The recent increase of train speed and frequency determines a rise of the loads transmitted to the superstructure. Therefore, railway components might experience service loads that have not considered at the design stage. Moreover, wear and backlash modification between components of a mechanical system might be able to modify the internal boundary conditions of the assembly. According to damage tolerance philosophy, an initial flaw is assumed to exist in the fatigue critical location of a structural component, and the analysis of the crack propagation life for such component needs accurate Stress Intensity Factor (SIF) evaluations. In this study, the effects of the boundary conditions on the crack propagation life have been evaluated for a semi-elliptical surface crack having semi-axes a and c and growing from the root of a shoulder fillet notch in a round bar loaded in bending. Two cases have been analyzed: - the shoulder is free from external forces; - the shoulder is in contact with an adjacent generic body. At first, the SIF distribution has been calculated with the Virtual Crack Closure Technique, considering or not the nonlinear effect induced by the contact forces arising from the interaction between the shoulder and the neighboring component. Successively, in both the above cases a two-parameters propagation law has been utilized to predict the evolution of both crack shape and crack depth when a cyclic bending load is applied to the rod. For this purpose, different values of the Stress Concentration Factor at the root of the fillet, and of the initial aspect ratio of the crack front, were considered in the calculations. It is found that the aspect ratio evolves to a unique asymptote, taking or not into account the non-linearity introduced by the contact at the shoulder, and this value depends on the notch severity. The ratio between the dimensionless SIFs obtained with and without the unilateral constraint at the shoulder, β cs / β fs , does not depend on the relative crack depth and crack shape. Also, the effect of the notch severity on the dimensionless SIF appears to be evident only for the portion of the crack front in the vicinity of the free surface.

This paper is a continuation of a previous study conducted at Kansas State University [8]. This paper demonstrates the influence of the thickness of concrete cover, compressive strength of concrete and the type of wire indentation on bond performance between steel and concrete in pre-stressed concrete ties using a consistent concrete mixture. A key objective of this research is to find the best parameters for pre-stressed concrete ties to prevent them from splitting/cracking in the field. This is very important for pre-stressed manufacturers, and especially for the railroad crosstie industry, so as to avoid failures in the field. The goal is to develop a qualification test with the capability to identify the compatible combinations of wire type and concrete mix before the ties are manufactured. A study took place at Kansas State University to understand and quantify the influence of variables such as the thickness of concrete cover, type of indents, and the compressive release strength on the bond behavior between steel and concrete. For the experimental testing three prisms with different cross sections were cast at the same time in series. Four pre-stressing wires were symmetrically embedded into each concrete prism and the spacing between wires was 2.0 inches. All prisms had the same length of 59.5in with square cross section. With the thickness of concrete cover of 3/4″ the first prism had a 3.5×3.5in square cross section, the second prism had a 5/8″ thickness of concrete cover and 3.25×3.25in square cross section and the third prism had a 1/2″ thickness of concrete cover and a 3.0×3.0in square cross section. All pre-stressing wires which were used in these tests had a 5.32mm diameter and were of different wire types. The indent pattern variations of the wire types included spiral, classical chevron shape, and the extreme case of smooth wire with no indentations. The wires were initially tensioned to 7000 pounds (31.14 KN) and then gradually de-tensioned after reaching the desired compressive strength. The different compressive (release strength) strength levels tested included 4500 psi (31.03 MPa) and 6000 psi (41.37 MPa). For this study, a consistent concrete mixture with 0.32 water-cement ratio was used for all prisms, except for prisms casted with WE wire. For these prisms a water-cement ratio of 0.38 was used. Prisms had almost identical geometrical and mechanical properties as pre-stressed concrete ties which are manufactured in the railroad industry. Each prism provided a sample of eight different independent splitting tests of concrete cover (four wire cover tests on each end) for a given release strength. All cracks which appeared after de-tensioning were observed and measured to identify the cracking field, and all sides of the prisms on the live and dead end were marked for identification. For all prisms, longitudinal strain profiles on the live end and dead end were measured along with the values of transfer lengths. The strain profiles were taken using an automated Laser-Speckle Imaging (LSI) system. All results, representing quantitative and qualitative assessment of cracking behavior, are given in this paper as a function of thickness of concrete cover and release strength of concrete. For each sample prism, crack length and crack width were measured, and crack area was calculated as a simple function of crack length and crack width. In the case where spalling occurred, the crack width used was arbitrary set at 0.2in. These tests reveal the influence of thickness of concrete cover, the indented wire type and the release strength of concrete on the bond between steel and concrete. This work represents a successful first step in the development of a qualification test to ensure adequate splitting resistance in pre-tensioned concrete railroad ties.

This paper reports on the ballast layer mesoscale behavior, tie-ballast interaction, and ballast-subgrade interaction under five crosstie support conditions, namely full support, lack of rail seat support, lack of center support, high center binding, and severe center binding condition. Discrete Element Method, an effective technique to study particulate natured unbound aggregate materials, i.e., ballast, was adopted in this study. The DEM simulations included one-tie spacing geometry, approximately 11,000 polyhedral particles. The ballast gradation used in DEM models was according to the AREMA No. 3 and No. 4A specifications. The shape properties of ballast particles in DEM models was consistent with field collected samples. The pressure distributions along tie-ballast interface under rail seat load of 10-kips predicted by DEM simulations were in good agreement with the results backcalculated from laboratory tests, which validated the DEM models. Next, DEM simulations considered rail seat loads of 20-kips and 25-kips. The predicted results indicated that support condition is a key factor for predicting normal stress distribution and force transmission within ballast layer. Ballast particles in shoulders and areas with poor support indicated low or negligible contact stresses. Extremely high normal stresses observed in some support conditions often exceeded single particle crushing load limit and thus would cause ballast particle breakage and layer degradation under repeated loading. Further, the tie-ballast pressure captured in some scenarios could be higher than allowable maximum pressure of 85-psi under concrete tie in AREMA standard. Finally, the pressure at bottom of the ballast layer obtained from the DEM simulations were compared with top of subgrade pressure calculated from analytical/empirical equations such as Talbot equation and AREMA manual.

The bond between wire and concrete is crucial for transferring the stresses between the two materials in a prestressed concrete member. Furthermore, bond can be affected by such variables as thickness of concrete cover, type of pre-stressing (typically indented) wire used, compressive (release) strength of the concrete, and concrete mix. This work presents current progress toward the development of a testing procedure to get a clear picture of how all these parameters can ruin the bond and result in splitting. The objective is to develop a qualification test procedure to proof-test new or existing combinations of pre-stressing wire and concrete mix to ensure a reliable result. This is particularly crucial in the concrete railroad crosstie industry, where incompatible conditions can result in cracking and even tie failure. The goal is to develop the capability to readily identify compatible wire/concrete designs “in-plant” before the ties are manufactured, thereby eliminating the likelihood that defectively manufactured ties will lead to in-track tie failures due to splitting. The tests presented here were conducted on pre-tensioned concrete prisms cast in metal frames. Three beams (prismatic members) with different cross sections were cast simultaneously in series. Four pre-stressing wires were symmetrically embedded into each concrete prism, resulting in a common wire spacing of 2.0 inches. The prisms were 59.5in long with square cross sections. The first prism was 3.5 × 3.5in with cover 0.75in, the second was 3.25 × 3.25in with cover 0.625in and the third prism in series was 3.0 × 3.0 in with cover 0.50in. All pre-stressing wires used in these initial tests were of 5.32 mm diameter and were of the same wire type (indent pattern) denoted by “WE”, which had a spiral-shaped geometry. This is one of several wire types that are the subject of the current splitting propensity investigation. Others wire types include variations of the classical chevron shape, and the extreme case of smooth wire with no indentions. The wires were initially tensioned to 7000 pounds (31.14 KN) and then gradually de-tensioned after reaching the desired compressive strength. The different compressive (release strength) strength levels tested included 3500 psi (24.13 MPa), 4500psi (31.03 MPa), 6000 psi (41.37 MPa) and 12000psi (82.74MPa). A consistent concrete mix with water-cement ratio 0.38 was used for all castings. Geometrical and mechanical properties of test prisms were representative of actual prestressed concrete crossties used in the railroad industry. Each prism provided a sample of eight different and approximately independent splitting tests of concrete cover (four wire cover tests on each end) for a given release strength. After de-tensioning, all cracks that appeared on the prisms were marked, and photographs of all prism end surfaces were taken to identify the cracking field. During the test procedure longitudinal surface strain profiles, along with live-end and dead-end transfer lengths, were also measured using an automated Laser-Speckle Imaging (LSI) system developed by the authors. Both quantitative and qualitative assessment of cracking behavior is presented as a function of cover and release strength. In addition to the identification of whether cracking took place at each wire end location, measurements of crack length and crack area are also presented for the given WE wire type. The influence of concrete cover and release strength are clearly indicated from these initial tests. The influence of indented wire type (indent geometry) will also be discussed in this paper, along with a presentation of some preliminary test results. This work represents a successful first step in the development of a qualification test for validating a given combination of wire type, concrete cover, and release strength to improve the reliability of concrete railroad crosstie manufacturing.

The lateral stability of the continuous welded rail (CWR) depends on a number of parameters which contribute to the progressive loss of the initial alignment of the track and its consequent predisposition to deform sideways, gradually or sharply, with serious risks for the safety both of passengers and operators. Different types of initial lateral defect, in terms of shape and size, are introduced by many authors in their own numerical and analytical model, but essentially all of them can be traced back — except for small “personalizations” — to the model proposed by Andrew Kish, who hypothesized the existence of a misalignment defect having the shape of a sine curve extended for half-wavelength, characterized by amplitude and wavelength values typical of the USA railroads. Moreover, all previous studies focused their attention on the introduction, in a geometrically perfect railway track, of a single defect confined in a zone of finite dimensions and having a rather simple geometry which qualitatively approximates the real defect, with the aim of simplifying the calculation of the buckling temperatures of the track associated with such geometry. In this paper, it was preliminarily analyzed the way the defect introduced in the track affects the critical temperature values. It started with a defect created artificially, applying to a geometrically perfect track and in the absence of thermal loads, a lateral displacement in the central transversal section of the track, and calculating, with the hypothesis of linear elastic behavior, the resulting deformed shape, which was assumed, after zeroing the corresponding stress field, as the input geometry for the subsequent buckling calculation. The deformed shape so obtained, being a Zimmermann deformed shape type, has no geometrical discontinuities near the defect and interprets in a natural way the defected geometry of the track, due to the dependence of its configuration on the flexural stiffness of the entire track in the lateral plane. Afterwards, modeling was carried out taking into account the real behavior of the track after the loss of its rectilinear configuration: the defect was created simulating the response of the track to a momentary lateral load — resulting, e.g., from train passages — which succeeded to cause a permanent displacement resulting from the elastic-plastic response of the track. The deformed shape of the track obtained in this way was used as the input geometry for the calculation of the buckling temperatures, once without resetting the stress field induced in the structure by the loading–unloading hysteresis cycle, and then considering the track free from internal stresses. The results show that both the numerical model that contemplate the defect introduced “plastically”, and that where the track is free from internal stresses, lead to more conservative results against the risk of thermal buckling in railway tracks made with CWR. A better approximation of the realistic representation of a generic defected railway track was pursued considering an indefinite number of defects distributed along the track, where each defect was characterized by different amplitude and wavelength values. The obtained results show that the presence of multiple defects further reduces the safety factor against the thermal track buckling phenomenon. The paper ends with the proposal of an evaluation criterion that takes into account the effects of multiple alignment defects on the critical buckling temperatures in continuous welded rail tracks.

The ballast layer serves as a major structural component in typical ballasted railroad track systems. When subjected to an external load, ballast particles present a complex mechanical response which is strongly dependent on particle to particle interactions within this discrete medium. One common test used to study the shear strength characteristics of railroad ballast is the Direct Shear Test (DST). However, it is often not feasible in standard geotechnical engineering laboratories to conduct direct shear tests on ballast particles due to significantly large specimen and test setup requirements. Even for the limited number of laboratories equipped to accommodate the testing of such large specimens, conducting repeated tests for parametric analysis of different test and specimen parameters on shear strength properties is often not feasible. Numerical modeling efforts are therefore commonly used for such parametric analyses. An ongoing research study at Boise State University is using the Discrete Element Method (DEM) to evaluate the effects of varying particle size and shape characteristics (i.e., flakiness, elongation, roundness, angularity) on direct shear strength behavior of railroad ballast. A commercially available three-dimensional DEM package (PFC3D ® ) is being used for this purpose. In numerical modeling, railroad ballasts can be simulated using spheres (simple approach) and non-breakable clumps (complex approach). This paper utilizes both approaches to compare the ballast stress-strain response as obtained from DST. Laboratory test results available in published literature are being used to calibrate the developed numerical models. This paper presents findings from this numerical modeling effort, and draws inferences concerning the implications of these findings on the design and construction of railroad ballast layers.

Fifty years ago, the railcar industry relied entirely on classical analysis methods using fundamental solid mechanics theory to establish design and manufacturing protocols. While this method produced working designs, the assumptions required by this type of analysis often led to overdesigned railcars. In the 1950s, the generalized mathematical approach of Finite Element Analysis (FEA) was developed to model the structural behaviors of mechanical systems. FEA involves creating a numerical model by discretizing a continuous system into a finite system of grid divisions. Each grid division, or element, has an inherent geometric shape and each element is comprised of points which are referred to as nodes. The connected pattern of nodes and elements is called a mesh. A solver organizes the mesh into a matrix of differential equations and computes the displacements using linear algebraic operations from which strains and stresses are obtained. The rapid development of computing technology provided the catalyst to drive FEA from research into industry. FEA is currently the standard approach for improving product design cycle times that were previously achieved by trial and error. Moreover, simulation has improved design efficiency allowing for greater advances in weight, strength, and material optimization. While FEA had its roots planted in the aerospace industry, competitive market conditions have driven simulation into many other professional fields of engineering. For the last few decades, FEA has become essential to the submittal of new railcar designs for unrestricted interchange service across North America. All new railcar designs must be compliant to a list of structural requirements mandated by the Association of American Railroads (AAR), which are listed in its MSRP (Manual of Standards and Recommended Practices) in addition to recommended practices in Finite Element (FE) modeling procedures. The MSRP recognizes that these guidelines are not always feasible to completely simulate, allowing the analyst to justify situations where deviations are necessary. Benefits notwithstanding, FEA has inherent challenges. It is understood that FEA does not provide exact solutions, only approximations. While FEA can provide meaningful insight into actual physical behavior leading to shorter development times and lower costs, it can also create bogus solutions that lead to potential safety and engineering risks. Regardless of how appropriate the FEA assumptions may be, engineering judgment is required to interpret the accuracy and significance of the results. A constant balance is made between model fidelity and computational solve time. The purpose of this paper is to discuss the FEA approach to railcar analysis that is used by BNSF Logistics, LLC (BNSFL) in creating AAR compliant railcar designs. Additionally, this paper will discuss the challenges inherent to FEA using experiences from actual case studies in the railcar industry. These challenges originate from assumptions that are made for the analysis including element types, part connections, and constraint locations for the model. All FEA terminology discussed in this paper is written from the perspective of an ANSYS Mechanical user. Closing remarks will be given about where current advances in FEA technology may be able to further improve railcar industry standards.

Ballast compaction and particle rearrangement cause ballast to rotate and move vertically and horizontally. Ballast movement, including translation and rotation, has a significant effect on track performance. Large movement of ballast particles leads to track geometry roughness, e.g., hanging ties, and thus increases potential of damage and deterioration to rails, ties and fastening components. This study investigated ballast particle movement at different locations beneath a crosstie. In the paper, a wireless device — “SmartRock” was utilized to monitor ballast movement under cyclic loading in laboratory tests. The SmartRock has a shape of a realistic ballast particle. Inside the SmartRock was imbedded a tri-axial accelerometer, tri-axial gyroscope, and tri-axial magnetometer with 9 degrees of freedom so that particle translation, rotation and orientation can be interpreted, relatively. The real-time measurements were recorded by the SmartRock and then sent to a computer via Bluetooth. In the laboratory tests, a ballast box was constructed. In the ballast box, a half section of a typical railroad track was constructed. Five hundred cyclic load repetitions were applied on the top of the rail. Translational and rotational accelerations of the particle were recorded by the “SmartRock”. Three ballast box tests were conducted. Two SmartRocks were placed beneath the middle of tie and the edge of tie, respectively but at different depths during each test — right under the tie, 12 cm beneath the tie and 25 cm beneath the tie. The results indicated that (1) ballast particles had translational as well as rotational modes under cyclic loading; (2) ballast particles had rotation together with horizontal translation; (3) particle rotation were higher beneath the edge of tie than those beneath the middle of tie; (4) Ballast movement were significantly reduced with depth. The paper also further confirmed that the SmartRock was capable of recording real-time translational and rotational accelerations, which would not have altered the motions of surrounding ballast particles due to its realistic shape of a particle, hence, provided a new means to monitor ballast particle movement in railroad engineering.

Rail grinding has become a widely accepted practice in the railroad industry to not only remove surface defects but to also shape the head of the rail to achieve a specific wheel/rail contact interface. Optimal wheel/rail contact is paramount for a variety of reasons including (but not limited to): reduced rail wear and fatigue defects with associated increased rail life, improved steering, and reduced wheel/rail noise. Historically, the process of profile grinding has been performed through the selection of a grinding pattern from a predefined library of preset motor angles and power settings, as well as grinder traveling speed. By applying calibrated metal removal equations created to estimate the metal removal of a single grinding stone on a specific grinding vehicle and a measured shape of the rail, it is possible to iteratively create a grinding pattern and estimate the post-grind rail shape in both real-time and offline. These dynamically created grinding patterns position and power grinding motors to remove metal only where it needs to be removed to achieve the desired rail shape (template). Over grinding and extraneous grinding is minimized using this grind optimization technique. The concept of dynamic rail grinding pattern generation was applied to optimize grinding for the VLI railroad in Brazil. Based on a survey of both wheel profiles and rail profiles throughout the VLI network (in which digital coordinates of the transverse cross-sections were recorded using handheld profile measurement devices), new rail grinding templates (desired rail shape dependent upon rail curvature and elevation) were developed to control the wheel/rail interface. These new rail grinding templates were aligned to measured rail profiles categorized by curvature and high/low rail types to create representative average metal removal curves. The resulting metal removal curves were then used as inputs to the dynamic grinding pattern generation functions to create custom grinding patterns specifically calibrated to the grinder in operation on VLI. Results showed that applying dynamic grinding pattern generation resulted in increased productivity of the grinder (faster speeds and less passes) with corresponding improvement of the post-grind rail shape, i.e., increased quality of the final rail shape. This paper will discuss the background theory behind dynamic pattern generation and how it can be applied to normal grinding operations. Additionally, an application of this process will be presented which utilized dynamic pattern generation to create targeted corrective grinding patterns for the VLI railroad.

Accurate unbiased assessment of transfer length for prestressed concrete railroad ties requires detailed knowledge of the longitudinal variation of geometrical cross-section parameters responsible for establishing the resulting surface strain profile. This is because the complex cross-sectional shape produces a non-uniform strain plateau region, which makes the accurate evaluation of transfer length more difficult. In particular, human judgment of a “plateau region” for assessment of the average maximum strain becomes subject to large uncertainty, and clearly this procedure cannot be used in any type of automated in-plant transfer length diagnostics. The important geometrical tie parameters include the cross-sectional area, centroid, moment of inertia, and the eccentricity of the prestressing wires. If a CAD drawing is available, this information can be digitally extracted from the CAD model representation of the crosstie. In fact, this digital extraction has been done and has already been in use for some time in assessing transfer length for one of the common crosstie manufacturer designs. However, current research efforts are investigating the characteristics of existing crossties which have been in track for many years, for which CAD drawings of the original designs are unlikely to be available. The objective of the current research is to develop a comprehensive understanding of the material characteristics that have caused splitting failures in prestressed concrete railroad ties, and those characteristics that have resulted in ties that have performed well after many years in track. As part of this effort, a three-dimensional (3D) Optical Scanning System is being used to accurately scan and quantify the surface geometry of previously manufactured ties that have been in service, so as to produce an accurate 3D CAD model for later analysis associated with the above long-term research objectives. For the initial phase of this work, a sample from the CXT crossties of known geometrical characteristics that were subjected to representative long-term loading at the TTCI Facility in Pueblo Colorado, was scanned so as to accurately map out detailed 3D tie surface geometry. These ties were cast using the same concrete materials but with different prestressing wires, and were all subjected to the same extreme in-track loading for a period of several years. A commercially-available 3D Laser-Based Optical Scanning System, having a maximum spatial resolution of approximately 0.1mm, was used to perform the surface scanning operations presented in this paper. The CXT tie provides a useful initial evaluation of the accuracy and general feature capture capability of the scanning procedure, since a 3D CAD model for this tie has been provided by the manufacturer. A detailed qualitative and quantitative analysis is presented which compares the 3D CXT CAD model geometry with the 3D geometry of the experimentally scanned ties. Illustrations as to how this 3D technique can reveal such features as abrasion and wear, along with the longitudinal variation of the above mentioned cross-section parameters associated with longitudinal surface strain and transfer length assessment, are included in this paper.

Ballast aggregate settlement is generally a result of consolidation or rearrangment of ballast particles in the area underneath crossties. Excessive settlement negatively impacts track performance, resulting in increased risk of train derailment. The purpose of this paper is to compare two methods to evaluate ballast aggregate settlement with repeated loading in railroad: discrete element modeling and laboratory tests using “SmartRock”. In this study, ballast aggregates are considered as uniformly graded, angular shaped with crushed faces. For the discrete element modeling, digital imaging techniques are utilized to create the ballast aggregates. Aggregate settlement in railroad ballast and the effect of aggregate shape on the dynamic response of ballast are evaluated through the discrete element simulations. A wireless device, “SmartRock” is developed to study the relationship between individual ballast particle behavior and overall ballast performance. It has a shell of a typical ballast particle shape with force cells attached on the surface and embedded with a tri-axial gyroscope, a tri-axial accelerometer, and a tri-axial magnetometer. The device can move under train traffic like a real ballast particle and record inter-particle contact forces and particle motion in real time. For the laboratory tests, a model-scale track section is constructed and subjected to repeated loading similar to train traffic. The developed “SmartRock” are embedded below rail seats and in the track shoulders. The laboratory data using “SmartRock” can be compared with results from the discrete element modeling in the future. These comparisons will validate the discrete element modeling procedure as a means to analyze railroad ballast aggregate behavior and the potential of “SmartRock” in railroad applications.

The Federal Railroad Administration’s Office of Research and Development is conducting research into fuel tank crashworthiness. A series of impact tests are planned to measure fuel tank deformation under two types of dynamic loading conditions — blunt and raking impacts. This paper describes the results of the first set of blunt impact tests for two retired EMD F-40 locomotive fuel tanks, Tank 232 and Tank 202. On October 8, 2013 and October 9, 2013, the FRA performed impact tests on two conventional passenger locomotive fuel tanks at the Transportation Technology Center (TTC) in Pueblo, Colorado. Each fuel tank was emptied of fluid and mounted on a crash wall with the bottom surface exposed. A rail cart modified with a “rigid” indenter was released to impact the center of the bottom of each fuel tank at about 6 mph. A center-impact on Tank 232 was chosen to impact between two baffles. A center-impact on Tank 202 was chosen to impact on a baffle. In the first test, Tank 232 was impacted by the indenter at 4.5 mph. The maximum residual indentation on the bottom of the tank measured approximately 5 inches. The tank deformed across the middle longitudinal span of the tank forming a diamond-shaped indention. In the second test, Tank 202 was impacted by the indenter at 6.2 mph. The maximum residual indentation on the bottom of the tank measured approximately 1.5 inches. The bottom of the tank deformed with an “X” shape spanning out from the location of square indenter at the center of the tank. Post-test autopsies revealed the deformation of the interior structures, i.e. baffles and attachments. There was no damage to the baffles in Tank 232. Deformation to the interior structure of Tank 202 was limited to the baffle directly beneath the impact location, which folded in the area near the impact location. Material coupons were cut and tensile testing performed to determine the properties of the materials used in each tank. Prior to the test, computer models were developed from measurements taken on the test articles. Material properties were estimated based on Brinell hardness measurements. Computer analyses were conducted to determine the conditions for the test, i.e. instrumentation, location of impact, target impact speeds and to predict the deformation behavior of the tank. Post-test, the resulting stress-strain relationships for the bottom sheets and baffles of both tanks were used to update the finite element models of the two tanks. The models were also updated to reflect the actual geometry of the tanks as confirmed by measurements of the tank interiors. The results of the finite element (FE) models run at the test conditions with the updated tank details are compared with the results from the test itself. Specifically, the deformation progression and the residual dent depth are compared between the tests and the models. In accidents, fuel tanks are subjected to dynamic loading, often including a blunt or raking impact from various components of the rolling stock or trackbed. Current design practice requires that fuel tanks have minimum properties adequate to sustain a prescribed set of static load conditions. Current research is intended to increase understanding of the impact response of fuel tanks under dynamic loading.

This paper presents the development of a semi-implicit time integration scheme, originally developed for structural dynamics in the 1970’s, and its implementation for use in Discrete Element Methods (DEM) for rigid particle interaction, and interaction of elastic bodies that are modeled as a cluster of rigid interconnected particles. The method is developed in view of ballast modeling that accounts for the flexibility of aggregates and the arbitrary shape and size of granules. The proposed scheme does not require any matrix inversions and is expressed in an incremental form making it appropriate for non-linear problems. The proposed method focuses on improving the efficiency, stability and accuracy of the solutions, as compared to current practice. A critical discussion of the findings of the studies is presented. Extended verification and assessment studies demonstrate that the proposed algorithm is unconditionally stable and accurate even for large time step sizes. It is demonstrated that the proposed method is at least as computationally efficient as the Central Difference Method. Guidelines for the implementation of the method to ballast modeling are discussed.

Consisting of large sized aggregate particles with uniform size distribution, ballast is an essential component of the track substructure to facilitate load distribution and drainage. As freight tonnage accumulates with traffic, ballast will get fouled increasingly due to either aggregate breakdown and degradation or contamination by other materials such as coal dust and subgrade soil intrusion. Fouling affects shear strength and load carrying ability of ballast layer especially under wet conditions. According to Selig and Waters [1], ballast fouling is often due to aggregate degradation, which covers up to 76% of all the fouling cases. To investigate the effects of ballast aggregate breakdown and degradation on the mechanical behavior of fouled ballast, a series of Los Angeles abrasion tests were performed in this study to generate fouled ballast materials caused by particle breakage and abrasion under a well-controlled laboratory environment. The change of particle shape properties during the Los Angeles abrasion tests was quantified and studied through image analysis technology. Large-scale triaxial tests were performed on specimens of new ballast, degraded ballast coarse particle fraction (without fines), and full gradation of degraded ballast (with fines) under repeated load application using a triaxial test device recently developed at the University of Illinois specifically for ballast size aggregate materials. The large-scale triaxial results indicated that the specimen having those degraded coarse particles yielded higher permanent deformation trends from repeated load triaxial testing when compared to the specimen with the new ballast gradation. As expected, the highest permanent deformation was obtained from the degraded ballast specimen having fine particles and the Fouling Index (FI) value of approximately 40.

Automated in-plant diagnostic testing of prestressed concrete railroad crossties is now within reach due to recent progress in robust surface strain measurement techniques. The newly developed non-contact Laser Speckle Imaging (LSI) technique has been shown to provide rapid and accurate surface strain profile measurement, which is a key requirement for rapid transfer length assessment. Accurate determination of transfer length is critical for maintaining continuous production quality in the modern manufacture of prestressed concrete railroad crossties. Conventional assessment of transfer length generally presumes the underlying existence of a bilinear prestressing force distribution and a corresponding bilinear surface strain profile. Furthermore, it is well-known that this bilinear profile is smoothed due to the effects of finite gauge length during the process of measuring surface strain. In addition, recent extensive crosstie measurements in concrete railroad tie plants have shown significant departures from this simple bilinear profile, which bring to question the general validity and reliability of the traditional 95% AMS (95% Average Maximum Strain) method. Deviations from the simple bilinear profile shape were shown to be partially due to the non-prismatic shape of typical concrete railroad ties. In addition, extensive comparisons between predicted and measured surface strain profiles on numerous crossties suggest that the underlying strain distribution for crossties is best represented by an exponential strain profile, with an asymptotic approach to the fully-developed compressive strain. This is in contrast with extensive testing of prisms with fixed cross-section and fixed prestressing wire eccentricity, for which the surface strain appears to be best represented by the simple bilinear strain profile. Clearly, departures from non-prismatic behavior have added complexity to transfer length measurement. If accurate and reliable measurements of this important quality control parameter are to be realized, these issues of transfer length uncertainty need to be addressed. This paper provides an experimental comparison of several possible alternative transfer length assessment procedures, in an attempt to answer important uncertainty questions which need to be addressed if rapid real-time transfer length is to be achieved. It is shown that in spite of considerable differences in the transfer length processing methods, and significant departures from prismatic behavior, the averaged results are in many cases consistent with the simple bilinear underlying strain profile assumption. Bias in the measurement of crosstie transfer length due to non-prismatic behavior will also be investigated in this paper.

Two finite element models of CRTS II ballastless track superstructure were developed to simulate temperature deformations of the precast concrete track slab. One of the models considered the fully bonding interface between the track slab and the cement asphalt mortar layers, while the other uses cohesive zone elements to simulate possible interfacial separation and slip. Warp deformations of the track slab under various temperature loads were calculated using both models. Numerical results show that the track slab glued fully to the mortar layers can approximately hold regular geometry shape even if it bears extreme temperature gradient loads. Although deteriorated mortar layers could increase significantly temperature deformations of the track slab, warp deformations of the track slab are still quite small and the rail harmonic irregularities due to the temperature deformations are tiny. It can be concluded that temperature deformations of CRTS II track slab have little effect on running safety and ride comfort of high-speed trains.

This research program was sponsored by the Federal Railroad Administration (FRA) Office of Research and Development in support of the advancement of improved safety standards for passenger rail vehicles. In a train collision, the cab or locomotive engineer is in a vulnerable position at the leading end of the vehicle. As cars with increased crashworthiness are introduced into service, there is a greater potential to preserve the space occupied by the engineer following an accident. In particular, full-scale impact tests have demonstrated the engineer’s space can be preserved at closing speeds up to 30 mph. When sufficient survival space is preserved, the next objective is to protect the engineer from the forces and accelerations associated with secondary impacts between the engineer and the control cab. Given the hard surfaces and protruding knobs in a control cab, even a low speed collision can result in large, concentrated forces acting upon the engineer. Researchers have designed a passive (i.e., requiring no action by the operator) interior protection system for cab car and locomotive engineers. The occupant protection system will protect engineers from the secondary impact that occurs following a frontal train impact, when the engineer impacts the control console. The protection system will result in compartmentalization of a 95th percentile anthropomorphic test device (ATD), and measured injury criteria for the ATD’s head, chest, neck, and femur that are below those currently specified in Federal Motor Vehicle Safety Standard (FMVSS) 208 [1]. The system that has been developed to protect the engineer includes a specialized airbag and a knee bolster with energy absorbing honeycomb material and deformable brackets. Finite element and lumped mass-spring analyses show the effectiveness of the system in limiting the injury criteria to survivable limits. Component tests have measured the key characteristics of the airbag and the knee brackets and have provided test data necessary to validate the analyses. Two tests were conducted to validate the airbag model. A static deployment test of the airbag measured the inflation progression, the inflated shape and the internal pressure of the airbag. A drop tower test of the airbag measured the force-crush and energy absorbing characteristics of the airbag. The knee bolster assembly consists of two components. Separate quasi-static tests of the aluminum honeycomb and the knee bolster bracket measured the force-crush and energy absorbing characteristics. The component test results were used to improve the computer model and permit analysis of the entire system. This paper discusses the prototype design, including background research, baseline definition and prototype development. The initial prototype design is analyzed using computer models. The components are tested to verify and improve the computer models. The test and analysis results are presented. Future work is planned for fabrication of the cab desk and prototype system to be used in a sled test with a 95 th percentile ATD.

In May 2011, a derailment of a passenger train occurred in a tunnel in the northeast region of the United States. Fortunately, no serious injuries or fatalities resulted from this derailment. The probable cause of the derailment was determined to be a broken rail from a defect originating in the base of the rail. This internal rail base defect is characterized as having a crescent, thumbnail, or semi-elliptical shape. In addition, the formation and growth of this defect may have been exacerbated by corrosion. This paper describes engineering calculations to estimate the growth rate of this type of rail base defect. These engineering calculations are based on applying the principles of fracture mechanics and beam theory. Fracture mechanics principles are applied to determine stress intensity factors for the semi-elliptical shaped defect with different aspect ratios. Stress intensity factors are then used to estimate the growth of the defect under the accumulation of tonnage from repeated wheel passages. For this purpose, the rail is assumed to behave as a beam in bending.

Ensuring rail safety is a priority for the Federal Railroad Administration (FRA) and the railroad industry in North America. One such endeavor is to leverage Wireless Sensor Networks (WSN) to monitor and report in real-time the status of mechanical and electrical components for each railcar, and in conjunction with other railroad subsystems, ensure the safety, security and integrity of transported goods. The envisioned solution utilizes sensors installed on each railcar to form a train-based wireless network and collect real-time (or near real-time) information on different elements of a train and transmit aggregated information to the locomotive, dispatch centers or regional offices for early fault detection and accident prevention. The railroads have been interested in using a standards-based low-cost communication protocol for this purpose, such as IEEE 802.15.4, often referred to as ZigBee. Our results show, however, that ZigBee was designed for smaller wireless networks, such as a single railcar. It exhibits several critical problems associated with the unique network topology found on a freight train and the size of such a network. In essence, the network would take the shape of a very long chain of nodes. Some of the problems stemming from this topology are excessively long synchronization delays for establishing the network along the entire train, severe problems with route discovery and maintenance necessary for selecting the next relay node along the chain, aggregation of data errors and a resulting unacceptable packet loss rate, the lack of a traffic prioritization mechanism to protect important packets such as those containing critical alarms of equipment failure, and many more. In this paper, we describe our findings and experiences in our evaluation of ZigBee for railcar monitoring onboard freight trains, a detailed analysis of the identified problems and their impact on the envisioned railcar monitoring as well as discuss potential solutions to these problems.

This paper reviews forms of spiral geometry that have been used and proposed as transition shapes in railroad tracks and compares them with respect to dynamic performance and spiral length. The dynamic performances that are compared were calculated using the Gensys rail vehicle motion simulation program and a well known model of the Eurofima coach. The results illustrate: 1) the dynamic disturbance caused by the discontinuity of first derivatives of track curvature and twist with the conventional ‘linear’ spiral; 2) the better dynamic performance of other simple forms of spiral including the virtually perfect performance of two of those forms; and 3) the amounts by which lengths of alternative spirals exceed that of the conventional ‘linear’ spiral for the example transition..