Modern Applied Science; Vol. 6, No.

12; 2012 ISSN 1913-1844 E-ISSN 1913-1852 Published by Canadian Center of Science and Education

Analysis of Contact Stress between Cylindrical Roller and Outer Ring Raceway with Taper Error Using ANSYS
Xintao Xia1, Shichao Zhu1, Chenhui Jia1 & Rongjun Niu1
1

College of Mechatronical Engineering, Henan University of Science and Technology, Luoyang, China

Correspondence: Xintao Xia, Henan University of Science and Technology, 48 Xiyuan Road, Luoyang 471003, China. Tel: 86-379-6421-9460. E-mail: xiaxt1957@163.com, xiaxt@mail.haust.edu.cn Received: November 5, 2012 doi:10.5539/mas.v6n12p86 Abstract Taking a cylindrical roller bearing as a research object, the contact stress between the cylindrical roller and the outer ring raceway with a taper error is studied by means of ANSYS in order to obtain the allowable value of the upper bound of the taper error. The results show that the given load corresponds to the suitable upper bound of the taper error of the outer ring raceway and with the increasing taper error, the contact stresses between the cylindrical roller and the outer ring raceway increase observably and the distribution of the contact stresses presents more complex asymmetry and nonuniformity. Keywords: contact stress, cylindrical roller bearing, outer ring raceway, taper error, ANSYS 1. Introduction The rings are the important parts of a rolling bearing and their quality has an important impact on the bearing capacity, operation life, and performance reliability due to the additional contact stresses generated by the machining errors (Chandrasekara et al., 1983; Chiu et al., 1987; Sun et al., 2008; 2010; Wang et al., 2011; Xia et al., 2012, 2012a; Xia, 2012). The taper error of the outer ring raceway as one of the most common machining errors can lead to an abnormal contact stress distribution between the roller and the outer ring raceway. This brings about a bad rolling stability of the roller. But, it is little reported how the machining errors influence the bearing contact stresses (Andrey et al., 2012; Demirhan et al., 2008; Hanson et al., 2010; Slack et al., 2010). Therefore, taking the cylindrical roller bearing coded by N1015 as a research object, this work investigates the relationship between the contact stresses and the taper error of the outer ring raceway with the help of the finite element analysis software ANSYS to establish the finite element model of the cylindrical roller bearing, to analysis the static contact between the cylindrical roller and the outer ring raceway with the taper error, and to determine the allowable value of the upper bound of the taper error. 2. Finite Element Model of Cylindrical Rolling Bearing 2.1 Technical Parameter of Bearing The technical parameters of the cylindrical roller bearing N1015 are shown in Table 1. Table 1. Technical parameter Parameter Bore diameter of bearing, d/mm Outside diameter of bearing, D/mm Thickness of ring, T/mm Width of ring, W/mm Diameter of roller, /mm Length of roller, L/mm Number of roller, N Poisson ratio of bearing element material, v Elastic modulus of bearing element material, E/GPa Parameter value 75 115 5 20 10 11 22 0.3 208 Accepted: November 23, 2012 Online Published: November 28, 2012

URL: http://dx.doi.org/10.5539/mas.v6n12p86

86

www.ccsen net.org/mas

Modern n Applied Scienc ce

Vol. 6, No. 12; 2012

2.2 Meshin ng In order to o reduce compu uting time, on n one hand, the e model of one e quarter roller r at the bottom m of the bearing g that bears the m maximum load d is created; o on the other ha and, the chamf fers of the inn ner and outer r rings and the radial r clearance are ignored. The T Solid 45 is s adopted to m mesh the mode el. The meshes s of the contac ct part of the roller r and the in nner and outer rings are refin ned to enhanc ce the computi ing accuracy. The length of f the grid along the axial direc ction is 0.07 mm. m Because t the side length h of the finite element mode el mesh is, par rticularly, less than half of the e size of the minor m semi-ax xis, the calcula ation results a are precise eno ough. The half f-width size of o the contact are ea is 0.15 mm and the size of f the finite elem ment model is suitable, as sh hown in Figure e 1.

Figure 1. Finite element m model 2.3 Creatio on of Contact Pair P The creation of the cont tact pair and t the setting of t the contact par rameters are c crucial issues o on contact ana alysis (Andrey e et al., 2012; Demirhan et al. ., 2008; Hanso on et al., 2010 0; Slack et al., , 2010; Xia et t al., 2012, 2012a). Considerin ng the inner an nd outer rings whose surface es are bigger a and stiffness is s higher than t the roller, the inner and outer rings are set as a target surfa aces and the ro oller is set as a contact surf face. Then, th he contact pair rs are created re espectively. Bo oth the contac ct stiffness co oefficient and the tolerance of penetratio on are key co ontact parameters s. The smaller r contact stiffn ness coefficient t is favorable to be convergent. However, , the bigger co ontact stiffness c coefficient is favorable f to im mprove the pre ecision. By m many times of calculation, th he contact stif ffness coefficient t is set to 1.5 and a the toleranc ce of penetrati ion is set by a d default value. 2.4 Loadin ng Before the e load and the constraint are applied, the no odal coordinat te systems belo onging to the m middle plane of o the roller are a all converted to t the cylindri ical coordinate es and the disp placement of n nodes of the m middle plane of o the roller alon ng the circumfe erential directio on are all cons strained. Furth hermore, a sym mmetry constraint is applied to t the cross-secti ion of the roller and all th he degrees of freedoms of t the outer surf face of the ou uter ring nodes are constrained. The load is s applied to the e finite elemen nt model after r the radial fre eedom of the in nner surface of o the inner ring is coupled. 3. Analysi is of Results Through m modeling, mesh hing, constrain ning, loading, and solving, t the results of th he surface stre ess of the rolle er are obtained b by finite eleme ent analysis. Th he normal dire ection contact stress produce ed by the roller r and the outer r ring raceway is s used to analyze the effect of taper error r of the outer ring raceway on the stress. The normal roller r loaded 1/5 5 of the dynam mic load rating is analysed, as s shown in Fig gure 2. When t the abscissa x1 takes the valu ue 5.4 mm that c corresponds to o the surface o of the middle part of the ou uter ring racew way, the conta act stress takes the largest val lue 1235.6 MP Pa. When the value of the a abscissa x1 is greater than 5 5.4 mm or less s than 5.4 mm m, the contact str ress gradually decreases. Ob bviously, the st tress is distrib buted uniforml ly at the middl le of the outer r ring raceway. According g to the differen nt values of th he taper error o of the outer ring raceway, fin nite element mo odels are recre eated. The taper error of the ou uter ring racew way is from 1 m to 5 m a and the results s of contact st tresses are given in Figures 3-7. It can be s seen from Figu ures 2-7 that th he maximum s stress of the ro oller surface ch hanges with th he taper error of o the outer ring raceway along g the axial dire ection. The larg ger the taper er rror, the larger r the maximum m stress. If the taper error exce eeds 3 m, the e value of the e maximum st tress increases s sharply, as s shown in Figu ure 8, showing g the unusually additional con ntact stress gen nerated by the m machining erro or.
87

www.ccsen net.org/mas

Modern n Applied Scienc ce

Vol. 6, No. 12; 2012

Figure 2. Contact stress s between oute er ring raceway y with 0m tap per error and ro oller In addition n, when the ta aper error of th he outer ring r raceway is 0 m, the distrib bution of the c contact stress is i the most unifo orm, i.e., the maximum m stres ss appears at x1=5.4 mm tha at corresponds to the surface e of the middle e part of the roll ler. With the increasing i tape er error, the lo ocation of the e maximum str ress gradually y deviates from m the middle par rt of the roller r. If the taper error exceeds 3 m, the loc cation of the m maximum stress sharply dev viates from the m middle part of o the roller, a as shown in F Figure 9, show wing the abno ormal contact stress distribution generated by the machin ning error. g to contact mechanics m and d rolling bear ring theories, the unusuall ly additional contact stress and According abnormal c contact stress distribution ge enerated by the e machining er rror belongs to o an unfavorab ble condition. If I the rolling bea aring runs at great speed und der such a cond dition, the axis s of rotation of the roller wo ould appear ske ewed and the rol ller can therefo ore get stuck in n two raceway ys, resulting in a safety incide ent. As a resul lt, in the proce ess of manufa acturing the ra aceway of the outer ring, th he taper error m must be controlled within an appropriate ra ange, such as 0 m-3 m for r the cylindrical roller bearin ng studied in t this work, whi ich is called the suitable upper r bound of the taper error of t the outer ring r raceway.

Figure 3. Contact stress between outer r ring raceway y with 1 m tap per error and roller

r ring raceway y with 2 m tap per error and roller Figure 4. Contact stress between outer

88

www.ccsen net.org/mas

Modern n Applied Scienc ce

Vol. 6, No. 12; 2012

Figure 5. Contact stress between outer r ring raceway y with 3 m tap per error and roller

Figure 6. Contact stress between outer r ring raceway y with 4 m tap per error and roller

Figure 7. Contact stress between outer r ring raceway y with 5 m tap per error and roller

Figure 8. Effect of tape er error on larg gest contact str ress

89

www.ccsen net.org/mas

Modern n Applied Scienc ce

Vol. 6, No. 12; 2012

Figure 9. Influ uence of taper error on locati ion of maximu um stress 4. Conclus sions The given load correspo onds to the suit table upper bo ound of the tap per error of the e outer ring rac ceway and wit th the increasing taper error, the t contact str resses between n the cylindric cal roller and the outer ring g raceway increase observably y and the distri ibution of the c contact stresse es presents mor re complex asy ymmetry and n nonuniformity. In the pro ocess of manu ufacturing the raceway of t the outer ring, , the taper err ror must be c controlled with hin a appropriat te range, such as 0-3 m fo or the cylindri ical roller bear ring studied in n this work, w which is called d the suitable up pper bound of the taper error r of the outer ri ing raceway. Acknowle edgement The resear rch is financed d by the Nation nal Natural Sci ience Foundati ion of China (G Grant No. 5107 75123). Reference es Andrey, S. Ruslan, P., & A Alexandr, D. ( (2012). Stresse ed state in the roller contact zone ., Hongrui, A., Yuan, J. H., R (area) ) and the bearing b ring finite length h. Advanced Materials R Research, 482 2-484, 2017-2 2020. http:/ //dx.doi.org/10 0.4028/www.sc cientific.net/AM MR.482-484.2 2017 Chandrase ekara, M. C. S., & Ramam mohana, R. A A. (1983). Mechanics and b behaviour of hollow cylind drical memb bers in rolling contact. Wear, r, 87, 287-296. http://dx.doi.o org/10.1016/00 043-1648(83)9 90132-1 Chiu, Y. P P., & Hartnett t, M. J. (1987 7). A numeric cal solution fo or the contact problem invo olving bodies with cylind drical surface considering cylinder effec ct. Journal of f Tribology-Tr Transactions of f the ASME, 109, 479-4 486. http://dx.d doi.org/10.1115/1.3261478 Demirhan, , N., & Kanber r, B. (2008). Stress and displ lacement distri ibution on cyli indrical roller b bearing rings using u FEM. Mechanic cs Based Design of Structu ures and Machines, 36, 86-102. http:/ //dx.doi.org/10 0.1080/15397730701842537 Hanson, M M. T., & Keer, L. M. (1995). Mechanics o of edge effects on frictionless contacts. Int ternational Jou urnal of Sol lids and Struct tures, 32, 391-405. http://dx. .doi.org/10.1016/0020-7683( (94)00153-N Slack, T., & Sadeghi, F. F (2010). Exp plicit finite el lement modeli ing of subsurf face initiated spalling in ro olling conta acts. Tribology International, , 43, 1693-170 02. http://dx.do oi.org/10.1016/ /j.triboint.2010 0.03.019 Sun, H. Y Y., Chen, X. Y., Y Liu, C. H., & Yang, P. R R. (2008). Stud dy on thermal l EHL perform mance of Lund dberg profil le rollers and th he modificatio on of its crown ning value. Trib bology, 28, 68-72. (in Chines se) Sun, H. Y., Chen, X. Y., & Zhang, H. X X. (2010). Cro owning design for the logarit thmic profile ro oller according g to a therm mal elastohydro odynamic lubri ication (EHL) theory. Tribol logy, 30, 567-5 571. (in Chines se) Wang, L. Q Q., Ye, Z. H., & Gu, L. (201 11). The effect t of roller profi file modificatio on on roller be earing performance. 1210-1 Advan nced Materials ls Research, 23 30-232, 1215. http:/ //dx.doi.org/10 0.4028/www.sc cientific.net/AM MR.230-232.1 1210 Xia, X. T. . (2012). Forec casting method for product reliability alo ong with perfo ormance data, Journal of Fa ailure Analy ysis and Prevention, 12(10), 532-540. http: ://dx.doi.org/10.1007/s11668 8-012-9592-y Xia, X. T T., Zhu, S. C., Jia, C. H., & Niu, R. J. (2012). Con ntact stress analysis of cylin ndrical roller with f convexity e logari ithmic curve generatrix of excursion. Journal of Hen nan University ty of Science and Techn nology, 33(5), 41-44. 4 (in Chi inese) Xia, X. T., , Zhu, S. C., Ji ia, C. H., & Ni iu, R. J. (2012a). Study of in nterval of arc m modification le ength of cylind drical roller r using ANSYS S. Research Inventy: Interna ational Journal l of Engineerin ng and Science e, 1(1), 8-13.
90