the contrary the profile
43、of distribution of5. MECHANISM OF FAILURE the stresses is not symmetrical to the origin for theroller whose value of the coefficient of friction wasThe frictional force acting on the contacting surface larger than 0,1. The maximum value of the stressesoccurs on or below the surface, as shown Fig. 11
44、.of the roller and the stress occuring on and below thesurface were calculated by the author and shown m F_gures 12, 13 and 14 show the distribuhon of thereference /6/. maxtmum value of the equwalent stresses o k, thepositive and negative components of the reversed shearAs an example, the distributi
45、on of the equivalent stress q the positive and negative components of thestress Ok, shear stresss 5 4-5 “ and reversed resersed shear stres Ozy and shear stress c45c forumdmectional stress czy resulting from the combmabon of different values of fr_cbon coeffic,ent.the Hertz_an stress field and the f
46、rictional force field of, F_gure 11 shows the distrtbutton of the equivalentthe upper rollers are shown in Figure 10 and Ftgure 11 stresses The maxtmum value occurs at a depth “z“respectively. The value of the coefficient of friction is beneath the contact surface up to a coefhctent ofgiven from the
47、 measured value just before failure. friction of 0,2 When the value of the coefficient of6friction is larger than 0.2 the maximum value of theequzvalent stress occurs on the contact surface. F_gure12.a) Dl.s_,;r_.av, lr,a_ o_ stzsxa _,al Distribution _ s_ress OL ,.(_ Dis_zlbu_Aon ot I_Zmss _,_(I D_z
48、_rlba_Aon off st.resll _=ye_ :.s=_bu-.o_,e =re,=,_ F_gure 11 Olstrrbutlon of resultant stress act,ng on theupper roller at the incipient stage of fanlureFigure 10 Distrnbutlon of resultant stress acting on the coeffument of frrchcn 0 5upper roller at the mmptent stage of failure:coefficient of frict
49、ion 0.065 G,_iN/ram _0 -100 -200 -300 -400 -500 -600 -700 -800The reversed shear stresses actmg parallel to the 0 _ ,._1, i/“plane of the contact surface and its max,mum value _ ,.,os _-i _j-/Toccur at a depth “z“ beneath the contact surface and _ =_ _coeff,c,ent of fr,ct,on (F,gure 14). _ 0,067 “ _I/,are dependent on theThe maximum shear stress value occurs at a constant _ ! _;(Tf_“depth “z“ below the contact surface z2 at an angle of E i i45: to the plane at the surface a
