Constantine M.

Tarawneh1
e-mail: tarawneh@utpa.edu

Thermal Modeling of a Railroad Tapered-Roller Bearing Using Finite Element Analysis

In the railroad industry, distressed bearings in service are primarily identied using wayside hot-box detectors (HBDs). Current technology has expanded the role of these detectors to monitor bearings that appear to warm trend relative to the average temperatures of the remainder of bearings on the train. Several bearings set-out for trending and classied as nonveried, meaning no discernible damage, revealed that a common feature was discoloration of rollers within a cone (inner race) assembly. Subsequent laboratory experiments were performed to determine a minimum temperature and environment necessary to reproduce these discolorations and concluded that the discoloration is most likely due to roller temperatures greater than 232  C (450  F) for periods of at least 4 h. The latter nding sparked several discussions and speculations in the railroad industry as to whether it is possible to have rollers reaching such elevated temperatures without heating the bearing cup (outer race) to a temperature signicant enough to trigger the HBDs. With this motivation, and based on previous experimental and analytical work, a thermal nite element analysis (FEA) of a railroad bearing pressed onto an axle was conducted using ALGOR 20.3TM. The nite element (FE) model was used to simulate different heating scenarios with the purpose of obtaining the temperatures of internal components of the bearing assembly, as well as the heat generation rates and the bearing cup surface temperature. The results showed that, even though some rollers can reach unsafe operating temperatures, the bearing cup surface temperature does not exhibit levels that would trigger HBD alarms. [DOI: 10.1115/1.4006273] Keywords: railroad bearing thermal modeling, tapered-roller bearing heating, internal bearing temperatures, discolored rollers, excessive roller heating, thermal nite element analysis

Arturo A. Fuentes Javier A. Kypuros Lariza A. Navarro Andrei G. Vaipan

Department of Mechanical Engineering, The University of Texas-Pan American, Edinburg, TX 78539-2999

Brent M. Wilson
Amsted Industries Incorporated, 1700 Walnut Street, Granite City, IL 62040

Introduction
Tapered-roller bearings (see Fig. 1) are the most widely used bearings in railroad cars. When operated under satisfactory load, alignment, and contaminant free conditions, the service life is exceptionally long. As a general rule, bearings will outlast the wheel life, and survive several reconditioning cycles prior to being retired. At the end of their life, bearings will initiate fatigue, particularly subsurface fatigue, rather than wearing out due to surface abrasion. Fatigue failures, or spalling, can lead to material removal at the raceway surface which in turn will cause grease contamination and increased friction that manifests itself as heat within the bearing. Excessive heat will lower the viscosity of the lubricant, which reduces the thickness of the uid lm that separates the rolling surfaces. As a consequence, metal-to-metal contact occurs, which can hasten the onset of premature bearing failure. To identify distressed bearings in service, bearing health monitoring equipment is employed by the railroads to warn of impending failures as a method to ward off potentially catastrophic events, such as derailments. The most common method of monitoring bearing health is by conventional wayside hot-box detectors which are strategically located to record bearing cup temperatures as the train passes. These devices are designed to identify those bearings which are operating at temperatures greater than 105.5  C (190  F) above ambient conditions. An extension of this practice is the tracking of temperature data and comparing individual bearings against the averages of the remainder along a train (Karunakaran and Snyder [1]). Identifying those bearings which are trending above normal allows the railroads
1 Corresponding author. Manuscript received May 3, 2011; nal manuscript received January 21, 2012; published online July 12, 2012. Assoc. Editor: Chenn Zhou.

to track bearings which appear to be distressed without waiting for a hot-box detector (HBD) to be alarmed. As a diagnostic aid, bearings which are identied as hot are removed from service for later disassembly and inspection. In most cases, the cause of bearing overheating can be attributed to one of several known modes of bearing failure such as: spalling, water contamination, loose bearings, broken components, damaged seals, etc. However, in some cases, these early set-out bearings do not exhibit any of the commonly documented causes of bearing failure and are, therefore, classied as nonveried. Upon closer disassembly and inspection, it has been observed that many of these nonveried bearings contain discolored rollers in an otherwise normal bearing. The discoloration of the steel is visual evidence that these rollers have been exposed to temperatures greater than what is expected during normal operating conditions. Hence, initial work performed by the authors of this paper focused on determining conditions that would replicate the discoloration observed in the rollers. A laboratory furnace was used to heat numerous rollers to elevated temperatures in various

Fig. 1 Detailed component view of a typical railroad taperedroller bearing

Journal of Thermal Science and Engineering Applications C 2012 by ASME Copyright V

SEPTEMBER 2012, Vol. 4 / 031002-1

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

environments. Results indicated that the visual discoloration which best matched that observed in bearings removed from service were rollers heated in grease to temperatures over 232  C (450  F) for periods of at least 4 h (Tarawneh et al. [2]). The latter nding stirred many discussions and speculations among people in the railroad industry as to whether rollers can be at such elevated temperatures for these extended durations without generating sufciently high temperatures at the cup surface to trigger hotbox detectors. To this end, a number of dynamic laboratory tests were carried out to explore several defects and/or hypothetical scenarios that can lead to bearing overheating. Details regarding these dynamic tests can be found elsewhere (Tarawneh et al. [2]). Simultaneously, several static experiments were conducted using cylindrical heaters embedded in two rollers in order to better understand the heat transfer paths within the bearing assembly (Tarawneh et al. [3]). The data acquired from the static experiments were used to determine overall heat transfer coefcients for heat dissipation from the bearing cup and heat loss to the axle, which allowed for the derivation of analytical expressions for the temperatures of the bearing cup and the bearing/axle interface. However, even though the aforementioned work offered great insight into the thermal behavior of tapered-roller bearings, it had its limitations. Only external bearing cup temperatures could be recorded in the dynamic experiments since it is not feasible to measure the temperatures of the internal components of the bearing while it is rotating. In the static tests, monitoring the temperature of the rollers was possible, but the case studies were limited to a setup utilizing two cartridge heaters embedded in two rollers to provide the heating source. Additional embedded heaters could have been used, but it would have added greatly to the complexity of the experimental setup and instrumentation, not to mention the time and effort involved in conducting these experiments. Numerical simulations tend to be a more economical means of obtaining prompt results and an efcient way to overcome the experimental challenges and limitations. Hence, ALGOR 20.3TM was utilized to develop a nite element (FE) model for a class K taperedroller bearing mounted to an axle, as shown in Fig. 2. The ndings of the previous experimental and analytical work performed by Tarawneh et al. [2] and [3] were used to devise the FE model and validate its accuracy. The FE model was then utilized to run several different bearing heating simulations that provided denitive answers as to whether it is possible for rollers to heat to high temperatures without heating the cup surface to a sufcient temperature necessary to trigger any HBD alarms. The paper presented here provides a thorough summary of the results attained from the aforementioned steady-state thermal nite element analysis (FEA).

Literature Review
Many studies have utilized the FE method to develop models to analyze specic phenomena of interest concerning different types

Fig. 2 Mesh results for the solid model of the bearing-axle assembly that was used to perform the nite element analysis in this study

of bearing assemblies and their components. In most cases, experimentally acquired data are used to validate the accuracy of the verstam [4] performed FE simulations using derived FE models. O the MSC.MARC code to study the effect of bearing geometry on the residual stress-state in cold drawn wires, and veried the simulation results with data acquired from full-scale industrial experiments. Wei et al. [5] used FE simulations to demonstrate that their proposed design of a deep end-cavity roller would enable a straight-prole-roller bearing to perform similar to a logarithmicprole-roller bearing by eliminating the sharp edge-stresses at the two apexes of the rollers. Furthermore, the FE analysis shows that the deep end-cavity roller design reduces the centrifugal force acting on the outer race of the bearing while saving material and reducing the weight of the bearing. Demirhan and Kanber [6] used ANSYS to investigate the stress and displacement distributions on cylindrical roller bearing rings. The study concluded that the stresses and displacements have different distribution characteristics on the inner and outer faces of the rings and are not uniformly distributed along the height of the rings because of large stresses at the contact points. Vernersson [7] developed a FE model for heat transfer from the rolling wheel into the rail where a lm with thermal contact resistance is placed at the wheel-rail contact interface. The proposed FE model, which is validated by experimental results, can be used to efciently design tread braking systems for both freight and passenger trains. Other works that utilized the FE method to study magnetic and journal bearings are reported in Schmidt and Weiland [8], Awasthi et al. [9], and Sukumaran Nair and Prabhakaran Nair [10]. Theoretical investigations of bearing overheating were carried out by Dunnuck [11] and Wang [12] who explored two abnormal operating conditions: (1) a jammed roller bearing and (2) a stuckbrake situation. A partially jammed roller bearing is one that is rotating with velocities greater than zero but less than the epicyclic speed, whereas, a fully jammed roller bearing is one that has no angular velocity with respect to the cage. The theoretical studies revealed that the maximum temperature within the bearing assembly can reach 268  C for a jammed roller and 126  C for a stuck-brake, compared to the normal operating condition temperature of 81  C. However, the study also concluded that the stuckbrake heating scenario resulted in only a small increase in the external surface temperature of the bearing cup, which suggests that it would be difcult to detect a stuck-brake situation from the cup surface temperatures. In a related study, a dynamic model of the torque and heat generation rate in tapered-roller bearings under excessive sliding conditions was developed using the program SHABERTH (Wang et al. [13]). The investigation focused on jammed roller bearings, and the model was run with an assumed ambient temperature of 25  C (77  F), a load per row of 80,000 N (18,000 lb), and a rotational speed of 560 rpm which corresponds to a train speed of 97 km/h (60 mph). The study concluded that the heat generated in the bearing was proportional to the number of jammed rollers. In yet another theoretical work, thermally induced failures in railroad class F (61=2 12) tapered-roller bearings observed in laboratory experiments when the bearings were operating at high speeds were modeled using nite element analysis with ABAQUS and FORTRAN (Kletzli et al. [14]). The experiments were conducted by the Association of American Railroads (AAR), and showed that new (defect-free) bearings failed after 200300 h of operation at a speed of 161 km/h (100 mph), but none failed at 129 km/h (80 mph). The study concluded that the increase in the heat generation is a direct consequence to the grease starvation mechanism caused by the high operating speeds, which results in a larger friction coefcient. A few studies were conducted using simplied experimental setups designed to mimic the operation of roller bearings (Farneld [15] and Hoeprich [16]). The main outcome of some of these experiments was that bearing temperatures predominantly depended on speed and not on load, and that increasing the amount of lubrication slightly raises bearing temperatures. Transactions of the ASME

031002-2 / Vol. 4, SEPTEMBER 2012

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

Theoretical modeling of the thermal effects in plain journal bearings was reported in Wang and Zhu [17], Fillon and Bouyer [18], Wang et al. [19], Li et al. [20], and Ma and Taylor [21] with limited experimental validation. In addition, Briot et al. [22] looked into the thermal transport conductance between the rings of a roller bearing. The thermal and dynamic behavior of railroad tapered-roller bearings has been explored extensively through several experimental and theoretical studies conducted by the authors of this paper. First, in a series of ve papers, Tarawneh et al. ([3], [23], and [24]) and Cole et al. ([25] and [26]) examined the heat transfer paths within tapered-roller bearings, and heat transfer to the bearing from an adjacent hot railroad wheel. Experimentally validated analytical expressions were developed to describe the surface temperature of the bearing cup, the temperature at the cone-axle interface, and the temperature along the wheel web. Additionally, the aforementioned studies resulted in the determination of heat transfer coefcients for heat dissipation from the bearing and wheel surfaces which can be used to devise reliable FE models. Second, in a series of four papers, Tarawneh et al. ([2], [27][29]) investigated the warm bearing temperature trending problem and were able to identify the root cause of this troubling phenomenon. It was concluded that vibration induced roller misalignment is the likely cause for the bearing temperature trending phenomenon seen in service (Tarawneh et al. [27]), and an on-track eld test conducted by the authors in collaboration with The Union Pacic Railroad, Rail Sciences Incorporated, and Amsted Rail Industries validated the results obtained from the laboratory testing (Tarawneh et al. [28]). Furthermore, vibration signatures of temperaturetrended bearings in eld and laboratory testing are provided in Tarawneh et al. [29]. From the literature review provided here, the need for reliable numerical models that can predict internal bearing temperatures becomes apparent. With this motivation, the authors utilized their earlier experimental results and analytical models to develop a FE model that can be used to simulate numerous normal and abnormal operating scenarios, that are otherwise very complex and time consuming to duplicate in a laboratory setting, and acquire internal and external bearing temperatures. The developed model can prove to be a very useful tool in future thermal research of railroad tapered-roller bearings. Knowing the temperature distribution within the internal components of the bearing during normal and abnormal operating conditions can help bearing manufacturers explore possible design modications to their bearings to dissipate the heat more efciently or select appropriate lubricants that can withstand the internal temperatures experienced by the bearings.

Finite Element Analysis

The main objective of the work presented here is to study how roller heating affects the temperature of the bearing assembly components, especially the external surface of the cup, which is the part of the bearing scanned by the infrared wayside hot-box detectors. To this end, a FE model was created and utilized to run several roller heating simulations in order to attain answers as to whether it is possible for certain rollers to heat to temperatures above 232  C (450  F) within the bearing and go undetected by the HBDs. This study would be very challenging to perform experimentally due to the constraints imposed by placing conventional thermal sensors inside a rotating assembly. The following two subsections, Bearing Modeling and Boundary Conditions, provide a detailed description of the developed FE model including the boundary conditions (BCs) used. Bearing Modeling. A solid model of the bearing-axle assembly that was generated utilizing the three-dimensional graphing software Pro/EngineerTM was used to develop the FE model for this study. The axle is rendered as a simple cylinder with a 0.1572 m diameter and a 2.2 m length, which is sufcient in size to act as a heat sink. The bearing is modeled after a class K railJournal of Thermal Science and Engineering Applications

road tapered-roller bearing with some minor modications that simplied the geometry and resulted in a signicant reduction to the computational time. First, no cages, seals, wear rings, or grease were included in the model; this was done because the thermal resistances of the polyamide cages and grease are large compared to the rest of the bearing assembly, so the majority of the heat will ow from the rollers to the bearing cup and inner cones. Recent advances in bearing seal technology have resulted in minimal contact low friction seals. Furthermore, these seals constitute a small fraction of the total weight of the bearing and are separated from the rest of the internal bearing components by a combination of air and grease which both have high thermal resistances. The wear rings are in contact with the axle, which constitutes a very large body of metal and acts as a heat sink. Hence, the omission of the aforementioned components from the FE model will not have a signicant effect on the results acquired from this study. The latter statement is validated both experimentally and theoretically as will be shown later in this paper. The second assumption is concerned with the contact area between the rollers and the cup and cone raceways. Under normal operating conditions, only the upper hemisphere of a railroad bearing is loaded; therefore, larger contact areas exist between the rollers and the cup and cones in this region. However, since the rollers of a bearing enter and exit the loaded zone continuously as they execute multiple revolutions in a second, an average contact area was applied to all 46 rollers in the bearing. Hence, in the model, the contact area between each roller and the cup and cone raceways is 123.96 mm2, which represents 3.68% of the total surface area of the roller. The contact area was obtained by rst estimating the initial contact area of the rollers with the cup and cones using the Hertzian line contact theory (Bra ndlein et al. [30] and Harris and Kotzalas. [31]), and then using FE analysis to determine the load distribution on the upper hemisphere of a fully-loaded bearing (full load corresponds to a force of 159,000 N (35,750 lb) per bearing applied through the bearing adapter), which was then used to calculate the amount of roller compression. A detailed description of this methodology is provided elsewhere (Alnaimat [32]). Note that the contact area used in the model represents the average contact area of all the rollers within a cone assembly when fully-loaded. A sensitivity analysis that was carried out to investigate the effect of the contact area revealed that the results changed by less than 4% when the contact area was varied by 610%. Once the model was completed, it was imported into ALGOR 20.3TM and discretized into 68,086 elements with a mesh size of 4.0 mm, which is the largest element size that can be used without jeopardizing the convergence of the FE model while keeping the computational time relatively low ($20 min on a DELL OPTIPLEX 755 Minitower, Core 2 Duo E8200/2.66 GHz processor). The convergence analysis run on the model revealed that the attained results varied by less than 0.7% when the mesh size was changed by 610%. The nal meshed model is illustrated in Fig. 2. The ALGOR bricks and tetrahedral solid mesh type option was used since it generates the most accurate mesh utilizing the fewest elements. Even though the use of brick and hexahedral elements can sometimes yield better results [33], the generated mesh contains many more elements, which substantially increases the analysis time. In this study, the use of hexahedral elements resulted in an insignicant improvement on the acquired results (<0.5%) while more than tripling the analysis time. Note that, while the mesh contained six-node wedge, ve-node pyramid, and four-node tetrahedral elements toward the center of the model, the majority of the solid mesh consisted of eight-node brick elements. The benet of using a small number of nonbrick elements (i.e., wedge, pyramid, tetrahedral) is to signicantly reduce the number of elements needed to accomplish the solid mesh, thus, decreasing computational time. Additionally, surface knitting (i.e., the process of recognizing where two parts have surfaces in contact and then splitting the surfaces) was used in order to properly apply convection loads. SEPTEMBER 2012, Vol. 4 / 031002-3

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

For the nite element thermal analysis, different solvers were considered to work out the system of equations including the iterative algebraic multigrid (AMG) and the Sparse solver. While there were no signicant differences in the performance of the different solvers, the iterative AMG solver with a convergence tolerance of 0.000001was selected based on the analysis time for our model (the iterative AMG solver takes 30% less time to run than the Sparse solver). Furthermore, the application of radiation to the model required a nonlinear iterative approach in order to solve the steady-state heat transfer analysis. To start the nonlinear iterative process, the ambient temperature was used as the default nodal temperatures. Then, in each successive iteration, the radiation heat transfer rate was estimated based on the prior estimate of the nodal temperatures. The resulting temperatures were then compared to the prior estimate of the nodal temperatures. The iterative process was stopped when the relative norm (i.e., the relative change in temperature between iterations) was less than a relative tolerance of 0.001. If the difference was not acceptable based on the relative tolerance, the radiation heat rates were recalculated and the process repeated. Note that, the convergence tolerance determines how accurate of a solution is found to the matrix of equations (i.e., the smaller the tolerance, the more accurate the solution), whereas, the relative tolerance is used to prescribe the quality of convergence of the temperatures due to the radiation loads and to stop the iterative process. At this point, a comment is in order regarding the FE model developed for this study. In service, a railroad bearing is mounted at the end of the axle next to the wheel and is coupled to the sideframe of the railcar via an adapter. Thus, the bearing is connected to a semi-innite body of metal and can conduct heat to the axle through the inner cones and to the side-frame through the cup surface in contact with the adapter. The model utilized in this investigation uses a bearing that is pressed onto the middle of a 2.2 m axle in order to provide comparable heat conduction paths to the ones the bearing experiences in service, without the added complexity that the actual setup would impose on the FEA. Furthermore, the bearing overall heat transfer coefcients used in this study were acquired utilizing an experimental setup similar to the one depicted by the FE model (Fig. 2) with a full-load applied to the bearing through the adapter using a hydraulic cylinder (Tarawneh et al. [3]). Hence, based on the above discussion, the FE model devised for this investigation is assumed to approximate the actual setup of a bearing in service. Boundary Conditions. The validity of the FE model depends greatly on the correctness of the BCs applied when running the simulations. With this in mind, the BCs used for this study were derived from previously conducted experimental efforts (Tarawneh et al. [2], [3], and [24]), a well-established textbook in the eld of heat transfer (Incropera et al. [34]), and from material specications provided by the bearing manufacturer. Four major BCs, which are described in this section, were utilized; namely: conduction, convection, radiation, and heat ux. The heat conduction coefcients for the bearing assembly and axle were provided by the bearing manufacturer. For the bearing steel, AISI 8620 with a thermal conductivity of 46.6 W m1 K1 was used, and for the axle steel, AISI 1060 with a thermal conductivity of 51.9 W m1 K1 was chosen. A sensitivity analysis was performed to ensure that variations in the thermal conductivity values due to temperature have a marginal effect on the reported results; the results differed by less than 1% when both thermal conductivity values were changed by 610%. Note that the thermal conductivity values for both AISI 8620 and AISI 1060 do not change by more than 7% over the range of temperatures reported in this study. Convection and radiation BCs for the bearing were acquired from the results of previous experimental testing conducted by the authors. Tarawneh et al. [3] provides an overall heat transfer coefcient Ho 8.32 W K1 for the bearing cup, which takes into 031002-4 / Vol. 4, SEPTEMBER 2012

Fig. 3 Solid model of the bearing-axle assembly showing how the boundary conditions (BCs) were applied for the FE analyses conducted for this study. Note: (1) heat ux was applied to the circumferential surface of the rollers only, (2) bearing overall heat transfer coefcient was applied to the bearing external surfaces (i.e., bearing cup external surface and side walls, and bearing cone side walls), and (3) axle heat transfer coefcient and radiation were applied to all exposed surfaces of the axle.

account forced convection generated by a 5 m s1 airstream and radiation to an ambient at a temperature of 25  C. However, since the software used for the FE simulations requires convection coefcients to be entered in units of W m2 K1, the external surface area of the bearing cup Acup 0.1262 m2 was used to obtain the heat transfer coefcient in the appropriate units (ho 65.9 W m2 K1). The latter overall heat transfer coefcient was applied to the external (exposed) surface of the bearing only, as illustrated in Fig. 3. For the axle, the cylinder-in-cross-ow correlation, Eq. (1), and the ow over a at surface correlation, Eq. (2), given in Incropera et al. [34] (Chap. 7, p. 427 and 410, respectively), were utilized to calculate the convection heat transfer coefcient from the circumferential surface and the two ends of the axle, respectively. Both correlations yielded a convection coefcient value of haxle 25 W m2 K1 corresponding to a 6 m s1 airstream and a 25  C ambient temperature. "  5=8 #4=5 0:62Re1=2 Pr1=3 Re Nu 0:3 h 1 i 1=4 282; 000 1 0:4=Pr2=3 hLc ; Pe ! 0:2 k 8 9 > < 0:3387 Re 1=2 Pr 1=3 > = hL c Nu 2 h ; i 1=4 > > k : 1 0:0468=Pr2=3 ; where Re VLc ; v Pe Re Pr (3) (1)

Pe ! 100

(2)

In Eqs. (1)(3), the characteristic length, Lc, is equal to the diameter, D, of the axle. Table 1 provides the numerical values for the properties and parameters appearing in Eqs. (1)(3). Note that, even though the correlations given in Eqs. (1) and (2) are for a stationary cylinder, the acquired convection coefcient is still valid for a rotating cylinder since the air-ow across the cylinder is moving relatively fast (Re 55811 ) 100). The literature review conducted by the authors revealed three references (Jones et al. [35], Badr and Dennis [36], and Kendoush [37]) which conclude that the effect of convection produced by a rotating cylinder need only be considered when the air-ow across the cylinder is Transactions of the ASME

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

Table 1 Property Symbol Units Value

a

Numerical values for the properties and parameters appearing in Eqs. (1)(3) Thermal conductivitya k W m1 K1 27.0 103 Kinematic viscositya  m2 s1 16.9 106 Velocity V m s1 6.0 Characteristic length Lc m 0.1572 Prandtl numbera Pr None 0.706

Film temperature Tf K 310

All thermal properties were obtained at Tf from Ref. [34], Appendix A, p. 941.

Table 2 Summary of the performed nite element (FE) simulations. The average cup temperatures provided in the table were obtained by averaging six nodes simulating the thermocouples placed around the circumference of the middle of the cup 60 deg apart. Case No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Qtotal (W) 529.0 672.5 712.0 1086.0 908.5 754.0 1484.5 1287.4 1438.2 1249.0 979.8 1163.8 2208.0 Qroller (W) 11.5 155.0 42.0 290.0 28.0 124.0 489.3 264.3 238.8 731.5 21.3 25.3 48.0 Maximum average roller temperature ( C) 55.0 120.8 75.9 218.2 83.3 110.7 291.8 232.0 233.4 388.6 80.4 90.8 149.5 Average cup temperature (8C) 50.2 57.5 59.0 79.2 68.6 59.7 90.2 88.5 96.8 80.0 71.9 80.7 130.5

Description of heating scenario Normal operation. All 46 rollers are heated equally to produce Tcup 50  C (see Fig. 5) One welded roller. 45 rollers heated to normal operation conditions; one roller abnormally heated Six welded rollers. 40 rollers heated to normal operation conditions; six rollers on one cone assembly abnormally heated (see Fig. 6) Twisted cage bar. 44 rollers heated to normal operation conditions; two adjacent rollers abnormally heated (see Fig. 7) Added debris. One cone assembly (23 rollers) heated to normal operation conditions; the other cone assembly abnormally heated Two hot rollers. 44 rollers heated to normal operation conditions; two rollers, one on each cone assembly, abnormally heated Two hot rollers. 44 rollers heated to normal operation conditions; two rollers, one on each cone assembly, abnormally heated (see Fig. 8) Three hot rollers. 43 rollers heated to normal operation conditions; three consecutive rollers abnormally heated (see Fig. 9) Four hot rollers. 42 rollers heated to normal operation conditions; four consecutive rollers abnormally heated Misaligned roller. 45 rollers heated to normal operation conditions; one roller abnormally heated Abnormal operation. All 46 rollers are heated equally to produce Tcup 72  C Abnormal operation. All 46 rollers are heated equally to produce Tcup 80  C Abnormal operation. All 46 rollers are heated equally to produce Tcup 130.5  C (see Fig. 10)

relatively slow (i.e., ows with a Re 100). Furthermore, a sensitivity analysis was carried out to assess the effect of the axle convection coefcient on the reported results; the results differed by less than 5% when the axle convection coefcient was increased by 20%. The only parameter needed to calculate radiation from the axle to the ambient was emissivity, and it was measured to be about 0.96 from previous experimentation (Tarawneh et al. [24]). Again, a sensitivity analysis was performed on the emissivity value used in this study, which revealed that the results differed by less than 1% when the emissivity value was lowered by 20%. Finally, to simulate heat generation within the bearing assembly, heat ux was applied to the circumferential surface of the rollers, as depicted in Fig. 3. The appropriate heat ux value was determined through a trial-and-error process starting with an overall heat input of 11.5 W per roller (normal operation conditions) and increasing this input until the desired external cup temperature was achieved. The acquired heat input per roller was then divided by the surface area of the roller to obtain the heat ux value. Here, it is assumed that the rollers are the source of heat within the bearing which is justied considering the mass of the roller (0.145 kg) relative to the mass of the bearing cup (11.53 kg) and cone (3.9 kg). To illustrate this, consider an abnormal operation condition in which debris gets wedged between a roller and an adjacent cage bar causing the roller to become fully or partial Journal of Thermal Science and Engineering Applications

jammed and resulting in excessive sliding friction between the roller and the cup and cone raceways. Since the mass of the roller is very small compared to the mass of the cup and cone, it is safe to assume that it will heat at a much faster rate than the other two components, thus, becoming the heat source.

Discussion of Results
Thirteen different bearing heating scenarios, summarized in Table 2, were simulated for this study. The rst ve cases are based on previous experimental work conducted by Tarawneh et al. [2], whereas, the remaining eight cases simulate hypothetical heating scenarios that can result from abnormal roller operation leading to excess frictional heating. The results of both, the FE analysis and the previously validated lumped-capacitance theoretical model (Tarawneh et al. [3] and [23]), are presented and compared hereafter. An in-depth discussion is reserved for six of the thirteen simulated cases, with the purpose of highlighting pivotal information that can help answer the question posed earlier in this paper; i.e., is it possible to have rollers reaching 232  C (450  F) within a cone assembly without heating the bearing cup to a temperature that will trigger the HBDs? FE Model Validation. As stated earlier, the rst ve heating scenarios listed in Table 2 were intended to replicate ve of the SEPTEMBER 2012, Vol. 4 / 031002-5

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

dynamic experiments performed for a previous study (Tarawneh et al. [2]) aimed at identifying possible causes of warm bearing temperature trending. These dynamic tests were performed in a laboratory with an ambient temperature of 25  C utilizing a dynamic bearing tester that maintained a speed setting of 536 rpm corresponding to a train traveling with a velocity of approximately 91.7 km h1 (57 mph) at full load (which corresponds to a force of 159,000 N (35,750 lb) per bearing). The source of heating in those ve dynamic tests is known, which makes them uniquely invaluable for validating the FE model by providing a basis for comparison with the lumped-capacitance analytical model. A brief description of the rst ve heating scenarios follows. Scenario number one, normal operation, simulates a bearing running under normal operating conditions. This model represents a healthy bearing with an average cup temperature of approximately 50.0  C (this temperature is the average of six thermocouple readings positioned 60 deg around the circumference of the cup at the middle of the bearing). The aforementioned temperature was acquired from the dynamic testing described earlier in this section. Scenario number two, one welded roller, represents the case in which one roller in the cone assembly was welded to the steel cage causing it to slide on the cup raceway instead of roll, thus, generating excess heat through friction. The remaining 45 rollers were assumed to be operating normally. In eld service, this scenario simulates a case in which a wear particle or debris becomes wedged between the roller and the cage or one of the raceways causing it to slide on the raceway rather than roll. The experimental results for the welded roller case indicate that the average cup temperature is about 57.4  C. Scenario number three, six welded rollers, is similar to the previous scenario but replicates the test in which six rollers in one cone assembly were welded to the steel cage, thus, producing sliding friction. The remaining 40 rollers were assumed to be operating normally. Again, in service, this case might occur if multiple wear particles from a spalled raceway, for instance, get lodged between the rollers and the cage. The average cup temperature for this case is 58.9  C, which is only slightly higher than that of the one welded roller case. Scenario number four, twisted cage bar, simulates the case in which a cage bar, in one cone assembly, was bent toward the cone race causing the two adjacent rollers to rub the cage generating excess heat. The latter case can occur in the eld from a sudden large impact caused by a bad joint in the track. The average cup temperature obtained from the laboratory test data is 79.2  C indicating that this case produced signicant frictional heating. Scenario number ve, added debris, as the name suggests, replicates the test in which debris (a mixture of sandblaster sand and metal lings) was added to the raceway of one cone assembly through a hole in the adjacent seal. The added debris causes the 23 rollers in that cone assembly to operate abnormally which results in increased frictional heating produced by the sliding of rollers. In eld service, the latter case can occur in a number of ways, namely; debris generated from spalled raceways, contaminants introduced to the bearing assembly by accident when the seals are removed to conduct a visual inspection, or in extreme weather conditions when the train goes through sandstorms. The average cup temperature attained experimentally for this case is 68.6  C. The other eight heating scenarios, described in Table 2, were derived from the rst ve cases and were intended to provide a good understanding of the effects of abnormal roller operation, and demonstrate that it is possible to have one or more rollers heating to temperatures above 232  C without triggering wayside HBDs. At this point, a comment is in order on the cup temperature. A rotating bearing with one or two hot rollers would have the hot rollers sweeping round and round the cup, uniformly heating the cup, even at modest train speeds. Our FE model is nonrotating, so the cup temperature is high near the hot rollers and lower around 031002-6 / Vol. 4, SEPTEMBER 2012

Fig. 4 A picture of the dynamic bearing tester used to conduct the laboratory experiments

the circumference of the cup. To simulate the averaging of cup temperature that a rotating bearing would produce, a spatialaverage of the acquired cup temperatures was used to obtain a single value that represents the average cup temperature reported in this study. The spatial averaging simulates the time averaging that a rotating bearing would yield. A numerical example will provide a framework for understanding this averaging effect. In a train moving at a leisurely 11.2 m s1 (25 mph), a 91.4 cm (36 in) diameter wheel sees 234 revolutions per minute (rpm). Tarawneh et al. [3] show that the thermal response time of the bearing cup is measured in tens of minutes, indicating that thousands of roller orbits would spread the heating over the entire bearing cup on a moving train. To validate and quantify the accuracy of the results acquired from the FE model, both an experimental and a theoretical technique were employed. The experimental method required the use of a dynamic bearing tester (pictured in Fig. 4) powered by a 22 kW (30 hp) motor that is managed by a smart controller. The controller has outputs that can be monitored via a data acquisition system to record the axle speed and the power required to run the bearings. In an experiment aimed at replicating the normal operation of bearings, four fully assembled healthy class K bearings were run utilizing the aforementioned dynamic bearing tester at a speed of 91.7 km h1 (57 mph) with a full load. The steady-state power exerted by the motor to rotate all four bearings at the set speed and load was found to be approximately 2100 W, which corresponds to 525 W per bearing. The average cup temperature of each of the four bearings was about 50.0  C in a room with an ambient temperature of 25.0  C. The latter power input acquired experimentally is within 1% of the value obtained by the FE model for the case of normal operation given in Table 2. Here, it is important to note that the power input per bearing obtained experimentally by the authors is much lower than the 1307 W theoretical value reported in Dunnucks [11] thesis for similar operating conditions. In fact, the only way to attain a power value similar to that of Dunnucks was to run the dynamic bearing tester at 136.8 km h1 (85 mph) and 135% load setting, which are far from normal operating conditions. The theoretical technique for validating the devised FE model involved the systematic comparison of the average axle surface temperature (this is the average temperature of the surface of the axle covered by one cone assembly) attained from each of the simulations to that obtained from the analytical method developed previously by Tarawneh et al. [3]. The theoretical expressions derived for the axle surface temperature are the following: T2 t where Transactions of the ASME Qin T t Ta Ho 2 (4)

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

Table 3

Comparison of temperature values between the analytical model and the FE analysis Finite element (FE) model Analytical method Axle temperature ( C) 51.4 58.6 60.6 79.2 70.4 62.7 99.1 89.3 96.8 87.4 73.9 83.1 135.3 % difference T2 1.2 1.4 1.4 1.9 1.8 1.3 1.2 1.8 1.8 1.3 1.6 1.7 1.6

Simulations 1 2 3 4 5 6 7 8 9 10 11 12 13 Normal operation One welded roller Six welded rollers Twisted cage bar (two hot rollers) Added debris (one raceway) Two hot rollers Two hot rollers Three hot rollers Four hot rollers Misaligned roller Abnormal operation (Tcup 72  C) Abnormal operation (Tcup 80  C) Abnormal operation (Tcup 130.5  C)

Total heat input Qtotal (W) 529.0 672.5 712.0 1086.0 908.5 754.0 1484.5 1278.4 1438.2 1249.0 979.8 1163.8 2208.0

Axle temperature ( C) 52.0 59.4 61.4 80.7 71.6 63.4 100.3 90.8 98.6 88.5 75.1 84.5 137.4

2 T2

3 

1Rr

p
1Rr2 4r1RRJ

2 J 2J r 6 7 q 4 t 5 e 1 R RJ 2 2 1 R r 1 R r 4r1 R RJ 1 R r 4r1 R RJ 2 3  p 1Rr

2J r 6 7 q 4 5e 2 2 1 R r 1 R r 4r1 R RJ 1 R r 4r1 R RJ

1Rr2 4r1RRJ 2

(5)

is the In Eqs. (4) and (5), T2 is the axle surface temperature, T2 normalized axle surface temperature, Ta is the ambient temperature, Qin is the total heat input generated by the bearing, t is the normalized time, J is the Bessel function, r is the normalized thermal diffusivity, Ho is the cup side overall heat transfer coefcient, H1 is the axle side overall heat transfer coefcient, and R is the axle to cup side overall heat transfer coefcient ratio, H1/Ho. Here, it is important to highlight that the parameters Ho and H1 were determined by comparing bearing temperature data acquired experimentally to the analytical expression obtained from theory utilizing nonlinear regression routines available in the MATLABTM optimization toolbox. Detailed relations for the terms in Eq. (5) can be found elsewhere (Tarawneh et al. [3]). The FE model is validated by inputting the value of the applied total heat rate, Qtotal, from each simulation into Eq. (4) and calculating the corresponding axle surface temperature, T2, which is then compared to the average axle surface temperature predicted by the FE model. The results of the comparison are presented in Table 3. Note that the percent difference in the axle surface temperature between the analytical method and the FE model is less than 2%, which indicates a very good agreement between the two methods. Hence, based on the experimental and theoretical validations provided here, the developed FE model can be used to run bearing heating simulations and acquire reliable predictions of the temperature of the internal components of the bearing assembly without the complexity and limitations imposed by laboratory testing and instrumentation.

Selected Cases of Interest. For the sake of brevity, six of the thirteen heating scenarios were chosen for an in-depth discussion as these cases provide the reader with a comprehensive understanding of the conducted study. Note that in Table 2, the column labeled Qtotal indicates the total heat input to the bearing, whereas, the column labeled Qroller species the heat input to the abnormal roller(s), with all other normally operating rollers within the bearing assembly producing only 11.5 W. Journal of Thermal Science and Engineering Applications

The rst scenario to be discussed is that of normal operation (simulation 1 in Table 2), which is a fundamental heating scenario that represents a healthy bearing in service. Dynamic testing of numerous healthy bearings in a laboratory maintained at an ambient temperature of 25  C revealed that their average cup temperature is approximately 50  C (122  F), and the power input per bearing is about 525 W. The FE model simulation replicating the latter normal heating scenario indicates that the total heat input, Qtotal, needed to attain the 50  C cup temperature is about 529 W (within 1% of the experimentally obtained value), which translates into a roller heat input, Qroller, of 11.5 W (for each of the 46 rollers), and a maximum average roller temperature of 55  C. Thus, under normal operating conditions, the temperature of the rollers is only about 5  C (9  F) hotter than the average cup temperature. Figure 5 shows the results of the simulation along with the temperature distribution. Note that the axle was suppressed from the visual results to provide a better temperature visualization of the bearing surface. In the heating scenario six welded rollers (simulation 3 in Table 2), the FE model simulation presented in Fig. 6 replicates the laboratory dynamic test in which six rollers in one cone assembly were welded to the steel cage bars causing them to slide on the cup raceway rather than rotate. The results indicate that the operating temperature of the welded rollers was about 76  C, which is 21  C ($38  F) hotter than the temperature of a roller in normal operation, yet, the average cup temperature of the bearing with six welded rollers was only 9  C above that of a normally operating bearing. The latter provides initial proof that a number of rollers within a bearing may experience considerable heating events without affecting the average cup temperature signicantly since the bearing cup averages all the roller temperatures. Further evidence of the aforementioned can be seen in the following FE model simulation. The twisted cage heating scenario (simulation 4 in Table 2), shown in Fig. 7, simulates the test in which one of the cage bars in a steel cone assembly was bent toward the cone race forcing two adjacent rollers to misalign and rub against the cage bar producing SEPTEMBER 2012, Vol. 4 / 031002-7

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

Fig. 5 Thermal FE analysis results for normal operation conditions. Axle was suppressed from the visual results to provide a better temperature visualization of the bearing surface (heating scenario 1 in Table 2).

Fig. 8 Thermal FE analysis results for two hot rollers (one on each cone assembly) (heating scenario 7 in Table 2)

Fig. 6 Thermal FE analysis results for six welded rollers in one cone assembly (heating scenario 3 in Table 2)

Fig. 9 Thermal FE analysis results for three consecutive hot rollers (heating scenario 8 in Table 2)

Fig. 7 Thermal FE analysis results for twisted cage bar (heating scenario 4 in Table 2)

Fig. 10 Thermal FE analysis results for all rollers heated equally to produce a 130.5  C average cup temperature (heating scenario 13 in Table 2)

excessive frictional heating. The simulation results illustrate how the two misaligned rollers reach an operating temperature of 218.2  C (425  F), whereas, the average cup temperature does not exceed 80  C, which is only 30  C above that of a bearing in normal operation. Therefore, even though the latter bearing contains two hot rollers operating at unsafe temperatures, conventional track-side HBDs will not trigger an alarm since these devices will only alert when a bearing cup temperature is 105.5  C (190  F) above ambient temperature. Rollers operating at very high temperatures will degrade the lubricant and can cause grease starvation, which may lead to bearing seizure and eventual catastrophic failure. The abovementioned FE model simulations are based on dynamic bearing testing that was previously conducted in the lab031002-8 / Vol. 4, SEPTEMBER 2012

oratory. The FE model results provide internal temperature data that could not be obtained experimentally due to instrumentation limitations associated with placing temperature sensors inside a rotating bearing. To further investigate the effect of hot rollers on the temperature of the bearing cup, which is the surface scanned by the infrared wayside HBDs, several hypothetical heating scenarios were explored; three of which are discussed hereafter. The question posed earlier in this paper as to whether it is possible for certain rollers to heat to temperatures above 232  C (450  F) within the bearing and go undetected by the HBDs can be answered by looking at the results of simulations 710. Simulation 7 in Table 2, shown in Fig. 8, models the case in which two rollers, one on each cone assembly, are operating abnormally. The Transactions of the ASME

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

Table 4

Relevant temperatures obtained through the FE simulations Finite elementtemperature results

Simulations 1 2 3 4 5 6 7 8 9 10 11 12 13 Normal operation One welded roller Six welded rollers Twisted cage bar (two hot rollers) Added debris (one raceway) Two hot rollers Two hot rollers Three hot rollers Four hot rollers Misaligned roller Abnormal operation (Tcup 72  C) Abnormal operation (Tcup 80  C) Abnormal operation (Tcup 130  C)

Cup 1st raceway ( C) 50.3 58.7 60.6 86.2 71.9 59.8 90.4 96.6 107.9 81.8 72.0 80.9 130.8

Cup 2nd raceway ( C) 50.3 56.2 57.5 73.5 65.5 59.8 90.4 81.2 87.6 76.9 72.0 80.9 130.8

Cup average ( C) 50.2 57.5 59.0 79.2 68.6 59.7 90.2 88.5 96.8 80.0 71.9 80.7 130.5

Roller average ( C) 55.0 61.7 63.7 80.5 71.1 65.7 100.6 88.9 94.9 89.1 80.4 90.8 149.5

Roller max ( C) 55.0 120.8 75.9 218.2 83.3 110.7 291.8 232.0 233.4 388.6 80.4 90.8 149.5

motivation behind this simulation is to determine the roller temperature and heat rate associated with a bearing cup temperature of 90  C, which is still about 40  C below the hot-box alarm threshold assuming an ambient temperature of 25  C. The results indicate that, in order to produce a 90  C bearing cup temperature, the two hot rollers must generate a heat rate of 489 W if the remaining 44 rollers are assumed to be operating normally (producing 11.5 W each). The roller temperature associated with this heat rate is about 292  C, which is hot enough to produce distinct roller discoloration without triggering the HBDs. In the heating scenario three hot rollers (simulation 8 in Table 2), shown in Fig. 9, it is assumed that three adjacent rollers are caught misaligned while entering the loaded zone of the bearing, thus, heating abnormally to an elevated temperature of 232  C. The main goal of this simulation is to determine the bearing cup temperature associated with this hypothetical heating scenario. The results of this simulation demonstrate that the average bearing cup temperature is 88.5  C even though there are three hot rollers operating abnormally at an elevated temperature that can cause distinct discoloration in these rollers. Again, the 88.5  C bearing cup temperature is well below the HBD threshold for an ambient temperature of 25  C and, therefore, this bearing will most likely continue to operate abnormally while undetected by conventional wayside bearing health monitoring equipment. Simulations 9 and 10 in Table 2 are two other hypothetical heating scenarios that demonstrate how certain rollers can reach unsafe operating temperatures without heating the bearing cup anywhere close to the hot-box alarm threshold. Finally, simulation 13 in Table 2, shown in Fig. 10, provides an insight into the operating conditions that would lead to a bearing cup temperature of 130.5  C, which would trigger the HBD alarm. The results reveal that all 46 rollers within the bearing have to reach an operating temperature of 149.5  C, generating a total heat input of 2208 W, in order for the bearing cup to reach 130.5  C. This heating scenario demonstrates the extreme operating conditions that must occur before conventional wayside detectors tag that bearing for removal from service.

Conclusions
The purpose of the work presented in this paper is to demonstrate that it is possible to have certain rollers within the cone assemblies operating at unsafe elevated temperatures without heating the bearing cup to levels that would trigger the HBD alarm. A theoretical approach was sought because of instrumentation limitations associated with monitoring internal temperatures of a rotating bearing. To this end, a FE model of a railroad tapered-roller bearing pressed onto an axle was developed. The Journal of Thermal Science and Engineering Applications

boundary conditions used for the FE model were derived from previously conducted experimental and theoretical work. The model was validated experimentally and theoretically by comparing the axle temperature obtained from the FE results to that calculated from an analytical expression derived from theory developed in a previous study. The systematic comparison of the axle temperature values revealed that the results agreed to within 2%. Thirteen different heating scenarios were investigated in this study; ve of which were intended to duplicate previously performed dynamic bearing tests. The studied cases varied from normal operation to bearings having certain rollers misbehaving to bearings heating to levels that would trigger HBDs. In each case, the temperatures and heat generation rates within the bearing were determined for a specied external surface cup temperature. The ambient temperature used in all the simulations listed in Table 2 is 25  C (77  F). The FE model results revealed that rollers in a bearing operating normally are only 5  C hotter than the bearing cup temperature; however, abnormally operating rollers such as stuck or misaligned rollers can reach temperatures that are signicantly higher than the bearing cup temperature without heating the cup to levels that will trigger an alert. The latter is of concern because rollers operating at elevated temperatures will have adverse effects on the material properties of the bearing raceways, the cages, and also the grease condition. Considering the fact that most lubricants used in railroad bearings start to degrade when operated at temperatures above 125  C for prolonged periods, having a few rollers operate at temperatures at or above 232  C will most likely contribute to the accelerated deterioration of the grease, and in extreme conditions, can result in grease starvation and premature bearing failure. Since conventional wayside bearing monitoring equipment will only alert the railroad if the bearing cup temperature reaches 105.5  C (190  F) above ambient, it is likely that certain rollers will continue to operate at unsafe temperatures and go undetected until they cause enough damage to the internal components of the bearing that will raise the bearing cup temperature to alarm levels. At that time, however, it might be already too late to avoid catastrophic bearing failure. The aforementioned raises the question about the need for continuous bearing condition monitoring systems as opposed to conventional wayside detection equipment. In summary, a validated FE model was developed for a railroad tapered-roller bearing that can provide insight into the operating temperatures of the internal components of the bearing for a specied external bearing cup temperature, which has proven to be a very arduous task to accomplish experimentally. The usefulness of the devised FE model is demonstrated in Table 4, which provides relevant temperature results acquired from the simulations SEPTEMBER 2012, Vol. 4 / 031002-9

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

that can aid the railroad industry and researchers in future thermal analyses of railroad bearings.

Acknowledgment
The authors wish to thank Amsted Rails bearing division, BRENCO, Inc., for support of this work and permission to publish.

Nomenclature
A D Ho H1 h ho J k Lc Nu Pe Pr Q R Re T1 T2 Ta T t t V surface area, m2 diameter, m cup side overall heat transfer coefcient, W K1 axle side overall heat transfer coefcient, W K1 convection heat transfer coefcient per unit area, W m 2 K 1 cup side overall heat transfer coefcient per unit area, W m 2 K 1 Bessel function thermal conductivity, W m1 K1 characteristic length, m Nusselt number Peclect number Prandtl number heat input, W axle to cup side overall heat transfer coefcient ratio, H1/Ho Reynolds number average surface temperature of the cup,  C average temperature of the axle surface covered by one cone assembly,  C ambient temperature,  C normalized temperature time, s normalized time velocity, m s1

Greek Symbols
a m t r

thermal diffusivity, m2 s1 dynamic viscosity, Pa s kinematic viscosity, m2 s1 normalized thermal diffusivity

Subscripts
axle cup in roller total surface area of axle not covered by the bearing bearing cup input heat input to the roller(s) that are operating abnormally total heat input

Acronyms
AMG BC FE FEA HBD algebraic multigrid boundary condition nite element nite element analysis hot-box detector

References
[1] Karunakaran, S., and Snyder, T. W., 2007, Bearing Temperature Performance in Freight Cars, Proceedings of the Bearing Research Symposium Sponsored by the AAR Research Program in Conjunction With the ASME RTD 2007 Fall Technical Conference, Chicago, Illinois, Sept. 1112. [2] Tarawneh, C. M., Wilson, B. M., Cole, K. D., and Reed, M., 2008, A Metallurgical and Experimental Investigation Into Sources of Warm Bearing Trending, Proceedings of the 2008 IEEE/ASME Joint Rail Conference, Wilmington, DE, Apr. 2223, paper No. JRC200863028. [3] Tarawneh, C. M., Cole, K. D., Wilson, B. M., and Alnaimat, F., 2008, Experiments and Models for the Thermal Response of Railroad TaperedRoller Bearings, Int. J. Heat Mass Transfer, 51, pp. 57945803.

[4] Overstam, H., 2006, The Inuence of Bearing Geometry on the Residual Stress State in Cold Drawn Wire Analyzed by the FEM, J. Mater. Process. Technol., 171(3), pp. 446450. [5] Wei, Y., Qin, Y., Balendra, R., and Jiang, Q., 2004, FE Analysis of a Novel Roller Form: A Deep End-Cavity Roller for Roller-Type Bearings, J. Mater. Process. Technol., 145(2), pp. 233241. [6] Demirhan, N., and Kanber, B., 2008, Stress and Displacement Distributions on Cylindrical Roller Bearing Rings Using FEM, Mech. Based Des. Struct. Mach., 36(1), pp. 86102. [7] Vernersson, T., 2007, Temperatures at Railway Tread Braking. part 1: Modeling, Proc. Inst. Mech. Eng., F J. Rail Rapid Transit., 221(2), pp. 167182. [8] Schmidt, E., and Weiland, T., 2006, Application of a Computationally Efcient Air-Gap Element Within the Finite Element Analysis of Magnetic Bearings, IEEE Trans. Magn., 42(4), pp. 12631266. [9] Awasthi, R. K., Jain, S. C., and Sharma, S. C., 2006, Finite Element Analysis of Orice-Compensated Multiple Hole-Entry Worn Hybrid Journal Bearing, Finite Elem. Anal. Des., 42(14/15), pp.12911303. [10] Sukumaran Nair, V. P., and Prabhakaran Nair, K., 2004, Finite Element Analysis of Elastohydrodynamic Circular Journal Bearing With Micropolar Lubricants, Finite Elem. Anal. Des., 41(1), pp.7589. [11] Dunnuck, D. L., 1992, Steady-State Temperature and Stack-Up Force Distributions in a Railroad Roller Bearing Assembly, M.S. thesis, University of Illinois at Urbana-Champaign, Urbana, IL. [12] Wang, H., 1996, Axle Burn-Off and Stack-Up Force Analyses of a Railroad Roller Bearing Using the Finite Element Method, Ph.D. thesis, University of Illinois at Urbana-Champaign, Urbana, IL. [13] Wang, S., Cusano, C., and Conry, T. F., 1993, A Dynamic Model of the Torque and Heat Generation Rate in Tapered Roller Bearings Under Excessive Sliding Conditions, Tribol. Trans., 36(4), pp. 513524. [14] Kletzli, D. B., Cusano, C., and Conry, T. F., 1999, Thermally Induced Failures in Railroad Tapered Roller Bearings, Tribol. Trans., 42(4), pp. 824832. [15] Farneld, N. E., 1972, Thermal Investigations of Roller Bearings, Tribology, 5(3), p. 104. [16] Hoeprich, M. R., 1996, Rolling-Element Bearing Internal Temperatures, Tribol. Trans., 39(4), pp. 855858. [17] Wang, X. L., and Zhu, K. Q., 2006, Numerical Analysis of Journal Bearings Lubricated With Micropolar Fluids Including Thermal and Cavitating Effects, Tribol. Int., 39(3), pp. 227237. [18] Fillon, M., and Bouyer, J., 2004, Thermohydrodynamic Analysis of a Worn Plain Journal Bearing, Tribol. Int., 37, pp. 129136. [19] Wang, X. L., Zhu, K. Q., and Wen, S. Z., 2001, Thermohydrodynamic Analysis of Journal Bearings Lubricated With Couple Stress Fluids, Tribol. Int., 34, pp. 335343. [20] Li, X. K., Davies, A. R., and Phillips, T. N., 2000, A Transient Thermal Analysis for Dynamically Loaded Bearings, Comput. Fluids, 29, pp. 749790. [21] Ma, M. T., and Taylor, C. M., 1996, An Experimental Investigation of Thermal Effects in Circular and Elliptical Plain Journal Bearings, Tribol. Int., 29(1), pp. 1926. [22] Briot, J. M., Bourouga, B., and Bardon, J. P., 1997, Thermal Transport Conductance Between the Rings of a Roller Bearing, Rev. Gen. Therm., 36(8), pp. 610623. [23] Tarawneh, C. M., Cole, K. D., Wilson, B. M., and Freisen, K., 2007, A Lumped Capacitance Model for the Transient Heating of Railroad Tapered Roller Bearings, Proceeding of the 2007 ASEE-GSW Annual conference, March 2830, Paper No. T2C5. [24] Tarawneh, C. M., Fuentes, A. A., Wilson, B. M., Cole, K. D., and Navarro, L., 2009, Thermal Analysis of Railroad Bearings: Effect of Wheel Heating, Proceedings of the 2009 ASME Joint Rail Conference, Pueblo, CO, March 35, Paper No. JRC200963055. [25] Cole, K. D., Tarawneh, C. M., and Wilson, B. M., 2009, Analysis of FluxBase Fins for Estimation of Heat Transfer Coefcient, Int. J. Heat Mass Transfer, 52, pp. 9299. [26] Cole, K. D., Tarawneh, C. M., Fuentes, A. A., Wilson, B. M., and Navarro, L., 2010, Thermal Models of Railroad Wheels and Bearings, Int. J. Heat Mass Transfer, 53, pp. 16361645. [27] Tarawneh, C. M., Wilson, B. M., Cole, K. D., Fuentes, A. A., and Cardenas, J. M., 2008, Dynamic Bearing Testing Aimed at Identifying the Root Cause of Warm Bearing Temperature Trending, Proceedings of the 2008 ASME RTD Fall Technical Conference, Chicago, IL, Sept. 2426, Paper No. RTDF200874036. [28] Tarawneh, C. M., Kypuros, J. A., Wilson, B. M., Snyder, T. W., Gonzalez, B. A., and Fuentes, A. A., 2009, A Collaborative On-Track Field Test Conducted to Verify the Laboratory Findings on Bearing Temperature Trending, Proceedings of the 2009 ASME Joint Rail Conference, Pueblo, CO, March 35, Paper No. JRC200963056. [29] Tarawneh, C. M., Kypuros, J. A., Fuentes, A. A., Wilson, B. M., Gonzalez, B. A., Rodriguez, G., and Maldonado, R. K., 2009, Vibration Signatures of Temperature Trended Bearings in Field and Laboratory Testing, Proceedings of the 2009 ASME RTD Fall Technical Conference, Ft. Worth, TX, Oct. 2021, Paper No. RTDF2009-18038. [30] Bra ndlein, J., Eschmann, P., Hasbargen, L., and Weigand, K., 1999, Ball and Roller Bearings: Theory, Design and Application, 3rd ed., Wiley, New York.

031002-10 / Vol. 4, SEPTEMBER 2012

Transactions of the ASME

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms

[31] Harris, T. A. and Kotzalas, M. N., 2006, Advanced Concepts of Bearing Technology: Rolling Bearing Analysis, 5th ed., Taylor & Francis, London. [32] Alnaimat, F., 2007, Thermal Analysis of Railroad Tapered Roller Bearings, M.S. thesis, University of Texas-Pan American, Edinburg, TX. [33] Melosh, R. J., 1963, Structural Analysis of Solids, ASCE J. Struct. Div., 89(4), pp. 205223. [34] Incropera, F. P., DeWitt, D. P., Bergman, T. L., and Lavine, A. S., 2007, Fundamentals of Heat and Mass Transfer, 6th ed, Wiley, New York.

[35] Jones, J., Poulikakos, D., and Orozco, J., 1988, Mixed Convection From a Rotating Horizontal Heated Cylinder Placed in a Low-Speed Wind Tunnel, Int. J. Heat Fluid Flow, 9(2), pp. 165173. [36] Badr, H. M., and Dennis, S. C. R., 1985, Laminar Forced Convection From a Rotating Cylinder, Int. J. Heat Mass Transfer, 28(1), pp. 253264. [37] Kendoush, A. A., 1996, An Approximate Solution of the Convective Heat Transfer From an Isothermal Rotating Cylinder, Int. J. Heat Fluid Flow, 17(4), pp. 439441.

Journal of Thermal Science and Engineering Applications

SEPTEMBER 2012, Vol. 4 / 031002-11

Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 04/28/2013 Terms of Use: http://asme.org/terms