While the support bearings are key elements of rotating machinery systems, the overall system performance and design depend on a close integration of several disciplines. For prescribed life and operating environment the applied static loads and speeds on each of the system component may be generally determined by an equilibrium analysis. Conventional rotor dynamics models may be used to model overall system dynamics, rotor response and dynamic loads imposed on the support bearings. As a function of these applied conditions bearing response and dynamic performance is determined by integration of the equations of motion of each bearing element; in the case of rolling bearings, the available bearing dynamics models, such as ADORE (Advanced Dynamics Of Rolling Elements) provide integration of the classical differential equations of motion to model real-time performance of the bearing. With prescribed bearing geometry and applied operating conditions, it is well established that lubricant properties and mechanics play a major role in determining the stability of bearing elements and overall system performance. The rolling-sliding interactions in the concentrated contacts between the bearing elements produce heat, which travels through the bearing and the overall system. This affects temperature of the bearing elements, which in turn changes the bearing geometry and material behavior including the lubricant. Thus overall system design and performance simulation requires a close iteration between the various models at varying levels of sophistication.

A numerical three-dimensional contact model is presented to investigate the contact behavior of multilayered elastic-perfectly plastic solids with rough surfaces. The surface displacement and contact pressure distributions are obtained based on the variational principle with fast Fourier transform (FFT)-based scheme. Von Mises yield criterion is used to determine the onset of yield. The effective hardness is modeled and plays role when the local displacement meet the maximum displacement criterion. Simulations are performed to obtain the contact pressures, fractional total contact area, fractional plastic contact area, and surface/subsurface stresses. These contact statistics are analyzed to study the effects of the layer-to-substrate ratios of stiffness and hardness, surface roughness, and layers thickness of rough, two-layered elastic/plastic solids. The results yield insight into the effects of stiffness and hardness of layers and substrates, surface roughness, and applied load on the contact performance. The layer parameters leading to low friction, stiction, and wear are investigated and identified.

The tribological and electrical conduction behaviors of copper under dry sliding against copper-chromium alloys were investigated on the basis of simulation tests. The experimental results show that electrical current and tribological parameters have combined effects on both electrical conduction ability and tribological properties. High sliding velocity and electrical current density deteriorate both the tribological and electrical conduction properties in tribo-electrical sliding system. In the end, the worn surface analysis suggests that friction, resistance heat and electrical arc spoiling are the main factors to influence the tribological and electrical conduction properties.

Scuffing damage was found on Space Shuttle Power Drive Unit gears for the Rudder Speed Brake. Scuffing simulation tests rule out a predicted scuffing scenario. Scuff initiation is caused by tooth interference under unique transient high speed and torque conditions. Simulation tests for subsequent operation following a scuffing event show limited wear much less than the thickness of the gear case hardened layer.

The topography of the sheet metal and tool surface influences the contact behavior between the tool and the sheet metal. With a simulation on the micrometer scale the topographical deformation of the sheet metal is elasto-plastically modeled using an explicit finite element method. The model of the lubricant in between is based on a simplified method calculating the pressure and velocity of the liquid with Darcy’s law. A flat rigid tool is pressed on the sheet metal surface simultaneously with a constant lateral motion. Simulations were performed with real and modeled surfaces. The topographies show different tribological behavior depending on their texture.

The need for rapid development of high efficiency jet engines and other mechanical systems has forced the pace of development of simulation tools so that the detailed behavior of the rolling element bearings can be simulated. This includes modeling of the motions, the exchanged forces, and heat between all the bearing components, i.e. rings, cage and rolling elements. The computer program BEAST is a multi-body computer program with sophisticated and robust modeling of the tribological contacts in the bearing where the exchange of forces between the bearing components takes place.

The production of nanocrystalline wear debris containing components from the worn specimen, from the counterface and from the environment does not support any of the better known wear models or wear equations based on adhesion, delamination, fatigue or oxidation. In this presentation, plastic deformation, mechanical mixing and patterns of flow determined from experiments will be compared with molecular dynamics (MD) simulations and a continuum analysis of two ‘fluids’ shearing in opposite directions. Together, these suggest generic behavior that needs to be included in any realistic sliding wear model.

Many tribological problems are expensive to solve due to the requirement of heavy and iterative computation. In this study, a direct mapping between design variables and merits is obtained using a neural network. Data from limited numerical simulations of the behaviors of fluid film of a slider in the thermohydrodynamic regime were used to train the network, which replaces further lengthy simulations. A balance between the modeling accuracy and generalization capability of the network is achieved by optimizing the network size using an efficient optimization scheme, DIviding RECTangles (DIRECT). Performance comparison of the optimization method is based on a hybrid learning strategy that combines the steepest descent and Levenberg-Marquardt method.

The stress field within machine components is an important indicator for contact failures. Since both thermal stresses due to frictional heating and plasticity are significant in engineering application, it is critical to predict the total stress field. In this work, the steady- state thermal effect is considered and a thermo-elastic-plastic contact model is developed. The model is applicable for rolling and/or sliding contact problem, as far as small equivalent plastic strain hypothesis is respected. Influence coefficients for surface normal displacement, temperature, and strain and stress tensors are used with the discrete convolution and fast Fourier transform algorithm. The single-loop conjugate gradient iteration scheme is also applied to achieve fast convergence speed. Simulations are presented for several academic examples ranging from elastic to thermoelastic-plastic.

The topography of a machined surface significantly influences the performance of both lubricated and unlubricated contacts. In this work the unlubricated frictionless contact of rough surfaces is the subject of study. The numerical deterministic contact mechanics model incorporates linear elastic surface material and the dry contact is simulated using classes of surface profiles randomly generated from a given Abbott curve (Abbott-Firestone bearing ratio curve). In this way all the height information of the surface profile is preserved and not only a few parameters, like R a , R q , R z , R sk , etc. The aim of this work is to investigate how classes of surfaces based on a single Abbott curve perform in terms of contact mechanical parameters like the real area of contact. The result shows that surfaces taken from a class of random surfaces generated from a specific Abbott curve behaves similar in a contact mechanics simulation. That is, the distribution of for example the real area of contact within such a class is compact, having a small deviation from its mean. This implies that it is possible to simulate classes of surfaces based on Abbott curves and to use the results to predict contact mechanical properties of real surface topographies.

This work was initially focused to analyze the onset of systematic fractures of car body parts (Ford Focus model) during stamping. Fractures were attributed to unexpected high values of friction between the tools and the steel sheets. All the sheets that failed fulfilled the required standard mechanical properties and topographical surface requirements (Ra between 1.0 and 1.7 μm and peak count Pc >50 peaks/cm, according to the German SEP 1940 standard), and were processed with the same lubricant. A detailed study of the surface features showed that, having similar and correct values of the conventionally used parameters for surface texture requirements, such as Ra and peak count Pc, those sheets that failed had some different asperity features, related to the asperity peak shape. Experimental work (measurement of real values of surface contact Sc and friction behavior under boundary lubrication conditions) confirmed that these differences in the asperity shape have an important effect on the friction behavior, controlling the rate of surface roughness flattening and therefore determining the real values of surface contact area and the remaining valley areas able to carry lubricant. The differences in asperity shape were reproduced in single roughness models and the flattening process of the asperities was simulated by finite element analysis. The agreement between simulation and experimental observation was excellent. The influence of asperity shape was found to be very important, justifying the real differences found in the values of real surface in contact and hence in the friction behavior during stamping real car body parts.

Statistical and fractal approaches for characterizing surface topography have been used widely in contact mechanics. In the present study, an attempt is made to compare between contact mechanics results obtained with statistical and fractal approaches to characterize surface topography. Specifically, a three-dimensional fractal surface was generated and statistical surface parameters were extracted using different sampling resolutions. Contact mechanics simulations were performed using the simulated fractal surface and statistical surfaces represented by the extracted statistical surface parameters. Purely elastic contact is studied in order to eliminate the influence of the individual asperity mechanical response on the obtained results. Therefore, differences in the simulated contact area and load can be related solely to the different approach employed for surface characterization. The influence of surface characterization approaches on the contact parameters and the resulting predictions is discussed in the contents of the presented results.

The equations of motion for a thin strip of flexible material are used to describe the in-plane transverse motion of an axially moving magnetic tape in a tape transport. The relative contribution of the tension term compared to the bending stiffness term is analyzed for typical tension values and the frequency range of interest. The bending stiffness term is found to be 10–20 times larger than the tension term in the frequency range of interest in typical tape drives. The full equation of motion for the displacement of a single span of tape between two reels is solved numerically subject to boundary and initial conditions and the result is compared to experimental measurements. The agreement between measurement and numerical simulation is very good.

The area of molecular contact between surfaces plays a central role in friction and adhesion. Traditionally it is calculated using continuum contact mechanics, which is known to break down as the contact width approaches atomic dimensions. Yet contact mechanics is being applied at ever smaller lengths, even to atomic force microscope (AFM) tips containing a few atoms. Molecular simulations are used to test the limits of contact mechanics under ideal non-adhesive and adhesive conditions. A simple geometry of sphere-on-flat is considered. One surface is an atomically flat crystal and is deformable. The approximately spherical curved surface is rigid and is produced by bending a crystal, or cutting a crystal or amorphous material. While these methods of producing the curved surface differ only in the amount of atomic-scale roughness, they produce very different behavior in some quantities. The normal stiffness is affected very little, the contact area can be modified by a factor of two, and the friction and lateral stiffness can be modified by an order of magnitude. The implications for AFM experiments are discussed.

A finite element model of the path of 1/2 inch magnetic tape media as found in a Small Format Tape Evaluation System (SFTES) was created and validated. The model was used to predict tape geometry within the system, specifically in the neighborhood of an inclined-wall vacuum column and an inclined capstan. Parametric study via simulation and testing was used to improve the quality of tape handling within the SFTES.

Particles are often presented at contact interfaces. In this study, a three-body microcontact model Considering of hard particles and rough surfaces is proposed in order to understand the effects of particles between surfaces on contact characteristics. Both transitional surface-to-surface and particle-to-surface two-body microcontact simulations can be obtained according to the simplification of this model. In the three-body contact situation, the curves of contact area ratio versus dimensionless load are located in the range between two straight lines. The surface-to-surface two-body contact situation is the upper bound and the particle-to-surface two-body contact situation is the lower bound. The contact situation will approach to the pure particle-to-surface two-body contact situation as the value of particle diameter over roughness value of surface increases. The total contact area ratio increases when particle size or particles density decreases, and contact load increases. Contrary to the results for contact area ratio, the dimensionless separation increases when particle density increases or contact load decreases.

The determination of the mechanical properties using the nano or macroindentation test results was developed in order to predict the mechanical properties of materials such as thin or thick coatings [1]. In the present paper a new method using a continuous FEM supported simulation of the microindentation results through a special tester, is proposed in order to predict the stress–strain curve of various ferrous or non ferrous materials that are being used as constructive materials in several machines or devices. The elasticity modulus is determined using a large spherical indenter tip and the stress strain curve using smaller steel balls. In the case of three deferent ball diameter indenter tips, the adopted experimental-analytical methodology present similar results regarding the constitutive law of the examined steel or aluminum specimens.

This paper describes two approaches for the simultaneous identification of the coulomb and viscous parameters in kinematical joints. One is a time-domain method (TDM) and the other is a frequency-domain method (FDM). Simulation shows that both of the two methods have good performances in identifying friction at high SNR (90dB). But at low SNR (20dB), the estimation accuracy of the frequency-domain method is higher than that of the time-domain method. A field experiment employing a linkage mechanism driven by motor is also carried out. The experimental results obtained by the two approaches are almost identical under different experiment conditions. It has been concluded that the presented identification methods of friction in kinematical joints are correct and applicable.

An effective simulation technique for describing the spreading properties of molecularly thin lubricant films on magnetic disks has been developed. We propose a molecular precipitation method that can simulate initial molecule arrangement of the films dip-coated onto the disks. Reptation and Rouse models as the model of the molecular motion, and molecular insertion and molecular precipitation methods as the method for putting molecules in initial positions were compared. From the results of the spreading profiles and diffusion coefficients, it has been revealed that the molecular precipitation method combined with the Rouse model is effective in simulating the spreading of the lubricant films.

Liquid lubricants break down at extreme temperatures and promote stiction in micro/nano-scale environments. Consequently, using flows of solid granular particles as a “dry” lubrication mechanism in sliding contacts was proposed, because of their ability to carry loads and accommodate surface velocities. Granular flows are highly complex flows that in many ways act similar to fluids, yet are difficult to predict because they are not well understood. Granular flows are composed of discrete particles which display fluid and solid lubricant behavior with time. This work describes the usefulness of employing lattice-based cellular automata (CA) as a tool for modeling granular flows in tribological contacts. The granular kinetic lubrication (GKL) continuum modeling approach has been successful at predicting trends gleaned from experiments conducted with granules in a couette shear cell. These results are used as a benchmark for determining the effectiveness of the CA modeling results. While the CA model was constructed entirely from rule-based mathematics, velocity and solid fraction results from the simulations were in good agreement with those from the GKL model benchmark results.