Trains.com

Final TSB Report on Lac Megantic Wreck

29784 views
239 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 21, 2014 10:00 AM

NP Eddie

As a retired professional railroader, I must ask this question: What does then president Ed Burkhart have to say about this?  The culture of safety begins from the president on down.

 

Ed Burns

Retired

Shortly after the disaster, Mr. Burkhardt said that the train ran away because the air brakes leaked off.  So he must have erroneously believed that air brakes are an essential part of the securement.     

 

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Thursday, August 21, 2014 10:07 AM

My original comments regarding one of schlimm's posts have been redacted, as has the original terminology that provoked them.  The issue should have been 'taken private' early on, and this has now been done.  Consider the situation a misunderstanding that has been amicably resolved.  Sorry for anyone who now has wasted popcorn money  on further comments in public.

As a peripheral note, I would like to see us move on and discuss the TRB report and its various implications without any judgment or speculation that goes beyond its findings.

  • Member since
    October 2001
  • From: US
  • 591 posts
Posted by petitnj on Thursday, August 21, 2014 10:18 AM

One of the report's statements is "railroads require brakes be fully released before application of the hand brakes". I have never heard of that. In fact, that is shown to be a very bad idea by the study in the report that shows hand brakes are more effective when applied during an air brake application. I can often take another turn or pull on the handbrake after the locomotive separates and the train air brakes go into emergency. 

On freight cars doesn't the hand brake do the same thing as the air brake cylinder?  

The locomotive is a different case. The air brake pushes on the bottom of brake show lever and the hand brake chain on the top. If the air brake cylinder is "out" or pressurized, the chain will tighten but loosen when the air drains out of the cylinder. (That seems like a formula for disaster -- the brakes feel tight but they are not!). 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 21, 2014 10:29 AM

dehusman

Euclid

From what I have read, I gather that no push-pull test was performed.  If an engineer were very well experienced, and reliable, why would he ignore the requirements of Rule 112? 

There was a push pull test performed.

Read section 1.1, page two, top paragraph.

The LE performed a hand brake effectiveness test (whether he performed it well or correctly, he did perform a test, and the train passed the test he did perform).

I don’t understand how the engineer or his supervisors did not immediately recognize the peril of leaving the entire locomotive consist shut-down. 

What is the peril?  I don't understand why there is a peril.

They all know air brakes.  They know that at least one engine running to pump air is necessary to preventing set air brakes from leaking off and releasing.  They all knew that the only running locomotive unit had been shut down during the fire department phase.  They should have known that the train would run away.

Since he performed the securement test with independent brakes set, would that not be a faulty test?  I do not see how observing that the train does not move with automatics released and the independent set can prove that sufficient hand brakes are set.  It only proves that the train is secured with a combination of hand brakes and independent brakes.

It seems to me that the engineer and his supervisors who knew that the only running engine had been shut down would also know that there was a high probability that the independent brakes would release.  And the engineer would certainly know that the train would lose a large portion of its securement if the independent brakes released. 

The peril I referred to is the peril of leaving a train secured by relying on air brakes, which are likely to release if the air pumping ceases due to an engine shut down.  It is extreme peril and it ultimately played out as a disaster.  And it is extreme peril that should have been obvious to the engineer.  The peril should have also been obvious to the engineer’s supervisors if they were accurately informed about how the train was secured and tested. 

Actually, the peril should have been obvious to the engineer even before the fire department was called and the engine was shut down.  He knew that the one running locomotive was having serious mechanical problems with its engine, thus posing a significant likelihood that it might stop running.

 

  • Member since
    March 2002
  • 9,265 posts
Posted by edblysard on Thursday, August 21, 2014 10:57 AM

“At around 2250, the train arrived at Nantes, was brought to a stop using the automatic brakes, and was parked for the night on a descending grade on the main track. The LE applied the independent brakes to the locomotive consist. He then began to apply the hand brakes on the locomotive consist and the buffer car (7 cars in total), and shut down the 4 trailing locomotives. Subsequently, the LE released the automatic brakes and conducted a hand brake effectiveness test without releasing the locomotive independent brakes. The LE then contacted the rail traffic controller (RTC) responsible for train movements between Farnham and Megantic Station (Megantic), who was located in MMA’s yard office in Farnham, to indicate that the train was secured.”

A little clarification for you Bucky…

Note the portions I highlighted of the initial summary…the locomotives air brakes (independent brakes) were left applied, which means when the locomotive that was running was shut off, the air in the independent brake system began to leak down, slowly equalizing and releasing the locomotive air brakes, which apply to all the wheels on the locomotive.

Once that retarding force was gone, the only brakes holding the train in place were the handbrakes on the locomotives, which only apply to one axel, and the few handbrakes on the cars.

Mr. Burkhardt’s statement is partially correct in that shutting off the locomotive is the physical event that started the process of the runaway.

I am pretty sure that by the time he made that statement, he was or had been made aware that the independent brakes on the locomotives had been used, incorrectly, to help secure the train, and I am certain he knew that the rules required that “sufficient” handbrakes on the cars alone is the requirement for proper securement.

 From an operational perspective, this disaster started long before that when the management failed to teach its employees the “how and why” of the securement rules and failed to test and see that the rules where complied with.

 

23 17 46 11

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 21, 2014 11:29 AM

edblysard
A little clarification for you Bucky…

Note the portions I highlighted of the initial summary…the locomotives air brakes (independent brakes) were left applied, which means when the locomotive that was running was shut off, the air in the independent brake system began to leak down, slowly equalizing and releasing the locomotive air brakes, which apply to all the wheels on the locomotive.

Once that retarding force was gone, the only brakes holding the train in place were the handbrakes on the locomotives, which only apply to one axel, and the few handbrakes on the cars.

Mr. Burkhardt’s statement is partially correct in that shutting off the locomotive is the physical event that started the process of the runaway.

 

What I highlighted in blue has been my exact understanding all along here. 

Regarding the comment by Burkhardt that I referred to:

I understand your point that he was stating a true fact when he said that the brakes leaked off.  My point is that that statement of fact reveals a misunderstanding of the securement rules or the acceptance of willful violation of the rules.  I am sure he believed what he was saying since he was using it to make a very critical point about the fire department causing the disaster. 

Now we learn that the engineer did indeed execute the securement according to the same misunderstanding or willful violation of the rules that was indicated by the President Burkhardt.  I believe that Burkhardt’s comments about the air leaking off and causing the runaway reveals the likelihood of a company culture that routinely compromised the securement rules.  The engineer executing exactly the faulty securement that Burkhardt described seems to confirm the possibility of a culture of securement compromise. 

  • Member since
    August 2008
  • From: Calgary AB. Canada
  • 2,298 posts
Posted by AgentKid on Thursday, August 21, 2014 2:33 PM

NP Eddie
What does then president Ed Burkhart have to say about this?

Mr. Burns, Ed Burkhart's reaction to the TSB report was reported in the following story from Tuesday's TRAINS Newswire:

http://trn.trains.com/en/Railroad%20News/News%20Wire/2014/08/Burkhardt%20responds%20to%20Lac%20M%C3%A9gantic%20report.aspx

Bruce

So shovel the coal, let this rattler roll.

"A Train is a Place Going Somewhere"  CP Rail Public Timetable

"O. S. Irricana"

. . . __ . ______

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Thursday, August 21, 2014 3:32 PM

Euclid

Statement 1:  From what I have read, I gather that no push-pull test was performed. 

Statement 2:   Since he performed the securement test with independent brakes set, would that not be a faulty test? 

Not performing a test and performing a test incorrectly are two different things and have two different causes and may have two different solutions.

 

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 21, 2014 3:59 PM

Yes, I understand that. My statement 1 was just my assumption ahead of the report, and I had not yet read it in the report. I agree that he performed a test, but it was a faulty test. So I assume that the test would not qualify as a test according to the rules.  So he did not perform a proper test.

But my main point was to explain that the peril of relying on air brakes to secure the train should have been obvious to the engineer if he was as reliable as people have said.  It should also have been obvious to the engineer's supervisors if they knew how the train had been secured. 

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Thursday, August 21, 2014 4:39 PM

Euclid

We have had a lot of discussion about the role of the automatic brake and how it might have been inadvertently released.  It is interesting to learn now that when the engineer left the train standing at Nantes, he left the automatic air brake fully released. 

 I agree.  I was one of those who assumed from the media reports that the train air had been applied, and I couldn't understand how the train brakes could have leaked off so quickly absent a malfunction or some other event that actually caused a release.    But the report makes clear that the train air wasn't even applied - only the independent air in the loco consist was applied.  That explains how it could have leaked off so quickly.  The report actually says at one point that if the train air had been applied, it probably would not have leaked off before morning.

Just shows how important it is to await authoritative information tbefore making any definitive conclusions. 

  • Member since
    October 2008
  • From: Calgary
  • 2,047 posts
Posted by cx500 on Thursday, August 21, 2014 5:49 PM

The Toronto Globe & Mail has published a transcript of the conversations that night between the engineer and dispatcher.  It gives some insight into the human elements in play.

http://www.trainorders.com/discussion/warning.php?forum_id=15&url=http%3A%2F%2Fwww.theglobeandmail.com%2Fnews%2Fnational%2Fdispatches-from-a-disaster%2Farticle20148699%2F%23dashboard%2Ffollows%2F

John

  • Member since
    March 2012
  • 493 posts
Posted by DwightBranch on Thursday, August 21, 2014 6:24 PM

Edit: issue resolved.

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 21, 2014 7:14 PM

If the engineer had done the proper securement test, we know that the test would have failed.  We know this just by the fact that the train ran away when the securement was reduced to only handbrakes; while the rules call for 100% securement by handbrakes.  The proper test would have indicated that the engineer had not set enough handbrakes.  So then he would have to set the independent brakes to hold the train, and go out and wind up some more handbrakes.  Then he would have to test that by releasing the independent brakes to see if the train held in place or rolled.     

After he had the train 100% secured by handbrakes, he could leave the independent and/or automatic brakes set or released.  Those air brake functions are not relevant to holding a train secured from rolling away.

One thing to consider is that these sort of lackadaisical securement methods probably are done very often, and most of the time, they work just fine.  Trains have to be secured just when parked on level ground.  Even in many hilly locations, a runaway would not cause a disaster.  

But the tie-up at Nantes perched a train above Lac Megantic.  And that town had the track curving to the extent that the speed was much limited.  Anyone would know that a runaway originating from Nantes would be moving way too fast for the Lac Megantic curve before it got there.   

I think this is where the ball got dropped.  Tying up oil trains at Nantes with a “bailing wire” approach to securement was a death sentence.  Somebody should have seen it coming.

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Thursday, August 21, 2014 8:53 PM

Euclid

 

I think this is where the ball got dropped.  Tying up oil trains at Nantes with a “bailing wire” approach to securement was a death sentence.  Somebody should have seen it coming.

a

In the real world - Hindsight is 20/20 - Foresight is 20/400 or worse.

Never too old to have a happy childhood!

              

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Thursday, August 21, 2014 8:58 PM

jeffhergert

While I know how air brakes work, I guess I don't know the more technical details as well as I thought.

The report said the brake pipe reduced at an average rate of 1 psi/minute and the cars didn't set up.  (According to the time line from the event recorder, the brake pipe dropped 15 psi in 8 minutes for a rate of 1.875 psi/min.)  I know it only takes a 1 to 1.5 psi difference in pressure for the control valve to operate, but evidently it has to happen at a certain psi/minute rate.  So what is the minimum psi/min rate it needs to "see" something is happening?  What is the rate of a service reduction, say a 6 to 8 psi minimum initial set as made by the engr's brake valve?

Trying to find a good diagram of the control valve (informally aka, triple valve) during various phases has been hard.  Most are very simplistic.  I was always under the impression once charged the control valve was also in a lap position, no air moving through it like it is after a brake application.  However, I've seen one diagram that shows a feed port still allowing air from the brake pipe to the auxiliary reservoir after fully charged.  If this is so, does that mean that if the brake pipe reduces at 1 psi/min, the auxiliary reservoir also reduces at the same rate backwards through that port?  If that's the case, then I could see there would be no pressure differential between the two sides of the control valve.  It couldn't "see" a drop in brake pipe pressure no matter how much the brake pipe drops because the aux res also has been dropping.

I've read reference somewhere else from Canada where after fully securing the train with hand brakes, they would leave the automatic in release.  In theory, if secured it shouldn't be a problem.  (After all, you definitely aren't depending on air brakes to hold the train.) The problem comes when you get a false positive on the securement test.  You think you have enough hand brakes set, but you don't.  

Had he set 20 psi, it probably wouldn't have run away after the brake pipe leaked off.   At least until the air leaked off individual brake cylinders and that could take a few days, if not longer.  Of course that's through my understanding of how things work and I've been wrong before   

Jeff

 

  The report actually mentions that, had the train brakes been set, they probably would have remained set until at least the next morning (when, presumably, the relief crew would have arrived).  But air brakes (whether independent or train) shouldn't be use to secure a train, because there are any number of things that can happen to cause them to unintentionally either bleed off or release.

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Thursday, August 21, 2014 9:15 PM

dehusman

Euclid

From what I have read, I gather that no push-pull test was performed.  If an engineer were very well experienced, and reliable, why would he ignore the requirements of Rule 112? 

There was a push pull test performed.

Read section 1.1, page two, top paragraph.

The LE performed a hand brake effectiveness test (whether he performed it well or correctly, he did perform a test, and the train passed the test he did perform).

I don’t understand how the engineer or his supervisors did not immediately recognize the peril of leaving the entire locomotive consist shut-down. 

What is the peril?  I don't understand why there is a peril.

They all know air brakes.  They know that at least one engine running to pump air is necessary to preventing set air brakes from leaking off and releasing.  They all knew that the only running locomotive unit had been shut down during the fire department phase.  They should have known that the train would run away.

According to the TSB report the automatic brakes were RELEASED.  Nobody was counting on the engines keeping the train brakes set because the air brakes were released on the train by the engineer when he did the hand brake test.

All of the engines had handbrakes set.

The LE told the RTC that the train had been properly secured.

 


REPLY

  The engineer WAS relying on the air brakes to keep the train secured, although he may not have fully understood this.  True, the TRAIN (air) brakes were released when he performed the push-pull test.  But, according to the report, the independent (air) brakes on the locomotive consist were applied. The train would have failed the test had the independent (air) brakes been released, as shown by the fact that the train ran away when the independent brakes bled off. 

  • Member since
    December 2007
  • 1,307 posts
Posted by Falcon48 on Thursday, August 21, 2014 9:27 PM

Euclid

NP Eddie

As a retired professional railroader, I must ask this question: What does then president Ed Burkhart have to say about this?  The culture of safety begins from the president on down.

 

Ed Burns

Retired

Shortly after the disaster, Mr. Burkhardt said that the train ran away because the air brakes leaked off.  So he must have erroneously believed that air brakes are an essential part of the securement.     

 

  The air brakes shouldn't have been an "essential part" of the securement.  But, based on the report, the way the train was actually secured, they were.  Because of this, the train ran away when the (independent) air brakes "leaked off".  So, Mr. Burkhardt was technically correct.  But the root cause wasn't the "leaking off" of the air brakes.  It was relying on the air brakes to secure the train in the first place.  If it had been secured properly, the "leaking off" of the air brakes would have been a non-event. 

  • Member since
    November 2005
  • From: Hope, AR
  • 2,061 posts
Posted by narig01 on Thursday, August 21, 2014 11:05 PM
By cx500
on Thursday, August 21, 2014
The Toronto Globe & Mail haspublished a transcript of the conversations that night between the engineer and dispatcher. It gives some insight into the human elements in play.

http://www.trainorders.com/discussion/warning.php? forum_id=15&url=http%3A%2F%2Fwww.theglobeandmail.com%2Fnews%2Fnational%2 from-a-disaster%2Farticle20148699%2F%23dashboard%2Ffollows%2F

John

Here is the direct link to the Toronto Globe & Mail

http://m.theglobeandmail.com/news/national/dispatches-from-a-disaster/article20148699/?service=mobile

http://www.theglobeandmail.com/news/national/dispatches-from-a-disaster/article20148699/

If someone could get the links.

Rgds IGN
  • Member since
    October 2008
  • From: Calgary
  • 2,047 posts
Posted by cx500 on Thursday, August 21, 2014 11:51 PM
  • Member since
    September 2013
  • 918 posts
Posted by Kyle on Friday, August 22, 2014 4:43 AM

You also have to blame the LE for leaving a locomotive with an obvious problem (spewing oil) running in the first place.  Wouldn't you leave a mechanically sound locomotive like one of the other five running instead of a faulty one?  I also have to wonder why the train air brakes weren't applied, even if they shouldn't be relied upon to hold the train in place, they would add another layer of protection.  

I also agree that the LE didn't Release the independent brakes when he preformed the test, it would have revealed that not all the hand brakes were set properly, and more were needed.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Friday, August 22, 2014 6:10 AM

narig01
John

Here is the direct link to the Toronto Globe & Mail

Link to Globe and Mail transcript and audio of conversations

The other link IGN provided is optimized for use on mobile devices:

Link optimized for mobile devices


  • Member since
    August 2005
  • 964 posts
Posted by gardendance on Friday, August 22, 2014 8:12 AM

Perhaps this should be in a separate thread, but with all this talk about hand, air, independent and train brakes, I have to wonder what would have happened if they had just placed a wooden chock on each side of one wheel?

Patrick Boylan

Free yacht rides, 27' sailboat, zip code 19114 Delaware River, get great Delair bridge photos from the river. Send me a private message

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Friday, August 22, 2014 8:16 AM

gardendance

Perhaps this should be in a separate thread, but with all this talk about hand, air, independent and train brakes, I have to wonder what would have happened if they had just placed a wooden chock on each side of one wheel?

 
One wheel?
 
You would have a piece of flattened wood.

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    September 2003
  • From: Omaha, NE
  • 10,621 posts
Posted by dehusman on Friday, August 22, 2014 8:20 AM

Kyle

You also have to blame the LE for leaving a locomotive with an obvious problem (spewing oil) running in the first place.  Wouldn't you leave a mechanically sound locomotive like one of the other five running instead of a faulty one?

It is very common for older engines to blow oil out the stack.  Decades ago I worked at a yard with lots of GP7's and GP18's.  I did not own a shirt without oil spots on the shoulders.

Dave H. Painted side goes up. My website : wnbranch.com

  • Member since
    June 2011
  • 1,002 posts
Posted by NP Eddie on Friday, August 22, 2014 9:19 AM

I chuckled about the answer that you would have a piece of flattened wood. I was the clerk to the BNSF's Terminal Supt at Northtown. A newer switchman got excited when chocking a railcar with a piece of wood but used his radio as a chock instead! He brought the radio to my office in pieces!!!. I issued him a new radio and forwarded the deceased radio to our Radio Shop.

 

Ed Burns

Tags: final report
  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,919 posts
Posted by blue streak 1 on Friday, August 22, 2014 3:22 PM

All this talk about securement overlooks one item.

A split rail derailer near the location where the siding came back to the main would have allowed the runaway to run into a berm that will stop a low speed runaway.

The collision of the Colonial into the PC loco consist at Gun pow would have been avoided if Amtrak had not removed the splitt rail there a few years earlier.  Actually it was a full switch that ran into a berm.

SOU RR had runaway tracks on Saluda that would allow a runaway to go up a high slope runaway track.   IMHO split rail derailers need to be anywhere a train may be secured on a slope.  ( amount unknown )

Notice that many industrial spurs now have a regular derailer on mild slopes and split rails on steeper slopes.

  • Member since
    June 2007
  • 228 posts
Posted by RDG467 on Friday, August 22, 2014 3:37 PM

That's a good point , Blue Streak 1, although a better one, which wasn't asked (or answered) in the report is 'Why were the empty cars for Tafisa sitting in the Nantes siding instead of down in the yard?"  If the siding was clear, the train could've pulled into it and had derail protection.....[I believe I read somewhere that there was a derail installed on the siding in front of the cut of cars.] ..instead of being left on the main....

  • Member since
    December 2007
  • From: Southeast Michigan
  • 2,983 posts
Posted by Norm48327 on Friday, August 22, 2014 4:39 PM

We can sit here and carve up the TSB report all day and nothing will come of it. Better to accept the facts as written by TSB and let it drop." What ifs" are just that. They carry no weight.

Norm


  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Friday, August 22, 2014 5:06 PM

blue streak 1

All this talk about securement overlooks one item.

A split rail derailer near the location where the siding came back to the main would have allowed the runaway to run into a berm that will stop a low speed runaway.

Had you read the report - the Train was parked on the Main.  Cars occupied the siding and there was a derail in place on the siding in the pictures that were shown.

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Friday, August 22, 2014 6:13 PM

BaltACD

blue streak 1

All this talk about securement overlooks one item.

A split rail derailer near the location where the siding came back to the main would have allowed the runaway to run into a berm that will stop a low speed runaway.

Had you read the report - the Train was parked on the Main.  Cars occupied the siding and there was a derail in place on the siding in the pictures that were shown.

I think blue streak's basic point is that a derail could have prevented this runaway.  It is true that the train was parked on the mainline, but the mainline could have had a derail.

Join our Community!

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

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy