Passenger Equipment Safety Standards; Front End Strength of Cab Cars and Multiple-Unit Locomotives
Federal RegisterJan 8, 2010
Ask Donna
What actually matters in this document.
Text
DEPARTMENT OF TRANSPORTATION
Federal Railroad Administration
49 CFR Part 238
[Docket No. FRA-2006-25268, Notice No. 2]
RIN 2130-AB80
Passenger Equipment Safety Standards; Front End Strength of Cab Cars and Multiple-Unit Locomotives
AGENCY:
Federal Railroad Administration (FRA), Department of Transportation (DOT).
ACTION:
Final rule.
SUMMARY:
This final rule is intended to further the safety of passenger train occupants by amending existing regulations to enhance requirements for the structural strength of the front end of cab cars and multiple-unit (MU) locomotives. These enhancements include the addition of requirements concerning structural deformation and energy absorption by collision posts and corner posts at the forward end of this equipment. The requirements are based on standards specified by the American Public Transportation Association (APTA). FRA is also making clarifying amendments to existing regulations for the structural strength of passenger equipment and is clarifying its views on the preemptive effect of this part.
DATES:
Effective Date:
This final rule is effective March 9, 2010. Petitions for reconsideration of this final rule must be received not later than February 22, 2010.
ADDRESSES:
Any petition for reconsideration of the final rule should reference Docket No. FRA-2006-25268, Notice No. 2, and be submitted by any of the following methods:
•
Federal eRulemaking Portal.
Go to
http://www.regulations.gov
. Follow the online instructions for submitting comments.
•
Mail:
Docket Management Facility, U.S. Department of Transportation, 1200 New Jersey Avenue, SE., West Building Ground Floor, Room W12-140, Washington, DC 20590.
•
Hand Delivery:
Docket Management Facility, U.S. Department of Transportation, 1200 New Jersey Avenue, SE., West Building Ground Floor, Room W12-140, Washington, DC, between 9 a.m. and 5 p.m. Monday through Friday, except Federal holidays.
•
Fax:
202-493-2251.
Instructions:
Note that all petitions for reconsideration received will be posted without change to
http://www.regulations.gov
, including any personal information provided. Please
see
the Privacy Act heading, below.
Docket:
For access to the docket to read background documents, comments, or petitions for reconsideration received, go to
http://www.regulations.gov
anytime, or to the Docket Management Facility, U.S. Department of Transportation, West Building Ground Floor, Room W12-140, 1200 New Jersey Avenue, SE., Washington, DC, between 9 a.m. and 5 p.m., Monday through Friday, except Federal holidays. Follow the online instructions for accessing the dockets.
FOR FURTHER INFORMATION CONTACT:
Gary G. Fairbanks, Specialist, Motive Power and Equipment Division, Office of Railroad Safety, RRS-14, Mail Stop 25, Federal Railroad Administration, 1200 New Jersey Avenue, SE., Washington, DC 20590 (telephone 202-493-6282); Eloy E. Martinez, Program Manager, Equipment and Operating Practices Division, Office of Railroad Development, Federal Railroad Administration, 55 Broadway, Cambridge, Massachusetts 02142 (telephone 617-494-2599); or Daniel L. Alpert, Trial Attorney, Office of Chief Counsel, Mail Stop 10, Federal Railroad Administration, 1200 New Jersey Avenue, SE., Washington, DC 20590 (telephone 202-493-6026).
SUPPLEMENTARY INFORMATION:
Table of Contents for Supplementary Information
I. Statutory Background
II. Proceedings to Date
A. Proceedings To Carry Out the Initial 1994 Rulemaking Mandate
B. Key Issues Identified for Future Rulemaking
C. RSAC Overview
D. Establishment of the Passenger Safety Working Group in May 2003
E. Establishment of the Crashworthiness/Glazing Task Force in November 2003
F. Development of the NPRM Published in August 2007
G. Development of This Final Rule
III. Technical Background
A. Predominant Types of Passenger Rail Service
B. Front End Frame Structures of Cab Cars and MU Locomotives
C. Accident History
D. FRA and Industry Standards for Front End Frame Structures of Cab Cars and MU Locomotives
E. Testing of Front End Frame Structures of Cab Cars and MU Locomotives
1. FRA-Sponsored Dynamic Testing in 2002
a. Test Article Designs
b. Dynamic Impact Testing
c. Analysis
2. Industry-Sponsored Quasi-Static Testing in 2001
a. Test Article Design
b. Quasi-Static Testing
c. Analysis
3. FRA-Sponsored Dynamic and Quasi-Static Testing in 2008
a. Test Article Design
b. Dynamic Testing of a Collision Post
c. Quasi-Static Testing of Collision and Corner Posts
d. Analysis
F. Approaches for Specifying Large Deformation Requirements
G. Crash Energy Management and the Design of Front End Frame Structures of Cab Cars and MU Locomotives
H. European Standard EN 15227 FCD, Crashworthiness Requirements for Railway Vehicle Bodies
IV. Discussion of Specific Comments and Conclusions
A. Technical Comments
1. Crash Energy Management
2. Dynamic Performance Requirements
3. Alternative Corner Post Requirements for Designs With Stepwells
4. Use of Testing and Analysis To Demonstrate Compliance
5. Submission of Test Plans for FRA Review
6. Whether the Requirements Affect Vehicle Weight
7. System Safety
8. Other Comments
B. Preemption
1. Whether FRA Characterized Its Views on Preemption as the RSAC Consensus
2. Whether FRA's Views Are Consistent With 49 U.S.C. 20106, as Amended
3. Whether FRA's Views on Preemption Affect Safety
4. Whether FRA's Views on Preemption Affect Recovery for Victims of Railroad Accidents
5. How a State May Act as the Owner and Not the Regulator of a Railroad
6. How State Regulation of Push-Pull Operations Is Preempted
7. Whether It Was Necessary To Discuss Preemption in the NPRM
8. Whether FRA Has Authority To Express Its Views on Preemption
9. What Impelled FRA's Views on Preemption
10. Whether FRA's Views on Preemption Affect FELA
11. Whether Preemption Applies Under the Locomotive (Boiler) Inspection Act
V. Section-by-Section Analysis
VI. Regulatory Impact and Notices
A. Executive Order 12866 and DOT Regulatory Policies and Procedures
B. Regulatory Flexibility Act and Executive Order 13272
C. Paperwork Reduction Act
D. Federalism Implications
E. Environmental Impact
F. Unfunded Mandates Reform Act of 1995
G. Energy Impact
H. Trade Impact
I. Privacy Act
I. Statutory Background
In September of 1994, the Secretary of Transportation (Secretary) convened a meeting of representatives from all sectors of the rail industry with the goal of enhancing rail safety. As one of the initiatives arising from this Rail Safety
Summit, the Secretary announced that DOT would begin developing safety standards for rail passenger equipment over a five-year period. In November of 1994, Congress adopted the Secretary's schedule for implementing rail passenger equipment safety regulations and included it in the Federal Railroad Safety Authorization Act of 1994 (the Act), Public Law 103-440, 108 Stat. 4619, 4623-4624 (November 2, 1994). Congress also authorized the Secretary to consult with various organizations involved in passenger train operations for purposes of prescribing and amending these regulations, as well as issuing orders pursuant to them. Section 215 of the Act is codified at 49 U.S.C. 20133.
II. Proceedings to Date
A. Proceedings To Carry Out the Initial 1994 Rulemaking Mandate
The Secretary delegated these rulemaking responsibilities to the Administrator of the Federal Railroad Administration,
see
49 CFR 1.49(m), and FRA formed the Passenger Equipment Safety Standards Working Group to provide FRA with advice in developing the regulations. On June 17, 1996, FRA published an advance notice of proposed rulemaking (ANPRM) concerning the establishment of comprehensive safety standards for railroad passenger equipment.
See
61 FR 30672. The ANPRM provided background information on the need for such standards, offered preliminary ideas on approaching passenger safety issues, and presented questions on various passenger safety topics. Following consideration of comments received on the ANPRM and advice from FRA's Passenger Equipment Safety Standards Working Group, FRA published an NPRM on September 23, 1997, to establish comprehensive safety standards for railroad passenger equipment.
See
62 FR 49728. In addition to requesting written comment on the NPRM, FRA also solicited oral comment at a public hearing held on November 21, 1997. FRA considered the comments received on the NPRM and prepared a final rule establishing comprehensive safety standards for passenger equipment, which was published on May 12, 1999.
See
64 FR 25540.
After publication of the final rule, interested parties filed petitions seeking FRA's reconsideration of certain requirements contained in the rule. These petitions generally related to the following subject areas: Structural design; fire safety; training; inspection, testing, and maintenance; and movement of defective equipment. To address the petitions, FRA grouped issues together and published in the
Federal Register
three sets of amendments to the final rule. Each set of amendments summarized the petition requests at issue, explained what action, if any, FRA decided to take in response to the issues raised, and described FRA's justifications for its decisions and any action taken. Specifically, on July 3, 2000, FRA issued a response to the petitions for reconsideration relating to the inspection, testing, and maintenance of passenger equipment, the movement of defective passenger equipment, and other miscellaneous provisions related to mechanical issues contained in the final rule.
See
65 FR 41284. On April 23, 2002, FRA responded to all remaining issues raised in the petitions for reconsideration, with the exception of those relating to fire safety.
See
67 FR 19970. Finally, on June 25, 2002, FRA completed its response to the petitions for reconsideration by publishing a response to the petitions for reconsideration concerning the fire safety portion of the rule.
See
67 FR 42892. (For more detailed information on the petitions for reconsideration and FRA's response to them, please
see
these three rulemaking documents.) The product of this rulemaking was codified primarily at 49 CFR part 238 and secondarily at 49 CFR parts 216, 223, 229, 231, and 232.
Meanwhile, another rulemaking on passenger train emergency preparedness produced a final rule codified at 49 CFR part 239.
See
63 FR 24629 (May 4, 1998). The rule addresses passenger train emergencies of various kinds, including security situations, and requires the preparation, adoption, and implementation of emergency preparedness plans by railroads connected with the operation of passenger trains. The emergency preparedness plans must include elements such as communication, employee training and qualification, joint operations, tunnel safety, liaison with emergency responders, on-board emergency equipment, and passenger safety information. The rule requires each affected railroad to instruct its employees on the applicable provisions of its plan, and the plan adopted by each railroad is subject to formal review and approval by FRA. The rule also requires each railroad operating passenger train service to conduct emergency simulations to determine its capability to execute the emergency preparedness plan under the variety of emergency scenarios that could reasonably be expected to occur. In addition, in promulgating the rule, FRA established specific requirements for passenger train emergency systems,
e.g.,
to mark all emergency window exits and all windows intended for rescue access by emergency responders, to light or mark all door exits intended for egress, to mark all door exits intended for rescue access by emergency responders, and to provide instructions for the use of such exits and means of rescue access.
B. Key Issues Identified for Future Rulemaking
Although FRA had completed these rulemakings, FRA had identified various issues for possible future rulemaking, including those to be addressed following the completion of additional research, the gathering of additional operating experience, or the development of industry standards, or all three. One such issue concerned enhancing the requirements for corner posts on cab cars and MU locomotives.
See
64 FR 25607. FRA requirements for corner posts were based on conventional industry practice at the time, which had not proven adequate in then-recent side swipe collisions with cab cars leading.
Id.
FRA explained that those requirements were being adopted as an interim measure to prevent the introduction of equipment not meeting the requirements, that FRA was assisting APTA in preparing an industry standard for corner post arrangements on cab cars and MU locomotives, and that adoption of a suitable Federal standard would be an immediate priority.
Id.
In broader terms, this issue concerned the behavior of cab car and MU locomotive end frames when overloaded, as during an impact with maintenance-of-way equipment or with a highway vehicle at a highway-rail grade crossing, and thus concerned collision post strength as well. FRA and interested industry members also began identifying other issues related to the passenger equipment safety standards and the passenger train emergency preparedness regulations. FRA decided to address these issues with the assistance of FRA's Railroad Safety Advisory Committee (RSAC).
C. RSAC Overview
In March 1996, FRA established RSAC, which provides a forum for developing consensus recommendations to FRA's Administrator on rulemakings and other safety program issues. The Committee includes representation from all of the agency's major stakeholders, including railroads, labor organizations, suppliers and manufacturers, and other
interested parties. A list of member groups follows:
• American Association of Private Railroad Car Owners (AARPCO);
• American Association of State Highway and Transportation Officials (AASHTO);
• American Chemistry Council;
• American Petroleum Institute;
• APTA;
• American Short Line and Regional Railroad Association (ASLRRA);
• American Train Dispatchers Association;
• Association of American Railroads (AAR);
• Association of Railway Museums;
• Association of State Rail Safety Managers (ASRSM);
• Brotherhood of Locomotive Engineers and Trainmen (BLET);
• Brotherhood of Maintenance of Way Employes Division;
• Brotherhood of Railroad Signalmen (BRS);
• Chlorine Institute;
• Federal Transit Administration (FTA);*
• Fertilizer Institute;
• High Speed Ground Transportation Association (HSGTA);
• Institute of Makers of Explosives;
• International Association of Machinists and Aerospace Workers;
• International Brotherhood of Electrical Workers (IBEW);
• Labor Council for Latin American Advancement;*
• League of Railway Industry Women;*
• National Association of Railroad Passengers (NARP);
• National Association of Railway Business Women;*
• National Conference of Firemen & Oilers;
• National Railroad Construction and Maintenance Association;
• National Railroad Passenger Corporation (Amtrak);
• NTSB;*
• Railway Supply Institute (RSI);
• Safe Travel America (STA);
• Secretaria de Comunicaciones y Transporte;*
• Sheet Metal Workers International Association (SMWIA);
• Tourist Railway Association, Inc.;
• Transport Canada;*
• Transport Workers Union of America (TWU);
• Transportation Communications International Union/BRC (TCIU/BRC);
• Transportation Security Administration (TSA);* and
• United Transportation Union (UTU).
*Indicates associate, non-voting membership.
When appropriate, FRA assigns a task to RSAC, and after consideration and debate, RSAC may accept or reject the task. If the task is accepted, RSAC establishes a working group that possesses the appropriate expertise and representation of interests to develop recommendations to FRA for action on the task. These recommendations are developed by consensus. A working group may establish one or more task forces to develop facts and options on a particular aspect of a given task. The individual task force then provides that information to the working group for consideration. If a working group comes to unanimous consensus on recommendations for action, the package is presented to the full RSAC for a vote. If the proposal is accepted by a simple majority of RSAC, the proposal is formally recommended to FRA. FRA then determines what action to take on the recommendation. Because FRA staff play an active role at the working group level in discussing the issues and options and in drafting the language of the consensus proposal, FRA is often favorably inclined toward the RSAC recommendation. However, FRA is in no way bound to follow the recommendation, and the agency exercises its independent judgment on whether the recommendation achieves the agency's regulatory goal, is soundly supported, and is in accordance with policy and legal requirements. Often, FRA varies in some respects from the RSAC recommendation in developing an actual regulatory proposal or final rule. Any such variations would be noted and explained in the rulemaking document issued by FRA. If the working group or RSAC is unable to reach consensus on a recommendation for action, FRA moves ahead to resolve the issue(s) through traditional rulemaking proceedings or other action.
D. Establishment of the Passenger Safety Working Group in May 2003
On May 20, 2003, FRA presented, and RSAC accepted, the task of reviewing existing passenger equipment safety needs and programs and recommending consideration of specific actions that could be useful in advancing the safety of rail passenger service. RSAC established the Passenger Safety Working Group (Working Group) to handle this task and develop recommendations for the full RSAC body to consider. Members of the Working Group, in addition to FRA, include the following:
• AAR, including members from BNSF Railway Company (BNSF), CSX Transportation, Inc., and Union Pacific Railroad Company;
• AAPRCO;
• AASHTO;
• Amtrak;
• APTA, including members from Bombardier, Inc., LDK Engineering, Herzog Transit Services, Inc., Long Island Rail Road (LIRR), Metro—North Commuter Railroad Company (Metro-North), Northeast Illinois Regional Commuter Railroad Corporation (Metra), Southern California Regional Rail Authority (Metrolink), and Southeastern Pennsylvania Transportation Authority (SEPTA);
• BLET;
• BRS;
• FTA;
• HSGTA;
• IBEW;
• NARP;
• RSI;
• SMWIA;
• STA;
• TCIU/BRC;
• TWU; and
• UTU.
Staff from DOT's John A. Volpe National Transportation Systems Center (Volpe Center) attended all of the meetings and contributed to the technical discussions. In addition, staff from the NTSB met with the Working Group. The Working Group has held 13 meetings on the following dates and locations:
• September 9-10, 2003, in Washington, DC;
• November 6, 2003, in Philadelphia, PA;
• May 11, 2004, in Schaumburg, IL;
• October 26-27, 2004 in Linthicum/Baltimore, MD;
• March 9-10, 2005, in Ft. Lauderdale, FL;
• September 7, 2005 in Chicago, IL;
• March 21-22, 2006 in Ft. Lauderdale, FL;
• September 12-13, 2006 in Orlando, FL;
• April 17-18, 2007 in Orlando, FL;
• December 11, 2007 in Ft. Lauderdale, FL;
• June 18, 2008, in Baltimore, MD;
• November 13, 2008, in Washington, DC; and
• June 8, 2009, in Washington, DC.
At the meetings in Chicago and Ft. Lauderdale in 2005, FRA met with representatives of Tri-Rail (the South Florida Regional Transportation Authority) and Metra, respectively, and toured their passenger equipment. The visits were open to all members of the Working Group and FRA believes they have added to the collective understanding of the Group in identifying and addressing passenger equipment safety issues.
E. Establishment of the Crashworthiness/Glazing Task Force in November 2003
Due to the variety of issues involved, at its November 2003 meeting the Working Group established four task forces—smaller groups to develop recommendations on specific issues within each group's particular area of expertise. Members of the task forces included various representatives from the respective organizations that were part of the larger Working Group. One of these task forces was assigned the job of identifying and developing issues and recommendations specifically related to the inspection, testing, and operation of passenger equipment as well as concerns related to the attachment of safety appliances on passenger equipment. An NPRM on these topics was published on December 8, 2005,
see
70 FR 73069, and a final rule was published on October 19, 2006,
see
71 FR 61835. Another of these task forces was established to identify issues and develop recommendations related to emergency systems, procedures, and equipment, and helped to develop an NPRM on these topics that was published on August 24, 2006,
see
71 FR 50276, and a final rule that was published on February 1, 2008,
see
73 FR 6370. Another task force, the Crashworthiness/Glazing Task Force (Task Force), was assigned the job of developing recommendations related to glazing integrity, structural crashworthiness, and the protection of occupants during accidents and incidents. Specifically, this Task Force was charged with developing recommendations for glazing qualification testing and for cab car and MU locomotive end frame optimization. (Glazing and cab car/MU locomotive end frame issues are being handled separately, and glazing is not a subject of this final rule.) The Task Force was also given the responsibility of addressing a number of other issues related to glazing, structural crashworthiness, and occupant protection and recommending any research necessary to facilitate their resolution. Members of the Task Force, in addition to FRA, include the following:
• AAR;
• Amtrak;
• APTA, including members from Bombardier, Inc., General Electric Transportation Systems, General Motors-Electro-Motive Division, Kawasaki Rail Car, Inc., LDK Engineering, LIRR, LTK Engineering Services, Maryland Transit Administration, Massachusetts Bay Transportation Authority (MBTA), Metrolink, Metro-North, Northern Indiana Commuter Transportation District (NICTD), Hyundai Rotem Company, Saint Gobian Sully NA, San Diego Northern Commuter Railroad (Coaster), SEPTA, and STV, Inc.;
• BLET;
• California Department of Transportation (Caltrans);
• NARP;
• RSI; and
• UTU.
While not voting members of the Task Force, representatives from the NTSB attended meetings and contributed to the discussions of the Task Force. In addition, staff from the Volpe Center attended all of the meetings and contributed to the technical discussions.
The Task Force held seven meetings on the following dates and locations:
• March 17-18, 2004, in Cambridge, MA;
• May 13, 2004, in Schaumburg, IL;
• November 9, 2004, in Boston, MA;
• February 2-3, 2005, in Cambridge, MA;
• April 21-22, 2005, in Cambridge, MA;
• August 11, 2005, in Cambridge, MA; and
• September 9-10, 2008, in Cambridge, MA.
F. Development of the NPRM Published in August 2007
The NPRM was developed to address concerns raised and issues discussed about cab car and MU locomotive front end frame structures during the Task Force meetings and pertinent Working Group meetings. Minutes of each of these meetings have been made part of the docket in this proceeding and are available for public inspection. Except for one issue, which is discussed below, the Working Group reached consensus on the principal regulatory provisions contained in the NPRM at its meeting in September 2005. After the September 2005 meeting, the Working Group presented its recommendations to the full RSAC body for concurrence at its meeting in October 2005. All of the members of the full RSAC in attendance at its October 2005 meeting accepted the regulatory recommendations submitted by the Working Group. Thus, the Working Group's recommendations became the full RSAC's recommendations to FRA.
After reviewing the full RSAC's recommendations, FRA agreed that the recommendations provided a good basis for a proposed rule, but that test standards and performance criteria more suitable to cab cars and MU locomotives without flat forward ends or with energy absorbing structures used as part of a crash energy management design (CEM), or both, should be specified. The NPRM therefore provided an option for the dynamic testing of cab cars and MU locomotives as a means of demonstrating compliance with the rule. However, FRA made clear that the proposal was not the result of an RSAC recommendation. Otherwise, FRA adopted the RSAC's recommendations with generally minor changes for purposes of clarity and formatting in the
Federal Register
.
The NPRM was published in the
Federal Register
on August 1, 2007,
see
72 FR 42016, and FRA solicited public comment on it. FRA notified the public of its option to submit written comments on the NPRM and to request a public, oral hearing on the NPRM. FRA also invited comment on a number of specific issues related to the proposed requirements for the purpose of developing the final rule.
G. Development of This Final Rule
This final rule is the product of FRA's review and consideration of the recommendations of the Task Force, Working Group, and full RSAC, and the written comments to the docket. FRA received written comments in response to the publication of the NPRM from a wide array of interested parties. Specifically, FRA received three separate comments from members of the U.S. Congress: (1) From Senator Kent Conrad, Senator Byron Dorgan, and Congressman Earl Pomeroy; (2) from Congressman James Oberstar, Chairman, House Committee on Transportation and Infrastructure, and Congressman Bennie Thompson, Chairman, House Committee on Homeland Security; and (3) from Congressman Adam Schiff. FRA also received comments from the AAR and APTA, which represent freight and passenger railroads, respectively, as well as comments from Caltrans and the Peninsula Corridor Joint Powers Board (Caltrain), which are involved in providing passenger rail service. The BLET and UTU submitted comments on behalf of the railroad employees whom they represent. In addition, FRA received comments from rail car manufacturers Bombardier and Colorado Railcar Manufacturing (CRM), as well as from the firm of Raul V. Bravo + Associates, Inc. (RVB). FRA also received comments from other interested parties: the American Association for Justice (AAJ), formerly known as the Association of Trial Lawyers of America, and the California Public Utilities Commission (CPUC). All Aboard Washington (AAWA), an advocacy organization for promoting
rail service in the Pacific Northwest, and a private citizen also commented on the NPRM. At about the same time as the written comment period closed on October 1, 2007, management of DOT rulemaking dockets was transitioning from DOT to the Federal Docket Management System at
http://www.regulations.gov
. This transition led to some delay in the posting of comments to the Web site; however, FRA has considered all such comments in preparing this final rule.
FRA notes that Congressman Adam Schiff made a request that FRA hold public hearings to receive oral comment on the NPRM in Los Angeles or Glendale, CA, so that those who have a “deeply-felt” concern for rail safety could be heard. As stated in a January 30, 2008 letter to Congressman Schiff, FRA discussed this request with the Congressman's staff and was informed that the Congressman had decided to reserve his request that FRA convene public hearings on the NPRM. (A copy of this letter has been placed in the public docket for this rulemaking.) No public hearing was held in response to the NPRM.
Throughout the preamble discussion of this final rule, FRA refers to comments, views, suggestions, or recommendations made by members of the Task Force, Working Group, and full RSAC. FRA does so to show the origin of certain issues and the nature of discussions concerning those issues at the Task Force, Working Group, and full RSAC level. FRA believes this serves to illuminate factors that it has weighed in making its regulatory decisions, as well as the logic behind those decisions. The reader should keep in mind, of course, that only the full RSAC makes recommendations to FRA and that it is the consensus recommendation of the full RSAC on which FRA acts. However, as noted above, FRA is in no way bound to follow the recommendation, and the agency exercises its independent judgment on whether the recommended rule achieves the agency's regulatory goal, is soundly supported, and is in accordance with policy and legal requirements.
III. Technical Background
Transporting passengers by rail in the U.S. is very safe. Since the beginning of 1978, about 12.5 billion passengers have traveled by rail, based on reports filed monthly with FRA. The number of rail passengers has steadily increased over the years, and since the year 2000 has averaged more than 525 million passengers per year. On a passenger-mile basis, with an average of about 16.1 billion passenger-miles per year since 2000, rail travel is about as safe as scheduled airline service and intercity bus transportation, and it is far safer than private motor vehicle travel. Passenger rail accidents—while always to be avoided—have a very high passenger survival rate.
Yet, as in any form of transportation, there are risks inherent in passenger rail travel. For this reason, FRA continually works to improve the safety of passenger rail operations. FRA's efforts include sponsoring the research and development of safety technologies, providing technical support for industry specifications and standards, and engaging in cooperative rulemaking efforts with key industry stakeholders. FRA has focused in particular on enhancing the crashworthiness of passenger trains.
In a passenger train collision or derailment, the principal crashworthiness risks that occupants face are the loss of safe space inside the train from crushing of the train structure and, as the train decelerates, the risk of secondary impacts with interior surfaces. Therefore, the principal goals of the crashworthiness research sponsored by FRA are twofold: First, to preserve a safe space in which occupants can ride out the collision or derailment, and, then, to minimize the physical forces to which occupants are subjected when impacting surfaces inside a passenger train as the train decelerates. Though not a part of this final rule, other crashworthiness research focuses on related issues such as fuel tank safety, for equipment with a fuel tank, and the associated risk of fire if the fuel tank is breached during the collision or derailment.
The results of ongoing research on cab car and MU locomotive front end frame structures help demonstrate both the effectiveness and the practicality of the structural enhancements in this final rule to make this equipment more crashworthy. This research is discussed below, along with other technical information providing the background for this rulemaking.
A. Predominant Types of Passenger Rail Service
FRA's focus on cab car and MU locomotive crashworthiness should be considered in the context of the predominant types of passenger rail service in North America. The first involves operation of passenger trains with conventional locomotives in the lead, typically pulling consists of passenger coaches and other cars such as baggage cars, dining cars, and sleeping cars. Such trains are common on long-distance, intercity rail routes operated by Amtrak. On a daily basis, however, most passenger rail service is provided by commuter railroads, which typically operate one or both of the two most predominant types of service: Push-pull service and MU locomotive service.
Push-pull service is passenger train service typically operated, in one direction of travel, with a conventional locomotive in the rear of the train pushing the consist (the “push mode”) and with a cab car in the lead position of the train; and, in the opposite direction of travel, the service is operated with the conventional locomotive in the lead position of the train pulling the consist (the “pull mode”) and with the cab car in the rear of the train. (A cab car is both a passenger car, in that it has seats for passengers, and a locomotive, in that it has a control cab from which the engineer can operate the train.) Control cables run the length of the train, as do electrical lines providing power for heat, lights, and other purposes.
MU locomotive service is passenger rail service involving trains consisting of self-propelled electric or diesel MU locomotives. MU locomotives may operate individually but typically operate semi-permanently coupled together as a pair or triplet with a control cab at each end of the consist. During peak commuting hours, multiple pairs or triplets of MU locomotives, or a combination of both, are typically operated together as a single passenger train in MU service. This type of service does not make use of a conventional locomotive as a primary means of motive power. MU locomotive service is very similar to push-pull service as operated in the push mode with the cab car in the lead.
By focusing on enhancements to cab car and MU locomotive crashworthiness, FRA seeks to enhance the safety of the two most typical forms of passenger rail service in the U.S.
B. Front End Frame Structures of Cab Cars and MU Locomotives
Structurally, MU locomotives and cab cars built in the same period are very similar. Both are designed to be occupied by passengers and to operate as the lead units of passenger trains. The principal distinction is that cab cars do not have motors to propel themselves. Unlike MU locomotives and cab cars, conventional locomotives are not designed to be occupied by passengers—only by operating crewmembers. Concern has been raised about the safety of cab car-led and MU locomotive train service due to the closer proximity of the engineer and
passengers to the leading end of the train than in conventional locomotive-led service.
The principal purpose of cab car and MU locomotive front end frame structures is to provide protection for the engineer and passengers in the event of a collision where the superstructure of the vehicle is directly engaged and the underframe is either not engaged or only indirectly engaged in the collision. In the event of impacts with objects above the underframe of a cab car or MU locomotive, the end frame members are the primary source of protection for the engineer and the passengers. There are various types of cab cars and MU locomotives in current use. As discussed below, flat-nosed, single-level cab cars have been used for purposes of FRA-sponsored crashworthiness research. (The cab cars were originally constructed as MU locomotives but had their traction motors removed for testing.) Flat-nosed designs are representative of a large portion of the cab car and MU locomotive fleet.
In a typical flat-nosed cab car, the end frame is composed of several structural elements that act together to resist inward deformations under load. The base of the end frame structure is composed of the end/buffer beam, which is directly connected to the draft sill of the vehicle. For cars that include stepwells, the side sills of the underframe generally do not directly connect to the end/buffer beam. There are four major vertical members connected to the end/buffer beam: two collision posts located approximately at the one-third points along the length of the beam; and two corner posts located at the outermost points of the beam. These structural elements are also connected together through two additional lateral members: a lateral member/shelf located just below the window frame structure; and an anti-telescoping plate at the top. The attachment of the end frame structure to the rest of the vehicle typically occurs at three locations. The first location is at the draft sill at the level of the underframe. This is the main connection where a majority of any longitudinal load applied to the end frame is reacted into the underframe of the vehicle. There are two other connections at the cant/roof rail located at each side of the car just below the level of the roof. When a longitudinal load is applied to the end frame, it is reacted by the draft sill and the cant rails into the main car body structure. A schematic of a typical arrangement is depicted in Figure 1 (although not every cab car or MU locomotive necessarily has every component shown).
ER08JA10.000
C. Accident History
In a collision involving the front end of a cab car or an MU locomotive, it is vitally important that the end frame behaves in a ductile manner, absorbing some of the collision energy in order to maintain sufficient space in which the engineer and passengers can ride out the event. Several collisions have occurred where the superstructure of a leading cab car has been loaded but the underframe of the car has not. These collisions demonstrate a need for better protecting the cab engineer and passengers from external threats. One example of a collision where the end frame did not effectively absorb collision energy occurred in Portage, IN, in 1998 when a NICTD train consisting of MU locomotives struck a tractor-tandem trailer carrying steel coils that had become immobilized on a grade crossing.
1
The leading MU locomotive impacted a steel coil at a point centered on one of its collision posts, the collision post failed, and the steel coil penetrated into the interior of the locomotive, resulting in three fatalities. Little of the collision energy was absorbed by the collision post, because the post had failed before it could deform in any significant way.
1
National Transportation Safety Board, “Collision of Northern Indiana Commuter Transportation District Train 102 with a Tractor-Trailer Portage, Indiana, June 18, 1998,” RAR-99-03, 07/26/1999. This report is available on the NTSB's Web site at:
http://www.ntsb.gov/publictn/1999/RAR9903.pdf
.
There are additional examples of incidents where the end frame of a cab car or an MU locomotive was engaged during a collision and a loss of survivable volume ensued due to the failure of end frame structures. In a collision in Secaucus, NJ, in 1996, a cab car-led New Jersey Transit Rail Operations (NJTR) train impacted a conventional locomotive-led NJTR
train.
2
At the collision interface, the conventional locomotive pushed in or tore loose the collision and corner posts of the cab car. The underframe of the cab car was not loaded. The engineers of both trains and one passenger in the cab car were fatally injured. Also in 1996 in Silver Spring, MD, a collision occurred between a cab car-led Maryland Area Rail Commuter (MARC) train and a conventional locomotive-led Amtrak train. In the collision, the front left collision and corner posts of the cab car were pushed in and torn loose. The underframe of the cab car was not loaded.
3
Three crewmembers and eight passengers on the MARC train were fatally injured as result of the collision and ensuing fire. Earlier, on January 18, 1993, near Gary, IN, two NICTD trains collided corner-to-corner on intersecting tracks that shared a bridge. One of the trains was at rest and the other had a speed estimated to be 32 mph. The left front corner posts and adjacent car body sidewall structures were destroyed on the leading MU locomotive of each train. Seven passengers were fatally injured.
4
2
National Transportation Safety Board, “Near Head-On Collision and Derailment of Two New Jersey Transit Commuter Trains Near Secaucus, New Jersey, February 9, 1996,” RAR-97-01, 03/25/1997. This report is available on the NTSB's Web site at:
http://www.ntsb.gov/publictn/1997/RAR9701.pdf
.
3
National Transportation Safety Board, “Collision and Derailment of Maryland Rail Commuter
MARC
Train 286 and National Railroad Passenger Corporation
AMTRAK
Train 29 Near Silver Spring, Maryland, on February 16, 1996,” RAR-97-02, 06/17/1997. This report is available on the NTSB's Web site at:
http://www.ntsb.gov/publictn/1997/RAR9702.pdf
.
4
National Transportation Safety Board, “Collision between Northern Indiana Commuter Transportation District Eastbound Train 7 and Westbound Train 12 Near Gary, Indiana, on January 18, 1993,” RAR-93-03, 12/7/1993.
The preceding collisions were used to characterize types of loading conditions, which led to the development of a simplified, generalized test scenario, in furtherance of the goal of establishing methods for measuring the crashworthiness performance of end frame structures and developing strategies for incrementally improving their survivability under a range of impact conditions. Although the speeds associated with certain past events are greater than the speed at which full protection can currently be provided, and even though enhancements to passenger train emergency features and other requirements unrelated to crashworthiness, such as fire safety, may overall do as much or more to prevent or mitigate the consequences of these types of events, these collisions do provide indicative loading conditions for developing structural enhancements that can improve crashworthiness performance.
FRA also notes that on January 26, 2005, in Glendale, CA, a collision involving an unoccupied sport utility vehicle (SUV) (that was intentionally parked on the track by a private citizen), two Metrolink commuter trains, and a standing freight train resulted in 11 fatalities and numerous injuries. Eight of the fatalities occurred on a cab car-led commuter train, which derailed after striking the SUV, causing the cab car to be guided down a railroad siding, which resulted in an impact at an approximate speed of 49 mph with the standing freight train. After the collision with the standing freight train, the rear end of the lead cab car buckled laterally, obstructing the right-of-way of an oncoming, conventional locomotive-led commuter train. The rear end of the cab car raked the side of the conventional locomotive-led train, which was moving at an approximate speed of 51 mph, crushing occupied areas of that train. This incident involved enormous quantities of kinetic energy, and the underframe of the leading cab car crushed more than 20 feet inward. Because the strength of the end frame ultimately depends on the strength of the underframe, which failed here, stronger collision posts and corner posts on the front end of the leading cab car would have been, in themselves, of little benefit in absorbing the collision energy. For this reason, as discussed below, FRA has been exploring other crashworthiness strategies, such as CEM, to help mitigate the effects of collisions involving higher impact speeds. Nevertheless, CEM will also require proper end frame performance in order to function as intended.
D. FRA and Industry Standards for Front End Frame Structures of Cab Cars and MU Locomotives
Both the Federal government and the passenger railroad industry have been working together to improve the crashworthiness of cab cars and MU locomotives. As noted above, in 1999, after several years of development and in consultation with a working group comprised of key industry stakeholders, FRA promulgated the Passenger Equipment Safety Standards final rule. The rule included end frame structure requirements and additional crashworthiness-related requirements for cab cars, MU locomotives, and other passenger equipment. In particular, the final rule provided for strengthened collision posts for new cab cars and MU locomotives (
i.e.
, those ordered on or after September 8, 2000, or placed in service for the first time on or after September 9, 2002).
APTA also issued industry standards in 1999, in furtherance of its initiative to continue the development and maintenance of voluntary industry standards for the safety of railroad passenger equipment. In particular, APTA Safety Standard (SS)-C&S-013-99, Standard for Corner Post Structural Strength for Railroad Passenger Equipment, and SS-C&S-014-99, Standard for Collision Post Structural Strength for Railroad Passenger Equipment, included provisions on end frame designs for cab cars and MU locomotives. (Copies of these standards have been placed in the public docket for this rulemaking.) Specifically, these APTA standards included increased industry requirements for the strength of cab car and MU locomotive vertical end frame members—collision posts and corner posts. The 1999 APTA standards also included industry requirements for the deformation of these end frame vertical members, specifying that they must be able to sustain “severe deformation” before failure of the connections to the underframe and roof structures occurs.
In January 2000, APTA requested that FRA develop information on the effectiveness of APTA's then-recently introduced
Manual of Standards and Recommended Practices for Rail Passenger Equipment
, which included APTA SS-C&S-013-99 and APTA SS-C&S-014-99, and FRA's then-recently issued Passenger Equipment Safety Standards rule. This review was intended to look in particular at what increase in crashworthiness was obtained for cab cars and MU locomotives through the combination of these standards and regulations. FRA shared APTA's interest and included full-scale impact tests and associated planning and analysis activities in its overall research plan to gather this information. FRA then developed the details of the testing process in conjunction with APTA's Passenger Rail Equipment Safety Standards (PRESS) Construction and Structural (C&S) Subcommittee.
Around this same time, questions arose in the passenger rail industry in applying the APTA standards for collision posts and corner posts to new cab cars and MU locomotives. Views differed as to what the standards actually specified—namely, the meaning of “severe deformation” in the provisions calling for corner and collision posts to sustain “severe deformation” before failure of the posts' attachments. Consequently, there was not common agreement as to whether
particular designs met the standards. On May 22, 2003, APTA's PRESS Committee accepted the recommendation of its C&S Subcommittee to replace these provisions in the standards concerning “severe deformation” with a recommended practice that the corner and collision post attachments be able to sustain minimum prescribed loads with negligible deformation. APTA SS-C&S-013-99 and SS-C&S-014-99 were then incorporated in their entirety into APTA SS-C&S-034-99, Rev. 1, Standard for the Design and Construction of Passenger Railroad Rolling Stock. (A copy of APTA SS-C&S-034-99, Rev. 1, has been placed in the public docket for this rulemaking. As discussed below, the latest revision, Rev. 2, of APTA SS-C&S-034-99 is available on APTA's Web site at
http://www.aptastandards.com/portals/0/PRESS_pdfs/Construcstruct/construcstruct%20reaffirm/APTA%20SS-CS-034-99%20Rev%202-Approved.pdf
. The larger compilation of standards and recommended practices for rail passenger equipment of which this standard is a part, APTA's
Manual of Standards and Recommended Practices for Rail Passenger Equipment
, is available on APTA's Web site at
http://aptastandards.com/PublishedDocuments/PublishedStandards/PRESS/tabid/85/Default.aspx
.)
When the decision to turn the provisions concerning “severe deformation” into a recommended practice was made, ongoing research from full-scale impact tests was showing that a substantial increase in cab car and MU locomotive crashworthiness could be achieved by designing the posts to first deform and thereby absorb collision energy before failing.
5
As discussed below, in August 2005, APTA's PRESS C&S Subcommittee accepted a revised “severe deformation” standard for collision and corner posts. The standard includes requirements for minimum energy absorption and maximum deflection. The standard thereby eliminates a deficiency in the 1999 APTA standards by specifying test criteria to objectively measure “severe deformation” (or large deformation).
5
Mayville, R., Johnson, K., Tyrell, D., Stringfellow, R., “Rail Vehicle Cab Car Collision and Corner Post Designs According to APTA S-034 Requirements,” American Society of Mechanical Engineers, Paper No. IMECE2003-44114, November 2003. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2003/rail_cw_2003_11.pdf
. All of the published Volpe Center papers and reports on rail equipment crashworthiness can be found at:
http://www.volpe.dot.gov/sdd/pubs-crash.html
.
The NPRM in this rulemaking was based on APTA SS-C&S-034-99, Rev. 1, and proposed dynamic performance requirements in the alternative to the quasi-static, large deformation criteria in the APTA Standards. In response to the NPRM, members of industry disagreed with including FRA's proposed dynamic performance requirements in the rule and requested that FRA demonstrate actual compliance with both the quasi-static and the dynamic large deformation requirements that were proposed. As detailed below, these tests were performed in the spring and summer of 2008. FRA has sought to retain the dynamic performance requirements as an alternative to the quasi-static requirements, in particular because the dynamic performance requirements facilitate evaluation of equipment without a flat front-end or traditional corner or collision posts. After discussion within the Task Force, consensus was reached on including dynamic performance requirements in appendix F to part 238 as an alternative to the enhanced collision and corner post requirements in §§ 238.211 and 238.213 of this final rule. As discussed below, the enhanced requirements in §§ 238.211 and 238.213 essentially codify the current APTA standards.
E. Testing of Front End Frame Structures of Cab Cars and MU Locomotives
This section summarizes the work done by FRA and the passenger rail industry on developing the technical information to support regulations requiring that corner and collision posts in cab car and MU locomotive front end frames fail in a controlled manner when overloaded. Due to the collaborative work of FRA with the passenger rail industry, APTA's current passenger rail equipment standards include deformation requirements, which prescribe how these vertical members should perform when overloaded quasi-statically.
1. FRA-Sponsored Dynamic Testing in 2002
Two full-scale, grade-crossing impact tests were conducted in June 2002 as part of an ongoing series of FRA-sponsored crashworthiness tests of passenger rail equipment carried out with the support of the Volpe Center at FRA's Transportation Technology Center (TTC) in Pueblo, CO. The purpose of these two tests was to evaluate incremental improvements in the crashworthiness performance, in highway-rail grade-crossing collision scenarios, of modern corner and collision post designs when compared against the performance of older designs. The grade-crossing tests were intended to address the concern of occupant vulnerability to bulk crushing resulting from offset/oblique collisions where the primary load-resisting-structure is the equipment's end frame design.
a. Test Article Designs
Two end frame designs were developed. The first end frame design was representative of typical designs of passenger rail vehicles in the 1990s prior to 1999. The first end frame design is referred to as the “1990s design.” The second end frame design incorporated all the enhancements required beginning in 1999 by FRA's Passenger Equipment Safety Standards in part 238 and also recommended beginning in 1999 by APTA's standards for corner post and collision post structures, respectively, SS-C&S-013-99 and SS-C&S-014-99. The second end frame design is referred to as the State-of-the-Art (SOA) design. The two end frame designs developed were then retrofitted onto two Budd Pioneer passenger rail cars for testing.
The SOA design differed principally from the 1990s design by having higher values for static loading of the end frame structure and by specifically addressing the performance of the collision and corner posts when overloaded. As noted above, the 1999 APTA standards for cab car and MU locomotive end frame structures included the following statement for both corner and collision posts:
[The] post and its supporting structure shall be designed so that when it is overloaded * * * failure shall begin as bending or buckling in the post. The connections of the post to the supporting structure, and the supporting car body structure, shall support the post up to its ultimate capacity. The ultimate shear and tensile strength of the connecting fasteners or welds shall be sufficient to resist the forces causing the deformation, so that shear and tensile failure of the fasteners or welds shall not occur, even with severe deformation of the post and its connecting and supporting structural elements.
(
See
paragraph 4.1 of APTA SS-C&S-013-99, and paragraph 3.1 of APTA SS-C&S-014-99.) Although the term “severe deformation” was not specifically defined in the APTA standards, discussions with APTA technical staff led to specifying “severe deformation” in the SOA design as a horizontal crush of the corner and collisions posts for a distance equal to the posts' depth. Some failure of the parent material in the posts was allowable, but no failure would be
allowed in the welded connections, as the integrity of the welded connections prevents complete separation of the posts from their connections.
An additional difference in the designs was the exclusion of the stepwells for the SOA design, to allow for extended side sills from the body bolster to the end/buffer beam. By bringing the side sills forward to support the end/buffer beam directly at the corners, the end/buffer beam can be developed to a size similar to the one for the 1990s design. In fact, recent cab car procurements have provided for elimination of the stepwells at the ends of the cars.
As compared to the 1990s design, the SOA design had the following enhancements: more substantial corner posts; a bulkhead sheet connecting the collision and corner posts, extending from the floor to the transverse member connecting the posts; and a longer side sill that extended along the engineer's compartment to the end beam, removing the presence of a stepwell. In addition to changes in the cross-sectional sizes and thickness of some structural members, another change in the SOA design was associated with the connection details for the corner posts. In comparison to the corner posts, the collision posts of both the 1990s and SOA designs penetrated both the top and bottom flanges of both the end/buffer beam and the anti-telescoping plate. This was based upon typical practice in the early 1990s for the 1990s design, and a provision in the APTA standard for the SOA design. Yet, the corner posts differed in that the corner posts for the 1990s design did not penetrate both the top and bottom flanges of the end/buffer and anti-telescoping beams, while those in the SOA design did. The SOA design therefore had a significantly stiffer connection that was better able to resist torsional loads transferred to the anti-telescoping plate.
b. Dynamic Impact Testing
As noted, two full-scale, grade-crossing impact tests were conducted in June 2002. In each test a single cab car impacted a 40,000-pound steel coil resting on a frangible table at a nominal speed of 14 mph. The steel coil was situated such that it impacted the corner post above the cab car's end sill. The principal difference between the two tests involved the end frame design tested: In one test, the cab car was fitted with the 1990s end frame design; in the other, the cab car was fitted with the SOA end frame design.
Prior to the tests, the crush behaviors of the cars and their dynamic responses were simulated with car crush and collision dynamics models. The car crush model was used to determine the force/crush characteristics of the corner posts, as well as their modes of deformation.
6
The collision dynamics model was used to predict the extent of crush of the corner posts as a function of impact velocity, as well as predict the three-dimensional accelerations, velocities, and displacements of the cars and coil.
7
Pre-test analyses of the models were used in determining the initial test conditions and instrumentation test requirements.
6
Martinez, E., Tyrell, D., Zolock, J., “Rail-Car Impact Tests with Steel Coil: Car Crush,” American Society of Mechanical Engineers, Paper No. JRC2003-1656, April 2003. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2003/rail_cw_2003_4.pdf.
7
Jacobsen, K., Tyrell, D., Perlman, A.B., “Rail Car Impact Tests with Steel Coil: Collision Dynamics,” American Society of Mechanical Engineers, Paper No. JRC2003-1655, April 2003. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2003/rail_cw_2003_3.pdf.
The impact speed of approximately 14 mph for both tests was chosen so that there would be significant intrusion (more than 12 inches) into the engineer's cab in the test of the 1990s design, and limited intrusion (less than 12 inches) in the test of the SOA design. This 12-inch deformation metric was chosen to demarcate the amount of intrusion that would leave sufficient space for the engineer to ride out the collision safely.
During the full-scale test of the 1990s design, the impact force transmitted to the end structure exceeded the corner post's predicted strength, and the corner post separated from its upper attachment. Upon impact, the corner post began to hinge near the contact point with the coil; subsequently, tearing at the upper connection occurred. The intensity of the impact ultimately resulted in the failure of the upper connection of the corner post to the anti-telescoping plate. More than 30 inches of deformation occurred and the survivable space for the engineer was lost.
By contrast, during the test of the SOA end frame design, the corner post remained attached. The maximum rearward deformation measured was approximately 9 inches. The results of this test showed that the SOA end frame design is sufficient to prevent the engineer from being crushed in such an impact.
c. Analysis
The SOA design performed very closely to pre-test predictions made by the finite element and collision dynamics models.
See
Figure 2, below. As noted, the SOA design crushed approximately 9 inches in the longitudinal direction.
ER08JA10.001
Pre-test analyses for the 1990s design using the car crush model and collision dynamics model were in close agreement with the measurements taken during the actual testing of the cab car end frame built to this design. The pre-test analyses also nearly overlay the test results for the force/crush characteristic of the SOA design. As a result, FRA believes that both sets of models are capable of predicting the modes of structural deformation and the total amount of energy consumed during a collision. Careful application of finite-element modeling allows accurate prediction of the crush behavior of rail car structures.
Both the methodologies used to design the cab car end frames and the results of the tests show that significant increases in rail passenger equipment crashworthiness can be achieved if greater consideration is given to the manner in which structural elements deform when overloaded. Modern methods of analysis can accurately predict structural crush (severe deformation) and consequently can be used with confidence to develop structures that collapse in a controlled manner. Modern testing techniques allow the verification of the crush behavior of such structures.
2. Industry-Sponsored Quasi-Static Testing in 2001
While FRA's full-scale, dynamic testing program was being planned and conducted with input from key industry representatives, several passenger railroads were incorporating in procurement specifications the then-newly promulgated Federal regulations and industry standards issued in 1999. Specifically, both LIRR and Metro-North had contracted with Bombardier for the development of a new MU locomotive design, the M7 series. Bombardier conducted a series of qualifying quasi-static tests on a mock-up, front-end structure of an M7, including a severe deformation test of the collision post. In addition to the severe deformation test, the other end frame members were also tested elastically at the enhanced loads specified in the APTA standards. The severe deformation qualification test was conducted on February 20, 2001.
a. Test Article Design
The mock-up test article was developed for the front end of an M7 cab car. The first 19.25 feet of the car was fabricated with great fidelity between the car's body bolster and the extreme most forward end. The mock-up contained all structural elements, but did not contain the corner post rub plates, the plymetal floor, any interior finishing, windows, doors, bonnet, or similar components.
b. Quasi-Static Testing
Load was applied at incrementally increasing levels with hydraulic jacks while being measured by load cells at the rear of the longitudinal end frame members. Initially, the elastic limit was determined for the post, and then the large deformation test was conducted. The test was stopped, for safety considerations, prior to full separation of the collision post with the end/buffer beam.
The maximum deflection in the collision post before yielding occurred at a position 42 inches above the end beam, near the top of the plates used to reinforce the collision post. The plastic shape the collision post acquired during testing was `V'-shaped, with a plastic hinge occurring at 42 inches above the end beam. Some cracking and material failure occurred at the connection of the post with the end beam. The anti-telescoping plate was pulled down roughly three inches, and load was shed to the corner post via the shelf member and the bulkhead sheet. The shape that
the collision post experienced was very similar to what was observed from the dynamic testing of the SOA corner post, as discussed above.
c. Analysis
Under FRA sponsorship, the Volpe Center, with cooperation from Bombardier, conducted non-linear, large deformation analyses to evaluate the performance of the cab car corner and collision posts of the SOA end frame design and the Bombardier M7 design under dynamic test conditions. One of the purposes of this research was to determine whether the level of crashworthiness demonstrated by the SOA prototype design could actually be achieved by a general production design—here, the M7 design. Pre-test analysis predictions of the dynamic performance of the SOA corner post closely matched test measurements.
8
A similar analysis of the corner post was performed on the M7 design, and the results compared closely with the SOA design test and analysis results. Overall, the crashworthiness performance of the collision posts of the SOA and M7 designs were found to be essentially the same, and the M7 corner post design was even found to perform better than the SOA corner post design. This latter difference in performance was attributable to the sidewall support in the M7 design, which was not present in the SOA design.
8
Martinez, E., Tyrell, D., Zolock, J. Brassard, J., “Review of Severe Deformation Recommended Practice Through Analyses—Comparison of Two Cab Car End Frame Designs,” American Society of Mechanical Engineers, Paper No. RTD2005-70043, March 2005. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2005/rail_cw_2005_03.pdf
.
Having established the fidelity of the models and modeling approach, a number of comparative simulations were conducted of both the SOA end frame and the M7 end frame under both dynamic and quasi-static test conditions to assess the equivalency of the two different tests for demonstrating compliance with the severe deformation criteria. For both sets of tests, the modes of deformation were very similar at the same extent of longitudinal displacement, and the locations where material failure occurred were also similar. In addition, the predicted force-crush characteristics showed reasonable agreement within the repeatability of the tests. Figure 3, below, shows a comparison of the deformation mode for the M7's collision post, as observed from the quasi-static testing that was conducted and as predicted for the dynamic loading condition.
ER08JA10.002
3. FRA-Sponsored Dynamic and Quasi-Static Testing in 2008
In 2008, a full-scale dynamic test and two quasi-static tests were performed on the posts of an SOA end frame. These tests were designed to evaluate the dynamic and quasi-static methods for demonstrating energy absorption of the collision and corner posts. The tests focused on the collision and corner posts individually because of their critical positions in protecting the engineer and passengers in a collision where only the superstructure, not the underframe, is loaded.
a. Test Article Design
The SOA design was originally developed for the Budd Pioneer car for the 2002 dynamic impact testing. For the testing in 2008, only a Budd M1 car was available, so the design had to be modified to fit a Budd M1. The design
of an end frame for retrofit onto the cab end of a Budd Pioneer car was modified to account for differences between the two car designs. In addition, reinforcements to the M1 car body and connections from the end frame to the car body were designed and fabricated.
The design of the SOA end frame itself required only a few modifications to adapt to the M1 car body. Due to the rounded nature of the M1 car body as compared to the Pioneer car body, the lateral extent of the anti-telescoping beam was changed slightly so that it extended beyond the corner post by 1.5 inches, as compared to 1.0 inches for the Pioneer car.
b. Dynamic Testing of a Collision Post
For this test, a 14,000-pound cart impacted a standing car at a speed of 18.7 mph. The cart had a rigid coil shape mounted on the leading end that concentrated the impact load on the car's collision post. The test was conducted against the NPRM's proposed requirements for protecting the engineer's space—namely, that there be no more than 10 inches of permanent, longitudinal deformation and none of the attachments of any of the structural members separate.
During the test, the collision post deformed approximately 7.4 inches and absorbed approximately 138,000 foot-pounds of energy. The attachment between the post and the anti-telescoping beam remained intact. The connection between the post and the buffer beam did not completely separate; however, the forward flange and both side webs fractured. The post itself did not completely fail. There was material failure in the back and the sides of the post at the impact location. Overall, the end frame was successful in absorbing energy and preserving space for the engineer and the passengers. Figure 4 depicts three deformation states from the dynamic test: initial contact of the crash cart with the end frame, the greatest intrusion of the end frame, and the final deformation state.
ER08JA10.003
c. Quasi-Static Testing of Collision and Corner Posts
A quasi-static collision post test was run to compare the quasi-static and the dynamic performance requirements proposed in the NPRM and to demonstrate the efficacy of the quasi-static test method. The NPRM proposed that the collision post absorb at least 135,000 foot-pounds of energy in no more than 10 inches of longitudinal, permanent deformation. Load was applied with the same fixture for the dynamic test. This fixture had a diameter of 48 inches and a width of 36 inches. The fixture was made of a thick, stiff material and reinforced so that it did not deform or absorb energy. Longitudinal string potentiometers at several locations recorded the deformation of the post. Four load cells, connected in parallel, measured the load being applied into the post. The force and the displacement were cross-plotted and the integral was used to calculate the energy absorbed during the test.
The test car was coupled to a reaction car. As the load from the hydraulic ram was introduced to the car through the collision post, it was reacted through the couplers. The mode of deformation in the quasi-static collision post test was very similar to the mode of deformation seen in the dynamic collision post test. The collision post pulled down on the anti-telescoping beam. The post was loaded past 15 inches of deformation and did eventually fail completely in the middle. The collision post fractured as it separated from the buffer beam. After 11 inches of crush, the post had absorbed 110,000 foot-pounds of energy. Based on the unloading characteristic measured during the test, 11 inches of crush is approximately equal to 10 inches of permanent deformation. Since the collision post and end frame were supposed to absorb 135,000 foot-pounds of energy in 10 inches of permanent deformation, but only absorbed 110,000 foot-pounds of energy for that distance, the test article did not pass the test requirements.
Design details warranted a closer look in determining why the test was unsuccessful. The specimens taken at the location of the fracture revealed that an internal gusset on the post coincided with an exterior shelf tab. The gusset locations were within specification for these posts. However, there is some flexibility with the location of the gusset relative to the location to the shelf tab. In both the dynamic and quasi-static tests, the fracture occurred at the location of both the gusset and the shelf welds. The rigid gusset did not allow the post to oval as it deformed, causing the fracture at the back of the post.
Attention turned to conducting a test of the corner post. The NPRM proposed that the corner post absorb at least 120,000 foot-pounds of energy with no more than 10 inches of permanent, longitudinal deformation. The same fixture was used for this test as for the collision post testing. The fixture was centered on the corner post. In response to the results of the quasi-static test of the collision post, the shelf was redesigned so that the tab was removed and the depth of the shelf was decreased. This reduced the number of welds at the corner and back of the post. However, because the corner post was not designed with internal gussets, gusset design details did not need to be addressed.
In the quasi-static corner post test, the end frame deformed as expected and absorbed energy while deforming. The anti-telescoping beam was pulled down significantly and the shelf and bulkhead were deformed. The connection between the corner post and the buffer beam fractured, but the post did not separate completely. Also, the connection between the shelf and the post fractured, but the post itself did not fracture. The post and end frame absorbed 136,000 foot-pounds of energy in 11 inches of crush. After elastic recoil, 11 inches of crush is the equivalent of 10 inches of permanent deformation; thus, the test was successful.
The testing program demonstrated repeatable methods for assessing the energy-absorbing capability of end frame structures. These methods include both dynamic and quasi-static tests where energy absorption and permanent deformation are used as limiting criteria. The tests also show the improved crashworthiness of the SOA design.
d. Analysis
Analysis is a crucial part of conducting a full-scale test. Based on the results of the 2002 full-scale dynamic test in which a heavy steel coil impacted the corner post of an SOA end frame design, some fracture was expected in certain key end frame components during the 2008 tests. For this reason, a material failure model, based on the Bao-Wierzbicki fracture criterion, was implemented in the finite element model of the car end frame using ABAQUS/Explicit. The finite element model with material failure was used to assess the effect of fracture on the deformation behavior of car end structures during quasi-static loading and dynamic impact and, in particular, the ability of such structures to absorb energy.
The material failure model was implemented in ABAQUS/Explicit for use with shell elements. A series of preliminary calculations was first conducted to assess the effects of element type and mesh refinement on the deformation and fracture behavior of structures similar to those found on cab car and MU locomotive end frames, and to demonstrate that the Bao-Wierzbicki failure model can be effectively applied using shell elements.
Model parameters were validated through comparison to the results of the 2002 testing. Material strength and failure parameters were derived from test data for A710 steel. The model was then used to simulate the three full-scale tests that were conducted during 2008 as part of the FRA program—dynamic impact testing of a collision post, and quasi-static load testing of a collision post and a corner post. Analysis of the results of the two collision post tests revealed the need for revisions to both the design of some key end frame components and to key material failure parameters. Using the revised model, pre-test predictions for the outcome of the corner post test were found to be in very good agreement with the actual test results.
Overall, the results of the tests in comparison with their pretest analyses show that, at this time, actual testing is necessary to demonstrate performance. However, as modeling methods improve and are shown to predict failure and energy absorption more accurately, there is the potential that use of analysis alone will in the future be acceptable for demonstrating crashworthiness performance.
F. Approaches for Specifying Large Deformation Requirements
As discussed above, APTA's initial “severe deformation” standard for corner and collision posts, published in 1999, did not contain specific methodologies or criteria for demonstrating compliance with the standard. Consequently, the dynamic tests performed by FRA and the Volpe Center, static tests performed by members of the rail industry, and analyses conducted by the Volpe Center and its contractors all helped to develop the base of information needed to identify the types of analyses and test methodologies to use. Further, evaluation of the test data, with the analyses providing a supporting framework, allowed development of appropriate criteria to demonstrate compliance.
The principal criteria developed involve energy absorption through end frame deformation and the maximum amount of that deformation. As shown by FRA and industry testing, energy can be imparted to conventional flat-nosed cab cars and MU locomotives either dynamically or quasi-statically. As shown by Volpe Center analyses, currently available engineering tools can be used to predict the results of such tests. Given the complexity of such analyses, and commensurate uncertainties, there is a benefit to maintaining dynamic testing as an alternative for evaluating compliance with any “severe deformation” standard.
There are tradeoffs between quasi-static and dynamic testing of cab car and MU locomotive end frames. Both sets of tests prescribe a minimum amount of energy for end frame deformation. However, the manner in which the energy is applied is different, and the setup of the two types of tests is different. As demonstrated by the tests conducted by Bombardier, quasi-static tests can be conducted by rail equipment manufacturers at their own facilities. Dynamic tests require a segment of railroad track with appropriate wayside facilities; there are few such test tracks available. Nevertheless, dynamic tests do not require detailed knowledge of the car structure to be tested, and allow for a wide range of structural designs. Quasi-static tests require intimate knowledge of the structure being tested, to assure appropriate support and loading conditions, and development of quasi-static test protocols requires assumptions about the layout of the structure, confining structural designs. In addition, dynamic tests more closely approximate accident conditions than quasi-static tests do.
In August 2005, APTA's PRESS C&S Subcommittee accepted a revised “severe deformation” standard for collision and corner posts. The standard includes requirements for minimum energy absorption and maximum deflection. The form of the standard is largely based on the testing done by Bombardier, and therefore is quasi-static. The standard eliminates a deficiency of the 1999 standard by specifying test criteria to objectively measure “severe deformation.” The standard can be readily applied to conventional flat-nosed cab cars and MU locomotives but is more difficult to apply to shaped-nosed cab cars and MU locomotives or those with CEM designs, or both.
In addition, APTA as well as several equipment manufacturers have expressed an interest in maintaining the presence of a stairwell on the side of the cab car or MU locomotive opposite from where the locomotive engineer is situated. This feature enables multi-level boarding from both low and higher platforms. As such, FRA and the APTA PRESS C&S Subcommittee worked together to develop language associated with providing a safety equivalent to the requirements stipulated for cab car and MU locomotive corner posts in terms of energy absorption and end frame deformation. The Subcommittee agreed that for this arrangement there is sufficient protection afforded by the presence of two corner posts (an end corner post ahead of the stepwell and an internal corner post behind the stepwell) that are situated in front of the occupied space. The load requirements stipulated for such posts differ in that the longitudinal requirements are not equal to the transverse requirements.
This in effect changes the shape of these posts so that they are not equal in both width and height. For the end corner post ahead of the stepwell, the longitudinal loading requirements are smaller than the transverse ones. The opposite is true for the corner post behind the stepwell. It was agreed to allow for the combined contribution of both sets of corner posts, together, to provide an equivalent level of protection to that required for corner posts in standard cab car and MU locomotive configurations.
See
the discussion in the Section-by-Section Analysis on the structural requirements for cab cars and MU locomotives with a stairwell located on the side of the equipment opposite from where the locomotive engineer controls the train.
G. Crash Energy Management and the Design of Front End Frame Structures of Cab Cars and MU Locomotives
Research has shown that passenger rail equipment crashworthiness in train-to-train collisions can be significantly increased if the equipment structure is engineered to crush in a controlled manner. One manner of doing so is to design sacrificial crush zones into unoccupied locations in the equipment. These zones are designed to crush gracefully, with a lower initial force and increased average force. With such crush zones, energy absorption is shared by multiple cars during the collision, consequently helping to preserve the integrity of the occupied areas. While developed principally to protect occupants in train-to-train collisions, such crush zones can also potentially significantly increase crashworthiness in highway-rail grade-crossing collisions.
9
9
Tyrell, D.C., Perlman, A.B., “Evaluation of Rail Passenger Equipment Crashworthiness Strategies,” Transportation Research Record 1825, pp. 8-14, National Academy Press, 2003. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2003/rail_cw_2003_12.pdf.
The approach of including crush zones in passenger rail equipment is termed CEM, and it extends from current, conventional practice. Current practice for passenger equipment operated at speeds not exceeding 125 mph (
i.e.,
Tier I passenger equipment under part 238) requires that the equipment be able to support large loads without permanent deformation or failure, but does not specifically address how the equipment behaves when it crushes. CEM prescribes that car structures crush in a controlled manner when overloaded and absorb collision energy. In fact, for passenger equipment operating at speeds exceeding 125 mph but not exceeding 150 mph (
i.e.,
Tier II passenger equipment under part 238), the equipment must be designed with a CEM system to dissipate kinetic energy during a collision,
see
§ 238.403, and Amtrak's Acela Express trainsets were designed with a CEM system complying with this requirement.
FRA notes that Metrolink is in the process of procuring a new fleet of cars utilizing CEM technology. As part of its response to the Glendale, CA train incident on January 26, 2005, Metrolink determined that CEM design specifications should be included in this planned procurement, and, in coordination with APTA, approached FRA and FTA to draft such specifications. In turn, FRA and FTA formed the ad hoc Crash Energy Management Working Group in May 2005. This working group included government engineers and participants from the rail industry, including passenger railroads, suppliers, labor organizations, and industry consultants, many of whom also participated in the Crashworthiness/Glazing Task Force. The working group developed a detailed technical specification for crush zones in passenger cars for Metrolink to include in its procurement specification, as well as for other passenger railroads to include in future procurements of their own. Metrolink released its specification as part of an invitation for bid, and then awarded the contract to manufacture the equipment to Rotem, a division of Hyundai, now Hyundai Rotem Company (Rotem).
Rotem has developed a shaped-nose, CEM design for new Metrolink cab cars. Because of the shaped-nose, it is more difficult to engineer structural members identifiable as full-height collision posts and corner posts that extend from the underframe to the cantrail or roofline at the front end, as specified in the current APTA standard. Consequently, to meet the APTA standard, Rotem has to locate the collision and corner posts inboard of the crush zone, rather than place them at the extreme front end of the cab car. Further, as currently written, the APTA quasi-static standard does not expressly take into account the energy-absorption capability of the crush zone, even if the crush zone would likely be engaged in a grade-crossing impact. Although the APTA standard acknowledges the use of shaped-nose and CEM designs, there remains uncertainty in the standard associated with demonstration of compliance by such designs. (The APTA standard does provide that on cars with CEM designs, compliance may be demonstrated either through analysis or testing as agreed to by the vehicle builder and purchaser, but no test methodology or criteria are provided.)
Dynamic performance criteria place fewer constraints on the layout of the cab car or MU locomotive end structure and allow the energy-absorption capability of the crush zone(s) to be expressly taken into account in the design of the collision and corner post structures. As noted, this final rule allows for the application of dynamic performance requirements for collision and corner post structures of cab cars and MU locomotives. FRA believes that the results of the crashworthiness research discussed above provide strong support for including dynamic performance requirements as alternatives to the quasi-static requirements for collision and post requirements in this rule, and that it is particularly necessary to address what FRA believes will be a growing number of cab cars and MU locomotives utilizing CEM designs.
H. European Standard EN 15227 FCD, Crashworthiness Requirements for Railway Vehicle Bodies
In the NPRM, FRA discussed that then-preliminary European standard prEN 15227 FCD, Crashworthiness Requirements for Railway Vehicle Bodies, included four collision scenarios. This standard is no longer preliminary and is consequently referred to throughout this document as EN 15227, without the preliminary “pr” designation. Collision Scenario 3 of the European standard involves a “train unit front end impact with a large road vehicle on a level crossing.” The standard requires commuter and intercity trains to be able to sustain an impact with a deformable object weighing 33 kips (15,000 kg) at a speed up to 68 mph (110 kph). Calibration tests on components and numerical simulations of the scenario are recommended for showing compliance.
FRA has noted key differences between the European standard and the dynamic testing collision scenarios that FRA proposed for both collision posts and corner posts, below, including the amount of energy involved and the character of the object. Assuming that the mass of the train is more than about 25 times as great as the mass of the object (in that the mass of the train roughly corresponds to the mass of a commuter train made up of a cab car, four coaches, and a locomotive; or made up of six MU locomotives), then the total energy dissipated in an EN 15227 Collision Scenario 3-impact is 5.0 million foot-pounds. The total energy absorbed in the collision scenarios included in this final rule are 135,000
foot-pounds for the collision post and 120,000 foot-pounds for the corner post. However, in the European standard, the impacted object is deformable and potentially absorbs a significant amount of the available energy; in the collision scenarios included in this final rule, the object is rigid, and virtually all of the energy is absorbed by the cab car or MU locomotive.
A recent paper describes the performance of the SOA end frame in both the FRA and the EN grade-crossing collision scenarios.
10
Specifically, testing and analysis of the SOA end frame's performance in appendix F's collision post test scenario was compared to an analysis of the SOA end frame's performance in EN15227's Collision Scenario 3.
10
Llana, P., “Structural Crashworthiness Standards Comparison: Grade-Crossing Collision Scenarios,” American Society of Mechanical Engineers, Paper No. RTDF2009-18030, October, 2009. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2009/09-18030.pdf.
Table 1
Table 1 summarizes a few key crashworthiness parameters and results from the testing and analysis conducted. Application of the FRA scenario involved only one car; whereas the EN 15227 scenario involved a complete consist or train unit. The difference in weight of one car, 80 kips, versus that of a complete consist, 767 kips, was an order of magnitude. In the FRA scenario, the 14-kip impact object was tested striking the car at 19 mph, resulting in 170 ft-kips of initial kinetic energy. Whereas in the EN 15227 scenario, the 767-kip consist was analyzed striking the deformable lorry at 53 mph, resulting in 72,000 ft-kips of initial kinetic energy. The difference in the amount of initial kinetic energy involved between the two scenarios was two orders of magnitude. Similarly, the impacting objects were quite different. As noted earlier, the FRA scenario provides for a rigid impact object; whereas in the EN 15227 scenario, the impact object is deformable. In the FRA scenario, this resulted in the energy being mostly absorbed by the impacted collision post, with virtually no energy absorbed by the impact object. Whereas in the EN 15227 scenario, both the first car and the impact object absorbed large amounts of energy, with very little energy absorbed by one collision post.
Table 1—Comparison of SOA End Frame Performance Applying Appendix F Collision Post Standard and EN 15227 Collision Scenario 3
Parameter
Application of Appendix F collision post standard
Application of EN 15227 collision scenario 3
specification
Type of Train
Single car: 80 kips
Complete train unit: 767 kips.
Impact Object
Rigid cart: 14 kips
Deformable lorry: 33 kips.
Impact Speed
19 mph (cart)
53 mph (consist).
Initial Kinetic Energy
170 ft-kips
72,000 ft-kips.
Energy Absorbed
End frame: 138 ft-kips; Cart: ~0; Collision post: 105 ft-kips
Leading car: 1370 ft-kips; Lorry: 950 ft-kips; Collision post: 89 ft-kips.
Pass/Fail Criteria
Intrusion <= 10 in., no separation
Preserve survival spaces, mean deceleration < 7.5g.
As the table shows in summary form, the key parameters of these two scenarios are very different, though they are both grade-crossing collision scenarios involving rail vehicles with impact objects. Additionally, comparing the complexity of the analysis required for each scenario, application of the FRA scenario is simpler to analyze. In analyzing the FRA scenario, fewer vehicles are involved, initial kinetic energy is lower, deformations are less, and the deformations that result are virtually all in the car and not the impact object.
Overall, FRA believes that the following conclusions can be drawn about the standards in appendix F and those specified in EN 15227's Collision Scenario 3. The appendix F standards concentrate the load on a single post, above the underframe; can be applied to both CEM and non-CEM equipment; and can potentially be used to demonstrate compliance either through analysis or testing. The EN 15227 grade-crossing collision specification distributes the load across the entire end structure; imparts a significant amount of load in the underframe and roof structure; assumes the use of CEM equipment; and can be used to demonstrate compliance through analysis only. Moreover, FRA believes that its dynamic collision scenario is not only easier to analyze, but easier to test than the EN 15227 scenario and imparts more energy to the impacted post than in the EN 15227 scenario.
IV. Discussion of Specific Comments and Conclusions
As noted above, FRA received written comments on the NPRM from representatives of government; various organizations, including railroad labor; railroads; railroad car manufacturers; railroad engineering firms; and as well as private citizens. The comments can principally be divided into two groups: comments of a technical nature affecting the substance of the requirements proposed, and comments as to the preemptive effect of the requirements proposed. FRA found that these groupings serve the organization of this final rule, even though some comments do not fit neatly into either grouping. Please note that certain comments are not discussed in either of these two groupings; instead, they are discussed directly in the Section-by-Section Analysis or in the Regulatory Impact and Notices portion of this final rule.
A. Technical Comments
This section contains the discussion of technical comments on the NPRM, as well as comments closely associated with these technical comments. FRA has endeavored to group the comments together by issue to the extent possible, rather than by commenter. Please note that the order in which the comments are discussed, whether by issue or by commenter, is not intended to reflect the significance of the comment raised or the standing of the commenter.
Please also note that following the submission of these written comments, FRA convened the Task Force and Working Group to consider and discuss the comments and to help achieve consensus on recommendations for this final rule. As a result, certain of these comments have been superseded by changes made in the rule text from the NPRM to this final rule, and they should not necessarily be understood to reflect the positions of the commenters with
respect to the requirements of the final rule. Nevertheless, FRA is setting out all of the comments received and is responding to each of them here so that FRA's positions are clearly understood.
1. Crash Energy Management
Caltrans raised concern with FRA's mention of CEM designs in the NPRM, believing that no rail equipment that features a CEM design has been built, that including CEM in the preamble implied that the NPRM included a CEM requirement, and that the implication that CEM designs may provide for a higher level of safety would expose those railroads not employing CEM designs to litigation for not selecting the “safer” design as identified by FRA.
FRA notes that Amtrak's Acela Express trainsets use CEM, and CEM is used in European and other vehicles. FRA does believe that, all other things being equal, CEM designs are superior in crashworthiness to conventional designs. Yet, as FRA stated in the preamble to the NPRM, FRA's recognition that fuller application of CEM technologies to cab cars and MU locomotives could enhance their safety would not nullify the preemptive effect of the standards arising from the rulemaking. FRA continually strives to enhance railroad safety, has an active research program focused on doing so, and sets safety standards that it believes are necessary and appropriate for the time that they are issued with a view to amending those standards as circumstances change. FRA has imposed, and will continue to impose, the requirements that it deems necessary for the safe operation of cab cars and MU locomotives in all of the configurations in which they will be operated. FRA is not requiring CEM in this final rule.
RVB also raised concerns with the NPRM for its application to CEM designs. RVB asked why the “static strength” requirements had to be met if the CEM requirements for energy absorption are met. RVB stated that the required amount of energy can be absorbed by CEM structures using considerably smaller collision and corner posts.
FRA understands that there are potential alternative arrangements using CEM that may place the end frame structure outboard of the crush elements or behind the crush elements. If the end frame is situated outboard of the crush elements (or crash energy absorbers), then the end frame will likely serve as the means for assuring planar introduction of the load into the crush elements, allowing them to react in a progressive, controlled collapse. To accomplish this energy transfer to the crush elements, the end frame must be very rigid, which can make meeting the severe deformation requirements for the end frame more difficult to achieve. Nonetheless, as long as the system of structural and CEM elements protecting the occupied volume performs well under the dynamic performance requirements provided in appendix F of this final rule, FRA is confident that sufficient protection is provided to passengers and crewmembers alike. For end frame members inboard of the crush elements, it is likely that they will serve as the reaction points for the crush elements. As in the case of end frame members outboard of the crush elements, to support the load introduced by the crush elements the end frame may have to be very rigid. As a result, meeting the severe deformation requirements for the end frame may also be more difficult to achieve. Yet, again, as long as the system of structural and CEM elements protecting the occupied volume performs well under the dynamic performance requirements provided in appendix F of this final rule, FRA is confident that the system provides sufficient protection for passengers and crewmembers.
Additionally, FRA would like to make clear that the energy-absorption requirements in this rulemaking should not be confused with energy absorption as part of a CEM approach. While inclusion of energy-absorption requirements is consistent with FRA's approach to incrementally build on traditional crashworthiness requirements, and whereas CEM is an advanced crashworthiness approach, FRA did not intend that the energy-absorption requirements in this rulemaking be considered part of a CEM approach. Instead, FRA's inclusion of energy-absorption requirements in this rulemaking is intended to address traditional cab car and MU locomotive designs that have very strong underframes with relatively weaker superstructures, for which it is vitally important to provide protection to crewmembers and passengers in the event that the superstructure is impacted. FRA is incorporating mature technology and design practice to extend from linear-elastic requirements to elastic-plastic requirements together with descriptions of allowable deformations without complete failure of the system.
RVB additionally commented that in the NPRM the collision and corner posts must be designed for yield strength in the case where the posts are behind the CEM structure and used as support for the CEM structure. RVB believed that this proposed requirement conflicted with the allowance in the NPRM for the posts to resist loads up to their ultimate strength. RVB believed that, by requiring yield strength in such case, the ultimate strength of the post would be much greater than the amount specified.
FRA understands the complexities introduced by using a CEM design that behaves significantly differently than a conventional cab car or an MU locomotive because of its crush zone(s). This is one of the reasons FRA proposed the option to test such designs dynamically, and one of the reasons why FRA has included alternative, dynamic performance requirements in this final rule. FRA has modified the dynamic performance requirements in the final rule from those proposed in the NPRM, and FRA believes that these modifications will help to address concerns with applying the requirements to CEM designs.
RVB also commented that since, by definition, a CEM system requires a structure that facilitates controlled collapse of the crush zone(s), the proposal would result in a much higher load imparted to the underframe than by the 800,000-pound compression load requirement, exceeding the yield strength of the structure. RVB claimed that this was another area of significant over-design that was unaddressed in the NPRM. RVB added that by disallowing correction of static strength requirements as they are taken up by CEM systems, a vehicle would be heavier than it needs to be, use more energy to operate, and exert more force on wheels and rails that would increase maintenance costs for equipment and track.
FRA believes that the commenter is incorrect in its assertions. FRA agrees that for CEM designs the overall average load that the structure must resist may exceed 800,000 pounds. However, this load is typically spread over a significantly larger area than just the line of draft of the vehicle, as specified for vehicles not utilizing CEM designs. Because the capacity of a vehicle incorporating a CEM design to resist compression loads elastically may be taken into account, FRA does not believe that this will result in over-design of the vehicle. In addition, FRA wishes to dispel the belief that a heavier vehicle would be necessary to meet the requirements proposed in the NPRM and those contained in this final rule. Crashworthiness features from clean-sheet designs can occupy the same space as other material and not weigh in excess of the structure(s) being replaced. There is considerable leeway in
designing such systems so that no additional weight is required. Moreover, the vehicle body structure itself typically accounts for only between 25 to 35 percent of the final weight of a vehicle, which minimizes the significance of any weight added to the vehicle to comply with the requirements of this final rule.
RVB further commented that one means of recognizing a CEM vehicle addressing the static end strength requirements would be for this part 238 to specify the minimum amount of energy that must be absorbed by each end of a vehicle in a train in a specified collision scenario. According to RVB, dynamic testing of the entire crush zone or testing of the critical crush zone elements, in conjunction with suitable analysis, would be required to confirm compliance, and acceptance criteria would include verification that (i) the required minimum energy has been absorbed, (ii) the occupied volume is not compromised, and (iii) climbing/telescoping does not occur under the collision scenario. For a CEM vehicle, RVB believed that this should be in place of the specific strength requirements for the collision and corner posts, and allow evaluation of the car ends as a system.
FRA recognizes the possibilities raised by the commenter. FRA intends to work with the APTA PRESS C&S Subcommittee to consolidate knowledge gained from the Metrolink CEM design effort to support development of such criteria. Inclusion of such criteria in this part 238 would be the subject of a separate rulemaking activity, however, and such criteria are not included in this final rule.
RVB additionally commented that the NPRM suggested that a manufacturer with a CEM system may choose to conduct two dynamic tests instead of conducting quasi-static tests on the individual components. RVB believed the practical situation is that the structure needed to support the CEM system would almost certainly meet the quasi-static requirements proposed in the NPRM. According to RVB, if a dynamic test were to be conducted for a CEM system, it would seem to serve the public better to conduct a dynamic test that verifies the performance of the entire CEM system, not just for how it protects against a steel coil.
As noted above, FRA plans on working with the industry to address the issue of more comprehensive requirements for CEM systems. However, with regard to specific application of the requirements of this final rule, a dynamic test of a CEM structural system as contemplated by the commenter may not in itself demonstrate that the superstructure has the strength to protect against the collision scenarios addressed in this rulemaking. In such a dynamic test of a CEM structural system, the entire end structure of the vehicle would potentially absorb all of the collision load. Yet, this final rule specifically targets grade-crossing collision scenarios where only portions of the superstructure are loaded. It is therefore believed that analysis and component testing, not a full-scale test alone, would be necessary to verify the design of a complete CEM system.
In its comments, RVB stated that the NPRM introduced requirements that would make manufacturers design to the actual strength of some components rather than rely on the yield stress as a measure of strength. RVB believed that this approach is sensible, particularly as CEM systems are introduced, in that such systems rely on controlled (plastic) deformation and operation at the maximum strength (load) capacity of structural members in collisions. Nevertheless, RVB believed that there are still numerous transportation requirements that are based on yield strength and that these impose constraints on the design of CEM members that may not be sensible, including the anti-climbing arrangement and the collision and corner post load cases for application points well above the underframe. According to RVB, FRA should consider moving to a true strength approach for all components as it stated is being done in much of the structural engineering community.
FRA notes that the commenter is focused on CEM systems for which the rule will probably not be applied for some time, and, if sooner, for systems FRA would have to review individually because such systems are sufficiently different from conventional designs. The requirements based on yield strength work well for non-CEM designs and facilitate their testing and use.
RVB also commented on FRA's statement in the NPRM that an energy- absorption requirement of 5 megajoules (MJ) will effectively prevent a cab car from being used in the lead position for Tier II equipment. RVB believed that this magnitude of energy absorption is feasible for cab cars.
FRA recognizes that advancements have been made in the ability of CEM systems to absorb energy. However, FRA continues to believe that for operational speeds in excess of 125 mph, as a rule of general applicability for our nation's railroads, no passengers should be allowed in the lead vehicle. Tier II passenger equipment can operate at speeds where the amount of energy required to be dissipated is too large for any vehicle design to survive a direct impact. Yet, with use of advanced system designs such as Positive Train Control (PTC) and CEM, the risk may potentially be minimized, and FRA would consider such cases individually in the context of the particular environment in which the equipment would operate.
In its comments on the NPRM, Caltrain stated that it would be far more appropriate for FRA to define a risk assessment methodology and prescriptions for addressing risk, letting designers provide alternatives such as CEM that deliver the required performance. Caltrain asked why a collision post inboard of a CEM system would be required to resist the same load as a collision post where there is no CEM system. Caltrain stated that presumably the load would be reduced as the CEM system performs its function, so that a substantially lighter collision post could be used to protect the passenger space, if the CEM system does not otherwise eliminate altogether the need for an interior collision post. Caltrain believed that if it is the intent of FRA to provide this level of flexibility, FRA should make this clear.
It is indeed FRA's intent to provide flexibility for vehicle designs with CEM features. In the final rule, FRA has added appendix F to part 238 to provide dynamic performance requirements as alternatives to both the collision and corner post quasi-static requirements. These dynamic performance requirements specify the performance of the end frame, were prepared with CEM designs in mind, and provide the designer leeway in choosing how that performance will be achieved. Nonetheless, FRA is not defining a risk assessment methodology and prescriptions for addressing risk, as an alternative to the collision and corner post quasi-static requirements. FRA believes that appendix F to part 238 provides the flexibility needed while assuring safety with more certainty than by performance of a risk assessment alone.
2. Dynamic Performance Requirements
FRA received a number of comments on its proposal to include dynamic performance requirements as an option to demonstrate compliance with the severe deformation requirements for collision and corner posts. In addition to inviting general comment on the proposal, FRA invited specific comment on the dynamic testing collision scenarios included in the proposed rule, including suggestions for any alternative
collision scenario or way to address possible future designs. FRA also invited specific comment whether this final rule should provide for all cab cars and MU locomotives to be tested dynamically to demonstrate compliance—whether or not they have a shaped-nosed design or a CEM design—and, if so, whether the collision scenarios included in the proposed rule are appropriate or whether another collision scenario would be.
CPUC supported FRA's intent to allow full-scale crash testing as an alternative to quasi-static testing to determine the crashworthiness of a prototype cab car or MU locomotive. APTA expressed support for FRA's approach to bring the Federal structural requirements for cab cars and MU locomotives up to current industry standards, including quasi-static tests with specific pass/fail requirements to demonstrate the ability of collision and corner posts to undergo severe deformations prior to failure. (APTA did advise that FRA make sure to reference in the preamble and section-by-section analysis APTA's most current industry standard, APTA SS-C&S-034-99, Rev. 2—not Rev. 1.) APTA appreciated FRA's concern that future vehicles utilizing CEM designs may require different treatment in Federal structural regulations than those with traditional flat-nosed designs. However, APTA had several concerns about including the proposed dynamic test option to accommodate such designs in the final rule. Noting that FRA has conducted an extensive full-scale collision test program to gain confidence in predictive, finite element analysis models and to support development of industry standards and rulemaking, APTA believed that FRA should not include a dynamic test scenario in the regulation unless and until similar testing supports it. APTA urged FRA to conduct appropriate testing and defer inclusion of dynamic testing in the regulation, even as an option, until those test results are available and validate the model.
As discussed in the “Technical Background” portion of this preamble, the testing described by APTA has been completed. In 2008 a full-scale dynamic test and two full-scale quasi-static tests were performed on the posts of an SOA end frame. These tests were designed to evaluate the dynamic and quasi-static methods for demonstrating energy absorption by—and graceful deformation of—the collision and corner posts. FRA believes that these tests support inclusion of the quasi-static and dynamic performance requirements of this final rule and address APTA's concerns.
APTA also mentioned that in the NPRM FRA stated that alternative, dynamic performance requirements are necessary because shaped-nose designs may not have readily identifiable, full-height corner and collision posts. APTA stated that, although FRA referred to the CRM and Rotem designs as potential examples of shaped-nose designs, both these designs include easily identifiable, full-height collision and corner posts behind the shaped nose. According to APTA, all evidence points to having collision and corner posts up to their full height as key design features to protect the engineer and passengers from front-end collisions.
FRA believes that the dynamic performance requirements in this final rule allow in particular for innovative designs that protect the occupied volume for its full height, even without what would be identified as full-height collision and corner posts. Whether or not the Rotem and CRM designs have full-height collision and corner post structures does not address FRA's underlying concern that the requirements in this final rule would otherwise be too restrictive without the alternative standards based on dynamic testing. For instance, the Stadler Rail equipment procured by the Capital Metropolitan Transportation Authority (CMTA) in Austin, TX, has no readily identifiable collision or corner post structures and yet has been found to behave well under analysis using the dynamic performance requirements in this final rule. By not allowing for such design innovation, potential use of alternative designs that could demonstrate compliance would be unnecessarily restricted.
Further, APTA questioned the safety implications of allowing such key features as full-height collision and corner posts to be optional. APTA stated that all the full-scale testing done by FRA, all the model-validation testing, and all the knowledge gained of how the end frame performs in collisions pertain to equipment with these design features. Until such safety implications are better understood, APTA believed the inclusion of alternative, dynamic performance requirements to be premature. Overall, APTA was not convinced that the proper foundation has been established for adding these dynamic performance requirements to the final rule, nor was APTA convinced that a single dynamic test demonstrates full equivalency for the range of protections provided by traditional full-height collision and corner posts.
As provided in the final rule, FRA makes clear that the occupied volume must be protected for its full height, utilizing either the quasi-static or the dynamic performance requirements. FRA expects that for traditional flat-nosed designs, the occupied volume will be protected for its full height by means of full-height collision and corner posts. Yet, for other designs, this protection of the occupied volume for its full height could be achieved by the performance of the entire end frame acting together to prevent intrusion and absorb energy. FRA believes that there are many potential ways of providing protection for the full height of the occupied volume, and this is reflected in the final rule.
In its comments on the NPRM, RVB stated that use in dynamic testing of a proxy object that is essentially a steel coil has a historical basis resulting from only a few accidents. RVB believed that the European approach of using a proxy vehicle would be more sensible and that it was not clear why FRA would resist adopting aspects of that approach that are in widespread use in Europe and other countries.
As discussed earlier, FRA notes that use of a proxy object that deforms (a deformable lorry,
e.g.
) adds undue complexity to the analysis of impacts. In addition, development of a proxy object with a repeatable crush response is, in itself, a daunting task, and the cost of developing such an object for each car manufacturer is not cost beneficial. Nevertheless, FRA has modified from the NPRM the manner in which the dynamic testing is conducted, to address related concerns about use of the proxy object. Further, FRA believes that the grade-crossing collision scenarios on which the dynamic testing is based challenges the end frame members in a way that can clearly demonstrate the ability of the end frame to resist significant impact loads.
RVB also commented that it was unclear why FRA decided to position the proxy object 19 inches from the car center in the collision post dynamic test. RVB stated that not all collision posts are located 19 inches from the centerline, and believed it would seem better to center the proxy object at the post itself.
FRA notes that the location of the collision posts is dictated by the need to place the posts at the one-third points laterally, along the end of the vehicle. With this in mind, positioning the proxy object 19 inches from the car center is intended to engage the end frame where the collision post structure will be. Nevertheless, because the alternative, dynamic performance requirements
more fully test the end frame as an integrated whole rather than as individual structural elements, and are not intended to test the strength of an individual element quasi-statically, it is not necessary to specify that the impact be centered on the collision post structure.
RVB further commented that the NPRM seemed to impose essentially the same energy-absorption requirements on both the collision and the corner posts in the alternative, dynamic performance requirements, and RVB was unclear if this was FRA's intent. RVB claimed that there is practically no difference between the 20 and 21 mph impact speeds that were proposed for the dynamic performance requirements, asserting that the target speeds used for actual testing would need to be higher than these values to ensure that the speeds are achieved.
FRA notes that in conducting a dynamic test there are alternative means of imparting impact energy into the front end of the cab car or MU locomotive. Speed is only one of the elements that make up impact energy. FRA has taken this fact into account in preparing the final rule and restated the dynamic performance requirements in terms of the amount of collision energy imparted. No specific test speeds are stated. Yet, the amount of collision energy is specific for each test of the two types of post structures, and each amount of collision energy was carefully chosen based upon input from industry stakeholders. FRA makes clear that it is not necessary to impart higher levels of energy than specified in this final rule to assure that the requirements are met. Of course, these requirements are minimum standards and may be exceeded by the manufacturer.
Additionally, RVB commented that the top of the deformable anti-climber of the FRA CEM-design is approximately 24 inches above the top of the underframe. RVB believed that an impact with a circular proxy object centered 30 inches above the top of the underframe, as proposed in the NPRM, could result in a ramp and alter the trajectory of the object in an undesirable manner. As a result, RVB believed it unclear how much energy would actually be imparted as intended to the structural elements, and that it may not be prudent to conduct a dynamic test in this manner for such a design to demonstrate its compliance.
FRA notes that the FRA CEM-design is intended to act as a complete system so that even if a ramp were to form on the deformable anti-climber, the end frame structure would be able to resist intrusion by the proxy object into the occupied space of the vehicle. The deformable anti-climber can absorb a significant amount of energy prior to bottoming out even when loaded in an offset manner. Nevertheless, to minimize the potential for off-axis rotations, FRA has reconsidered use of the standing proxy object specified in the NPRM to be struck by a moving cab car or MU locomotive, and has specified instead use of a proxy object connected to a moving crash cart to strike a standing cab car or MU locomotive.
In its comments on the NPRM, Caltrain raised concern with the testing performed by FRA to validate the effectiveness of the proposed collision and corner post requirements. Caltrain stated that the 1998 NICTD grade-crossing accident in Portage, IN, was recreated with a 40,000-pound steel coil at an impact test speed of 14 mph. Caltrain stated that the test speed used to recreate this accident was far lower than in most grade-crossing accidents, and that the test did not actually compare the proposed design to one that was compliant with part 238. Caltrain believed that data from a higher-speed test, using equipment that is compliant with part 238, would be more useful in evaluating potential solutions.
As discussed earlier, the SOA design is compliant with part 238 and has been tested. Further, the test cited by the commenter was carefully designed to overload only the structure of interest, and was not intended to replicate the actual collision speed. Moreover, FRA emphasizes that in this rulemaking the agency is taking an incremental approach to improving safety by enhancing the current end frame design of cab cars and MU locomotives. As noted, FRA is separately exploring the application of CEM to provide protection against even higher speed events.
In its comments on the NPRM, Caltrans stated that any dynamic testing requirement, even as an option, should be founded in actual testing and validation of the variables and proposed design criteria. Caltrans mentioned that although FRA has conducted tests that simulate a collision with a highway vehicle carrying a roll of coiled steel, the actual tests as conducted had significantly lower impact speeds and greater allowable deformation requirements. Caltrans maintained that until a real-time crash test has been conducted and analyzed by FRA that uses identical testing variables, inclusion of a standard for dynamic testing of end frame designs is premature.
FRA notes that the energy involved in the earlier testing supporting the NPRM was in fact equivalent to that proposed in the NPRM. Nevertheless, additional dynamic testing has been performed in support of the requirements in this final rule. Specifically, as discussed in the “Technical Background” section, a dynamic test was successfully conducted on April 16, 2008, and the dynamic performance requirements in this final rule are based on the actual test conditions and amount of collision energy imparted.
Caltrans also commented that FRA needs to clarify whether full-height collision and corner post tests are required if the alternative, dynamic performance requirements are used, and if not, whether FRA has performed a structural analysis showing that safety may be maintained in the absence of full-height posts. Caltrans cited FRA's statement that dynamic testing is essential as an option for validating car designs that feature non-flat front ends. Yet, Caltrans believed that current car designs that feature non-flat front ends, CRM's diesel MU locomotive and Metrolink's new Rotem cab car, both feature full-height collision and corner posts.
FRA makes clear that the fact that testing collision and corner posts dynamically is provided as an alternative in the final rule does not mean that protecting the full height of the occupied volume is optional under such circumstances. For traditional end frame designs (
i.e.,
flat-nosed designs) tested dynamically, full-height collision and corner posts are certainly not optional. Yet, FRA believes that the rule must continue to allow flexibility for other design approaches that may use different shapes and structures to protect the full height of the occupied volume. For example, FRA notes that novel designs may effectively prevent intrusion into the occupied volume through application of the concept of deflection—to deflect objects away from the vehicle. For such design approaches, full-height collision and corner posts are not necessarily required, provided, of course, that the occupied volume is nonetheless protected for its full height. FRA has conducted analysis to show that safety can be maintained in the absence of full-height collision and corner posts. Manufacturers attempting to meet the requirements of this final rule must perform the detailed structural analyses to show that safety is maintained in the absence of these structures.
In its comments on the NPRM, Bombardier raised a number of concerns with the proposal to include an option for a dynamic method of demonstrating compliance with the proposed severe-
deformation requirements for collision and corner posts. Bombardier believed the proposal to be contrary to the recommendation of the Task Force in developing the NPRM. Bombardier stated that it supported the general industry consensus that such dynamic performance requirements should not be included as an option, contending that the proposed dynamic tests were impractical, had not been fully validated, did not adequately test a realistic production design end structure, raised safety concerns, and would be costly. FRA will address each comment in turn.
Bombardier stated that due to the significantly higher static load design requirements for collision posts (compared to corner posts), collision posts would be much more substantial in size and strength than corner posts. However, because the proposed dynamic test defined only a 1.0 mph difference between the impact speeds to test both collision and corner post structures, Bombardier believed this illustrated the sensitivity in the size of the post required to resist such a small increase in impact velocity. According to Bombardier, a 1.0 mph difference in test speeds would approach the accuracy achievable for a full-scale impact test, and, from a practical perspective, would create various technical and commercial problems, most likely require re-testing if the actual test speed were only marginally above or below the target speed. For instance, Bombardier claimed that if the actual impact speed during the test of a corner post were 1.0 mph above the target speed for corner posts (
i.e.,
at the impact speed required to qualify a collision post) there would be a high probability that the corner post would fail and a re-test of another production end frame would be required. Similarly, Bombardier maintained that if the post were tested at a speed slightly below the target value, it may not absorb the energy required in the proposed regulation and, again, a re-test would likely be required to verify compliance.
FRA notes that the dynamic performance requirements proposed in the NPRM were intended to be both practical and achievable, as illustrated by the fact that the proposed quasi-static requirements would have required the same levels of energy absorption. These levels of energy absorption were chosen after comparing the performance of the FRA-developed, SOA end frame with a production model tested by the commenter. Moreover, the commenter worked in conjunction with FRA and the Volpe Center to assess the degree of incremental improvement that is reasonably achievable for collision and corner posts, and a paper was published on this topic. (
See
“Review of Severe Deformation Recommended Practice Through Analyses—Comparison of Two Cab Car End Frame Designs,” cited above.) There are various ways to achieve the impact speeds with the precision required for either the proposed collision post or corner post tests, and the speeds were intended to be minimum speeds that could be exceeded by the manufacturers (as FRA's requirements are safety minimums). Nonetheless, FRA has revised the dynamic performance requirements in this final rule to state the requirements in terms of collision energy rather than collision speed. Like the collision speeds proposed in the NPRM, the specified levels of collision energy may also be exceeded.
Bombardier also commented that, while FRA had conducted analysis to determine the severe deformation characteristics of a collision post, no dynamic testing had been conducted to verify the acceptability or practicality of the dynamic test proposed for collision posts. Bombardier stated that, while a dynamic test had been conducted on the SOA corner post, that test used a significantly different proxy object mass (40,000 lbs vs. 10,000 lbs) and different impact speed (14 mph vs. 21 mph) than that proposed in the NPRM. Bombardier maintained that, although FRA analysis showed these to be “equivalent” tests, the actual qualification test proposed in the NPRM had never been validated. Bombardier compared this situation to the proposed changes to the large-object impact test for forward-facing glazing, which the Task Force separately considered, stating that FRA predicted that a test based on energy using a different mass and impact speed would be equivalent to the current glazing requirements but that subsequent tests that were conducted at the request of industry to validate the proposed requirement confirmed that the proposed tests were not equivalent. Therefore, Bombardier contended that until FRA conducts and validates the proposed dynamic tests for both a collision post and a corner post on a production-model end frame, it would be premature to include such requirements in this part.
As discussed in the “Technical Background” section, FRA makes clear that the testing cited by the commenter was completed successfully on April 16, 2008, following submission of these comments. The collision post and the entire SOA end frame performed well under the impact conditions prescribed and maintained the requisite safe volume for the locomotive engineer. Equivalency of the testing has been validated.
11
With regard to glazing, FRA believes that a fuller discussion of glazing is necessary in a separate forum, including a discussion of the glazing testing cited by the commenter and the current glazing test requirements. Nevertheless, FRA does not believe that the agency is required to conduct such testing on a production design. FRA does have the responsibility to demonstrate that the rules to be imposed on the industry are achievable and do not impose undue economic costs. Yet, this can be accomplished in different ways, including engineering analysis, prototype testing, and analysis of information provided by the industry on its production designs. This process was followed in the development of the proposed performance standards supporting this final rule.
11
Priante, M., Llana, P., Jacobsen, K., Tyrell, D., Perlman, A.B., “A Dynamic Test of a Collision Post of a State-of-the-Art End Frame Design,” American Society of Mechanical Engineers, Paper No. RTDF2008-74020, September 2008. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2008/08-74020.pdf.
In addition, Bombardier commented that on several occasions industry members pointed out to FRA that, while the full-scale test of the SOA corner post design was valuable to validate specific design features and characteristics, the SOA design did not fully represent a production design. Bombardier stated that on a production-version end frame (flat-nosed), the corner post is set back from the collision post in the longitudinal direction by about 6 inches to accommodate car clearance during curve negotiation, and both the collision and corner posts are connected laterally by the lateral shelf and bulkhead. According to Bombardier, this arrangement would cause the proxy object to impact the structure between the collision and corner posts, rather than directly impact the corner post, in a dynamic test of a production-model corner post. Bombardier likewise believed that for a flat-nosed cab car, the proxy object would impact the structure between the collision and corner posts at 18 inches from the outside of the vehicle, instead of on the corner post (stating,
e.g.,
that the coil would contact the sheathing on a flat-nosed cab car about 4
1/2
; inches ahead of the corner post), and that this would be greater for a non-flat-nosed car. According to Bombardier, this would result in both the collision and corner posts sharing the impact load and that it would therefore be possible to design a structure with a weaker corner post than
would be required to meet the quasi-static requirements.
As FRA has noted, FRA intends that the dynamic performance requirements be applicable to end frame designs that may not have identifiable corner post or collision post structures. For such designs, it is expected that the end frame will act more as an integrated whole in resisting an impact load, rather than having one structural element to resist the load by itself. Nonetheless, the final rule directs that the impact loads be applied to the end frame at the corner post and collision post locations. FRA does note that use of a crash cart to impart these loads is not specifically required by this final rule (even though FRA generally assumes that a cart will be used for purposes of the discussion in this preamble and in the examples provided in the rule text). Use of a crash cart is intended to help achieve a more repeatable testing methodology and better focus the impact loads than through use of the proxy object proposed in the NPRM, but allowance is provided for variation in the test set-up so that a car builder may tailor a test in a way that is best suited for a particular design within the requirements specified.
Bombardier further commented that, as FRA noted in the NPRM, industry members had raised concerns regarding the safety of conducting full-scale, dynamic testing of collision and corner posts. While these members acknowledged that all testing, including that required for quasi-static testing, requires attention to safety, Bombardier believed that it is much easier to manage the safety of a quasi-static test, which is conducted in a controlled lab/shop environment, than the type of dynamic tests proposed in the NPRM. Noting that during the dynamic test of the SOA corner post one side of the vehicle completely lifted off the rail, Bombardier raised concern about the potential likelihood and consequence of a derailment occurring in a dynamic test of a production-design vehicle at a higher speed, especially one with a shaped-nose. Bombardier believed that there would be particular safety concern in conducting the proposed dynamic test because the 10,000-pound proxy object would be positioned between the rails directly in front of the test vehicle and fall directly in front of the vehicle. Bombardier therefore stated that it would be premature to include the proposed dynamic tests in a Federal regulation, until FRA conducts and validates the safety of these tests on a collision post and a corner post for both a flat-nosed and a shaped-nose, production-model end frame.
As discussed earlier, FRA has modified the alternative, dynamic performance requirements in this final rule so that the testing methodology is safer and more repeatable. Specifically, FRA has modified the testing methodology so that the proxy object is set in motion to strike a standing cab car or MU locomotive. The resultant speed of the cab car or MU locomotive from being struck by the object is expected to be approximately 3 mph. Even if a cart connected to the proxy object should derail during the test, the cart is much lighter than a cab car or MU locomotive, and would present a much lesser safety hazard than would a derailment of those heavier vehicles. FRA believes that this revised test methodology sufficiently addresses the safety concerns raised by the commenter.
Bombardier also commented that while the NPRM indicated that a dynamic test option is needed to address cars with shaped noses or CEM designs, or both, all of the analysis and testing that had been conducted had been directed to assure that flat-nosed cab end structures undergo “graceful,” severe deformation and maximize the energy absorbed by the post structure before total failure of the top or bottom post connections occurs. Bombardier believed that utilizing a dynamic test to validate a shaped-nose design significantly deviates from the original intent of the severe-deformation requirements. According to Bombardier, shaped-nose designs would inherently be much stiffer than flat-nosed designs, and as a result would have a much greater tendency to deflect the proxy object rather than absorb the energy through severe structural deformation. Bombardier therefore maintained that the proposed dynamic test option would not be a measure of the severe-deformation performance of shaped-nose designs. Additionally, Bombardier stated that CEM designs would have well-defined, severe-deformation requirements that typically require significantly more energy absorption than that defined in the NPRM for collision and corner posts, and as such, requiring the proposed dynamic (severe-deformation) test option would be redundant. Consequently, Bombardier recommended that the proposed requirements for the dynamic test option be deleted and that the proposed quasi-static test requirements for the collision and corner posts be retained for only flat-nosed designs.
FRA notes that the goal of dynamic testing is preservation of a survivable space for the train crew and passengers. Flat-nosed designs must be able to absorb energy and deform gracefully because these designs are inherently required to interact with objects that threaten the superstructure of the car. Yet, FRA disagrees with not allowing the industry the alternative to use dynamic performance requirements. A dynamic test does not have to be conducted—it is provided as an alternative to demonstrate compliance. There are certain designs for which it would be difficult, if not impossible, to test quasi-statically, such as the Stadler Rail equipment procured by the CMTA. Moreover, for a quasi-static test in which the front end of the car is not flat, or the post is not centered on the specified impact point, applying a high force could cause the impactor shape to shift vertically or laterally, when all it should do is move longitudinally. The benefit of a dynamic test as an alternative is that the force would be applied quickly and the test could be conducted properly, even if the cart moved laterally or vertically and derailed.
Bombardier also commented that it did not agree with the justifications outlined in the NPRM for including alternative, dynamic performance requirements. Bombardier stated that there was significant discussion in the NPRM about CEM and European standard EN 15227, Crashworthiness Requirements for Railway Vehicle Bodies, and its four collision scenarios. Bombardier believed that extreme care must be taken when comparing such a European standard with the severe-deformation requirements proposed in the NPRM and in the current APTA standards. According to Bombardier, FRA must clarify that EN 15227 is a standard for the qualification of a CEM system, where a large quantity of energy is absorbed, and not a severe deformation standard for collision and corner posts where a very small amount of energy absorption is required. However, Bombardier did agree that the approach in the European standard should be taken into consideration at the time when CEM standards are developed for North American application.
FRA believes that it was appropriate in the NPRM to reference the European standard and its adoption of dynamic test standards. FRA did not intend to indicate that the European standard was comparable to the dynamic performance requirements proposed in the NPRM, and FRA did highlight several differences between them. As noted above, FRA has made a more technical comparison of the European deformable-lorry requirements and the dynamic performance requirements in this final rule. This effort involved
taking FRA's prototype end frame design and using finite element analysis to compare its performance with the European specification and the final rule's requirements. Significant differences were found between the rule's dynamic performance requirements and those described in the European standard, including: the safety of conducting such testing, the repeatability of the results obtained, the ease of analysis, and the focus on the performance of the superstructure of the cab car or MU locomotive. The FRA dynamic performance requirements entail lower amounts of collision energy designed to provide repeatable results under conditions that are readily analyzable with a clear means of assessing adequate performance. The same was not found to be true of the European standard.
In its comments on the NPRM, CRM raised concern with actual dynamic testing of collision and corner posts using curved-shaped equipment, believing that the curved shape can be addressed in a quasi-static test but that the results would likely differ with those from a dynamic test.
FRA notes that, although the manner of load application can vary, dynamic testing provides immediate feedback as to how the tested structure will perform in an actual collision. Quasi-static testing of a shaped structure has to simplify for how the load enters the structure and reacts; consequently, the test results may not be truly reflective of actual performance. For this reason, FRA believes that the alternative, dynamic performance requirements in this final rule are better applicable to non-traditionally-shaped cab cars and MU locomotives.
CRM also commented that the dynamic testing proposed for the corner post of an aerodynamically-shaped car would impart larger lateral and vertical loads on the corner post than on the collision post.
As FRA has noted, the dynamic performance requirements included in this final rule facilitate testing of end frame designs without readily identifiable collision or corner post structures. In this light, instead of focusing on whether an individual corner post or collision post structure is capable of resisting an applied load, the focus is more appropriately placed on the ability of the end frame structure as an integrated whole to withstand the impact. In fact, the end frame may be intentionally shaped to deflect a striking object, which would be an acceptable means of complying with the dynamic performance requirements.
Additionally, CRM raised concern about the repeatability of energy-absorbing testing, stating that it has found that physical properties, such as yield, can be 30-percent higher than the published minimum. CRM asked if FRA has experience in the repeatability of identical energy-absorption tests with substantially-varying material properties, noting that repeatability studies it had seen were for multiple test samples made with both the same heat and physical properties.
FRA recognizes that material variability is a concern. Manufacturers may need to request that specific material testing be conducted when ordering materials for constructing cab cars and MU locomotives in compliance with this rule. Nevertheless, differences in yield strength are not as important as differences in the elongation to failure of the material, because most of the performance of interest is associated with plastic deformations. FRA has conducted dynamic and quasi-static tests of nominally the same design with varied results in energy absorption. This experience has demonstrated the importance of validating analysis through testing. Small design details can have dramatic effects and should be considered carefully in highly loaded areas.
3. Alternative Corner Post Requirements for Designs With Stepwells
The BLET raised concern with the proposed corner post requirements for cab cars and MU locomotives utilizing low-level passenger boarding on the non-operating side of the cab end. The BLET believed that the proposed requirements for corner post resistance were significantly lower than those for the operating side. The BLET stated that it has consistently voiced the position that current crashworthiness protection for this equipment is so low that the only practical recourse a locomotive engineer has after realizing a collision is impending is to place the train's brakes in emergency and flee the operating cab, running through the car toward the rear. While the BLET did believe that the standards proposed in § 238.213(b) would mark a significant improvement for the engineer's immediate worksite, it believed that lesser, non-operating side requirements in § 238.213(c) would still create a Hobson's choice for a locomotive engineer in the seconds immediately preceding a collision. Claiming that there would be a much greater potential for the non-operating side of the car to deform in such a way as to provide insufficient survivability, the BLET stated that both sides of the equipment should be required to withstand the same level of force. The BLET added that it is noteworthy that the non-operating side of the equipment is typically located on the “railroad” side of the train and that, as a result, impacts on that side are more likely to involve railroad equipment, producing higher collision forces. Similarly, in a frontal raking collision between two trains made of up this equipment, the BLET believed that the two “weaker” corners would meet, with potentially catastrophic consequences for passengers and crewmembers alike. The BLET also stated that the Volpe Center had researched and tested stepwell configurations and determined that it was viable to design a stepwell that was capable of supporting the end/buffer beam so that the non-operating side of the cab could comply with proposed § 238.213(b).
FRA notes that, after a review and analysis of technical information, both FRA and APTA's PRESS C&S Subcommittee determined that the proposed alternative arrangement would provide a level of safety equivalent to that on locomotive engineer's side of the cab end. Moreover, the analysis did not show that an impact on the non-operating side of the cab end would be more likely to spread damage across the full width of the cab end as described by the commenter. Nevertheless, in light of the comments raised, FRA conducted a further review and analysis of the available technical information. That review and analysis reaffirmed FRA's determination that the engineer and other occupants would not be placed at greater risk as a result of the corner post arrangement on the non-operating side of the cab end. FRA has therefore decided to retain this provision in the final rule. However, the final rule contains an additional requirement that FRA review and approve plans for manufacturing cab cars and MU locomotives with this corner post design arrangement. Each plan must detail how the corner post requirements will be met, including what the acceptance criteria will be to evaluate compliance. FRA believes that this close oversight will help to alleviate concerns that the manufactured designs are in any way less safe for crewmembers and passengers to occupy.
Another commenter on the NPRM, Caltrans, expressed its support of the proposed requirement that car designs featuring low-level passenger boarding in an end vestibule opposite from the engineer's seating location have two corner posts on that non—operating side of the car. However, Caltrans stated that the rule must make clear that this requirement applies only to those cars
with a passenger loading stepwell in the same vestibule as the engineer's control location. Caltrans believed that this provision should not encompass its car design where the engineer is located on the second level of the car and the side door is on the opposite side on the lower level.
FRA agrees with the comment raised by Caltrans and makes clear that the provision does not apply to a design where the stepwell and engineer's cab are not in the very same vestibule.
APTA's comments on the NPRM expressed support for the proposal to allow vehicle designs with two corner posts on the non-engineer's side of the cab end. According to APTA, this proposal would allow vehicles to continue to have stepwells for low-platform boarding, which APTA noted is an operational necessity for many passenger railroads. APTA did raise concern that neither the preamble nor the proposed rule text specifically acknowledged that the corner post ahead of the stepwell be allowed to fail when applying the loads to the corner post behind the stepwell. APTA believed that allowing a structural member to fail as part of a test or analysis is an unusual concept for a Federal regulation and that it warrants clear discussion in the preamble.
FRA agrees that testing a post all the way through to complete failure has safety implications and should not be done without thorough analysis first. As noted, FRA has modified this provision to require FRA review and approval of a plan, including acceptance criteria, to evaluate compliance with these corner post requirements. FRA believes that this oversight will help to address the concern raised by the commenter.
4. Use of Testing and Analysis To Demonstrate Compliance
FRA requested specific comment on whether and under what circumstances analysis and scale model or fixture testing might be acceptable to demonstrate compliance with the alternative, dynamic performance requirements. A number of comments were received in response to this request, and in addressing them FRA discusses their application to both the quasi-static and the dynamic performance requirements, as appropriate.
Bombardier commented that the severe-deformation requirements proposed in the NPRM (for either the quasi-static or the dynamic performance requirements) would result in a significant, added cost for cab cars and MU locomotives, particularly as a percentage of the overall procurement cost for small orders. Bombardier contended that if these severe-deformation requirements were truly considered to be safety requirements, then it is imperative that they be required for all new equipment, regardless of the size of the order. Bombardier noted that since the proposed quasi-static requirements were also contained in an APTA standard (APTA SS-C&S-034-99, Rev. 2), the quasi-static requirements would not impose a greater cost burden on the industry than what it already accepts. However, Bombardier maintained that the actual cost to conduct dynamic testing, which would be expected to be done at a location offsite of the manufacturer's facility, would most likely be much greater than for quasi-static testing. Consequently, before any dynamic performance requirements are included in the regulation, Bombardier believed that a proper cost-benefit analysis would be needed and that it was not evident from the information in the public docket that a valid cost-benefit analysis had been conducted. Bombardier noted that the section-by-section analysis seemed to imply that verification of compliance with either the quasi-static or dynamic performance requirements would require an actual test, while the preamble did state that modern methods of analysis can accurately predict structural crush (severe deformation) and consequently can be used with confidence to develop structures that collapse in a controlled manner. Bombardier added that the proposed rule text was itself silent as to whether an actual test would be required or whether analysis could be used to verify compliance with the severe-deformation requirements. Bombardier therefore believed that FRA should clarify what would be required to demonstrate compliance with the severe-deformation requirements and should include the associated costs in the cost-benefit analysis.
FRA notes that it did ask the commenter and other members of the Task Force to provide FRA with estimated costs for each testing alternative for FRA to review. FRA did not receive this specific cost information. FRA agrees with Bombardier that the cost of meeting the quasi-static test requirements is likely not to add to the costs of manufacturing or purchasing new passenger equipment. However, FRA does not agree that the costs of dynamic testing would be greater than the costs of quasi-static testing. Based upon the testing program sponsored by FRA at the TTC in Pueblo, CO, the overall cost of conducting either quasi-static or dynamic testing should be comparable. But even more important, FRA believes that dynamic testing provides at least the same level of confidence in the safety of the equipment tested as through quasi-static testing, and a manufacturer or railroad could voluntarily choose to conduct dynamic testing. The voluntary act of a manufacturer or railroad would provide sufficient evidence that dynamic testing does not materially add to costs, and no specific benefit-cost analysis is needed to provide a voluntary alternative. As FRA has noted, FRA does agree that actual physical testing should be required and that large orders, as well as small orders alike, should undergo actual testing. Yet, as discussed elsewhere in this preamble, FRA does not believe that actual physical testing of a complete, production-design vehicle is required, and FRA recognizes in particular the potential cost of doing so for small car orders.
CRM also raised concerns as to the cost of demonstrating compliance with the regulation to manufacturers of small orders of cab cars or MU locomotives. CRM believed that consideration needs to be given to these manufacturers to protect them from undue financial and schedule hardships.
FRA has taken into account the costs of this final rule to manufacturers of small orders of cab cars or MU locomotives. As noted, FRA believes that for both large and small orders, the manufacturer must perform actual physical testing. However, FRA does not believe that actual physical testing of a complete, production-design vehicle is required. FRA recognizes in particular the potential cost of doing so for small order sizes. Compliance may be demonstrated by a combination of engineering analysis and physical testing on a smaller scale.
CRM further commented that destructive testing could be very expensive. CRM stated that its customers generally order in small quantities, often in the range of two to three cars. According to CRM, producing a 19.25-foot long section of the end of a car for destructive testing would represent a considerable, additional expenditure. CRM therefore requested that FRA clarify that the test sample need not be a large end section of the car, noting that as the NPRM is focused on the post structure and its attachments, the test sample should be limited to just that. CRM nonetheless estimated the costs of quasi-static testing to be approximately $250,000 for each design after a capital expenditure of $75,000 for test fixtures.
FRA agrees that the entire car need not be tested. Bombardier has conducted quasi-static end frame tests where the end of the car was tested only to the body bolster; this would be appropriate. (
See
“Review of Severe Deformation Recommended Practice Through Analyses—Comparison of Two Cab Car End Frame Designs,” cited above.) There are a variety of ways of testing the end frame structure that would not require production of a test specimen of the 19.25-foot size described. Current testing of end frames (both dynamically as well as a quasi-statically) is intended to ensure that the superstructure with some supporting structure can deform gracefully while not allowing permanent deformations in the car body structure too much of a distance behind the connection points. As a result, considerably smaller test articles may be used, provided of course that both the collision post and corner post structures are subject to actual testing. In addition, FRA believes that the costs estimated by CRM for testing are too high, absent more specific cost information from the commenter, and that any expenditure for test fixtures should be a one-time cost that could be spread over many orders.
In addition, CRM proposed that analysis be allowed in lieu of actual testing for orders of less than 50 cars, provided that the analysis methods have been validated by actual testing. In its comments on the NPRM, Caltrain also requested clarification whether actual testing is required to demonstrate compliance, or whether analysis would be acceptable. Caltrain believed that it had been decided that for purposes of complying with the APTA collision and corner posts standards on which this rulemaking is based, current computer finite element modeling methods were adequate to verify design performance, in part due to the cost associated with destructive testing.
FRA believes that there is no substitute for conducting actual testing, as we have seen from the quasi-static test of the collision post that did not meet the energy-absorbing requirement due to the location of a rigid gusset, even though the modeling showed that it would.
12
In particular, because there are always some uncertainties associated with new designs and materials, some degree of testing is required whether for material characterization or sub-assembly testing to confirm that the modes of deformation and failure are modeled appropriately. FRA recognizes that after several designs have been tested and approved, perhaps future designs that are very similar to the older designs could be accepted through analysis only. The individual car builder would still have to demonstrate good experience conducting large deformation analyses, including material failure.
12
Muhlanger, M., Llana, P., Tyrell, D., “Dynamic and Quasi-Static Grade Crossing Collision Tests,” American Society of Mechanical Engineers, Paper No. JRC2009-63035, March 2009. This document is available on the Volpe Center's Web site at:
http://www.volpe.dot.gov/sdd/docs/2009/09-63035.pdf
.
APTA stated that FRA asked for specific comment on whether and under what circumstances analysis and scale model or fixture testing might be acceptable to demonstrate compliance with the dynamic performance requirements. APTA stated that this was a key question, noting that the rule text proposed that compliance “be demonstrated.” APTA believed that either a test or analysis could apparently fulfill the requirement and that there was no indication or guidance of when analysis would suffice in lieu of testing. APTA recommended that, until the industry, in partnership with FRA, can reasonably describe under what circumstances a test must be done and when analysis alone is sufficient, the option for dynamic testing should not be included.
FRA notes that due to uncertainty associated with progression of material failure, some level of actual physical testing is necessary. But this uncertainty is not limited to demonstrating compliance with the dynamic performance requirements; it would also apply for demonstrating compliance with the quasi-static requirements. In this preamble to the final rule, FRA is providing additional guidance in response to similar comments received on the need for and extent of actual physical testing. In general, FRA believes that a combination of actual physical testing and analysis is appropriate to demonstrate compliance with the requirements in this final rule, and FRA encourages manufacturers to approach FRA should they have any questions or concerns about demonstrating the compliance of cab cars or MU locomotives they manufacture with this final rule's requirements.
5. Submission of Test Plans for FRA Review
In part because FRA recognized that questions may arise in applying the proposed dynamic performance requirements in situations not clearly anticipated today, FRA requested comment on whether this final rule should include either an option or a requirement that the test methodology be submitted for FRA review prior to the conduct of destructive testing.
APTA commented that it believed such pre-approval to be unwise. APTA stated that delay awaiting FRA approval would impact schedules, extend the already extensive procurement process, and expose car builders to liquidated damages should FRA review be delayed. Instead, if FRA were to impose a requirement to submit a test plan, APTA recommended that FRA include a presumption that the plan is approved by some reasonable time after submittal to FRA, to avoid increasing the commercial risk to car builders. Caltrans' comments raised similar concern with the inclusion of a requirement that test plans be submitted to FRA for approval, asserting a great possibility of project delay while the railroad or its contract equipment supplier is awaiting FRA's response. In addition, CRM commented that, while its involvement with Volpe Center staff in the analysis and testing of its equipment has been very informative and helpful, it did not recommend mandating the submittal of test plans. CRM believed that doing so would require FRA to budget for a staff to support this effort in a timely manner so that delivery schedules remain unaffected. Nonetheless, CRM recommended that FRA publish guidelines for preparing analyses and conducting tests so that manufacturers know to follow an approach with which FRA agrees.
In response to these comments, FRA makes clear that it welcomes the submittal of test plans for its review. For instance, if a manufacturer were to conduct a test without using appropriate instrumentation or without applying a load at the appropriate location, a new test would likely be costly and would likely have been avoided had a test plan been submitted to FRA for review. Nevertheless, FRA agrees with the commenters and, in general, is not imposing new submittal requirements. As noted, however, FRA is requiring the submission and approval of plans to ensure compliance with the alternative corner post requirements for the non-engineer's side of the cab end of vehicles with stepwells for low-level platform boarding.
See
§ 238.213(c) and
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.