Effects of Independently Rolling Wheels on Flange Climb Derailment

The effects of independently rolling wheels (IRW) on flange climb derailment have been investigated through simulations using Transportation Technology Center, Inc. (TTCI)’s *NUCARSTM dynamic modeling software. Simulations of single wheelsets and hypothetcal light rail vehicles equipped with IRWs show that flange angle and flange length parameters play an important role in preventing derailments. That role is especially critical for independent rolling wheels due to their lack of self-steering capability. The speed contour concept was proposed for engineers to adopt the flange angle and flange length in a logical way for wheel profile design in new vehicles and wheel profile maintenance. It is also shown that the sensitivity of IRW to flange climb is also very dependent on particular vehicle designs.

A Simple Vibration Analysis for Running Boards and Other Attached Components

Vibration of attached components such as running boards, hand grabs, brake components, etc. has become a serious problem. This paper sets out a simple analysis method for ensuring the survival of these components. A simple mass spring model is used to develop a transfer function into the car body. The frequency components of a wheel flat and 39/33 foot jointed track are then established and the excitation amplitudes for components attached to the car body calculated. The response of these components at their natural frequency is then used to calculate their resulting stress levels. Simple methods for performing this analysis are described

Development of Crash Energy Management Designs for Existing Passenger Rail Vehicles

As part of the passenger equipment crashworthiness research, sponsored by the Federal Railroad Administration and supported by the Volpe Center, passenger coach and cab cars have been tested in inline collision conditions. The purpose of these tests was to establish baseline levels of crashworthiness performance for the conventional equipment and demonstrate the minimum achievable levels of enhancement using performance based alternatives. The alternative strategy pursued is the application of the crash energy management design philosophy. The goal is to provide a survivable volume where no intrusion occurs so that passengers can safely ride out the collision or derailment. In addition, lateral buckling and override modes of deformation are prevented from occurring. This behavior is contrasted with that observed from both full scale tests recently conducted and historical accidents where both lateral buckling and/or override occurs for conventionally designed equipment. A prototype crash energy management coach car design has been developed and successfully tested in two full-scale tests. The design showed significant improvements over the conventional equipment similarly tested. The prototype design had to meet several key requirements including: it had to fit within the same operational volume of a conventional car, it had to be retrofitted onto a previously used car, and it had to be able to absorb a prescribed amount of energy within a maximum allowable crush distance. To achieve the last requirement, the shape of the force crush characteristic had to have tiered force plateaus over prescribed crush distances to allow for crush to be passed back from one crush zone to another. The distribution of crush along the consist length allows for significantly higher controlled energy absorption which results in higher safe closing speeds.

A Study of Control Methods for the Braking System on Passenger Multiple Units

As well known, the mechanical (friction, pneumatic) brake system on trains contains some non-linear elements. So it has been difficult to control the speed or acceleration of trains according to desired patterns. This paper reviews our research on the control method of the physical performance of train running such as acceleration (deceleration) by mechanical braking devices. One of our approaches is the introduction of the feedback control into the brake control system. Mathematical models of non-linear elements in the brake system and some effective methods of controller design are proposed with both simulation and experimental results. Another approach is the real time estimation of the friction forces between a brake shoe and a wheel tread. Friction has severe non-linearity; however it can not be measured easily on running trains. We propose the introduction of the onboard real-time estimation method of friction coefficients using the speed information which can be obtained easily in the existing brake system.

Railcar Wheel Flat Impact Analysis

Wheel flats are common railcar wheel defects that cause bearing failures and train derailments. In an effort to better understand the impacts created by a wheel flat, a nonlinear finite element model using LS-DYNA was created of a single 914.4 mm (36 in) railcar wheel with a wheel flat and a 3.24 m (127.5 in) half track with one rail, six wood ties and tie plates. Discrete element springs and dampers simulated the ballast and subgrade. Goals of this model were to accurately simulate a wheel flat impact at varying speeds and to monitor the plastic deformation that occurs at the sharp edges of the wheel flat during rotation. Producing and validating this model then could be used to test modifications to the track and wheel to lower the severity of wheel impacts. Results indicated an accurate simulation; however, improvements with the material properties, suspension model and mesh sensitivity can be made.

Modeling Friction Wedges: Part II — An Improved Model

Friction wedges are a critical but imperfectly understood component in the typical North American three-piece freight car truck. A companion paper has reviewed the “state-of-the-art” in modeling this crucial suspension element. This paper proposes a new and considerably more complex wedge model. Among the features of the model are the ability to represent shear compliance of the wedge faces and varying normal and tangential pressure distributions on the faces, as well as explicit modeling of column toe-in and toe-out. The model is implemented using the VAMPIRE® vehicle dynamics package developed by AEA Technology plc. Sample results are presented for a standard metal wedge and a design with a resilient pad on the slope surface. Directions for further developments are proposed.

Modeling Friction Wedges: Part I — The State-of-the-Art

The friction wedge is a critical component in the three-piece truck. This paper describes the current approach for modeling friction wedges and compares its implementation in the commercially available NUCARS™ and VAMPIRE® vehicle dynamics codes. NUCARS™ is a software package developed by Transportation Technology Center, Inc., while VAMPIRE® is a package developed by AEA Technology plc. Sample results from both codes are presented based on standalone test cases. Shortcomings of the “state-of-the-art” model are described and directions for future work are proposed.

Infinite Beams as Models of Track Transient Vibrations

In the paper, an attempt to build solution of transient vibrations of railway track is presented. Excitations in this system generated by track structure, treated roughly as periodic ones, are often disturbed by track transient imperfections. Obtained solutions are based on Ritz-Galerkin method and then on Hermite’s polynomials application to the problem. Such a model doesn’t require a lot of computing and can be easily adapted e.g. for nonlinear track model. Our approach can take continuous foundation of Winkler or Pasternak layer type, or discrete rail support. For discrete not equi-spaced supports only finite number of them is considered. As the starting point for that solution, the response to Dirac’s impulse is employed. The method described in the paper is relatively efficient (beyond the resonances and near them) to analyse of infinite beams resting on continuous or discrete foundation, even with nonlinear characteristics.

Finite Element Estimation of the Residual Stresses in Roller-Straightened Rail

The purpose of this paper is to develop models to accurately predict the residual stresses due to the roller straightening of railroad rails. Several aspects of residual stress creation in rail due to roller straightening are addressed. The effect of the characteristics of the loads applied by the roller-straightener on the stress profile is examined. In addition, the analysis attempts to discern the relative influence of bending and contact on the residual stresses. The last goal is to determine how the heat treatment of rail alters the predicted roller-straightening residual stress field. The loads for the simulation are estimated from available data. To identify the most credible values, a baseline loading case is defined and modeled. These straightening loads are parameterized by considering alternative loading scenarios. Residual stresses and deformations are calculated using these loads. To separate the effects of bending and contact on the residual stress induced by the roller loads, each credible load case is analyzed with two models. One is a 2-dimensional generalized plane strain (GPS) model that accounts only for the flexural stresses. The other is a fully 3-dimensional analysis that includes roll-on-rail contact to make estimates of the true residual stress field. Comparison of the residual stress results from both models reveals the relative influence of local roll-rail contact and bending on the final profile. Comparison of the 2- and 3-dimensional residual stress results reveals that the magnitude of the contact loads is a decisive influence on the stress field, even in portions of the rail web located far from the contact interface. Therefore, it is critical to obtain accurate estimates of the straightening loads to make accurate roller straightening residual stress estimates. Heat treatment of the rail prior to roller straightening primarily affects the longitudinal residual stress in the web, causing a positive shift in the stress values.

Accurate Rail Vehicle Dynamic Simulations by DynaRail

A general purpose rail vehicle dynamic simulation tool has been developed at CAM (Center for Automated Mechanics). This tool allows the user to start any rail vehicle dynamic simulation with measured wheel and rail profile data without a need for generating pre-cooked wheel/rail contact geometry tables. Using the constrained multibody methods, the tool robustly solves the three dimensional wheel/rail interaction problem and computes the longitudinal, lateral and vertical locations of 1st and 2nd contacts, the sizes of all Hertzian contact areas, rolling radii, contact angles, normal reactions, the creepages and their associated forces and moments online and makes all the details available as outputs to the user for subsequent analysis and validation of the solution. The purpose of this study is to demonstrate the ability of the tool for predicting the detailed responses of the rail vehicle systems. For all simulations, in addition to the detailed contact reactions, the precise positions of all contacts in three-dimensional space are readily solved for and available for further considerations. This is an application paper and intends to introduce the railroad engineers and consultants to a tool that is capable of fulfilling their simulation and modeling needs for complex rail/vehicle systems. Such needs are not commonly provided by the specialized codes and very hard for the railroad engineers to fulfill them using other general purpose codes. DynaRail is particularly useful to the wheel/rail profile designers who may be concerned with the continuous movement of the contact point at an unprecedented accuracy. The online prediction and imposition of the contacts makes it possible to systematically include and consider the effect of the contact movements on its associated contact geometry variables such as the size of contact ellipses, rolling radii, contact angles and, ultimately the effect of all these on the stability and/or the resulting motion of wheel on rail.