Derailment and fire in Quebec

[dave1905]

With all this discussion of the use of brakes, I have a (silly?) question for those with practical experience of North American railroads - I'm not asking should or would this be done, but could a train be held on a hill by putting the locomotive(s) into 'plug mode', perhaps against brakes? (For those not familiar with the term, plug mode is putting the locomotive into reverse when the tendency is for it to move forward. It is used in some places to slow down a unit train which is being loaded or unloaded, and the track slopes down out of the loader/unloader).

 

Bad things would happen.

[highpeak]

TSB apparently held a press conference today, though didn't have much to say.   Perhaps the most interesting is that in addition to looking at the causes of the crash, they will also be looking for institutional problems at MMA that may have contributed.

http://www.montrealgazette.com/news/montreal/accident+ever+caused+thing/8653105/story.html

I am glad to hear that TSB will go beyond proximate causes for the accident, but frankly I would prefer to see a judicial inquiry of the kind that was done after the Dryden plane crash (which is why I referred to that in my first post). A typical inquiry into that crash would have blamed the pilot for attempting to take off with contaminated flight surfaces, harrumphed about the chain of command and communications and noted the lack of a ground power unit at Dryden. Judge Moshansky found a lot of other things to question including accepted industry practices and regulatory failures. Sound at all familiar?

TSB published a report into a collision in Alberta in January 2012 (reference in post 184) that included the following risk findings:

1. When clear instructions are not provided detailing the manner in which the hand brake effectiveness test must be conducted, there is a risk that insufficient hand brake force will be applied and uncontrolled movements will occur.
2. When sidings are frequently used to set out cars and special derails are not installed, there is an increased risk of uncontrolled movements entering the main track, especially where sidings are located on steep grades.
3. Applying only the minimum number of hand brakes, as set out in Section 7.12 of Canadian National's General Operating Instructions, to cars held on tracks with significant grades may not provide adequate securement.

Canadian National did subsequently install a special derail (meaning it is normally locked clear of the track rather than locked in place on the rail) on the siding. Did MMA read that report? How widely read are these reports within the industry? Were any steps taken to change rules and procedures? Or are we happy to note that in North America we don't put derails on sidings and leave it at that?

 

Chris Leigh and Mike Stationmaster have asked some pertinent questions and countervailing points have been made. One of these issues involved the age and reliability of the motive power. It is certainly true that brand new equipment of any kind can be defective, out of the box. It is also true that old equipment can be as reliable as the newest. But come on, is anybody really going to suggest that some of the roads are meticulously maintaining these engines? They are wrecks on wheels, patched up and sent out for another run. Was the equipment a direct cause of the accident? Was it an indirect cause of the accident?

 

Here's a thought process: how seriously would anybody take a business proposal spelling out exactly what MMA did to take on this business?

 

I realise I am being a bit abrasive and persistent with this, but 50 people have died and a town levelled in an event that in many ways can't be called an accident and is the latest in a depressing sequence of similar events, different because of location but not so much because of causes. For the rail industry's own good, it can't be business as usual.

 

[highpeak]

  I can concieve of a couple instances where shutting down the engines  could, in a curious set of circumstances, cause a train that was "properly secured", to start to roll.  The problem with those scenarios is it would have happened pretty fast after the FD shut down the engine and the FD would have seen the movement. 

Dave, can you elaborate on that?

One thing that emerged early on (in a statement made by the Nantes fire chief in response to MMA's suggestion that shutting down the engine was a cause of the accident) was that the fire chief's understanding of the brake system was flawed. I don't have the link any more, but he seemed to believe that the automatic brake would fail safe.

My main interest in the scenarios you have in mind is whether or not the FD would have been able to do anything to prevent the movement of the train.

[Gerald Henriksen]

 

Looking for blame in Lac-Megantic: http://www.thestar.com/business/2013/07/12/looking_for_blame_in_lacmgantic_olive.html

 

[pH]

Bad things would happen.

Dave, could you expand on this too, please? As I said in my earlier post, I know this is used in loading unit coal trains at certain mines. These can take hours to load, so it's not something that causes problems immediately. Is it something that will cause problems with time? Is the problem that the train would be stopped, whereas in loading, it does move forward slowly?

Edited July 12, 2013 by pH

[Gerald Henriksen]

TSB has released data regarding number of incidents with all railways in Canada: http://bst.gc.ca/eng/stats/rail/r13d0054/r13d0054.asp

[The Stationmaster]

If the engine air brakes were being used to hold the train that becomes a failure of the process.  NONE of the air brakes should be relied upon.  The locomotives should have had handbrakes applied.  The whole point is that every engine could have been shut down and every psi of air bled out of every air system on every car and locomotive and the train should still have stayed put. 

 

On some railroads, the crews are supposed to secure the train with handbrakes, then test the train by RELEASING the air brakes (on cars and engines), to make sure the handbrakes hold the train.

Er yes - I've already explained how that should be done - and I've hinted at 'work arounds' (for want of a better expression) which show how there are potentially ways to avoid going back and making sure the handbrakes are on.  I think the situation was probably very simple - as is so often the case - in that there are ways jobs should be done and ways they are done.  We don't as yet know what actually happened but it could well be that Ed Burkhardt has given us a big clue - inadvertently or not.

 

If the train was held on the loco straight air brakes (to save applying handbrakes) then the brakes would leak off if the engine stopped - I don't know the rate at which US brake cylinders leak off but certainly on some British locos the bogie brakes could start to leak off and rub in about an hour, sometimes longer - it varied from loco to loco.   

 

The simple fact is that if the train was properly secured when it first arrived at Nantes it would not have moved until it was mea not to move unless somebody deliberately tampered with it.  If it was properly secured with handbrakes why would an engine be left running - all the US roads are very canny when it comes to fuel consumption so why leave an engine running for no apparent reason? (unless it was possibly to avoid having to 'pump up' the air when the train needed to move?).

 

What we do know is that the train was stationary when the Engineer left it; it was stationary when the fire crew arrived and it was still stationary when they left. So clearly something happened after the firecrew left which reduced the amount of brake force sufficiently to allow the train to start moving and/or overcome what other brake force was present - as yet we don't know what that was although, as I've said, Burkhardt gave more than a hint when pressed by the media (not that he was necessarily correct of course).

[Gerald Henriksen]

7 more people identified, 4 more bodies found.  Totals so far - 28 confirmed dead, 22 assumed dead.

 

Province of Quebec has not ruled out holding an inquiry

 

Gas venting from the soil is making some areas inaccessible to the police.

 

http://www.cbc.ca/news/canada/montreal/story/2013/07/12/lac-megantic-quebec-train-explosion-investigation.html

[pH]

TSB has released data regarding number of incidents with all railways in Canada: http://bst.gc.ca/eng/stats/rail/r13d0054/r13d0054.asp

I can't see figures per million train miles in there, but I did hear them quoted in a CBC news report. (I was in the dentist's chair at the time, so the decimals may not be right, but the overall numbers were so amazing that they stuck.) They said CP had 1.6 accidents per million train miles, CN had 2.8, the US national average was 14.6, and MMAs was 36.1. Either the main Canadian railways are way safer than US ones, or different standards are being used in reporting.

[Ozexpatriate]

Looking for blame in Lac-Megantic: http://www.thestar.com/business/2013/07/12/looking_for_blame_in_lacmgantic_olive.html

I sure hope that is an op-ed piece!

[jamie92208]

I wonder if anyone else has read the leading article in this months issue of Trains magazine.  The central thread is that the US/Canadian railroad industry is booming because of revenues from ransporting crude oil.  Apparently a pipleine has had its planning consent refused and the oil will have to continue to be carried by train.  There are some interesting statistics showing how much safer rail is than pipelines in terms of barrels of oil spilled.   One of the major fields that is not connected to the natinal pipleine system is teh North bracken in the Dakotos where teh Lac Magantic train came from.   There is a horrible irony that this terrible crash should occur just when rail is being touted as the safest way to trasnport crude oil.

 

Jamie

[Ozexpatriate]

Apparently a pipleine has had its planning consent refused and the oil will have to continue to be carried by train.

Which one? Keystone XL? (Phases 3 and 4 of Keystone.)

 

Wikipedia has a list of oil spills. Most of these are tanker (ship) or pipeline related.

Edited July 12, 2013 by Ozexpatriate

[jamie92208]

Which one? Keystone XL? (Phases 3 and 4 of Keystone.)

 

Wikipedia has a list of oil spills. Most of these are tanker (ship) or pipeline related.

It was one of the keystone ones.  The article also has the statistics for oil spills which together with problems at refineries separating different batches of product point to rail ahving a significant advantage plus being inherently more flexible in its use of capital and also being able to respond to market changes much quicker.

 

Jamie

[Ozexpatriate]

Regrettably, there was a lot of oil spilled (and of course burned).

 

Using 72 cars at 113,000 litres of crude, and

 

Estimating the oil at 0.827 kg/l, gives us something up to around 6,500 tonnes of crude.

 

This is not Deepwater Horizon or Exxon Valdez territory but it is larger than many North American pipe spills.

Edited July 13, 2013 by Ozexpatriate

[Gerald Henriksen]

Chris Leigh and Mike Stationmaster have asked some pertinent questions and countervailing points have been made. One of these issues involved the age and reliability of the motive power. It is certainly true that brand new equipment of any kind can be defective, out of the box. It is also true that old equipment can be as reliable as the newest. But come on, is anybody really going to suggest that some of the roads are meticulously maintaining these engines? They are wrecks on wheels, patched up and sent out for another run. Was the equipment a direct cause of the accident? Was it an indirect cause of the accident?

Why are they wrecks on wheels? Just because MMA (like some other operators) declined to spend money on giving them a new paint job? Would you therefore be all happy if a railroad ran the exact same equipment but with shiny new paint jobs?

 

My local shortline is running a GP9, which is likely 15 years older than what MMA is running. The class I's are running equipment that is older than what MMA has. The key is they are maintained properly.

 

Would it surprise me if MMA wasn't maintaining their equipment properly? Given some of what I have read in the last week not really. But if they aren't maintaining old equipment, then they also wouldn't be maintaining new equipment.

 

In other words, the issue isn't with the age of the locomotives, or the fact they don't have new paint jobs. The issue is likely company policies/procedures on how stuff is maintained, and what happens to equipment that is showing signs of trouble.

 

Here's a thought process: how seriously would anybody take a business proposal spelling out exactly what MMA did to take on this business?

 

Sadly, likely quite seriously. Face it, a business proposal is all about numbers and whether they make the bankers happy. It has nothing to do with shortcuts or any cost cutting that might be detrimental. There are far too many examples in any number of industries to believe otherwise.

 

As for the anybody, when MMA was created a 12.77% share was purchased by Caisse de dépôt et placement du Québec (aka the Quebec public pension fund).

 

 

I realise I am being a bit abrasive and persistent with this, but 50 people have died and a town levelled in an event that in many ways can't be called an accident and is the latest in a depressing sequence of similar events, different because of location but not so much because of causes. For the rail industry's own good, it can't be business as usual.

A lot of what we call accidents in life really aren't, but we still call them accidents.

 

I am certainly not trying to discard what has happened, nor am I saying that no changes need to be made.  But am also not going to try blaming it on the fact that the locomotive in question was 30 years old.

 

I am also cynical enough to think that by the time the TSB releases its report this will have been forgotten enough that combined with industry lobbying (and not just by the railroads) the politicians won't implement all the changes that are necessary.

 

 

[Gerald Henriksen]

I can't see figures per million train miles in there, but I did hear them quoted in a CBC news report. (I was in the dentist's chair at the time, so the decimals may not be right, but the overall numbers were so amazing that they stuck.) They said CP had 1.6 accidents per million train miles, CN had 2.8, the US national average was 14.6, and MMAs was 36.1. Either the main Canadian railways are way safer than US ones, or different standards are being used in reporting.

 

This article mentions a bunch of the stats, showing that despite everything the railways in Canada are getting safer as the years go by: http://o.canada.com/2013/07/12/despite-calls-for-reviews-canadian-railway-safety-is-getting-better-not-worse/

[dave1905]

Dave, could you expand on this too, please? As I said in my earlier post, I know this is used in loading unit coal trains at certain mines. These can take hours to load, so it's not something that causes problems immediately. Is it something that will cause problems with time? Is the problem that the train would be stopped, whereas in loading, it does move forward slowly?

 

The way you explained this is putting the engine in reverse of the direction its moving.  That means either the wheels turn in reverse (sliding the wheels and grinding them down) or turns them (and the motor) against the way they are being powered creating tremendous heat in the traction motors.  The choice between damaging the motors or damaging the wheels is not a really good choice.

 

Why not just put it in dynamic braking?

 

As far as holding a train by putting it in reverse, that means that the force on the wheels has to balance the weight of the train.  So that means putting enough power into the motors to balance the weight of the train.  Pushing that much power through a motor that isn't turning will cook it off ASAP.

[dave1905]

Er yes - I've already explained how that should be done - and I've hinted at 'work arounds' (for want of a better expression) which show how there are potentially ways to avoid going back and making sure the handbrakes are on.  I think the situation was probably very simple - as is so often the case - in that there are ways jobs should be done and ways they are done.  We don't as yet know what actually happened but it could well be that Ed Burkhardt has given us a big clue - inadvertently or not.

 

If the train was held on the loco straight air brakes (to save applying handbrakes) then the brakes would leak off if the engine stopped - I don't know the rate at which US brake cylinders leak off but certainly on some British locos the bogie brakes could start to leak off and rub in about an hour, sometimes longer - it varied from loco to loco.   

 

The simple fact is that if the train was properly secured when it first arrived at Nantes it would not have moved until it was mea not to move unless somebody deliberately tampered with it.  If it was properly secured with handbrakes why would an engine be left running - all the US roads are very canny when it comes to fuel consumption so why leave an engine running for no apparent reason? (unless it was possibly to avoid having to 'pump up' the air when the train needed to move?).

 

What we do know is that the train was stationary when the Engineer left it; it was stationary when the fire crew arrived and it was still stationary when they left. So clearly something happened after the firecrew left which reduced the amount of brake force sufficiently to allow the train to start moving and/or overcome what other brake force was present - as yet we don't know what that was although, as I've said, Burkhardt gave more than a hint when pressed by the media (not that he was necessarily correct of course).

 

The engines or an engine will be left running to maintain air on the cars.  If the cars do not have air on them for more than 4 hours, a full initial terminal brake test must be performed.  Also on many older engines the batteries may not be reliable enough to start a shut down engine.  In the winter time, engines are kept running because NA diesels do not use anti-freeze so the engines are kept running to keep from dumping the water or freezing the block.

[BR60103]

It would be interesting to watch a brake test performed by a one man crew.

[Guest Belgian]

Forgive me if this has already been covered, but this is already a long thread and I've only a little time today and have been away all week.

 

Reading the last few postings however, there seems to be a common theme that the brakes 'leaked off'. How is this possible? I have always understood that the basic premise of both vacuum and air  brakes was 'fail safe' and that the pressure was applied to release the brakes - any leak causing all brakes to engage. Thus why is there talk of air leakages causing the brakes to to disengage? Have railroads gone mad (for financial reasons or whatever) and reversed the process?

 

JE

Edited July 13, 2013 by Belgian

[Edwin_m]

This has already been covered.  The brakes are failsafe in the sense that they will stop the train if the pipe is vented, but the brakes are held applied by air from the reservoirs and if no air supply is available these will leak off after a period of usually hours. 

[Oldddudders]

Slightly off-topic, but this article makes it clear that oil and gas movements in the US will be increasing in the near future

 

http://cs.trains.com/trn/b/fred-frailey/archive/2012/11/13/what-39-s-bigger-than-coal-in-rail-39-s-future.aspx

 

If the present tragedy prompts some tightening of regulations about oil and gas train-handling, those deaths may not have been entirely in vain, maybe preventing others in the future.

[Glorious NSE]

It would be interesting to watch a brake test performed by a one man crew.

 

Does a 'smart' EOTD allow for that function to be done from the cab?

[Glorious NSE]

TSB published a report into a collision in Alberta in January 2012 (reference in post 184) that included the following risk findings:

1. When clear instructions are not provided detailing the manner in which the hand brake effectiveness test must be conducted, there is a risk that insufficient hand brake force will be applied and uncontrolled movements will occur.
2. When sidings are frequently used to set out cars and special derails are not installed, there is an increased risk of uncontrolled movements entering the main track, especially where sidings are located on steep grades.
3. Applying only the minimum number of hand brakes, as set out in Section 7.12 of Canadian National's General Operating Instructions, to cars held on tracks with significant grades may not provide adequate securement.

Canadian National did subsequently install a special derail (meaning it is normally locked clear of the track rather than locked in place on the rail) on the siding. Did MMA read that report? How widely read are these reports within the industry? Were any steps taken to change rules and procedures? Or are we happy to note that in North America we don't put derails on sidings and leave it at that?

 

Lots of 'devils advocating' in the following, but:

 

Assuming they did read that report, would they have automatically linked what they were doing at Nantes to the circumstances there? The CN was setting out freightcars on a passing siding, the MMA was parking a whole train overnight on their main line. Even if they had suddenly jumped up and overnight fitted Nantes siding (or all their sidings!) with a derail then the runaway would still have occurred as the train wasn't parked in Nantes siding...

 

I'm not saying this to put the point of view that this is 'a good thing' - but if the intention of that CN accident report was to make a fundamental change in the way things are done then it needed to firstly mandate that change (or at least, mandate the railroad assesses the risks related) - and secondly put some kind of usage context.

 

Derails/Traps at Nantes would have prevented this **only if** the railroad had made a conscious decision to park that train in the siding behind their protection, and not on the main outside it - effectively it would become a specific place designed/designated for tying down parked trains.

 

Similarly, the rulebook covers testing the handbrakes effectiveness on a tied-down train before you walk off and leave it (and says 'add more handbrakes and test it again' if it fails that test,) I would be gobsmacked if MMAs rules turn out to be fundamentally different from the industry standard in that - so the issue should not be lack of instruction, or of industry minimum numbers of brakes, but whether the rules were known, and followed.

 

Don't get one-tracked on this either, there's plenty of other things that could have prevented this from happening if the railroad had assessed the risks differently, one obvious, simple one to my mind would have been to have made a conscious decision not to have left this particular train with it's substantially greater risk to sit there unattended in the first place...

 

 

 

But come on, is anybody really going to suggest that some of the roads are meticulously maintaining these engines? They are wrecks on wheels, patched up and sent out for another run. Was the equipment a direct cause of the accident? Was it an indirect cause of the accident?

 

Sorry - My point here is you're making a huge assumption, seemingly based largely on some of them being well overdue for a (non-safety-critical) coat of paint. Your assumption may turn out to be correct or incorrect, I have no hard knowledge either way. 

With all the loco's remaining substantially undamaged through the incident I've no doubt the TSB will already have a good idea of their overall condition and whether they would pass inspection or not.

 

Either way, the reliability of the loco's is not what should have been securing that train. It should have been irrelevant...

 

 

 

I realise I am being a bit abrasive and persistent with this, but 50 people have died and a town levelled in an event that in many ways can't be called an accident and is the latest in a depressing sequence of similar events, different because of location but not so much because of causes. For the rail industry's own good, it can't be business as usual.

 

I totally agree with you there. It needs a serious look, and it should be from a viewpoint wider than Lac Megantic and even the MMA...

Edited July 13, 2013 by Glorious NSE

[Guest Belgian]

This has already been covered.  The brakes are failsafe in the sense that they will stop the train if the pipe is vented, but the brakes are held applied by air from the reservoirs and if no air supply is available these will leak off after a period of usually hours. 

So, if I'm not mistaken, brakes are not 'fail safe' if they rely on an air reservoir to maintain the braking effect. Safety has gone backwards!

 

JE