Derailment and fire in Quebec

[dibber25]

It would be interesting to know how many Class 2 railways are allowed one-man operation. I understand that CP and CN were refused permission for single-manning. I'm trying to imagine a long shift, driving at 15mph (or is it 25mph?) I've seen both speeds quoted as the maximum on the MMA. Hour after hour, on your own, crawling along like that, into the evening. Into the dark. Doing pretty much what you've always done, without thinking too much about it.....

In addition to asking about the wisdom of single-manning, particularly in those sort of circumstances I'd be asking about a number of practices which seem to be OK in Canada, but about which I have grave doubts.

1 Leaving a train 'parked up' on the main line for an extended period.

2 Security-wise, leaving a train parked in an unsecured area (Canadian lines have no fences etc and two people were arrested recently for allegedly plotting to attack cross-border railways)

3 Dramatic increases in traffic (28,000% in oil by rail in five years) without, apparently any upgrading of some of the routes taking this extra traffic.

4 The use of old, umpteenth-hand locomotives on heavy trains of hazardous material. (Any picture of an MMA train with 5 or 6 locos will have 4 or 5 different liveries and locos that are in excess of 30 years old)

5 The lack of provision of catch or trap points on loops, even those where there is an obvious danger of vehicles running away down hill.

6 From a PR perspective, the lack of an action plan in the event of an emergency. Who is authorised to say what to the press, to prevent mis-information, conjecture and the kind of reaction which Burkhardt seems to have received.

CHRIS LEIGH

[Talltim]

I kinda agree, but apart from the single manning, much of that applies in the US as well as Canada

[Adrian Wintle]

1 Leaving a train 'parked up' on the main line for an extended period.

2 Security-wise, leaving a train parked in an unsecured area (Canadian lines have no fences etc and two people were arrested recently for allegedly plotting to attack cross-border railways)

 

1. Think of this as effectively a 'one engine in steam' situation and it makes a bit more sense.

 

2. A lot of Canadian railroads are fenced, but the fence is purely a demarcation of the railroad property.

 

Adrian

[highpeak]

I don't think you could stable a train at Lac-Megantic without leaving it out in the woods on the main line. Anywhere in town would block a grade crossing. If you stopped it short there would be an idling loco at the edge of town while if you stopped it beyond the town the locos would be out in the woods. The Nantes loop is alonside the main road and not in a built up area.

 

Adrian

I agree that Megantic probably would not be a good place to leave an oil train idling for a few hours, but the town was a significant railroad town back in the day. It was one end of the Moosehead Subdivision of the International of Maine, the other end being Brownville Jct, and was a junction with the Quebec Central line from Tring Jct on the Quebec-Sherbrooke main line. It was also half the name of the 2ft gauge Franklin and Megantic, though I doubt the perpetually broke road ever had any serious plan of crossing the border.

The fourth picture down on this page http://users.silcon.com/~lgoss/morecpr.htm shows Megantic with the Scoot getting ready to leave for Brownville Jct.

[dave1905]

1 Leaving a train 'parked up' on the main line for an extended period.

2 Security-wise, leaving a train parked in an unsecured area (Canadian lines have no fences etc and two people were arrested recently for allegedly plotting to attack cross-border railways)

3 Dramatic increases in traffic (28,000% in oil by rail in five years) without, apparently any upgrading of some of the routes taking this extra traffic.

4 The use of old, umpteenth-hand locomotives on heavy trains of hazardous material. (Any picture of an MMA train with 5 or 6 locos will have 4 or 5 different liveries and locos that are in excess of 30 years old)

5 The lack of provision of catch or trap points on loops, even those where there is an obvious danger of vehicles running away down hill.

 

 1.  What's the big deal with parking a train on the main track.?  I don't see a problem with it.  Trains stop on the main track all the time.

 

2.   Over 95% of N America is "unsecured area".  Ever been to Wyoming? 

 

3.  How do you know the track hasn't been upgraded or there was anything "wrong" with the track?

 

4.  Why would a new engine have made any difference to the outcome?  Some of the newest engines made have chronic problems with fuel lines breaking.

 

5.  Because in N American practice derails are not put on "running tracks" main tracks and sidings, only storage tracks and industrial tracks.

[Gerald Henriksen]

It would be interesting to know how many Class 2 railways are allowed one-man operation.

In Canada, the answer is 2 regardless of class. MMA and QNS&L are the only railways in Canada to have been given an exemption from the rules to allow single person operation.

 

I'm trying to imagine a long shift, driving at 15mph (or is it 25mph?) I've seen both speeds quoted as the maximum on the MMA. Hour after hour, on your own, crawling along like that, into the evening. Into the dark. Doing pretty much what you've always done, without thinking too much about it.....

I'm not saying I agree with him, but Ed Burkhardt yesterday in a press conference in Lac-Megantic said he believes single person operation is safer because the crew member doesn't have a co-worker distracting him/her.

 

In addition to asking about the wisdom of single-manning, particularly in those sort of circumstances I'd be asking about a number of practices which seem to be OK in Canada, but about which I have grave doubts.

 

1 Leaving a train 'parked up' on the main line for an extended period.

What relevance does this have? This is a railway that has little in the way of traffic, there was nothing else coming, and even if the engineer had taken the time to place the train into the siding (assuming nothing else is being stored on it) given the siding is double ended the train still would have runaway into Lac-Megantic.

 

About the only thing you may be able to question would be parking a train on a grade, but how much flat territory does the average railway in the area have?

 

2 Security-wise, leaving a train parked in an unsecured area (Canadian lines have no fences etc and two people were arrested recently for allegedly plotting to attack cross-border railways)

Short of having security guards, what else would you expect? The amount of graffiti on freight cars in North America demonstrates that not even the class 1's fully secure their trains despite the money spent since 9/11.

 

3 Dramatic increases in traffic (28,000% in oil by rail in five years) without, apparently any upgrading of some of the routes taking this extra traffic.

Logistics.

 

While not exact, Google shows that Farnham QC to Saint John NB (where the MMA picked up the train to the ultimate destination - I'm not sure where MMA would hand the train off) is 695km by road. Rail may be shorter, depending on the terrain and other factors.

 

If we take that figure of 695km, that is slightly more than the distance London - Glasgow (663km / Google). How long, and how much money, do you think it would take to upgrade that amount of track? Assuming of course that the railway has the money to do it all.

 

4 The use of old, umpteenth-hand locomotives on heavy trains of hazardous material. (Any picture of an MMA train with 5 or 6 locos will have 4 or 5 different liveries and locos that are in excess of 30 years old)

How old are the locomotives used on UK nuclear flask trains?

 

How old are some of the locomotives carrying passengers in the UK, and how many more years are they expected to be in use?

 

The age of the locomotives is a non-issue. The class 1's still have lots of old locomotives on their rosters (the last 4-axle freight loco stopped production 20 years ago), and the big reason they have continued buying newer stuff is that they can afford the up-front costs (apparently $2 - $3 million per unit) to get the better fuel efficiency combined with reduction in units per train (say 4 new units replaces 6 older units).

 

5 The lack of provision of catch or trap points on loops, even those where there is an obvious danger of vehicles running away down hill.

Don't know that these have every been used over here. Maybe they should, but it can be a significant issue given the length of some trains over here and the need for a crew member to then walk the length of the train to set/unset them (and even more of an issue in the winter).

 

6 From a PR perspective, the lack of an action plan in the event of an emergency. Who is authorised to say what to the press, to prevent mis-information, conjecture and the kind of reaction which Burkhardt seems to have received.

It's easy to say you should have one, its another thing to spend the money on creating one when your company is losing money. Besides, an action plan doesn't help if your company head decides to speak to the press anyway...

[highpeak]

And I wouldn't mind betting that at times what actually happens might not be too far removed from my 'work around' alternative'

I'm sure you are right regarding a workaround being used or at least some kind of "this has always been OK" short-cut procedure as opposed to the full procedure called for of apply handbrakes/test that they hold/repeat until train is secure without air.

The point of the study I quoted is that you can't depend on simply setting some given number of handbrakes. The number most often mentioned here seems to be 11 cars, which from most statements by professional railroaders including Dave1905 on here does not appear to be enough. The study suggests that variations in brake rigging and exactly how tight you wound the chain up, plus variations in weight of train mean that you can't really always set x number of brakes and expect the same results. The full test procedure is the only reliable way and a "set x brakes and call it quits" approach is asking for trouble. It will be interesting to see what MMA instructed its engineers to do and what they did to check on that.

But when you've done it that way and never had a problem, it becomes the daily procedure. And given that you are on your own after 11 hours of mind-numbing creeping along with a heavy train, you are no doubt ready to tie up as quickly as you can and get some sleep. In reality, it's only worked because you've always had the air as the actual restraining force on the train. And then you have some chain of events like this (and this is speculation but it's areas I'd be looking at): a locomotive that needed attention but was sent out anyway because it was felt to be good enough for another trip and there might not have been an alternative; the gamble doesn't pay off and the fuel leak gets worse resulting in a fire; the fire department has to shut the engine down (which I think will turn out to be a red herring, surely the air shouldn't have bled off that quickly? I've seen statements from railroaders to the effect you can still have an effective brake on a cut of cars several days later), a fireman inadvertently releases the air while climbing out of the cab (a post on railroad.net by a railroader suggests that's easier to do than you'd think.)  And now the holes in the cheese have lined up.

Ed Burkhardt said in one of his news conferences that you can't go back in time and bring Lac-Megantic and 50 people back. But you can make sure history doesn't repeat itself if you try to understand all the aspects of this accident.

[Gerald Henriksen]

the gamble doesn't pay off and the fuel leak gets worse resulting in a fire; the fire department has to shut the engine down (which I think will turn out to be a red herring, surely the air shouldn't have bled off that quickly? I've seen statements from railroaders to the effect you can still have an effective brake on a cut of cars several days later), a fireman inadvertently releases the air while climbing out of the cab (a post on railroad.net by a railroader suggests that's easier to do than you'd think.)  And now the holes in the cheese have lined up.

And now you also have wet rail from the fire department, which maybe just reduces the effectiveness of the locomotives brakes just enough to allow the slide downhill to start.

[highpeak]

This report into a 2012 accident involving a runaway cut of cars provides some insight into procedures expected to be followed when securing cars with handbrakes. [ur;]http://www.bst-tsb.gc.ca/eng/rapports-reports/rail/2012/r12e0004/r12e0004.asp[/url]

The report also has an interesting analysis of the effective brake shoe force measured from two cars which survived the accident and an estimate that to secure the 13 loaded coal cars on the 1% grade handbrakes would have been needed on 5.1 cars. Only one car had had the handbrake applied.

[highpeak]

And now you also have wet rail from the fire department, which maybe just reduces the effectiveness of the locomotives brakes just enough to allow the slide downhill to start.

A very good point. If you were just on the limit of the holding ability of the handbrakes the reduction in coefficient of friction might be enough to let the train move. Not relevant here, but maybe you could run into that with dew or rain?

[Ozexpatriate]

2 Security-wise, leaving a train parked in an unsecured area (Canadian lines have no fences etc and two people were arrested recently for allegedly plotting to attack cross-border railways)

Perhaps one of our Canadian observers can comment but I've heard it said that many Canadians don't routinely lock their front door.

 

Perhaps this is a myth and it is certainly not true in the big cities but I suspect there are plenty of Canadians in small towns and rural areas who will say they don't.

Edited July 12, 2013 by Ozexpatriate

[Removed a/c]

My Mum never does, nor does she lock her car. 

[Ozexpatriate]

Burkhardt during an interview on his heated reception in Lac-Megantic:

CNN asked the railway executive what he believed was the cause of the disaster.
 
"Well, there's no question that it's a brake failure on the train. The train rolled away. That speaks for itself, doesn't it?" he answered. "There's no sugarcoating that."

 

"What caused the brake failure is a bit complex it's more than one factor," he continued, saying that the company is still "trying to pull all the pieces together on that."

[brossard]

Well I always lock my doors.  I live in a suburb on the south shore of Montreal.  Mind you, I've never had a problem.  On a related issue, I left my car at Dorval Airport while I took a business trip a few years ago.  When I got back I opened the drivers door to find to my dismay a ticket for $50 for not having locked the car (thought I had of course).  So mixed feelings - POed at the ticket but happy the nice policeman had locked my car for me.

 

John

[Gerald Henriksen]

Contrast that with Clapham, where BR admitted liability within a couple of days - the evidence of poor workmanship in the rewiring at Clapham having been found. There was an immediate sense of relief, and the bereaved were able to proceed.

 

It seems unlikely that Mr Burkhardt would have made his public statements without firm evidence from company advisers. He would also have discussed the issue of liability with his insurers. I believe holding up the corporate hand is the right thing to do.

 

Part of the problem, at least in the US and sadly increasingly elsewhere, is doing the "right" thing can cost the company big time.

 

See the following article that discusses the possible legal ramifications from the statements made, as well as the possible dollar amounts:

http://www.theglobeandmail.com/report-on-business/industry-news/the-law-page/a-candid-executives-next-problem-the-litigators/article13173158/

[Gerald Henriksen]

Perhaps one of our Canadian observers can comment but I've heard it said that many Canadians don't routinely lock their front door.

 

Perhaps this is a myth and it is certainly not true in the big cities but I suspect there are plenty of Canadians in small towns and rural areas who will say they don't.

 

Growing up in a small town northwest of Toronto (76-86) some of our neighbours didn't lock doors, and it was certainly common to leave car doors unlocked when in the village.

 

More recently we had a neighbour in our apartment building who seldom locked their door (and the buildings doors were often propped open) and never had any problems.

 

The bus terminal at the mall next to us the drivers leave the doors open with the bus running as they run into the mall to get coffee.

[brossard]

Part of the problem, at least in the US and sadly increasingly elsewhere, is doing the "right" thing can cost the company big time.

 

See the following article that discusses the possible legal ramifications from the statements made, as well as the possible dollar amounts:

http://www.theglobeandmail.com/report-on-business/industry-news/the-law-page/a-candid-executives-next-problem-the-litigators/article13173158/

 

I have to admit that I was surprised at Burkhardt's comments.  They didn't come across as if they'd been prepared.  I thought to myself that his corporate lawyers would be having kittens.  I know from where I used to work that nothing went out for public consumption before the lawyers had reviewed and diluted it.

 

Canada isn't as litigious as the US but I can imagine that there will be claims from the Provincial Gov't (who have chipped in $60M - they'll want that back), Town and Victims.

 

John

[Gerald Henriksen]

Canada isn't as litigious as the US but I can imagine that there will be claims from the Provincial Gov't (who have chipped in $60M - they'll want that back), Town and Victims.

 

John

 

The assumption so far is that the victims will attempt to sue in Chicago (where the company that owns MMA is based).

[Mike Boucher]

Perhaps one of our Canadian observers can comment but I've heard it said that many Canadians don't routinely lock their front door.

 

Perhaps this is a myth and it is certainly not true in the big cities but I suspect there are plenty of Canadians in small towns and rural areas who will say they don't.

 

I don't live in Canada,  but I am French-Canadian by descent  My parents have lived in the same house, about 35 miles north of Boston, since 1978.  Their house has one door that doesn't even have a lock.

 

I suspect there are plenty of people in rural sections of the UK who don't lock their door either.

[highpeak]

My girlfriend who lives out in the boonies here in CT doesn't bother locking any doors on her house. Her rationale is if you want to break in, the door lock won't stop you. The pit bull might make you change your mind though.

[Gerald Henriksen]

A good news type story after all of this.

 

Despite everything, including the fact that the office is in the red zone and many of the staff are temporarily homeless, the local paper managed to publish their weekly edition:

http://www.thestar.com/news/canada/2013/07/11/lac_megantic_disaster_cant_stop_newspaper_from_publishing.html

 

For those who want to check it out, the website (French only):

http://www.echodefrontenac.com/

 

 

One of the stories already on the website may be of interest to some here.  The Sunday after the runaway a reporter "found" another unattended MMA train on the east side of Lac-Megantic with an engine left running:

http://www.youtube.com/watch?v=oOwPYr38A1c (the reporter provides both French and English in the description, and if you enable subtitles you get English subtitles).

 

There is also a second video showing more of the train, as well as a distant view of the lake:

http://www.youtube.com/watch?v=RGg0tW8PgTc

 

Probably a safe guess that this would have been the westbound train that the runaway train's engineer would have taken back to Farnham had things gone normally.

[Edwin_m]

In reality, it's only worked because you've always had the air as the actual restraining force on the train. ... the fire department has to shut the engine down (which I think will turn out to be a red herring, surely the air shouldn't have bled off that quickly? I've seen statements from railroaders to the effect you can still have an effective brake on a cut of cars several days later)

Just to be clear, if the air brake was applied on the train, then there would have been no pressure in the train air pipe.  So, regardless of whether the loco compressor was running, the train brakes would have leaked off over time.  The only brake that is affected by the engine shut-down is the air brake on the loco itself, and possibly the air brake on the coupled locos if these were also linked by a second (reservoir) pipe (I think it is US/Canadian practice to do this). 

 

However, given that the consist of locos is able to start the train on the steepest gradient, the air brakes on the same locos on would usually be enough on their own to hold the train on the same incline.  Certainly if insufficient wagon handbrakes were applied the leaking off of loco air brakes could well make the difference between the train staying in place and rolling away.  What happens with loco parking brakes?  Do these apply on all wheels like the loco air brake?  If so these could possibly also have held the whole train. 

 

An air brake in good condition will stay applied for several hour before leaking off, and days is not impossible.  However if the equipment is in less good condition it could leak off within an hour or so, which I think is the sort of time after which the train must be secured by other means.  Given what we hear about this operation I'd say it was quite likely that the leaking-off time would be close to the minimum. 

[dibber25]

This is about precautions and 'best practice'. My comments were mainly about precautions. Sure they cost in either time or money or both and that's the problem for railways which are trying to wring a profit out of lines which the Class 1s have ditched - hence the old locomotives etc. For instance, if a train is not parked on a main line - if it is shut away in a loop with trap points correctly set, it is all but impossible for it to run away on the main line. If you park your car on the highway, leave the engine running and the handbrake off, you'll probably get away with it, but you'll be much more likely to find you car in one piece if you park it on your driveway, switch off the engine and put the handbrake on. There may be people out in the sticks in the UK who don't lock their doors (I doubt there are many, after the Tony Martin affair) but if they get burgled its a pretty safe bet their insurance company will have something to say.

As to the subject of old locomotives in the UK, well, we don't have many locomotives in passenger use. I think the oldest in daily use would be the Class 91s at 20+ years but much more recently rebuilt. There are a handful of old diesels used on nuclear flask traffic, again totally rebuilt and kept in first class condition. I accept that conditions in Canada are very different from the UK but I still believe any one of the provisions I mentioned in my earlier post might well have mitigated this disaster. As so often happens in major accidents, it isn't just one failing but a chain of seemingly minor failings which all come together at the wrong time. We'll never eliminate them but that doesn't mean that we should not look at changes which might help. The words "we don't do it that way - we never have" are one of the biggest dangers.

CHRIS LEIGH

[The Stationmaster]

Just to be clear, if the air brake was applied on the train, then there would have been no pressure in the train air pipe.  So, regardless of whether the loco compressor was running, the train brakes would have leaked off over time.  The only brake that is affected by the engine shut-down is the air brake on the loco itself, and possibly the air brake on the coupled locos if these were also linked by a second (reservoir) pipe (I think it is US/Canadian practice to do this). 

 

However, given that the consist of locos is able to start the train on the steepest gradient, the air brakes on the same locos on would usually be enough on their own to hold the train on the same incline.  Certainly if insufficient wagon handbrakes were applied the leaking off of loco air brakes could well make the difference between the train staying in place and rolling away.  What happens with loco parking brakes?  Do these apply on all wheels like the loco air brake?  If so these could possibly also have held the whole train. 

 

An air brake in good condition will stay applied for several hour before leaking off, and days is not impossible.  However if the equipment is in less good condition it could leak off within an hour or so, which I think is the sort of time after which the train must be secured by other means.  Given what we hear about this operation I'd say it was quite likely that the leaking-off time would be close to the minimum. 

Think about it this way Edwin - the train auto brakes are applied but are maintained in that position by the reservoir pressure on each vehicle, those reservoirs gradually lose pressure and the brakes leak off, the rate of leakage will vary but once the brake itself starts to release the vehicle will move if gravity wins.

 

On the engines (assuming they're similar to just about everywhere else) the brakes are kept on by air pressure and that pressure is supplied by the compressors with - in many cases very limited reservoir capacity.  Once the compressor stops pumping leakage is almost inevitable and about an hour for the brakes to leak off wouldn't strike me as unusual, especially on a poorly maintained loco.  If there were several locos on the train (I understand there were) the brakes applied on all of them represent quite a brake force, but once that brake force begins to reduce it is again up to gravity and while a 'rubbing' brake can hold vehicles it won't hold several thousand tons.

 

My scenario might be wrong - and it is speculation I'm the first to admit, but it is speculation born of a bit of experience of what can happen when folk come up with a 'work arounds' or cut corners or take chances - however one cares to put it.  And human nature is fairly consistent across the planet in some respects in my experience.

 

We then look at other factors such as those listed by Chris - the last time one of my Drivers left a train unattended on a running line he got 3 days suspension - if it had been my decision he would have got 5 days off.  Leaving trains unattended on running lines is asking for trouble, leaving a train unattended on a steep gradient on a running line is just plain foolhardy and any sort of risk assessment would rule it out, experienced railwaymen wouldn't even need the risk assessment (and in this country we learnt not to do it back in the 19th century!).  

 

As far as single manning is concerned I'm now so used to it on a diesel/electric powered railway I hardly need to question it.  There is clear evidence that having a second person in the cab can be a distraction but equally running at low speed over considerable distances must be fairly soul destroying and again is something that should be risk assessed (and would be in any sensible organisation) but that also takes us back to the state of the track and a quick look on Google shows some of it on that railway to be in a parlous state.  Ropey track and hazardous goods do not mix well - one is injecting unnecessary risk as any experienced railwayman will tell you - but that takes us back to the cost vs profit vs damage costs debate.

 

In due course we'll find out what happened and I suspect a lot of what we hear won't surprise at least some of us.  This incident was however - even at this stage - very clearly not an accident, it had a cause which could or should have been managed out.

Edited July 12, 2013 by The Stationmaster

[Glorious NSE]

As to the subject of old locomotives in the UK, well, we don't have many locomotives in passenger use. I think the oldest in daily use would be the Class 91s at 20+ years but much more recently rebuilt.

 

The oldest loco's in daily passenger use would be the HSTs, which would be roughly comparable in age with the GE that played a part here (and in usage terms will have done many times the mileage of that GE, although most have had at least 2 rebuilds!)...although there are 47s that are more than a decade older than that, and are definately not immune from assorted failures and even fires etc which are also irregularly used in public passenger service (such as Anglia's summer Yarmouth trains...) and on charter passenger work.

 

I think some concepts are being run away with here - just because there was a loco fire and the loco's look scruffy does not neccesarily mean the loco's were not being properly maintained, as Dave says loco fires can happen if something breaks on even a new loco (google 70018!) - and I don't believe the state of the track is an issue either - I can't see anything to do with this accident that suggests the track was not maintained well enough to support the speeds trains were allowed to run. Now it's entirely possible that either or both of those suppositions are true - but I think it's wrong to assume them from the info we have here.

 

Regarding leaving the train - I do wonder whether it was tied up before midnight as that is the result of them risk assessing having an engineer trying to run at 10mph continuously through the small hours of the morning without making an error or falling asleep... 

 

Mike - I disagree with your point about leaving the train on the running line. In a North American context it would have had zero extra protection if it had run away from a yard or the passing siding. 

Edited July 12, 2013 by Glorious NSE