Volpe Center: Information Resources - Volpe Journal Spring 98

Volpe Journal Spring 98

Previous Section | Contents | Next Section

Rail Equipment Crashworthiness Research

On a passenger per mile basis, railroads are a safe way to travel. But proposed high-speed trains will change the operating conditions for riders on the rails, and Congress has appropriated funds for research into ways of making both high-speed and conventional trains even safer. Led by Senior Engineer David Tyrell, the Volpe Center is studying ways to ensure that passengers survive a railroad crash.

Throughout the country, passenger train speed and passenger traffic are increasing. Amtrak is purchasing high-speed trainsets that will travel at speeds up to 150 mph on the same tracks as lower-speed commuter and freight trains. The Massachusetts Bay Transportation Authority (MBTA) has recently reopened the Old Colony line from Boston to the south shore. Maryland Area Rail Commuter (MARC) Service is purchasing equipment to run commuter trains at 125 mph. The state of Washington has purchased Talgo trainsets originally developed for service in Spain. The state of Florida is developing a high-speed rail system, based on the French Train Grande Vitesse (TGV) system, in which the trains will run at speeds up to 200 mph. (View Photo: Amtrak High-Speed Train)

BACKGROUND
Crashworthiness of rail equipment is an important issue because of increased traffic, which can increase the likelihood of train collisions; increased equipment speed, which can increase the severity of train collisions; and the application of equipment developed for foreign operating environments, which generally includes smaller and lighter freight equipment.

In support of the Federal Railroad Administration's (FRA) Office of Research and Development, the Volpe Center has been conducting research into rail equipment crashworthiness. It has proposed strategies to improve crashworthiness and to apply analytic tools and test techniques that evaluate the effectiveness of those strategies. The FRA and Amtrak have used information from this research to develop the crashworthiness requirements for the high-speed trainsets, and the FRA also has used this information to develop initial regulations for passenger equipment. The American Public Transit Association (APTA) currently is using this information to develop commuter and intercity industry standards and recommended practices for passenger equipment crashworthiness. The FRA is planning to expand and update its passenger equipment regulations in several years, and the information now being developed through this research is expected to provide the technical basis for the updated crashworthiness regulations.

INTRODUCTION
Equipment crashworthiness principally depends on two factors:

First, a rail car must maintain sufficient space for the occupants to survive the collision, i.e., the occupants must not be crushed by the collapse of the rail car or by an object penetrating the occupant volume from outside.

Second, the forces and decelerations imposed on the occupants must remain within survivable limits.

For nearly 100 years, proposals have been made to include crush zones in rail equipment to control occupant volume deceleration, to increase occupant volume strength through car body structural design strategies, to prevent occupants' direct impact with the interior through passengers' use of lap and shoulder belts, and to implement other concepts for improved rail equipment crashworthiness. Until recently, fully developing and evaluating effective concepts has been time consuming and expensive. Rail equipment is expensiveit costs about $2 million to purchase a single rail passenger coach carwhich has precluded the widespread use of destructive experimental techniques such as those used by the automotive industry. The aerospace industry has developed detailed computer modeling programs that can simulate transportation equipment collisions and other conditions causing large structural deformations. These programs have been available for more than ten years, but affordable computer equipment capable of exercising detailed models of train collisions has been available only for about five years.

Modern computers and computer-aided engineering tools evaluate the effectiveness of proposals to increase rail equipment crashworthiness and the development of those concepts that increase crashworthiness. Crush zones are now being incorporated in passenger equipment, including the French double deck TGV high-speed trainset, the British 465 Networker commuter train, and the American FlyerAmtrak's high-speed trainset.

Researchers at the Volpe Center have focused on three objectives:

Determining the collision conditions that are of greatest concern.
Evaluating the effectiveness of current equipment in preserving the occupant volume and in limiting the forces and decelerations the occupants experience.
Developing modifications to existing equipment and new design approaches for improved crashworthiness.

To implement its approach to rail equipment crashworthiness research, the Center has been extending and applying modern analytic tools and experimental techniques. Figure 1 shows a finite-element analysis mesh for the end of a passenger coach car. Models such as these have been used in DYNA3D and ABAQUS, powerful finite-element solvers originally developed for the aerospace industry. These finite-element solvers can account for large deformations of structures and nonlinear material properties, which allow them to be used to determine the crush behavior of individual cars. (View Figure 1)

Figure 2 shows a computer simulation model of an occupant seated in a passenger coach car. Volpe Center research-ers used this simulation in a program called MADYMO, which was initially developed to evaluate the collision-related characteristics of automobile interiors. In addition to using MADYMO to predict how interior configurations affect train occupants, Center researchers also made actual measurements using instrumented crash-test dummies known formally as Hybrid Ill Anthropomorphic Test Devices (ATD). The dummies were seated in a portion of a coach car's interior that was mounted on a pneumatic "sled" and pulsed to simulate a collision. The effects measured on the dummies during dynamic sled testing agreed closely with the predictions made with MADYMO. (View Figure 2)

COLLISION CONDITIONS OF CONCERN
Train collisions can be grouped into four categories:

Derailments, such as that which occurred in August 1994, when an Amtrak train derailed in Batavia, NY and rolled down an embankment.
Grade crossing collisions, such as occurred in May 1995 when an Amtrak train collided with a heavy highway vehicle in Sycamore, SC.
Collisions between structurally similar trains, such as occurred in January 1993 when two Northern Indiana Commuter Transportation District commuter trains collided in Gary, IN.
Collisions between structurally different trains, such as occurred in August 1981 when an MBTA commuter train collided with a freight train in Beverly, MA.

Of particular concern have been accidents such as the Gary, IN, accident, the February 1996 collision between a MARC commuter train and an Amtrak train; and the February 1996 collision between two New Jersey Transit (NJT) Corporation trains in Secaucus, NJ.

In all of these accidents, a cab car was at least one of the colliding cars. A cab car is similar to a coach car, with passenger seats throughout the car, but with an operator's control stand at one end. MARC and NJT practice, as well as MBTA practice and that of most other operating authorities with similar equipment, is to pull the equipment away from the city center with the operator in the locomotive, and to push the train into the city center with the operator in the cab car.

Figure 3 shows the NJT cab car after the Secaucus accident. The accident occurred when the cab car operator ran past a stop signal and started through a switch onto the mainline track. The cab car made it part way through the switch when it was struck by a locomotive-led train coming in the opposite direction on the mainline track. The collision ripped off the right front corner of the cab car, damaging the operator's compartment and crushing the first five rows of seats. The operator of the cab car was killed, as was a passenger who was seated in one of the first five rows. The other train's operator, who was seated in the locomotive, was killed when a piece of the cab car penetrated the locomotive's windshield. (View Figure 3)

STRUCTURAL CRASHWORTHINESS
The Volpe Center has been leading efforts to evaluate concepts for improved cab car crashworthiness.

A study of the effectiveness of design modifications to an existing cab car end structure, shown in Figure 4, was conducted by the Center with contractor support. Mr. Tyrell, with the support of Kristine Severson and A. Benjamin Perlman, developed a collision dynamics model that describes the motions of a cab carled train and a locomotive-led train when they collide. (View Figure 4)

A team of engineers at Arthur D. Little, Inc. (ADL) led by Dr. Ron Mayville determined the crush behavior of a cab car. Using the data developed from the ADL portion of the effort in the collision dynamics model, the Volpe Center was able to determine the effectiveness of the design modifications in preserving the operator and passenger volumes under collision conditions similar to the Secaucus collision. The principal conclusion of the study was that the end structure of a cab car could be modified to protect all the passengers in such a collision, but, with strength limitations of the main car structure, the end structure could not be modified to protect the operator.

By utilizing contractor support and in-house expertise effectively, the Center was able to provide the FRA with the information it required in a timely fashion. ADL, with direction from Mr. Tyrell, is currently conducting a study to determine the effectiveness of completely different cab car structure designs.

The Volpe Center is directing several other studies of rail equipment structural crashworthiness:

ADL is studying the mechanics of how one car can override another during a train collision. When this happens, the stronger underframe of the overriding car can crush a large portion of the weaker superstructure of the overridden car. The study has two aspects: means of preventing override, such as locking mechanisms that engage when the cars bang together; and means of limiting the damage caused by override, such as strong end structures that transmit the loads to the underframe of the overridden car.

Another project currently underway, also with the assistance of ADL, is a preliminary design analysis of rail car structures that incorporate crush zones. The research is determining limitations on crush zone length, and estimating the amount of energy the crush zones can absorb.

Center technical staff is currently planning tests of rail car structural elements, substructures, and cars. The Southeastern Pennsylvania Transportation Authority, NJT, and MARC all have contributed retired commuter cars for testing purposes.

INTERIOR CRASHWORTHINESS
In addition to its work on improving the crashworthiness of the rail car structure, the Volpe Center has been investigating ways to reduce the forces and decelerations occupants experience during train collisions. This effort involves modeling and testing occupant motions in different interior configurations during simulated train collisions. Mr. Tyrell and Ms. Severson developed the test plan and alternated monitoring the tests. Figure 5 is a photograph of the setup used for the dynamic sled tests. (View Figure 5) Other organizations have contributed significantly to this effort:

The Center's Crashworthiness Division, which performs automobile crashworthiness research for the National Highway Traffic Safety Administration (NHTSA), assisted in arranging the testing of coach car seats at MGA Research. The Crashworthiness Division also provided advice in setting up the computer models of occupant motions during train collisions.
Amtrak donated the seats used in testing.
Federal Aviation Administration personnel shared their experiences in developing a methodology to evaluate the crashworthiness of transport aircraft passenger seats.
NHTSA loaned instrumented test dummies.

The Center is currently directing a number of studies of interior crashworthiness:

Testing the seats that Amtrak plans to use on the high-speed trainset.
Analyzing how occupant size affects the likelihood of fatality for occupants seated facing forward.
Awarding a contract to Foster-Miller, Inc., to study the implementation of lap and shoulder belts in passenger trains, both intercity and commuter.

INFORMATION DISSEMINATION
The information developed under this research program is disseminated through a wide range of channels, including a number of government technical reports; technical papers published by the American Society of Mechanical Engineers (ASME) and the Transporta-tion Research Board (TRB); and verbal presentations at ASME and TRB meetings, the World Railroad Research Congress, and FRA outreach meetings.

Volpe Tech Day, which was held March 10, 1997, at the Canon Congressional Office Building in Washington, DC, included a display on the Center's rail safety program for rail equipment, including a description of the approach, methodology, and tools used to perform this research. In June 1996, the Center hosted the Symposium on Rail Vehicle Crashworthiness, which it had organized with support from Camber Corporation. Invited speakers included representatives from foreign governmental agencies, including Socit Nationale des Chemin de Fer Franais (French National Railways) and British Rail Research; from agencies within the U.S. Department of Transportation, including the Federal Railroad Administration, the National Highway Transportation Safety Administration, and the Federal Aviation Administration. Among the nongovernment speakers were representatives from foreign and domestic rail equipment suppliers, including General Motors ElectroMotive Division, General Electric Transportation Systems, Bombardier, and GEC Alsthom; as well as domestic rail transportation operators, including Amtrak.

In addition to performing crashworthiness studies, the Volpe Center is providing support to the FRA in a broad range of technical areas, including fire safety, emergency preparedness, gradecrossing safety, track safety, equipment component safety, vehicle-track interaction, human factors, and train control. The Center provides the FRA with the information it needs by drawing not only on the capabilities of its talented staff, but also on the collective expertise of contractors, academics, and staff contacts within the industry, foreign governments, and other agencies of the Federal Government. Volpe Center participants in the rail equipment crashworthiness studies are as follows:

David Tyrell, a senior mechanical engineer with the Volpe Center's Structures and Dynamics Division, leads the research in rail equipment crashworthiness. Mr. Tyrell develops research plans, carries out the plans with staff and contractor support, formulates recommendations to the sponsor based on research results, and presents and discusses recommendations with industry and government groups.
Kristine Severson, a mechanical engineer in the same division, develops and exercises models of human/interior interaction, rail car crush, and train collision mechanics. Ms. Severson also evaluates test data and assists in planning the research and presenting the results to the sponsor and the industry.
Brian Marquis, a mechanical engineer in the division, develops and exercises analytic models.
A. Benjamin Perlman, a part-time Volpe Center employee and full-time professor at Tufts University, Medford, MA, assists in developing the models and in verifying the analytic predictions and test results.
Dr. Herbert Weinstock, chief of the division, provides general supervision and guidance.

Resources
Bing, A., 1993, Collision Avoidance and Accident Survivability, Volume 1: Collision Threat, DOT/FRA/ORD-93-02.I, FRA, U.S. Department of Transportation.

Harrison, J., 1993, Collision Avoidance and Accident Survivability, Volume 1: Collision Threat, DOT/FRA/ORD-93-02.I, FRA, U.S. Department of Transportation.

Galganski, R. A., 1993, Collision Avoidance and Accident Survivability, Volume 3: Accident Survivability, DOT/FRA/ORD-93-02.III, FRA, U.S. Department of Transportation.

Mayville, R. A., Stringfellow, R. G., Rancatore, R. J., Hosmer, T. P., 1995, Locomotive Crashworthiness Research, Volumes 1 through 5, DOT/FRA/ ORD-95/8.1-8.5.

Tyrell, D. C., Severson, K. J., Marquis, B. J., 1995, "Evaluation of Selected Crashworthiness Strategies for Passenger Trains," Transportation Research Record 1489, Rail, pp. 5058.

Tyrell, D. C., Severson, K. J., Marquis, B. J., 1995, "Analysis of Occupant Protection Strategies in Train Collisions," ASME International Mechanical Engineering Congress and Exposition, AMD-Vol. 210, BED-Vol. 30, pp. 539557.

Tyrell, D. C., Severson, K. J., Marquis, B. J., 1995, "Train Crashworthiness Design for Occupant Survivability," ASME International Mechanical Engineering Congress and Exposition, AMD-Vol. 210, BED-Vol. 30, pp. 5974.

Tyrell, D. C., Severson, K. J., 1996, "Crashworthiness Testing of Amtrak's Traditional Coach Seat," DOT/FRA/ ORD-96/08, FRA, U.S. Department of Transportation.

Tyrell, D. C., Severson, K. J., Mayville, R. A., Stringfellow, R. G., Berry, S., Perlman, A. B., 1997, "Evaluation of Cab Car Crashworthiness Design Modifications," Proceedings of the 1997 IEEE/ASME Joint Railroad Conference, IEEE Catalog Number 97CH36047.

Contributor: David Tyrell

Return to Top