railroad accident report - wikimedia commons · 2018-01-16 · pb97-916302 ntsb/rar-97/02 national...
TRANSCRIPT
/
PB97-916302NTSB/RAR-97/02
NATIONALTRANSPORTATIONSAFETYBOARD
WASHINGTON, DC 20594
RAILROAD ACCIDENT REPORT
COLLISION AND DERAILMENTOF MARYLAND RAIL COMMUTER MARC TRAIN 286 ANDNATIONAL RAILROAD PASSENGER CORPORATIONAMTRAK TRAIN 29NEAR SILVER SPRING, MARYLANDON FEBRUARY 16,1996
6664C
\U.S. GOVERNMENT PRINTING OFFICE: 1989 0-942-865
Abstract: On February 16, 1996, Maryland Rail Commuter (MARC) train 286 collided with National RailroadPassenger Corporation (Amtrak) passenger train 29 near Silver Spring, Maryland. All 3 CSXT operatingcrewmembers and 8 of the 20 passengers on MARC train 286 were killed in the derailment and subsequent fire.Eleven passengers on MARC train 286 and 15 of the 182 crewmembers and passengers on Amtrak train 29 wereinjured.
The major safety issues discussed in this report are the performance and responsibility of the MARC train286 crewmembers, the oversight of CSXT signal system modifications, the Federal oversight of commuter railoperations, the lack of positive train separation control systems, and the adequacy of passenger car safetystandards and emergency preparedness. In addition, the Safety Board examined the use of the reverser duringan emergency brake application, the effectiveness of the computer-aided train dispatching recordkeeping, thecrashworthiness of locomotive fuel tanks, and the contents of the CSXT and MARC operating agreement.
As a result of its investigation, the Safety Board issued recommendations to the FRA, the FTA, the CSXT,the MTA/MARC, the U.S. Department of Transportation, the Federal Emergency Management Agency, theGovernor and the General Assembly of Maryland, the Association of American Railroads, the MontgomeryCounty Emergency Management Agency, the Baltimore County Emergency Management Agency, theBaltimore City Emergency Management Agency, the Metropolitan Washington Council of Governments, theJefferson County Commissioners, the Berkeley County Commissioners, the American Short Line RailroadAssociation, the Brotherhood of Locomotive Engineers, the United Transportation Union, the InternationalBrotherhood of Teamsters, and the American Public Transit Association. In addition, the Safety Boardreiterated safety recommendations to the FRA, the General Electric Company, and the Electro-Motive Divisionof General Motors.
The National Transportation Safety Board is an independent Federal agency dedicated to promoting aviation,railroad, highway, marine, pipeline, and hazardous materials safety. Established in 1967, the agency is mandatedby Congress through the Independent Safety Board Act of 1974 to investigate transportation accidents, determinethe probable causes of the accidents, issue safety recommendations, study transportation safety issues, andevaluate the safety effectiveness of government agencies involved in transportation. The Safety Board makespublic its actions and decisions through accident reports, safety studies, special investigation reports, safetyrecommendations, and statistical reviews.
Information about available publications may be obtained by contacting:
National Transportation Safety BoardPublic Inquiries Section, RE-51490 L’Enfant Plaza, SWWashington, DC 20594(202) 314-6551
Safety Board publications may be purchased, by individual copy or by subscription, from:
National Technical Information Service5285 Port Royal RoadSpringfield, Virginia 22161(703) 605-6000
COLLISION AND DERAILMENT
OF MARYLAND RAIL COMMUTERMARC TRAIN 286
AND NATIONAL RAILROAD PASSENGER CORPORATIONAMTRAK TRAIN 29
NEAR SILVER SPRING, MARYLAND
FEBRUARY 16, 1996
RAILROAD ACCIDENT REPORT
Adopted: July 3, 1997Notation 6664C
NATIONALTRANSPORTATION
SAFETY BOARD
Washington, DC 20594
iii
CONTENTS
EXECUTIVE SUMMARY .........................................................................................................vii
INVESTIGATIONAccident Narrative ........................................................................................................................................1Injuries….......................................................................................................................................................7Damages........................................................................................................................................................7
Track and Signals ...................................................................................................................................8MARC Train 286....................................................................................................................................8Amtrak Train 29 .....................................................................................................................................8
MARC Train 286 Crew Information.............................................................................................................8Engineer..................................................................................................................................................8Conductor ...............................................................................................................................................9Assistant Conductor..............................................................................................................................10
Amtrak Train 29 Crew Information ............................................................................................................11Engineer................................................................................................................................................11Assistant Engineer ................................................................................................................................11
Train Information ........................................................................................................................................12MARC Train 286..................................................................................................................................12Amtrak Train 29 ...................................................................................................................................14
Postcollision Train Information ..................................................................................................................14MARC Train 286..................................................................................................................................14Amtrak Train 29 ...................................................................................................................................17
Track and Signal Information .....................................................................................................................17Tracks ...................................................................................................................................................17Signals ..................................................................................................................................................20
Operations Information ...............................................................................................................................22CSXT Train Operations........................................................................................................................22Operational Efficiency Testing.............................................................................................................24MARC Train Radio Communication....................................................................................................24MARC Train 286 Passenger Load Data ...............................................................................................25CSXT and MARC Operating Agreement.............................................................................................25
Meteorological Information ........................................................................................................................26Medical and Pathological Information........................................................................................................26
MARC Train 286..................................................................................................................................26Amtrak Train 29 ...................................................................................................................................26
Toxicological Information ..........................................................................................................................28Emergency Response ..................................................................................................................................28
Montgomery County Fire and Rescue Services ...................................................................................28Montgomery County Disaster Plan ......................................................................................................29Amtrak, MARC, and CSXT .................................................................................................................29
Survival Aspects..........................................................................................................................................30MARC Train 286 Occupant Evacuation ..............................................................................................30Firefighter Evacuation Action ..............................................................................................................32
iv
Tests and Research......................................................................................................................................32Locomotive Event Recorders ...............................................................................................................32MARC Train 286 and Amtrak Train 29 Movement.............................................................................32Sight Distance.......................................................................................................................................33Signals ..................................................................................................................................................34Stopping Distance.................................................................................................................................34Passenger Train Air Brakes ..................................................................................................................35Cab Control Car 7752 Doors ................................................................................................................35MARC Passenger Coach Car Interior Materials ..................................................................................35Fuel Oil Presence..................................................................................................................................36
Postaccident Actions ...................................................................................................................................36National Transportation Safety Board..................................................................................................36Federal Railroad Administration ..........................................................................................................37Maryland Mass Transit Administration and Maryland Rail Commuter ..............................................38CSX Transportation Inc........................................................................................................................38National Railroad Passenger Corporation ............................................................................................39Cooperative Organization Activities ....................................................................................................39
Other Information........................................................................................................................................39Maryland Mass Transit Administration................................................................................................39MARC Train Accidents........................................................................................................................40Maryland Department of Labor and Industry.......................................................................................40Brunswick Line Signal Modifications..................................................................................................41Federal-Grant-Requested Signal Project ..............................................................................................41CSXT Signal System Modification Design..........................................................................................43CSXT Traincrew Roster Consolidation ...............................................................................................44Locomotive Fuel Tanks ........................................................................................................................45Cellular Service Communication..........................................................................................................45
ANALYSISGeneral Factors ...........................................................................................................................................46Accident Narrative Review.........................................................................................................................46MARC Train 286 Engineer Performance ...................................................................................................47MARC Train 286 Crewmember Responsibility .........................................................................................49Traincrew Voice Recording........................................................................................................................51CSXT Signal System Modification Oversight............................................................................................52Federal Commuter Rail Operations Oversight............................................................................................54Positive Train Separation Control Systems ................................................................................................55Passenger Car Safety Standards..................................................................................................................58
Emergency Lighting .............................................................................................................................61Emergency Windows............................................................................................................................62Exterior Emergency Door Release .......................................................................................................62Interior Materials ..................................................................................................................................63Cab Control Car Structural Crashworthiness .......................................................................................64
Emergency Preparedness.............................................................................................................................66Reverser Use during Emergency Braking...................................................................................................69Computer-Aided Recordkeeping.................................................................................................................70Locomotive Fuel Tank Crashworthiness ....................................................................................................70CSXT and MARC Operating Agreement ...................................................................................................72
v
FINDINGS ................................................................................................................................................73Conclusions.................................................................................................................................................73Probable Cause............................................................................................................................................75
RECOMMENDATIONS .............................................................................................................75
APPENDIXESAppendix A--Investigation and Hearing.....................................................................................................81Appendix B--Chronology of Signal Sequence............................................................................................83Appendix C--Chronology of Accident ........................................................................................................87Appendix D--Excerpts from CSXT Operating Rules .................................................................................89Appendix E--Organization Charts of CSXT and MDOT/MTA/MARC.....................................................97Appendix F--Signal Incidents, Complaints, and False Proceed Indications .............................................103Appendix G--Fire Testing Results of MARC Car Interior Materials .......................................................105Appendix H--Issuance of Urgent Safety Recommendations.....................................................................109Appendix I-- Federal Railroad Administration Emergency Order No. 20................................................115Appendix J--MARC Train Service Schedule for Brunswick and Camden Lines.....................................135Appendix K--FRA Correspondence on Regulatory Authority of Commuter Railroad Operations..........137Appendix L--National Transportation Safety Board and Positive Train Control Systems.......................143Appendix M--Acronyms and Abbreviations.............................................................................................145
vii
EXECUTIVE SUMMARY
About 5:39 p.m. on February 16, 1996,Maryland Rail Commuter (MARC) train 286collided with National Railroad PassengerCorporation (Amtrak) passenger train 29 nearSilver Spring, Maryland. En route fromBrunswick, Maryland, to Union Station inWashington, DC, MARC train 286 wastraveling under CSX Transportation Inc.(CSXT) operation and control on CSXT tracks.MARC train 286 passed an APPROACH signalbefore making a station stop at Kensington,Maryland; proceeded as if the signal had beenCLEAR; and, then, could not stop for the STOPsignal at Georgetown Junction, where itcollided with Amtrak train 29. All 3 CSXToperating crewmembers and 8 of the 20passengers on MARC train 286 were killed inthe derailment and subsequent fire. Elevenpassengers on MARC train 286 and 15 of the182 crewmembers and passengers on Amtraktrain 29 were injured. Estimated damagesexceeded $7.5 million.
The National Transportation Safety Boarddetermines that the probable cause of this ac-cident was the apparent failure of the engineerand the traincrew because of multiple distrac-tions to operate MARC train 286 according tosignal indications and the failure of the FederalRailroad Administration (FRA), the FederalTransit Administration (FTA), the MarylandMass Transit Administration (MTA), and theCSXT to ensure that a comprehensive humanfactors analysis for the Brunswick Line signalmodifications was conducted to identify po-tential sources of human error and to provide aredundant safety system that could compensatefor human error.
Contributing to the accident was the lackof comprehensive safety oversight on theCSXT/MARC system to ensure the safety ofthe commuting public. Contributing to the se-verity of the accident and the loss of life wasthe lack of appropriate regulations to ensureadequate emergency egress features on therailroad passenger cars.
The major safety issues discussed in this re-port are the performance and responsibility ofthe MARC train 286 crewmembers, the over-sight of CSXT signal system modifications,the Federal oversight of commuter rail opera-tions, the lack of positive train separation con-trol systems, and the adequacy of passengercar safety standards and emergency prepared-ness. In addition, the Safety Board examinedthe use of the reverser during an emergencybrake application, the effectiveness of thecomputer-aided train dispatching recordkeep-ing, the crashworthiness of locomotive fueltanks, and the contents of the CSXT andMARC operating agreement.
On March 12
Emergency Management Agency, theMetropolitan Washington Council ofGovernments, the Jefferson CountyCommissioners, the Berkeley CountyCommissioners, the American Short LineRailroad Association, the Brotherhood ofLocomotive Engineers, the UnitedTransportation Union, the InternationalBrotherhood of Teamsters, and the AmericanPublic Transit Association. In addition, theSafety Board reiterates safetyrecommendations to the FRA, the GeneralElectric Company, and the Electro-MotiveDivision of General Motors.
INVESTIGATION
Accident Narrative
On Friday, February 16, 1996, at 5:39 p.m.,an eastbound Maryland Rail Commuter(MARC) train 286, operated by the CSX Trans-portation Inc. (CSXT) for the Maryland MassTransit Administration (MTA) collided with thewestbound National Railroad Passenger Corpo-ration (Amtrak) passenger train 29, CapitolLimited. (See figure 1.) The accident occurredduring a blowing snowfall near milepost (MP)BA 8.49� at a railroad location, referred to asGeorgetown Junction, about 1 mile west of Sil-ver Spring, Maryland. (See figure 2.) The snowaccumulation on the ground at the accident sitewas 5 inches. (See figure 3.) Both trains wereoperating on the double main tracks owned andmaintained by the CSXT.
The MARC train 286 was a “push-pull”�
commuter train consisting of a locomotive uniton the rear end, two passenger cars, and a pas-senger coach cab control car in the lead. Theengineer was operating the train from the cabcontrol car in the push mode at the time of thecollision. The MARC train 286 was in sched-uled commuter service proceeding eastward ontrack 2 between Brunswick, Maryland, and Un-ion Station, Washington, DC. The train departedBrunswick at 4:30 p.m. eastbound for a scheduled5:30 p.m. arrival at Union Station with threeCSXT operating crewmembers� (an engineer, a
1The prefix BA stands for the Baltimore Division,Metropolitan Subdivision. All MP designations inthis report are for that subdivision; henceforth the BAwill not be shown.
2Train arrangement in which the motive power isat one end and a control cab car is at the other end,permitting the engineer to operate from either end ofthe train. The power is at the rear of the train whenthe train is configured for the push mode and at thefront of the train when it is configured for the pullmode.
3Traincrews are employed and contracted by theCSXT to operate MARC commuter trains between
conductor, and an assistant conductor) on board.MARC train 286 had to make a “flag” stationstop� at the Kensington, Maryland, station toboard two waiting passengers. The train carried20 passengers at the time of the collision. Be-fore MARC train 286 stopped at Kensington,the engineer on the westbound MARC train 279stated that as the two trains passed each other onadjacent tracks, he had heard a portion of theradio communication of the MARC train 286engineer acknowledging the wayside signal1124-2,� located about 1,000 feet west of Kens-ington station.
The westbound Amtrak train 29, with a 2-unit locomotive and 15 cars, departed UnionStation about 5:25 p.m. en route to Chicago, Illi-nois. The Amtrak operating crew consisted of anengineer, an assistant engineer, a conductor, andan assistant conductor. Thirteen on-board serv-ice employees, a mechanical rider who was inthe second unit, and 164 passengers were alsoon board the Amtrak train. The Amtrak train 29had been routed onto track 2 from Union Stationto Georgetown Junction to pass a stopped west-bound CSXT freight train that occupied track 1east of Georgetown Junction. The engineer ofAmtrak train 29 stated he was operating on a
Washington and Brunswick.4The station is not a regularly scheduled stop. A
train makes a flag stop to pick up passengers who arestanding on the platform and are visible to the engi-neer. A train also makes a flag stop to discharge pas-sengers who have requested the stop and notified theconductor when boarding.
5The signal would have been displaying an AP-PROACH indication, logically related to the STOPindication being displayed at Georgetown Junctionbecause the crossover at the interlocking was alignedfor a train movement from track 2 to track 1 for thewestbound Amtrak train 29.
2
Figure 1--View of trains at point of impact.
MEDIUM CLEAR signal (see table 1) and wasbeginning to negotiate the crossover from track2 to 1 at Georgetown Junction when thecollision occurred. (See appendix B for thesignal sequence chronology.)
The locomotive event recorder of MARC train286 indicated that the train accelerated upon de-parting the Kensington station, slowed for aposted speed restriction, and then acceleratedagain. The engineer of MARC train 286 placedthe train brakes in emergency about 2.18 milesfrom the Kensington station at a speed of ap-proximately 66 mph. About 11 seconds after theemergency brake application and 7 seconds be-fore impact, the train control lever (reverser) wasrecorded as being moved from reverse to forward.(The locomotive had been operating in the re-verse position as a pusher locomotive before
this.) The impact speed with Amtrak train 29 wasabout 38 mph.
The collision between the lead Amtrak unitand the MARC cab control car tore away thefront left quadrant of the cab control car. The fueltank on the lead unit was ruptured in the collisionand sprayed fuel on the cab control car. All threeMARC cars and the MARC locomotive derailedin the accident. Both Amtrak units and the firsteight cars of train 29 derailed. The derailed Am-trak equipment consisted of six material handlingcars, one baggage car, and a transition-sleepercar, which was the only occupied Amtrak car toderail. (See figure 4.)
3
Table 1.--Key to CSXT signal indications
CLEAR MEDIUM CLEAR APPROACH STOP
Rule C-281 Rule C-283 Rule C-285 Rule C-292
Proceed. Medium speed(30 mph) through turn-outs, crossovers, sid-ings, and over power
operated switches; thenproceed.
Proceed prepared to stopat the next signal. Trainsexceeding medium speedmust immediately begin
reduction to mediumspeed as the engine
passes the APPROACHsignal.
Stop.
Ten survivors from cab control car 7752 es-caped after the accident through an openingbetween the diaphragms of the first and secondcar interior passageway of train 286. Two othersurvivors from the MARC train exited throughan emergency window in the second car.
The CSXT AU� dispatcher was contactedabout 5:41 p.m. by the traincrew of CSXT trainK951, which was stopped on the adjacent trackeast of Georgetown Junction, that Amtrak train29 had derailed while crossing over at George-town Junction. Also about 5:41 p.m., the Mont-gomery County [Maryland] Fire and RescueServices (MCFRS) 911 dispatcher received ap-proximately 12 telephone calls reporting the de-railment and fire. At 5:44 p.m. the AU dispatcher
6The CSXT designates the train dispatcher thatcontrols train movements on the Brunswick Line asthe AU train dispatcher.
contacted the MCFRS to notify it of the accidentand was informed that it was already aware of thederailment. About 5:46 p.m. the first units arrivedon scene.
During the emergency response activities,CSXT freight train Q401 and MARC train 281had been operating, respectively, behind MARCtrain 286 and Amtrak train 29. These two trainshad been stopped from approaching the accidentscene by railroad wayside signal indication. Uponrequest to the CSXT AU dispatcher, the engineerof CSXT train Q401 received authorization fromthe dispatcher to separate the locomotive unitsfrom his train and to approach the accident siteprepared to stop because of the emergency re-sponse personnel in the area. The locomotivefrom CSXT Q401 stopped about 775 feet west ofthe derailed locomotive of MARC train 286 about6:20 p.m. The MARC train 281 engineer had alsomade a similar request to pull up behind
7
Amtrak train 29 and lend assistance. Approvalwas not received until about 6:56 p.m. but notfrom the dispatcher; instructions were receivedfrom the CSXT/MARC manager of passengeroperations to move as close as possible to Amtraktrain 29. Neither train movement was coordinatedwith the MCFRS incident commander (IC). TheIC stated that when a citizen reported that a trainwas approaching the accident site, he had his staffactivate radio tones and air horns to warn emer-gency responders to evacuate the accident site.(See appendix C for the accident chronology.)
The first firefighters to reach cab control car7752 reported that the car was fully involved infire and that they did not observe any survivors.They made several attempts to enter the cab con-trol car. The fire was extinguished within 10minutes, after which the firefighters were able to
enter cab control car 7752. They were later as-sisted by members of the Montgomery CountyPolice Department (MCPD) in the recovery of 11victims for coordinating the identification andnotification process with the Maryland medicalexaminer. At 3:50 a.m. on Saturday, February 17,1996, the last victim was recovered, and the ICturned over the accident to the MCPD. At 4 a.m.the recovery operations were secured.
In the collision and subsequent fire, all 3CSXT crewmembers and 8 passengers onMARC train 286 were fatally injured; 11 of the12 MARC train survivors sustained from seriousto minor injuries. The Amtrak engineer, assis-tant engineer, and mechanical rider sustainedserious injuries; the conductor, the assistantconductor, two on-board service crewmembers,and eight passengers sustained minor injuries.
Injuries*
Table 2.-- Injuries sustained in Silver Spring railroad accident
Type MARCTrain 286
Operating Crew
MARCTrain 286
Passengers
AmtrakTrain 29
Operating Crew
AmtrakTrain 29
Employees
AmtrakTrain 29
Passengers
Total
FATAL 3 8 0 0 0 11
SERIOUS 0 3 2 1 0 6
MINOR 0 8 2 2 8 20
NONE 0 1 0 11 156 168
TOTAL 3 20 4 14 164 205
*Based on the injury criteria (49 Code of Federal Regulations [CFR] 830.2) of the International Civil Aviation Organiza-tion, which the Safety Board uses in accident reports for all transportation modes.
Damages
The CSXT, the MTA/MARC, and Amtrak esti-mated the damages, and the MTA/MARC andNational Transportation Safety Board estimatedreplacement costs, as follows:
Cars $4,963,624 (includes replacement cost for MARC equipment)Locomotives 2,350,000 (includes replacement cost for Amtrak lead unit)Track/signals 200,000Total $7,513,624
8
Track and Signals-- Three turnouts,� includingtheir signal operating components, were de-stroyed. The turnout leading to the Georgetownindustrial track and the westernmost crossoverfrom track 1 to 2 were replaced and reposi-tioned. A total of about 663 feet of track wasreplaced in main tracks 1 and 2. The track workwas completed on April 12, 1996. Because ofthe turnout and track repositioning for theGeorgetown industrial track, the eastbound ab-solute signal (EAS) for track 2 was replaced andrelocated adjacent to track 2.
MARC Train 286-- The locomotive of train 286had minimal damage. The three MARC passen-ger cars, including cab control car 7752, sus-tained heavy damage and were considered totallydestroyed by MARC.
Amtrak Train 29-- The collision and derailmentaffected only the two locomotive units and thefirst eight cars of the Amtrak train, with the re-maining cars in the consist sustaining no reportedstructural damage. The left rear side of the leadunit car body exterior, truck assembly, and fueltank received substantial fire damage. The lead-ing truck assembly of this unit was detached fromthe unit. Amtrak considered the unit totally de-stroyed. The second unit and the derailed passen-ger cars sustained minimal damage; they were tobe repaired and returned to service.
MARC Train 286 Crew Information
Engineer-- The 43-year-old engineer beganworking for the Baltimore & Ohio (B&O) Rail-road on September 3, 1970, in the signal de-partment but became a brakeman on February16, 1971. He was promoted to engineer on April15, 1974, and qualified over the territory thatincluded Baltimore, Maryland, to Washingtonand Washington to Brunswick shortly thereafter.He operated freight trains and some passenger
7An arrangement of switch point rails, crossoverrails (frog), and closure rails that permit trains to bediverted from one track to another.
trains (from the extra board) before working as ayard engineer in Baltimore. In 1980, he becamea CSXT employee through a merger.
In August 1994, he requalified over the ter-ritory in passenger service after leaving the yardengineer position and worked several differentjob assignments that included Washington toBrunswick trips. He became the regular engi-neer on the assignment that included MARCtrain 286 on January 1, 1996. The CSXT man-ager of passenger operations, who was his su-pervisor, testified that he was “one of the mostprofessional engineers that we had out here. Hedid a real good job in everything that he did. Hewas consistent, he was concise.”
His most recent CSXT operating rules classwas completed on February 25, 1995. His mostrecent engineer certification was December 31,1995; the class designation was Train ServiceEngineer with no restrictions noted. His servicerecord showed five instances of discipline; how-ever, none were for operational failure. SeventyCSXT operational efficiency tests had been per-formed on the engineer during the previous 12months; he had no reports of failure.
The most recent railroad physical for the en-gineer of MARC train 286 was conducted onOctober 21, 1995, and it found him medicallyqualified. His distant vision was 20/20 uncor-rected; his near vision was 20/35, although hehad no difficulty reading and did not wearglasses nor was he required to do so to operatetrains. His color perception was normal, and hehad no hearing loss. The engineer’s father notedhis son showed no signs of illness when he lastsaw him on February 14, and did not complainof any illness when they last spoke on February15. In addition, the engineer expressed no con-cerns or health complaints when he spoke to afemale acquaintance from Brunswick about 4p.m. on February 16.
For each weekday since his most recent as-signment began on January 1, 1996, the engi-neer kept essentially the same work schedule.According to CSXT records he also kept this
9
same schedule during the week before the acci-dent. (See table 3.) His work schedule was 11hours 35 minutes on duty and 12 hours 25 min-utes off duty. He worked on four train runs eachday with a break between each run. Thesebreaks were scheduled to be 45 minutes, then 1hour 15 minutes, and finally 2 hours, althoughthese could vary depending on train delays.
On Friday, February 16, the day of the acci-dent, the engineer reported for duty at 10:10a.m. and departed from Camden Station withMARC train 251 as scheduled at 11 a.m. Thetrain arrived 45 minutes late in Washington at12:55 p.m. because of weather-related switchproblems at Savage, Maryland. The next train heoperated, MARC train 273 to Brunswick, de-parted 24 minutes late at 1:24 p.m. Snow de-layed that run and he arrived 33 minutes late in
Brunswick at 2:53 p.m. He departed Brunswickwith MARC train 286 at 4:30 p.m.
Conductor-- The 48-year-old conductor washired by the B&O on October 27, 1969. He be-came a CSXT employee through a merger in1980. He was regularly assigned to a yard job inBaltimore. According to his wife, he has alwaysworked in the Baltimore rail yards (Curtis Bayor Locust Point) and not in regular freight serv-ice. His seniority was sufficiently high thatwhen he worked in the yards, he was always offon weekends. During the last 2 years, he hadalso worked in passenger service on MARCtrains when called. His passenger service as-signments also provided weekends off, exceptfor an occasional weekend train to the baseballstadium in Baltimore.
Table 3.--MARC train 286 engineer’s daily work schedule
Time Activity
10:10 a.m. Report for duty at CSXT Riverside yard in Baltimore.Drive to Camden Station to operate MARC train 251.
11 a.m. Depart with train 251 en route to Union Station.12:10 p.m. Arrive at Union Station.
Stay in station until his next run. 1 p.m. Depart station with MARC train 273 en route to Brunswick. 2:20 p.m. Arrive at Brunswick terminal.
Remain at Brunswick until time to operate return train 286 to Union Station.Telephone and talk to friend usually each day during this time.
4:30 p.m. Depart Brunswick with train 286 en route to Union Station. 5:50 p.m. Arrive at Union Station.
Remain at terminal until time to operate (fourth) train 260. 8 p.m. Depart with train 260 from Union Station to Baltimore. 9:13 p.m. Arrive at Baltimore. 9:30 p.m. Off duty. 9:45 p.m. Arrive home after 15-minute drive.
10
He was promoted to freight conductor onJanuary 1, 1992, and to passenger conductor onNovember 19, 1993. He successfully completedhis most recent operating rules class on February13, 1995. His service record showed three in-stances of discipline, two of which were for oper-ating rules infractions. The first operating rulesinfraction was on May 11, 1981, for violation ofrules 104-C and 105 by fouling a track beforeswitches connected with the train movement werelined and not being properly prepared to stopwithin half the range of vision. A 5-day overheadsuspension� was given. The second recorded in-stance was on July 27, 1981, when the employeewas cited for a violation of rule 103-G, whichresulted in a derailment and in a formal repri-mand. (See Appendix D, Excerpts from CSXTOperating Rules.) Twelve CSXT operational effi-ciency tests had been performed on the conductorduring the previous 12 months; no failures werereported.
The most recent railroad physical for the con-ductor of MARC train 286 was on June 17, 1993,which showed that he was medically qualifiedfor his job; his visual acuity, degree of visualfield, color sense, and hearing were all withinacceptable ranges. His wife stated that he al-ways slept well at night and had done so thenight before the accident. She reported she didnot know whether her husband had taken anymedication on Friday, February 16, but he hadnot reported any illness to her. He did not haveany medical conditions or take any medicationon a regular basis.
During the 4 days before the accident, theconductor worked a daylight “yard job” at CurtisBay. He awoke about 6 a.m. daily; worked at therail yard until about 1:30 p.m. on Wednesday,February 14, and until 3 p.m. on Thursday, Feb-ruary 15; and retired about 10 p.m. on both days.Between 9 and 10 p.m. on Thursday, he was noti-fied by the CSXT that he would be in passenger
8Punishment that is held in abeyance provided theemployee has no other infractions.
service the next day. On Friday, February 16, heawoke at 8 a.m. and left for work about 9 a.m.
Assistant Conductor-- The 53-year-old assistantconductor went to work for the B&O on Sep-tember 1, 1965. He remained with that sameorganization, which ultimately became a part ofthe CSXT. He worked first as a brakeman andlater as a conductor in both yard and roadfreight service. His last assignment before en-tering passenger service was as a brakeman atthe Jessup Yard.
The assistant conductor completed his mostrecent operating rules training class on June 18,1995. His service record showed two instances ofdisciplinary action. He received a 10-day over-head suspension on September 27, 1969, for vio-lation of rule 804 (approaching a hazardous mate-rials derailment). The second incident occurredon May 30, 1986, and entailed violation of rules106 (safe train operation) and 450 (radio usage),which resulted in a 10-day overhead suspension.Both occurred while the assistant conductor wasin yard service. Twenty-nine CSXT operationalefficiency tests had been performed on the assis-tant conductor during the previous 12 months; nofailures were reported.
The most recent railroad physical for the as-sistant conductor of MARC train 286 was onSeptember 12, 1994, and showed him medicallyqualified for his job. His distant vision was20/70 in each eye but improved to 20/20 withglasses, and his near visual acuity, degree ofvisual field, color sense, and hearing were allwithin acceptable ranges. His wife reported hewas wearing his glasses when he went to workon February 16. She said that her husband hadslept well at night during the week before theaccident and that he took a daily capsule of themedication Prilosec for an ulcer.
His wife characterized the relationship of herhusband and the engineer of MARC train 286 asfriendship. They had worked together previouslyat the Curtis Bay rail yard. She said that onmany previous occasions her husband haddescribed him as one of the best engineers, that
11
her husband liked all of his coworkers, and thatthey worked well together. She reported he hadalso worked with the conductor previously andconsidered him a good conductor, although hehad not really talked much about the conductor.
For the last 4 months he had worked the sameweekday assignment on MARC passenger trains.On Wednesday, February 14, and Thursday, Feb-ruary 15, the assistant conductor awoke about7:30 a.m. and arrived at work about 10 a.m.During the day, his assignment allowed him freetime between trains while in Brunswick, and hegenerally called his wife between 2:30 and 3:30p.m. He returned home each night about 10:15p.m. and retired at 11:30 p.m. On Friday, Febru-ary 16, he awoke and arrived at work at hisusual times, and according to his wife, he calledher at 3:20 p.m. She described the conversationas typical and said that he had expressed noparticular concerns when they talked.
Amtrak Train 29 Crew Information
The operating crewmembers were employedby Amtrak.
Engineer-- The 49-year-old engineer began hisrailroad employment with the B&O on June 14,1977. He was originally hired as a brakemanand worked in Pittsburgh, Pennsylvania. Hewent into engine service on June 8, 1978, whenhe became a fireman and was promoted to engi-neer on May 20, 1979. He operated Amtrak pas-senger trains at times as a B&O employee. OnAugust 20, 1986, he transferred to Amtrak andoperated its passenger trains full time.
As a B&O engineer, he was qualified overthe territory from Cumberland, Maryland, toPittsburgh; as an Amtrak engineer, he becamequalified over the entire territory from Wash-ington to Pittsburgh in 1986. From 1986 until1993, he operated over various routes, often asthe relief engineer from Washington to Pitts-burgh, and that route became his regular as-signment in 1993.
The engineer successfully completed his lastlocomotive engineer’s certification examinationon February 1, 1996, with no restrictions. He waslast examined on the CSXT operating rules onOctober 3, 1995. His service record shows oneinstance of discipline for violating CSXT oper-ating rule 292 in Halifax, South Carolina, onAugust 27, 1992. The incident occurred when hisassistant engineer, who was operating the train,violated a STOP signal. He received a 30-daysuspension from Amtrak and was barred fromoperating over the CSXT until March 1993. Inthe 3 years before the accident, the engineer wassubject to 20 efficiency tests and was found tobe in compliance in each instance.
By happenstance, the engineer’s supervisorconducted an efficiency test of the engineer at1:04 p.m. on the day before the accident. Thesupervisor observed the engineer cross overfrom track 1 to 2 at Georgetown Junction (east-bound) on a MEDIUM CLEAR signal. Such acrossover has a speed limit of 30 mph; and thesupervisor determined with a radar gun that theengineer crossed over at 27.9 mph.
The engineer of Amtrak train 29 reported hewas in good health and was not ill on the day ofthe accident. He stated he takes three medica-tions each day: Procardia (30 mg) and one aspi-rin, since a heart attack in 1990, and Mevacor, acholesterol reduction medication. He passed hislast railroad physical on November 1, 1995. Hisvision and hearing capabilities were found to bewithin acceptable range without any restrictions.He wore reading glasses but was not required towear them while operating the train, and he wasnot wearing them at the time of the accident. Hewas wearing railroad-supplied clear safetyglasses.
Assistant Engineer-- The 38-year-old assistantengineer of Amtrak train 29 was hired by theB&O as part of a track gang in June 1977. InAugust 1977 he began working as a conductorand remained as a conductor with the B&O untilAugust 1986. He transferred to Amtrak inAugust 1986 and continued to work as a con-ductor until 1989 when he went into engine
12
service as a fireman. He was promoted to engi-neer on May 27, 1992, and his most recent certi-fication was February 9, 1996, which had a vi-sion restriction requiring corrective lenses. Hewas qualified on the CSXT railroad operatingrules, had been qualified over the territory for 4years, and had worked with the engineer previ-ously. His last CSXT operating rules exam wasOctober 3, 1995, and he had no recorded in-stances of discipline.
The assistant engineer of Amtrak train 29had his most recent physical examination onApril 25, 1995, and was found medically quali-fied for his job. He reported that he was in goodhealth, had no long term illnesses, was not sickon the day of the accident, took no medication,and was wearing his prescription glasses at thetime of the accident.
Train Information
MARC Train 286-- The train consisted of passen-ger coach cab control car 7752, passengercoaches 7709 and 7720, and diesel electric loco-motive (GP39-H2) unit 73. These cars were des-ignated by MARC as MARC II type cars. (Seefigure 5a.) Neoprene diaphragms were attachedto both ends of the cars, which provided aweather-resistant barrier between two coupledcars. All cars were equipped with a safety bar onboth ends of the cars. The safety bar latched inboth the horizontal and the stored vertical posi-tion.
MARC train 286 was configured to operate inthe push mode with locomotive unit 73 at the rearof the train. Passenger coach cab control car 7752was very similar in arrangement to the otherpassenger coaches in train 286 except that theleading "F" end had an operating cab with con-trols in the front right corner of the car. (Seefigure 5b.) This enabled the engineer to operate
9Manufactured by Nippon Sharyo Seizo KaishaLtd., of Toyokawa, Japan, and delivered under State ofMaryland contract SRA 2108-003 and subsequentchange orders.
from the front of the train in the push mode asthe locomotive unit pushed from the rear of thetrain. This was the configuration at the time ofthe accident.
Required Federal Railroad Administration(FRA) scheduled maintenance of MARC pas-senger coaches and cab control cars is per-formed under contract by Amtrak personnel attheir Washington Ivy City maintenance facility.Required FRA scheduled maintenance of theMARC locomotives is performed under contractby the CSXT mechanical personnel at the River-side yard near Baltimore. CSXT and Amtrakpersonnel may also perform daily inspection,servicing, and running repair of MARC equip-ment depending on where the equipment isstored or turned around.� The CSXT also per-forms inspection, servicing, and running repairat its Brunswick facility.
The night of February 15, 1996, at the CSXTRiverside yard, locomotive unit 73 received adaily inspection. It received a locomotive airbrake test as well as a train air brake test withthe three passenger coaches as MARC train 243the next morning. The unit and the cars werethen designated as MARC train 271, and Am-trak performed another train air brake test inWashington before the train departed forBrunswick. The inspection records showed noanomalies. That train departed Brunswick des-ignated as MARC train 286 and was the consistin the collision.
The CSXT inspection records for cab controlcar 7752 showed that the speed indicator wasaccurate at 70 mph and that no exceptions weretaken on the air brake inspection tests. The twopredeparture inspection sheets on the day of theaccident both indicated that the radio of cabcontrol car 7752 was “good,” and no previousradio malfunction had been reported. The radiowas destroyed in the collision.
10The three MARC cars involved in this accidenthad been continuously together since October 3,1995.
14
Amtrak Train 29-- The train consisted of a two-unit (F40PH-255 and P40-811) locomotive fol-lowed by six material handling cars, a baggagecar, a transition-sleeper (from one level tobilevel), two sleeping cars, a dining car, a loungecar, two coaches, and a coach-dormitory car. Theradios and antennas from both Amtrak units 255and 811 were tested at the accident site and thenbench tested. All radios and antennas werefound in good working order. The throttle on thelead unit was found in the eighth notch position.
Postcollision Train Information
MARC Train 286-- The cars and locomotive ofthe train remained coupled and were situated ap-proximately in a linear orientation, derailed butupright, resting parallel to track 2 on the ballastand the displaced track of the siding. The carswere displaced laterally to the south between 6and 20 feet from the track 2 centerline. Both thefront and rear sanders of locomotive unit 73were 3/4 full. The sand box on the front left sideof cab control car 7752 had been sheared away,and the sand box on the right (engineer's) sidehad been crushed but still contained sand. Be-cause the running gear and batteries of thecoaches were damaged in the derailment, theemergency lighting and public address systemwere inoperable.
After colliding with Amtrak train 29, cabcontrol car 7752 derailed and came to rest tiltedapproximately 10 degrees clockwise. (See figure6.) The leading truck assembly of the car hadseparated from its mounting and was found up-side down and lodged under the car midsection tothe right of centerline. The car body plymetalfloor above where the truck assembly came torest was indented about 2 feet in depth. The left-side front anchor bracket assembly of the truckhad fractured and was torn from its attachment tothe car.
The interior of cab control car 7752 (see fig-ure 7) was gutted by fire, leaving the charredmetal shell, internal fittings, and combustion resi-due. Damaged remnants of insulation remained
in the ceiling and side walls. The rubber dia-phragm at the rear passageway door was burnednear the roof. The rear safety bar was horizon-tally positioned. A 50-pound portable fire extin-guisher, normally secured by a metal strap onthe front bulkhead of a vestibule, was missingand later discovered in the debris. The glasscover of the emergency toolbox was shattered,and the tools were missing.
The rear interior door of the car was openand slid within the wall pocket between the ves-tibule and the lavatory. The fire destroyed thecontrols for the left- and right-side rear doorsand burned the position indicator lenses. Theright-side rear door was open 0.5 inch, and itswindow was missing. The door contacted thefront of the car body shell, into which it slides,12 inches above the vestibule floor. Its interioremergency release handle was in a securedcabinet in the lavatory. (During an inspection ofthe car on March 14, 1996, at the Middle River,Maryland, MARC facility, the door did not openwhen this handle was pulled fully down.) Theright-side rear "T" handle (exterior emergencydoor release mechanism) was missing from itscable, which, when pulled downward by pliers,moved the inside emergency release handledownward; however, the door did not open. Theleft-side rear door was open 1 inch, and its win-dow was damaged by the fire and outwardlydisplaced at the top. The door contacted thefront of the car body shell, into which it slides, 8inches above the vestibule floor. Its interioremergency release handle was damaged by thefire and could not be pulled down or operate thedoor. The cover to its cabinet was missing. Theleft-side rear T-handle was also missing from itscable, which, when pulled downward by pliers,did not move, and the door did not open.
The second car in the consist, passengercoach 7720, derailed and tilted approximately30 degrees clockwise with its front end lodgedagainst the rear end of cab control car 7752. Theleft-side front corner of the coach, which includedthe bulkhead, the vestibule door, and a
17
4.5-foot-long section of sidewall, was pressedinward resulting in a breach of the car body endand corner structure. Fire damage was limited tothe front diaphragm, which contacted the adja-cent cab control car. The left-side front door wasdestroyed, and the T-handle was imbedded inthe damage. The right-side front door wasclosed and could not be opened using the T-handle. The front interior door of the coach wasdisplaced rearward and closed; the rear interiordoor remained intact. The left- and right-siderear doors, respectively, were operable andclosed. Its emergency toolbox was empty, itsfirst aid kit was intact, and its fire extinguisherwas found in the debris.
The last car in the consist, passenger coach7709, derailed and tilted approximately 30 de-grees clockwise. Its corner was depressed nearthe frame of the left-side rear door, which wasclosed. The front end of the passenger coach7709 was lodged against the rear end of the pas-senger coach 7720. The truck assembly on therear end, which was lodged against locomotiveunit 73 and Amtrak locomotive unit 811, wasdamaged. Passenger coach 7709 had no firedamage. No emergency window exit decalswere found on or near the windows. On Febru-ary 29, 1996, with MARC officials present, aSafety Board investigator took several minutes,applying physical exertion, to remove the left-side front emergency exit window of passengercoach 7709. Then on March 14, a Safety Boardinvestigator attempted unsuccessfully to removethe right-side rear emergency exit window,which was later removed by another investigatorafter about 3 minutes of physical exertion. (Alubricant used to install these particular emer-gency windows was later found to have hard-ened over time.)
Amtrak Train 29-- The locomotive units, thefirst, and the fourth through eighth cars were allderailed (either one or both axles) but remainedupright. With the exception of the leading unit(ATK 255) and the fifth and sixth (material han-dling) cars, the equipment was found to be cou-pled and approximately in a linear orientation.The equipment was displaced laterally to both
sides of track 2 and the crossover or resting ontrack 1 and its displaced ballast.
Lead unit ATK 255 derailed and receiveddamage to the hood, fuel tank, and running gear.(See figure 8.) The left side of the unit super-structure received fire damage behind the con-trol compartment, which remained intact. Thelavatory was destroyed. The right-side wind-shield was shattered, and the left-side wind-shield was not present. The side windows wereshattered. Unit ATK 811 derailed and receivedsubstantial damage to its superstructure andrunning gear.
Of the six material handling cars, the firstthrough sixth in the consist positioning, only thesecond car remained on the rail. The baggagecar, the seventh car in the consist, and the fronttrucks of the eighth car, a transition/dormitorycar, derailed. The passenger-occupied 9th
through 15th cars remained on the track and werenot damaged. All Amtrak cars in the consist re-mained upright and parallel to the track, exceptfor the fifth and sixth cars. No damage wasnoted to passenger compartments.
All emergency lighting had been illuminatedimmediately after the collision according to theconductor. The emergency lights were illumi-nated in all cars, except in the 15th car, whenSafety Board investigators arrived on sceneabout 1 hour after the collision. The public ad-dress system remained operable after the colli-sion and was used by the conductor for emer-gency broadcasts. All emergency tools andequipment remained in place.
Track and Signal Information
Tracks-- The collision occurred on the CSXTBaltimore Division, Metropolitan Subdivision,which was double main track territory. Thetracks were designated as track 1 and 2, and theywere spaced on 12.5-foot centers at the collisionpoint. The point of collision was in a turnout ontrack 2 about MP 8.49, where the track crossedover to track 1, about 180 feet east of the EASfor track 2. (See figure 9.)
20
The MPs were in decreasing numerical orderin the timetable eastward direction. The trackspass through a residential area. An unpaved rail-road maintenance road is adjacent to track 1. Theright-of-way closest to the point of collision iselevated above the adjacent topography. A steepslope about 45 feet wide is on the south side, anda far lesser slope about 80 feet wide is on thenorth side. Both slopes had a dense cover of treesand shrubbery. Open parking lots for large resi-dential buildings are near the bottom of the slopeson both sides of the right-of-way adjacent to thecollision site. Highway overpass bridges areabout 1,000 and 800 feet, respectively, to the westand the east of the collision site.
Approaching the accident site at MP 8.49from the west, track 2 between MPs 8.95 and8.68 curves to the left through a 1° and 36' curveon a 0.43-percent ascending grade eastward. Thetrack alignment and the gradient between MPs8.68 and 8.27 are, respectively, straight and on a0.18-percent descending grade in the eastwarddirection.
The track and turnouts were 140-pound con-tinuous welded rail. Both tracks 1 and 2 and theturnouts making up the interlocking were in-spected and maintained by the CSXT to FRAclass 4 track standards, which provide specificminimum track geometry requirements forfreight and passenger train operations at maxi-mum allowable speeds of 60 and 80 mph, re-spectively. The before and after accident trackgeometry measurements for gage, alignment,and cross level and all other track inspection andmaintenance records were reviewed and indi-cated no anomalies. Postaccident inspectionfound the westernmost crossover switch rails ontrack 2 and the switch rails on track 1 were linedproperly for a crossover movement by Amtraktrain 29.
About 8 to 10 “snow pots” (oil burningswitch heaters) were in place and lit at all fourswitches of the Georgetown interlocking cross-overs. The snow pots, which were in the spacesbetween the switch and under the switch pointrails, were burning to keep snow from accumu-lating behind the switch points.
An adjacent track, the former Georgetownbranch track, is parallel to and intersects withtrack 2 east of the collision site. The siding wasnot involved in the collision but was skewed outof alignment by the derailment and lateral dis-placement of the MARC equipment.
Signals-- Train movements through the acci-dent area, including Georgetown Junction, arecontrolled by a traffic control signal (TCS) sys-tem�� on the two main tracks that is arranged fortrain movement in both directions with colorposition light signals���and the General RailwaySignal model 5H electric switch machines con-trolled from the operations center in Jackson-ville, Florida. The AU dispatcher controls thepower-operated switches at Georgetown Junctionfor crossover movements between tracks 1 and 2remotely from Jacksonville by leased telephonelines. East and west of Georgetown Junction thesystem consists of a TCS on two main tracksarranged for train movement in both directionswith color position light signals and electronictrack circuits.
The EAS at Georgetown Junction controlseastbound train movements on track 2, and it aswell as the eastbound signal 1124-2 at Kens-ington are mast-mounted color position lightsignals. (See figure 10.) Their method of opera-tion is by timetable, direct traffic control (DTC)block system,���and signal indication of a TCS.
A defect detector is at MP 11.7, about 0.5mile west of signal 1124-2 at Kensington, andafter the accident, it provided a paper tape print-out for train movements on track 2. (See table4.) The clock on the defect detector, measuring
11Train movements are authorized by block sig-nals whose indications supersede the superiority oftrains for both opposing and following train move-ments on the same track.
12Fixed signals that display aspects by the colorand position of two or more lights.
13Governed by the verbal authority of the traindispatcher, or a series of consecutive DTC blocksactivated by train movement on the track circuit.
21
in hours and minutes, is not synchronized witheither the event recorder at Georgetown Junc-tion or the system log in Jacksonville. The de-fect detector is approximately 16 minutes aheadof the Georgetown Junction event recorder and14 minutes ahead of the system log in Jackson-ville. The detector noted no defects on the threepassing trains.
The defect detector broadcasts an arrival anda trailing message, respectively, as a train ap-proaches and shortly after the last car passes.The arrival message is preceded by a 1/4 secondtone to alert the engineer, which is followed byan approximately 5-second message that identi-fies the defect detector MP location and tracknumber for an engineer to determine whetherthe message applies to him. If no defects aredetected (none were for either train in this acci-
dent), the approximately 10-second trailing mes-sage repeats the arrival message and then adds“no defects,” one of four brief safety messages,and a concluding “end of transmission.”
The Georgetown Junction interlocking isequipped with a signal event recorder to recordspecific events and signal relay changes corre-sponding to the dispatcher’s requests, trainmovements, signal indications, and interlockingconditions. Following the accident the signalevent recorder was secured and obtained infor-mation indicating that the last time the EAS wascleared for a train movement from track 1 to 1was at 3:14 p.m. for the eastbound MARC train284. About 4:05 p.m. the EAS recorded a STOPsignal and displayed this signal about 1 hour 33minutes until the accident at 5:39 p.m.
Figure 10--Signal 1124-2. The “D” was added after the accident.
22
Table 4.--MP 11.7 defect detector printout for track 2 train movements
Train Date Time Direction Axles Cars Speed Length (not recorded) (mph) (ft)
MARC 284-16 2/16 16:19 East 40 10 61 744
MARC 286-16 2/16 17:50 East 16 4 49 279
CSXT Q401-16 2/16 18:23 East 22 5 15 280
Operations Information
Train movements on the CSXT BaltimoreDivision, Metropolitan Subdivision, are governedby CSXT operating rules in the Operating Proce-dures Manual, effective January 1, 1995. Instruc-tions for the movement of trains or equipmentand related essential information are in BaltimoreDivision Timetable No. 5, issued on January 1,1996. Superintendent’s bulletins containing writ-ten special instructions about the movement andsafety of trains and employees are issued periodi-cally. The Metropolitan Subdivision is called theBrunswick Line for the MARC commuter trainoperations between Brunswick and Washington,approximately 50 miles. Eastbound MARCtrains on the Brunswick Line are normally oper-ated in the push mode and westbound MARCtrains in the pull mode.
Train movements through the accident areaare controlled by the CSXT AU train dispatcherfrom the operations center in Jacksonville. TheAU train dispatcher controls the movement ofapproximately 80 trains daily over three lines:Brunswick (Washington to Brunswick, whichincludes the Georgetown Junction crossover);Camden (Washington to Baltimore); and Penn(Baltimore to Brunswick). (See figure 11.) About28 unscheduled CSXT freight trains and 2scheduled Amtrak passenger trains as well asthe 18 weekday-scheduled MARC commutertrains on the Brunswick Line operate dailythrough the Georgetown Junction interlocking.
The Safety Board investigation of this acci-dent involved the activities of seven trains: Am-trak train 29, CSXT freight trains K951 andQ401, and MARC trains 279, 281, 284, and 286.About 4:05 p.m. MARC train 284 was the last
eastbound train to operate on track 2 through thecrossover switch at Georgetown Junction with theswitches lined in the “normal” or through posi-tion. MARC train 279 was the last westboundtrain to operate before the accident through thecrossover switch at Georgetown Junction, cross-ing from track 2 to 1 about 5:15 p.m. The CSXTtrain K951 was operating westbound towardGeorgetown Junction on track 1 and had stoppedeast of the Georgetown Junction (near the high-way overpass bridge) about 5:28 p.m. Amtraktrain 29 was operating westbound on track 2 pro-ceeding to make the same crossover as MARCtrain 279 had made and had been operating be-hind MARC train 279 before being struck byMARC train 286. The CSXT train Q401 andMARC train 281 had been operating, respec-tively, behind MARC train 286 and Amtrak train29 also before the collision.
CSXT Train Operations-- Trains�� through theaccident area operate under the authority of blocksignal indications of a TCS system. Authority formovement is governed by the CSXT operatingrules 265 through 271 for the TCS system rules.Passenger-specific operating rules are containedin the CSXT operating rules 620 through 635.The Baltimore Division Timetable No. 5 specifiesthe maximum authorized speed for the Metro-politan Subdivision as 79 mph for passenger and55 mph for freight trains. Trains can operate ineither direction on either track by signal
14The MARC trains began push/pull operations in1985 with the delivery of the MARC II cab controlcars.
23
Figure 11--Map of MARC operations for Brunswick, Camden, and Penn Lines.
indication. Between MP 11.0 at Kensington andMP 8.3 at Georgetown Junction, the maximumauthorized passenger train speed is 70 mph, ex-cept for a timetable speed restriction betweenMPs 10.6 and 9.5 for a reduction to 55 mph forpassenger trains.
The Safety Board reviewed event recordingsmaintained by the CSXT in three formats: com-puterized train sheets (FRA records of trainmovements),�� signal event logs, and train dis-
15Dispatcher records of train movements aremaintained by the CSXT in compliance with 49 CFRPart 228 and are required to identify the timetable, thetrain and engine, the traincrew with duty time, the
patcher voice communication recordings(including telephone). Train-to-train radio com-munications are indicated on the dispatcher’svoice recording only if the dispatcher is using theradio channel. All event records, with the excep-tion of the computerized train sheets, were vol-untarily maintained by CSXT. The Safety Board
dispatcher with duty time, the station and office, thelocation and date, the weather conditions in 6-hourintervals, the distances between stations, the directionof train movement, the time when the train passes eachreporting station, the train arrival and departure times ateach reporting station, and unusual events affectingtrain movement.
24
reviewed the train sheets for the previous 90 daysand found that the sheets significantly lacked therecording of unusual occurrences, such as thisaccident, and weather information, which areelements the FRA regulations require to bemaintained.
Operational Efficiency Testing-- The CSXTmaintains an operational test and observationprogram in compliance with 49 CFR Part 217 todetermine the extent of compliance with its oper-ating rules, timetables, and timetable special in-structions. The stated objectives of the programare to eliminate human error accidents and to im-prove employee compliance with and knowledgeof the operating rules and special instructions.
CSXT operating officers, following proce-dures and instructions contained in the CSXTEfficiency Test Manual dated April 1, 1993, testCSXT and Amtrak crews who operate overCSXT property. Of the 43,566 efficiency testsperformed on the Baltimore Division during thecalendar 1995, 1,334 test failures were recorded.The Metropolitan Subdivision had 2,357 of thoseefficiency tests and recorded 95 test failures. Themajority of the failures involved radio (37) andsafety rule (31) tests.
To comply with 49 CFR Part 240, the CSXTmaintains a Qualification and Certification Pro-gram of Locomotive Engineers. For 1995 theCSXT revoked 66 engineer certifications, whichranged from 30-day suspensions to permanentrevocations. Thirty-two, 14, and 6 suspensionsinvolved track authority, signal, and disarming asafety device violations, respectively.
The MTA/MARC neither has operating rulesfor its commuter trains nor performs operationalefficiency tests on traincrews because its trainsare operated by CSXT employees in accordancewith the CSXT operating rules or Amtrak em-ployees in accordance with the Amtrak operatingrules. Amtrak conducts operational efficiencytests on Amtrak traincrews operating on theCSXT for observance of the CSXT operatingrules as well as the Amtrak operating rules.
MARC Train Radio Communication-- Shortly be-fore the accident, MARC trains 286 and 279passed each other just west of the Kensingtonstation, and a radio transmission took place be-tween the engineers of the two trains. The engi-neer of MARC train 279 provided informationabout that interchange, which is contained in thefollowing synopsis:
The MARC train 279 conductor called hisengineer by radio when they were in SilverSpring to determine whether he had heard fromMARC train 286 yet. The engineer had not. Theconductor wanted the engineer to notify train286 that a passenger had boarded train 279 bymistake. The passenger wanted to go to Balti-more, not Brunswick. Since MARC train 286had not yet passed Silver Spring, the conductorsaid that train 286 could return the passenger toWashington from the Silver Spring station tocatch the proper train to Baltimore. The train279 engineer did not transmit that information totrain 286 at that time because the train was stilltoo far away to contact. He said that the twotrains usually passed each other around Kens-ington. Around Georgetown Junction, theMARC train 279 conductor called his engineerand again asked whether he had heard fromMARC train 286, and again, the engineer hadnot. The conductor reported that another pas-senger had also boarded the wrong train. He saidthat if they had not passed train 286 by the timethey arrived at Kensington station, he would letthe passenger off at that station for train 286 topick up and return to Washington.
Shortly thereafter, the MARC train 286 en-gineer contacted the MARC train 279 engineerand warned him about children in the GarretPark area who were throwing snowballs near thetrack. The train 279 engineer said that at first hehad not quite heard the other engineer and hadto ask him to repeat his message. The train 279engineer asked the children’s location, and thetrain 286 engineer responded that the childrenwere near the east end of the station platform.The train 279 engineer at that time beganwatching for the children or something on histrack. The train 279 engineer next told the train
25
286 engineer about the passengers at Kensing-ton and at Silver Spring waiting for train 286;the train 286 engineer had replied that that wasall right. The train 279 engineer, however, hadnot told the train 286 engineer on which side ofthe tracks the passengers had gotten off.
The MARC train 279 engineer stated that theradio communications took place near the defectdetector at MP 11.7 between Kensington (MP11.0) and Garrett Park (MP 12.4). He thoughtthat the two trains had passed each other nearMP 11.7 and that the rear of his train had possi-bly cleared the defect detector by four or fivecar lengths before the head end of his trainreached the head end of MARC train 286. Be-tween the radio communications of the two en-gineers, the defect detector was transmittingmessages to the trains over the same radio fre-quency. The train 279 engineer said that becauseof the defect detector broadcast, which is stillconveyed when engineers speak, he and the train286 engineer were not hearing each other’severy word. Nevertheless, he believed that theymanaged to get across their messages. Thecommunications began before the trains met andcontinued as they passed each other. At the endof the communications, the train 279 engineertold the train 286 engineer that everything wasall right behind him, which meant that no snow-ball-throwing children were behind his train.The engineer of MARC train 279 said he againheard the engineer of MARC train 286 use theradio to call out what he thought was the signalaspect at Kensington; however, he could notunderstand what signal aspect was called be-cause the defect detector radio was broadcastingits message at the same time.
MARC Train 286 Passenger Load Data-- MARCprovided the January and February 1996 passen-ger load information for MARC train 286. In-clement weather during January forced the can-cellation of 6 days of the train 286 scheduledservice. It operated 15 days and had a total rider-ship count for the month of 72 passengers. Thehighest single-day (Friday, January 26) total was21 passengers. Train 286 had operated 11 daysthrough Thursday, February 15, and had a total
ridership count of 79 passengers. The highest sin-gle-day (Friday, February 9) total was 25 passen-gers.
CSXT and MARC Operating Agreement-- An oper-ating agreement was first effected between theB&O, now the CSXT operations, and the Mary-land State Railroad Administration, now theMTA/MARC. Their last contracted agreement,dated November 1, 1985, was renewable for fiveadditional 5-year periods. The October 1990 op-erating agreement,��
� in effect at the time of theaccident, has been under negotiation since 1995for its second 5-year term. The contract providesthat the CSXT service will be “safe and efficient”and that the CSXT will maintain equipment tocomply with all applicable safety regulations ofregulatory bodies and manufacturers’ standards.The MTA owns the railroad equipment used forthe MARC service. The MTA/MARC has theright to review and audit equipment shop ordersand maintenance. The operating contract has noprovision regarding the delineation of responsi-bility for passenger handling. The CSXT directorof passenger services explained in an August1996 letter to the Safety Board that the contractobligates the:
CSXT to handle MARC’s commutertrains in a manner consistent with safeoperating rules and practices, to operatethem on schedule subject to operatingconditions and overriding safety con-siderations, and to operate them in acost efficient manner.
16The 1990 operating contract provides for limitedliability of the CSXT in the event of an accident in pas-senger service and affords that the MTA indemnify theCSXT for the first $150 million of exposure per acci-dent. The CSXT is not indemnified for any damagespayable to its employees from an accident; however,the compensation terms of the contract include a 20-percent additive to incurred labor costs intended toreimburse the CSXT for the risk of insuring its ownemployees.
26
He continued that the CSXT, as a freight railroad,does not have special expertise in the handling ofpassengers and “is responsible for matters associ-ated with train handling and MARC is responsi-ble for matters relating to the passengers them-selves.” The CSXT indicated that its crews werenot specifically trained in emergency proceduresfor passenger handling and that MARC, at itschoice, did not supply on-board train personnel.However, MARC informed Safety Board investi-gators that it believed that for MARC to provideits own on-board personnel would have been a“contradiction to CSXT’s labor agreements.”
(See Appendix E, Organization Charts of theCSXT and the MDOT/MTA/MARC.)
Meteorological Information
A strong low-pressure system, which was offthe central Atlantic coast on the morning of Feb-ruary 16, moved northeast during the day. Ac-cumulated snow, reduced visibility due to thesnow and fog, and strong winds were associatedwith the storm system.
The National Weather Service issued thefollowing short term forecast at 5:30 p.m. forMontgomery County:
Winter storm warning continues….Snow will continue this evening ac-cumulating 1 to 3 inches before ta-pering to flurries by 7 p.m…. Tem-peratures will drop into the lower 20sby 11 p.m. Winds will be north 15 to25 miles per hour and gusty. Therewill be blowing and drifting of snowas wind chills below zero.
Pathological and Medical Information
MARC Train 286-- On February 17, 1996,autopsies were performed in Baltimore on thethree MARC operating crewmembers and theeight MARC passengers who were fatally in-jured in the collision. The Maryland medicalexaminer recorded the postmortem examinationreports.
The engineer, who was found on the exteriorleft side of cab control car 7752, received fatalmultiple injuries and generalized body burns.The conductor, who was recovered from theaisle floor at row 10, sustained fatal multipleinjuries. The assistant conductor, who was dis-covered lying over a right side seat of row 9,received fatal smoke and soot inhalation andgeneralized body burns. The eight passengerfatalities were also in cab control car 7752: twosustained fatal smoke and soot inhalation inju-ries, three received fatal smoke and soot inhala-tion injuries plus general body burns, one sus-tained fatal generalized body burns, another re-ceived fatal soot inhalation injuries and gener-alized body burns, and the final victim sustainedfatal multiple injuries with generalized bodyburns. (See figure 12 for the postaccidentplacement of fatalities, which was based on in-formation obtained from the MARC train survi-vors, the MCPD, and the Maryland medical ex-aminer report.)
Amtrak Train 29-- The engineer had remained inthe locomotive control compartment and re-ceived blunt torso injuries and multiple contu-sions. The assistant engineer, who jumped fromthe locomotive unit before the collision, sus-tained pulmonary contusions and multiple faciallacerations. The mechanical rider, who was inthe control compartment of the second locomo-tive unit, received blunt torso injuries and aclosed head injury. He said that he had beenthrown against the back wall of the locomotive.He and the two operating crewmembers wereadmitted to a local hospital. The conductor inthe midsection of the train sustained a facialabrasion and stated that he had struck a table.The assistant conductor in the last car also re-ceived a facial abrasion but from striking a seat.Two on-board service crewmembers sustainedminor injuries. All four employees were treatedat and released from local hospitals. Eight Am-trak passengers received bruises and abrasionsand were taken to area hospitals, treated in theemergency rooms, and released.
28
Toxicological Information
The postaccident specimens, which wereacquired from the three deceased crewmembersof MARC train 286, consisted of blood for allthree and tissue and other body fluids, as ob-tainable under the circumstances. Postaccidentblood and urine specimens were also acquiredfrom the Amtrak engineer, assistant engineer,conductor and assistant conductor. The federallyapproved laboratory, Northwest Toxicology,Inc., in Salt Lake City, Utah, conducted the testsfor marijuana, cocaine, phencylidine (PCP),opiates (morphine and codeine), and ampheta-mines (amphetamine and methamphetamine)and for alcohol in accordance with the FRApostaccident testing program. No drugs or alco-hol were identified in either traincrew’s speci-mens.
The Maryland medical examiner during theautopsy process conducted a more comprehen-sive drug screening than the FRA-required test-ing. Neither other drugs nor alcohol were foundfor the engineer and assistant conductor. How-ever, acetaminophen, orphenadrine, and antipy-rine were detected in the conductor’s bile andantipyrine also in his blood. The nonprescrip-tion-analgesic acetaminophen relieves pain andreduces fever. Orphenadrine is a prescriptionmuscle relaxant. The prescription-analgesic-anesthetic antipyrine, applied within the ear,relieves the pain and swelling of some ear in-fections and additionally removes ear wax.
At the request of the Safety Board, furthertesting on the available specimens was con-ducted by the Center for Human Toxicology inSalt Lake City. The test results confirmed theFRA and medical examiner’s test results.
No postaccident toxicological testing speci-mens were obtained from the CSXT train dis-patcher who was on duty at the time of the acci-dent. FRA regulations require specimens beprovided by the dispatcher following an acci-dent and their testing to occur only should thedispatcher be “directly and contemporaneouslyinvolved in the circumstances of the acci-
dent/incident.” To determine whether the dis-patcher should be tested, the CSXT divisionmanager of operations reviewed the signal logfor the involved railroad line and the audio tapesof the dispatcher’s conversations before the ac-cident, interviewed the dispatcher, and con-ferred with the CSXT director of operatingpractices. The two CSXT officials noted that theaudio tapes indicated the dispatcher had notconversed with either traincrew or anyone elsein the area during the relevant timeframe beforethe accident. They also determined that the dis-patcher had lined the involved signal apparatuswell in advance of the accident and its positionhad remained unchanged. As a result of theseobservations, the CSXT officials found the dis-patcher not “directly and contemporaneouslyinvolved” in the accident and, thus, excluded thedispatcher from postaccident testing.
Emergency Response
Montgomery County Fire and Rescue Services--About 5:41 p.m. the MCFRS 911 dispatcherreceived at least 12 telephone calls from citizenswho reported the train collision and fire. At 5:42p.m. the MCFRS dispatcher notified fire andrescue personnel through a computer activateddispatch system. Also at 5:42 p.m., the Mont-gomery County 911 emergency operator re-ceived a cellular phone call from the conductoron Amtrak train 29.
At 5:46 p.m., the first firefighter arriving onscene reported a fully engulfed passenger carwhich he was unable to enter or open the doors.By 6:15 p.m., an emergency medical services(EMS) command was established as well as thefire sectors situated for the management of thelevel two incident.��
17This level of command is appropriate for use onserious, involved, or extended incidents in which fiveor more units are used. It is established at thedirection of the IC and requires a formal commandpost and the use of command communications andincident position designations.
29
The IC said that at 6:30 p.m. his communi-cation center advised him that another train wasapproaching the accident site from the directionof Kensington. The IC staff activated radiotones and air horns to warn emergency respond-ers to evacuate the scene, and it was cleared forabout 10 minutes until the IC determined thatthe approaching CSXT train Q401 was not ahazard to the emergency responders.
The last victim was removed from thewreckage by 3:50 a.m. on February 17 andemergency operations were suspended. Duringthe rescue operations, five on-scene triage siteswere available for the injured victims, who werelater transported to five area hospitals for treat-ment.
Under Montgomery County Emergency Op-erations Center (EOC) direction, two nearbyschools were used to provide victims with shel-ter, crisis counseling, and staging for transpor-tation. County, emergency management, andvolunteer agencies’ officials and membersstaffed the EOC; however, MARC, Amtrak, andthe CSXT were not represented at the EOC.State, county, and volunteer organizations sup-plied services at both the accident site and shel-ters.
Montgomery County Disaster Plan--Montgomery County has a disaster plan�� to en-sure maximum preparedness for response to andrecovery from any disaster that occurred withinthe county. The on-scene IC decided not to fullyimplement the disaster plan for this accidentbased on the information developed during theinitial response. The disaster command system(DCS), a part of the plan that provides for mul-tiagency response and integrates the incidentcommand system used in the field, was acti-vated. The DCS established a standardized op-erational system with unified command, com-prehensive resource management and inter-
18Approved by the Governor in 1988 and certifiedby the county executive and the county chief admin-istrative officer in November 1995.
agency coordination, and cooperation from theEOC. The Fire and Rescue Disaster Plan annexO was also activated. The plan outlines how anincident will be managed in the field and howthe IC will coordinate with the disaster managerin the EOC. Additionally, the Greater Metro-politan Washington Area Police andFire/Rescue Services Mutual Aid OperationalPlan��was implemented between Montgomery,Prince George's, and Howard Counties inMaryland, Fairfax County in Virginia, andWashington, DC.
Montgomery County simulated a Washing-ton Metropolitan Area Transit Authority(WMATA) train derailment in Rockville in Feb-ruary 1995 during which at least 24 people weretransported to hospitals and shelters were estab-lished with county agencies' participation. Ahazardous materials incident was also includedin the derailment exercise. In October 1995 an-other WMATA train derailment was simulatedwith casualties during a 30-inch snowfall.MCFRS units responded on January 6, 1996, toan actual WMATA train derailment during asnowstorm, which involved one fatality, at theShady Grove station in Gaithersburg, Mary-land.� The MCFRS had not participated in anydisaster drills that involved MARC, Amtrak, orthe CSXT and in any disaster training or famili-arization training with any freight or passengerrailroad that provides service in MontgomeryCounty.
Amtrak, MARC, and CSXT-- Amtrak reported toSafety Board investigators that about 6:15 p.m.one of its officers made four attempts within 10minutes to provide the passenger list and otherinformation to the MCFRS personnel at the
19Part of the Metropolitan Washington Council ofGovernments plan and last revised in September1990.
20Railroad Accident Report--Collision of Wash-ington Metropolitan Area Transit Authority Train T-111 with Standing Train at Shady Grove PassengerStation, Gaithersburg, Maryland, January 6, 1996(NTSB/RAR-96/04).
30
command center in the adjacent parking lot. Hewas finally told that the information was notneeded and he should wait. Later he askedMCFRS personnel at this command center towhich hospital the traincrew had been taken,and they told him that they did not know. Am-trak did not locate the traincrew in the hospitaluntil 10:30 p.m.
The assistant superintendent of the CSXTBaltimore Division told the Safety Board that ontwo occasions at the MCFRS command post andon one occasion at the accident site, he identi-fied himself and asked to assist. He was told thathis assistance was not needed but to stand by, incase he would be needed. Another CSXT offi-cial was later asked to participate in a mediainterview with the MCFRS public relations offi-cer.
Neither MARC nor the CSXT either had anaction plan for accidents or emergencies, in co-ordination with each other or Amtrak, or hadperformed, in conjunction with each other orAmtrak, any emergency disaster drills or exer-cises that involved the CSXT operation ofMARC passenger train equipment. In addition,neither MARC nor CSXT personnel had at-tended any classroom instructions or hands-onexercises that included the use of emergency fireand rescue equipment or procedures.
According to the MARC Interim SystemSafety Plan of April 5, 1995, MARC had madeits passenger equipment available in 1995 to theBrunswick Volunteer Fire Department and tothe Maryland Fire and Rescue Institute foremergency response training. On April 18,1995, the Amtrak manager of emergency pre-paredness had conducted 4 hours of classroomtraining and 2 hours of hands-on training at aMARC facility to members of the BaltimoreCounty Fire Department and the Cowenton[Maryland] Volunteer Fire Department. OnOctober 18, 1995, the Maryland Fire and Rescue
Institute provided a 3-hour class,�� which in-cluded classroom discussion, a photographicslide presentation, and a tour of MARC passen-ger equipment at the Brunswick CSXT railroadyard, for several fire and rescue departments ofFrederick County, Maryland. MARC reportedthat it participated in this training by providingthe equipment.
Survival Aspects
MARC Train 286 Occupant Evacuation-- Two pas-sengers had occupied the second car, passengercoach 7720, and exited it through an emergencywindow. One of these passengers was injuredand needed assistance. The other passenger losthis eye glasses after being thrown from his seaton the left side of the car in the fourth or fifthrow. He said that he had been waiting to buy aticket from the conductor when he heard an ex-plosion, the lights went out, and no emergencylights came on. As he was starting for the rearexit because displaced seats obstructed the exitat the forward end of the car, the injured pas-senger pointed out the right-side emergencywindow near the center of the car. He reportedthat he pulled the rubber molding around theemergency window and “then pulled on thewindow handle and nothing happened.” He de-cided to then put one foot on the seat and toplace his other foot on the wall beside the win-dow. He said:
Just so I can get better leverage and Ipull and then it starts coming unstuck,and the top comes unstuck. So I grabthe top of the actual window with myhands and pry the rest of it off andthen put that down.
He stated that removing the emergency windowtook about 3 minutes and that helping the in-jured passenger to exit and his own exit requiredabout 7 minutes more.
21Part of the 45-hour Rescue Specialist Trainingcourse that has been offered throughout the State tofull-time and volunteer emergency responders since1991.
31
Ten of the 18 passengers on board cab con-trol car 7752 survived the accident. The passen-gers were unsuccessful in opening the left andright rear exterior doors after the accident. Theten surviving passengers had exited the trainthrough an opening created by the misalignmentof the passageway between cab control car 7752and passenger coach 7720. The longitudinalmisalignment of both cars was about 15 degrees.Cab control car 7752 came to rest upright andhad rotated vertically clockwise about 10 de-grees; passenger coach 7720 had also rotatedvertically clockwise, but about 30 degrees.
Nine surviving passengers were U.S. De-partment of Labor Job Corps students, and theystated they had known nothing about the opera-tion of the emergency doors or windows. Somestated that they did not see any emergency plac-ards explaining the use of emergency exits andthat no crewmember had given any emergencyannouncements before the accident. One studentremarked that he had seen the emergency plac-ards but had paid no attention to them. The sur-vivors noted that about half of the students wereasleep before the collision. One stated that hehad been seated in the rear section of cab con-trol car 7752 when the conductor ran from thefront of the car shouting, “Brace yourself! Geton the floor!” The student dropped to the floorand held onto the seat on which he had beensitting. He said that windows had shatteredduring the collision, that the car was on fire af-terwards, and that he could feel the heat fromthe fire on his back. He reported that he crawledto the rear of the car and tried unsuccessfully toopen both side doors. He saw light through anopening between the cars and climbed throughthe space.
One student, who was sitting next to anemergency window on the last seat in the right-rear section of control cab car 7752, describedthe smoke as extending from about 2 feet abovethe floor to the car ceiling. This student said thathe did not have time to open the window andthat he believed he could escape faster by usingthe door through which he had entered the car.
Another student stated that the conductorwith another person came from the front of thecar shouting, "everybody run to the back" andthe conductor had reached the midpoint of thecar when the collision occurred. The studentreported that after the collision, he was thrownbetween the seats, the lights went out, smokecame into the car, and that other students werescreaming and running to the rear of the car. Hesaw students unsuccessfully attempting to openthe rear side doors, and he explained that be-cause of the fire, no one attempted to open theemergency windows. He slid to the floor be-cause he could not see or breathe with thesmoke. He observed that while he was on thefloor, one person exited “off the train through ahole or crack in the train.” He followed anotherperson through that opening and ran down a hillto a large residential building. He estimated thathe escaped within 5 minutes.
One student, who had been standing andtalking to another student, stated that he hadseen two crewmembers running rearwardthrough the car from the control compartmentand had heard them shouting, what he under-stood to be, “aboard, aboard.” He noted that thetwo crewmembers appeared to be expressingshock and disbelief that another train was ap-proaching on the same track. He said that he sawthe conductor jump behind one of the seats andthat he himself crouched behind and held ontothe last seat on the left side of the car in antici-pation of the collision. He recounted that heproceeded immediately after the crash to therear door exits, that he felt the heat from the fireon his back, and that smoke quickly filled thecar. He reported that he looked “high and low”for “handles or gadgets or something to open thedoors” but smoke obscured his vision and thathe could find nothing to open the doors. Headded that he pounded with this hands on thedoor windows but the glass would not break.Other students were running toward him,coughing from the smoke, and yelling for some-one to open the doors. He also heard bangingand kicking noises that suggested other passen-gers were attempting to open or break windows.
32
Firefighter Evacuation Action-- The first fire-fighters to reach cab control car 7752 observedthe car fully engulfed by fire and found no sur-vivors. The firefighters said that they attemptedto open the right-rear door of the car with an ax,a halligan tool, and a 10,000 psi hand-pumpedhydraulic tool. The door had opened about thewidth (0.5 foot) of the ax head, could not beopened any farther, and closed when the toolswere removed. One firefighter reported that hehad observed two victims within 0.5 foot of theright-rear vestibule door of cab control car 7752.He stated that he then endeavored to find an-other entrance into the car because the doorcould not be opened.
Subsequently, the firefighters broke the firstright-side window on passenger coach 7720 andentered that car through the broken window.They exited passenger coach 7720 through thedamaged left-front door and crawled down tothe ballast on the north side of the train. Twofirefighters crawled up on the couplers and en-tered cab control car 7752 through the dia-phragm opening that had been used as an escaperoute by the surviving passengers from the car.After the two entered the cab control car 7752, athird firefighter passed a hand line through asmall opening between the two cars. (Becauseof the angle at which the cars came to rest, noperson could pass through the vestibules fromone car to the other car.) The firefighters in cabcontrol car 7752 proceeded to suppress the fireand search for victims.
The firefighters stated that they had not ob-served either any exterior instructions detailinghow to open the emergency windows or the lo-cation of the emergency door release handles onthe car exterior.
Tests and Research
Locomotive Event Recorders-- MARC unit 73was equipped with a Pulse MTR magnetic tapeevent recorder and Amtrak units ATK 255 and811 were equipped with Pulse Train Trax®
solid state event recorders. The Pulse solid stateevent recorders on Amtrak units ATK 255 and811, respectively, used a FAK PAK® remov-able recording media and a nonremovable mem-ory module with battery-backed SRAM mem-ory.
The event recorders were removed fromMARC unit 73 and the Amtrak lead unit ATK255 under Safety Board supervision between 9and 10:30 p.m. on February 16, 1996. The eventrecorder from the Amtrak second unit ATK 811was not removed but was downloaded the nextmorning. The readouts of the three event re-corders were successfully performed between 1and 10 a.m. on February 17. A subsequent re-view of the raw electrical wave forms from theMARC unit 73 event recorder data was per-formed on March 26 and 27, 1996, using SafetyBoard laboratory hardware and WAVES soft-ware, which was required to establish more ac-curate timing of events than possible with themanufacturer readout software.
MARC Train 286 and Amtrak Train 29 Move-ment-- At 4:45 p.m., shortly after its eastbounddeparture from Brunswick on track 2, MARCtrain 286 was stopped behind the eastboundCSXT freight train Q401, which had been dis-abled beyond Point of Rocks, Maryland. TheMARC train 286 engineer radioed the AU dis-patcher, who directed him to make a reverse(westbound) move on track 2 to Point of Rocksand to cross over there onto track 1. MARC train286 departed Point of Rocks eastbound on track 1at 4:55 p.m., traveled eastward for 12 milesaround the disabled train, and returned at 5:11p.m. to track 2 at MP 30. The event recorder dataindicated that the train stopped at the Rockvillestation, accelerated to 74 mph, maintained forabout 1 minute a speed of between 72 and 74mph, and 3 minutes 15 seconds and 2.9 milesfrom that station initiated dynamic braking nearthe signal at MP 13.78. The train then deceler-ated under dynamic braking for 1 minute 33seconds and 1.33 miles until attaining a speed of20 mph near the Garret Park station (MP 12.4)and subsequently accelerated for 1 minute 13seconds and 0.8 mile until reaching 53 mph near
33
the defect detector at MP 11.7. WestboundMARC train 279, operating on track 1, passedeastbound MARC train 286, still operating ontrack 2 at 53 mph, near a dragging equipment de-tector 0.7 mile west of the Kensington station.The train 279 engineer then advised the train 286engineer that passengers were to be picked up atthe Kensington and Silver Spring stations.
MARC train 286 at this time decelerated atidle throttle for 26 seconds and 0.36 mile, thenits engineer initiated dynamic braking at a speedof 46 mph, and 6 seconds later the train passedsignal 1124-2 at 44 mph. It stopped 42 secondsand 0.2 mile later at the Kensington station andremained there about 50 seconds. The event re-corder data indicated that it had traveled about2.18 miles after leaving the station when anemergency brake application was made at 66mph. Dynamic and independent braking beganabout 6 seconds and 540 feet after the emer-gency brake application, and the traction motorcurrent for dynamic braking increased to 375amps. The train was traveling approximately 49mph when, about 6 seconds and 460 feet later(1,000 feet after the emergency brake applica-tion), the reverser on MARC unit 73 was movedfrom reverse to forward. The traction motor cur-rent decreased to 0 amps when the direction oftravel changed, and the axle generator drivewheel of MARC unit 73 stopped rotating about13 seconds and 500 feet later (1,500 feet afteremergency brake application). Rapid decelera-tion at a speed of about 38 mph, consistent withimpact with Amtrak train 29, occurred duringthis time.
Amtrak train 29 departed Union Station 1hour 20 minutes late (about 5:25 p.m.) because ofelectrical problems with Amtrak unit ATK 811.The train was routed westbound on track 2 be-hind MARC train 279, which had departed 9minutes earlier. The engineer of Amtrak train 29stated that at MP 6.5 he received an APPROACHMEDIUM signal indication, which limited thetrain speed to 30 mph approaching the next sig-nal, and he began reducing the train speed in theSilver Spring area. He noted that he neither hadradio communications nor overheard any trans-
missions during that time. The engineer said thathe received a MEDIUM CLEAR indication fromwestbound absolute signal (WAS)-2 for Amtraktrain 29 and looked again at the signal to confirmthe indication and, also, that CSXT train K951was stopped at WAS-1 on track 1 east of theGeorgetown Junction.
According to the event recorder data fromAmtrak units 255 and 811, Amtrak train 29 hadtraveled about 7.7 miles since its last stop andhad been moving for about 1 minute between 27and 32 mph when data variations, consistentwith impact, occurred at 32 mph. Eight secondsbefore the data ended, the throttle had beenmoved from position three to four; about 5 sec-onds and 250 feet before the data ended, a 10psi automatic brake pressure reduction had beenmade. The engineer stated that when he sawMARC train 286 moving toward his train, he de-cided that a collision was imminent and, thus, toremain at his operating position and to acceleratethe train in an attempt to prevent a head-on colli-sion. However, the postaccident investigationfound no recorded change in speed.
Sight Distance-- Tests were conducted on Feb-ruary 19, 1996, between 5 and 6 p.m. to deter-mine the optimum sight distance to the waysidesignals as they would have been observed by theMARC train 286 crew at Kensington andGeorgetown Junction. A similarly configuredMARC train was used in the push mode on east-bound track 2. The weather conditions could notbe replicated; however, the MARC train 279,CSXT train K951, and Amtrak train 29 engineershad reported no difficulties seeing signal indica-tions during the intermittent snow showers onFebruary 16. The eastbound signal 1124-2 atKensington, 885 feet west of the station, was dis-playing an APPROACH signal aspect (see table1). The optimum sight distance to this signal was4,299 feet. EAS-2, the next signal, was 13,786feet east of eastbound signal 1124-2, and a STOPsignal aspect (see table 1) was visible at 1,737feet.
34
Signals-- After being notified of the accident,the Safety Board had requested that all signalinstrument houses be sealed. A preliminary in-vestigation, which entailed a visual inspectionof all signals, began about 8:15 p.m. at theGeorgetown Junction control station and con-tinued until all involved signal cases were un-sealed, including the signal cases east and westof Georgetown Junction and Kensington. Theinvestigation documented the position of vitalrelays and tested for grounds or combinations ofgrounds at the energy bus. No grounds as wellas no evidence of tampering were found at anysignal instrument house.
Other tests performed to determine that cir-cuits were effective and functioning properlyincluded battery voltage readings; route, time,and switch indication locking; time release re-lays; and cable meggering.�� All test results in-dicated compliance with FRA regulations. Volt-ages of 9.8 and 10 volts d.c. were recorded ateastbound signal 1124-2, respectively, for thetop marker lamp and the APPROACH indica-tion (see table 1). No signal lamps were burned-out.
Operational tests pertaining to the involvedroutes were conducted, and switch and trackcircuit relays were manipulated because of thetrack structure absence. A route was establishedat Georgetown Junction for a westbound movefrom track 2 to 1 with switches 1 reverse and 3and 5 normal. WAS-2 was cleared and a ME-DIUM CLEAR indication was observed (seetable 1), and EAS-2 at Georgetown Junctiondisplayed a STOP indication (see table 1). Asthe two simulated trains proceeded toward eachother, EAS-2 was repeatedly requested for aPROCEED indication and never displayed otherthan a STOP indication. The Kensington east-bound signal 1124-2 never displayed other thanan APPROACH indication.
22Testing for the electrical resistance of wire andcable insulation.
During the time-locking tests, the time ele-ment relay on tracks 1 and 2 was, respectively, 8minutes 9 seconds and 8 minutes 13 seconds.The November 3, 1994, CSXT test recordsshowed that tracks 1 and 2 were 8 minutes 10seconds and 8 minutes 5 seconds, respectively.The Georgetown Junction signal event recorderdata and the Jacksonville system log were com-pared. The event recorder data indicated thatMARC train 286 passed eastbound signal 1124-2 about 5:34 p.m. and that the collision occurredabout 5:39 p.m. (total elapsed time of 4 minutes44 seconds). The system log recorded that thetrain passed the signal about 5:36 p.m. and thatthe collision occurred about 5:41 p.m. (totalelapsed time of 4 minutes 46 seconds). The col-lision occurrence was based on the Georgetownindustrial track switch being “out of correspon-dence” (not in proper position).
After the on-site investigation, a meeting washeld at the CSXT headquarters in Baltimore todocument complaints about the signal systemoperation. (See appendix F.) The Safety Boardproposed at this meeting that further testing beperformed at Georgetown Junction and specifi-cally requested the presence of the involvedparties for the investigation. After the damagedtrack and turnouts were replaced, additionaltesting for route and time locking were per-formed on May 28, 1996. During these tests,EAS-2 displayed only a STOP indication, andeastbound signal 1124-2 remained at an AP-PROACH indication until the track circuit eastof the signal was shunted and then displayed aSTOP and PROCEED indication.
Stopping Distance-- The Safety Board con-ducted these tests on November 23, 1996, usingtrain equipment similar to MARC train 286, todetermine whether differences were in the stop-ping distances between the push and pull modesof train operation and from moving the reverserfrom the reverse to forward position duringemergency braking. The tests were performedon relatively flat straight tracks on the Amtraknortheast corridor near Bowie, Maryland. Theweather was clear with the temperature at 40 oF.
35
Five tests were done each in the push and thepull modes with speeds ranging from 59 to 66mph. The stopping distances extended fromabout 1,200 feet at 59 mph with dynamic andemergency braking to about 1,900 feet at 66mph with only emergency braking. The test re-sults indicated no significant differences instopping distance between the push and the pullmodes of operation; however, use of the rever-ser did eliminate dynamic braking and, thus,increase stopping distance. MARC train 286 at66 mph (recorded speed before emergency brakeapplication), according to the test data, wouldhave required approximately 1,650 feet of stop-ping distance after an emergency brake applica-tion with dynamic braking until stopped com-pared with approximately 1,900 feet of stoppingdistance after an emergency brake applicationusing the reverser with no dynamic braking after49 mph. The test data also demonstrated thatretaining dynamic braking until impact, MARCtrain 286 would have collided with Amtrak train29 at 34 mph (recorded collision speed of 38mph), resulting in an additional 0.3 second ofelapsed time and in Amtrak train 29 advancingabout 14 feet farther into the crossover beforeimpact.
Passenger Train Air Brakes-- Some locomo-tives, such as those operated on Amtrak, auto-matically combine dynamic and pneumaticbraking, which is called blended braking, forgreater efficiency and less mechanical wear.MARC locomotives, such as the one involved inthe accident, are designed to apply the dynamicbrakes automatically during emergency braking,and the dynamic brakes supplement the pneu-matic braking. The dynamic brakes continue toapply in emergency until a wheel slip occurs, orthe reverser is moved, at which time the dy-namic braking discontinues and the braking isfully pneumatic. Moving the reverser duringemergency braking resets the electrical contacts,which interrupts the dynamic braking circuits,drops the electrical load, and eliminates the dy-namic braking; thus, moving the reverser duringemergency braking causes a loss of the addi-tional braking effort that the supplemental dy-namic braking provided.
The CSXT engineer training in place at thetime of the accident had advised that reverseruse on a locomotive in motion was inappropriateunless “the brakes fail when a locomotive isbeing moved at low speed without any cars at-tached” and other operational use would “in-crease braking distance due to wheel slide.” Thetraining material did not indicate that using thereverser would eliminate dynamic braking. TheCSXT training and mechanical officials statedthat they were unfamiliar with the automaticdynamic braking feature that occurred duringemergency braking on the MARC locomotives.
Cab Control Car 7752 Doors-- The tests anddisassembly of the rear interior door performedby the Safety Board on May 2, 1996, revealedthat the debris within the door wall pocket andfloor track had prevented the door from closing.The right-side rear door interior emergency han-dle and T-handle were pulled downward andwere functional. The right-side rear door wasmanually opened and slid into its car bodypocket. The left-side rear door contacted a loosesheet metal panel in its car body pocket whenthe door was manually opened and, therefore,could be opened only about 12 inches.
The left-side door opening mechanism wasremoved and tested in the Washington, DC,Safety Board laboratory. Test results showed nomechanical damage but all electrical compo-nents had thermal damage. Sealant melting andthen hardening on the pinion and ball bearing, asa result of the fire, caused the resistance to rota-tion of the motor shaft and to movement of thepinion bearing and the fused toggle spring andself-aligning spherical bearing, which contrib-uted to the inoperable emergency handle on theleft-side rear door.
MARC Passenger Coach Car Interior Materials--The Maryland Department of Transportation(MDOT) purchase contract for 11 passengercars, including cab control car 7752, containedthe following smoke and flammability specifi-cations:
36
All materials used in the interior of thecar (that is, all materials inboard of thestructural shell and including, but notlimited to, liners, floor panels, thermaland acoustic insulation, seats andcushions, floor covering materials,wainscots, carpeting, glazing materi-als, and light fixture lenses) shall havethe highest degree of fire resistanceand lowest smoke emission consistentwith the other qualities required. As aminimum, all materials used in the in-terior of the car shall meet the re-quirements of the U.S. Department ofTransportation’s [DOT] “ProposedGuidelines for Flammability andSmoke Emission Specifications.”��
To determine the compliance of the MARCcar interior materials with the flammability andsmoke requirements, tests were conducted onmaterials from an exemplar MARC passengercar by the University of Maryland Departmentof Fire Protection Engineering at College Park,Maryland. (See appendix G.) These tests weregoverned by the FRA recommendations fortesting the flammability and smoke emissioncharacteristics for commuter and intercity railvehicle materials.
The tested materials, consisting of the majorcombustible items in the car, included the up-holstered portion, but not the rigid plastic side-rail and back components, of the seats (see fig-ure 13); the ceiling lining, which was similar, ifnot identical, to the other wall lining and parti-tion materials; and the window mask material.The pad cushion materials of the seat passed the
23Developed by the DOT in 1979 and used by thetransit industry in general on a voluntary basis, theseguidelines were superseded by the “RecommendedFire Safety Practices for Rail Transit Materials Se-lection,” published in 1984 by the Urban MassTransportation Administration, now the Federal Tran-sit Administration. The recommended practicesformed the basis for the FRA recommendations fortesting the flammability and smoke emissions char-acteristics of commuter and intercity rail vehicle ma-terials, published in January 1989.
smoke criterion but failed the flammability cri-terion. The fabric upholstery seat coveringpassed the flammability test but failed thesmoke test in the flaming mode. The vinyl seatcoverings passed the flammability test but failedthe smoke test in the nonflaming mode. Theceiling panel passed both the smoke and flam-mability criteria; however, the window maskmaterial failed both criteria. Floor materialswere not included in the tests because the cabcontrol car 7752 floor was not destroyed by thefire.
Fuel Oil Presence-- On March 18, 1996, firedebris samples were collected from MARC cabcontrol car 7752 and second car 7720 to docu-ment the presence of diesel fuel. The chroma-tographic results from three samples from cabcontrol car 7752 revealed weathered diesel fuelmixed with heavier unknown petroleum hydro-carbon.
Postaccident Actions
National Transportation Safety Board-- OnMarch 12, 1996, during its investigation of thisaccident, the Safety Board issued the followingfour urgent safety recommendations (see appen-dix H) to improve the safety of the rail com-muting public:
--to the Maryland Mass Transit Administra-tion:
Install removable windows or kickpanels for emergency exits in interiorand exterior passageway doors. (R-96-4)
Install an easily accessible interioremergency quick-release mechanismadjacent to all exterior doors. (R-96-5)
Install retro-reflective signage on carinteriors and exteriors at emergencyexits that contains easily understoodinstructions and clearly marks allemergency exits (doors and windows).(R-96-6)
37
Figure 13--Photograph of typical seats in MARC passenger cars.
--and to the Federal Railroad Administration:
Inspect all commuter rail equipment todetermine whether it has: (1) easily ac-cessible interior emergency quick-release mechanisms adjacent to exteriorpassageway doors; (2) removable win-dows or kick panels in interior and ex-terior passageway doors; and, (3)prominently displayed retro-reflectivesignage marking all interior and exte-rior emergency exits. If any commuterequipment lacks one or more of thesefeatures, take appropriate emergencymeasures to ensure corrective actionuntil these measures are incorporatedinto minimum passenger car safetystandards. (R-96-7)
Federal Railroad Administration-- The FRAstated that because this and the recent train acci-dent in Secaucus, New Jersey,�� had caused seri-ous concerns about certain aspects of the safety ofpassengers and railroad employees, it issued onFebruary 20, 1996, emergency order (EO) 20 toall railroads. (See appendix I.) The MARC op-erations were covered by EO 20, which requiredthe CSXT to modify operating rules coveringdelays in block and crew communications andMARC to address the operation and inspection ofand the instructions on emergency exits as well asto provide interim system safety plans.
24Railroad Accident Report--Near head-on Colli-sion and Derailment of Two New Jersey CommuterTrains near Secaucus, New Jersey, on February 9,1996 (NTSB/RAR-97/01).
38
Maryland Transportation Administration andMaryland Rail Commuter-- Either the AU dis-patcher or the chief dispatcher at the CSXT op-erations center in Jacksonville relays nonemer-gency information to MARC and Amtrak througha passenger coordinator desk, which provides thisinformation to the MARC operations center�� atBaltimore Washington International (BWI) air-port, Maryland. To manage emergency situationsthat require immediate attention, the dispatcherand the MARC operations center use a direct line.
When the CSXT dispatcher became aware ofthe collision between MARC train 286 and Am-trak train 29, the CSXT manager of operationssupport (MOS) in Jacksonville tried telephoningthe MARC BWI operations center about 5:45p.m., but the phone line was busy. While trying toreach the MARC operations center, the MOS didinform the CSXT passenger services director inJacksonville and both CSXT managers of passen-ger operations in Baltimore. The MOS stated thathe was unable to get through by telephone to theMARC operations center until 6 p.m., at whichtime he was informed that “they knew” about thecollision and were very busy. Some MARC offi-cials who had the Amtrak alphanumeric pagerswere made aware of a MARC train involvementin the accident only by reading messages meantfor the Amtrak personnel. Since the accident, theCSXT and MARC have established direct com-munication procedures to be followed in theevent of an emergency situation and have avail-able a dedicated phone line for emergency noti-fication. In addition, the CSXT has access to theMARC officers’ alphanumeric pagers.
CSX Transportation Inc.-- In response to theFRA EO 20, the CSXT installed D markers (seefigure 9) for use in passenger train operations to
25Its consoles mirror the CSXT AU dispatcher dis-play screens and monitor the AU dispatcher radiocommunications on channels only in use by the dis-patcher. From this information, the commuter trainpassengers are advised by a station public address sys-tem of train arrivals and delays and passenger boardingdirections.
remind train operators of the requirements in therevised CSXT operating rule 269 (TCS system),which states:
When a train has passed a signal per-mitting it to proceed (other than a Re-stricting, a Stop and Proceed at Re-stricted Speed, or Grade aspect) and isstopped in the block, the train mustproceed prepared to stop at the nextsignal. This must be done until it canbe seen that the next signal permits thetrain to proceed.
Exceptions:
A. This rule does not apply topassenger trains, making stationstops, this includes push-pull, orMU [multiple-unit] passengertrains:
1. Equipped with auto-matic train control/cabsignal apparatus
or2. Operating immedi-ately preceding an in-termediate signal, how-ever, rule 98-G doesapply.
Passenger trains in push-pull, or MUservice not equipped with automatictrain control/cab signal apparatus, oroperating immediately preceding anabsolute signal be governed by excep-tion (B) below.
B. If a passenger train in push-pull, or MU service has passeda block signal that permitted itto proceed, (other than an ap-proach, a restricting, a stop andproceed at restricted speed orgrade aspect), and
1. Is stopped in theblock
or
39
2. The train speed is re-duced below 10 mph.
The train must proceed preparedto stop at the next signal not ex-ceeding 40 mph. The train mustnot exceed the indication per-mitted by the previous signal.This must be done until,
a. The next signalis clearly visible,b. That signal dis-plays a proceed in-dication, andc. The track to thatsignal is clear.
National Railroad Passenger Corporation-- InOctober 1996 as the first of 20 planned installa-tions, Amtrak made the OREIS�� software avail-able to the Montgomery County fire departmentand rescue communications center to assist firstresponders to train accidents. The software pro-vides schematics of Amtrak cars and locomo-tives with layout information about the seatingconfiguration, the emergency exit doors andwindows, and the electric and fuel sources.
Cooperative Organization Activities-- Alsosince the accident, MARC, in cooperation withthe CSXT and Amtrak, has developed and dis-tributed video training material for emergencyresponders. The MTA safety department, inconjunction with MARC, the CSXT, Amtrak,and the Maryland Fire and Rescue Institute, haseffected a training course for emergency re-sponders and scheduled several training dates.MARC and the MCFRS have initiated a trainingprogram to become familiar with emergencyequipment on the MARC passenger trains.
26Operation Respond Emergency Information System,developed by the Operation Respond Institute, Inc.,of Washington, DC.
Other Information
Maryland Mass Transit Administration-- As themodal administration of the MDOT, the MTAoperates the MARC train service as well as thelight rail, subway, and bus services for approxi-mately 370,000 passengers each day. The MARCtrain service averages 20,000 passenger tripsdaily compared with 50,000 and 20,000 passen-ger trips daily, respectively, for the subway andlight rail operations.
Regularly scheduled daily (Monday throughFriday) MARC commuter rail service is providedon the Camden and Penn Lines between Balti-more and Washington and on the Brunswick Linebetween Martinsburg, West Virginia, and Wash-ington. The CSXT and Amtrak, under contract,operate daily 40 and 39 trains, respectively. TheMARC weekday schedule for the Brunswick andCamden Lines was last revised on December 11,1995. (See appendix J.) Nine scheduled MARCpassenger trains are provided daily in each direc-tion on the Brunswick Line. Two morningMARC trains originate and three evening MARCtrains terminate in Martinsburg, and Brunswick isan intermediate stop for these trains. The re-maining MARC trains originate or terminate inBrunswick.
In addition to a chief transportation officer,the MARC operating staff consists of threetrainmasters, who address the station operations,such as the condition of the platforms and walk-ways, and ten staff personnel, who act as ticketagents, operations clerks, and customer servicerepresentatives. The MARC trainmasters alsorespond to passenger complaints and resolve pas-senger operational issues with the CSXT; theyhave no responsibility for or authority over trainhandling.
The CSXT and Amtrak supply and supervisethe traincrews, control the train movement dis-patching, and provide the equipment mainte-nance, depending on whose property is beingused. The CSXT, under its contractual agreement,allocates two managers of passenger operations,
40
who supervise the CSXT traincrews. MARC paysthe cost of these positions. The on-duty CSXTconductors and assistant conductors are requiredto wear MARC uniforms, which MARC supplies,to assist passengers in identifying the traincrew.The CSXT engineers are not required to wearthe MARC uniforms because they do not havedirect contact with the passengers.
The CSXT, sometimes in coordination withMARC, issues Passenger Service Bulletins tocover unique circumstances that occur during theoperation of MARC trains. These bulletins in-clude passenger safety alerts, schedule changenotices, and general guidance. The CSXT andMARC were developing a passenger conductor’sguidance manual that incorporated the PassengerService Bulletins, which MARC had been re-viewing at the time of the accident and is stillreviewing now.
MARC Train Accidents-- During the previous 5years, the CSXT reported two train accidents thatinvolved MARC trains operating on the CSXT.Both accidents involved nonrail equipment foul-ing the track on the Camden Line betweenWashington and Baltimore. In one accident,MARC train 244 collided with a backhoe thatwas fouling the track near Hanover, Maryland, inDecember 1992. Neither a derailment nor injurieswere reported, and the equipment damages were$15,000. The second MARC accident occurredwhen an empty MARC train, being moved in thepush mode to Washington by a CSXT crew, col-lided with a concrete pouring boom fouling thetrack near College Park, Maryland, in November1993. The cab control car derailed and the engi-neer and two private-contractor personnel sus-tained injuries; the damages were estimated at$475,000.
Maryland Department of Labor and Industry--The Maryland Department of Labor and Indus-try (MDLI) is accountable for occupationalsafety and health in the State of Maryland. It isresponsible for boiler, amusement ride, and ele-vator as well as railroad safety. The MDLI hasan agreement with the FRA to conduct inspec-tions, and three MDLI railroad safety inspectors
work, under the FRA state participation pro-gram, in the track, operating practices, and mo-tive power equipment disciplines by reportingthrough the FRA and its reporting system. TheMDLI is unable to perform independent inspec-tions, must conform to the FRA inspectionregulations, and must channel all activitiesthrough the FRA; consequently, the MDLI is notpermitted to have an independent inspectionprogram.
The MDLI has jurisdiction over some areaswhere the FRA would not be involved, such asprivate yards where it inspects the tracks, but ithas no jurisdiction over rapid rail systems,which include the two metro systems and thelight rail system in Baltimore. However, it doesinspect some of the track in the light rail systembecause freight operations are on some of thattrack. It has little direct relationship with theMTA regarding railroad issues, but it does allMARC inspection activity (operations, me-chanical, and track) that would be done with theoperating carriers such as Amtrak and CSXT.According to the MDLI, its inspection reportsare sent directly from Amtrak or the CSXT toMARC, and it does not work directly withMARC, unless specific issues arise that putthem in contact from the inspector level to theoperating level.
The MDLI has no signal inspector and hadno responsibility for oversight of the design andinstallation of the CSXT signal system modifi-cations on the Brunswick Line, and it stated thatit believed the FRA had the oversight responsi-bility. As recently as 1989, the MDLI had abouteight inspectors; currently they have only three,even though the number of passenger and com-muter trains has increased. The MDLI informedthe Safety Board that it is considered the moni-tor of carriers in the State and not the safetysystem for the State. The MDLI was aware ofthe FRA EO 20 but had not participated in thedevelopment of that order and had not seen it.Also the MDLI had not participated in the de-velopment of the MARC interim system safetyplan submission to the FRA.
41
Brunswick Line Signal Modifications-- About1985, MARC requested the CSXT to allowmore commuter trains on both the Camden andBrunswick Lines. (MARC wanted to add 22trains on a daily basis between 5 and 10 a.m.and between 4 and 9 p.m. in addition to theJanuary 1, 1990, schedule.) The CSXT providedcost estimates in 1986 after which an agreementwas signed. The development of a joint projectto increase the service capacity of the corridorfrom Baltimore to Washington to Brunswickwas begun in August 1988. The project was de-signed to permit signaled, dispatcher-controlledmovement between main tracks in either direc-tion. In April 1990, the Board of Public Worksawarded a project contract of $13.085 million tothe CSXT. The areas covered were the 40 milesfrom Camden Station to Union Station and the50 miles from Brunswick Station to Union Sta-tion. The Federal Transit Administration(FTA),��� formerly the Urban Mass Transit Ad-ministration (UMTA), the CSXT, and theMDOT were to pay, respectively, 75, 12.5, and12.5 percent. The MTA and the CSXT woulddivide any cost overruns of the modifications by87.5 and 12.5 percent, respectively.
The signal modification project was designedin 12 phases, and plans were subsequently is-sued in October 1991. Construction began southof Baltimore and proceeded towards Brunswick.Track changes during construction necessitatedchanges in signal design, which were approvedby MDOT. New crossovers were proposed forseveral locations which included MP 19.7 atDerwood, Maryland. The crossover installationat Derwood added a new control point (CP),which required respacing the signals betweenGeorgetown Junction and Derwood. As a result,additional safe braking distance for mixedfreight and passenger operation was achieved byreducing from four to three the number of in-termediate signals between Georgetown Junc-tion and Derwood.
27Established in 1968, under the Urban MassTransportation Act of 1964, to administer Federalgrants to urban mass transit projects for new acquisi-tion, construction, and operations and for the im-provement of existing facilities and equipment.
The project was completed in July 1993 at afinal total cost of $17.085 million. The MTApaid the cost overrun of $3.9 million plus $0.10million for the final inspection. The Federal andState interest in the fixed assets of the signalsystem will be 87.5 percent for the 15-year lifeof the improvements. At the end of this 15years, July 2006, the CSXT assumes full owner-ship.
Federal-Grant-Requested Signal Project-- Inaccordance with the FTA act under which Fed-eral assistance is allocated, the FTA providesgrants through section programs of discretionaryand formula funds. Discretionary funds are allo-cated at the discretion of the FTA administrator,and the formula funds are apportioned by astatutory formula based on population, popula-tion density, and various transportation data.The FTA obligated, for example in 1995, ap-proximately $2.6 billion of discretionary fundsand about $3.1 billion of formula funds.��
� Ac-cording to the FTA, local organizations andtransit systems submit grant applications to theappropriate FTA regional office. The granteesself-certify that they have the legal, financial,and technical capacity to engage in the projectfor which they are applying for FTA funding.
In May 1987 the MDOT applied to the FTAfor a rail modernization grant of approximately$9.8 million to assist in the financing of a fiscal-year 1987 capital project, generally described asthe “installation of centralized traffic controlsystem on MARC/CSXT Lines.” The MDOTState Railroad Administration (now the MTA)was delegated to administer this project. In ad-dition to the Federal funds, the MDOT and theCSXT would each provide about $1.6 millionfor the project. The grant application includedthe excerpted project justification:
28The MTA indicated that for a $940-million 6-year-period (1996 through 2001) capital program,approximately $667 million of Federal funds wereanticipated. MARC would account for $390 million,of which $294 million would be Federal funds, of this$940 million.
42
The centralized traffic control (CTC)system on the CSXT lines will modifythe method and efficiency of traincontrol and operations, thus providingincreased system capacity, and conse-quently allowing expanded use of theCSXT lines for commuter rail opera-tions. More specifically, the CTC willpermit:
the addition of trains in bi-directional service during peak pe-riods: head-ways could be reducedto twenty minutes in each direc-tion. This would provide the ca-pacity to double service levels onthe Brunswick Line and tripleservice on the Washington-Baltimore Line.
the addition of bi-directional serv-ice in off peak periods: Midday,evening, weekend, and specialevent service will be possible, en-hancing commuter as well as shop-per, tourist, and non-work relatedtrips by providing increased flexi-bility in scheduling trips.
continued safe, combined move-ments of freight and passengertrains on the CSXT lines. Byminimizing the possibility of con-flict between freight and passengertraffic through computerizedmonitoring and control of move-ments, the CTC will enhance thesafety of MARC and CSXT opera-tions.
When questioned whether it would conduct asafety analysis of the project, given the abovestatements in the project justification, the FTAindicated that a safety analysis, or at least a pre-liminary safety analysis, would be done at thelocal level as the entities develop their trans-portation improvement programs. Under thoseauspices FTA employees would not review thegrant application because the grantee has certi-
fied the technical capacity to undertake the proj-ect. For this particular grant, according to theFTA, the budget included $450,000 for engi-neering, which would be the design of the signalsystem upgrade, and that would be conducted bythe grantee or the grantee contractor, in this casethe CSXT. Because this grant application in-volved an upgrade to the signal system, SafetyBoard investigators questioned whether the FTAhad in-house signal expertise or would look tothe FRA for any guidance on the benefits orshortcomings of the MDOT proposed signalproject. According to the FTA, it had no in-house signal expertise and no communicationwith the FRA regarding this project because itwas not required to have the FRA involved. TheFTA has indicated that subsequent to the SilverSpring accident, the FTA administrator has re-quested that the FRA become more involved inreviewing operational plans and safety proposalsfor grantees, including commuter railroads.
The MDOT grant project (exhibit A) de-scription included:
The improvements envisioned in thisprogram provide the foundation for thenext “generation” of train control sys-tems: advanced train control system(ATCS). This system will permit dis-crete train identification and provideremote override operation of the trainitself. The technology for ATCS isprogressing with significant commer-cial applications starting in 5-10 years.
The FTA or the MTA did not follow up theMDOT pursuit of this technology. According toMARC officials, the industry had no consensuson ATCS or commercial applications at thetime.
Following the approval of the grant, theCSXT and the MDOT/MTA signed an agree-ment in March 1990 regarding the installation ofthe CTC system on CSXT lines. The signalsystem, according to the agreement, would bedesigned in compliance with all applicable Fed-
43
eral and State requirements and installed by theCSXT. The MTA would have the right to re-view and comment on the design and construc-tion plans. However, the CSXT, according tothis agreement, would have exclusive controlover the design, installation, and operation ofthe signal system. The agreement stated that “atthe end of 15 years, which is calculated fromJuly 1, 1991, or the operational start date,whichever comes first, CSXT will assume fullownership of the fixed assets and UMTA’s andSRA’s [State Railroad Administration] net sal-vage credit shall be reduced to zero.”
CSXT Signal System Modification Design-- TheCSXT uses several equal factors in the determi-nation of signal spacing and traffic density.First, if it is necessary to run trains on shortheadway,�� the signals will need to be spacedclose enough to allow the trains to move safelybehind one another. The aspects displayed bythe signals must inform the train engineer whatactions need to be taken to safely advance thetrain. Another factor is the train braking distancerequired to stop or slow a train to the prescribedspeed, conforming to the signal indication withrespect to the operating rules. The CSXT stan-dard braking distance for freight trains is 11,000feet for 55-mph operation or greater than 13,000feet for 60-mph operation.
When signals must be spaced closer thanthree-aspect breaking will allow, a fourth aspectneeds to be employed so that trains would beoperating at less than maximum authorizedspeed should conditions require the train to stop.In a two-block three-aspect signal system, anormal signal progression would be CLEAR,APPROACH, and STOP. As example of a three-block four-aspect signal system, the normal sig-nal progression would be CLEAR, AP-PROACH-MEDIUM, APPROACH, and STOP.
29Time separation between two trains traveling inthe same direction on the same track, which is meas-ured from the instant the head end of the leading trainpasses a given reference point until the head of thetrain immediately following passes the same referencepoint.
When freight and passenger trains both operateon the same set of tracks, the more restrictivebraking data for freight trains is used.
The proposal to install the TCS operation onthe Brunswick and Camden Lines was con-ceived, designed, and constructed to increase thecapacity of these subdivisions to accommodateadditional traffic proposed by MARC. The op-erating headway to meet the MARC schedulingrequest was a 15-minute headway during themost dense scheduling periods and a signifi-cantly more-than-15-minute headway in othernonpeak periods. To increase the capacity on theMetropolitan Subdivision, the TCS operationwas proposed with crossovers at MPs 19.7 and29.9 as well as a power turnout to access thePotomac Electric Power Company plant at MP37.
To accommodate the braking requirementsfor freight trains, the intermediate signal spacingneeded to be no less that 11,000 feet. Automaticsignals 99 and 100 at MP 10 were part of a sin-gle direction, automatic block signal system inservice between CP Rocks at MP 42.8 and CPGeorgetown Junction at MP 8.5. The controlcircuitry for these signals was inconsistent withthe controls required for TCS operation becauseit had been designed for one direction operation.The last automatic wayside signal before theCSXT signal modification project had been sig-nal 100, which had been east of the Kensingtonstation platform. Additionally, signal 100 wasless than the 11,000 feet minimum braking re-quirement from the EAS-2 signals at CPGeorgetown Junction. As a result of the signalmodifications, signal 100 was replaced by signal1124-2, which was now the last automatic way-side signal before EAS-2 for Georgetown Junc-tion and was west of the Kensington stationplatform.
Neither the FRA nor the MDOT/MTA re-viewed the CSXT design of the proposed signalsystem modification before installation. TheCSXT did not have to apply to the FRA to re-view and approve the proposed CSXT signalmodifications on the Brunswick Line. The FRA
44
made changes in 1984 to the regulations (49CFR Part 235). An extensive list of changes wasdeveloped that the FRA believed should not re-quire prior approval to implement a signal modi-fication;� therefore, the installation of the TCSsystem to replace an automatic block signalsystem did not require prior approval. The FRAstated that the reason a railroad would changefrom an automatic block signal to a TCS systemis safety and efficiency. In addition, the FRAregional inspectors were aware of the signalmodifications through routine inspections, andthe signal installation on the Brunswick Linewas FRA-inspected on three separate occasionswith no exceptions taken. The FRA performednormal inspections of the signal system modifi-cations after installation for compliance withFederal regulations and the CSXT signal plans.
CSXT Traincrew Roster Consolidation-- CSXThad consolidated its seniority rosters on January30, 1996, which combined six rosters into one.That change lowered the roster position of thethree crewmembers of MARC train 286 and wasreported to be a topic of conversation on the dayof the accident.
The roster change began in late 1993 whenthe CSXT initiated an effort through the U.S.Interstate Commerce Commission (ICC) to con-solidate the train and engine service work forcesof the former B&O, Western Maryland, andRichmond, Fredericksburg, and Potomac rail-roads. The approximately 1,300 employees whowere involved worked in the territory boundedby Philadelphia, Pennsylvania; Richmond, Vir-ginia; and Parkersburg, West Virginia. The la-bor unions objected to the consolidation, and anarbitration hearing was held in March 1995,with a decision rendered 1 month later. Neitherthe CSXT nor the labor unions were satisfied
30The changes were made to clarify the meaningof a material modification, a discontinuance, a cata-strophic occurrence, and track change. These changeswere based on information acquired through the expe-rience of investigating applications for signal changesin signal and train control systems.
with the arbitrator’s decision, and both subse-quently appealed it to the ICC, which decided infavor of the consolidation in December 1995.The labor unions then appealed to the courts,and the U.S. Circuit Court in Washington re-fused to stay the ICC decision; therefore, theCSXT went forward with its consolidationplans. The labor unions then advised the CSXTof their intention to strike over the issue, afterwhich the CSXT obtained a court-issued re-straining order. The consolidation was subse-quently implemented on January 30, 1996, al-though litigation has continued.
Before the consolidation, the MARC train286 engineer had been the number 11 engineerof 126 engineers on the Baltimore West EndEngineer Seniority Roster. After the consolida-tion, he was the number 50 engineer of 428 en-gineers. The MARC train 286 conductor andassistant conductor had been on two prior ros-ters. As trainmen, they were on the BaltimoreWest End Trainman Seniority Roster, and theconductor and assistant conductor, respectively,had been numbers 34 and 11 of 246 trainmen.After the consolidation, the conductor and as-sistant conductor, respectively, became numbers110 and 36 of 695 trainmen. On the BaltimoreWest End Conductor Seniority Roster, the con-ductor and assistant conductor, respectively, hadbeen numbers 133 and 99 of 255 conductors.After the consolidation, the conductor and as-sistant conductor, respectively, became numbers328 and 273 of 664 conductors.
The regular conductor for MARC train 286stated that the engineer did not show muchemotion about the roster change. However, hesaid that the engineer was aware of and wasconcerned about the possibility that as the low-est seniority engineer in passenger service, hecould be the first to be bumped from his as-signment by a higher seniority engineer. Theregular conductor for MARC train 286 charac-terized the assistant conductor as “very dis-traught” and vocal over the seniority issue. Onthe day before the accident, the topic was dis-cussed at a safety meeting in Brunswick, whichthe assistant conductor, the engineer, and he had
45
attended (the accident-day conductor was not atthe meeting). The regular conductor said, “Itupset him [the assistant conductor] enough thaton our run on the [MARC train] 286 goingfrom...Brunswick to Washington that he stood inthe cab and talked about it the whole way downthe road. So he was really upset about it.”
On the day of the accident, the assistant con-ductor again discussed his concerns with a fel-low conductor and a United Transportation Un-ion (UTU) representative. The conversationtook place at the Riverside Yard after the assis-tant conductor reported for work in the morning.The UTU representative reported that the assis-tant conductor was “very upset” about themeeting that had occurred the day before. Theengineer, the assistant conductor, and the con-ductor also all discussed the seniority change onthe day of the accident at the yard office inBrunswick. They spoke to another engineer whosaid, “nobody seemed to like it.” The assistantconductor and the engineer of MARC train 286also spoke to a fellow conductor about theirconcerns during their lunch break about 3:30p.m. MARC train 286 left Brunswick about 1hour later.
The wife of the assistant conductor wasasked about his comments concerning the sen-iority change, and she said that they had dis-cussed it previously in general terms. She indi-cated her husband was not too worried about hisjob and had told her there was nothing he coulddo about the seniority change anyway and thathe just had to go along with it. Still, no one hadbumped him from his assignment.
The wife of the conductor was asked abouther husband's comments concerning the recentalteration in the seniority structure at work. Shesaid that they had discussed it, although not inthe last week. She said he did not know what hisseniority position would ultimately be, but hewas not worried about it because he was "stillworking." He also told her that no changes werelikely before November because of the litigationbetween the CSXT and the various labor unions.
She added that her husband told her that nochange would take place in the conductor orengineer seniority rosters for passenger serviceat all and that it would remain the same as be-fore.
Locomotive Fuel Tanks-- The postaccident in-spection of the fuel tank of the MARC unit 73indicated that it was not breached during the de-railment. The fuel tank of the Amtrak lead loco-motive unit ATK 255 was almost completelyseparated from the unit and was found lodgedagainst and partially underneath the left side ofthe unit, adjacent to its normally mounted loca-tion. The left side-plate of the tank was cata-strophically ruptured open and showed substan-tial shredding deformation and impact striations.The circumstances suggest that the fuel tank ofthe lead Amtrak unit ATK 255 ruptured on im-pact with the MARC cab control car 7752 andthat the diesel fuel therein ignited and engulfedthe cab control car. There are no regulatory re-quirements or industry design specifications forthe locomotive fuel tanks involved in this colli-sion; however, the Association of American Rail-roads (AAR) Recommended Practice (RP)-506,Performance Requirements for Diesel ElectricLocomotive Fuel Tanks, effective September 1,1995, is the current industry standard for all lo-comotives built after that date.
Cellular Service Communication-- During theemergency response activity of this accident, thearriving fire and rescue personnel were unableto communicate by cellular phone with off-siteaccident support personnel because of an appar-ent saturation of the cellular phone network. Theon-scene news media quickly engaged andwould not release the existing cell sites in thearea. This situation prevented the rescue per-sonnel from making any connections and ham-pered the rescue efforts according to severalrescuers, who stated that this problem had oc-curred previously during other high-visibilityaccidents in which the news media were present.
46
ANALYSIS
General Factors
The witness statements, inspection reports,and locomotive event recorder readouts pro-vided no evidence of equipment failure. Thetrack inspections and measurements before andafter the accident indicated no defects or devia-tions from the FRA track safety standards. TheSafety Board investigation revealed that theCSXT AU dispatcher conducted the dispatchingactivities properly.
Results of the postaccident toxicologicaltests found that the three MARC train 286crewmembers and the two Amtrak train 29 lo-comotive crewmembers were not impaired byalcohol or drugs. The investigation also exam-ined the experience, sleep/wake cycle, andhealth of the three MARC train 286 crewmem-bers and two Amtrak train 29 crewmembers todetermine whether those factors influenced theaccident. Except for the MARC engineer’s nor-mal operational experience with MARC train286, which will be discussed later, these factorshad no effect on the accident.
Therefore, the Safety Board concludes that thetrain equipment and the track functioned asdesigned and that the AU dispatcher conductedthe dispatching activities properly. Neither thethree MARC train 286 crewmembers nor the twoAmtrak train 29 locomotive crewmembers wereimpaired by alcohol or drugs. All traincrewmembers were in good health, had noevidence of fatigue, and were experienced in andqualified for their duties.
Because of the reports of blowing snowfallin the accident area, the Safety Board reviewedthe locomotive event recordings, AU dispatchervoice recordings, and statements from othertraincrews for any negative impact that theweather might have had on the visibility ofKensington signal 1124-2. No traincrew oper-ating in the accident area reported being unableto distinguish signal indications. The MARC
train 279 engineer stated that he heard the radiotransmission from the MARC train 286 engineeracknowledge signal 1124-2, as required byCSXT operating rule 34-A; however, the way-side equipment defect detector had distorted thattransmission. The Safety Board concludes thatthe weather conditions did not impair the abilityof the MARC train 286 crewmembers to distin-guish the indication of Kensington signal 1124-2.
The Safety Board staff conducted a postac-cident signal examination that revealed the sig-nal indications for MARC train 286 before thecollision would have been an APPROACH atKensington signal 1124-2 and a STOP atGeorgetown Junction signal EAS-2. Addition-ally, all postaccident signal tests indicated thatthe proper codes were being transmitted toKensington from the electronic track circuit unitat Georgetown Junction. The signal systemtransmitted codes in a logical progression whentested, and the signals displayed were not inconflict with each other. All tests of signal1124-2, EAS-2, and WAS-2 indicated that con-flicting signals could not be displayed. There-fore, the Safety Board concludes that the signalsystem functioned as designed.
Accident Narrative Review
The APPROACH indication of signal 1124-2required the MARC train 286 engineer to slowhis train to not more than 30 mph after passingthe signal and to be prepared to stop at theGeorgetown Junction signal. The collision oc-curred because the engineer did not operateMARC train 286 in conformity with the signalindication when he stopped at Kensington sta-tion and then proceeded towards GeorgetownJunction, attaining a speed of about 66 mph. Theengineer’s actions after departing the Kensing-ton station were appropriate had signal 1124-2been CLEAR, but his actions were inappropriatefor an APPROACH aspect.
47
The Safety Board determined from thestopped position of the MARC train 286 loco-motive and its event recorder information thatthe engineer placed the train into emergencybraking 1,407 feet before the collision at a speedof about 66 mph. The engineer made the emer-gency brake application about 510 feet afterpassing the optimum sight distance location,about 1,227 feet from the EAS-2 or 5.27 sec-onds later. The delay is understandable and rea-sonable considering the engineer’s apparent be-lief that he was operating under a CLEAR signalindication.
There is no reason to suppose that theMARC train 286 engineer would be looking forthe Georgetown Junction signal as soon as itwas physically visible. If the engineer thoughtthat his last signal (1124-2) was CLEAR, noneof the signals he could have normally expectedat Georgetown Junction would have been sorestrictive as to demand his immediate action.Hence, he had no reason to try to see the signalas soon as possible. In addition, there was noradio conversation between train engineers andthe dispatcher that could have provided theMARC train 286 engineer with a clue on theother trains operating in the area. Disbelief waslikely once he or the other crewmembers or bothobserved the STOP signal at Georgetown Junc-tion. The crew would have then consumed sometime trying to reconcile the restrictive STOPindication with an expected CLEAR indication,which had been the norm for them at George-town Junction. One of the passengers stated, “Icould see the look, like bend over and check tosee if something’s coming, then they jump backlike in shock, then they went forward again justto double check,” which would attest to disbe-lief on the part of the traincrew.
About 407 feet before impact, the reverser ofMARC train 286 was moved when the trainspeed was about 49 mph. The speeds at impactwere 38 mph and 32 mph, respectively, forMARC train 286 and Amtrak train 29. Aftercolliding with Amtrak train 29, MARC train 286came to rest with its cab control car 7752 about93 feet beyond the point of impact.
The Safety Board identified the followingsafety issues in its investigation of this accident:the performance and responsibility of theMARC train 286 crewmembers, the oversight ofCSXT signal system modifications, the Federaloversight of commuter rail operations, the lackof positive train separation control systems, andthe adequacy of passenger car safety standardsand emergency preparedness. In addition, theSafety Board considered the use of a reverserduring an emergency brake application, the ef-fectiveness of the computer-aided train dis-patching recordkeeping, the crashworthiness oflocomotive fuel tanks, and the contents of theCSXT and MARC operating agreement.
MARC Train 286 Engineer Performance
The actions of the MARC train engineerprompted two questions that would need to beanswered to understand the accident events:Why did he behave as he did? How could awell-respected, experienced engineer forget asignal?
Although it is possible that the engineer didnot see signal 1124-2, this does not appear to belikely because the engineer of the passingMARC train 279 overheard a partial radiotransmission that he believed was the MARCtrain 286 engineer acknowledging the signalindication as required by CSXT operating rules.(The MARC train 279 engineer did not hear,however, what signal aspect was acknowl-edged.) In addition, based on witness statementsand the normal operating practice of the assis-tant conductor to ride this trip segment fromBrunswick to Washington in the cab control car,at least one other crewmember was, and perhapsthe other two were, in the cab control car duringthe time that the signal was visible. The CSXToperating rule 34 requires that crewmemberscall out the signal as it becomes visible. That atleast one other crewmember was present arguesagainst the signal not being seen.
Another possible explanation for the engi-neer’s behavior is that he saw the signal, forgot
48
the aspect after passing it, and realized he didnot remember the aspect. However, this is notlikely because the engineer could have simplyasked the other crewmember or crewmembers inthe cab control car about the aspect. If he or theyalso did not know the aspect, the engineershould have immediately slowed his train to therestricted speed,�� as required by CSXT generaloperating rule S, which states, “In case of doubtor uncertainty, the safe course must be taken.”
However, it is more likely that the engineerobserved and correctly identified the signal andcorrectly repeated its indication over the radio,but he then forgot its aspect because other in-formation interfered with its retention.�� Thespecific information and tasks that demandedthe engineer’s attention and interfered with hisretention of the signal information came fromseveral sources, some routinely encountered andsome unique to the trip. These sources includedthe important mental and physical tasks requiredto stop the train at Kensington station, the vari-ous radio conversations with the engineer ofMARC train 279, the defect detector broadcastsand disrupted radio conversations, and the pres-ence of the crewmembers in the cab control car,possibly discussing an important topic.
The physical and mental tasks associatedwith stopping the train at Kensington stationprovided the primary source of interference. Theattention demanding tasks included reducing thethrottle, applying the train brakes, and stoppingat the proper platform location at a station wherethe engineer seldom stopped. He was alsoprobably watching through the snow to ensurethat the passenger waiting at the at-grade plat-form was clear of the path of the train.
31Defined by the CSXT operating rules as “Aspeed that will permit stopping within one-half therange of vision. It will also permit stopping short of atrain, a car, an obstruction, a stop signal, a derail oran improperly lined switch. It must permit looking outfor broken rail. It will not exceed 15 MPH.”
32Psychological literature refers to the process asretroactive interference: new information interferingwith the retention of previously learned or sensedinformation.
Other sources of interference for the MARCtrain 286 engineer included the various radioconversations held with the MARC train 279engineer and the defect detector broadcasts thatneeded to be listened to but at the same timedisrupted the radio conversations. The defectdetector broadcasts caused both engineers torepeat information back and forth to ultimatelyattain it, which was a process that required in-creased attention for a longer period of timethan otherwise necessary. The radio conversa-tions that had to be repeated between the twoengineers included the information about thechildren throwing snowballs and their location.As both trains continued, the engineers ex-changed more information concerning the pas-senger at Kensington and the one at SilverSpring to be picked up by MARC train 286,which was important information because nei-ther station is a regular station stop for the train.
As these conversations took place, bothtrains passed over adjacent defect detectors,which also provided information important tothe safe operation of the train. Indeed, “if/then”CSXT operating rules 58A-G, 60, and 60A pre-scribe what an engineer must do under variouscircumstances if a defect detector locates a de-fect. Many of the prescribed actions may in-volve immediately stopping and inspecting thetrain and, thus, require the immediate attentionof the engineer. With the trains so close to eachother, the engineers needed to attend to bothmessages, first to determine who the messageapplied to and secondly because of its potentialsafety implications. The MARC train 279 engi-neer stated that these conversations occurredabout MP 11.7, which is less than 0.5 mile fromsignal 1124-2. Because the signal could havebeen visible from about 0.8 mile, these commu-nications had the potential to delay perceptionof that signal or to interfere with the aspect re-tention or both.
In addition, the CSXT operating rules requirethat when two trains pass each other, each engi-neer is required to conduct a visual inspection ofthe other train as it passes, which would requirethe engineer’s attention during the passing.
49
When a train passing event and a defect detectorbroadcast occur concurrently, an engineer bene-fits by listening to the broadcast associated withthe other train to cue him where to look for spe-cific problems. This is a practice of many engi-neers. The defect detector, however, broadcastson the same radio frequency used by engineersto talk to each other or the dispatcher or both.Consequently, when the defect detector began toautomatically broadcast, radio communicationbetween the two trains was disrupted. This dis-ruption made understanding each other difficultand required repetition and clarification of in-formation as well as additional attention.
The last source of interference may havecome from the presence of a crewmember orcrewmembers in cab control car 7752 with theengineer and the possible discussion of an im-portant topic. The other crewmembers have es-sentially no function in the cab control car, butthey may have been talking to the engineer.Their presence, however, can be beneficialwhen they assist the engineer in observing thetrack and signals and the conversation is limitedto essential operational exchanges. Listening toor participating in a conversation with anothercrewmember would further divide the engi-neer’s attention and interfere with his retentionof the signal aspect. A conversation betweencrewmembers likely did occur in cab control car7752 on the accident day because of the practiceof the assistant conductor to ride this segment ofthe trip in the cab control car and to talk duringthe trip. Conversation creates a potential fordistraction and interference with the engineer’sretention of information, in this case the signalinformation.
Although the information provided by thesignal 1124-2 aspect was critical for directingthe engineer’s actions after the station stop, ithad no relevance to the task immediately athand, which was stopping at Kensington station.The locomotive event recorder showed that theengineer started reducing speed about 0.5 milebefore the signal. The speed reduction is be-lieved to have been in response to the stoppingfor passengers and not to the APPROACH sig-
nal indication because compliance with the sig-nal would not have required the engineer to re-duce his speed as soon as he did.
Ample sources of competing information andtasks were present and capable of interferingwith the engineer’s retention of the signal 1124-2 information. One or more of those sourcesinterfered with his retention of that information,which is believed to have caused the signal indi-cation to be completely forgotten as if it had notexisted. Nevertheless, when the engineer leftKensington station, he had to have some guid-ance, which is normally supplied by the signals,on how to operate over the next block of track.However, because signal 1124-2 was before thestation and his stop, the engineer needed to re-call the signal aspect after intervening informa-tion and tasks. The regular conductor for MARCtrain 286 could not recall ever having to stop atGeorgetown Junction, which would mean thatthe engineer would not have normally observedany other indication but CLEAR at signal 1124-2. On the accident day, the engineer may haverecalled the expected CLEAR indication, ratherthan what was actually there, and operatedMARC train 286 accordingly. The Safety Boardtherefore concludes that the MARC train 286engineer apparently forgot the signal aspect,which required him to be prepared to stop atGeorgetown Junction, due to interference causedby various events, including performing an un-scheduled station stop, that occurred between thepresentation of the APPROACH aspect at signal1124-2 and the STOP signal at GeorgetownJunction.
MARC Train 286 CrewmemberResponsibility
As MARC train 286 departed the Kensingtonstation, its engineer was required to comply withthe signal indication that had been displayed onsignal 1124-2. An APPROACH indicationrequired that the engineer limit the train speed to30 mph or less and be prepared to stop atGeorgetown Junction. According to the eventrecorder data, however, the engineer acceleratedthe train speed to 59 mph in the next 1.15 miles,
50
gradually reduced the speed, and then accelerateduntil an emergency brake application was made at66 mph. The acceleration rate from the stationstop was constant for the first 0.8 mile, reaching aspeed of about 50 mph in just over 1 minute. Theengineer exceeded the 30 mph speed restrictionwithin the first minute of the departure. As aresult of this evidence, the Safety Boardexamined why other crewmembers in the cabcontrol car apparently did not take action asrequired by the CSXT operating rules.
The responsibilities of the various crew-members in the operating cab control car arequite different. The engineer’s role of physicallyoperating the train requires him to be in the lo-comotive or cab control car of the train. The jobresponsibilities of conductors in passengerservice require them to be primarily in the pas-senger cars. Consequently, the nature of eachjob generally separates crewmembers and keepsthem in different places on the train. (Neitherthe CSXT nor MARC prohibits other crew-members being in the operating cab control car.)The CSXT recognizes that crewmembers otherthan the engineer may have occasion to be in theoperating cab control car and has assigned re-sponsibilities to them when they are. The CSXToperating rules 34 and 34C basically requireother crewmembers to acknowledge signals andto take action should the engineer fail to operatehis train in accordance with a signal indication.
Under the usual passenger train operations,the conductor and assistant conductor would be inthe passenger cars performing the assigned dutiesof collecting tickets, responding to passengerqueries, preparing for station stops, andmonitoring radio transmissions from the engineer.However, because of the light passenger loadduring this segment of the trip on MARC train286, the conductor and assistant conductordivided the work; one of them stayed in the cabcontrol compartment with the engineer while theother performed the requisite duties and thenreturned to the compartment. The CSXToperating rules expect that when crewmembersare in the operating compartment of thecontrolling locomotive (in this case, the cab
control car) they have the responsibility forcalling and acknowledging the wayside signalsand for taking the appropriate action should theengineer fail to operate the train according to thesignal indication.
The assistant conductor is believed to havebeen in the cab control car with the engineer forthe entire trip from Brunswick to the accidentsite, based on his normal habit of riding this tripsegment in the cab control car and, also, thestatements of surviving passengers. Conse-quently, he would have been in the cab controlcar when the Kensington signal was visible andwhen the engineer acknowledged it. Whetherthe assistant conductor also acknowledged thesignal is unknown; however, he did have theopportunity both to see and to hear what it was.
In addition, the conductor is believed to havebeen in the cab control car during much of thetrip, and may have been in the cab control carwhen the Kensington signal was visible and theengineer acknowledged it. However, the witnesstestimony is unclear on his exact activities atthat time. He could have been walking to orwaiting at the doors between cab control car7752 and car 7720 to board passengers atKensington station and not have seen the signal;although, he might have heard the engineer ac-knowledge the signal over his portable radio.Still, the conductor may have neither seen thesignal indication nor heard the engineer ac-knowledge it and, therefore, not have knownwhat the signal indication was. However, had henot heard the signal being called, the conductorwas in the cab control compartment with theengineer later according to witnesses and couldhave then questioned the signal indication thatshould have been acknowledged.
Accident survivors who had been in the pas-senger compartment of the cab control car placedall three crewmembers in the cab control car afterthe Kensington station stop. Clearly, both crew-members were in the cab control compartmentwith the engineer for most, if not all, of the tripsegment after signal 1124-2 from Kensington toGeorgetown Junction. Additionally, the survivors
51
did not note any unusual activity from the crew-members until the accident appeared imminent.MARC train 286 had not been operated in accor-dance with the APPROACH indication duringthat segment of the trip. The other crewmembersapparently had not taken action when the engi-neer did not operate the train according to thelast signal indication.
The explanation for their behavior may befound in the normal operating practice. Rule 34,to acknowledge the signals, is a required activityfor crewmembers that occurs many times eachday. As such, the assistant conductor may haveacknowledged the signal when the engineer did,and the conductor may also have done so. Oncethe signal is called and acknowledged, however,only the engineer has a physical activity to per-form. The other crewmembers have only a pas-sive oversight responsibility to do somethingunder rule 34C should the engineer fail in hisresponsibility. Since adhering to signal indica-tions is fundamental to safe rail operations, suchfailures are rare. Personnel practices furthersanctify signal compliance, as engineers whofail to adhere to signals are usually disciplinedor fired. The other crewmembers have the re-sponsibility to be vigilant for a rare event and totake corrective action. Actually taking a correc-tive action, such as using the emergency brakehandle, is even more rare and not likely to occurin a crewmember’s entire career. Thus, becausethe engineer has an active task to perform withevery signal and does so routinely, and the othercrewmembers have only a passive responsibilitythat is rarely, if ever, exercised, it is not difficultto conceive that the crewmembers in this acci-dent deferred to the engineer and did not moni-tor his compliance with the signal indicationduring the interval from the Kensington stationstop to the emergency brake application.
It is possible that one crewmember or bothtold the engineer that he was not complying withthe APPROACH signal. However, they mayhave deferred to the engineer’s recollection ofthe signal aspect as CLEAR in the absence of anindependent source of information, such as cabsignals, to advise them otherwise. This is per-
haps an unlikely possibility because a prudentengineer would probably accept the most re-strictive signal aspect and act accordingly. Nev-ertheless, no actions were taken by the conduc-tor or assistant conductor to counteract the ac-tions of the engineer as required by rule 34C.This occurred even though the conductor andassistant conductor were competent, experi-enced personnel, which calls into questionwhether it is reasonable to rely on the vigilanceof a person to compensate for the error of an-other person in the same circumstances. TheSafety Board concludes, therefore, that neitherthe conductor nor the assistant conductor while inthe cab control compartment appeared to haveeffectively monitored the engineer’s operation ofMARC train 286 and taken action to ensure thesafety of the train. The Safety Board believes thatthe CSXT should inform all operating traincrewmembers of the circumstances of this acci-dent and emphasize the crew responsibility whilein the operating compartment for the safe opera-tion of the train.
Traincrew Voice Recording
The 35-year experience using cockpit voicerecordings (CVRs) to assist in determining thecause of commercial aviation accidents hasshown that evidence about the operating com-munications among crewmembers is frequentlyimportant in accident investigations. Initiallyinstalled on aircraft carrying 40 or more passen-gers, CVRs only recorded the last 30 minutes ofcockpit and radio conversation before an acci-dent. In the early 1990’s a change was made toestablish design standards that increase the re-cording length from 30 minutes to 2 hours. Al-though required for all newly manufacturedtransport aircraft registered in Europe after1995, a 2-hour CVR is not required for any air-craft operating in the United States. However,the Safety Board is of the opinion that a 2-hourCVR should be required on all newly manufac-tured commercial aircraft registered in theUnited States.
52
Over the years the CVR recordings havebeen a key tool in documenting the circum-stances leading up to an accident and a valuableassistance to the Safety Board in determiningthe probable cause of aviation accidents. TheCVR has been most useful in the type of acci-dent that has not been caused by mechanicalfailure onboard an aircraft. The CVR recordinghas shown to be an almost necessary tool indocumenting the operational decisions or mis-takes of the crew that lead up to the accident.The Safety Board had repeatedly recommendedto the Federal Aviation Administration (FAA),using this argument, that the FAA require CVRson smaller commuter aircraft, and the FAA hasnow required that CVRs be installed on allmulti-engine turbine-powered aircraft having aseating configuration of six or more and forwhich two pilots are required by certification oroperating rules.
The Safety Board understands that appropri-ate protections of the privacy of such communi-cations have been established in aviation andcould also be adopted by the railroad industry.The communications that occurred from theGarret Park station up to the collision atGeorgetown Junction would have been ex-tremely valuable to this investigation. In par-ticular, knowing the signal aspect acknowledgedby the MARC train 286 engineer at signal 1124-2 would have facilitated investigative activities.Although Safety Board investigators conductedan exhaustive attempt to reconstruct traincrewactivities, they could not document the MARCtrain 286 engineer's acknowledgment of theKensington signal or the communications, ifany, among crewmembers as the train ap-proached Georgetown Junction.
Current locomotive event recorders havegreat utility but only provide mechanical re-sponse data, which cannot answer some ques-tions asked in an accident investigation aboutthe traincrew’s knowledge and actions. Thevoice recordings maintained by the CSXT op-erations center included no communicationsbetween trains or among MARC train 286crewmembers. Had a voice recording fromMARC train 286 existed, the signal aspect ac-
knowledged and the communications in the lastfew minutes before the collision would havebeen available to this investigation. A few yearsago the FRA contemplated issuing a rule re-quiring voice recorders in locomotive compart-ments but rejected the idea because it did notconsider them as a necessary safety measure.The FRA could have included traincrew voicerecording requirements in the 1993 regulationsfor locomotive event recorders as part of theminimum parameters to be recorded. The SafetyBoard, consequently, concludes that had theFRA required the recording of the train crew-members’ voice communications, the essentialdetails about the circumstances of this accidentcould have been provided. Therefore, the SafetyBoard believes that the FRA should amend 49CFR Part 229 to require the recording of traincrewmembers’ voice communications for exclu-sive use in accident investigations and with ap-propriate limitations on the public release ofsuch recordings.
CSXT Signal System ModificationOversight
The CSXT and MARC had operationalreasons to modify the Brunswick Line signalsystem: improve passenger safety and freighttrain operations by changing the method thatCSXT dispatched and monitored trains, upgradethe system capacity to operate more trains withincreased peak and midday service, increase theMARC labor and equipment productivity, andreduce the CSXT operating costs. Identifiableimprovements, such as total trains, traincrewuse, cost savings, and CTC operations, could bequantified and measured; however, the signalsystem modifications did not address the overallsafety of the signal system for traincrew use.The adequacy of the system safety could only havebeen addressed with a total system review thatincluded a human factors analysis of such issues ashuman information processing capabilities.
A total system review examining human ca-pabilities and limitations might have resulted inthe installation of a redundant system, such as anautomatic train control system or automatic cabsignals (ACS), which would have produced an
53
audible indication to alert the engineer and avisible reference to identify when the cab signaldisplay changed to a more restrictive aspect.The MTA application for FTA funding for theproject indicated that the funding approval forthe project would later address advanced traincontrol systems. Yet, when MTA and MARCofficials were queried about this subject, theyhad no current plans.
The Safety Board investigators questionedthe removal of signal 100, which, located east ofthe Kensington station, had been the last signalon track 2 for eastbound trains traveling towardsGeorgetown Junction. As a result of the modifi-cation and respacing of the signals, the last sig-nal on track 2 for trains traveling towardsGeorgetown Junction was now signal 1124-2,which is west of Kensington station and about1.25 miles west of the former signal 100 loca-tion. The spacing of signals is FRA regulatedunder 49 CFR Part 236.24, which requires sig-nals to be adequately spaced to provide properdistances for reducing speeds or stopping by useof other than an emergency brake applicationbefore reaching the point where reduced speedor stopping is required. The FRA determinedduring routine signal inspections that theBrunswick Line signal system complied with theregulation for the spacing of roadway signals.
The CSXT signal system modification, how-ever, did not adequately account for the operat-ing characteristics of passenger trains stoppingat the Kensington station, as evidenced by thisaccident. The removal of signal 100 relocated asource of vital information for passenger trainengineers stopping at the Kensington stationfrom a position close to where it would be actedupon to a position farther away. In this case, thephysical distance the signal was moved was notthe critical element; but rather, the relocationcreated the potential for other information andtasks to intervene and interfere with the reten-tion of the signal indication, thus permitting it tobe forgotten before it was required to be used.Of course, the potential for interference to leadto an operational error did not necessarily exist
to the same degree for all trains, and the rightset of circumstances had to exist. Nevertheless,the potential for an operational error to occur asa result of the relocated signal could have beenforeseen. Had the design of the signal systemreceived input from knowledgeable human fac-tors specialists, the potential pitfall of the relo-cation could have been addressed and redun-dancy provided for an engineer forgetting a sig-nal. Such redundancy could have been accom-plished through a delayed-in-block rule change,as the FRA did with EO 20 issued following thisaccident; with a repeater signal; with cab sig-nals; or with a positive train separation (PTS)control system, as the Safety Board has longadvocated.
The FRA EO 20 contained specificinformation for all railroads on every push-pulloperation without benefit of cab signal,automatic train stop, or automatic train controland whose speed exceeds 30 mph. The new rulerequired that all trains stopped or delayed in ablock immediately preceding interlockings andcontrolled points must reduce their speed inaccordance with applicable operating rules, but,in no case, may speed exceed 40 mph. Inaddition, EO 20 added another measure thatrequired appropriate signs be installed at eachaffected signal and at the departure end ofstations. The Safety Board concludes that hadthe FRA and the FTA required the CSXT toperform a total signal system review of theproposed signal changes that included a humanfactors analysis within a comprehensive failuremodes and effects analyses, this accident mayhave been prevented.
The information obtained during the SafetyBoard investigation of this accident, includingits public hearing, raised questions about theoversight by Federal agencies of federallyfunded transit projects and, specifically, theFTA grant application and approval process.Although Federal funding provided most of thefunds for the design and installation of the CTCsystem on the CSXT Brunswick and CamdenLine signal modification, the Federal Govern-ment apparently did not perform an in-depth
54
analysis or evaluation of this project from asafety standpoint primarily because the applyingagency self-certified that it had the technicalcapacity to undertake the project. Furthermore,the project justification statement indicated thatsafety would be enhanced by the installation ofthis upgraded signal system; however, the avail-able evidence indicates that the project was un-dertaken for economic reasons and that a totalsystem safety review, including a human factorsanalysis of the upgraded signal system, was notconsidered at either the State or Federal Gov-ernment level.
The Safety Board recognizes that the FTAmay not have the necessary expertise in all proj-ect areas for which transit agencies seek fund-ing. In this particular instance, the FTA indi-cated that it did not have any in-house signalexpertise with which to judge the safety benefitsof the proposed signal modifications. However,the FTA could have either requested assistancefrom other modal administrations that have thetechnical expertise or required a total systemsafety analysis by an independent contractor, asa condition for grant approval. Subsequent tothe Silver Spring accident, the FTA has re-quested that the FRA become more involved inreviewing operational plans and safety proposalsfor grantees, including commuter railroads. Al-though the FTA request for FRA involvement iscommendable, it does not formally address whatthe FTA specifically expects from the FRA.
The Safety Board concludes that Federalfunds granted for the signal modifications on theCSXT Brunswick Line to accommodate an in-crease in the number of MARC trains did notensure that the safety of the public was ade-quately addressed. Therefore, the Safety Boardbelieves that the FRA should require compre-hensive failure modes and effects analyses, in-cluding a human factors analysis, for all signalsystem modifications and that the FTA shouldrevise the grant application process to requirethe same such analyses be provided for all fed-erally funded transit projects that are directlyrelated to the transport of passengers.
Federal Commuter Rail OperationsOversight
The Safety Board requested the FRA to re-spond to its concerns about commuter railroadoperations within FRA regulatory authority,which included a 5-year reportable accidentlisting for MARC and the Virginia Railway Ex-press (VRE) and a 5-year accident, inspection,and defect history for commuter railroad opera-tions. The FRA responded that 17 commuter railoperations are under its regulatory authority ofwhich six, including VRE, are operated by Am-trak; two, including MARC operations, are op-erated by Amtrak and CSXT together; seven areoperated by the commuter railroad, itself; twoare operated by a freight railroad; and one isoperated by a contractor. The FRA reported thatit does not have a 5-year accident, inspection,defect, or violation record database for individ-ual commuter railroads and that before 1996,any data regarding accidents/incidents was re-ported as it related to the railroad operating thecommuter service. This data could not be read-ily separated because it was commingled withthe freight and intercity rail reporting data. TheFRA advised that it is currently working withthose railroads to ensure that commuter rail op-erations data will be reported separately in thefuture. (See appendix K.)
The Safety Board, consequently, concludesthat without a separate collection database spe-cific to commuter rail inspections and acci-dent/incidents, it is difficult for the Federal Rail-road Administration to evaluate its own effec-tiveness of inspections and to identify problem-atic trends. Records should include reporting oninspections and accident/incidents to determinethe effectiveness of its own as well as that ofparticipating state organizations and to identify,and to take corrective action in, those areas in-adequate to public safety. Therefore, the SafetyBoard believes that the FRA should develop andmaintain separate identifiable data records forcommuter and intercity rail passenger opera-tions.
55
Positive Train Separation Control Systems
The Safety Board has long advocated a PTScontrol system and since 1970�� has issuedsafety recommendations concerning train colli-sion prevention. (See appendix L.) A PTS con-trol system can prevent trains from colliding byautomatically interceding in the operation of atrain when an engineer does not comply with therequirements of the signal indication.
Following its investigation of a head-on col-lision on the Burlington Northern Railroad nearLedger, Montana,�� the Safety Board issued inJuly 1993 the following safety recommendationto the FRA:
In conjunction with the Association ofAmerican Railroads and the Railway Pro-gress Institute, establish a firm timetablethat includes at a minimum, dates for fi-nal development of required advancedtrain control system hardware, dates foran implementation of a fully developedadvanced train control system, and acommitment to a date for having the ad-vanced train control system ready for in-stallation on the general railroad system.(R-93-12)
The Safety Board additionally issued a similarrecommendation to the AAR and to the RailwayProgress Institute (RPI), respectively, in SafetyRecommendations R-93-13 and -15.
The Safety Board classified Safety Recom-mendation R-93-12 “Open--Acceptable Re-sponse” after the FRA took the measure to seekthe “final system definition, migration path, andtimetable” for a PTS control system by Decem-
33Railroad Accident Report--Head-on Collisionbetween Penn Central Trains N-48 and N-49 atDarien, Connecticut, August 20, 1969 (NTSB/RAR-70/03).
34Railroad Accident Report--Head-on Collisionbetween Burlington Northern Freight Trains 602 and603 near Ledger, Montana, on August 29, 1991(NTSB/RAR-93/01).
ber 1994. The Safety Board also classifiedSafety Recommendations R-93-13 and -15, re-spectively, to the AAR and the RPI “Open-Acceptable Response” based on their responses.Because of the AAR and the RPI participationand support in current test projects of the UnionPacific/Burlington Northern Santa Fe Railroad(UP/BNSF) in the Pacific Northwest and ofAmtrak tests in Michigan, the Safety Board re-classifies R-93-13 and -15, respectively, as“Closed--Superseded” and “Closed--AcceptableAlternate Action.” Neither the AAR nor the RPIis in a position to either establish timetables orto implement a PTS control system for the rail-road industry. The FRA and the railroad indus-try share the responsibility for the developmentand implementation of a PTS control system.Under its regulatory authority, the FRA can or-der a railroad to install a PTS control system,and the FRA can issue emergency orders, as itdid following this accident, where an unsafecondition or practice causes an emergencysituation involving a hazardous death or injury.
The Safety Board, however, believes that theAAR can assist the railroad industry with thedevelopment of PTS control systems through acontinuing review of nonrailroad technologyand assess its adaptability to railroad communi-cation-based control systems. In addition, theSafety Board believes that the AAR can assistthe railroad industry with the development ofPTS control systems by acting as a clearing-house for information on the status and resultsof pilot projects and by disseminating that in-formation to the railroad industry and the Fed-eral and participating State transportation or-ganizations. Finally, the Safety Board believesthat the AAR can assist the railroad industrywith the installation and operation of PTS con-trol systems by maintaining industry standardsto ensure interoperability of equipment and anopen architecture for train control systems.
Citing the recent train accidents in Secaucusand Silver Spring, the FRA stated in EO 20 thatit had a particular concern for operations thatinvolve lead cars carrying passengers on track
56
segments that have neither cab signals nor anautomatic train stop or automatic train control.EO 20 required that commuter and intercity pas-senger railroads modify services operatingabove 30 mph that lack cab signals or automatictrain stop or automatic train control protectionsand that permit passengers to occupy the leadcar, either cab control cars in the forward posi-tion push-pull mode or self-propelled locomo-tives with passenger seating (MU [multiple-unit]locomotives). The FRA also exercised its over-sight responsibility for operating rules by con-cluding that certain current conditions and prac-tices on commuter and intercity passenger rail-roads posed an imminent and unacceptablethreat to public and employee safety. The EO 20specifically addressed the delayed-in-block ruleand the exclusion granted to passenger trainsunder certain conditions. The FRA recognizedthat unacceptable threats to public and employeesafety exist where protection is not provided bycab signal or automatic train stop or automatictrain control systems. The FRA addressed sev-eral public safety issues that required immediateattention in EO 20; however, it did not addressthe other critical risks posed by reliance on crewalertness in complying with operating rules.
The Safety Board has investigated numeroustrain collisions in which the probable cause orcontributing cause was the inattention of thetraincrew to wayside signals. In its investigationof the head-on collision of two freight trainsnear Kelso, Washington,�� the Safety Board at-tempted to determine again why one traincrewdid not comply with the signal indication of anintermediate signal. The Safety Board reportedits concerns about a systemic safety issue: theadequacy of passive wayside signals to reliablycapture traincrews’ attention when competingsources of attention are present, and it urged therailroad industry to recognize that human vigi-lance has limits and that wayside signals do not
35Railroad Accident Report--Head-on Collisionand Derailment of Burlington Northern Freight Trainwith Union Pacific Freight Train, Kelso, Washington,on November 11, 1993 (NTSB/RAR-94/02).
ensure safe train operations. The FRA EO 20,notice no. 2, concluded that “certain currentconditions and practices on commuter and inter-city passenger railroads pose an imminent andunacceptable threat to public and employeesafety. Of greatest concern are push-pull andMU operations lacking the protection providedby cab signal, automatic train stop, or automatictrain control systems.” After its investigation ofthe Thedford, Nebraska,�� accident, the SafetyBoard stated that had a PTS control system beenin place it could have detected that the engineerwas not responding appropriately to the signalindications and could have slowed and stoppedthe train, thus preventing the collision.
The FRA newly required rule for calling sig-nals has basically the same instructions as theexisting CSXT operating rule 34. The signalcalling that the FRA requires likely occurred inthe Silver Spring accident, and at least onecrewmember was in the cab control car with theengineer and is believed also to have seen thesignal. The accident still happened because sucha rule does not adequately compensate for hu-man capabilities and crew interaction, as notedearlier in this report. Therefore, the SafetyBoard concludes that the FRA reliance on theneed for increased vigilance of wayside signalsand special actions in operating rules, such asthe crew communication rule of EO 20, does notadequately safeguard the public.
The full development of a PTS control sys-tem is still underway; however, current technol-ogy exists for cab signal, automatic train stop, orautomatic train control systems. At the SafetyBoard public hearing for this accident also, acommuter railroad official from METRA�� ex-plained an alternate system that METRA has
36Railroad Accident Report--Collision and De-railment Involving Three Burlington NorthernFreight Trains near Thedford, Nebraska on June 8,1994 (NTSB/RAR-95/03).
37Chicago Commuter Rail Service Board, for-merly Northeast Illinois Railroad Corporation, oper-ates its own suburban services and controls commutertrains, operated by freight railroads, such asUP/BNSF, in the Chicago area.
57
designed, in cooperation with the operating car-rier and labor unions, to caution the engineer ofhis last signal indication, which METRA be-lieves may not require FRA approval. TheSafety Board concurs with the FRA EO 20, no-tice no. 1, that:
Since most train collisions on the railroadresult from human factors, the most ef-fective preventive measure is a highly ef-fective train control system. Cab signalsystems serve an important safety purposebecause they provide a constant displayof the governing signal indication. Thisprovides a corrective measure should anengineer fail to note, forget, or misread arestrictive wayside signal indication.Even greater security is provided by atrain control system capable of interven-ing should the engineer fail to observesignals and operating rules for whateverreason….Such systems are referred to asautomatic train control or automatic trainstop.
Although all MARC locomotives and cab con-trol cars have cab signal equipment, the Bruns-wick Line was not equipped with a train controlsystem to implement those devices. A train con-trol system, which would have been recognizedby the MARC cab control car cab signalequipment, could have provided the engineerwith a visual reminder of the 1124-2 signal as-pect, required him to acknowledge and complywith the APPROACH signal indication, or en-forced the requirements of the signal indicationby stopping the train. The Safety Board con-cludes that had a train control system that couldutilize the cab signal equipment on the MARCcab control car been a part of the signal systemon the Brunswick Line, this accident may nothave occurred. Therefore, the Safety Board be-lieves that the FRA should require, in the in-terim of a PTS control system being available,the installation of cab signals, automatic trainstop, automatic train control, or other similarredundant systems for all trains where commuterand intercity passenger railroads operate.In addition, the Safety Board believes that the
FTA should cooperate with the FRA for requir-ing, in the interim of a positive train separationcontrol system being available, the installation ofcab signals, automatic train stop, automatic traincontrol, or other similar redundant systems forall trains where commuter and intercity passen-ger railroads operate.
The 1987 MDOT grant application to theFTA for the CSXT signal system modificationon the Brunswick Line stated that the improve-ments envisioned in this program provide thefoundation for the next “generation” of traincontrol systems: ATCS. However, neither theFTA nor the MTA followed up on the MDOTpursuit of this technology. In the MTA/MARCgrant application to the FTA, the future installa-tion of an ATCS, such as cab signals, was partof the justification for awarding the grant for thesignal modifications being proposed. At the timeof this accident, no advanced train control hadbeen installed.
Since the collision at Georgetown Junction,MARC has undertaken a project, for which theMTA has hired a consultant and providedfunding, to develop and evaluate an intermittentcab signaling system (ICSS) that features bothcivil speed enforcement and positive train stoptechnology. In addition, the supplier of trackcircuit equipment is estimating the cost for up-grading the equipment to continuously inductiveACS that will be compatible with the automatictrain control equipment currently installed onMARC locomotives and cab control cars. TheCSXT is also involved in the project because itswayside signal equipment and locomotives willbe directly affected by the installation of anychanges proposed to the current signal system.
The Safety Board is encouraged by the ef-forts of the MTA/MARC project to develop andevaluate an ICSS; however, ICSS should onlybe an interim solution until a PTS control sys-tem can be fully implemented. A PTS controlsystem is a major step for the railroad industryto provide a redundant system where an unac-ceptable threat to public and employee safetyexists. Pending the FRA issuance of regulations
58
that require a PTS control system installation,railroads remain responsible for a PTS controlsystem development and installation. Conse-quently, the Safety Board believes that theCSXT should develop and install a PTS controlsystem on its track segments that have com-muter and intercity passenger trains. The SafetyBoard also believes that the FTA and the MTAshould cooperate with the CSXT in the devel-opment and installation of a PTS control systemwhere MARC equipment operates on CSXTtracks.
In the FRA July 1994 report to Congress re-garding advanced train control, the FRAplanned to begin a 2-year “Corridor Risk As-sessment” study in 1995 to identify and evaluatewhich conventional rail corridors would beprime candidates for advanced train control im-plementation. The study was to contain a geo-graphic information system (GIS) platform toprovide the analysis, which would include PTScontrol system preventable accidents plotted onthe GIS. The initial results of the study are to bepresented to the FRA Railroad Safety AdvisoryCommittee in June 1997 for review and furtheranalysis.
An Amtrak project in Michigan is complete,but another in Illinois has not started. TheMichigan project tested an incremental traincontrol system on the Amtrak line betweenKalamazoo and Green Beach.�� The Michiganterritory is similar to the Brunswick Line, in thatboth passenger and freight trains operate.
The Silver Spring accident is the latest in aseries of collisions that could have been pre-vented had a PTS control system been in place.A PTS control system could have detected that
38The DOT, Michigan DOT, Amtrak, and HarmonIndustries began a project in 1995 to implement ahigh-speed positive train control system on a 71-mileportion of the Detroit, Michigan, and Chicago, Illi-nois, corridor. Testing was done on the 20-mile seg-ment between Dowagiac and Niles, Michigan. Thesystem works by allowing grade crossings, signals,and locomotives to communicate with each other us-ing data radio.
the MARC train 286 engineer was not respond-ing appropriately to signal indications and thenslowed and stopped the train, thus, preventingthe collision. The Safety Board concludes that afully implemented PTS control system wouldhave prevented this accident by recognizing thatMARC train 286 was not being operated withinallowable parameters, based on other authorizedtrain operations, and would have stopped thetrain before it could enter into the unauthorizedtrack area. The Safety Board therefore believesthat the FRA should require the implementationof PTS control systems for all trains wherecommuter and intercity passenger railroads op-erate. The Safety Board also believes that theFTA should cooperate with the FRA for requir-ing the implementation of positive train separa-tion control systems for all trains where com-muter and intercity passenger railroads operate.In addition, the Safety Board reiterates SafetyRecommendations R-87-16 and R-93-12 to theFRA. The Safety Board intends to closelymonitor the progress made on this importantissue and to continue discussing the benefits of aPTS control system in all reports of accidentsthat such a control system could have prevented.
Passenger Car Safety Standards
During the investigation of this accident, theSafety Board identified problems with emer-gency egress from the passenger cars that con-tributed to the number of fatalities. The SafetyBoard concluded that the emergency egress ofpassengers was impeded because the passengercars lacked readily accessible and identifiablequick-release mechanisms for the exterior doors,removable windows or kick panels in the sidedoors, and adequate emergency instruction sign-age. The Safety Board further concluded that theabsence of comprehensive Federal passenger carsafety standards resulted in the inadequateemergency egress conditions. Therefore, theSafety Board issued four urgent safety recom-mendations on March 12, 1996. One was issuedto the FRA; three were issued to the MTA, asfollows:
59
Install removable windows or kick panelsfor emergency exits in interior and exte-rior passageway doors. (R-96-4)
Install an easily accessible interior emer-gency quick-release mechanism adjacentto all exterior doors. (R-96-5)
Install retroreflective signage on car inte-riors and exteriors at emergency exits thatcontains easily understood instructionsand clearly marks all emergency exits(doors and windows). (R-96-6)
The MTA responded in a November 7, 1996,letter that it has: 1) entered into an engineeringcontract for designing removable windows foremergency exits in car end and side doors, butthe work will not be completed until the end of1997; 2) completed the design of a quick-releasemechanism adjacent to all exterior doors withthe mechanism installation to be completed inMarch 1997�� (see figure 14); and 3) installedluminescent interior as well as enhanced retrore-flective exterior signage on all MARC cars as ofAugust 30, 1996. Based on this MTA response,the Safety Board classifies the urgent SafetyRecommendations R-96-5 and -6 “Closed--Acceptable Action” and “Closed--Exceeds Rec-ommended Action,” respectively. The SafetyBoard also recognizes the complexity of redes-igning and installing removable windows foremergency exits in car end and side doors andthe MTA efforts to complete the work aspromptly as possible. Because this work cannotbe completed within the 1-year timeframe re-quired for an urgent safety recommendation, theSafety Board had reclassified Safety Recom-mendation R-96-4 to regular safety recommen-dation status “Open--Acceptable Action.”
Additionally, as a result of this accident in-vestigation, the following urgent safety recom-mendation was issued to the FRA:
39MARC informed the Safety Board on March 3,1997, that the installation was complete on February28, 1997.
Inspect all commuter rail equipment todetermine whether it has: (1) easily ac-cessible interior emergency quick-releasemechanisms adjacent to exterior pas-sageway doors; (2) removable windowsor kick panels in interior and exterior pas-sageway doors; and, (3) prominently dis-played retro-reflective signage markingall interior and exterior emergency exits.If any commuter equipment lacks one ormore of these features, take appropriateemergency measures to ensure correctiveaction until these measures are incorpo-rated into minimum passenger car safetystandards. (R-96-7)
After reviewing the FRA response of June 6,1996, the Safety Board classified Safety Rec-ommendation R-96-7 “Open--Acceptable Ac-tion” in March 1997. The FRA had stated in thatresponse that it had inspected 1,250 pieces ofrailroad equipment on 16 commuter organiza-tions, that any variation in the inspection resultswas indicative of equipment age and of differ-ence by commuter agencies, and that emergencyquick-release mechanisms were typically acces-sible to passengers; although, many were notadjacent to exterior passageway doors. In addi-tion, the FRA had stated that marking emer-gency exits inside cars with luminescent ratherthan retroreflective materials may be better. TheFRA, however, has taken no immediate action tomake corrections. The Safety Board acknowl-edges the FRA efforts concerning its urgentsafety recommendation and recognizes the com-plexities of developing practical solutionswithin the requested 1-year period; therefore,the Safety Board classifies Safety Recommen-dation R-96-7 “Closed--Superseded” and is sub-stituting it with three separate safety recommen-dations for longer-term action. Consequently,the Safety Board believes that the FRA should:
60
Figure 14--Photograph showing quick-release mechanisms on MARC cars.
Require all passenger cars to have easilyaccessible interior emergency quick-release mechanisms adjacent to exteriorpassageway doors and take appropriateemergency measures to ensure correctiveaction until these measures are incorpo-rated into minimum passenger car safetystandards.
Require all passenger cars to have eitherremovable windows, kick panels, or othersuitable means for emergency exitingthrough the interior and exterior passage-way doors where the door could impedepassengers exiting in an emergency andtake appropriate emergency measures toensure corrective action until these meas-ures are incorporated into minimum pas-senger car safety standards.
Issue interim standards for the use of lu-minescent or retroreflective material orboth to mark all interior and exterioremergency exits in all passenger cars assoon as possible and incorporate intominimum passenger car safety standards.
The Safety Board first began addressing pas-senger car safety with the FRA after an accidentat the Botanical Gardens station in New York,New York, in January 1975.�� The Safety Boardcontinued its efforts on passenger car crash-worthiness following Amtrak accidents inMarch 1980 near Glacier Park, Montana, and in
40Railroad Accident Report--Collision of TwoPenn Central Commuter Trains at Botanical GardenStation, New York, New York, January 2, 1975(NTSB/RAR-75/08).
61
June 1982 at Gibson, California. �� The SafetyBoard issued the following safety recommenda-tions, respectively, to the FRA:
Promulgate regulations to establish mini-mum standards for the interior of com-muter cars so that adequate crash injuryprotection and emergency equipment willbe provided passengers. (R-75-38)
Promulgate regulations to establish mini-mum safety standards for the inspectionand maintenance of railroad passengercars. (R-80-31)
Expedite the development of passengercar safety standards which were mandatedby Congress in October 1980 (reiteratedJanuary 14, 1983) including in the stan-dards: (A) Criteria for the location andintensity of emergency lights within thecars to ensure adequate visibility for es-cape from smoke filled cars; (B) Re-quirements for emergency evacuationplans, for training of personnel for emer-gencies, and for emergency systems, suchas emergency exits and doors, smoke de-tector systems, etc., specifying the num-bers, type, location, and markings; (C)Acceptable levels of flame spread rate,smoke emissions, and toxic fumes for in-terior materials, and (D) Requirements forthe installation of a sprinkler system towhich water can be supplied by emer-gency equipment through externallymounted standard standpipes. (R-83-76)
Safety Recommendation R-75-38 was classifiedin April 1984 “Closed--Superseded” by SafetyRecommendation R-84-46, which was subse-quently classified “Closed--No Longer Applica-ble” in August 1991, when Amtrak informed the
41Railroad Accident Reports--Derailment of Am-trak Train No. 7 the Empire Builder on BurlingtonNorthern Track, Glacier Park, Montana, March 14,1980 (NTSB/RAR-80/06) and Fire Onboard AmtrakPassenger Train No. 11, Coast Starlight, Gibson,California, June 23, 1982 (NTSB/RAR-83/03).
Safety Board of research in progress whicheliminated the need for FRA studies. SafetyRecommendations R-80-31 and R-83-76 wereboth classified “Closed--Unacceptable Action,”based on FRA responses, in January 1986. InJune 1996, the FRA finally reacted to the con-cerns raised by the Safety Board, the GeneralAccounting Office, and others and issued a no-tice of initiation for rulemaking on rail passen-ger equipment safety standards to comply withthe Federal Railroad Safety Authorization Actof 1994.
Current FRA regulations for passenger carsafety standards are inadequate. They do notaddress passenger car safety standards for self-contained emergency lighting; inspection, re-moval, and maintenance of emergency windows;exterior emergency door releases; interiorflammability and smoke standards; and struc-tural crashworthiness. The Safety Board is en-couraged by the current FRA position in devel-oping rulemaking and expects that the passengercar safety standards will not only address thesafety of passengers in newly built passengercars but also in existing passenger cars. TheFRA indicated that the group working on thedevelopment of passenger car safety standardscompleted their work in December 1996, and anotice of proposed rulemaking (NPRM) is ex-pected by the end of 1997.
In addition during the investigation of thisaccident, the Safety Board identified severalareas of safety deficiencies that should be ad-dressed by passenger car safety standards forimproved passenger safety. The identified areasare the power source of emergency lighting, thedifficulty in removing emergency windows, themissing or inaccessible exterior emergency doorrelease handles, the failure of interior materialsto meet flammability and smoke standards, andthe structural crashworthiness of cab controlcars.
Emergency Lighting-- The two passengersin coach car 7720 stated that no emergencylighting was available after the accident. Onepassenger's injuries and the other's loss of eye
62
glasses compounded the reported difficulty inmoving in the darkness. Additionally, the tiltedposition of the car contributed to their disorien-tation and hindered their mobility. Postaccidentinspection of the car revealed that batteries sup-plying power to the emergency lighting system,located below the floor level, had been damagedin the derailment. After Safety Board investiga-tors alerted MARC to the failure of the emer-gency lighting system, MARC projected plansto complete by late 1997 the fleetwide installa-tion of battery pack ballasts, designed to providepower should head-end or car battery power belost, in selected fluorescent lighting fixtures.
The MARC passenger cars are not unique inthe location of the battery supplying power tothe emergency lighting system being below thecar floor, which makes them susceptible in aderailment to damage from contact with the cartrucks, the rails, or the ground. The installationof battery pack ballasts or other self-containedindependent power sources in the car interiorwould provide reliable power to emergencylighting in derailments when the batteries underthe car are most likely to be damaged. TheSafety Board concludes that a need exists forFederal standards requiring passenger cars beequipped with reliable emergency lighting fix-tures with a self-contained independent powersource when the main power supply has beendisrupted to ensure passengers can egress safely.Therefore, the Safety Board believes that theFRA should require all passenger cars to containreliable emergency lighting fixtures that areeach fitted with a self-contained independentpower source and incorporate the requirementsinto minimum passenger car safety standards.
Emergency Windows-- One passenger incoach car 7720 stated that he exerted much ef-fort to open an emergency window in that car.Postaccident inspection of the car emergencywindows revealed that much effort was alsonecessary to open two other emergency win-dows in the same car because the window lubri-cant had hardened. The Safety Board deter-mined that no periodic preventive maintenancehad been performed on the emergency windows
of the MARC passenger equipment and thatnone had been required.
The FRA EO 20 requires all passenger rail-roads to inspect emergency window exits forproper operation, marking, and instructions aspart of routine equipment maintenance. TheFRA informed the Safety Board in October1996 that changes to the regulatory text of thedraft NPRM on Passenger Train EmergencyPreparedness were being made because the in-spection cycle requirements of EO 20 werevague. The February 24, 1997, FRA-issuedNPRM proposes that a railroad test a represen-tative sample of emergency window exits on itscars a least once every 180 days to verify properoperation and that it repair a defective unit be-fore returning the car to service. As a result ofthe Safety Board urgent recommendations andthe FRA EO 20, MARC has performed func-tional tests on all existing emergency exit win-dows and initiated a program to install addi-tional emergency exit windows in all their pas-senger equipment. The Safety Board, however,is concerned that the FRA EO 20, pertaining tothe inspection cycle, does not require a pre-scribed test cycle and that until the proposedrulemaking becomes effective, emergency win-dows on other passenger rail carriers may not beadequately inspected. The Safety Board con-cludes that prescribed inspection and mainte-nance test cycles are needed to ensure reliableoperation of emergency windows in all long-distance and commuter rail passenger cars.Therefore, the Safety Board believes, pendingthe issuance of the final rule on passenger carsafety standards, that the FRA should promptlyprovide a prescribed inspection and maintenancetest cycle to ensure the proper operation of allemergency exit windows as well as provide thatthe 180-day inspection and maintenance testcycle is prescribed in the final rule.
Exterior Emergency Door Release--The MARC II passenger cars have provisionsfor a T-handle, which is used to disconnect theinterior release mechanism allowing the exteriorside doors to be opened from the outside, ateach of the four exterior doors. During postacci-
63
dent inspection of cab control car 7752, investi-gators found that the exterior emergency doorrelease T-handles were missing. In addition,when MARC last performed general servicingof cab control car 7752, the T-handles werenoted as missing. MARC mechanical officersexplained that because of the difficulty in re-ceiving replacements from the original equip-ment supplier, they needed time to manufactureand install substitute T-handles, but MARC didnot restrict the car from being placed back intopassenger service.
The postaccident inspection of the passengercoach 7720, immediately behind cab control car7752, revealed that the exterior emergency doorrelease T-handles on the right side were buriedin ballast when the car derailed and were, there-fore, inaccessible to emergency responders. TheT-handles were designed to be below the floorline of the car and about 3 inches inboard. Theoriginal design plans for the MARC II passengercars showed the emergency door release T-handles above the bottom of the car body and ina glass-covered recessed pocket adjacent to eachvestibule door. However, the T-handles had notbeen installed there because of an approved de-sign change during car building. Emergency re-sponders may have been able to quickly locatethe T-handles and gain access to passengercoach interior had the handles been locatedwhere planned.
The Safety Board concludes that the exterioremergency door release T-handles for theMARC cars were not either in place or accessi-ble to firefighters because no requirements fortheir maintenance or accessibility exist. There-fore, the Safety Board believes that the FRAshould require that all exterior emergency doorrelease mechanisms on passenger cars be func-tional before a passenger car is placed in reve-nue service, that the emergency door releasemechanism be placed in a readily accessible po-sition and marked for easy identification inemergencies and derailments, and that these re-quirements be incorporated into minimum pas-senger car safety standards.
Interior Materials-- The analysis of firedebris indicated that diesel fuel from thebreached fuel tank of Amtrak unit 255 sprayedinto the breached opening of the MARC 286 cabcontrol car. Positive residues were found onsome passenger seats and on a sheetmetal panelnear the opening. The analysis revealed that thediesel fuel played a significant role in the earlyfire growth within the car, and within 3 to 5minutes after the collision based on witness tes-timony, flashover developed in the cab controlcar with the fire accelerating over the breachedarea because of that sprayed diesel fuel. Haddiesel fuel not sprayed into the cab control car,the fire likely would not have spread as quicklyas it did. For the passengers to quickly exit thecar became even more critical because of therapid growth of fire. Except for those passengerswho died from blunt trauma injuries, others mayhave survived the accident, albeit with thermalinjuries, had proper and immediate egress fromthe car been available. The Safety Board con-cludes that the catastrophic rupture of the Am-trak unit 255 fuel tank in the collision with theMARC cab control car 7752 released fuel,which sprayed into the interior of the cab con-trol car, and resulted in the fire and at least 8 ofthe 11 fatalities.
The Safety Board recognizes that the materi-als taken from an exemplar MARC car may nothave been identical to the materials that wereinstalled on cab control car 7752 and that otherfactors, such as wear, can affect the perform-ance of the car materials; nevertheless, the mate-rials taken from the exemplar car were signifi-cantly similar to the materials in the accidentcar. Some of the interior materials from the ex-emplar MARC car failed current flammabilityand smoke emissions testing criteria, and thematerials in the cab control car 7752 also mostlikely would have failed. Had the materials metcurrent performance criteria, however, the out-come would not have been any different becauseof the presence of diesel fuel as an ignitionsource. The fire would have spread quicklywhether the interior materials of the MARCpassenger cars had met current performancecriteria regarding flammability and smoke emis-
64
sions characteristics; still, the Safety Board isconcerned that the interior materials in theMARC passenger cars did not meet existing per-formance criteria for flammability and smokeemissions characteristics.
Performance criteria is based on guidelinesdeveloped by the FRA and the FTA, and thepassenger commuter rail industry could refer-ence them with or without modification whenordering new equipment. The Safety Board con-cludes that because other commuter passengercars may also have interior materials that maynot meet specified performance criteria forflammability and smoke emission characteris-tics, the safety of passengers in those cars couldbe at risk. Therefore, the Safety Board believesthat the FRA should require that a comprehen-sive inspection of all commuter passenger carsbe performed to independently verify that theinterior materials in these cars meet the ex-pected performance requirements for flamma-bility and smoke emissions characteristics.
The current FRA information on the flam-mability and smoke emissions characteristicsand the testing of commuter and intercity railvehicle materials is based on guidelines thathave not changed significantly in the past 30years. Consequently, the Safety Board con-cludes that the Federal guidelines on the flam-mability and smoke emissions characteristicsand the testing of interior materials do not pro-vide for the integrated use of passenger car inte-rior materials and, as a result, are not useful inpredicting the safety of the interior environmentof a passenger car in a fire. Therefore, theSafety Board believes that the DOT should re-view the testing protocols within the variousmodal administrations regarding the flammabil-ity and the smoke emissions characteristics ofinterior materials and coordinate the develop-ment and implementation of standards for mate-rial performance and testing with the FRA andthe FTA.
Cab Control Car Structural Crashwor-thiness-- The left side of cab control car 7752was ripped open in the collision from its front to
midsection. (See figure 15.) The cab control carwas equipped, as required by FRA standards,with structural components serving as collisionposts at each end doorway; however, the mag-nitude of the impact and collision forces weremore than the collision posts could resist. TheSafety Board first addressed structural crash-worthiness of passenger cab control cars after itsinvestigation of the August 1981 head-on colli-sion between a Boston & Maine Corporationfreight train and a Massachusetts Bay Trans-portation Authority commuter train in Beverly,Massachusetts,��
�urging the FRA to:
Expedite implementation of Safety Boardrecommendations to study structural pro-tection for occupants of control cars andlocomotive operating compartments. (R-82-34)
Safety Recommendation R-82-34 was classified“Closed-Acceptable” based on the FRA re-sponse that it had completed the study, Analysisof Locomotive Cabs.
The Safety Board investigation of the Janu-ary 1993 collision between two MU-locomotivepassenger trains in Gary, Indiana,�� resulted in asafety recommendation asking the FRA to ad-dress the collision protection afforded by thecorner post structures of passenger cars:
In cooperation with the Federal TransitAdministration and the American PublicTransit Association, study the feasibilityof providing car body corner post struc-
��Railroad Accident Report--Head-on Collision ofBoston & Maine Corporation Extra 1731 East andMassachusetts Bay Transportation Authority TrainNo. 570, Beverly, Massachusetts, August 11, 1981(NTSB/RAR-82/01).
��Railroad Accident Report--Collision BetweenNorthern Indiana Commuter Transportation DistrictEastbound Train 7 and Westbound Train 12 nearGary, Indiana, on January 18, 1993(NTSB/RAR-93/03).
66
tures on all self-propelled passenger carsand control cab locomotives to afford oc-cupant protection during corner collision.If feasible, amend the locomotive safetystandards accordingly. (R-93-24)
Similar Safety Recommendations R-93-25 and -26, respectively, were issued to the FTA and theAmerican Public Transit Association. SinceApril 1994, Safety Recommendations R-93-24through -26 have been classified “Open--Acceptable Action.”
The car body design of the trains involved inthe Gary accident was a lightweight monocoquestructure�� and almost identical to the design ofthe MARC II cars involved in this accident. Thedamage to the cars in the Gary collision was notas extensive as to cab control car 7752 in theSilver Spring collision, and no fire was in-volved. The Silver Spring accident serves tounderscore the vulnerability of certain passengercar and cab control car designs in major colli-sions and the importance of addressing occupantcollision protection in car body design.
The FRA, which is principally responsiblefor oversight of the passenger rail transportationindustry, initiated a formal study of the crash-worthiness issue in response to the Rail SafetyEnforcement and Review Act. That study re-sulted in the comprehensive Locomotive Crash-worthiness and Cab Working Conditions-Reportto Congress in September 1996, which includedan engineering evaluation of cab control cars.The findings identified several crashworthinesssafety features, such as implementation ofstronger collision posts and full height cornerposts, that merited further FRA action in coop-eration with private industry.
��Metal structure in which the covering absorbs alarge part of the stresses to which the body is subjected.
Emergency Preparedness
The CSXT, as well as a local resident, con-tacted the MCFRS 911 dispatcher about thecollision at 5:41 p.m. Within 1 minute, the firstfire engine company was notified, and it arrivedon scene within 5 minutes of notification. AllMCFRS commands had been established by6:15 p.m. Therefore, the Safety Board concludesthat even though the MCFRS personnel re-sponded promptly to the emergency, they coulddo nothing to save any of the accident victimsbecause passenger coach cab control car 7752was already completely engulfed in flames whenthe first firefighter arrived on scene. The inves-tigation identified problems with the MCFRSpreparedness for railroad passenger train acci-dents; the CSXT, the MTA/MARC, and theMCFRS contingency planning; and the interac-tion between these three agencies.
The MCFRS activated the fire annex sectionof the Montgomery County Emergency Man-agement Agency (MCEMA) disaster plan; how-ever, the fire annex section did not provide forinterchange with the CSXT or MARC, whichwas evident when interactions were not consis-tently maintained between the supervisors ordispatchers or both of the MCFRS, the CSXT,and MARC. The MCFRS personnel were notreceptive to the CSXT offers for assistance. Thefire annex section did not provide for railroadrepresentatives to respond to strategic locationsto contribute their expertise and assistance.
The CSXT attempted to assist in evacuatingpassengers by moving trains closer to the acci-dent site and only complicated the emergencyresponse efforts. Although this attempt was con-sistent with the CSXT practice for controllingtrain movements and providing assistance, theMCFRS was not familiar with railroad opera-tions because no procedures had been coordi-nated between the CSXT and the MCFRS. As aresult, the accident scene was evacuated about6:30 p.m. by emergency responders who fearedthat another train was entering the accident site.
67
The MCEMA disaster plan did not contain pro-cedures for responding to railroad passengertrain accidents. The Safety Board concludes thatthe MCEMA disaster plan lacked procedures forresponding to railroad passenger train accidents,such as simulating the accident response withcoordinated management, which could haveemphasized the importance of being familiarwith passenger cars and of coordinating activi-ties between the MCFRS, the MTA/MARC, andthe CSXT. To be familiar with the means ofemergency egress from a passenger train and tocoordinate activities with the railroads are ex-tremely essential procedures needed for emer-gency response. Therefore, the Safety Boardbelieves that the MCEMA should develop com-prehensive procedures for responding to railroadpassenger train accidents and include these pro-cedures in its disaster plan.
Although MARC knew of the train accident,it was not immediately aware that the collisioninvolved one of its commuter trains. The MARCchief transportation officer said that for 45 min-utes after the accident, the MARC operationscenter had been in a state of confusion. In addi-tion, the CSXT operations center in Jacksonvillewas unable to contact the MARC operationcenter by telephone from 5:45 until 6 p.m. be-cause the telephone line was busy and no back-up communications was available. Some MARCofficials who had the Amtrak alphanumericpagers were only made aware of the MARCtrain involvement in the collision by reading themessages that were meant for Amtrak personnel.Since the accident, a dedicated emergencyphone line between the MARC operations cen-ter and the CSXT operations center has beeninstalled, and the CSXT has been provided withthe alphanumeric pager numbers of MARC su-pervisors.
The confusion between the CSXT and theMCFRS at the accident site and the untimelynotifications between the CSXT and MARC ofthe collision resulted because neither the CSXTnor MARC had a formal emergency manage-ment plan available that contained proceduresfor dispatchers and traincrews to notify emer-
gency responders of train movements near anaccident site. When the AU dispatcher author-ized the engineer of CSXT train Q401 to movehis locomotive closer to assist in evacuatingpassengers, the MCFRS IC was not advised thatthe train would be approaching the accident site.The movement of trains toward an accident areashould have been addressed by the CSXT andMCRFS dispatchers. The Safety Board con-cludes that the confusion during the initialemergency response resulted because the CSXTand MARC lacked a formal emergency man-agement plan to follow. The implementation ofan emergency management plan that addressedcommunications and training would have elimi-nated the confusion between the CSXT andMARC. Therefore, the Safety Board believesthat CSXT should develop and implement aformal emergency management plan that con-tains procedures specific to employee responsi-bilities and interaction with emergency responseagencies and other transportation entities. TheSafety Board also believes that theMTA/MARC should develop an emergency planthat will provide a detailed description of emer-gency response procedures as well as a protocolto coordinate activities with the emergency re-sponse organizations and other transportationentities when an accident occurs.
The CSXT traincrews of MARC passengertrains had minimal guidance, compared with theAmtrak manual of on-train instructions for con-ductors and assistant conductors, to properlyperform passenger train functions. Since theCSXT operation in 1985 of the MARC passen-ger service, the CSXT had not maintained acomprehensive passenger program that wouldprovide guidance to traincrews for passengertrain functions. The CSXT produced PassengerService Bulletins as needed, but it offered littleguidance on responding to passenger trainemergencies. The CSXT passenger traincrewsreported that they had not received any emer-gency training in passenger train operations andin passenger responsibility in emergencies. TheSafety Board concludes that the CSXT person-nel operating MARC passenger trains are notadequately trained to understand and therefore
68
execute their responsibilities for passengers inemergencies. The CSXT and MARC havebeen working since 1993 to complete the Pas-senger Conductor’s Manual, which was unfin-ished at the time of the accident. A review ofthis unfinished manual shows that it is muchless comprehensive than the Amtrak manual ofon-train instructions for conductors and assistantconductors. The Safety Board believes that theCSXT and MARC should develop and imple-ment, in cooperation, a complete training agendafor all CSXT passenger traincrews that providesexperience in the correct use of emergencyequipment, in emergency communications pro-cedures, and in passenger evacuation and assis-tance in an emergency and also includes thedistribution of a comprehensive employee guid-ance manual.
Since the accident MARC has informed theSafety Board that it, in cooperation with Amtrakand the CSXT, has developed video materialsfor training emergency responders and the Am-trak and CSXT traincrews who operate MARCcommuter trains.�� However, such passivetraining may not be as effective as training thatrequires traincrews to actively participate andpractice what is being demonstrated. To achievethe protocols and procedures described in anyemergency management plan, emergency drillsshould be performed in conjunction with localemergency management agencies and with therailroad to reinforce training, to test communi-cations, and to determine whether proceduralchanges are needed. Therefore, the Safety Boardbelieves that the CSXT and the MTA/MARC, incooperation with the emergency managementagencies of Baltimore County, of the city ofBaltimore, of the Metropolitan WashingtonCouncil of Governments, and of Jefferson andBerkeley Counties in West Virginia, shouldconduct periodic disaster drills to assess theiremergency management plans, to reinforce and
45Before the accident, Amtrak was providingtraining for MARC traincrews on the Penn Line andhands-on as well as audio-visual training foremergency responders in areas near the Penn Line.
evaluate their emergency training, and to test thecommunications with the organizations.
The Safety Board has found in other accidentinvestigations��� that emergency responders canbe hampered in their search and rescue, as wellas extrication, efforts because of the lack ofemergency plans, inaccessible terrain along rail-road property, accounting for number of passen-gers, difficult extrication caused by rescue toolsinadequate for the construction of and materialsin passenger equipment, coordination and com-munication with railroads and emergency re-sponders, and infrequent disaster drills foremergency responders. With the exception ofthe nationwide Amtrak service, only certain lo-calities have commuter passenger railroad serv-ice. Many rural emergency response agenciesmay never have had the opportunity to respondto a rail disaster involving fire. With the adventof high-speed passenger rail service, increaseddevelopment of commuter railroad systems, andwidespread rail transportation of hazardousmaterials, however, the likelihood of morecommunities being involved in a railroad emer-gency of this type has increased. The SafetyBoard therefore concludes that the lack of ap-propriate training for emergency responders inthe areas of emergency planning, coordinationand communications, rescue methods, inacces-sible terrain along railroad property, familiaritywith railroad equipment, and disaster drillsmay become a recurrent problem for other
46Railroad Accident Reports--Rear-end Collision ofAmtrak Passenger Train 94, the Colonial, and Con-solidated Rail Corp. Freight Train ENS-121 on theNortheast Corridor, Chase, Maryland, January 4,1987 (NTSB/RAR-88/01); Head-on Collision of Na-tional Railroad Passenger Corporation (Amtrak) Pas-senger Trains Nos. 151 and 168, Astoria, Queens, NewYork, New York, July 23, 1984 (NTSB/RAR-85/09);Derailment and Subsequent Collision of Amtrak Train82 with Rail Cars on Dupont Siding of CSX Trans-portation Inc. at Lugoff, South Carolina, on July 31,1991 (NTSB/RAR-93/02); and Derailment of AmtrakTrain No. 2 on the CSXT Big Bayou Canot Bridge nearMobile, Alabama, on September 22, 1993(NTSB/RAR-94/01).
69
emergency response organizations unless a na-tional effort is made to address emergency re-sponse training for railroad accidents. Conse-quently, the Safety Board believes that the Fed-eral Emergency Management Agency shouldinclude in its training at the U.S. Fire Admini-stration National Fire Academy a curriculumthat addresses the needs of State and localemergency management agencies to respond to amajor railroad accident and that familiarizesemergency response organizations with railroadequipment and appropriate rescue methods forrailroad accidents.
In December 1993, the FRA published itsreport Recommended Emergency PreparednessGuidelines for Passenger Trains,�� in which itstated that it
has recognized the need for intercity andcommuter passenger train system opera-tors to engage in careful advance planningto respond effectively to emergencies.This advance planning should addressemergency response procedures, trainingof system operating and other emergencyresponse organization personnel, and pro-vision and use of emergency equipment.
The report also noted that "This document isadvisory in nature. The recommended guidelinescontained herein do not have the force and ef-fect of law or regulation."
On February 24, 1997, the FRA publishedthe NPRM Passenger Train Emergency Prepar-edness, which proposes requiring minimumFederal safety standards for the preparation,adoption, and implementation of emergencypreparedness plans by railroads connected withthe operation of passenger trains, including
��Prepared for the FRA by the DOT, Research andSpecial Programs Administration, John A. VolpeNational Transportation Systems Center, Final ReportNo. DOT/FRA/ORD-93/24 November 1993.
freight railroads hosting the operations of railpassenger service. The proposed rule also re-quires each affected railroad to instruct its em-ployees on the provisions of the plan. TheSafety Board encourages the FRA to move for-ward in the rulemaking process for the devel-opment of comprehensive emergency manage-ment plans that affect the safety and well beingof passengers, railroad employees, and thecommunity in an emergency.
Reverser Use During EmergencyBraking
The Safety Board stopping distance tests in-dicated that had the MARC train 286 engineernot used the reverser, thereby retaining dynamicbraking until impact, MARC train 286 wouldhave impacted Amtrak train 29 at a speed ofabout 34 mph as opposed to the actual impactspeed of about 38 mph. The additional decel-eration of MARC train 286 would have resultedin an additional 0.3 seconds of elapsed time be-fore impact, which in turn would have resultedin Amtrak train 29 moving approximately 14feet farther into the crossover before impact.Thus, with MARC train 286 operating at thespeed of 66 mph and going into emergencybraking 1,407 feet before impact, a collision wasinevitable regardless of the reverser use by theMARC train 286 engineer.
Despite the CSXT instructions that the rever-ser only has limited utility and its intentions thatthe reverser be used only under specific condi-tions, the use of the reverser having a retardingeffect is implied in the instructions. The MARCtrain 286 engineer may have drawn from thatimplication and used the reverser about 1,000feet into his emergency braking sequence out ofdesperation when he realized emergency brakingwould not prevent the impending collision.Nevertheless, because the reverser use elimi-nated the additional braking provided by thelocomotive dynamic brakes, the Safety Boardconcludes that the MARC train 286 engineer’suse of the reverser during the emergency brakeapplication resulted in a marginally increasedstopping distance for MARC train 286.
70
Therefore, the Safety Board believes that theCSXT should inform its engineers, and theAAR, the American Short Line Railroad Asso-ciation, the Brotherhood of Locomotive Engi-neers, the UTU, the International Brotherhoodof Teamsters, and the American Public TransitAssociation should inform its membership, ofthe circumstances of this accident and cautionthem not to use the reverser during emergencybrake applications for those trains on which theuse of reverser will eliminate the dynamicbraking, thus increasing stopping distance.
Computer-Aided Recordkeeping
Federal regulations under 49 CFR Part228.17 require that each carrier keep records oftrain movements that are made under the direc-tion and control of a dispatcher who uses a tele-phone, radio, or any other electrical or mechani-cal device to dispatch, report, transmit, receive,or deliver orders pertaining to train movements.These records are to contain such important in-formation as the identification of the timetablein effect; the train location and the date; the dis-patcher identification and duty time; the weatherconditions at 6-hour intervals; the identificationof the trains and the traincrews and their dutytimes; the station names and the office designa-tions; the distances between stations; the direc-tion of train travel and the times of arrival, de-parting, and passing all reporting stations; andany unusual events affecting the movement oftrains and the identity of affected trains. Beforethe computerization of dispatching offices, theserecords were recorded manually and retained onfile for review by railroad operating officers,Federal regulators, and accident investigators.
The CSXT operates its CTC system with acomputed-aided dispatching (CAD) system, asdo many major railroads in the country. TheCAD generates “computerized train sheets” (acomputer file that can print out the requestedinformation), which was designed by the CSXTfor the recordkeeping of each train (about 3,000per month). The Safety Board examined trainsheets covering about 90 days, with particularattention given to the accident trains, and found
that the train sheets were incomplete in so far asrecording the information delineated under 49CFR Part 228.17. For example, the computer-generated train sheet for MARC train 286 didnot show the train activity after its departurefrom Point of Rocks, when it was required tomake a reverse movement to crossover and goaround a disabled freight train; the weather thatday; and the accident occurrence. The SafetyBoard investigation revealed that the train sheetsbeing maintained by CSXT lacked the informa-tion required by FRA regulation and, therefore,were of little value in determining unusualevents the day of the accident. The Safety Boardconcludes that the FRA has not addressed theuse of CAD system records to provide informa-tion for the identification and evaluation of po-tential safety-related trends for corrective action.The FRA must consider, when alternative meth-ods of recordkeeping are employed, that the cur-rent minimum data requirements are critical togood event recording and that alternative shouldbe comparable or better. Consequently, theSafety Board believes that the FRA should up-date 49 CFR Part 228.17, Train Dispatcher’sRecord of Train Movements, to include thesame parameters for electronic recordkeeping ofthe dispatcher’s record of train movements.
Locomotive Fuel Tank Crashworthiness
In 1992 the Safety Board conducted compre-hensive Safety Study (NTSB/SS-92/04), Loco-motive Fuel Tank Integrity, which addressed,among other issues, the potential in railroad ac-cidents for diesel fuel fires to fatally injure thetrapped crewmembers. The report primarilycovered the risk in freight locomotive opera-tions, although a passenger train collision casestudy was cited. However, the risk of collisiondamage is effectively the same for both types ofoperations because fuel tanks for freight loco-motives are configured essentially the same aspassenger locomotives, such as the one involvedin this accident. Specific safety issues discussedin the study were the adequacy of current fueltank design, factors that affect fuel tank design,
71
and the sufficiency of research to improve fueltank integrity or fuel containment. The studyfindings concluded that the evaluation of theextent of locomotive fuel tank damage and spillson an annual basis is difficult because of thelimited data available. Furthermore, althoughthe railroad industry has explored changes infuel tank design, no evidence was found that theindustry has performed a systematic engineeringanalysis to determine the feasibility of providingbetter crash protection for fuel tank systems.
As a result of this safety study, the SafetyBoard issued Safety Recommendation R-92-10to the FRA:
Conduct, in conjunction with the Asso-ciation of American Railroads, the Gen-eral Electric Company, and the Electro-Motive Division of General Motors, re-search to determine if the locomotive fueltank can be improved to withstand forcesencountered in the more severe locomo-tive derailment accidents or if fuel con-tainment can be improved to reduce therate of fuel leakage and fuel ignition.Consideration should be given to crash orsimulated testing and evaluation of recentand proposed design modifications to thelocomotive fuel tank, including increasingthe structural strength of end and sidewall plates, raising the tank higher abovethe rail, and using internal tank bladdersand foam inserts.
The Safety Board also issued the similar SafetyRecommendations R-92-14, -16, and -17, re-spectively, to the AAR, the General ElectricCompany (GE), and the Electro-Motive Divi-sion of General Motors (EMD). The SafetyBoard classified Safety Recommendations R-92-10, -16, and -17 to the FRA, GE, and the EMD,respectively, “Open--Acceptable Response” af-ter noting the industry efforts to address the is-sue. However, the crash or simulation testingand evaluation of recent and proposed designmodifications has not yet been addressed.Therefore, the Safety Board reiterates to theFRA, GE, and the EMD, respectively, Safety
Recommendations R-92-10, -16, and -17. In ad-dition the Safety Board has classified SafetyRecommendation R-92-14 “Closed--AcceptableAction” to the AAR after it reported the adop-tion of RP-506 for all locomotive units built af-ter September 1, 1995. RP-506 addressed struc-tural and puncture resistance properties of thefuel tank to reduce the risk of spillage underderailment or minor collision conditions.
Also as a result of the safety study, theSafety Board issued Safety Recommendation R-92-11 to the FRA:
Establish, if warranted, minimum per-formance standards for locomotive fueltanks based on the research called for inSafety Recommendation R-92-10.
Because the Safety Board believed that AARRP-506 was an adequate direction to take and, atthe time, realized no need for Federal regula-tions, Safety Recommendation R-92-11 wasclassified “Closed--No Longer Applicable.”
The locomotive in the Silver Spring accident,Amtrak unit ATK 255, was built in 1977 andwas under the FRA standard in 49 CFR Part229.71, stipulating the minimum clearance fromthe top of the rail. The current FRA regulationsdo not address the design, size, locations, or per-formance of locomotive fuel tanks, and do notrequire a regularly scheduled or periodic in-spection of fuel tanks to ensure no safety haz-ards are present. In this collision, substantialcompressive forces bearing on the lead truck ofAmtrak unit ATK 255 caused it to be displacedinto the front plate of the fuel tank, located im-mediately behind the lead truck. The front of thefuel tank compressed inward substantially dam-aging the fuel tank. The left side of the fuel tanksubsequently was ruptured catastrophically bythe raking action of the body bolster of theMARC cab control car 7752 as both trains con-tinued to move forward. The fuel tank providedlittle impact resistance to the compressive forcesof the lead truck and the shearing action of theprojecting body bolster of cab control car 7752.The released fuel sprayed into the exposed inte-
72
rior of cab control car 7752, and with multiplesources of ignition, the fuel quickly ignited andengulfed the car.
The Safety Board has previously expressedits concern to the FRA about locomotives builtbefore the effective date of September 1, 1995,noting that many of these locomotives will re-main in service for several years, and this acci-dent only reinforces that concern about this is-sue. The Safety Board will continue to addressthe crashworthiness issue in future accident in-vestigations and to monitor the FRA progressfor improving locomotive fuel tank crashwor-thiness.
CSXT and MARC Operating Agreement
The FTA grant process and FRA/MDLI op-erations oversight suggest areas of needed im-provement for the involved government agen-cies; however, the CSXT is inevitably the mostimportant agent in safe, or unsafe, operations. Inthis regard, the indemnification arrangement forCSXT/MARC service is a significant, but notlikely uncommon, issue: the actual operator ofthe service in question was well-insulatedagainst damages that might arise from both itsemployees and the passengers on the line. Riskof liability from injury to its employees wascompensated by a substantial surcharge to theCSXT billings for labor expense, and the Stateof Maryland assumed the risk, through self-insurance and the purchase of commercial cov-erage, for the first $150 million of passenger-related liabilities.
The exposure to liability may induce cau-tious behavior; however, in the absence of thisexposure, it is not so obvious that one will bereadily able to identify the precautions that were
forsworn when the financial risks were elimi-nated, particularly in a regime, like railroading,where so much of the activity is required andinspected under regulatory authority. Neverthe-less, the manner in which the protection ofMARC passengers was approached, and subse-quently explained, almost assuredly reflects adisconnect between activity and responsibility.The CSXT manager for these commuter servicesindicated that at the request of MARC, theCSXT crews wore MARC uniforms in the pas-senger cars. The CSXT crews were not trainedby MARC or by the CSXT in emergency proce-dures. Indeed, no specific carrier-initiated safetyassessment of the emergency passenger equip-ment or its operation had been undertaken, andno emergency preparedness plan had been de-veloped. In postaccident questioning, CSXTpersonnel allowed that passenger safety, as op-posed to train handling safety, was an issue forMARC. The CSXT was, of course, aware thatMARC had done little in this regard and had noon-train personnel. But when asked whether thematter had ever been discussed between the twoprincipals, the CSXT indicated that MARC hadnever broached the subject. It appears that theCSXT had not raised the issue as well. TheSafety Board concludes that the CSXT/MARCsystem lacked comprehensive safety oversight toensure the safety of the commuting public. TheSafety Board is concerned by this apparenthands-off approach and is convinced that it can-not be divorced from the environment of dimin-ished liability in which it arose. Therefore, theSafety Board believes that the Governor and theGeneral Assembly of the State of Marylandshould instruct and empower an appropriateState agency to provide continual, effective, andindependent safety oversight of all aspects ofthe MARC operations.
73
FINDINGS
Conclusions
1. The train equipment and the track functionedas designed and the AU dispatcher conducted thedispatching activities properly. Neither the threeMARC train 286 crewmembers nor the two Am-trak train 29 locomotive crewmembers were im-paired by alcohol or drugs. All train crewmem-bers were in good health, had no evidence of fa-tigue, and were experienced in and qualified fortheir duties.
2. The weather conditions did not impair theability of the MARC train 286 crewmembers todistinguish the indication of Kensington signal1124-2.
3. The signal system functioned as designed.
4. The MARC train 286 engineer apparently for-got the signal aspect, which required him to beprepared to stop at Georgetown Junction, due tointerference caused by various events, includingperforming an unscheduled station stop, that oc-curred between the presentation of theAPPROACH aspect at signal 1124-2 and theSTOP signal at Georgetown Junction.
5. Neither the conductor nor the assistant con-ductor while in the cab control compartment ap-peared to have effectively monitored the engi-neer’s operation of MARC train 286 and takenaction to ensure the safety of the train.
6. Had the Federal Railroad Administration re-quired the recording of the train crewmembers’voice communications, the essential detailsabout the circumstances of this accident couldhave been provided.
7. Had the Federal Railroad Administration andthe Federal Transit Administration required theCSX Transportation Inc. to perform a total sig-nal system review of the proposed signalchanges that included a human factors analysiswithin a comprehensive failure modes and ef-fects analyses, this accident may have been pre-vented.
8. Federal funds granted for the signal modifi-cations on the CSXT Brunswick Line to ac-commodate an increase in the number of Mary-land Rail Commuter trains did not ensure thatthe safety of the public was adequately ad-dressed.
9. Without a separate collection database spe-cific to commuter rail inspections and acci-dent/incidents, it is difficult for the Federal Rail-road Administration to evaluate its own effec-tiveness of inspections and to identify problem-atic trends.
10. The Federal Railroad Administration reli-ance on the need for increased vigilance of way-side signals and special actions in operatingrules, such as the crew communication rule ofemergency order 20, does not adequately safe-guard the public.
11. Had a train control system that could utilizethe cab signal equipment on the Maryland RailCommuter cab control car been a part of thesignal system on the Brunswick Line, this acci-dent may not have occurred.
12. A fully implemented positive train separa-tion control system would have prevented thisaccident by recognizing that MARC train 286was not being operated within allowable pa-rameters, based on other authorized train opera-tions, and would have stopped the train before itcould enter into the unauthorized track area.
13. The emergency egress of passengers wasimpeded because the passenger cars lackedreadily accessible and identifiable quick-releasemechanisms for the exterior doors, removablewindows or kick panels in the side doors, andadequate emergency instruction signage.
14. The absence of comprehensive Federal pas-senger car safety standards resulted in the in-adequate emergency egress conditions.
74
15. A need exists for Federal standards requiringpassenger cars be equipped with reliable emer-gency lighting fixtures with a self-containedindependent power source when the main powersupply has been disrupted to ensure passengerscan safely egress.
16. Prescribed inspection and maintenance testcycles are needed to ensure reliable operation ofemergency windows in all long-distance andcommuter rail passenger cars.
17. The exterior emergency door release T-handles for the MARC cars were not either inplace or accessible to firefighters because norequirements for their maintenance or accessi-bility exist.
18. The catastrophic rupture of the Amtrak unit255 fuel tank in the collision with the MARCcab control car 7752 released fuel, whichsprayed into the interior of the cab control car,and resulted in the fire and at least 8 of the 11fatalities.
19. Because other commuter passenger cars mayalso have interior materials that may not meetspecified performance criteria for flammabilityand smoke emission characteristics, the safety ofpassengers in those cars could be at risk.
20. The Federal guidelines on the flammabilityand smoke emissions characteristics and thetesting of interior materials do not provide forthe integrated use of passenger car interior ma-terials and, as a result, are not useful in predict-ing the safety of the interior environment of apassenger car in a fire.
21. Even though the Montgomery County Fireand Rescue Services personnel respondedpromptly to the emergency, they could do noth-ing to save any of the accident victims becausepassenger coach cab control car 7752 was al-ready completely engulfed in flames when thefirst firefighter arrived on scene.
22. The Montgomery County Emergency Man-agement Agency disaster plan lacked proceduresfor responding to railroad passenger train acci-dents, such as simulating the accident responsewith coordinated management, which couldhave emphasized the importance of being fa-miliar with passenger cars and of coordinatingactivities between the Montgomery County Fireand Rescue Services, the Maryland Rail Com-muter, and the CSX Transportation Inc.
23. The confusion during the initial emergencyresponse resulted because the CSX Transporta-tion Inc. and Maryland Rail Commuter lacked aformal emergency management plan to follow.
24. The CSX Transportation Inc. personnel oper-ating Maryland Rail Commuter passenger trainsare not adequately trained to understand and,therefore, execute their responsibilities for pas-sengers in emergencies.
25. The lack of appropriate training for emer-gency responders in the areas of emergencyplanning, coordination and communications,rescue methods, inaccessible terrain along rail-road property, familiarity with railroad equip-ment, and disaster drills may become a recurrentproblem for other emergency response organi-zations unless a national effort is made to ad-dress emergency response training for railroadaccidents.
26. The MARC train 286 engineer’s use of thereverser during the emergency brake applicationresulted in a marginally increased stopping dis-tance for MARC train 286.
27. The Federal Railroad Administration has notaddressed the use of computer-aided dispatchingsystem records to provide information for theidentification and evaluation of potential safety-related trends for corrective action.
28. The CSX Transportation Inc./Maryland RailCommuter system lacked comprehensive safetyoversight to ensure the safety of the commutingpublic.
75
Probable Cause
The National Transportation Safety Boarddetermines that the probable cause of this acci-dent was the apparent failure of the engineer andthe traincrew because of multiple distractions tooperate MARC train 286 according to signalindications and the failure of the Federal Rail-road Administration, the Federal Transit Ad-ministration, the Maryland Mass Transit Ad-ministration, and the CSX Transportation Inc. toensure that a comprehensive human factorsanalysis for the Brunswick Line signal modifi-cations was conducted to identify potential
sources of human error and to provide a redun-dant safety system that could compensate forhuman error.
Contributing to the accident was the lack ofcomprehensive safety oversight on the CSXTransportation Inc./Maryland Rail Commutersystem to ensure the safety of the commutingpublic. Contributing to the severity of the acci-dent and the loss of life was the lack of appro-priate regulations to ensure adequate emergencyegress features on the railroad passenger cars.
RECOMMENDATIONS
As a result of its investigation, the NationalTransportation Safety Board makes the followingrecommendations:
--to the Federal Railroad Administration:
Amend 49 Code of Federal RegulationsPart 229 to require the recording of traincrewmembers’ voice communications forexclusive use in accident investigationsand with appropriate limitations on thepublic release of such recordings. (R-97-9)
Require comprehensive failure modes andeffects analyses, including a human fac-tors analysis, for all signal system modifi-cations. (R-97-10)
Develop and maintain separate identifi-able data records for commuter and inter-city rail passenger operations. (R-97-11)
Require, in the interim of a positive trainseparation control system being available,the installation of cab signals, automatictrain stop, automatic train control, orother similar redundant systems for alltrains where commuter and intercity pas-senger railroads operate. (R-97-12)
Require the implementation of positivetrain separation control systems for alltrains where commuter and intercity pas-senger railroads operate. (R-97-13)
Require all passenger cars to have easilyaccessible interior emergency quick-release mechanisms adjacent to exteriorpassageway doors and take appropriateemergency measures to ensure correctiveaction until these measures are incorpo-rated into minimum passenger car safetystandards. (R-97-14)
Require all passenger cars to have eitherremovable windows, kick panels, or othersuitable means for emergency exitingthrough the interior and exterior passage-way doors where the door could impedepassengers exiting in an emergency andtake appropriate emergency measures toensure corrective action until these meas-ures are incorporated into minimum pas-senger car safety standards. (R-97-15)
Issue interim standards for the use of lu-minescent or retroreflective material orboth to mark all interior and exterioremergency exits in all passenger cars assoon as possible and incorporate the in-
75
terim standards into minimum passengercar safety standards. (R-97-16)
Require all passenger cars to contain reli-able emergency lighting fixtures that areeach fitted with a self-contained inde-pendent power source and incorporate therequirements into minimum passenger carsafety standards. (R-97-17)
Provide promptly a prescribed inspectionand maintenance test cycle to ensure theproper operation of all emergency exitwindows as well as provide that the 180-day inspection and maintenance test cycleis prescribed in the final rule. (R-97-18)
Require that all exterior emergency doorrelease mechanisms on passenger cars befunctional before a passenger car isplaced in revenue service, that the emer-gency door release mechanism be placedin a readily accessible position andmarked for easy identification in emer-gencies and derailments, and that theserequirements be incorporated into mini-mum passenger car safety standards. (R-97-19)
Require that a comprehensive inspection ofall commuter passenger cars be performedto independently verify that the interiormaterials in these cars meet the expectedperformance requirements for flammabil-ity and smoke emissions characteristics.(R-97-20)
Update 49 Code of Federal RegulationsPart 228.17, Train Dispatcher’s Record ofTrain Movements, to include the same pa-rameters for electronic recordkeeping ofthe dispatcher’s record of train move-ments. (R-97-21)
--to the Federal Transit Administration:
Revise the grant application process torequire a comprehensive failure modesand effects analyses, including a hu-man factors analysis, be provided forall federally funded transit projects
that are directly related to the transportof passengers. (R-97-22)
Cooperate with the Federal RailroadAdministration for requiring, in theinterim of a positive train separationcontrol system being available, the in-stallation of cab signals, automatictrain stop, automatic train control, orother similar redundant systems for alltrains where commuter and intercitypassenger railroads operate. (R-97-23)
Cooperate with the Federal RailroadAdministration for requiring the im-plementation of positive train separa-tion control systems for all trainswhere commuter and intercity passen-ger railroads operate. (R-97-24)
Cooperate with CSX TransportationInc. in the development and installa-tion of a positive train separation con-trol system where Maryland RailCommuter equipment operates onCSX Transportation Inc. tracks. (R-97-25)
--to the CSX Transportation Inc.:
Develop and install a positive trainseparation control system on tracksegments that have commuter and in-tercity passenger trains. (R-97-26)
Develop and implement a formalemergency management plan thatcontains procedures specific to em-ployee responsibilities and interactionwith emergency response agencies andother transportation entities. (R-97-27)
Develop and implement, in coopera-tion with Maryland Mass Transit Ad-ministration/Maryland Rail Com-muter, a complete training agenda forall CSX Transportation Inc. passengertraincrews that provides experience inthe correct use of emergency equip-
76
ment, in emergency communicationsprocedures, and in passenger evacua-tion and assistance in an emergencyand also includes the distribution of acomprehensive employee guidancemanual. (R-97-28)
Conduct, in cooperation with Mary-land Mass Transit Administra-tion/Maryland Rail Commuter, theBaltimore County Emergency Man-agement Agency, the City of Balti-more Emergency ManagementAgency, the Metropolitan WashingtonCouncil of Governments, the JeffersonCounty Commissioners, and the Ber-keley County Commissioners, periodicdisaster drills to assess their emer-gency management plans, to reinforceand evaluate their emergency training,and to test the communication with theorganizations. (R-97-29)
Inform all operating train crewmembersof the circumstances of this accidentand emphasize the crew responsibilitywhile in the operating compartment forthe safe operation of the train. (R-97-30)
Inform your engineers of the circum-stances of this accident and cautionthem not to use the reverser duringemergency brake applications forthose trains on which the use of thereverser will eliminate the dynamicbraking, thus increasing stopping dis-tance. (R-97-31)
--to the Maryland Mass Transit Administration:
Cooperate with CSX TransportationInc. in the development and installa-tion of a positive train separation con-trol system where Maryland RailCommuter equipment operates onCSX Transportation Inc. tracks. (R-97-32)
Develop an emergency plan that willprovide a detailed description ofemergency response procedures aswell as a protocol to coordinate ac-tivities with the emergency responseorganizations and other transportationentities when an accident occurs. (R-97-33)
Develop and implement, in coopera-tion with CSX Transportation Inc., acomplete training agenda for all CSXTransportation Inc. passenger train-crews that provides experience in thecorrect use of emergency equipment,in emergency communications proce-dures, and in passenger evacuation andassistance in an emergency and alsoincludes the distribution of a compre-hensive employee guidance manual.(R-97-34)
Conduct, in cooperation with the CSXTransportation Inc., the BaltimoreCounty Emergency ManagementAgency, the City of Baltimore Emer-gency Management Agency, the Met-ropolitan Washington Council of Gov-ernments, the Jefferson County Com-missioners, and the Berkeley CountyCommissioners, periodic disaster drillsto assess their emergency managementplans, to reinforce and evaluate theiremergency training, and to test thecommunication with the organizations.(R-97-35)
--to the U.S. Department of Transportation:
Review the testing protocols withinthe various modal administrations re-garding the flammability and thesmoke emissions characteristics of in-terior materials and coordinate the de-velopment and implementation ofstandards for material performanceand testing with the Federal RailroadAdministration and the Federal TransitAdministration. (R-97-36)
77
--to the Federal Emergency Manage-ment Agency:
Include in your training at the U.S.Fire Administration National FireAcademy a curriculum that addressesthe needs of State and local emergencymanagement agencies to respond to amajor railroad accident and that fa-miliarizes emergency response organi-zations with railroad equipment andappropriate rescue methods for rail-road accidents. (R-97-37)
--to the Governor and the General As-sembly of the State of Maryland:
Instruct and empower an appropriateState agency to provide continual, ef-fective, and independent safety over-sight of all aspects of the MarylandRail Commuter operations. (R-97-38)
--to the Association of American Railroads:
Assist the railroad industry with thedevelopment of positive train separa-tion control systems through a con-tinuing review of nonrailroad technol-ogy and assess its adaptability to rail-road communication-based controlsystems. (R-97-39)
Assist the railroad industry with thedevelopment of positive train separa-tion control systems by acting as aclearinghouse for information on thestatus and results of pilot projects andby disseminating that information tothe railroad industry and the Federaland participating State transportationorganizations. (R-97-40)
Assist the railroad industry with theinstallation and operation of positivetrain separation control systems bymaintaining industry standards to en-sure open architecture and aninteroperability of equipment for traincontrol systems. (R-97-41)
Inform your membership of the cir-cumstances of this accident and cau-tion them not to use the reverser dur-ing emergency brake applications forthose trains on which the use of thereverser will eliminate the dynamicbraking, thus increasing stopping dis-tance. (R-97-42)
--to the Montgomery County Emer-gency Management Agency:
Develop comprehensive proceduresfor responding to railroad passengertrain accidents and include these pro-cedures in your disaster plan. (R-97-43)
--to the Baltimore County Emergency Manage-ment Agency, the Baltimore City EmergencyManagement Agency, the Metropolitan Wash-ington Council of Governments, the JeffersonCounty Commissioners, and the BerkeleyCounty Commissioners:
Conduct, in cooperation with the CSXTransportation Inc. and the MarylandMass Transit Administra-tion/Maryland Rail Commuter, peri-odic disaster drills to assess theiremergency management plans, to rein-force and evaluate their emergencytraining, and to test the communica-tion with the organizations. (R-97-44)
--to the American Short Line Railroad Associa-tion, the Brotherhood of Locomotive Engineers,the United Transportation Union, the Interna-tional Brotherhood of Teamsters, and theAmerican Public Transit Association:
Inform your membership of the cir-cumstances of this accident and cau-tion them not to use the reverser dur-ing emergency brake applications forthose trains on which the use of thereverser will eliminate the dynamicbraking, thus increasing stopping dis-tance. (R-97-45)
78
Also, as a result of its investigation, the Na-tional Transportation Safety Board reiterates thefollowing recommendations:
--to the Federal Railroad Administration:
Promulgate Federal standards to requirethe installation and operation of a traincontrol system on main line tracks thatwill provide for positive separation ofall trains. (R-87-16)
Conduct, in conjunction with the As-sociation of American Railroads, theGeneral Electric Company, and theElectro-Motive Division of GeneralMotors, research to determine if thelocomotive fuel tank can be improvedto withstand forces encountered in themore severe locomotive derailmentaccidents or if fuel containment can beimproved to reduce the rate of fuelleakage and fuel ignition. Considera-tion should be given to crash or simu-lated testing and evaluation of recentand proposed design modifications tothe locomotive fuel tank, including in-creasing the structural strength of endand side wall plates, raising the tankhigher above the rail, and using inter-nal tank bladders and foam inserts. (R-92-10)
In conjunction with the Association ofAmerican Railroads and the RailwayProgress Institute, establish a firmtimetable that includes at a minimum,dates for final development of requiredadvanced train control system hard-ware, dates for an implementation of afully developed advanced train controlsystem, and a commitment to a date forhaving the advanced train control sys-tem ready for installation on the generalrailroad system. (R-93-12)
--to the General Electric Company:
Conduct, in conjunction with the Fed-eral Railroad Administration, the As-sociation of American Railroads, andthe Electro-Motive Division of Gen-eral Motors, research to determine ifthe locomotive fuel tank can be im-proved to withstand forces encoun-tered in the more severe locomotivederailment accidents or if fuel con-tainment can be improved to reducethe rate of fuel leakage and fuel igni-tion. Consideration should be given tocrash or simulated testing and evalua-tion of recent and proposed designmodifications to the locomotive fueltank, including increasing the struc-tural strength of end and side wallplates, raising the tank higher abovethe rail, and using internal tank blad-ders and foam inserts. (R-92-16)
--to the Electro-Motive Division of GeneralMotors:
Conduct, in conjunction with the Fed-eral Railroad Administration, the As-sociation of American Railroads, andthe General Electric Company, the re-search to determine if the locomotivefuel tank can be improved to withstandforces encountered in the more severelocomotive derailment accidents or iffuel containment can be improved toreduce the rate of fuel leakage and fuelignition. Consideration should begiven to crash or simulated testing andevaluation of recent and proposed de-sign modifications to the locomotivefuel tank, including increasing thestructural strength of end and side wallplates, raising the tank higher abovethe rail, and using internal tank blad-ders and foam inserts. (R-92-17)
79
BY THE NATIONAL TRANSPORTATION SAFETY BOARD
JAMES E. HALLChairman
ROBERT T. FRANCIS IIVice Chairman
JOHN A. HAMMERSCHMIDTMember
JOHN J. GOGLIAMember
GEORGE W. BLACK, JRMember
July 3, 1997
81
APPENDIX A
INVESTIGATION AND HEARING
The Safety Board was notified of the collision at 6 p.m. on February 16, 1996, and dispatched a majorrailroad accident investigation team. Investigative groups studied the operations, track, signals, mechanical,survival factors, and human performance aspects of the accident.
The CSX Transportation Inc., the National Railroad Passenger Corporation, the Maryland MassTransit Administration, the Federal Railroad Administration, the Brotherhood of Locomotive Engineers,the United Transportation Union, and the Montgomery County Department of Fire and Rescue Servicesassisted in the Safety Board investigation.
As part of its investigation, the Safety Board held a 3-day public hearing in Rockville, Maryland,between June 26 and 28, 1996, at which 27 witnesses testified. Parties to the hearing included the CSXTransportation Inc., the National Railroad Passenger Corporation, the Maryland Mass TransitAdministration, the Federal Railroad Administration, the Brotherhood of Locomotive Engineers, theUnited Transportation Union, and the Montgomery County Department of Fire and Rescue Services.
APPENDIX B
CHRONOLOGY OF SIGNAL SEQUENCE
The times recorded at Georgetown Junction differ about 2 minutes from the Jacksonvillesystem log. The discrepancy results from the delay that occurs in the order the information isreceived from field locations and the actual setting of the clock. The following information wasrecorded at Georgetown Junction:
1. Train No. K951-16: “Rock Runner” - a westbound ballast train.2. Train No. P279-16: Westbound MARC passenger train ahead of Amtrak passenger train.3. Train No. P28 4-16: Eastbound MARC passenger train ahead of MARC passenger train.4. Train No. P029-16: Westbound Amtrak passenger train involved in accident.5. Train No. P286-16: Eastbound MARC passenger train involved in accident.
P is the CSXT designation for a passenger train.Signal Identification corresponding to report:Signal No. 2 Eastbound Absolute Signal track 1 (EAS-1)Signal No. 4 Westbound Absolute Signal track 1 (WAS-1)Signal No. 6 Eastbound Absolute Signal track 2 (EAS-2)Signal No. 8 Westbound Absolute Signal track 2 (WAS-2)
Present Conditions at 1500:Switch No. 1 - NormalSwitch No. 3 - NormalSwitch No. 5 - NormalSignal No. 4 ClearSignal No. 2 at StopSignal No. 8 at StopSignal No. 6 at StopTrack 1 Interlocking Track Circuit EnergizedTrack 2 Interlocking Track Circuit EnergizedTrack Circuit West of Interlocking - Track No. 1 EnergizedTrack Circuit West of Interlocking - Track No. 2 EnergizedTrack Circuit East of Interlocking - Track No. 1 EnergizedTrack Circuit East of Interlocking - Track No. 2 EnergizedTrack Circuit West of Interlocking - From Georgetown Branch Energized
83
,,
APPENDIX B
Time Train Track Direction Event
15:14:04 P284-16 2 East Signal 6 ClearLast time signal No. 6 is cleared at Georgetown Junction on track No. 2 for an eastboundmove.15:21:3815:26:04
15:26:0415:26:3215:27:3815:28:16
15:32:1816:02:1016:05:39
16:05:39
E107-16E107-16
E107-16E107-16E107-16E107-16
E107-16P284-16P284-16
P284-16
11
122
2
WestWest
WestWestWestWest
West
East
Occupies track cast of GeorgetownOccupies Georgetown interlockingswitches 1N, 3N, 5NSignal 4 at StopOccupies track west of GeorgetownUnoccupied track east of GeorgetownUnoccupied Georgetown interlockingswitches 1N, 3N, 5NUnoccupied track west of GeorgetownOccupies track west of GeorgetownOccupies Georgetown interlockingswitches 1N, 3N, 5NSignal 6 at stop
Signal No. 6 remains at stop up to and beyond the time of the accident,about 1 hour and 33 minutes.16:05:5016:05:5516:06:1416:08:0616:32:5116:33:0816:33:2316:40:3716:05:39
16:46:4516:47:0516:47:079
16:47:1516:47:37
16:47:45
16:50:2016:50:3616:54:1017:03:4817:04:0417:04:09
P284-16P284-16P284-16P284-16
P275-16P275-16P275-16
P275-16P275-16P275-16
P275-16P275-16
P275-16
P277-16
17:04:04 P277-16
2222
l&2l&2
222
221
12
1
l&2l&2
1l&2l&2
2
EastEast
WestWestWest
WestWestWest
WestWest
West
West
West
Unoccupied track west of GeorgetownOccupies track east of GeorgetownUnoccupied Georgetown interlockingUnoccupied track cast of GeorgetownSwitch 1 out of correspondenceSwitch 1 reverseSignal 8 ClearOccupies track cast of GeorgetownOccupies Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at StopUnoccupied track east of Georgetownoccupies Georgetown interlockingswitches 1R 3N, 5NOccupies track west of GeorgetownUnoccupied Georgetown interlockingswitches lR, 3N, 5NUnoccupied Georgetown interlockingswitches 1R, 3N, 5Nswitch 1 out of correspondenceSwitch 1 normalUnoccupied track west of GeorgetownSwitch 1 out of correspondenceSwitch 1 ReverseSignal 8 at Clear
17:09:5117:14:45
17:14:4517:15:0317:15:0317:15:1117:15:38
17:15:42
17:16:4917:18:3117:21:1617:25:25
17:25:2517:25:4117:25:50
17:25:5917:26:18
17:26:22
17:26:5917:28:0717:31:3817:33:5717:35:4617:38:22
17:38:221738:41
17:38:42
17:38:4517:38:47
17:38:48
P277-16P277-16 West
P277-16P277-16P277-16 WestP277-I 6P277-16
P277-16 West
P279-16P279-I 6P279-16P279-16
P279-16P279-16P279-16
P279-16P279-16
P279-16
P029-16K951-16P279-16P286-16P029-16P029-16
P029-16P029-16P286-16P029-16P286-16P029-16P029-16P286-16P029-16P286-16
22
21212
1
2222
221
12
1
212222
2222222222
West
WestWest
WestWest
WestWestWestWest
WestWestWest
Westwest
West
WestWestWestEastWestWest
WestEast
WestEastWest
Occupies track cast of GeorgetownOccupies Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at StopOccupies Georgetown interlockingUnoccupied track cast of GeorgetownOccupies track west of GeorgetownUnoccupied Georgetown interlockingswitches 1R, 3N, 5NUnoccupied Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at ClearUnoccupied track west of GeorgetownOccupies track east of GeorgetownOccupies Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at StopUnoccupied track east of GeorgetownOccupies Georgetown interlockingswitches 1R, 3N, 5NOccupies track west of GeorgetownUnoccupied Georgetown interlockingswitches 1R, 3N, 5NUnoccupied Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at ClearOccupies track east of GeorgetownUnoccupied track west of GeorgetownOccupies track west of GeorgetownOccupies track East of GeorgetownOccupies Georgetown interlockingswitches 1R, 3N, 5NSignal 8 at StopSwitch 3 out of correspondence
switch 5 out of correspondence
Switch 5 Normal
switch 5 out of correspondence
*****Switches out of correspondence due to derailment*****
85
APPENDIX C
CHRONOLOGY OF ACCIDENT
Times indicated with (j) are from the Jacksonville signal log and are about 2 minutes aheadof time shown. All other times are from the Jacksonville AU dispatcher tape, published timetables, or the reporting agency.
Time
10:10 a.m.11:00 a.m.
12:59 p.m.1:24 p.m.
1:34 p.m.(j)
2:50 p.m.2:53 p.m.4:05 p.m.4:05 p.m.(j)
4:27 p.m.4:30 p.m.
4:33 p.m.(j)
4:47 p.m.(j)5:15 p.m.(j)5:18 p.m.
5:25 p.m.5:27 p.m.5:28 p.m.(j)5:30 p.m.(j)5:30(j) to5:34 p.m.(j)(approx.)
5:32 p.m.(j)5:34 p.m.(j)
5:36 p.m.(j)
Event
CSXT/MARC engineer reports to work in Baltimore.CSXT/MARC crew began first trip (train 251) from Camden station toWashington.MARC train 251 arrives in Washington 49 minutes behind schedule.CSXT/MARC crew began second trip (train 273) from Washington toBrunswick 24 minutes late.MARC train 273 travels westbound (WB) through Georgetown Jct. from track2 to 1.Amtrak train 29 crew reports to work.MARC train 273 ar r ives in Brunswick 33 minutes late.Scheduled departure time for Amtrak train 29 from Union Station.MARC train 284 travels eastbound (EB) through Georgetown Junction fromtrack 2 to 2.Scheduled departure time for Amtrak train 29 from Rockville station.CSXT/MARC crew began third trip (train 286) from Brunswick toWashington.AU Jacksonville dispatcher makes request and crossovers at GeorgetownJunction are lined for WB movements from track 2 to 1.MARC train 275 travels WB through Georgetown Juntion from track 2 to 1.MARC train 277 travels WB through Georgetown Junction from track 2 to 1.Scheduled time for MARC train 286 to depart (flag stop) from Rockvillestation.Amtrak train 29 departs Union Station 1 hour 20 minutes late.MARC train 279 travels WB through Georgetown Junction from track 2 to 1.MARC train 286 recorded by Rockville intermediate wayside signal on track 2.CSXT train K951, WB, stops for Georgetown Junction STOP signal on track 1.
Engineers of MARC train 279 and MARC train 286 communicate over radionear equipment defect detector at MP 11.7. Engineer of MARC train 279 hearsengineer of MARC train 286 acknowledge the wayside signal beforeKensington but did not hear what signal showed.MARC train 279 WB by Kensington signal on track 1.MARC train 286 EB by Kensington signal on track 2. MARC train 286 stops(approx. 50 seconds) at Kensington station (flag stop) and departs.Amtrak train 29 passes Takoma Park signal on track 2.
87
APPENDIX C
Time Event
5:38:41 p.m.
5:41 p.m.
5:44 p.m.5:45 p.m.
5:46 p.m.5:47 p.m.5:48 p.m.5:51 p.m.
5:53 p.m.
6:00 p.m.
6:02 p.m.
6:08 p.m.
6:15 p.m
.6:20 p.m.6:30 p.m
6:56 p.m.
Accident occurred: Train 286 (MARC) collided with train 29 Amtrak
First communication with CSXT AU dispatcher from CSXT traincrew K951.reporting accident. Reports fire and derailment of Amtrak train 29.CSXT AU dispatcher notifies MCFRS, which is already aware of accident.CSXT AU dispatcher attempts to call MARC operations center, but lineis busy.First MCFRS emergency responders on scene.CSXT train Q401 passes Rockville signal EB on track 2.CSXT AU dispatcher notifies WMATA operations center.Second call from CSXT train K951 to inform AU dispatcher of fire andevacuation, emergency personnel on scene.Amtrak train 29 conductor contacts AU dispatcher that Amtrak train 29 hadhad a MEDIUM CLEAR signal indication and about Amtrak crew condition.CSXT AU dispatcher contacted MARC operations center, which was alreadyaware of accident.AU dispatcher contacts CSXT manager of passenger operations in Baltimore,who already knew about accident.CSXT train 401 (following MARC train 286) with a CSXT road foreman onboard contacts AU dispatcher requesting permission to cut away and close upbehind MARC train 286. AU dispatcher authorizes CSXT train 401 to goahead and to be prepared to atop, looking out for emergency personnel.MARC train 281 (following Amtrak train 29) contacts AU dispatcher askingpermission to come up behind Amtrak train 29 and assist in passengerevacuation AU dispatcher gives OK.MCFRS command sectors established for level two incident.CSXT train Q401 stops near bridge.MCFRS IC receives message from his communications center about trainapproaching accident scene from Kensington and orders evacuation of site.MARC train 281 (following Amtrak train 29) instructed to move west of SilverSprint station to allow MARC train 285 into station to unload passengers.
88
APPENDIX D
EXCERPTS FROM CSXT OPERATING RULES
Rule 27
A signal imperfectly displayed must be regarded as the most restrictive indicationthat can be conveyed by that signal.
Exceptions:1. When the arms of a semaphore signal can be seen, they will govern.2. When one colored light is displayed in the cluster of lights of a color positionlight signal, it will mean the same as two lights in the cluster; or3. when one or more lower units of a color light signal aspect is dark, the aspectwill be observed as though the lights that should be displayed were displaying red.This does not apply to Rule C-290(a).
A signal imperfectly displayed must be reported promptly to the dispatcher.
Rule 34 (IN PART)
Crew members must maintain a lookout for signals or conditions along the trackthat affect the movement of their train. Crew members located in the operatingcab of an engine must clearly communicated the following information to eachother:
1. Concerning signals:
a) The name of each block and interlocking signal governing themovement of their train;
b) The number of the track to which the signal applies if in multiple trackterritory;
Rule 34-A
A crew member located in the operating cab of a radio-equipped engine mustannounce by radio the following conditions that affect the movement of thetrain:1. The name and location of each block and interlocking signal.2. The train entry into each DTC Block when entering the block.3. In multiple track territory, a crew member must announce the number of thetrack.
On radio-equipped freight trains, each of these announcements must beacknowledged by a crew member if located on the rear of the train, or by anycrew member who may be in a trailing unit of the traim.
89
APPENDIX D
Rule 34-C
Should the engineer fail to control the train in accordance with a signal indicationor a restriction imposed upon his train, other members of the crew must:1. Caution the engineer and, if necessary,2. Take action to ensure the safety of the train, (including stopping themovement).
Rule 98-G
When a train is stopped or its speed is reduced to 15 mph or less after passing asignal governing either the approach to a railroad crossing at grade, or thebeginning of TCS territory, the train must approach the next signal prepared tostop. It must do so until it can be seen that the indication of the signal permits thetrain to proceed.Exception: This does not apply where special instructions specify and governthe approach to locations with time-out features.
Rule C-292
Block, interlocking, and other fixed signal aspects and indications. STOP ANDPROCEED; stop then proceed at restricted speed.
Rule 103-G
When humping operations are being conducted in a hump yard equipped withremotely controlled switches, train or engine service employee may be required tocouple an air hose or adjust a coupling device. When this requires the employee toplace himself between rolling equipment located on a bowl track, the followingprotection must be provided against cars being released from the hump into thetrack involved.
1. The employee controlling any remotely controlled switch that provides accessfrom the apex of the bump to the track on which the rolling equipment is locatedmust be notified.2. Upon such notification, the operator of such remotely controlled switch mustline the switch against movement to the affected bowl track. The operator must
p l y , or must have applied a locking or blocking device to the control for thatswitch, and3. The operator must then notify the employee that the required protection hasbeen provided. The operator will remove the locking or blocking device only afterhe has been notified by the employee that the protection is no longer required onthat track.
APPENDIX D
Rule 104-C
Employees lining switches must ascertain that: 1. The route is lined for themovement. 2. The switch points fit properly. And 3. The switch lever is secured.
A main track switch must not be lined for a diverging movement of an approachingtrain until the employee attending the switch is assured by the affected train that themovement is to use the turnout or crossover.
Employees must ascertain that the switches that they are to use are in properposition. When such switches are being changed ahead of an approachingmovement, they must know that the movement is under control to prevent operatingover or through an improperly lined switch.
When kicking cars, a switch must not be lined for a following car going to anothertrack, until it is known that the preceding car will clear the route.
Rule 105
Trains may use tracks, other than main tracks, signaled tracks or sidings, withoutpermission. (see rule 46.)
Rule 106 (from Chessie System Operating Rules, effective September 1, 1985.)
Both the conductor and engineer are responsible for the safe and efficientoperation of their train and the observance of the rides. Train crew members mustcomply with the instructions of the conductor. When there is no conductor, or theconductor is not available to directly supervise, crew members will comply withthe instructions of the engineer. The train dispatcher must be advised as promptlyas practicable, by the engineer or conductor, of any condition that will delay thetrain or prevent it from making the usual speed. The conductor and engineer mustsee that members of their crew are familiar with their duties and instruct themwhen necessary in the observance of the rules and safe performance of their work.
Rule 450 (from Chessie System Operating Rules, effective September 1, 1985.)
Radio communications must not be used instead of hand signals when conditionsexist for continuous direct visual contact between the engineer and the signal ofthe employee(s) directing the movement. Radio communication and hand signals,except stop signals, must not be used simultaneously by a crew to direct train orengine movements. When changing from one mode of signaling to another, allcrew members involved must be notified and acknowledge their under-standingbefore the change is made.
Rule 804 (from Chessie System Operating Rules, effective September 1, 1985.)
When trying to determine the extent of damage in a derailment involving ahazardous material or a car leaking a hazardous material, the employees shouldalways approach the incident from up-wind in order to avoid any fumes. If fumesarc detected, the employees should withdraw to a safe distance until qualifiedpeople have arrived and determined the extent of the danger.
91
APPENDIX F
SIGNAL INCIDENTS, COMPLAINTS, AND FALSE PROCEED INDICATION S
A review of CSXT signal incident reports for the period between August 1995 and February1996 indicated approximately 134 signal and 21 highway/rail grade crossing incidents betweenWashington and Brunswick on the Metropolitan Subdivision. The majority of the signalincidents were weather-related (snow/lightning/rain), A.C. power interference by Amtrak at “F”Tower, commercial power outages, and burned-out signal lamps. All reported incidents were inthe fail safe mode.
During the public hearing, the FRA indicated that it had one signal complaint filed in 1994and none in 1995 on the CSXT. To follow up on reported traincrew complaints about theoperation of the signal system, the Safety Board provided a survey form to the Brotherhood ofLocomotive Engineers (BLE) and the UTU requesting a description of any unusual signaloccurrences. A total of 95 complaints were received from both organizations dating fromFebruary 1993; 3 of the 95 reports were at Kensington.
One of the complaints for March 13, 1996, involved multiple complaints. Three trainsreported the eastbound signal at Kensington “pumping” from CLEAR to STOP and PROCEED.The FRA had repros of the same condition. According to the FRA and the CSXT, the failurewas caused by one of the lamps in the green aspect being burned-out, and the filament in theother green aspect was broken and intermittently touching. The signal system is designed thatwhen both filaments in the yellow or green aspect burn out, the signal will go to STOP andilluminate the red aspect. This will prevent a train from passing a CLEAR signal to a dark signal.The signal should have displayed a CLEAR signal, but because of the condition of the signallamps, intermittently displayed a STOP and PROCEED signal. The signal was considered tohave failed in a fail-safe mode as designed. The FRA continues to investigate other BLE andUTU complaints on the Brunswick and Camden Lines. To date, no signals have failed in a morefavorable state, which would cause a false proceed signal indications1 on these lines.
In 1994, the CSXT had 13 false proceed signal indications. In 1995, the CSXT reportedseven confirmed false proceed signal indications, and one reported to May 1996. For 1995 and1996 these were attributed to vandalism (4), defective pole line (2), wiring short (l), and sunlightreflection (1).
By May 15, 1996, at a meeting held at CSXT headquarters in Baltimore, the FRA hadreceived over 30 complaints, which were under investigation. All complaints were reports ofsignal malfunctions which resulted in a more restrictive aspect than should have been displayed.T h e majority of the complaints were stated by one labor union as a lack of understanding on howthe signal system operates. Suggestions were made that during the employees annual operatingrules test, an overview of the signal system be presented. The FRA is in the process of finalizingits report and is to have a follow-up meeting with the parties.
1 A more favorable aspect than intended or other condition hazardous to the movement of a train.
103
APPENDIX G
FIRE TESTING RESULTS OF MARC CAR INTERIOR MATERIALS
Table 1. Test Results for Cushion Materials
Material Test
Cushions: ASTM D 3675(flammability)
3b. Seat
4c. Pad
4b. Padcover3b. or 4c. ASTM 662
(smoke)
4b.
~
Parameter
Ds(l.5)1
Ds(4.0)1
Ds(l.5)1
Ds(4.0)1
9.6
4.6
1498
54
105
43
92
2
6.6
5.3
1105
54
122
50
109
1. he maximum value under flaming or non-flaming is given.
6.5
5.9
833
50
117
51
109
Average
8
5
1145
115
48
103
ri?i!rcriterion225
21002
22002
0 0 2
22002
2. The 1979 Guidelines excepted cushions, we list the 1982 criteria for reference.
105
. . . .,
APPENDIX G
Table 2. Results for Upholstery Coverings
Material
Upholsterycovering
3a. Fabric
4a. Vinyl
3a. Fabric
4a. Vinyl
Test
FAR 28.853(flammability)
ASTM E 662(smoke)
Parameter
Flame time
Burnlength
Flamingdrips
Flame time
Burnlength
FlamingdripsD,(4.0)’
1. The maximum value under flaming or non-fl
1
1
1.75
0
1
2.2
0
254
359
ming is
4
2
0
0
2.6
0
245
351
iven.
2
1.5
0
0
1.9
0
205
321
Average
2.3
1.8
0
0.33
2.2
0
235
344
-
criterion
210s
‘6 in.
0s
210 s
26 in.
0s
2100
APPENDIX G
Table 3. Results for Wall and Ceiling Linings
Material
Walls/Ceilings:
1. Ceiling panel
2. Windowmask
1. Ceiling panel
2. Window mask
. - .
Test
ASTM E 162(flammability)
ASTM E 662(smoke)
Parameter
Is
Ds (1 .5)1
Ds (4.0)1
Ds(l.5)1
Ds (4.0)1
r
5.8
51
23
79
92
234
2
2.5
57
1
51
73
278
3
63.5
79
30
109
105
373
Average
24
62
18
80
90
295
1979criterion23 5
2100
2200
2100
22001. The maximum value under flaming or non-flaming is given. For large continuous sheets usedin applications, such as the passenger rail car, the edge effects should be different and, likely,less. The edge effect is very likely responsible for the Is value of 63.5 for the ceiling material.This is very likely attributed to the way that the cut sample responded in its metal frame holderof the apparatus.
107
APPENDIX H
ISSUANCE OF URGENT SAFETY RECOMMENDATIONS
National Transportation Safety BoardWashington, D.C. 20594
Safety Recommendation
Date: March 12, 1996
In Reply Refer To: R-96-4 through -6
Mr. John A. Agro, Jr.AdministratorMass Transit AdministrationWilliam Donald Schafer Tower6 Saint Paul StreetBaltimore, Maryland 21202-1614
About 5:38 p.m. on February 16, 1996, eastbound Maryland Rail Commuter (MARC)train 286 collided with westbound National Railroad Passenger Corporation (Amtrak) train 29,the Capitol Limited, at milepost 8.55 on CSX main track near Silver Spring, Maryland. TheMARC train was operating in the push mode in revenue service between Brunswick, Maryland,and Washington, D. C.; it consisted of a locomotive and three commuter cars. The Amtrak train,operating in revenue service between Washington, D. C., and Chicago, Illinois, consisted of 2locomotives and 15 cars.
The left front quadrant of the MARC cab car (the leading passenger car) separated andwas destroyed as a result of the collision. The fuel tank of the Amtrak lead locomotive rupturedon impact and the diesel fuel ignited. Fire engulfed the rear superstructure of the locomotive.Fuel spilled onto the MARC cab car, ignited, and destroyed the car.
One hundred sixty-four passengers, 13 on-board service personnel, 4 operating crew, and1 mechanical rider were aboard the Amtrak train. The engineer, assistant engineer, and conductorreceived minor-to-moderate injuries.
Three operating crewmembers and 20 passengers were on board the MARC train. Twocrewmembers and 7 passengers died of smoke inhalation, and 1 crewmember and 1 passengerdied as a result of impact injuries; 11 of the 12 survivors were injured.
APPENDIX H
2
Safety Board investigators interviewed six U.S. Department of Labor Job Corps studentswho were passengers in the MARC cab car and two individuals who were passengers in thesecond MARC car.
The students stated that after the impact, the car quickly filled with smoke, making it verydifficult to see. One student, who said he was sitting in the last seat on the right rear of the firstcar next to an emergency window, described the smoke as extending from the ceiling of the carto 2 feet above the floor. The student also stated that he did not have time to open the windowand that he believed it would be faster to exit using the door through which he had entered.Another student said that he had to crawl to the rear of the car on his hands and knees becauseof the smoke. Two other students, seated near the rear of the first car, proceeded through therear end interior door and made several unsuccessful attempts to open the left and right exteriorside doors. They stated that no instructions were provided concerning the operation of the doorhandle in an emergency.
These two students made their way into the vestibule of the second car and escapedthrough an opening in the damaged left front corner of the second car. All surviving students,as well as a 26-year-old passenger who was also seated in the cab car, followed.
The upper half of the side exterior doors on the MARC Sumitomo cars are fitted withfreed polycarbonate windows. When opened, the single-panel exterior side doors slide into apocket in the car body sidewall. Construction of the interior end doors, which have upper-halffreed windows, is similar to that of the exterior doors. These doors also slide into a wail pocketwhen opened; they are not equipped with emergency release mechanisms. The four exterior sidedoors are electrically operated and may be opened manually in an emergency by pulling anemergency handle located in one of four secured cabinets (two at each end of the passengercompartment). Each cabinet door is secured by two fasteners, which require a screwdriver orcoin to open. Instructions for opening the cabinet doors are on the door’s exterior. Instructionsfor operating the emergency handles to release the exterior doors are inside the cabinet.
The Safety Board is concerned that emergency quick-release mechanisms for the exteriordoors are located in a secured cabinet some distance from the door they control. Emergencycontrols for each door should be readily accessible and identifiable. Therefore, the Safety Boardbelieves that well-marked emergency quick-release mechanisms for exterior doors on MARC carsshould be relocated so that they are immediately adjacent to the door they control and readilyaccessible for emergency escape purposes.
Examination of the first and second cars revealed that the left and right rear exterior sidedoors of the former, as well as the front interior end door and the right front exterior door ofthe latter, were jammed. None of the doors had removable windows or pop-out emergencyescape panels (kick panels) for use in an emergency. The left front exterior door of the secondcar was destroyed. Thus, if the opening in the damaged car body of the second car had notprovided an escape route for the surviving passengers of the cab control car, the loss of life inthis accident could have been far greater.
APPENDIX H
3
Several students stated that they were unaware of the locations of the emergency exits,and none knew how to operate them. The Safety Board found that the interior emergencywindow decals were not prominently displayed and that one car had no interior emergencywindow decals. The Safety Board noted that the exterior emergency decals were often faded orobliterated and that the information on them, when legible, directed emergency responders toanother sign at the end of the car for instructions on how to open emergency exits, which is atime-consuming process. The Safety Board believes that all emergency exits should be clearlyidentified and provided at the exit with easily understood operating instructions. Theseinstructions should be prominently located on the car interior for use by passengers and on theexterior for use by emergency responders.
Therefore, the National Transportation Safety Board recommends that the Mass TransitAdministration of the Maryland Department of Transportation
Install removable windows or kick panels for emergency exits in interior andexterior passageway doors. (Class I, Urgent Action) (R-96-4)
Install an easily accessible interior emergency quick-release mechanism adjacentto all exterior doors. (Class I, Urgent Action) (R-96-5)
Install retroreflective signage on car interiors and exteriors at emergency exits thatcontains easily understood instructions and clearly marks all emergency exits(doors and windows). (Class I, Urgent Action) (R-96-6)
The National Transportation Safety Board is an independent Federal agency with thestatutory responsibility “to promote transportation safety by conducting independent accidentinvestigations and by formulating safety improvement recommendations” (Public Law 93-633).The Safety Board’s investigation of the MARC accident is continuing. The Safety Board isinterested in any action taken as a result of its safety recommendations. Therefore, it wouldappreciate a response from you regarding action taken or contemplated with respect to therecommendations in this letter. Please refer to Safety Recommendations R-96-4 through -6 inyour reply. If you need additional information, you may call (202) 382-6840.
Chairman HALL, Vice Chairman FRANCIS, and Members HAMMERSCHMIDT,GOGLIA , and BLACK concurred in these recommendations.
&u●
By Jim H
APPENDIX H
Transportation SafetyWashington, D. C. 20594
Safety Recommendation
Board
Date March 12, 1996
In Reply Refer To: R-96-7
Honorable Jolene M. MolitorisAdministratorFederal Railroad AdministrationWashington, D.C. 20590
About 5:38 p.m. on February 16, 1996, eastbound Maryland Rail Commuter (MARC)train 286 collided with westbound National Railroad Passenger Corporation (Amtrak) train 29,the Capitol Limited, at milepost 8.55 on CSX main track near Silver Spring, Maryland. TheMARC train was operating in the push mode in revenue service between Brunswick, Maryland,and Washington, D. C.; it consisted of a locomotive and three commuter cars. The Amtrak train,operating in revenue service between Washington, D. C., and Chicago, Illinois, consisted of 2locomotives and 15 cars.
The left front quadrant of the MARC cab car (the leading passenger car) separated andwas destroyed as a result of the collision. The fuel tank of the Amtrak lead locomotive rupturedon impact and the diesel fuel ignited. Fire engulfed the rear superstructure of the locomotive.Fuel spilled onto the MARC cab car, ignited, and destroyed the car.
One hundred sixty-four passengers, 13 on-board service personnel, 4 operating crew, and1 mechanical rider were aboard the Amtrak train. The engineer, assistant engineer, and conductorreceived minor-to-moderate injuries.
Three operating crewmembers and 20 passengers were on board the MARC train. Twocrewmembers and 7 passengers died of smoke inhalation, and 1 crewmember and 1 passengerdied as a result of impact injuries; 11 of the 12 survivors were injured.
APPENDIX H
2
Safety Board investigators interviewed six U.S. Department of Labor Job Corps studentswho were passengers in the MARC cab car and two individuals who were passengers in thesecond MARC car.
The students stated that after the impact, the car quickly filled with smoke, making it verydifficult to see. One student, who said he was sitting in the last seat on the right rear of the firstcar next to an emergency window, described the smoke as extending from the ceiling of the carto 2 feet above the floor. The student also stated that he did not have time to open the windowand that he believed it would be faster to exit using the door through which he had entered.Another student said that he had to crawl to the rear of the car on his hands and knees becauseof the smoke. Two other students, seated near the rear of the first car, proceeded through therear end interior door and made several unsuccessful attempts to open the left and right exteriorside doors. They stated that no instructions were provided concerning the operation of the doorhandle in an emergency.
These two students made their way into the vestibule of the second car and escapedthrough an opening in the damaged left front comer of the second car. All surviving students,as well as a 26-year-old passenger who was also seated in the cab car, followed.
The upper half of the side exterior doors on the MARC Sumitomo cars are fitted withfixed polycarbonate windows. When opened, the single-panel exterior side doors slide into apocket in the car body sidewall. Construction of the interior end doors, which have upper-halffreed windows, is similar to that of the exterior doors. These doors also slide into a wall pocketwhen opened; they are not equipped with emergency release mechanisms. The four exterior sidedoors are electrically operated and may be opened manually in an emergency by pulling anemergency handle located in one of four secured cabinets (two at each end of the passengercompartment). Each cabinet door is secured by two fasteners, which require a screwdriver orcoin to open. Instructions for opening the cabinet doors are on the door’s exterior. Instructionsfor operating the emergency handles to release the exterior doors are inside the cabinet.
The Safety Board is concerned that emergency quick-release mechanisms for the exteriordoors are located in a secured cabinet some distance from the door they control. Emergencycontrols for each door should be readily accissible and identifiable. Therefore, the Safety Boardbelieves that well-marked emergency quick-release mechanisms for exterior doors on MARC carsshould be relocated so that they are immediately adjacent to the door they control and readilyaccessible for emergency escape purposes.
Examination of the first and second cars revealed that the left and right rear exterior sidedoors of the former, as well as the front interior end door and the right front exterior door ofthe latter, were jammed. None of the doors had removable windows or pop-out emergencyescape panels (kick panels) for use in an emergency. The left front exterior door of the secondcar was destroyed. Thus, if the opening in the damaged car body of the second car had notprovided an escape route for the surviving passengers of the cab control car, the loss of life inthis accident could have been far greater.
A P P E N D I X H
3
Several students stated that they were unaware of the locations of the emergency exits,and none knew how to operate them. The Safety Board found that the interior emergencywindow decals were not prominently displayed and that one car had no interior emergencywindow decals. The Safety Board noted that the exterior emergency decals were often faded orobliterated and that the information on them, when legible, directed emergency responders toanother sign at the end of the car for instructions on how to open emergency exits, which is atime-consuming process. The Safety Board believes that all emergency exits should be clearlyidentified and provided at the exit with easily understood operating instructions. Theseinstructions should be prominently located on the car interior for use by passengers and on theexterior for use by emergency responders.
The Safety Board’s investigation of the MARC accident is continuing. However, theBoard is concerned that the unsafe conditions identified on MARC’s Sumitomo cars may existon other commuter lines subject to Federal Railroad Administration (FRA) oversight. Nocomprehensive passenger car safety standards are currently in place. Consequently, the SafetyBoard believes that FRA safety inspection personnel should determine whether the unsafeconditions identified on MARC’s Sumitomo cars exist on other lines. The Safety Board furtherbelieves that the FRA should issue emergency orders to correct such unsafe conditions, asnecessary, and incorporate the emergency measures into minimum passenger car safety Standards.
Therefore, the National Transportation Safety Board recommends that the FederalRailroad Administration:
Inspect all commuter rail equipment to determine whether it has: (1) easilyaccessible interior emergency quick-release mechanisms adjacent to exteriorpassageway doors; (2) removable windows or kick panels in interior and exteriorpassageway doors; and (3) prominently displayed retroreflective signage markingall interior and exterior emergency exits. If any commuter equipment lacks oneor more of these features, take appropriate emergency measures to ensurecorrective action until these measures are incorporated into minimum passengercar safety standards. (Class 1, Urgent Action) (R-96-7)
Chairman HALL, Vice Chairmann FRANCIS, and Members HAMMERSCHMIDT,GOGLIA, and BLACK concurred in this recommendation.
W*By: imH
APPENDIX K
FRA CORRESPONDENCE ON REGULATORY AUTHORITYOF COMMUTER RAILROAD OPERATIONS
NATIONAL TRANSPORTATION SAFETY BOARD
w(q Office of Surface Transportation Safety
‘%& Washington, D. C. 20594
Honorable Jolene M. Molitoris
400 Scvcnth Street S.W.Washington, D.C. 20590
Dear Ms. Molitoris:
The collision and derailment of the MARC commuter train with Amtrak train 29 at SilverSpring on February 16, 1996 has focused attention on commuter railroad options. The SafetyBoard needs to obtain accident and inspection statistics for commuter operations regulated by theFederal Railroad Administration (FRA) along with the inspection and compliance history for theseoperations. This information will assist the Safety
(s) of Board in its investigation of the recent accident
at Silver Spring by identifying the operator commuter operations and the accident history of thethose organizations and the commuter rail industry.
The Safety Board requests the following information:
● Listing of commuter railroad operations within FRA’s regulatory authority.
€ A five year accident history and rate per million train miles for commuter operations for thosecommuter railroad operations under FRA's regulatory authority.
Thank you for FRA's cooperation in this investigation.
Ed Dobranetski P.E.
Chief Major InvestigationsRailroad Division
137
APPENDIX L
NATIONAL TRANSPORTATION SAFETY BOARDAND POSITIVE TRAIN CONTROL
Railroads rely on train crews to comply with operating rules to prevent collisions. Theoperating rules explain the meaning of each signal, the proper response to a particular signal aspect,the procedure for conveying track warrants, the individual duties of the train crew members, andother vital information needed to safely operate trains. The operating rules provide all necessaryguidelines to prevent collisions providing crews understand and obey them.
An increasing majority of the train accidents investigated by the Safety Board have been theresult of human error. The best efforts by the railroads to train and test train crews for compliancewith operating rules has not guaranteed that individuals will take the correct action or that accidentswill not happen. Highly trained individuals still have accidents. A PTS control system provides theback-up to the engineer that ensures a train is properly operated.
After its investigation of a May 1986 rear-end collision at Brighton, Massachusetts,2 the SafetyBoard issued the following Safety Recommendation to the FRA:
R-87-16
Promulgate Federal standards to require the installation and operation of a train controlsystem on main line tracks that wi l l provide for positive separation of all trains.
In June of 1993 the FRA responded that the 1992 “Rail Safety Enforcement and Review Act”requires them to conduct a comprehensive review and safety inquiry into ATC. They reported thatseveral major railroads are beginning installation of the ATC communications platform or areactively considering installation. In addition the FRA stated that they were to explore possible trialapplications on one or more corridors selected for funding under the President’s high speed railground transportation initiative. The recommendation is still classified "Open--AcceptableResponse.” PTS control systems are still on the Safety Board’s list of most wanted transportationsafety improvements.
Following the accident involving the head-on collision between two trains in Kelso,Washington on November 11, 1993, the Safety Board reiterated Safety Recommendations R-87-16and R-93-12 to the FRA and issued the following safety recommendations to the FRA:
As part of your monitoring and oversight activities on the Burlington Northern and theUnion Pacific Railroad's train control demonstration project, identify and evaluate allpotential safety and business benefits of the train control system currently proposed for thenorthwest region of the United States. Consider the value of these benefits in your overallassessment of the system.
R-94-14
2 Railroad Accident Report--Rear End Collision Between Boston and Maine Corporation Commuter Train No.5324 and Consolidated Rail Corporation Train TV-14, Brighton, Massachusetts, May 7, 1986 (NTSB/RAR-87/02).
143
APPENDIX L
In conjunction with the AAR, identify and evaluate all of the potential benefits of positivetrain separation and include them in any cost benefit analysis conducted on positive trainseparation control systems.
Based on the FRA's response and the awarding of a financial grant to the Washington State DOT todevelop high-speed train control technology, through the use of computer modeling, as anassessment tool of the benefits of PTS, both recommendations were classified as “Open-Acceptable Response” in November of 1995.
From the same Kelso, Washington accident the Safety Board issued the following SafetyRecommendation to the AAR:
R-94-16
In conjunction with the FRA, identify and evaluate all of the potential benefits of positivetrain separation and include them in any cost benefit analysis conducted on positive trainseparation control systems.
The AAR responded in August of 1995 that its membership continue to believe that PTS must bejustified on the basis of safety benefits only. However, the Safety Board recognized the cooperativeefforts that were taking place on the UP/BNSF PTS pilot project and will revisit therecommendation once the results of the project are released. The recommendation is classified as“Open-Acceptable Response.”
144
AARACSAmtrak
ATCSB & OCADCFRCPCSXTCTCCVRDCSDTCDOTEASEMSEOEOCFAAFRAFTAGISICICC
ICSSMARC
APPENDIX M
ACRONYMS AND ABBREVIATIONS
Association of American Railroadsautomatic cab signalsNational Railroad Passenger
Corporationadvanced train control systemBaltimore and Ohio Railroadcomputer-aided dispatchingCode of Federal Regulationscontrol pointCS X Transportation Inc.centralized traffic controlcockpit voice recorderdisaster command systemdirect traffic controlU.S. Department of Transportationeastbound absolute signalemergency medical servicesemergency orderemergency operations centerFederal Aviation AdministrationFederal Railroad AdministrationFederal Transit Administrationgeographic information systemincident commanderU.S. Interstate Commerce
Commissionintermittent cab signaling systemMaryland Rail Commuter
MCEMA
MCFRS
MCPD
MDLI
MDOT
MOSMPMTA
MUNPRMPTSRPRPITCSUMTA
UTUVREWASWMATA
Montgomery County EmergencyManagement Agency
Montgomery County Fire andRescue Services
Montgomery County PoliceDepartment
Maryland Department of Laborand Industry
Maryland Department ofTransportation
manager of operations supportmilepostMass Transit Administration
(Maryland)multiple unitnotice of propose d rulemakingpositive train separationrecommended practiceRailway Progress Institutetraffic control signalUrban Mass Transit
AdministrationUnited Transportation UnionVirginia Railway Expresswestbound absolute signalWashington Metropolitan Area
Transit Authority
145