The results will be documented in reports that will be disseminated to industry. Some of the results will provide the basis for future FRA inspection and maintenance regulations.
Project Descriptions
HAZMAT TRANSPORTATION SAFETY
The safe transport of hazardous materials is dependent on a broad range of factors: the integrity of structural components such as the stub sill and tank shell; the integrity of non-structural components such as fittings and gaskets; the likelihood and susceptibility of the tank car operations to human error; and the behavior under potential accident, collision and derailment scenarios. The goal of this project is to improve the safety and efficiency of the transport of hazardous materials by rail. To achieve this goal, the effect of operating, design, manufacturing, inspection, and repair practices on tank car safety must be understood. Furthermore, these aspects must be considered as a system to ensure that resources are directed toward the improvements that will have the greatest impact on safety.
Hazmat Shipment Routing
This project will investigate the broad area of hazmat shipments on U.S. railroads. The focus of this work will be to move forward from the extensive body of knowledge that exists within the railroad industry as a result of an ongoing effort to reduce the risk component of such shipments. The risk implications of restricting such shipments to certain classes of track, certain levels of train control systems, and low population density areas will be studied. Such studies will also identify the broader effects of such operational changes, such as the effect on system or line capacity for the shipment of all goods. The goal will be to develop a better understanding of the complex risk tradeoffs between various operational and technological approaches to reducing risk exposures from such commodity movements. For example, the concept of rerouting “high risk” movements around major metropolitan areas rather than through them via urban tunnels will be investigated. Specifically for cities such as New York and Baltimore, with known hazardous material exposure, alternatives that would either reroute traffic and/or build new track and tunnels will be studied.
Improved Understanding of Tank Car Operating Environment
A better understanding of the in-service loads to which tank cars are subjected is needed for several reasons. Severe loads that occasionally occur during service can lead to incidents of tank car structural damage or complete failure. The frequency and magnitude of such events is presently not well understood. Additionally, the accumulation of damage over time from many relatively small cyclic forces can also lead to structural damage to tank cars. The tank car industry is in the process of implementing damage tolerance fatigue analysis as a performance based strategy to prevent fatigue failures. Sensitivity studies have suggested that an accurate representation of the loads encountered by tank cars is essential to damage tolerance analyses. Yard impacts and forces on tank cars during in-train accidents are also not presently documented in terms of stresses throughout the load carrying members of the tank car.
Recent advances in microprocessor and telecommunications technology will be used to develop an instrumented car that is capable of being placed into revenue service to measure, record and transmit forces. This car will be used to develop additional data that will be combined with existing Freight Equipment and Environmental Sample Test (FEEST) data to provide a better understanding of the actual load spectrum seen by tank cars. The aim of this project is to gather and analyze such data. This information will be used to improve tank car structural integrity.
Similar research is required for forces developed under extreme impacts in accident conditions. This type data, if it exists, has not been compiled for ready reference. This data is important for use by designers in determining the likely sites of damage and action required. This project will gather this data from literature, industry databases, and application of engineering principles. This data will need to be updated periodically; therefore this will be an on-going project.
Railroad Transportation of Spent Nuclear Fuel
With the renewed interest in finding a repository for long-term storage of radioactive materials (RAM) and spent nuclear fuel, there is a need to revisit issues related to railroad transportation of these materials. Research will focus on the safe transportation by railroad using available accident environment analyses to determine potential forces that may be encountered by the railroad spent nuclear fuel casks if involved in an accident. Current research is focusing on a risk assessment of the safety of transporting spent nuclear fuel in regular freight service versus dedicated train service. This is an on-going project.
Tank Cars Greater than 263,000 pounds Gross Rail Load
The current regulations prohibit transportation of hazardous materials in tank cars with a gross rail load (GRL) greater than 263,000 pounds. However, the railroad industry is moving rapidly in the construction and operation of other railcars with a GRL of 286,000 pounds. Tank car builders and owners are presently submitting applications to the USDOT for the use of these increased GRL cars. The industry’s fatigue analysis is performed on these cars using the FEEST-1 or FEEST-2 loading spectrums. The scaling up method used to account for the increased GRL is a linear method. A peer consensus is that longitudinal and vertical coupler loads must be scaled up or changed to account for the increased gross weight of the cars. Before such a requirement may be considered or put in place, a study of appropriate scaling factors must be made. A more thorough understanding of changes in in-train forces with changes in GRL may be developed by a review of past FRA research involving the train operation simulator models, and the ‘ADAMS’ computer models, and translating those studies to the study of increased gross weight tank cars. The issue of buckling of adjacent light, normal, and heavy tank cars, will also be investigated. The study results will be used as a guide in evaluating future requests for the use of increased gross weight tank cars and railcars, or as a minimum, will direct R&D efforts for the safe transportation of such loads. This research is planned to continue through FY 2004.
Figure 4.8.1 Tank Car Simuloader
Assessment and Validation of the Methodologies for Engineering Reliability
Accidents, tank car structural failures, and the existence of defects in structural components of railroad tank cars, lead the FRA to believe that measures of reliability for tank car components must be defined and detailed reliability assessments on a variety of components should be performed. This type of assessment should be performed for each unique component design to define and document boundaries for the reliable use of each tank car. This project is planned to continue through FY 2003.
The reliability of the tank car may be defined as the probability that, when operating under stated environmental conditions, the tank car will perform its intended function adequately for a specified interval of time. To assess tank car reliability, different modes of tank car failure must be defined and categorized. Although complete and catastrophic failure is easily recognized, tank car performance as a safe packaging of hazardous materials can deteriorate and elements contributing to this deterioration (e.g., corrosion, cracks, pitting, fatigue, changes in material properties) need to be documented. Reliability functions, expected life, hazard functions, and failure rates must to be defined for tank cars. Since the reliability of the tank car will be a function of several design variables and parameters, developing a methodology for combining these random variables into a tank car “strength” function is necessary. The results of this assessment can provide tank car owners with quantitative information to define and document boundaries for the reliable use for each tank car, enabling them to implement guidelines for the maintenance and use of tank cars.
Tank Car Pressure Relief Devices
In the 1980's, the FRA and the tank car industry developed an analytical program for calculating the effects of fire on railroad tank cars for the purpose of selecting pressure relief device type, sizing, capacity, and pressure settings. These procedures were used to predict various parameters, such as the start-to-discharge time to failure, the residual amount of lading and pressure in the tank at time of failure, the time to reach this pressure level, etc. Further work to enhance the program was made in 1992. Additional work is being conducted to further refine and expand the software. This research should be completed in FY 2002.
Flow capacity and rating of tank car pressure relief devices, especially non-pressure or general-purpose cars, requires an understanding of the thermodynamic behavior of the product in fire conditions and the performance of the car. This also is the case for cars carrying “poison by inhalation” material. This project will identify parameters and rules to apply in the formulation of proper relief properties for this type of lading, similar to what has been done for pressure tank cars.
Shelf Coupler Redesign/Height Mismatch Effects
One safety feature found on all tank cars used in hazardous materials service is the double-shelf coupler. These couplers are designed to remain engaged when subjected to forces that occur during switching operations and train accidents.
Before the use of double-shelf couplers, a coupler would become uncoupled in accidents and during normal switching operations, ride up over the draft assembly of a tank car, and puncture the head of the tank. To understand the transmission of forces through the train and the effects of these forces on tank cars, both in normal operation and in an accident environment, a study of the double-shelf coupler is planned. This study, which would continue for several years, would include stress analysis to understand the load paths through the coupler and a look at the coupler/coupler interface and coupler/draft assembly interface as a dynamic system. After such a study, appropriate design and material changes to the shelf couplers may be based upon the known system.
Spills from tank car closures are significant safety concerns of tank car handlers, shippers, owners, and operators. This is mostly a human factor’s problem, but should nevertheless, be researched to diminish its effect on tank car integrity. Work will be initiated in FY 2003 to look at unintentional leaks and will continue for several years.
TANK CAR STRUCTURAL INTEGRITY
Structural integrity of tank cars is an important factor in hazmat transportation safety; as the most severe hazmat releases tend to result from a loss of structural integrity, either suddenly in an accident, impact, or derailment, or gradually due to damage accumulated in normal service. The goal of this project is to ensure that the capability of hazmat carriers to maintain structural integrity is understood. This understanding will allow performance-based requirements for design, manufacture, operations, and inspection to be incorporated into tank car regulations and practice, thereby improving the safety of hazmat transport by rail.
Models for Fatigue Damage in Tank Car Structures
One potential failure mechanism for tank car structures is the growth of a fatigue crack to a critical size. Fatigue crack failures can be prevented through periodic inspections. However, the prediction of fatigue crack growth rates is essential in order for appropriate inspection intervals to be determined. Fracture mechanics approaches to predicting fatigue crack growth rates have been employed by other industries, especially the aerospace industry, and can be adapted to the tank car industry. However, additional testing must be performed, and some technical advances might be needed to adapt available methodologies specifically to the tank car industry. For example, baseline fatigue crack growth properties, essential for predicting in-service crack growth rates, are not yet available for all steels used in the manufacture of tank car shells and stub sills. Models for predicting the load interaction effects on crack growth rates have been developed for the aerospace industry, but the tank carload spectrum is very different from the load spectrum for aerospace applications. The existing models will need to be evaluated, and perhaps extended, before they can be applied with confidence to tank car applications. The effect of manufacture, especially welding, on crack growth behavior must also be evaluated.
Effect of Welding on Structural Integrity
Tank car structures are joined by welding. Despite many advances in welding practice over the past decades, cracking due to fatigue or sudden impact is most likely to occur in the vicinity of a weld. Damage in the weld, heat affected zone, or parent material near the weld can be affected by changes in microstructure, residual stresses, or initial flaws that can be induced during manufacture or repair. Furthermore, welds often occur near edges and corners of the structure, or induce changes in cross section geometry, where stresses are naturally high. The effect of welding and stress relief practice on the location and magnitude of residual stresses in tank car structures will be studied through laboratory experiments and numerical modeling techniques. Laboratory tests on the effect of welds on fatigue crack growth rates will be performed. These results will be incorporated into models for crack growth in tank car structures. Advances in nanotechnology will also be evaluated for application to tank car materials. This project will continue through FY 2005.
Models for Damage in Accidents and Over-Speed Impacts
The structural integrity of a tank car shell must be maintained during extreme conditions such as impacts, collisions, derailments and over-speed impacts. Historically, punctures caused by impacts to tank car heads have been the cause of many hazardous materials releases. Results from full scale and part scale impact tests conducted in the 1970s and early 1980s led to regulations for head shields for certain hazardous material tank cars. Further work was conducted in the mid-1980s on tank cars for transporting chlorine and for aluminum tank cars, resulting in the development of a model to predict the puncture of a tank car head. These improvements have greatly reduced the occurrences of tank car head punctures.
Further efforts are needed to evaluate the current models for puncture resistance of tank car heads. The ability to use finite element models to simulate the non-linear (material and geometric) deformations of the tank car shell, jacket, and head shield during impact has improved significantly in recent years. This allows the relative puncture resistance of different designs to be evaluated. However, the deformation at which failure actually occurs is less well understood. Failure models for steels have been developed for the U.S. Navy and the nuclear power industry, including the effect of lading (liquid or pressure) on global deformation behavior. Adapting these models to the tank car environment would lead to a better understanding of failure mechanisms that occur during head impacts and other accident scenarios, and improve the capability to assess the condition of tank cars damaged in accidents. This project is expected to continue for several years.
Tank Car Steels
Materials authorized for use in the construction of tank cars are specified in the regulations. The objective of this project is to develop a performance-oriented specification for steel used in tank car construction. This project will develop a parametric model for the determination and selection of steel for tank car construction. Past research showed that tank car steel properties, for tank car steels in similar service, were not consistent, in that they did not follow definite patterns. A compendium report will be prepared. This project will continue through FY 2004.
Intermodal Tank Integrity
Use of intermodal portable tanks for hazmat is increasing. The containment integrity of such tanks will be reviewed for safety when they are used in railroad service. Enhancements have been made to the integrity of intermodal tanks, but intermodal tank design criteria and performance need to be evaluated in terms of international transportation requirements and size and weight limitations. This research is planned for FY 2002 and is expected to continue for 3-4 years.
Thermal Integrity of Tank Cars
As tank cars age, thermal protection systems deteriorate and insulation materials slump. Current regulations require qualification of thermal systems periodically. However, research is necessary to ensure there is no negative effect on the overall tank car thermal integrity. This research is planned to begin in FY 2003 and continue for several years.
DAMAGE ASSESSMENT AND IMPROVED INSPECTION SYSTEMS
An accurate assessment of the condition of a tank car is essential to the safe transport of hazardous materials. Damage or wear that has resulted from normal operation must be identified at scheduled inspections before such damage leads to failures or releases. Also, the condition of any car that has been damaged in an accident or derailment must be known before a determination of the most appropriate action to ensure the safety of the public, railroad employees, and emergency responders can be made. The goal of this project is to develop, improve, and quantify the capability to assess the condition of tank cars, in repair shops and at accident sites.
In its report on Ensuring Tank Car Safety
, the TRB recommended that USDOT should continue to work closely with industry to identify methods for verifying the structural integrity of in-service tank cars, including nondestructive evaluation (NDE) test methods to supplement or replace existing test requirements. Further, the TRB recommended that results from the inspections and tests should be routinely collected to monitor tank car condition, improve test and inspection methods, and enhance tank car design, maintenance, and repair standards. The following projects address these issues.
Quantification of Effectiveness of Localized Non-Destructive Inspection Techniques
Under the FRA regulations, tank car owners are required to employ periodic structural integrity inspections, including tank shell thickness tests and inspections of tank car welds. By limiting the required inspection to known areas of crack initiation, RSPA and FRA expect an increase in the probability of defect detection, as well as an improvement in the reliability of the inspection results and a reduction in the inspection costs. As part of this rule, five NDE methods are presently authorized: dye penetrant, radiography, magnetic particle, ultrasonic, and aided visual inspection. Other NDE methods may be used by RSPA exemption, such as acoustic emission and direct visual inspection. This rule also requires tank car repair facilities to document the sensitivity and reliability of the nondestructive evaluation methods used for the structural integrity inspections. The FRA is working with the AAR in evaluating the probability of detection (POD) for the various non-destructive inspection techniques under controlled laboratory settings. The capability of these techniques will be assessed by studying statistical trends of cracks found in repair shops using the various non-destructive inspection techniques.
Figure 4.8.2 Tank Car NDE Inspection
Acoustic Emission Inspection
With the large number of welds to be inspected on tank cars to meet the tank car structural integrity requirements, development of improved global NDE will significantly improve the probability of finding defects. Presently, acoustic emission evaluation of tank cars is providing not only information about the welds that must be inspected under the new requirements, but also any defects on any part of the tank shell. Further evaluations of acoustic emission technologies, as well as the development of other highly sensitive global NDE methods will increase the number of defects detected.
Acoustic emission tank car inspection techniques will continue to be investigated, both for periodic inspection, and inspection after damage or after involvement in accidents. Present focus will be on the source location, improved detection interpretation capability, and quantification of the effectiveness of the procedure. This effort is planned to continue through FY 2002.
Tank Car Damage Assessment/Improved Emergency Response
A tank car a
