Ulf Sellgren

A finite element-based model of normal contact between rough surfaces

Engineering surfaces can be characterized as more or less randomly rough. Contact between engineering surfaces is thus discontinuous and the real area of contact is a small fraction of the nominal contact area. The stiffness of a rough surface layer thus influences the contact state as well as the behavior of the surrounding system. A contact model that takes the properties of engineering surfaces into account has been developed and implemented using finite element software. The results obtained with the model have been verified by comparison with results from an independent numerical method. The results show that the height distribution of the topography has a significant influence on the contact stiffness but that the curvature of the roughness is of minor importance. The contact model that was developed for determining the apparent contact area and the distribution of the mean contact pressure could thus be based on a limited set of height parameters that describe the surface topography. By operating on the calculated apparent pressure distribution with a transformation function that is based on both height and curvature parameters, the real contact area can be estimated when the apparent contact state is known. The model presented is also valid for cases with local plastic flow in the bulk material.

Particle emissions from rail traffic : a literature review

Particle emissions are a drawback of rail transport. This work is a comprehensive presentation of recent research into particle emissions from rail vehicles. Both exhaust and nonexhaust particle emissions are considered when examining particle characteristics such as PM10, and PM2.5 concentration levels, size, morphology, composition, and adverse health effects, current legislation, and available and proposed solutions for reducing such emissions. High concentration levels in enclosed rail traffic environments are reported and some toxic effects of the particles. The authors find that only a few limited studies have examined the adverse health effects of nonexhaust particle emissions and that no relevant legislation exists. Thus further research in this area is warranted.

Application of a constitutive model for micro-slip in finite element analysis

Micro-slip is a phenomenon that occurs between contacting surfaces when a frictional load, less than that necessary to produce macro-slip, is applied to the contacting surfaces. Micro-slip is an el ...

Using finite element analysis to predict the brake pressure needed for effective rotor cleaning in disc brakes

The general trend toward increased use of computer models and simulations during product development calls for accurate and reliable product models. The function of many products relies on contact interfaces between interacting components. Simulating the behavior of such products requires accurate models of both components and interfaces. Depending on the purpose of the simulation, interface models of different degrees of detail are needed. In simulating very large systems with many interfaces, it might be computationally expensive to integrate detailed models of each individual interface. Condensed models, or abstractions, that describe the interface properties with the fewest degrees of freedom are therefore required. This thesis deals with the modeling and simulation of mechanical interfaces in a systems context. The five appended papers discuss the issue from both the simulation and tribological points of view. The aim is to study how friction and wear can be modeled in the behavioral simulation of technical systems and to discuss the convenience and applicability of using different types of models as building blocks of a system model in simulations. Paper A reviews existing friction models of sliding contacts under different running conditions. Paper B uses a simplified contact model, the elastic foundation model, to model friction in a boundary-lubricated rolling and sliding contact. The model is integrated into a dynamic rigid body model of a mechanical system, and the system behavior is simulated. Paper C discusses the application of the elastic foundation model to rough surface contact problems and investigates how the error in its results depends on surface roughness. Papers D and E address how the wear of the contact surfaces at the pad-to-rotor interface in a passenger car disc brake can be simulated using finite element analysis (FEA).

Behavior modeling in mechanical engineering — A modular approach

Many engineering activities are confronted with the relation between shape and behavior. Modern ‘state-of-the-art’ CAD systems can support a dynamic design process with flexible and umambiguous geometric modeling of artifacts. The Finite Element (FE) method is a general method to model and simulate physical behavior. CAD and FE integration enables numerical prediction of the physical behavior of artifacts. To deal with the complexity of modern products and the dynamic character of the design process, the integration must be addressed from both a systems and a process point of view. The models must thus be flexible, scalable, and reusable. A modularized modeling approach based on four stages is proposed, and exemplified with behavior modeling of a turn-key grinding machine.

A stiffness modeling methodology for simulation-driven design of haptic devices

Efficient development and engineering of high performing interactive devices, such as haptic robots for surgical training benefits from model-based and simulation-driven design. The complexity of the design space and the multi-domain and multi-physics character of the behavior of such a product ask for a systematic methodology for creating and validating compact and computationally efficient simulation models to be used in the design process. Modeling the quasi-static stiffness is an important first step before optimizing the mechanical structure, engineering the control system, and performing hardware in the loop tests. The stiffness depends not only on the stiffness of the links, but also on the contact stiffness in each joint. A fine-granular Finite element method (FEM) model, which is the most straightforward approach, cannot, due to the model size and simulation complexity, efficiently be used to address such tasks. In this work, a new methodology for creating an analytical and compact model of the quasi-static stiffness of a haptic device is proposed, which considers the stiffness of actuation systems, flexible links and passive joints. For the modeling of passive joints, a hertzian contact model is introduced for both spherical and universal joints, and a simply supported beam model for universal joints. The validation process is presented as a systematic guideline to evaluate the stiffness parameters both using parametric FEM modeling and physical experiments. Preloading has been used to consider the clearances and possible assembling errors during manufacturing. A modified JP Merlet kinematic structure is used to exemplify the modeling and validation methodology.

An Optimization Approach Toward a Robust Design of Six Degrees of Freedom Haptic Devices

This work presents an optimization approach for the robust design of six degrees of freedom (DOF) haptic devices. Our objective is to find the optimal values for a set of design parameters that maximize the kinematic, dynamic, and kinetostatic performances of a 6-DOF haptic device while minimizing its sensitivity to variations in manufacturing tolerances. Because performance indices differ in magnitude, the formulation of an objective function for multicriteria performance requirements is complex. A new approach based on Monte Carlo simulation (MCS) was used to find the extreme values (minimum and maximum) of the performance indices to enable normalization of these indices. The optimization approach presented here is formulated as a methodology in which a hybrid design-optimization approach, combining genetic algorithm (GA) and MCS, is first used. This new approach can find the numerical values of the design parameters that are both optimal and robust (i.e., less sensitive to variation and thus to uncertainties in the design parameters). In the following step, with design optimization, a set of optimum tolerances is determined that minimizes manufacturing cost and also satisfies the allowed variations in the performance indices. The presented approach can thus enable the designer to evaluate trade-offs between allowed performance variations and tolerances cost.

Robust pre-synchronization in heavy truck transmissions

Transmission performance directly influences vehicle behaviour, energy efficiency, and perceived quality. A modern transmission is an aggregate of interacting sub-systems, e. g. gears, bearings, sy ...

Fast hybrid genetic clustering algorithm for Design Structure Matrix

Module formation is the step in which a products architecture is established in such a way that complex interactions are intra-modular and inter-modular interactions are more simple. If a matrix r ...

A Verified and Validated Model for Simulation-Driven Design of Heavy Duty Truck Synchronizers

The strong market trend toward lower fuel consumption for heavy road transports requires more frequent gear shifting and increased gear shift performance, i.e. shorter shift time. Increased shift p ...