Whilst extensive work has been done on fault detection in bearings using sound, very little has been accomplished with other machine components and machinery partly due to the scarcity of datasets. The recent release of the Malfunctioning Industrial Machine Investigation and Inspection (MIMII) dataset opens the opportunity for research into malfunctioning machines like pumps, fans, slide rails, and valves. In this paper, we compare common features from audio recordings to investigate which best support the classification of malfunctioning pumps. We evaluate our results using the Area Under the Curve (AUC) as a performance metric and determine that the log mel spectrum is a very useful feature, at least for this dataset, but that other features can enhance detection performance when ambient noise is present (improving AUC from 0.88 to 0.94 in one case). Also, we find that mel Frequency Cepstral Coefficients (MFCC) perform substantially poorer as features than a sampled mel spectrogram.

Heavy (Class 8) truck fuel storage location and geometry has not significantly changed in several decades. Manufacturers have taken steps to improve their designs by eliminating cross over lines and making material property and thickness changes, among other changes, but there has been no mandate or significant effort to decrease the potential for post collision fuel fed fires in heavy trucks. Even with these design changes, FARS data indicates the number of fatal post-impact fires has not decreased over time. Several studies were conducted in the 1980’s and 1990’s that brought the unprotected design of the fuel storage on these vehicles to light. This paper combines these historical works with current FARS data on the subject and describes a different design approach that increases the impact protection of the fuel storage tank. This new approach uses the truck’s frame rails to guard the fuel storage tank and absorb and redirect impact energy. Currently, a heavy truck “saddle” mounted fuel tank’s integrity is tested through a 30 foot drop test prescribed by 49 CFR 393 and also listed in SAE Recommended Procedure J703. In this work, a crash test methodology used to test the integrity of a school bus side mounted fuel tank as prescribed in FMVSS 301S is discussed. Results of using this crash methodology on a current “saddle” tank design and a prototype of the new fuel storage system design are also presented.

During the hot rolling of carbon steel, austenite phase transforms into a pearlitic morphology, which essentially is a matrix of ferrite lamellae (α-Fe) and cementite (Fe 3 C). This transformation occurs at the cooling bed after an equalisation temperature of around 600 °C. Pearlitic steels find their use in ropes for bridges and elevators, rails, and tyre cords among others. Characterisation of microstructure has not been broadly applied to pearlitic steels because of their complex microstructures. Therefore, the characterisation of this morphology becomes inevitable, in order to identify potential weaknesses in the matrix. In this study, hot-rolled reinforcement bars (rebars) produced from recycled steel and direct reduced iron (DRI), were used for microstructural examination using standard metallurgical procedures. Although the optical microscope (OM) and scanning electron microscope (SEM) were used to obtain qualitative microstructure, they could not characterise the pearlite morphology quantitatively because of their three-dimensional (3D) limitation. Hence, the image analyser - Gwyddion Software, was used to quantify the pearlite morphology of these Y16 rebars. The results indicate that the pearlite colony is characterised by 3D single interpenetrating crystals of ferrite and cementite running parallel to each other due to their common growth during the transformation process of austenite. It was further observed that, the dimensional properties of the phases in the morphology in terms of their width and Interlamellar spacing ( S ), including the roughness of the pearlite colony can vary significantly. These results could be used to enhance the processing methodology of the industrial production processes.

High quality products are primarily dependent on the accurate translation motion of each linear sliding guide rail (LSGR) of CNC machine tools in the advanced manufacturing industry. The existing researches about LSGR precision pay much more attentions on the components dimensional variation and deformations, while micro-geometric form errors (such as surface roughness) have been ignored in most previous studies. Therefore, the true contact interface property is so crucial that it will affect macro-mechanical performance, such as friction, wear, assembly relation, tightness, fatigue strength, etc. In order to obtain the contact behaviors of LSGR accurately, a novel finite element model contact analysis is proposed in this paper. Instead of adopting the conventional statistic characterization method, such as Greenwood-Williamson (GW) model, the rough slider surface generated by the fractal function get closer to the real surface topography. In this study, the true contact area, contact pressure, contact stress and deformation are all investigated. Furthermore, the contact properties results of the present model are in a good agreement with other analytical solutions. In conclusion, the proposed finite element model combining with the fractal theory may provide an accurate contact analysis for LSGR. It is also great of guiding significance for the prediction of assembly quality and operation performance in high-precision measurement region as well as other precise engineering applications.

The study established a three-dimensional, thermal-kinetic-mechanical finite element (FE) model to simulate an additive manufacturing process with a laser powder deposition (LPD) approach for repairing a 75-lb rail that is broadly used for light rail transportation in the US. A worn rail specimen is repaired using 304L stainless steel powders as the deposition material for lab tests. The researchers incorporated an element-birth-and-kill technique to activate the deposition elements step-by-step, according to the build-up strategy along which the laser heat source is translated simultaneously. The laser power attenuation and solid-state transformation expressions are described using external user-defined subroutines for thermal and kinetic analysis, respectively. A set of equations for calculation of hardness for both of the rail and deposition materials are also defined and developed in the FE model. The microstructure distribution coming from kinetic analysis output is employed for hardness calculation. The estimation of the width and depth of the dilution zone in thermal analysis is compared with the experimental results of the repaired specimen to validate the thermal model. Scanning Electron Microscope (SEM) and Optical Microscope (OM) analyses, along with a Rockwell B-scale hardness test are performed to validate the outgoing microstructure and hardness results of the FE model. Mechanical analysis results showed that residual thermal stresses can significantly reduce the safety margin against shearing off the deposition part under a dynamic train load. The validated model demonstrated great potential for investigating the effects of the variation of different LPD process parameters on the final mechanical and metallurgical properties of the repaired rail.

Sit-to-stand-walk (STW) is a complex task that sequentially transitions an individual from sitting through standing to walking. In this study we evaluate the unrestricted, natural pattern of movement of the STW task from a hospital bed of 21 (5 Female, 16 Male) frail (MFS > 55) adults (68.0±11.2 years) with a total of 144 unique trials. Bed height (low, medium, high) and bed rail condition (no rails, Hill-Rom ® , Stryker ® ), were varied, generating 9 potential trial types per participant. A new STW phase, Stand Preparation, is defined specifically for the frail that occurs just prior to the Flexion Momentum Phase, also named here as the Stand Initiation Phase. In conjunction with the newly defined Stand Preparation Phase, movements used by the frail to maintain or regain balance during STW task are newly defined as corrective behaviors (CBs). These include hand, foot, leg and torso CBs. In 144 unique STW trials, 678 hand and foot CBs were observed and recorded. The most frequent CB type was the hand CB (335), followed by the foot CB (316). A coding system for use in the kinematic analysis of the natural STW task was developed that identifies CBs through visual observation. In addition, a 3D biomechanical model was generated from collected marker position data and will be used in future biomechanical analyses with the visually observed CB data. The Stand Initiation Phase contained the most CBs. Significant factors included bed height and phase, as well as their interaction (all with p-values ≤ 0.006). This is the first study to establish a more accurate and complete STW of the frail elderly, as well as to define CBs employed during their natural STW. The dataset from this coding system, along with the newly established STW phases of the frail, are currently being used for further analyses to determine the exact timing and position of fall initiations during STW of the frail.

This study presents the design and development of a novel, low-cost load-deflection test setup providing the testing of flexible links and compliant mechanisms. Test bench consists of two stepper motors, lead screw, rail system, two carts, two clamps, bearings and a force sensor. Clamps are designed in a way to attach various types of compliant members such as pinned-pinned buckling beam, fixed-fixed beam and 3D printed links. Mechanism enables to calculate the stiffness of compliant and 3D printed flexible systems. Sliders are displaced quasi-statically to slowly stretch or compress the flexible members attached in between two clamps. Displacement of the carts and deflection of the midpoint of the buckling beams are captured using machine vision measurement. Force applied from one of the carts to the end of the attached link is recorded using the force sensor. Stiffness of 3D printed flexible translational vibratory mechanisms is obtained using the displacement of the carts and load deflection curve of buckling beams are obtained using deflection curve and load data. Experimental results are compared with the same simulations performed by FEA analysis.

The U.S. Department of Transportation (DOT) Federal Railroad Administration (FRA) began promulgating regulations for the structural crashworthiness of passenger rail equipment at 49 Code of Federal Regulations (CFR) Part 238 on May 12, 1999. These Passenger Equipment Safety Standards (PESS) [1] include requirements affecting the designs of sidewall structures on passenger rail equipment. The FRA’s Office of Research, Development and Technology and the DOT’s Volpe National Transportation Systems Center are conducting research to evaluate the side impact strength of Tier I passenger rail equipment designs that have been constructed according to the current side structure regulations in §238.215 and §238.217. Following a fatal 2011 accident in which a highway semitrailer truck impacted the side of a passenger train that was transiting a grade crossing in Miriam, NV, the National Transportation Safety Board (NTSB) recommended that the FRA “develop side impact crashworthiness standards (including performance validation) for passenger railcars that provide a measurable improvement compared to the current regulation for minimizing encroachment to and loss of railcar occupant survival space” [2]. This paper describes the status of the current FRA research related to side structure integrity and describes the planned next stage of the research program which will include analyzing the performance of generalized passenger railcar structures in side impact collision scenarios. A discussion of the technical challenges associated with analyzing side impacts on passenger rail equipment is also presented.

The Hybrid-III Rail Safety (H3-RS) anthropomorphic test device (ATD), also known as a crash test dummy, was developed by the Rail Safety and Standards Board (RSSB), DeltaRail (now Resonate Group Ltd.), and the Transport Research Laboratory (TRL) in the United Kingdom between 2002 and 2005 for passenger rail safety applications [1]. The H3-RS is a modification of the standard Hybrid-III 50th percentile male (H3-50M) ATD with additional features in the chest and abdomen to increase its biofidelity and eight sensors to measure deflection. The H3-RS features bilateral (left and right) deflection sensors in the upper and lower chest and in the upper and lower abdomen; whereas, the standard H3-50M only features a single unilateral (center) deflection sensor in the chest with no deflection sensors located in the abdomen. Additional H3-RS research was performed by the Volpe National Transportation Systems Center (Volpe Center) under the direction of the U.S. Department of Transportation, Federal Railroad Administration (FRA) Office of Research, Development, and Technology. The Volpe Center contracted with TRL to conduct a series of dynamic pendulum impact tests [2]. The goal of testing the abdomen response of the H3-RS ATD was to develop data to refine an abdomen design that produces biofidelic and repeatable results under various impact conditions with respect to impactor geometry, vertical impact height, and velocity. In this study, the abdominal response of the H3-RS finite element (FE) model that TRL developed is validated using the results from pendulum impact tests [2]. Results from the pendulum impact tests and corresponding H3-RS FE simulations are compared using the longitudinal relative deflection measurements from the internal sensors in the chest and abdomen as well as the longitudinal accelerometer readings from the impactor. The abdominal response of the H3-RS FE model correlated well with the physical ATD as the impactor geometry, vertical impact height, and velocity were changed. There were limitations with lumbar positioning of the H3-RS FE model as well as the material definition for the relaxation rate of the foam in the abdomen that can be improved in future work. The main goal of validating the abdominal response of the dummy model is to enable its use in assessing injury potential in dynamic sled testing of crashworthy workstation tables, the results of which are presented in a companion paper [3]. The authors used the model of the H3-RS ATD to study the 8G sled test specified in the American Public Transportation Association (APTA) workstation table safety standard [4]. The 8G sled test is intended to simulate the longitudinal crash accleration in a severe train-to-train collision involving U.S. passenger equipment. Analyses of the dynamic sled test are useful for studying the sensitivity of the sled test to factors such as table height, table force-crush behavior, seat pitch, etc., which help to inform discussions on revisions to the test requirements eventually leading to safer seating environments for passengers.

An optimization framework is developed to minimize structural weight of the front-frame of heavy-duty trucks while satisfying stress constraint. The shape of the frame is defined by a number of design parameters (which define the shape of the side-rail, position and width of the internal brackets, and width of the flanges). In addition, the thickness of the engine-mount, the side-rails, inner-brackets, radiator mount, shock absorber and cab-mount connector are also considered as design variables. Aluminum Alloy, 6013-T6 is chosen as the material and the maximum allowable stress is the yield stress (320 MPa). A quantity known as ‘Violation’ is defined as the ratio of area in the front-end module where stress constraint is violated to the total area of the frame is introduced to implement stress constraints. For optimization, the penalty method is used where the objective is to minimize the total weight while keeping the value of the ‘Violation’ parameter less than 0.1 %. The Particle Swarm Optimization Algorithm is implemented using parallel computation for optimizing the structure. Commercial FEA software MSC.PATRAN is used for creating the geometry and the mesh whereas MSC.NASTRAN is used to perform static analysis. Six design load conditions, each corresponding to a road condition are used for the problem.

Metro-Rail transit systems are large-scale networks in numerous modern urban areas that play prominent direct and supportive roles in providing efficient mobility for sustaining communities and local economies. Any event leading to failure of a metro-rail network could have serious societal consequences, such as dramatic effect on the safety and wellbeing of commuters in addition to direct and indirect costs from its diminished performance that lead to resilience loss. Potential performance losses might exhibit complexity and pose a challenge for measurement and prediction. Hence, measuring the resilience of such a network enables its efficient enhancement in a cost-effective manner. Enhancing resilience highly depends on identifying recovery strategies with special attention not only to restoring connectedness but also on reducing associated failure and recovery costs. An effective recovery strategy must demonstrate rapid optimal restoration of a disrupted system while minimizing the cost of the disruption. The objective of this paper is to identify effective recovery strategies to reduce the performance loss and to minimize the total cost of a network during and after a disruptive event, using Washington D.C. Metro with its 91 stations and 140 links as a case study. Method of measuring performance loss in this paper, illustrates that the best recovery sequence typically reflects the order of components ranked based on their degree of vulnerability in the network. Also, the proposed cost model provides a basis to decision makers to identify an optimal recovery strategy according to both paramount recovery sequence and minimum cost consideration.

Fixed workstation tables in passenger rail coaches can pose a potential injury hazard for passengers seated at them during an accident. Tables designed to absorb impact energy while minimizing contact forces can reduce the risk of serious injury, while helping to compartmentalize occupants during a train collision. The Rail Safety and Standards Board (RSSB) in the U.K. issued safety requirement GM/RT2100, Issue 5 [1] and the American Public Transportation Association (APTA) in the U.S. issued safety standard APTA PR-CS-S-018-13, Rev. 1 [2] with the goals of setting design and performance requirements for energy-absorbing workstation tables. The U.S. Department of Transportation, Federal Railroad Administration (FRA) Office of Research, Development and Technology directed the Volpe National Transportation Systems Center (Volpe Center) to evaluate the performance of the Hybrid-III Rail Safety (H3-RS) anthropomorphic test device (ATD), also known as a test dummy, in the APTA sled test in order to incorporate a reference to the H3-RS in the safety standard. The Volpe Center contracted with the manufacturer of the H3-RS, Transport Research Laboratory (TRL), in the U.K. to conduct a series of sled tests [3] with energy-absorbing tables, donated by various table manufacturers. The tables were either already compliant with the RSSB table standard or were being developed to comply with the APTA table standard. The sled test specified in Option A of the APTA table standard involves the use of two different 50th percentile male frontal impact ATDs. The H3-RS and the standard Hybrid-III (H3-50M) ATDs performed as expected. The H3-RS, which features bilateral deflection sensors in the chest and abdomen, was able to measure abdomen deflections while the H3-50M, which features a single sensor measuring chest compression, was not equipped to measure abdomen deflection. This study attempts to validate a finite element (FE) model of the APTA 8G sled test with respect to the thorax response of the H3-RS and H3-50M. The model uses a simplified rigid body-spring representation of one of the energy absorbing tables tested by TRL. The FE models of the H3-RS ATD and the H3-50M ATD were provided by TRL and LSTC, respectively. Results from the sled tests and FE simulations are compared using data obtained from the chest accelerometer, the chest and abdomen deflection sensors, and the femur load cells. Using video analysis, the gross motion of the dummies and table are also compared. Technical challenges related to model validation of the 8G sled test are also discussed. This study builds on previous analyses conducted to validate the abdomen response of the H3-RS FE model, which are presented in a companion paper [4].

In order to investigate the performance and emissions behavior of a high compression ratio Compression Ignition (CI) engine operating in Partially Premixed Charge Compression Ignition (PPCI) mode, a series of experiments were conducted using a single cylinder naturally aspirated engine with a high-pressure rail fuel injection system. This included a moderately advanced direct injection strategy to attempt PPCI combustion under low load conditions by varying the injection timing between 25° and 35° Before Top Dead Center (BTDC) in steps of 2.5°. Furthermore, during experimentation the fuel injection pressure, engine speed, and engine torque (through variance of the fuel injection quantity) were kept constant. In-cylinder pressure, emissions, and performance parameters were measured and analyzed using a zero-dimensional heat release model. Compared to the baseline conventional 12.5° BTDC injection, in-cylinder pressure and temperature was higher at advanced timings for all load conditions considered. Additionally, NO x , PM, CO, and THC were higher than conventional results at the 0.5 N-m load condition. While PM emissions were lower, and CO and THC emissions were comparable to conventional injection results at the 1.5 N-m load condition between 25° and 30° BTDC, NO x emissions were relatively high. Hence, there was limited success in beating the NO x -PM tradeoff. In addition, since Start of Combustion (SOC) occurred BTDC, the resulting higher peak combustion pressures restricted the operating condition to lower loads to ensure engine safety. As a result, further investigation including Exhaust Gas Recirculation (EGR) and/or variance in fuel Cetane Number (CN) is required to achieve PPCI in a high compression ratio CI engine.

A simplified roll-plane model is proposed to assess the effect of the vertical position of the center of gravity of the body-cargo system, on the rail fatigue life. A set of assumptions are made to simplify the analysis, including neglecting the bogie’s dynamic contribution to the wheel-rail forces. Three performance measures are defined to assess the effect of different dedicated railway cars on the rail fatigue life, including the fourth-power law, the load dispersion, and the rail fatigue. The simulation results suggest that the vertical position of the center of gravity of the body-cargo set, severely affects the fatigue life of the railway material, with the two-stack car being the most aggressive. For example, twice as aggressive as the gondola car.

The complicated steel wheel and rail interaction on curve causes side wear on rail head. Thus, the cost of maintenance for the track on curve is significantly higher than that for track on a tangent. The objective of this research is to develop 3D printing technology for repairing the side wear. In this paper, the study examines induced residual thermal stresses on a rail during the cooling down process after 3D printing procedure using the coupled finite volume and finite element method for thermal and mechanical analysis respectively. The interface of the railhead and additive materials should conserve high stresses to prevent any crack initiation. Otherwise, the additive layer would likely shear off the rail due to crack propagation at the rail/additive interface. In the numerical analysis, a cut of 75-lb ASCE (American Society of Civil Engineers) worn rail is used as a specimen, for which a three-dimensional model is developed. The applied residual stresses, as a result of temperature gradient and thermal expansion coefficient mismatch between additive and rail materials, are investigated. At the beginning, the worn rail is at room temperature while the additive part is at a high initial temperature. Then, additive materials start to flow thermal energy into the worn rail and the ambient. The thermal distribution results from thermal analysis are then employed as thermal loads in the mechanical analysis to determine the von-Mises stress distribution as the decisive component. Then, the effect of preheating on residual stress distribution is studied. In this way, the thermo-mechanical analysis is repeated with an increase in railhead’s initial temperature. In thermal analysis, the temperature contours at different time steps for both the non-preheated and preheated cases indicate that preheating presents remarkably lower temperature gradient between rail and additive part and also represents a more gradual cooling down process to allow enough time for thermal expansion mismatch alignment. In mechanical analysis, the transversal von-Mises stress distribution at rail/additive interface is developed for all cases for comparison purposes. It is shown that preheating is a key factor to significantly reduce residual stresses by about 40% at all points along transversal direction of interface.

Commonly-used sloshing models are either unable to capture changes in the continuous distribution of the fluid free surface, or are not suited for the integration with high fidelity computational multibody system (MBS) algorithms. The objective of this investigation is to address this deficiency by developing a new continuum-based liquid sloshing approach that accounts for the effect of complex fluid and tank geometry and can be systematically integrated with MBS algorithms in order to allow for studying complex motion scenarios. A unified geometry/analysis mesh is used from the outset to examine the effect of liquid sloshing on railroad and highway vehicle dynamics during various maneuvers including braking and curve negotiation [1,2]. Using a non-modal approach, the geometry of the tank and fluid is accurately defined, a continuum-based fluid constitutive model is developed, and a fluid-tank contact algorithm using the penalty approach is employed. In order to examine the effect of liquid sloshing on vehicle dynamics during curve negotiation, a general and precise definition of the outward inertia force is defined, which for flexible bodies does not take the simple form used in rigid body dynamics. During maneuvers, the liquid may experience large displacements and significant changes in shape that can be captured effectively using absolute nodal coordinate formulation (ANCF) finite elements. For rail systems, the liquid sloshing model is integrated with a three-dimensional MBS vehicle algorithm, in which the three-dimensional wheel/rail contact force formulation is used to account for the longitudinal, lateral, and spin creep forces that influence vehicle stability. The effects of fluid sloshing on vehicle dynamics in the case of a tank partially filled with liquid are studied and compared with the equivalent rigid body model in braking and curve negotiation. The results obtained in the study of the rail vehicle model show that liquid sloshing can exacerbate the unbalance effects when the rail vehicle negotiates a curve at a velocity higher than the balance speed, and can significantly increase coupler forces during braking. Analysis of the highway vehicle model shows that the liquid sloshing changes the contact forces between the tires and the ground — increasing the forces on certain wheels and decreasing the forces on other wheels — which in cases of extreme sloshing, can negatively impact the vehicle stability by increasing the possibility of wheel lift and vehicle rollover.

Poles are regularly placed along highways and are used to support signs, lights and electrical lines. The Midwest Guardrail System (MGS) is a standard W-beam guardrail system used throughout the United States to redirect vehicles that leave the roadway away from dangerous roadside obstacles, like ravines, water hazards, and bridge piers. Placing poles near a guardrail may affect its ability to safely contain and redirect vehicles. The compatibility of poles placed in the proximity of the MGS is studied using nonlinear finite element analysis. Computer simulations were conducted with vehicles impacting the MGS with varying lateral pole offsets between the back of the system and the front face of the pole, and varying longitudinal pole location from being placed directly behind a post to directly behind the unsupported rail half-way between posts. Results show that poles placed within 16 inches behind the MGS may cause concern in regard to acceptable crash test performance for guardrail systems. Additional simulations and full-scale crash testing is required before guidelines can be recommended.

In this paper, a 3-D fluid-structure interaction (FSI) analysis on the performance of the high-pressure fuel pump for diesel engines is presented. The fluid and structure are two-way coupled and several complex factors are taken into accounts in the FSI model. For instance, the fluid model includes not only the high-pressure fuel pump but also the rail and pressure-control valve which are used to maintain a stable delivery pressure of the pump; Gap boundary condition is adopted to simulate the opening and closing of the valve; The flow is assumed to be nonisothermal and the physical properties of the fuel such as dynamic viscosity and density are functions of pressure and temperature. While in the structure model, the spring force on the valve and the contacts between the valve and the valve seat as well as the top block are considered. The calculated volumetric efficiency losses agree well with the experiments, which indicates that the FSI model established in this study could well predict the physical phenomenon taking place in the high-pressure fuel pump. Several new conclusions can be drawn from the discussions on the results such as the suction efficiency loss due to the delay closing of the inlet valve is extremely small while the suction loss due to the expansion of the high-pressure fuel entrapped in the dead volume is very large.

A computer numerically controlled (CNC) milling center is a machine tool for the production of parts with planar, cylindrical and shaped surfaces. The milling center analyzed here includes an open frame — a structure resembling the shape of the letter C. The main cutting motion is performed by a tool clamped in the spindle. Secondary motion can be linear, rotary or a combination of these. Linear movements in three axes are performed by the tool by means of linear motion components (i.e. motion screws and linear guide rails). Rotary motion is performed only when the workpiece is clamped to the rotary table which is mounted on the mounting plate. The basic demands placed on the structure of a milling center include high static and dynamic stiffness during machining processes. This article is primarily aimed at evaluating the response of the frame of the CNC milling machine to the excitation caused by the fluctuation of cutting forces due to step changes in the number of engaged cutting edges. To ensure optimum machining conditions it is important to set suitable cutting conditions for a frame structure with sufficient stiffness. Unsuitable cutting conditions and low stiffness of the machine frame may lead to dimensional inaccuracies of the workpiece, to decreased quality of the machined surfaces or even to the destruction of the tool cutting edges. The aims of the study include the determination of the static deformation, modal analysis to assess the dynamic properties of the frame, and harmonic response analysis, taking into consideration the amplitudes of the loading forces specified in accordance with the recommended operating conditions of the individual tools. Finite element method (FEM) analyses of the frame were performed using MSC.Marc software. Due to the high structural complexity of the computer aided design (CAD) model, the computational model for the FEM analysis had to be simplified. Only the major structural parts and the connecting parts were meshed in detail, combining both structured and unstructured mesh. Geometrically complicated cast parts with large changes of thickness were meshed with linear tetrahedral elements (tetra4) with full integration. Rotationally symmetrical parts, plates and linear guide rails components were meshed with linear brick elements (hex8) with full integration. The overall number of elements was approximately 1,400,000. Tools, including the clamping head and the spindle, are represented by approximate meshes of brick elements. However, a detailed FEM model of the spindle and the tool would be needed for the analysis of the self-excited oscillations during machining, which is the subject of a large number of scientific publications. Increased attention was paid to the incorporation and set-up of the springs between corresponding pairs of nodes of the meshed linear motion components. As a computational model for modal and harmonic response analyses needs to be strictly linear, only linear elastic material properties and linear springs were defined in the analyses presented here.

Research to develop new technologies for increasing the safety of passengers and crew in rail equipment is being directed by the Federal Railroad Administration’s (FRA’s) Office of Research, Development, and Technology. Crash energy management (CEM) components which can be integrated into the end structure of a locomotive have been developed: a push-back coupler and a deformable anti-climber. These components are designed to inhibit override in the event of a collision. The results of vehicle-to-vehicle override, where the strong underframe of one vehicle, typically a locomotive, impacts the weaker superstructure of the other vehicle, can be devastating. These components are designed to improve crashworthiness for equipped locomotives in a wide range of potential collisions, including collisions with conventional locomotives, conventional cab cars, and freight equipment. Concerns have been raised in discussions with industry that push-back couplers may trigger prematurely, and may require replacement due to unintentional activation as a result of service loads. Push-back couplers are designed with trigger loads meant to exceed the expected maximum service loads experienced by conventional couplers. Analytical models are typically used to determine these required trigger loads. Two sets of coupling tests are planned to demonstrate this, one with a conventional locomotive equipped with conventional draft gear and coupler, and another with a conventional locomotive equipped with a push-back coupler. These tests will allow a performance comparison of a conventional locomotive with a CEM-equipped locomotive during coupling. In addition to the two sets of coupling tests, car-to-car compatibility tests of CEM-equipped locomotives, as well as a train-to-train test are also planned. This arrangement of tests allows for evaluation of the CEM-equipped locomotive performance, as well as comparison of measured with simulated locomotive performance in the car-to-car and train-to-train tests. This paper describes the results of the coupling tests of conventional equipment. In this set of tests, a moving locomotive was coupled to a standing cab car. The coupling speed for the first test was 2 mph, the second test 4 mph, and the tests continued with the speed incrementing by 2 mph until the last test was conducted at 12 mph. The damage observed resulting from the coupling tests is described. The lowest coupling speed at which damage occurred was 6 mph. Prior to the tests, a one-dimensional lumped-mass model was developed for predicting the longitudinal forces acting on the equipment and couplers. The model predicted that damage would occur for coupling speeds between 6 and 8 mph. The results of these conventional coupling tests compare favorably with pre-test predictions. Next steps in the research program, including future full-scale dynamic tests, are discussed.