Passenger Trains > NTSB Request to FRA re Amtrak Collision

This document was sent to me in a pdf, and I can’t find a direct link at the moment, however here is the text...

The National Transportation Safety Board (NTSB) is investigating a head-on collision that occurred on February 4, 2018, about 2:27 a.m. eastern standard time on the CSX Transportation (CSX) Columbia Subdivision in Cayce, South Carolina. Southbound Amtrak train 91, operating on a track warrant, diverted from the main track through a reversed hand-thrown switch into a siding and collided head-on with stationary CSX local freight train F777 03.

The engineer and conductor of the Amtrak train died as a result of the collision. At least 92 passengers and crewmembers on the Amtrak train were transported to medical facilities. The engineer of the stopped CSX train had exited the lead locomotive before the Amtrak train entered the siding, ran to safety, and was not injured. The conductor of the CSX lead locomotive saw the Amtrak train approaching in the siding and ran to the back of locomotive. The conductor was thrown off the locomotive and sustained minor injuries.

The normal method of train operation on the subdivision was a traffic control system with wayside signals. Signal indications authorize movement in either direction. On the day before the accident, February 3, 2018, CSX signal personnel suspended the traffic control signal system to install updated traffic control system components for implementing positive train control (PTC) on the subdivision. During this time, scheduled to last through February 4, 2018, the signals would not operate and dispatchers would use track warrants to move trains through absolute blocks in the work territory. Although the installation was only partially complete, the signal personnel stopped work at the accident location at 7:00 p.m., and the signal suspension remained in place.

Previous Investigation

On March 14, 2016, NTSB investigated a similar collision, which involved two Union Pacific Railroad (UP) freight trains in Granger, Wyoming. Westbound UP freight train KG1LAC-13 (5718 West) traveled from the main track through a switch into a controlled siding and collided head on with stopped eastbound UP local freight train LCK41-14 (5155 East). At the time, UP was installing and testing PTC on the main track. While this work was in progress, UP employees suspended signals and established absolute blocks to ensure that trains could move safely through the areas without signals (the suspension).

When 5718 West entered the limits of the suspension on main track 1, it was traveling about 46 mph. The crew saw that the switch at CP G844 was lined so that their train would enter a siding instead of continuing on main track 1. Typically, switches were not lined until the dispatcher had decided the next movement through the switch; therefore, the switch was still lined for the previous train movement. The engineer immediately applied the emergency brakes, and the train slowed to about 30 mph and collided with the stopped 5155 East.

The NTSB determined that the probable cause of the accident was that the employee-in-charge incorrectly used information from a conversation with the train dispatcher as authorization to send a train into the signal suspension territory. Contributing to the accident was the failure of the conductor pilot at CP G844 to check the switch position before authorizing the train to enter the signal suspension territory.

Previous Recommendation to Federal Railroad Administration

On July 14, 2009, a Dakota, Minnesota & Eastern Railroad (DME) freight train was operating under track warrant authority in nonsignaled territory on the main track when it went into Bettendorf Yard in Bettendorf, Iowa, via a misaligned hand-operated switch and struck 19 loaded railcars on a yard track. The hand-operated switch had been left incorrectly lined from the main track onto the yard track by the crew of a BNSF Railway local train. The engineer and the conductor on the DME train sustained fatal injuries. The NTSB determined that the probable cause of the accident was, in part, the BNSF Railway local train crew releasing track warrant authority before returning the hand-operated switch to the correct position.

As a result of the Bettendorf, Iowa, accident, the NTSB issued the following recommendation to the Federal Railroad Administration (FRA):
R-12-29

Require that until appropriate switch position warning technology is installed on main track switches (in non-signaled territory not equipped with positive train control), when a main track switch has been reported relined for a main track, the next train to pass the location approach the switch location at restricted speed. That train crew should then report to the dispatcher that the switch is correctly lined for the main track before trains are allowed to operate at maximum authorized speed.

On April 18, 2013, NTSB classified Safety Recommendation R-12-29 Closed⸺Reconsidered because the FRA argued that implementing this recommendation, which would apply to 52% of US railroad route miles, would be too disruptive to transportation.

Ongoing Investigation

In the current accident in Cayce, South Carolina, as well as in the Granger accident, the evidence indicates that human decision making and actions likely played key roles in the accident scenarios. In both accidents, safe movement of the trains through the signal suspension depended on proper switch alignment, which, in turn, relied on error-free manual work. The risk of error in the manual work was not safeguarded, either by technology or supervision. Thus, the reliance on error-free human performance for safe train movement created a single point-of-failure in the operating practices currently used and in compliance with extant regulations.6 The NTSB concludes that additional measures are needed, such as restricted speed, to ensure safe operations during signal suspensions, especially during the movement of passenger trains, due to the likelihood of harm to the traveling public.

Therefore, the NTSB recommends that the Federal Railroad Administration (FRA) issue an Emergency Order directing railroads to require that when signal suspensions are in effect and a switch has been reported relined for a main track, the next train or locomotive to pass the location must approach the switch location at restricted speed. After the switch position is verified, require the train crew to report to the dispatcher that the switch is correctly lined for the main track before trains are permitted to operate at maximum-authorized speed.

This recommendation would apply only to areas subject to a signal suspension—a minute portion of the United States’ rail network, whereas Safety Recommendation R-12-29 applied to all railroad dark territory.

Recommendation

As a result of this report, the National Transportation Safety Board makes the following urgent safety recommendation:
To the Federal Railroad Administration:
Issue an Emergency Order directing railroads to require that when signal suspensions are in effect and a switch has been reported relined for a main track, the next train or locomotive to pass the location must approach the switch location at restricted speed. After the switch position is verified, the train crew must report to the dispatcher that the switch is correctly lined for the main track before trains are permitted to operate at maximum-authorized speed. (R-18-005) (Urgent)

BY THE NATIONAL TRANSPORTATION SAFETY BOARD
ROBERT L. SUMWALT, III
Chairman
T. BELLA DINH-ZARR
Member
EARL F. WEENER
Member

Adopted: February 13, 2018


Edited for formatting...



Edited 3 time(s). Last edit at 02/15/18 19:41 by ats90mph.

Meanwhile at CSX...

https://www.bizjournals.com/jacksonville/news/2018/02/14/as-csx-workforce-shrinks-accidents-pile-up-and.html

Here are the links to the NTSB's press briefing and the pdf containing the recommendation. Seems to me that this recommendation should apply to all non-signalled territories and conditions, not just signal suspensions.

https://www.ntsb.gov/news/press-releases/Pages/NR20180215.aspx

https://www.ntsb.gov/investigations/AccidentReports/Reports/RSR1801.pdf?utm_source=news&utm_medium=email&utm_content=railroad



Edited 1 time(s). Last edit at 02/15/18 20:23 by OTG.

OTG Wrote:
-------------------------------------------------------
> Seems to me that this recommendation should apply to all
> non-signalled territories and conditions, not just
> signal suspensions.

That would be a ridiculously impractical, knee-jerk reaction. There is a lot of dark territory in the U.S. and doing so would bring the railroads, and the economy, to a grinding halt.

trainjunkie Wrote:

>
> That would be a ridiculously impractical,
> knee-jerk reaction. There is a lot of dark
> territory in the U.S. and doing so would bring the
> railroads, and the economy, to a grinding halt.


Exactly. Typical knee-jerk "we must do something" reaction so common in today's world.

At one time main track switches were respected and always lined back.
The operating rules were clear that if a main track switch were left lined in the reverse position
it had to be tended by a crew member of the next train that was to use it. Leaving a main track switch reverse and
unattended was a SERIOUS rule infraction.

Then the railroads decided the caboose was unnecessary and switches could be left open willy-nilly.
Main track switches were no longer respected and now we are reaping the harvest of that policy.
A large portion of the work force has grown up under this nonsense.

Last summer I was driving along a non-signaled CSXT line and was shocked to see a main track switch
lined and locked for a yard track with nobody around. The local crew had gone in the yard and was taking
lunch. I realize they probably had a "work between" authority and there was no chance of another train coming.
But what happens when the overbearing CSXT trainmaster shows up and begins shouting orders about getting in
the car because a train needs to be recrewed somewhere. What about the switch? I think we lined it back. Clear
your authority with the DS, let's go.

Where were the NTSB and FRA when these policies began? Doing nothing!!!!!!!
Always closing the barn door after the cows have left. After the accidents and deaths they show
up pontificating and acting like they are the only ones competent to judge what is safe.

First the Rule Book and common sense in realigning switches for main track before clearing up, then Emergency Order 24 now this and still have wrecks.

I am glad not to be out there anymore, but when I started that was beat into your head to realign, lock then clear up no matter what. Railroad 101.

It seems it just gets dumbed down more and more and the answer is gobs of technology.

I don't miss it at all.

Once PTC is perfected to the point it's considered perfect, the trains will then be made autonomous and on-board people will be out of the equation . . .

RailNet802, owevaaaah

True this. No caboose=less safe operation.

kk5ol Wrote:
-------------------------------------------------------
> Once PTC is perfected to the point it's considered
> perfect, the trains will then be made autonomous
> and on-board people will be out of the equation .
> . .
>
> RailNet802, owevaaaah

And they will carry PTC equipped autonomous passenger robots only too :D And the whole thing will run on HO gauge :P

CardinalFang Wrote:
-------------------------------------------------------
> True this. No caboose=less safe operation.

Hogwash!

HotWater Wrote:
-------------------------------------------------------
> CardinalFang Wrote:
> --------------------------------------------------
> -----
> > True this. No caboose=less safe operation.
>
> Hogwash!

In dark territory with hand-throw switches, cabooseless train operations coupled with fewer crew members and modern velocity mandates definitely have the potential to lay the groundwork for issues and accidents.

trainjunkie Wrote:
-------------------------------------------------------
> OTG Wrote:
> --------------------------------------------------
> -----
> > Seems to me that this recommendation should
> apply to all
> > non-signalled territories and conditions, not
> just
> > signal suspensions.
>
> That would be a ridiculously impractical,
> knee-jerk reaction. There is a lot of dark
> territory in the U.S. and doing so would bring the
> railroads, and the economy, to a grinding halt.

How would one train approaching only the last switch lined prepared to stop slow down the whole railroad? If the order was for every train approaching every switch I would agree with you, but most dark territories have relatively light traffic levels anyway. Granted it would suck to be following or meeting a local on a dark subdivision.


Englewood Wrote:
-------------------------------------------------------
> trainjunkie Wrote:
>
> >
> > That would be a ridiculously impractical,
> > knee-jerk reaction. There is a lot of dark
> > territory in the U.S. and doing so would bring
> the
> > railroads, and the economy, to a grinding halt.
>
>
> Exactly. Typical knee-jerk "we must do something"
> reaction so common in today's world.
>
> At one time main track switches were respected and
> always lined back.
> The operating rules were clear that if a main
> track switch were left lined in the reverse
> position
> it had to be tended by a crew member of the next
> train that was to use it. Leaving a main track
> switch reverse and
> unattended was a SERIOUS rule infraction.
>
> Then the railroads decided the caboose was
> unnecessary and switches could be left open
> willy-nilly.
> Main track switches were no longer respected and
> now we are reaping the harvest of that policy.
> A large portion of the work force has grown up
> under this nonsense.
>
> Last summer I was driving along a non-signaled
> CSXT line and was shocked to see a main track
> switch
> lined and locked for a yard track with nobody
> around. The local crew had gone in the yard and
> was taking
> lunch. I realize they probably had a "work
> between" authority and there was no chance of
> another train coming.
> But what happens when the overbearing CSXT
> trainmaster shows up and begins shouting orders
> about getting in
> the car because a train needs to be recrewed
> somewhere. What about the switch? I think we
> lined it back. Clear
> your authority with the DS, let's go.
>
> Where were the NTSB and FRA when these policies
> began? Doing nothing!!!!!!!
> Always closing the barn door after the cows have
> left. After the accidents and deaths they show
> up pontificating and acting like they are the only
> ones competent to judge what is safe.


I'm trying to understand your point here. Are you calling this recommendation as overbearing, or as late?

Mainline switch violations are still treated a major violations, I know a guy who left a mainline switch open twice a few months apart and was terminated after the second one. However, we used to have 6 crew members per train, and cabooses. As we put more and more responsibility on fewer and fewer crew members we have to expect more and more mistakes, as there are fewer safeguards against them. The railroad has changed immensely over the past 40 years, the rules need to change with it. Old heads saying "40 years ago this would be unthinkable" need to upgrade their thinking or they're part of the problem.

Reality is something unthinkable 40 years ago can now be thinkable. Reality is that mainline switches are not being respected as they once were, and rules need to be changed to reflect that. That's not a knee-jerk, it's something that should have been anticipated and changed for a long time ago, before however many deaths have happened happened.

OTG Wrote:
-------------------------------------------------------
> How would one train approaching only the last
> switch lined prepared to stop slow down the whole
> railroad? If the order was for every train
> approaching every switch I would agree with you,
> but most dark territories have relatively light
> traffic levels anyway. Granted it would suck to
> be following or meeting a local on a dark
> subdivision.

Ever had enough Form As (temporary speed restrictions), or speed restricted high-wides over certain bridges or on certain curves on a trip that they killed the trip? I have. Add in 3 or 4 "approach switch at XYZ prepared to stop" and those plus your temporaries are going to make trains die all over the place. The bigger the train, the bigger the delays. Depends on where the switches are and how far in advance of them you can see the points.

Personally, I couldn't care less. Just more money for me as I collect OT while getting dog caught. But I was playing devil's advocate on behalf of the carriers who are sure to scream bloody murder on this one. They are already having capacity issues, they're short on crews, and PTC is already slowing things down enough. This would break certain subdivisions IMHO, and all for very little return. Statistically, considering successful ton-miles handled in dark territory, this is a non-issue to the carriers.

I still think they should make all Atk engineers have at least 610 road time slips (no yard, no local, etc.) Road is pool freight. With that experience it may stop an engineer from falling into a trap.

I know the wreck where they side swiped a C&O in the pass, wasn't because of experience. But if he knew the road, he should have kept it down until all switches could be seen until he got on on tangent track.

If it had been me, going into dark territory, I would not be going 59 MPH. I would have known there were a few switches and other places to look out for before I picked up speed. The dispatchers can only go on what they are told. When the train on the pass left the switch open, the crew should have told him. If not the dispatcher should have questioned the train or crew and told Atk he didn't know the position of any switches.

Its all water under the bridge and 2 lives were lost. It could have been prevented and the Feds should change the rules to protect the crews and the public.

THERE IS NO TRAINING THAT REPLACES EXPERIENCE!

Flame On.

Nudge

spnudge Wrote:
-------------------------------------------------------
> I still think they should make all Atk engineers
> have at least 610 road time slips (no yard, no
> local, etc.) Road is pool freight. With that
> experience it may stop an engineer from falling
> into a trap.

This is not practical for Amtrak, unless you can find a freight carrier willing to donate their engineers for such service. What is practical is requiring a minimum number of trips as an AC on Amtrak before promoting to Conductor, a minimum number of trips as a Conductor before promoting to engineer, and a minimum number of trips as a student engineer before promoting to engineer. No more hiring engineers off the street.

spnudge Wrote:
-------------------------------------------------------
> I know the wreck where they side swiped a C&O in
> the pass, wasn't because of experience. But if he
> knew the road, he should have kept it down until
> all switches could be seen until he got on on
> tangent track.
>
> If it had been me, going into dark territory, I
> would not be going 59 MPH. I would have known
> there were a few switches and other places to look
> out for before I picked up speed. The dispatchers
> can only go on what they are told. When the train
> on the pass left the switch open, the crew should
> have told him. If not the dispatcher should have
> questioned the train or crew and told Atk he
> didn't know the position of any switches.

I've got just over ten years on the railroad, four of which have been in the seat, and have never run in dark territory on a mainline. However, twice recently (before the accident, the second time less than a week before) I ran during a signal suspension. I won't say I was prepared to stop at all facing point switches, but I certainly wasn't approaching them at 59 MPH, and I was ready to plug it if I was lined into something. The whole time I imagined my old head instructor old-heading me; "They're all lined for you, you have a track warrant, just go!"

To me it's common sense not to trust everybody else to do everything properly, but there is some weird, implicit trust expected of you from the same people that tell you not to trust anybody, and often I feel railroaders in general (especially younger and less experienced guys) are pressured to take everything at face value rather than slowing down and double checking everybody else's work in order to be safe. This is why I feel rules requiring this are a good thing.

spnudge Wrote:
-------------------------------------------------------
> Its all water under the bridge and 2 lives were
> lost. It could have been prevented and the Feds
> should change the rules to protect the crews and
> the public.
>
> THERE IS NO TRAINING THAT REPLACES EXPERIENCE!
>
> Flame On.
>
> Nudge


2 lives lost is not "Water under the bridge". The industry needs to learn from this, and rules need to change to prevent this from happening again. More important than experience is having a healthy safety culture; empowering employees to take the safest course, not pushing them to take the quickest course. If the FRA has to write rules requiring that because the industry can't do it themselves, then so be it.

I don't think your flame-retardant suit is too beat up. We're on the same page, just with two different approaches.

OTG Wrote:
-------------------------------------------------------
> This is not practical for Amtrak, unless you can
> find a freight carrier willing to donate their
> engineers for such service. What is practical is
> requiring a minimum number of trips as an AC on
> Amtrak before promoting to Conductor, a minimum
> number of trips as a Conductor before promoting to
> engineer, and a minimum number of trips as a
> student engineer before promoting to engineer. No
> more hiring engineers off the street.

On that note, the freight carriers need to quit hiring "conductors" off the street. Go back to hiring them as brakemen (or "conductor trainees") and give them some time in the field to learn how things work before trying to certify them and hand them all that responsibility.

I think it's outrageous that the Feds demanded conductors have 242 certification, then allowed the carriers to hire them off the street with no time to get any actual experience before turning them loose. Most new-hire conductors today have no concept of the level of responsibility they have, or the reason for half the rules and procedures they are expected to understand and follow.

Whenever I see a new class of "conductors" show up on the property I roll my eyes and hope to God most of them make it the first couple of years without screwing up and taking someone down with them. But I always know at least some of them are going to blow it. At least when they were hired as brakemen, we had a couple years to figure out whether or not they are a liability or an asset before they were handed all that responsibility. Now it's a crap shoot. The carriers should bear some responsibility for this but all they care about is cutting labor to the bone. Looks good on the department's balance sheet because the cost of injuries and accidents comes out of a different budget then theirs.

Englewood Wrote:
-------------------------------------------------------

>
> Exactly. Typical knee-jerk "we must do something" reaction so common in today's world.
>
> At one time main track switches were respected and always lined back.
> The operating rules were clear that if a main track switch were left lined in the reverse position
> it had to be tended by a crew member of the next train that was to use it. Leaving a main track
> switch reverse and unattended was a SERIOUS rule infraction.
>

Perhaps it is time to identify, by name, the individual CSX crew member responsible for properly lining this switch for the main after the train went into the yard.
Publicly bring him forward, identify him/her, and summarily FIRE them. Repeat as necessary when future infractions occur. The lesson will be learned.

Getting back to the original NTSB recommendation, this seems very reasonable:

Again:

"Recommendation

"As a result of this report, the National Transportation Safety Board makes the following urgent safety recommendation:
To the Federal Railroad Administration:
Issue an Emergency Order directing railroads to require that when signal suspensions are in effect and a switch has been reported relined for a main track, the next train or locomotive to pass the location must approach the switch location at restricted speed. After the switch position is verified, the train crew must report to the dispatcher that the switch is correctly lined for the main track before trains are permitted to operate at maximum-authorized speed. (R-18-005) (Urgent)"

OTG,

You did the right thing. I was on a Atk Special from SLO to LA. We got stopped with a Red Interlocking signal at the Allen Ave. X-Overs. Burbank called me and said I had permission to pass the signal, thru the x-over and run against the grain to the Top End of "A" yard. A yard got on the radio and said he had the flag and come on down. Well, I took off but took my time.I put 1st squirt in them and pulled, I was on a 1.1% grade down hill. The RFE that was riding the head end told me, "You know you can take it up to 59 mph." I told him I was comfortable with the speed I was going and just started into a right hand curve. Towards the end of the curve was the Burbank Switcher, coming out of an industry onto the main. They had left 3 cars on the westbound main. Well, I took another bite then plugged them. We stopped back enough to hold maybe 5 cars. Having the air under them helped. If I had been doing "59" there would have been about 5 of us killed or injured.

That is what experience teaches you. They can tell you how fast you can go. But they can't tell you how slow. "When In Doubt, the Safest Course will be taken."



Nudge

hazegray Wrote:
-------------------------------------------------------
>
> Perhaps it is time to identify, by name, the
> individual CSX crew member responsible for
> properly lining this switch for the main after the
> train went into the yard.
> Publicly bring him forward, identify him/her, and
> summarily FIRE them. Repeat as necessary when
> future infractions occur. The lesson will be
> learned.

Are you in management at CSXT? I am sure that is exactly what CSXT would like
to have happen. Nothing to see here, move along. Don't look deeper at the
safety "culture" (which usually means none) or what behind the scenes action
was taking place at management level. Perhaps there were short cuts taken at
upper levels concerning verifying switch positions or other aspects of operating
during signal suspension. I am NO fan of the NTSB but without some sort of public
investigation we will never know what happened. Keep an eye on the NTSB website about
6 months from now. The Docket Items will begin to appear which will include TRANSCRIPTS
of the NTSB interviews with the surviving employees involved. Then the truth will begin
to come out.