The Federal Railroad Administration (FRA) has been sponsoring research on rail integrity for several decades. This research has been chiefly managed and conducted by the Volpe National Transportation Systems Center (Volpe). Particular focus has been given in this research to rail head defects, known as detail fractures, since they are the most commonly encountered defect in continuous welded rail track [1]. Testing and analyses have been performed on railroad rails manufactured without head hardening. Modern rail, however, are now heat treated during the manufacturing process to harden the rail surface to increase its resistance to wear. As such, the heat treatment and nonuniform cooling induce complex residual stress patterns in the rail that can affect microstructure and fatigue crack growth rate behavior. This paper will describe research to examine defect growth behavior of modern rail steels. This research is a collaboration among several organizations: Thornton-Tomasetti, Arcelor-Mittal, Lehigh University, Harvard University, National Institute of Standards and Technology (NIST), Fraunhofer Institute, and Volpe. Arcelor Mittal donated rails with different grades of steel: advanced head hardened, head hardened, and standard strength (i.e. non-head-hardened). Lehigh conducted laboratory tests on specimens cut from these rails to perform various tests, which include: hardness measurements, mechanical testing to measure tensile properties, fracture toughness measurements, and fatigue crack growth rate tests. All of these tests were performed in accordance with applicable ASTM International standards. NIST and Fraunhofer performed preliminary neutron diffraction measurements of residual stresses on the different rails. Moreover, this paper will present results from the laboratory testing program. Implications of these results on detail fracture growth behavior will also be discussed.

For a given material and set of test conditions, fatigue crack propagation behavior can be described by the relationship between cyclic crack-growth rate, da/dN and the fluctuation of stress intensity factor, △K. Such test data are usually displayed in a log-log plot. At intermediate values of △K, fatigue crack-growth data fall along a straight line such that a power-law equation may be used as a curve-fit to the data. Various numerical techniques are applied in order to (1) derive the crack-growth rate and (2) determine the parameters for the power-law equation. Using data from laboratory tests conducted on rail steels, this paper explores the various numerical methods used to characterize fatigue crack-growth behavior. Tests were conducted using two different fracture-mechanics specimens (a standard compact tension specimen and a non-standard single edge notch specimen). Three different numerical techniques were applied to determine the fatigue crack-growth rate, da/dN from test data measuring crack length, a versus number of fatigue cycles, N: (1) secant method, (2) modified secant method, and (3) incremental polynomial method. Four different least squares regression analyses were then applied to determine the parameters for the power law. Moreover, the outcome of these analyses is to determine the combination of numerical techniques which yields the least amount of error when the crack-growth rate equation is integrated and compared to the original a versus N data. Fatigue life calculations performed by integrating the crack-growth rate equation demonstrate the sensitivity of predicted growth rates to the power-law parameters derived from the different regression analyses. This paper explores the various numerical methods and techniques employed to analyze fatigue crack growth data using test data on rail steels.

One of the Federal Railroad Administration’s (FRA’s) current areas of research within its rail integrity research program includes investigating the defect growth behavior of modern rail steels. The modern rail steel research is a collaboration among several organizations: Thornton-Tomasetti, Arcelor-Mittal, Lehigh University, Harvard University, and the Volpe National Transportation Systems Center (Volpe). A companion paper to this one will describe the results of recently-completed mechanical testing, fracture toughness testing, fatigue crack growth rate calculations, and residual stress field characterizations performed in Phase I of this research. The behaviors measured in Phase I were examined under laboratory conditions. The effects of the service load environment, including thermal loads, track support conditions, wheel loading, internal defect position and geometry will also need to be investigated for their effects on defect growth. A candidate approach that can be used to investigate these effects is to employ the finite element (FE) method to simulate a variety of conditions. Several of the types of measurements made in Phase I, such as residual stress distribution, serve as inputs to an FE model. Additional inputs, such as the wheel load and support conditions on the rail would be defined based on typical values encountered in the railroad environment. Stress intensity factors can be calculated around each simulated crack front for a given combination of material inputs, load conditions, and defect geometry. These stress intensity factors can then be used to estimate the fatigue crack growth rate under the given conditions. The modeling approach described above can result in a model that contains several complicated behaviors, including wheel-rail contact, discrete rail supports, and modeling techniques allowing the calculation of stress intensity factors. Further, several of these behaviors require specialized meshing techniques or analysis procedures. Thus, it is essential that the credibility of the model be established through a process of model validation. This paper lays out a framework for examining individual modeling techniques employed in the model, using a “building block” approach. Rather than trying to assess the entire model of a wheel on a discretely-supported rail containing an internal defect against a test measurement of the same conditions, the model is broken down into several key behaviors that must be verified. These distinct model behaviors, such as the method of discrete support, are then individually compared to known results to develop confidence in the simulation’s ability to produce physically-realistic results. In this way, confidence can be developed in the overall, complete model by developing confidence in several of the distinct modeling techniques that are employed in the overall model. The modeling techniques described in this paper include modeling the discretely-supported rail under a wheel load, modeling the internal defect as a crack, and using a submodeling technique to combine areas of coarse and fine mesh in a computationally-efficient manner.

Finite element analyses (FEA) were conducted in this paper to understand the underlying mechanisms contributing to a commonly observed failure mode of cut spikes used with the elastic fastening systems for wood ties. This failure mode features fatigue cracking development in the internal cross-sectional spike surfaces located approximately 1.5 inches below the top surface of a tie. Previous computational studies applied elastic material properties with “perfect” material behaviors. The study presented in this paper adopted post-elastic failure models for both the steel spike and wood tie materials, which proved key to reconstructing the observed failure mode in modeling. The commercial FE software Abaqus was employed in this study. Continuum FE models were developed for a single cut spike embedded in a wood tie. The steel spike was modeled to yield plastically upon reaching a yield strength limit. A user material subroutine documented by Abaqus was adopted to simulate the 3D orthotropic failure of the wood tie. Both the elastic properties and strength limits of the wood material were orthotropic, with the properties in the transverse direction significantly lower than those in the fiber direction. Different combinations of vertical, lateral and longitudinal forces were applied in the analyses, deforming the spike in various bending modes. The forces were increased in magnitude until the steel reached its yield strength (i.e., developed permanent plastic deformations), and the yielding locations were recorded and compared with the observed failure mode. The FEA showed that damage initiated in the wood tie being pressed by the spike with sufficiently large forces and that wood damage preceded steel yielding. The farther the wood material deteriorated from the top down, the lower the steel yielding location was in the spike shaft. Longitudinal forces were reacted to in the weaker transverse direction of the wood tie and therefore resulted in lower damage initiation forces and lower steel yielding locations than lateral forces did. It was concluded that the orthotropic wood tie failure condition and a substantial presence of the longitudinal force were necessary conditions for the spike to initiate failure at about 1.5 inches below the top surface of a tie. This corroborated the findings in a derailment investigation involving the spike failure. The lateral force alone unlikely caused this failure mode, but the presence of a lateral force on the spike appeared to decrease the magnitude of the longitudinal force needed to initiate damage in the spike.

The Federal Railroad Administration (FRA) routinely conducts investigations of railroad accidents to determine causation and any contributing factors to help the railroad industry implement corrective measures that may prevent similar incidents in the future. Over the past decade, FRA has investigated multiple broken rail accidents in which fractures in the rail web were identified. The common features observed in the recovered rail fragments from these accidents included welds and spots or burn marks on the web, indicating that the rails were joined together by pressure electric welding. Pressure electric welding uses a welding head that clamps around two opposing rail ends, pressing an electrode on each rail, then hydraulically pulling the rail ends together while arcing current through the electrodes into the rails, causing them to essentially melt together to form a continuous rail. Based on the similarities observed in the web fractures, FRA rail integrity specialists hypothesized that stray (i.e. inadvertent and unwanted) arcing during pressure electric welding can result in the formation of burns or pits on the rail where it makes contact with the electrodes. Moreover, these electrode-induced pits behave as stress raisers (also referred to as stress concentrations). Fatigue cracks often develop at locations of stress concentration. Once a fatigue crack initiates, the localized stress encourages the growth of the crack, which may potentially lead to rail failure. This paper describes the forensic evaluations of three railroad rails containing electrode-induced pitting. These evaluations include: magnetic particle inspection to nondestructively detect cracks emanating from the pitting; fractography to study the fracture surfaces of the cracks; metallography to study the microstructure; analysis of chemical composition; and measurements of tensile mechanical properties and fracture toughness of rail steel. Moreover, the results of these evaluations confirm the hypothesis postulated by FRA that stray arcing during pressure electric welding can cause electrode-induced pitting.

The fatigue crack growth rates of three vertical split rim (VSR) wheel fractures are calculated using the Paris-Erdogan fatigue crack growth equation. Initial crack length, final crack length were measured for three wheels that have experienced VSR failures. The results of the calculations indicate the lives after crack initiation are relatively short. The results of these calculations indicate the timely reduction of the number of VSR failures should emphasize the prevention of the initiation of the VSR cracks.

The bending of rail due to the repeated loading from railcar wheels is a known source of rail fatigue. If rail stresses are sufficiently high, they can initiate and propagate fatigue cracks after repeated cyclic loading such that they ultimately result in rail failure. Previous analyses of stresses from wheel loads have primarily focused on track beds for which the track stiffness is assumed uniform across a length of many cross-ties. In reality, however, spatial variations of track stiffness are known to exist and are affected by many factors such as the weather. These stiffness variations can lead to stresses that are locally higher than those predicted using models based on uniform average track stiffness alone. The work presented here is focused on the influence of spatial variations of track stiffness along the rail with respect to the maximum stresses generated. A computational model of a rail on a set of cross-ties with a statistically varying stiffness is used to study the maximum stresses generated when the track stiffness is not spatially uniform. The mean and standard deviation of the local track stiffness are varied and the maximum stresses at various positions within the rail are examined. This computational procedure is repeated for an ensemble of local track stiffness profiles to acquire the needed statistics of the corresponding stresses. These stresses are then related to crack initiation and the expected rate of crack propagation relative to the given the statistics of the track stiffness. This work is anticipated to have application for rail maintenance and the scheduling of rail defect inspections.

The KENTRACK program is a finite element based railway trackbed structural design program that can be utilized to analyze trackbeds having various combinations of all-granular and asphalt-bound layered support. It is applicable for calculating compressive stresses at the top of subgrade, indicative of potential long-term trackbed settlement failure. Furthermore, for trackbeds containing an asphalt layer, it is applicable for calculating tensile strains at the bottom of the asphalt layer, indicative of potential fatigue cracking. The program was recently expanded to include both English and international units. A procedure has been incorporated to provide a path to save results in a text formation in post-Windows XP operating systems. More importantly, properties of performance graded (PG) asphalt binders and the Witczak E* predictive model have been incorporated in the 4.0 Version of the program. Component layers of typical trackbed support systems are analyzed while evaluating the significance of layer thicknesses and material properties on design and predicted performance. The effect of various material parameters and loading magnitudes on trackbed design and evaluation, as determined and predicted by the computer program, are presented. Variances in subgrade modulus and axle loads and the incorporation of a layer of asphalt within the track structure have significant effects on subgrade vertical compressive stresses and predicted trackbed service lives. The parameter assessments are presented and evaluated using sensitivity analysis.

Increasing freight car loads demand higher performance tapered roller bearings. As the stress state on railway bearing applications continues to increase, further advancement in the modeling tools and methods used for subsurface contact stress evaluations are needed. Heat treat specifications and contact geometries for railway bearings were originally developed for ideal load conditions. However, in railroad applications, tapered roller bearings are exposed to a vast range of load conditions that are seldom perfect. Moreover, when comparing global rail markets, there are often differences in bearing loads, railcar wear conditions, maintenance practices, and reliability versus utilization expectations. Advanced modeling techniques need to be developed by bearing designers in order to meet the specific needs of each individual rail market. Prior research has shown that subsurface stresses, resulting from rolling contact, are the primary factor in the development of fatigue cracks in railway bearings. In addition, finite element modeling software has previously been used to analyze Hertzian contact stresses under rolling contact. Recent advancements in the technology and computational power of finite element methods can be used to numerically analyze more detailed simulations of complex geometries and biased load conditions in railway bearings. These improvements in the tapered roller bearing modeling methodology are necessary to determine the material, heat treat specifications, and geometry required to meet the demands of specific railway bearing applications. Furthermore, the specific risks associated with some common railway bearing design and modeling assumptions will be evaluated. An exploratory list of these assumptions include: line versus point contact, load deflection factor, zero contact angle, rigid body assumptions, linear material behavior, neglect for overload, and uniform loading on the bearing. Emphasis will be placed on potential improvements in the theoretical and finite element prediction of surface and subsurface stresses in railway bearings under rolling contact with a review of prior research on the subject.

Since the beginning of the railway industry, one of the most serious mechanical failures of critical components has been the fatigue failure of axles. About half of these fatigue failures initiate in the axle’s main body between the wheel seats due to surface damage and about half initiate in the wheel seat and journal filet areas due to corrosion and fretting. Fatigue cracks in the axle body almost always initiate at surface damage such as scratches, dents, and gouges (e.g., due to ballast impact, improper handling, etc.) where bending stresses are greatest. Efforts to improve axle fatigue life have focused on developing new, more fatigue-resistant steel grades; increasing the body diameter to reduce operating stresses; and modifying the properties (most notably near the axle surface) of the steel through heat treatment. The new process described in this paper improves an axle’s resistance to fatigue failure by imparting a refined microstructure and beneficial compressive residual stress to just the body without additional alloy enhancements or design changes. A 35% increase in fatigue strength and 47% increase in near-surface yield strength over standard Grade F axles have been obtained while also achieving modest improvements in ductility.

As part of an infrastructure subject to increased magnitude and frequency of loads, railroad track systems require regular inspection to assure high reliability for the safety of the public and passengers and the safe and efficient movement of goods. Unattended work hardened layers on the surface of the rails are operational hazards due to the inevitability of rolling contact fatigue cracks. A novel method utilizing the higher order surface waves, also called as Sezawa waves, has been used to detect and characterize the work hardened layer in rails. The technique involves generating Sezawa waves using contact ultrasonic transducers and monitoring the cut-off frequency of these waves. Results from hardness tests and metallographic analyses on the work hardened layers are also reported. The tests using the Sezawa wave technology demonstrated the ability to interrogate and resolve traffic-hardened layers in the depth ranges of 0–1 mm, 1–3 mm, and 4–7 mm. The minimum Brinell Hardness (HB) gradient, between the hardened layer and underlying rail, required to support Sezawa wave generation was also investigated and determined to be greater than 20 HB.

Acoustic emission monitoring of fracture-critical components of load-bearing structures such as railway bridges, provides a valuable input to condition-based maintenance planning and the allocation of maintenance resources. This paper presents a prototype wireless acoustic emission (AE) sensor network system for the structural health monitoring of aging steel railway bridges. It involves an AE sensor node, consisting of four AE channels and one strain gage input channel. Advances in data acquisition and communications instrumentation and the application of the fundamentals of fracture mechanics and artificial intelligence in signal processing have led to the development of a comprehensive, integrated system for acoustic emission monitoring. An active fatigue crack, when stressed due to the load of the on-coming freight train, emits AE. The strain gage channel senses the oncoming train and wakes up the AE channels from their hibernation mode. The wireless instrument digitizes the information, analyzes, compresses, and then transmits the filtered information to a central processing station for review by bridge engineers.

If the contact between the wheel and rail occurs at the flange, a high degree of slip is inherent. This can result in excessive wear and the potential for rolling contact fatigue cracking. Multi-body dynamics software is useful for predicting the bogie and track characteristics that may lead to flange contact occuring. However, there is a shortage of experimental tools available to validate such models or to assist in vehicle and track condition monitoring. In this study, the feasibility of a new approach is investigated. The method is based on an ultrasonic sensor mounted on the wheel. The concept is that the sensor emits an ultrasonic pulse which is designed to impinge on the wheel flange. If there is no contact the pulse is fully reflected back at the flange and picked up by the same sensor. If flange contact takes place, a proportion of the pulse amplitude will be transmitted into the rail. The signal reflected back to the sensor is therefore reduced. The amount by which this signal reduces indicates how much flange contact occurs. Test specimens were cut from sections of wheel and rail, and a 2MHz ultrasonic contact transducer was bonded onto the wheel in a position best suited to detect the flange contact. The specimens were pressed to together in a bi-axial loading frame to generate differing degrees of head and flange contact. The reflected signal was monitored as the normal and lateral loads were varied. It proved possible not only to detect the onset of flanging, but also to record a signal that varied monotonically with both normal and lateral applied load. A map of reflected ultrasound against the applied loading is presented.

North American freight car axles are one of the most robust mechanical components in railway service. Axles can be in service for many years (several decades in some cases) and undergo millions of fatigue cycles under heavy axle loads. Although for rotating bending fatigue the central axis is a zero stress location, larger internal central axis discontinuities will experience finite cyclic stress levels as the radial distance increases from the central axis. Larger internal discontinuities at the axle central axis, caused by forging bursts or unhealed pipe, are not a common occurrence; however they are irregularly shaped and represent a possible source of fatigue crack initiation. This paper describes efforts to quantify stresses at central axis discontinuities of various different sizes using finite element analysis (FEA) computer simulations. Also, the paper reviews the radial ultrasonic inspection of a large group of axles and subsequent destructive testing to examine the physical size of indications found using the ultrasonic techniques. Recommendations for radial ultrasonic testing of newly manufactured and second hand axles are provided, and changes to appropriate Association of American Railroads (AAR) axle specifications are outlined.

Broken joint bars have been identified as one of the major causes of main line derailments in the US. On October 2006, the US Federal Railroad Administration issued a federal regulation that mandates periodic inspections to detect cracks and other indications of potential failures in CWR joints [1]. The rule requires periodic on-foot inspection or an approved alternative procedure providing equivalent or higher level of safety. This paper describes a new machine vision-based system for joint bars inspection at speeds up to 70 mph. Four line-scan cameras mounted on a hi-railer or full size rail vehicle continuously capture high resolution images from both sides of each rail. An on-board computer system analyzes these images in real time to detect the joint bars. Each joint bar image is automatically saved and analyzed for visible fatigue cracks. The images can also be analyzed for missing bolts and other defects. When a potential defect is detected, the system provides audio warning, tags the image with GPS position, and displays the joint bar image with highlighted defects on the screen. The operator may confirm or reject defects. At the end of the survey, the operator can generate a survey report with the joint bar GPS location and types of all defects. This new system improves productivity and workers safety, inspecting joint bars from a moving vehicle instead of having to walk along highly transited tracks. It also allows the railroads to reduce the time between inspections, preventing defects to develop into hazards. Several tests have been performed on different rail roads showing system defect detection capabilities on both CWR and jointed track.