THE magazine of railroading

SEARCH TRAINSMAG.COM

Enter keywords or a search phrase below:

Casselton Oil Train Wreck

11112 views
305 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January, 2014
  • 3,749 posts
Casselton Oil Train Wreck
Posted by Euclid on Tuesday, February 07, 2017 5:57 PM

Video is in this link, but this may be older information:

https://www.ntsb.gov/investigations/Pages/casselton_nd.aspx

 

The final report was just released, but there is also prexisting reports in searches.  Here may be a better video link:

http://www.inforum.com/news/4213406-watch-feds-release-video-trains-fiery-2013-crash-near-casselton

 

It shows the locomotive video from the grain train as it derails the grain car and goes into emergency with the oil train approaching and passing. Then it shows the locomotive video from the oil train as it encounters and collides with the grain car.  Also included is the verbal communications between crews, etc. 

 

 

  • Member since
    June, 2001
  • From: US
  • 13,313 posts
Posted by Mookie on Tuesday, February 07, 2017 6:56 PM

Watched it twice.  The grain train never so much as jiggles at the derailment.  Looks like the only indication is the speed going down.  Then the thrill of the inevitable hitting of the grain car.  

I don't know how injuries are avoided in something that heavy doing a flop like that!

She who has no signature! cinscocom-tmw

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Tuesday, February 07, 2017 8:45 PM

That is quite a dramatic video with a lot of added information about the sequence of events as well as the actual audio.  I find it quite interesting how events unfolded as the oil train was approaching the grain train after it was known that the grain train had experienced an undesired emergency application of brakes. 

Apparently the grain train was blowing snow which obscured any visual indication of a dragging car to both its conductor and the crew of the oil train.  But it seems that with this impending meet, a UDE would have been of great concern.  Yet it seems to have taken 32 seconds before the grain train crew announced that they had gone into emergency. 

  • Member since
    May, 2003
  • From: US
  • 11,944 posts
Posted by BaltACD on Tuesday, February 07, 2017 9:12 PM

Euclid
Yet it seems to have taken 32 seconds before the grain train crew announced that they had gone into emergency.

The bigger part of that 'delay' is waiting for the emergency brake application to finish exhausting.  Brake air exhaust, especially from an emergency application, is loud - so loud that it will interfere with the intelligibility of a person speaking over the radio.  Remember the entire pressure of the trainline gets exhausted to atmosphere with an emergency application - that is a lot of air.

If you have an air tank at home that you use in maintaing your vehicles and have a 'blow nozzle' attachment - see how long it takes to 'vent' the tank to atmosphere - raise the sound and the volume of air several orders of magnitude and you begin to approach the volume and sound of the air being vented at the locomotive.

Never too old to have a happy childhood!

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Tuesday, February 07, 2017 9:22 PM

The exhaust from an emergency application is very short because every car opens a hole to atmosphere in the brake pipe.  A service application vents just one opening in the brake pipe, and takes a lot more time. 

  • Member since
    July, 2008
  • 100 posts
Posted by sandiego on Wednesday, February 08, 2017 12:11 AM

Exhaust from brake valve is no longer vented into cab; there are pipes from the brake valves (both the regular engineer's valve and the emergency brake valve on the conductor's side) to the area under the cab floor where the air brake equipment sits.

On the newer locomotives with electronic air brakes I'm not sure if the air is even vented through the engineer's brake valve or is instead vented by a magnet valve under the cab (controlled by air brake computer). There is little noise in the cab in any case as I recall. It's been 12 years since I was running so some of the active engineers out there would have better information.

Kurt Hayek

 

  • Member since
    December, 2001
  • From: Northern New York
  • 17,043 posts
Posted by tree68 on Wednesday, February 08, 2017 7:28 AM

I'm sure the crew of the derailed train was trying to figure out what was going on, not realizing they had a car now fouling the other track.  

As noted, cabs are a lot quieter than they used to be.  The chief indication of the emergency application would have been the brake line guage and the PCS indicator coming on.  

Even if they had declared an emergency the second the air started to drop, the end result was writ in stone.  The key train was going to hit that derailed car.

From what I could see in the video, only one car fouled the other track.  Had the car gone the other way, this incident would have been written off as a close call.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Wednesday, February 08, 2017 8:18 AM

The point is for a first train to notify an opposing train on the other track as soon as possible after the first train goes into emergency just in case the first train has derailed and fouled the track of the opposing train.  The un-commanded emergency application alone is a call to action to stop the opposing train even without knowing why the first train has gone into emergency.  It is a matter of taking evasive action just because of the possibility of the first train derailing and fouling the track of the opposing train. 

The oil train lost 32 seconds of warning about the emergency application of the grain train.  After the warning came, the oil train consumed another 27 seconds prior to impact.  So had they been warned as early as possible, the oil train had 59 seconds to stop before impact. 

Granted, the oil train may have been traveling faster than the grain train, but the grain train was able to stop only 38 seconds after going into emergency.  So an extra 32 seconds of lost warning for the oil train would have been of great help in reducing the collision impact. 

  • Member since
    March, 2002
  • 9,116 posts
Posted by edblysard on Wednesday, February 08, 2017 9:12 AM

And here we.....go!

23 17 46 11

  • Member since
    April, 2016
  • 376 posts
Posted by Shadow the Cats owner on Wednesday, February 08, 2017 9:55 AM

I just watched the videos in the NTSB report the grain train could NOT broadcast he was in Emergency due to the Signal Maintainer being on the same channel telling him he saw a bunch of debris from his train on the ground.  It is kinda hard to tell someone to look out when the radio is blocked from use by another person.  The engineer wanted to tell his dispatcher he had an UDE however the Signalman blocked him from doing so. 

  • Member since
    December, 2007
  • From: Southeast Michigan
  • 2,686 posts
Posted by Norm48327 on Wednesday, February 08, 2017 12:16 PM

Euclid

The point is for a first train to notify an opposing train on the other track as soon as possible after the first train goes into emergency just in case the first train has derailed and fouled the track of the opposing train.  The un-commanded emergency application alone is a call to action to stop the opposing train even without knowing why the first train has gone into emergency.  It is a matter of taking evasive action just because of the possibility of the first train derailing and fouling the track of the opposing train. 

The oil train lost 32 seconds of warning about the emergency application of the grain train.  After the warning came, the oil train consumed another 27 seconds prior to impact.  So had they been warned as early as possible, the oil train had 59 seconds to stop before impact. 

Granted, the oil train may have been traveling faster than the grain train, but the grain train was able to stop only 38 seconds after going into emergency.  So an extra 32 seconds of lost warning for the oil train would have been of great help in reducing the collision impact.

Let's analyze this from the point of view of the grain train Engineer and an objective observer: Train goes into emergency. The first logical reaction is to notify the dispatcher, which, I believe, is in compliance with the rules and best practices.

He keys up his mic and advises the dispatcher but his transmission is blocked by a signal maintainer's transmission telling him there is debris from his train on the track. The engineer doesn't know the dispatcher did not receive his message. Nor, unless the oil train is within visual range, does he know how long it will take before the oil train will encounter the derailed car if indeed he knows it is fouling the eastward track, which he likely doesn't.

Radio conversation at that point is likely to be heavy considering the dispatcher wants to know the particulars of the situation and his transmissions may block any transmission the grain train is making to advise the oil train.

There is a degree of panic that will set in for those who are not trained to observe and evaluate the situation like airline pilots are. They will seldom go into the panic mode due to their training even though they know in their minds the situation is dire. The engineer of the grain train has done his job in good faith and according to the rule book and best practices but circumstances have negated his best efforts to avoid catastrophy.

It falls under the category of "best laid plans of mice and men". Stuff happens that we have no control of despite our best efforts. All that could have been done to prevent the situation appears to have been done but, as fate would have it, nothing worked.

 

Norm


  • Member since
    January, 2002
  • From: Pennsyland.
  • 5,698 posts
Posted by zugmann on Wednesday, February 08, 2017 12:40 PM

Euclid
Granted, the oil train may have been traveling faster than the grain train, but the grain train was able to stop only 38 seconds after going into emergency. So an extra 32 seconds of lost warning for the oil train would have been of great help in reducing the collision impact.

You can play the what-if game until the cows come home.  But we don't live in a perfect world, so the endless word games we play on this forum aren't going to mean jack.   Best we can do is learn from it and move on. 

The opinions expressed here represent my own and not those of my employer or any other railroad, company, or person.

And why does the truth seem too hard to be true?

  • Member since
    May, 2003
  • From: US
  • 11,944 posts
Posted by BaltACD on Wednesday, February 08, 2017 12:42 PM

Couple of notes - Trains are required to monitor the Road Radio Channel for the territory they are operating on.  ie. Both the Grain train and the Oil train should have been monitoring the same channel - if they could hear each other from the first words of any converstation is unknown.

If the Oil train had begun stopping at the time the video identifies it (and I could not see it in the video at that point in time), it may have been able to stop prior to impacting the derailed car - or it may not have.  The Oil train according to the video was 'sighted' 15 seconds before the derailment was indicated as having occured and it took another 17 seconds for the UDE to happen - by which time the Oil train was virtually 'on top' of the Grain train.

Never too old to have a happy childhood!

  • Member since
    December, 2007
  • From: Southeast Michigan
  • 2,686 posts
Posted by Norm48327 on Wednesday, February 08, 2017 12:42 PM

zugmann
You can play the what-if game until the cows come home. But we don't live in a perfect world, so the endless word games we play on this forum aren't going to mean jack. Best we can do is learn from it and move on.

Thumbs Up

Norm


  • Member since
    December, 2007
  • From: Southeast Michigan
  • 2,686 posts
Posted by Norm48327 on Wednesday, February 08, 2017 12:46 PM

BaltACD

Couple of notes - Trains are required to monitor the Road Radio Channel for the territory they are operating on.  ie. Both the Grain train and the Oil train should have been monitoring the same channel - if they could hear each other from the first words of any converstation is unknown.

If the Oil train had begun stopping at the time the video identifies it (and I could not see it in the video at that point in time), it may have been able to stop prior to impacting the derailed car - or it may not have.

 

"If" is a word that conotates other possibilities. Prettty useless in an emergency.

Norm


  • Member since
    January, 2002
  • From: Pennsyland.
  • 5,698 posts
Posted by zugmann on Wednesday, February 08, 2017 12:52 PM

(nevermind)

The opinions expressed here represent my own and not those of my employer or any other railroad, company, or person.

And why does the truth seem too hard to be true?

  • Member since
    May, 2003
  • From: US
  • 11,944 posts
Posted by BaltACD on Wednesday, February 08, 2017 12:55 PM

Norm48327
BaltACD

Couple of notes - Trains are required to monitor the Road Radio Channel for the territory they are operating on.  ie. Both the Grain train and the Oil train should have been monitoring the same channel - if they could hear each other from the first words of any converstation is unknown.

If the Oil train had begun stopping at the time the video identifies it (and I could not see it in the video at that point in time), it may have been able to stop prior to impacting the derailed car - or it may not have.

"If" is a word that conotates other possibilities. Prettty useless in an emergency.

The point I was making was that the Oil train would have had to start stopping BEFORE the Grain train derailed, let alone before the Grain train crew could have 'suspected' they were derailed because of the UDE; and even then they may have struck the derailed grain car.

Never too old to have a happy childhood!

  • Member since
    June, 2001
  • From: Lombard (west of Chicago), Illinois
  • 12,883 posts
Posted by CShaveRR on Wednesday, February 08, 2017 2:49 PM

It's a pity that there was no sound from the cab video itself.  Some of those cameras do have sound (I never worked with them, but the oft-shared video of the UP train encountering a tornado had full sound from the unoccupied cab).  We could have heard when the emergency application occurred (if no sound from the application itself the PC valve would have shut the throttle).

The engineer of the grain train did not comply with radio procedures in the event of an emergency application.  He should have been saying "Emergency, emergency, emergency!" as soon as the train dynamited.  He had the time to get that out before the maintainer called him...or, if he was still grasping the situation at that point, have said the "Emergency" code instead of acknowledging the mantainer (he knew he was in emergency by then).  The engineer of the oil train said "Emergency, emergency, emergency", in a tone that suggests that he was chiding the engineer of the grain train for not saying it.  The quiet tone also suggests that he could not see the mayhem ahead of him at that point.

We don't have any idea of the speed, or time of brake application of the oil train, at least not from the video.  I'm with whoever said that prompt notification of the emergency might have allowed the oil train to slow down significantly, perhaps reducing the extent of the damage.

I'm sure that all of this information that we weren't privy to was available to the NTSB in the course of their investigation.

Carl

Railroader Emeritus (practiced railroading for 46 years--and in 2010 I finally got it right!)

CAACSCOCOM--I don't want to behave improperly, so I just won't behave at all. (SM)

  • Member since
    December, 2001
  • From: Winnipeg, Mb
  • 479 posts
Posted by traisessive1 on Wednesday, February 08, 2017 3:29 PM

This is why CN in Canada does it right. We tone and talk to the dispatcher on a serparate channel. 

We would announce the emergency broadcast on the standby channel, go over to the RTC channel and do the emergency tone. If the RTC answers right away, the rtc will then protect other trains that may not have heard our initial call. 

If the RTC doesn't answer we have to repeat the emergency call on the standby until the RTC answers. 

If for whatever reason the RTC doesn't answer at all, you go flagging.

Our rules state that you must stop if you hear a train call emergency on a track beside yours. We also have to slow down and be prepared to stop when passing a train that tripped a dragging equipment alarm. 

10000 feet and no dynamics? Today is going to be a good day ... 

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Wednesday, February 08, 2017 3:36 PM

 

CShaveRR
We don't have any idea of the speed, or time of brake application of the oil train, at least not from the video.  I'm with whoever said that prompt notification of the emergency might have allowed the oil train to slow down significantly, perhaps reducing the extent of the damage.

 

Carl,

 

I agree with your views on this.

 

The oil train struck the grain car 59 seconds after the grain train went into emergency, giving the first indication of a possible fouling of the track the oil train was on.   So, the earliest possible warning for the oil train was 59 seconds before impact.

 

However, 32 of those 59 seconds were lost by the delay in the oil train receiving the warning about the emergency application of the grain train.   So the oil train was left with only a 27-second warning of those original 59 seconds starting with the emergency application of the grain train.

 

I don’t know how existing rules pertain to this kind is situation, but there is certainly rule precedent for a requirement to immediately protect adjacent tracks any time a train goes into emergency.  As you know, the reason is that any time a train experiences an emergency application; there is a fair chance that it has been caused by a derailment.  If there has been a derailment and there are adjacent tracks, there is also a possibility of the derailment fouling an adjacent track, posing a risk to trains on the adjacent track.

  • Member since
    December, 2001
  • From: Northern New York
  • 17,043 posts
Posted by tree68 on Wednesday, February 08, 2017 4:12 PM

CShaveRR
I'm sure that all of this information that we weren't privy to was available to the NTSB in the course of their investigation.

I'd like to see the transcribed timeline, including radio transmissions and the "tale of the tape" from the event recorder...  Sound from the cab of the grain train would be handy,  too, as I doubt the two crew members were silent throughout.

From the linked report:

[quote]The NTSB investigation found that after 13 cars from the westbound grain train derailed, the train’s emergency brakes were applied. At that point, the eastbound oil train was 18 seconds away, traveling at 42 mph. The oil train was likely moving at about that speed when it hit the grain car lying across the track, the NTSB said.[/unquote]

I would opine that the crew of the grain train didn't know they'd derailed (ie, everything was still upright) until the derailed car apparently hit the switch (as reported by the signalman), at which time it went cattywhampus, breaking the trainline (initiating the emergency application) and fouling the adjacent track.  

The linked report itself says that at the time of the emergency application, the oil train was a mere 18 seconds away, doing 42 MPH.  The collision was going to occur. Period.  

Give the crew of the grain train several seconds to recognize the brake pipe was headed for zero, plus time to reach for the radio and make the emergency broadcast.  There is no way that broadcast would have changed the outcome, even if the oil train had dumped their air.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    December, 2007
  • From: Southeast Michigan
  • 2,686 posts
Posted by Norm48327 on Wednesday, February 08, 2017 4:44 PM

BaltACD
The point I was making was that the Oil train would have had to start stopping BEFORE the Grain train derailed, let alone before the Grain train crew could have 'suspected' they were derailed because of the UDE; and even then they may have struck the derailed grain car.

One can be reasonably certain the grain train crew had no idea a car had derailed and certain it was fouling the eastbound track. Trains go into emergency on a regular basis; Simple as a brake hose connection parting. "Some days are diamonds, some days are stones.". Emergency brake actions happen on a regular basis, and most of them are benign occurrences, just an inconvenience for dispatchers and conductors who have to walk the  train to find the problem.

Norm


  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Wednesday, February 08, 2017 4:54 PM

[quote user="tree68"]

 

From the linked report:

The NTSB investigation found that after 13 cars from the westbound grain train derailed, the train’s emergency brakes were applied. At that point, the eastbound oil train was 18 seconds away, traveling at 42 mph. The oil train was likely moving at about that speed when it hit the grain car lying across the track, the NTSB said.[/unquote]

I would opine that the crew of the grain train didn't know they'd derailed (ie, everything was still upright) until the derailed car apparently hit the switch (as reported by the signalman), at which time it went cattywhampus, breaking the trainline (initiating the emergency application) and fouling the adjacent track.  

The linked report itself says that at the time of the emergency application, the oil train was a mere 18 seconds away, doing 42 MPH.  The collision was going to occur. Period.  

Give the crew of the grain train several seconds to recognize the brake pipe was headed for zero, plus time to reach for the radio and make the emergency broadcast.  There is no way that broadcast would have changed the outcome, even if the oil train had dumped their air.

 

 

That timing that you quote from the report does not match what the video shows.  Clearly the video captions 32 seconds from the point the grain train went into emergency to the point where its engineer announced that on the radio.  The video continues counting time after the grain train stops and continues counting into the sequence shown from the oil train.  So after the engineer anounced being in emergency, the video counted out 27 more seconds to impact.  So the total time from the emergency application of the grain train to the point of impact for the oil train was 59 seconds.  I don't know where they get 18 seconds as the report states.

There may have been time to stop the oil train with the 59 seconds avaiable.  The grain train stopped in 32 seconds and it was only about 10 mph less speed that the oil train when the braking began. 

 

 

  • Member since
    December, 2001
  • From: Northern New York
  • 17,043 posts
Posted by tree68 on Wednesday, February 08, 2017 4:57 PM

Euclid
That timing that you quote from the report does not match what the video shows. 

So clearly, one or the other is wrong.  

That's why I'd like to see the official transcript/timeline.

Were the radio transmissions simply dubbed into the video?

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    April, 2016
  • 376 posts
Posted by Shadow the Cats owner on Wednesday, February 08, 2017 5:20 PM

I looked at the first video from 2013 derailment is at the start of the video Emergency brake is not even applied until 17 seconds after the derailment.  The collison is less than 20 seconds after that.  The slack would still have been running in on the oil train when they hit. 

  • Member since
    May, 2005
  • From: S.E. South Dakota
  • 11,253 posts
Posted by Murphy Siding on Wednesday, February 08, 2017 6:04 PM

tree68

 

 
Euclid
That timing that you quote from the report does not match what the video shows. 

 

So clearly, one or the other is wrong.  

That's why I'd like to see the official transcript/timeline.

Were the radio transmissions simply dubbed into the video?

 

Y.ikes! Surprise I just had a scary vision involving stopwatches and the Zapruder film!

Thanks to Chris / CopCarSS for my avatar.

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Wednesday, February 08, 2017 6:10 PM

DERAILMENT VIDEO EVENT SEQUENCE TIMES

 

Grain train derails and drags without parting the air hoses.

Derailment commences at time stamp: 1 second.

 

 

Grain train Derailed-dragging condition evolves to point of parting air hoses or damage opening the brake pipe and causes an Emergency application of brakes.

Emergency application occurs at time stamp: 16 seconds.

 

 

With Emergency braking applied to grain train, while train slows, engineer announces “train in emergency.”  

Announcement occurs at time stamp: 48 seconds.

 

 

Grain train stop occurs.

Stop occurs at time stamp:  54 seconds.

 

 

Video shifts from grain train to oil train.

Shift occurs at time stamp:  65 seconds.

 

 

Oil train initiates Emergency application of brakes.

Emergency application occurs at time stamp:  66 seconds.

 

 

Oil train impacts grain car.

Impact occurs at time stamp:  75 seconds 

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Thursday, February 09, 2017 8:49 AM

Shadow the Cats owner

I looked at the first video from 2013 derailment is at the start of the video Emergency brake is not even applied until 17 seconds after the derailment.  The collison is less than 20 seconds after that.  The slack would still have been running in on the oil train when they hit. 

The collision was not 20 seconds after the emergency application.

According to the video time sequence, the time span from the emergency application of brakes to the point of impact was 59 seconds.  This time span is also confirmed in the NTSB report called Event and On-board Image Recorders. 

  • Member since
    May, 2005
  • From: S.E. South Dakota
  • 11,253 posts
Posted by Murphy Siding on Thursday, February 09, 2017 10:46 AM

Euclid
 
Shadow the Cats owner

I looked at the first video from 2013 derailment is at the start of the video Emergency brake is not even applied until 17 seconds after the derailment.  The collison is less than 20 seconds after that.  The slack would still have been running in on the oil train when they hit. 

 

The collision was not 20 seconds after the emergency application.

According to the video time sequence, the time span from the emergency application of brakes to the point of impact was 59 seconds.  This time span is also confirmed in the NTSB report called Event and On-board Image Recorders. 

 

What's your point?

Thanks to Chris / CopCarSS for my avatar.

  • Member since
    January, 2014
  • 3,749 posts
Posted by Euclid on Friday, February 10, 2017 3:50 PM

This is the interview with the engineer of the oil train.  It is quite remarkable to hear his description of colliding with the grain car which he refers to as a “C-6.”

Open this PDF and scroll to page 13 for the start of the engineer’s statement about the collision, and the thoughts and conversation with the conductor as they rode out the crash and then tried to get away from the danger of the fire.  It captures the emotion of such an experience.  On page 16, he said he considered jumping prior to impact, but realized that all those tank cars would come piling in on them as soon as they hit.  He also said he knew that they were not going to just blow through that grain car and keep going.   

https://dms.ntsb.gov/pubdms/search/document.cfm?docID=425456&docketID=55926&mkey=88606

 

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Trains free email newsletter
NEWS » PHOTOS » VIDEOS » HOT TOPICS & MORE
GET OUR WEEKLY NEWSLETTER DELIVERED TO YOUR INBOX
Connect with us
ON FACEBOOK AND TWITTER

Search the Community