View Full Version : Runaway CSX train stopped after 2 hour uncontrolled trip!

May 16th, 2001, 01:39 AM
Runaway train stopped after uncontrolled 2 hours
May 15, 2001 Posted: 5:15 PM EDT (2115 GMT)

Officials are trying to halt a freight train out of control in northeast Ohio



KENTON, Ohio (CNN) -- A runaway freight train that barreled through 66 miles of northwestern Ohio with no one aboard was halted safely Tuesday by a railroad worker who jumped onto the moving train and pulled its brake.

The 47-car CSX train was slowed down by another engine in a coupling maneuver.

Two of the train's tank cars contained thousands of gallons of the hazardous material molten phenol acid, a toxic ingredient of paints and dyes harmful when it is inhaled, ingested or comes into contact with the skin.

Kathy Burns, a spokeswoman for CSX Transportation, said the train left its Toledo rail yard with no engineer at its controls. The engineer was "nearby" in the yard at the time, but not on the train, she said.

"It left the yard unmanned," she said. "Obviously, something went awry. That's what we're trying to get to the bottom of."

A runaway train in Ohio is slowed enough for CSX employee Jon Hosfeld to jump on and bring it to a halt (May 15)

Play video
(QuickTime, Real or Windows Media)

She said the railroad had dispatched mechanics and safety experts to the scene to investigate what happened.

Initial reports said the engineer apparently had a heart attack while in the locomotive and that he was aboard.

"There is no one on the train," said Kenton Police Detective Dennis Alexander after CSX worker Jon Hosfeld had climbed aboard and stopped the train.

Daring maneuver
Alexander said CSX slowed the runaway by latching a second engine to the end of the train. The second engine applied its brakes, slowing the first enough for Hosfeld to make the climb just Southeast of Kenton, about 55 miles northwest of Columbus. " That's how the train slowed down," Alexander said.

In a daring maneuver, Hosfeld, a 31-year CSX veteran who was waiting at a railroad crossing, jumped aboard the train, entered the locomotive and pulled its brake. The train was traveling about 10 mph when Hosfeld hopped aboard.

"We're just all happy that it ended without any injuries. That's always the number one issue for us: Safety first," said Corry Schiermeyer with the Federal Railroad Administration.

"Everything went smooth," said a spokeswoman for the Kenton Police Department said.

Fast pace
The CSX train averaged about 30-35 mph with no one at its helm for more than two hours. At one point, it was traveling at 47 mph.

The state highway patrol assisted in evacuating the area around any railroad crossings to make sure no one got in the way.

The train passed through three counties with officials unable to reach anyone aboard.

"They tried to stop it in Wood County, but it didn't work," said Kathy Palmerton with the Hancock County Sheriff's Department.

Palmerton said the train was pulling "flammable, combustible materials" and that it was headed south.

The CSX official said 25 cars were empty and 22 cars were loaded with material, including the two containing the hazardous materials.

Get more details, the photo of the lead unit #8888 :eek: and see the video at: http://www.cnn.com/2001/US/05/15/runaway.train.05/index.html


May 16th, 2001, 03:49 PM
On the TV news this morning, one of the guys interviewed said the "Hazardous Material" was Molten Sulfur. There was no mention of any other material. Molten sulfur is very hot, but not damgerous, we have used it for years to grout under heavy machinery, similar to the way we pour molten lead under machinery.

May 16th, 2001, 04:52 PM
Was the engine dead? If not, I thought there was a dead man's swith to prevent runaways. If the thing was in idle, I guess that device would not be useful. Why would anyone in their right mind step off an engine without setting the air brakes? Even so, the land looked fairly level, the train must have crept slowly away, and not taken off like a rocket. I wonder what is to happen to the engineer?

May 16th, 2001, 05:10 PM
i'm not sure if csx engines have a crewcall reset or not, many times yard engines don't. as was said this originated from a yard, therefore if they were flat switching they wouldn't be able to set the air because all of the cars air has been bled off and trainline seperated from engines so they can "kick" cars and let them go free therefore if the engine brakes got knocked off somehow it's so long.believe me it doesn't take much to get a car or a train rolloing you don't need much of a incline.

December 18th, 2001, 12:25 PM
This came to me by way of a friend who works for BNSF.

Memorandum on CSXT Runaway:

In accordance with your instructions, the following report summarizes
the work of the investigation committee which was assigned to review
analyze the events of May 15, 2001, in which a locomotive with cars
departed from Stanley Yard on the CSXT near Toledo, Ohio and traveled
south to Kenton, Ohio, with no crew member on board. I was assisted
this review by MPE Inspector Mike Lusher and OP Inspector Ed
both of whom responded initially to the incident, and by Chief
Harold Rugh, who provided essential technical information.

On May 15, 2001, at approximately 12:35 p.m., DST, an unmanned CSX
train consisting of one model SD-40-2 locomotive, 22 loaded, and 25
empty cars, 2898 gross trailing tons, departed Stanley Yard, which is
located in Walbridge, Ohio. The uncontrolled movement proceeded south
for a distance of 66 miles before CSX personnel were able to bring
movement under control. At the time of the incident, the weather was
cloudy with light rain. The ambient temperature was 55 degrees
Fahrenheit. There was no derailment of equipment or collision. There
were no reportable injuries as a result of the incident.

Circumstances Prior to Incident:
Yard crew Y11615, consisting of one engineer, one conductor and one
brakeman, reported for duty at Stanley Yard, Walbridge, Ohio, at 6:30
am, DST, May 15, 2001. After the normal job briefing with the
trainmaster, crew Y11615 performed routine switching assignments
approximately 11:30 am, at which time the crew received new
and a second job briefing. A few minutes before 12:30 p.m., the crew
entered the north end of track K12, located in the classification
with the intent to pull 47 cars out of K12 and then place these cars
departure track D10. Locomotive CSX 8888 was positioned with the
hood headed north. The engineer was seated at the
controls on the east side of the locomotive.

The locomotive coupled to the 47 cars on track K12, as instructed and
planned. The air hoses between the locomotive and the cars were not
connected, as is normal during this kind of switching operation. The
brakes on the cars were therefore inoperative. The brakeman notified
engineer by radio to pull north from K12. After the rear or 47th car
passed the brakeman's location, he walked west to position the switch
for the reverse movement to proceed into the assigned track D10.

The movement continued north out of K12 passing the conductor, who
positioned on the ground at the "Camera" switch. The conductor
the engineer by radio of the number of cars that had passed him and
received an acknowledgement from the engineer by radio.

The Incident:
With eight cars remaining to pass over the "Camera" switch, the
conductor notified the engineer by radio to prepare to stop. The
engineer did not respond to his communication. The conductor again
notified the engineer when four cars remained to clear the switch,
again there was no response from the engineer. The conductor then
ordered the engineer to stop movement, but again there was no
from the engineer and the movement continued.

In his interview, the engineer stated that as he pulled north out of
K12, he was notified by radio by the conductor that the trailing
switch for track PB9 off the lead was reversed. The engineer
that it would be necessary for the movement to be stopped short of
PB9 switch in order to line the switch for movement further along the
lead. Neither the conductor nor the brakeman were near the PB9
and the engineer intended to stop his train, dismount from the
locomotive, and align the switch to its normal position, if
The speed of the movement up the lead had now reached 11 mph. The
engineer observed the reversed switch, but due to the wet rail
conditions and the number of cars coupled to his locomotive, he
that he could not bring the equipment to a stop prior to passing
the misaligned switch. The engineer responded by applying the
locomotive's independent brake to full application. The independent
brake applies the brakes on the locomotive but does not apply the
on the individual freight cars. In addition, he reduced brake pipe
pressure with a 20 psi service application of the automatic brake
The automatic brake is pneumatic braking system designed to control
brakes on the entire train. Still certain he would not stop short of
switch the engineer attempted to place the locomotive in dynamic
mode. The dynamic brake utilizes the locomotive propulsion system to
brake the train. Dynamic braking is analogous to down shifting a
or automobile. Unfortunately, the engineer inadvertently failed to
complete the selection process to set up the dynamic brake. Under the
mistaken belief that he had properly selected dynamic brake, the
engineer moved the throttle into the number 8 position for maximum
dynamic braking. The engineer believed that the dynamic brake had
selected and that additional braking would occur.

However, since dynamic brake set up had not been established, the
placing of the throttle into the number 8 position restored full
locomotive power, instead of retarding forward movement of the train.

While the train was still moving at a speed of approximately 8 mph,
engineer dismounted the locomotive and ran ahead to reposition the
switch before the train could run through and cause damage to the
switch. The engineer was successful in operating the switch just
before the train reached it. The engineer than ran along side the
locomotive and attempted to reboard. However, the speed of the train
not decreased as the engineer had expected but had increased to
approximately 12 mph. Due to poor footing and wet grab handles on the
locomotive, the engineer was unable to pull himself up on the
locomotives ladder. He dragged along for approximately 80 feet until
released his grip on the hand rails and fell to the ground. Unable to
reboard and stop the movement of his train, the engineer ran to
a railroad employee, not a member of his crew but in possession of a
radio, located at the north end of the yard. This employee
notified the yardmaster of the runaway train. The yardmaster promptly
notified the Stanley tower block operator and the trainmaster. The
Toledo Branch train dispatcher located in Indianapolis was also
notified. The movement was now
proceeding southward on the Toledo Branch (Great Lakes Division)
governed by Traffic Control System (TCS) Rules. The time was
approximately 12:35 p.m.

The brakeman observed the train depart the yard but did not initially
see the engineer on the ground. The brakeman and another employee
personal vehicle to pursue the train to the next grade crossing to
attempt to board the train. Their immediate concern was for the
of the engineer, who they feared may have suffered a heart attack
at the controls of the locomotive. At the grade crossing, the two
employees were unable to board the train as the speed had increased
approximately 18 mph as it passed the milepost 4.

Local authorities and the Ohio State Police were notified of the
train at
approximately 12:38 p.m.

Attempts to Stop the Runaway:
At a siding called Galatea, near milepost 34, at approximately 1:35
p.m., the train dispatcher remotely operated the switch for the train
enter the siding. Previously a portable derail had been placed on the
track in an attempt to derail the locomotive and thereby stop the
movement. The portable derail was, however, dislodged and thrown from
the track by the force of the train passing over it, and the movement
the train was not impeded.

Northbound train Q63615 was directed by the dispatcher into the
at Dunkirk, Ohio. The crew was instructed to uncouple their single
locomotive unit from their train and wait until the runaway passed
location. at approximately 2:05 p.m., the runaway train passed
and the siding was lined for the crew of Q63615 to enter the main
and to pursue the runaway train.

At Kenton, Ohio, near milepost 67, the crew of Q63615 successfully
caught the runaway equipment and succeeded in coupling to the rear
at a speed of 51 mph. The engineer gradually applied the dynamic
of his locomotive, taking care not to break the train apart. By the
the train passed over Route 31 south of Kenton, the engineer had
the speed of the train to approximately 11 mph. Positioned at the
crossing was CSX Trainmaster Jon Hosfeld, who was able to run along
the unmanned locomotive and climb aboard. The trainmaster immediately
shut down the throttle, and the train quickly came to a stop. The
was 2:30 p.m. and the runaway train had covered 66 miles in just

An examination of the controls confirmed that the locomotive
brake had been fully applied, automatic brake valve was in the
zone, and the dynamic brake selector switch was not in the braking
All brake shoes had been completely worn to the brake beams.

The railroad was prepared to place an additional fully manned
ahead of the runaway south of Kenton, if necessary, to further slow
train. This rather hazardous option was fortunately not required.

Post Incident Investigation:
The engineer of Y11615 was slightly injured, but he declined medical
treatment. He was released from service with his crew at 5:30 p.m.
CSX did not require Drug or Alcohol testing of the crew, nor was
testing required.

The engineer first hired on the Pennsylvania Railroad in 1966, and he
was promoted to engine service in 1974. He received his most recent
check ride with a supervisor in January 2001. The engineer's
record is clean.

A Federal Railroad Administration Motive Power and Equipment
arrived at the location where CSX 8888 was stopped and performed a
mechanical inspection. He found all systems to function normally,
including sanders, headlight, auxiliary lights, bell, horn and the
alerter. The brake cylinder piston travel could not be determined,
because all brake shoes were completely burned off.

Air Brake System:
Locomotive CSX 8888 is a model SD40-2 manufactured by General Motors
Corporation (EMD). This unit is equipped with 26L type air brake
The Alerter system is connected directly to the air brake system
functions to provide an automatic full service penalty application of
the air brake system, and a power knock out (PC) caused by failure to
acknowledge the time out feature usually about 40 seconds. When the
Alerter time out has expired, the engineer must acknowledge by
the acknowledging switch which will reset the time out feature. The
Alerter system is nullified when locomotive brake cylinder pressure
20 psi is developed in the Independent Application and Release pipe.
This also prevents the P2A Application Valve from triggering a
brake application and PC action.

When the engineer of Y11615 placed the locomotive independent brake
valve into full application, a design pressure was developed in the
brake cylinders, depending on the type of relay valve, which
the Alerter System. Had the engineer not placed the independent brake
into the full applied position and caused a build up of brake
pressure, when the time out feature had expired, the system would
functioned causing not only an application of the locomotive brakes
a PC trip which would have resulted in a Power Knock Out bringing the
movement to a stop.

According to the interview of the engineer, he made a 20 psi brake
reduction with the automatic brake valve before dismounting the
locomotive. This by no means would have provided any braking power in
much as the brake pipe was not connected to the cars and the system
not charged.

Dynamic Brake:
General Motors Model SD-40-2 dynamic brake system is established by
placing the selector lever into Dynamic Brake Mode. This will convert
the traction motors to generators to produce voltage and amperage
is dissipated in the form of heat through the braking grids.
for the fields of the traction motors is developed through the main
generator and is regulated by increasing main generator outputby
increasing diesel engine rpm. This increase in engine rpm is
accomplished by increasing throttle position 1 (setup) through 8
positions. The effectiveness of the dynamic brake system is generally
maximized at speeds above 40 mph. At very speeds, below 10 mph, the
dynamic brake system is not effective. When the engineer failed to
properly move the selector lever into the dynamic brake mode, the
traction motors remained in the motoring mode. By placing the
handle into number 8 position in this set up, maximum locomotive
power was thus developed and diesel engine rpm would increase in a
manner to dynamic braking. Without first observing the load meter,
at low locomotive speeds, it may be difficult to determine
if the locomotive was in braking or power mode.

In the days following the incident, state and local officials
concern about the potential for this type of event to occur in the
future. While FRA does not dismiss any potential safety concern, the
exact circumstances that combined to cause this incident are highly
unlikely to recur. It is not uncommon, today, for our inspectors to
observe an engineer bring his locomotive to a full stop, dismount his
locomotive and operate a switch. This is done safely and in
with railroad safety rules and does not pose any special hazard to
employees or the general public. Railroad operating rules generally
prohibit any employee from dismounting, or mounting, moving
Engineers are required by railroad operating rules to apply a hand
and take other steps at the control stand to immobilize the
before dismounting.

FRA does have an initiative which will minimize the possibility of
runaway equipment resulting from unsecured equipment left unattended
a yard. This issue falls under the umbrella of the Switching
Fatality Analysis (SOFA) program, in my view, and we have taken steps
emphasize the securement of equipment in our ongoing SOFA activities.
For example, since the incident, State of Ohio and Federal inspectors
have visited CSX and NS terminals in Ohio, and elsewhere, to review
railroad managers policies and procedures relating to switching
including securement. We have not
been able to identify any systemic problems or shortcomings in
supervision, or operating practices that are cause for alarm.
do observe various local, non-systemic safety issues, and they are
addressed promptly with local managers, in accordance with our
policy and practice.

I might comment further on why this incident is unlikely to recur.
cause of the incident was multiple gross errors in judgement by the
locomotive engineer. For the incident to have occurred, each error
needed to be committed in sequence. First, the engineer was not
controlling the speed of his train on the lead, if he is unable to
for a switch improperly lined. This is covered by the railroad's
operating rules. Second, if the engineer cannot stop for a switch
improperly lined, the correct action to take is simply run through
switch and then stop without backing up, to avoid derailing the
Third, an engineer should never dismount his locomotive while it is
moving, except in extremely rare emergency circumstances, such as an
imminent collision. This is also covered by the railroad's operating
rules. Fourth, the engineer should not have relied on dynamic braking
low speed, since dynamic brakes are ineffective at speeds of less
ten mph, except on an AC locomotive. This is well known among
engineers. Fifth, the engineer seemed to believe, in error, that an
automatic brake application would improve braking power on single
locomotive with the independent brake fully applied. Sixth, the
misapplied the selector handle for "power" or "dynamic brake," an
that can only be understood if we assume the engineer acted with
haste and negligence. That all of these actions were taken by an
apparently well-qualified, fully rested employee with a good service
record is simply incredible.

It should be remarked that this incident could only have occurred
freight car switching and not during passenger car switching. Most
freight switching is done "without air," that is without air brakes
functional in the train. This is the
industry standard, and it has been so for many decades. Passenger
switching, on the other hand, is usually performed "with air," that
with air brakes functional in the train. In addition, it should be
that passenger coaches are rarely switched with passengers aboard,
because of concerns with passenger safety. In those rare
where passengers are on board during switching, the air brakes would
always be fully functional.

Finally, this incident could not have occurred to either a passenger
freight train involved in over-the-road operations. Before any
or freight train embarks on a run outside of the yard where the train
assembled, federal regulations require that the train receive a
inspection of its braking system and that the brakes be 100%
before the train is permitted to begin its journey.

December 19th, 2001, 12:57 AM
Mopac-- CSX does have a dead mans switch, and this was a road engine, but the Dynamics were on, and that will override the dead mans petal, or switch, whichever it was. The engineer thought he moved the brake handle to the applied position, but actually he moved the Throttle to run 8, and then the train was gone......

Hunter smile.gif

Mike Sheridan
December 20th, 2001, 12:17 AM
Unlikely to happen again, huh? Surely a 'dead-man device' should always idle the power and apply all the brakes. Does anyone here know why the system has this loophole in it? Obvious flaws like that are usually deliberate due to some operational need, though I'd have thought 'absence of driver' was operationally pretty significant ;) .

And doesn't this incident sound incredibly like the main premise in "Runaway Train" (the film)?

Incidentally, some years ago in the UK an electric passenger train (fortunately just vacated) that hadn't had the brakes set properly, ran away (the wrong way) down a main line for several miles. It just rolled - no power - on a 1:300 (about 0.3%) gradient thus illustrating the great advantage (?) of roller bearings. Its trip was ended when it hit another train that had been stopped and evacuated at the next (previous) station. Impact was at about 20mph I think and damage was slight (I think it was only a four car train of about 140 tons).

maintainance in the way
February 22nd, 2007, 03:35 AM
The "Cant Stop Express".....

Somehow "oooop's" dont seem strong enough.

February 22nd, 2007, 02:29 PM
Unlikely to happen again, huh? Surely a 'dead-man device' should always idle the power and apply all the brakes. Does anyone here know why the system has this loophole in it? Obvious flaws like that are usually deliberate due to some operational need, though I'd have thought 'absence of driver' was operationally pretty significant ;) .

If you read the transcript above, you would have noted that applying the independent brake nullifies the safety control feature (deadman, alerter, whatever). That way, when the locomotive is stopped with brakes applied, the engineer doesn't have to keep resetting the alerter or keep his/her foot on the deadman. The engineer had applied the independent brake in his attempt to stop, that nullifies the safety control.

The engineer lives about a mile from me. He felt so bad after it happened that he wanted to quit. He had a spotless record prior to this event.

Applying the automatic brake in addition to the independent will provide an additional 10-15 pounds of brake cylinder pressure, that's why he did that. But, Conrail had an option installed that reduced brake cylinder pressure in just such an instance to prevent or lessen the likelihood of sliding the wheels. That means that in this case, the automatic application actually reduced the brake cylinder pressure available for stopping.

maintainance in the way
February 22nd, 2007, 04:18 PM
The engineer lives about a mile from me. He felt so bad after it happened that he wanted to quit. He had a spotless record prior to this event.

God, he could have died. Switch points are easy to replace, unlike human limbs.

Big Al
February 22nd, 2007, 05:02 PM
Locomotives only achieve a braking power equal to +/- 40% of their weight. This allows a unit with full brake application to effectively overpower the brakes even with full cylinder pressure. The anit-slide feature is not universally used as some units will develop more cylinder pressure with a full service automatic brake reduction.

This was one of those occurrances that needed the 'perfect alignment of all the planets' to happen. Not common, but possible.

February 22nd, 2007, 05:04 PM
God, he could have died. Switch points are easy to replace, unlike human limbs.

CSX had started really cracking down on run-through switch incidents, and he didn't want to get nailed for running through one. We have heard that the FRA is going to make running through a switch a fine-able offense to the individual involved.

maintainance in the way
February 22nd, 2007, 10:31 PM
CSX had started really cracking down on run-through switch incidents, and he didn't want to get nailed for running through one. We have heard that the FRA is going to make running through a switch a fine-able offense to the individual involved.

Thats were I have a problem with the way it all works.

The whole "every incident", "every accident" is preventable. Well it
Probebly better if that were more of a goal than a threaght.

But heres a case were the "Or else" kicks in.
The consequences are out of line with the incident I mean,
at least pertaining to bent switch points.

Im not defending tearing up a switch or loosing his train.

But somewere along the line the threaght of the rules and the FRA
lent themselves to this being way worse.

Its sort of like a cop chasing down some guy when they know were he lives. Then blaming him 100% for the tragidy.

March 11th, 2007, 06:14 AM
Conrail had an option installed that reduced brake cylinder pressure in just such an instance to prevent or lessen the likelihood of sliding the wheels.

That means that in this case, the automatic application actually reduced the brake cylinder pressure available for stopping.

How does an Engineer ever "know" about these hidden features when
he/she hops aboard a unit for the first time?
Are there signs on board listing installed features?

Nick Leinonen
May 18th, 2008, 09:58 PM
How does an Engineer ever "know" about these hidden features when
he/she hops aboard a unit for the first time?
Are there signs on board listing installed features?

do a brake test as soon as you board the locomotive before you are coupled to a train, or just do a service application and then apply the independent brake and see if your pressures increase over the standard ind. pressure.

i've never seen one yet that will reduce the brake cylinder pressure...

the alerter system they had on that loco is garbage... on all of cn's power, the alerter continues to cycle even if you have in excess of 28psi brake if your reverser is thrown in fwd or rev or if your speed signal to the event recorder is anything but 0mph..

had that style of alerter system been on that loco, it would have stopped within a few minutes

May 19th, 2008, 12:19 AM
i've never seen one yet that will reduce the brake cylinder pressure...

Conrail locos do. I just had one today. My best guess is that the knockdown feature is there in case someone forgets to actuate.

May 25th, 2008, 03:05 AM
Wasn't it determined that the dead-man pedal showed evidence of being tampered with so the engineer wouldn't have to put up with it the alerter? Somewhere I had heard the safety device was removed or rigged so that it couldn't function properly.

May 25th, 2008, 04:58 PM
Dead-man or alerter. One or the other, but not both on locomotives. This engine is equipped with an alerter, but with the independent fully applied, the alerter doesn't do anything. The independent was fully applied and the throttle was in #8 (full). He thought he set it up for dynamic and had that fully applied as he ran out the door.

By the way, it now is a fine-able offense to run through switches, foul routes of improperly lined switches and crossovers, and not applying sufficient hand brakes per rules.