Rail and Component Structural Integrity
Rail Integrity
Defective rail is a leading cause of track related derailments. Rail and rail components such as joint bars experience large loads in service that can, in combination with residual stresses result in a sudden fracture under a train. Many defects begin inside the rail head and can only be detected with special ultrasonic or electromagnetic techniques. The prediction of where these defects may occur and how fast they will grow under service conditions can be used to determine appropriate inspection strategies. A typical internal rail head defect, called a detail fracture, appears in the photograph below.
In the course of its technical support for the FRA, the Volpe Center has overseen the development, validation and continued application of several engineering analysis tools:
Detail Fracture Growth (DFG) Model - The DFG model uses fracture mechanics principles to compute the safe crack-growth life of a rail head detail fracture under specified conditions (rail section, residual stress, train consist, track curvature, etc.) from the minimum detectable defect size to the size at which rail failure is expected under the next train. The life is computed in terms of million gross tons of traffic
Service Residual Stress (SRS) Model - The SRS model uses an advanced finite element technique to compute the residual stresses expected in a rail due to repeated application of service loads. The service loading is specified as an envelope of peak loads, and the model has the capability to compute residual stresses for peak loads having an unpredictable sequence of lateral locations (i.e., simulation of realistic wheel/rail contact effects). The most significant outputs of the SRS model are the location, magnitude, and extent of residual longitudinal tension in the rail head. These characteristics directly affect the safe crack-growth life of detail fractures as well as other types of transverse defects in the rail head
Rail Heat Treatment (RHT) Model - The RHT model uses conventional finite element technology to compute the temperature/time history and residual stresses developed by quenching heat treatments such as those employed in the manufacture of head-hardened rail. Rail metallurgical microstructure can be inferred from the temperature/time history output, and the residual stress output can be used as a specification of initial residual stresses for use with the SRS model
Rail Flaw Detection (RFD) Model - The RFD model uses the Monte Carlo method to simulate defect detection performance on a hypothetical railway line assumed to contain growing detail fractures subjected to periodic inspection by a rail flaw detection vehicle. The model inputs include seasonally adjusted defect growth rates, based on the DFG model, and a curve of detection probability versus defect size, based on empirical estimates from field experience. Limits on daily detector car mileage based upon track possession time and repair gang capability are also specified. The RFD model outputs include annual detected defects per track mile inspected, service/detected defect ratio, and detector car utilization (average miles inspected per day) for various specified annual tonnages and inspection frequencies
Mechanical Roller Straightening (MRS) Model - The MRS model simulates the mechanical effects of the roller straightening processes used for finishing rail to meet camber tolerances. Specifically, the MRS model computes the residual stresses due to roller straightening. The magnitude of these stresses can be used as an indicator of sudden fracture risk, and the detailed stress files can be used as input to the SRS model.
Computed results from these models have been correlated with available experimental data and field experience pertaining to past and present freight operations, past and present rail metallurgy, and present railroad practices of inspecting rail every 15 to 40 million gross tons.
Industry trends, however, have already begun to change which could affect rail integrity. Therefore, the efforts are underway to provide the technical basis to assist the FRA in developing pro-active initiatives to maintain safety in an environment of changing operational and track construction/maintenance practices. The engineering models are being applied to provide quantitative estimates of safety levels/risks associated with the changing conditions.
Heavy Axle Loads
Railroads are in the midst of transition from DC to AC/high-traction motive power, and some railroads also have increased the maximum freight car axle load from 33 to 39 tons. The higher freight car axle load has the potential to reduce defect safe crack-growth life (possibly necessitating more frequent inspection) by increasing the magnitude and/or extent of residual tension in the rail head. Loads from wheels of ac-powered locomotives have a similar potential due to the high tractive force component and also due to rail heating associated with wheel slip during train start-up. It is therefore prudent to assess the potential need for more frequent inspection of rail in service (in order to preclude the risk of excessive service defect rates) on track subjected to 39-ton axle loads and/or ac-powered locomotive operations.
The RHT, SRS, and DFG models will are applied to conduct a quantitative evaluation of the effects of these heavy axle loads on detail fracture safe crack-growth life. To the extent possible, correlations between crack growth characteristics observed under heavy axle load traffic at the Facility for Accelerated Service Testing (FAST) at the Transportation Technology Center, Inc. (TTCI) and the model predictions will be made.
Rail Grinding Practices
Some railroads have made significant extensions of rail wear service life by means of scheduled grinding programs that maintain profiles on worn rail heads. Current rail grinding practices cover the spectrum from frequent light grinding to infrequent heavy grinding, each railroad choosing its preferred practice, with no general industry consensus. An infrequent heavy grinding practice is followed mainly on high density main lines (100 million gross tons or more per year) in order to minimize disruption of revenue traffic.
Field experience on one such line circa 1990 strongly suggested that infrequent heavy grinding has the potential to decrease safe crack-growth life to an extent that the rail must be inspected for defects as much as five times more often than normal. (This extreme was in fact taken as a temporary action by the affected railroad.) An ad hoc analysis, conducted by the Volpe Center using the SRS model published in 1992, suggested that the rapid defect growth had resulted from high residual tension (located near the rail gage corner) due to a shift in the wheel/rail contact position following the grinding cycle. In view of the foregoing circumstances, it is prudent to examine the potential effect of grinding practice on safe crack-growth life in a comprehensive and consistent manner.
A quantitative evaluation of the effects of grinding using the SRS and DFG models is planned. As part of this work, a method of specifying wheel/rail contact loads and locations for different grinding patterns will be determined to formulate the load envelope inputs for the SRS model.
Alternative Inspection Procedures
Since 1994, two railroads have separately proposed test waivers for delayed remedial action on certain types of detected rail defects, including small detail fractures. The goal of delayed action is to improve safety by freeing the detector car from the constraints of the repair chase gang, so that the car can continue to search for medium and large defects. The Volpe Center exercised the DFG and RFD models to provide technical support for justification of these test waivers. The Volpe Center will provide similar technical support to the FRA in evaluating future waiver requests involving delayed remedial action or other alternative maintenance/inspection strategies.
Potential Benefit from Improved NDI
Current research sponsored by the FRA under the Small Business Innovative Research Program and projects at the Facility for Accelerated Service Testing (FAST) at the Transportation Technology Center, Inc. (TTCI) in Pueblo, CO, as well as other research being conducted independently by the railroad industry, is expected to lead to prototype equipment with improved NDI rail defect detection capabilities. The FRA and the AAR have developed the Rail Defect Test Facility for initial field trial of such equipment at TTCI.
The Volpe Center is observing these activities to stay abreast of hardware developments and provide technical support to FRA. Where possible, test data furnished by equipment suppliers or TTCI will be analyzed to estimate detection probability curves, and the capabilities thus implied will be evaluated by applying the RFD model. The Volpe Center will also participate in test planning and evaluation to assess the capabilities of new defect detection techniques and equipment.
The Volpe Center will provide technical expertise, as necessary, to evaluate and guide the development of promising new technology which may lead to new or improved techniques for rail inspection. A novel approach developed by an Australian railway for detecting rail surface fatigue from a moving vehicle and an electro-magnetic acoustic transducer (EMAT) system for the estimation of longitudinal force in continuous welded rail represent two examples of such technologies which may be included in this analysis.
Supplemental Evaluations of Rail Integrity
In 1999, TTCI initiated field testing at FAST to monitor the growth of detail fractures in modern rail which may have improved mechanical properties (e.g., better yield strength and hardness) than older rails. The Volpe Center provides technical support in coordinating and overseeing this testing at FAST.
The Volpe Center will coordinate work to determine residual stresses in the rails containing the defects and to determine basic fracture and fatigue properties of modern rail. At the conclusion of the FAST tests, the Volpe Center will correlate the defect growth data collected with results from the DFG model adjusted to include the effects the measured residual stresses and modern rail material properties.
Return to Top
Equipment and Component Structural Integrity
Engineering analyses and studies are conducted on critical components (e.g., wheels, bearings, and couplers) where structural failure could result in catastrophic consequences. The Volpe Center also provides ad hoc technical support in evaluating safety implications of rail vehicle equipment and component structural failures, evaluating industry proposals for revising equipment and component structural standards, and developing plans in response to requests for technical information following structural failure investigations.
Wheel Performance Studies
Wheel performance studies are conducted to provide information to the FRA's Working Group on Passenger Rail Equipment Safety Standards (PRESS) in developing specifications for braking systems and in evaluating the influence of braking operations on wheel crack development and life.
Research is conducted to develop predictive tools for estimating the residual stresses in railroad wheels in terms of design parameters (e.g., wheel diameter, rim thickness, plate type, and heat treatment) and operating variables (e.g., axle load, maximum speed, deceleration rate, and tread braking effort). Research includes the estimation of shakedown residual stresses in the wheel rim based on adaptation of an advanced finite element method originally developed under the Track Systems Research Program for the study of rail. The analysis tools have been applied to develop preliminary estimates of residual stresses in commuter vehicle wheels.
Estimation of Initial Manufacturing Stresses in Freight Wheels
The Volpe Center has developed computational tools to simulate the rim-quenching process used in the manufacture of commuter vehicle wheels. These simulations predict initial residual stresses that are introduced into wheels intentionally, as a part of the rim-quenching heat treatment during manufacture. This process establishes compressive residual circumferential (hoop) stress in the rim of the wheel making it resistant to the formation and propagation of radial cracks. Susceptibility to radial cracking increases if the stresses applied in service are severe enough to reverse the residual hoop compression to a state of neutral stress or hoop tension.
Preliminary results from the simulation of the quenching process for Multiple Unit (MU) vehicle wheels appeared to be in general agreement with experimental results for hoop stress measurements obtained using a "saw cut" technique. Modifications to the model were made to permit similar analyses of the manufacturing process undergone by freight car wheels. The results of these studies will serve as input to the shakedown residual stress software to estimate residual stresses in freight car wheels subjected to simulated service conditions.
Safe Performance Limits for Passenger Equipment Wheels
The shakedown analysis model will be integrated with the rim-quenching models and previously developed braking thermal stress models. These tools will be used to predict shakedown stress states in commuter vehicle wheels and a comparison of shakedown stress states for different combinations of design and operation will be conducted. These variants will be identified by a survey of US commuter railroad operations through a cooperative effort with the PRESS Wheel Design Working Group. The survey will yield the information required to apply the wheel residual stress models to investigate the range of operating parameters which exist on current US commuter fleets. These results will serve as baseline estimates from which the effects of potential changes to railroad operations (such as increased speed or stopping profile) may be evaluated. Performance criteria based on the appearance of tensile hoop residual stress in the rim (reversal of the as-manufactured residual hoop compression) will be applied to the results to derive rationally-based practical limits.
Application to Freight Car Wheels
With the advent of increased axle loads in the freight industry, the issue of service-induced residual stresses in freight wheels and their potential effects on premature failure must be acknowledged. The Volpe Center will extend the commuter wheel research program by conducting similar analyses for freight car wheels subjected to heavy axle load service. The base of information and methodologies developed from the MU wheel studies can be applied to evaluate safe operating limits for freight car wheels. The estimates of initial residual stresses following manufacture will be used as input to the analytical methodology.
Experimental Determination of Residual Stress in Railroad Wheels
Data for experimental calibration of the numerical results obtained in the above three tasks, was obtained from precision saw-cut tests on several passenger (MU) and freight wheels. Eight wheels (four MU wheels and four freight wheels, two of each in the as-manufactured condition, the other two having seen field service) were destructively tested. Moir interferometry was applied to obtain precise measurements of released strain. Displacement data of the cut opening (or closing) also was collected. The raw data was provided to the Volpe Center for use in model calibration. As part of the continuing joint research program between the Volpe Center and the Cracow University of Technology (CUT), an approach for reconstructing the full three-dimensional stress state in wheels has been developed. Application of this approach to the data collected during the destructive saw-cut experiments began in FY'98. A comprehensive report documenting application of the methodology to the actual laboratory data will be provided.
Supplemental Equipment and Component Safety Analyses
The Volpe Center supports FRA's Offices of Research and Development and Safety with scientific and engineering analysis to address safety concerns related to rail equipment and components on an as-needed basis. Examples of such activities include:
- Evaluation of Norfolk Southern Corporations request for waiver of the 125-trailer limit in RoadRailer® operations
- Evaluation of the Burlington Northern Santa Fe request for waiver of the commingling restriction on RoadRailer® operations.
Related Literature:
Return to Top
