Train Derails on Bridge in New Jersey

[John WR]

Paulsboro, NJ.  A freight train has derailed on a swing bridge causing several cars to fall off and some to wind up in Mantua Creek.  4 cars contain vinyl chloride and one of them broke open.  About  80 people were evacuated as the gas has potential for being poisonous.  However, in this case the reports are that there will be no long term effects from it.  A real cause for concern is that this is the second time in 3 years a train has derailed on the bridge.  

Here is a link with more information:  http://www.nj.com/gloucester-county/index.ssf/2012/11/paulsboro_train_derailment_to.html#incart_river_default

[CNW 6000]

http://www.nj.com/gloucester-county/index.ssf/2012/11/paulsboro_train_derailment_to.html#incart_river_default

Link activated.  Not a good situation.  At least nobody died.

 

[John WR]

Thanks for the activated link, Dan.  

Twice in 3 years strikes me as "not a good situation" too.  A lot of people got sick enough to go to the hospital but there are no serious illnesses reported.  Fortunately.  I just hope it gets fixed the right way this time.  

[Firelock76]

For the latest at this time (7:40 PM EST) check www.nj.com  "Paulsboro Train Derailment: How it Happened."  To much for me to repeat here.

[CShaveRR]

I just read the article, and it's the saddest journalistic experience I've had in a while.  I wouldn't let any of my stuff get into print looking like that.

"Pulled the emergency break"...really?

I suspect that the 24,000 pounds or so of chemical in the car with the breach (not breech) should have been 24,000 gallons.  "Hull breach"...sounds like a little too much Star Trek.  I've never heard "hull" used to describe either the tank or the shell of a tank car (at least they didn't call it a "tanker car"...<shudders>!).

The most interesting quote was the description of the effects of vinyl chloride:  "a toxic, colorless and flammable gas that causes irritation, headache, shortness or breath and dizziness.   High exposures to the gas, however[,] could be dangerous to one's health."  The foregoing symptoms, evidently, are a manifestation of normal, robust health in the chemical belt. 

Our dispatching friends here might be able to help on this, but it would appear to me that the cause of the derailment was the disregard of a signal that was intended to tell the crew that it was unsafe to cross the bridge.  The conductor's inspection did not reveal the cause of the signal.  But would not the dispatcher bear some responsibility for allowing a movement with hazmat like that to pass an absolute signal without any idea what was causing the faulty indication? 

[blue streak 1]

report says that it is a swing bridge.  I have often felt that they may be more prone to fail either one end or the otner especially with one full car and than an empty ? might have a failure of support causing a teter todder effect??.

[Firelock76]

Well, I didn't say the story was perfect, I just said it was there.  Going on that site I'm used to the typos and grammatical errors so I don't really notice 'em anymore. 

At any rate, by the time the official accident report is finished and published the news cycle will have more than moved on, and we'll never see it in the mainstream press.  We've just got to make the best of what we've got now.

[CShaveRR]

That is a possibility...it would suggest that the failure would lie in one of the piers supporting an end of the bridge in the closed position.  Having closely observed swing bridges (and bridgetenders) in my pre-employment career, I can say that once the bridge is lined up and all of the position holders put in place (at least one for each rail at each end...never did learn the proper term for these), only than can an attempt be made to clear the signal for a train to cross.  It wouldn't take much of a misalignment to prevent a signal from being cleared--could the span be anchored and still not accurately lined up?

Sorry, Firelock...I was in no way aiming at the messenger when I fired both barrels.

[ValleyX]

I disagree about the dispatcher, Carl.  Dispatchers give permission by stop signals all the time without knowing the cause of the circuit being down or failing to clear.  That's what restricted speed is all about.  

Not knowing, I would think that there are some employee timetable instructions that cover the operation of the bridge and what to do if a signal is not received.  

I have to say that I obtained permission by more than one stop signal that no one knew what the cause was, sometimes it would be something we would find, like a broken rail, but frequently, it was nothing we found, and I don't think they were all operating tests.  The bridge, I admit, is a special circumstance, but I'm guessing the dispatcher followed procedure.  I hope the conductor did.

[CNW 6000]

Carl,

The 1899 bridge in Oshkosh has been known to be physically locked in place and yet not able to be lined across. I know CN's B&B guys/gals will be glad to have a new bridge to work with in March.

[Paul_D_North_Jr]

Firelock76

For the latest at this time (7:40 PM EST) check www.nj.com  "Paulsboro Train Derailment: How it Happened."  To much for me to repeat here.

Direct link to that article:

http://www.nj.com/gloucester-county/index.ssf/2012/12/paulsboro_train_derailment_pla.html#incart_river_default 

From the article:

" "He received approval to cross the bridge and pass that red signal," said Hersman. "Two locomotives and several cars made it across the bridge and were one other side when crew stated they saw the bridge collapse and pulled the emergency break." [sic - should be "brake", of course, as Carl noted over in the Trackside Lounge thread]

Five cars made it over the bridge, which is 200 feet long, without incident. The sixth car, carrying plastic pellets, and the seventh car, carrying lumber, derailed onto land." 

Some of the comments to that article are interesting, and 1 or 2 appear to be prettty well-informed. 

I find it significant that the heavy locomotives and several cars made it across without derailing.  To me, that exonerates the conductor - the track that looked safe enough to him (likely a non-expert, though) evidently was.  I'll speculate that something apparently went wrong later, "under load", in the bridge's reaction to the dynamics of the train's motion and imposed weight on it.  That would not be too surprising for a bridge that is located in such a soft geological setting.   

ConRail also has a short news release on its website from Friday, 11-30 (nothing really new there, though):

http://www.conrail.com/11-30_Statement.htm

- Paul North. 

 

[Modelcar]

Reading all the current reports here, and many by experienced RR people...My thoughts {as a non rairoader},  would be, approaching said bridge, and with a known history of some trouble....and the engineer unable to do anything of change to turn the signal light from red to green....operation should have stopped right there at that situation.

Having a red light signal at such a potential trouble spot....and signaling that there {is trouble}....or perhaps trouble with the signal...{unknown at this time},  If possible, engineer should have had permission to wait and not cross}, until 100% certain the bridge had no visible shortcommings, and crossing it might be less than success.

[henry6]

First, give the train crew--both engineer and conductor--and the dispatcher their just due: they followed the rules, they are professionals, they have run this route, crossed this bridge, handled these trains often enough and efficiently enough to know what they were up against, what they saw, and what the knew.  The fly in the ointment to me is the loaded tank car...over loaded perhaps or sloshing so that too much weight was put on one truck tipping the swing deck when it landed on it.  Paperwork was all within bounds and overloading not reported; a surge of cargo to either end of the tank car would put a higher weight on the truck at that end.  This forward and/or back surge could have been caused by the starting of the train at the red block pushing the cargo first to the rear, then to the front, etc.  OR a slowing down on the bridge itself could cause a surge forward then back.  This could cause an overweight on the truck whether or not the tank was overfilled or not.  Nor could it be foreseen even by a veteran crew, especially if the cause and effect never happened before.

[switch7frg]

 One thing for sure ,    Mr. Cronkite is rolling  in his grave over this  "story" . I wonder if any of the ( reporters) teachers from grade school  has seen this account.   Inquireing minds want to know.

                                                                              Just curious  

                                                                                                                                                                  

 

 

 

[CShaveRR]

The crew and the dispatcher may have followed the rules, but I still think none of them followed what should have been instinct in this case.

I read somewhere (not this article) that there had been problems at this bridge for several days--or perhaps that something unusual had been reported in days prior to the derailment.  If that's true, and the bridge saw five to eight trains per day, and none of the preceding runs got the signal that wouldn't clear, they were very lucky.  Or, perhaps the crews were being "talked over" the bridge like this more or less routinely.  In that case, this crew was just the unlucky one.

Another thought:  For a couple of the cars to have derailed "onto land", a derailment would have had to have been dragged at least a couple of car lengths.  I'm surprised that couldn't have been felt in the way the train pulled...I've worked with engineers who could detect an axle on the ground with a 70-car hump shove by the shove's behavior.  And in this case, that would be one of the things I would be concerned about, even if the front part of the train had made it over.  The event recorder might detect a drop in speed that the crew wasn't aware of at the time of derailment.  At least twelve seconds of time would have elapsed for the derailed cars to make it off the bridge (figuring a steady 8 m.p.h. and about 120 feet of car).

Sloshing tank cars?  Overloaded?  Probably not, in both cases.  I don't believe I've heard of cars containing liquefied gases (such as vinyl chloride) sloshing--the commodity would expand to the volume of the tank, and under the pressure of such a load, would go nowhere (please correct me if I'm in error about this!)  There was a load of ethanol; that could have sloshed.  I doubt that anything was overloaded, as those tank cars are designed and sized for the commodities they handle, and probably with a safety factor.

This will be an interesting one to watch for a thorough investigation's results. 

[Anonymous]

Well there is that rule about when in doubt, the safest course must be taken.

It would be interesting to know what rules apply to passing that signal.  You can have a track defect anywhere but having one giving a red at a home  signal for a swing bridge when the bridge is supposedly lined, seems like it would pose a potentially higher risk to pass it as opposed to an unexplained red signal elsewhere. 

I can see a possibility that the track signal fault may have been no different than faults common elsewhere.  It is just this fault may have been caused by the settling foundation of the bridge.  So in that case, the signal fault would have been like the canary in the coal mine.  It would have alerted to a problem that was deadly, but invisible.   

[PNWRMNM]

Carl,

Vinyl Chloride Monomer is a liquefied compressed gas. That means that the vast bulk of the mass of the material in the tank car is in the liquid state, so yes it could slosh.

The sloshing theory is quite fanciful however. Tank cars are loaded so that they will not become "shell full" at a reference temperature. Liquids are loaded to 2% outage, or 98% innage by volume. Compressed gasses follow a different logic in deterermining how full is full, but I suspect they end up in the 2 to 3% outage range. A full slosh in a car with 2% outage would thus put an additional 1% of the load, or not more than 2,000# more, on the "heavy" end, which is next to nothing in terms of rail and bridge loadings.

Mac

Former Bureau of Explosives Inspector

[CShaveRR]

Thanks, Mac.

[Modelcar]

....From what I've read here....the approach signal being red, perhaps a check of the warning system could have been, to try and open the bridge to show it was in fact operational....And if then, the engineers {radio signal}, would not be effective in moving the bridge just a bit...it perhaps would have been an indicator something was not normal, and inspection of why the bridge wouldn't move should have been in order.  And not take the train across it.

I'm assuming the engineer would have had the authority to try to open the bridge a few feet to actually "test" it's integrity.

[n012944]

Not being able to get a signal across a movable bridge is more common than some of you may think.  The procedure to be followed when a signal can not be cleared across a bridge is in the timetable special instructions.  If those instructions were followed in this case, there is no cause for either the train crew or the dispatcher to have any responsibility in this incident

[cx500]

Somewhere I saw some aerial photos of the scene, and I could not see any guard rails between the running rails.  These can guide a derailed car more or less along the ties until it is past the bridge.  Guard timbers on the outside, while they do keep the ties in position, are not as effective as the inside rails. 

With a through truss bridge, and probably also this unusual design, a lot of the strength lies in the superstructure.  Once a column or diagonal is damaged, catastrophic failure of the bridge is likely.  If an already derailed car hits a structural member, the result will be very like what happened here.

I imagine the investigators already have a pretty good idea of what the cause or causes are.  Filling out all the details for the usual comprehensive report will of course take a lot longer.

John

[oltmannd]

ValleyX

I disagree about the dispatcher, Carl.  Dispatchers give permission by stop signals all the time without knowing the cause of the circuit being down or failing to clear.  That's what restricted speed is all about.  

Not knowing, I would think that there are some employee timetable instructions that cover the operation of the bridge and what to do if a signal is not received.  

I have to say that I obtained permission by more than one stop signal that no one knew what the cause was, sometimes it would be something we would find, like a broken rail, but frequently, it was nothing we found, and I don't think they were all operating tests.  The bridge, I admit, is a special circumstance, but I'm guessing the dispatcher followed procedure.  I hope the conductor did.

Up until several years ago, this bridge had a bridge tender.  Conrail automated it.   This particular bridge is  a weird one.  It is a swing bridge, but not like most.  It is hinged at one end an the other end swings away like a gate when it's open.  There is a large A frame over the hinge end with cables down to the open end of the movable piece to bear its weight when open.

The crew has said they couldn't get the signal to come in, but the conductor had taken a look at it and it was okay.  I wonder if the locking bars were in place but not locked in and the span "giggled"open as the train passed.

There are some good pictures of the bridge "before" here: http://www.flickr.com/photos/kevinhooa/2950926527/in/photostream/

[tree68]

I thought I saw it opined that perhaps Sandy had something to do with the incident, in that storm surge or other factors may have weakened the underpinnings.  I don't think the Philly area suffered direct storm surge (I could be wrong), but the accompanying rain might have caused a greater-than-normal increased flow.

I think the bridge is here:  N 39.83464 W 75.23661.  I don't see any guard rails, either.

A washout below the surface would have been hard to detect, and depending on how it happened, it could well have been the motion of the train over the bridge that was the straw the broke the camel's back.

[Firelock76]

There's been a lot of good thoughts and theories here, but I'm starting to think the simplest one is the best.  This was probably just a crummy old bridge that should have been replaced a long time ago, and for various reasons just wasn't.

I suppose they'll replace it now, huh?

[mudchicken]

Locking Pins = related to Mitre Rail (if this bridge wasn't stub)=Locking Device or dogs

[edblysard]

All the radio controlled switches/devices we have on my railroad function completely through their cycle before they can be “reset”

In other words, a radio switch has to open all the way and properly line for the other route before it will line back.

I find it odd that this bridge can be moved a “few feet” and then stopped…such a device would allow the bridge to be miss-aligned or left partially open, which I have never seen allowed.

I would suspect the engineer attempted to cycle the bridge through the open then closed function, and that the reporter just quite didn’t understand what was being implied.

I also find it hard to believe that a “slosher” would have any effect on the bridge, unless there was a major defect in the bridge.

Even the newer bridges are over engineered from a carried weight standpoint.

Sloshers are a normal part of railroading; they are in almost every train, so I would bet the bridge engineers were aware of this.

As for the crew or dispatcher being at fault, flagging and getting flagged by a red one is also normal, although at a bridge that had previous structural problems I personally would offer up a good faith challenge before I proceeded…they would have to order me to cross/proceed.

[CShaveRR]

Thanks, Ed!  I think you've told it like I would have seen it, too.

I looked at the coordinates Larry provided, and beyond (to the west)...what an amazing complex of tracks and equipment (including two industrial locomotives) that, until things are repaired or rebuilt, are isolated.

[edblysard]

Yeah, that all looks so familiar, my kinda railroading, lots of tracks to spot, pull and puzzle out!

I see a couple of refineries, a plastic plant too…

 

[Paul_D_North_Jr]

CShaveRR

Thanks, Ed!  I think you've told it like I would have seen it, too.

I looked at the coordinates Larry provided, and beyond (to the west)...what an amazing complex of tracks and equipment (including two industrial locomotives) that, until things are repaired or rebuilt, are isolated.

That complex is a refinery that used to be owned by Mobil some years ago - I did a little trackwork in there, but not much.  It's currently owned by PBF Energy, after Valero - see: http://www.pbfenergy.com/refineries - what with all the shuffling that's gone on since then. 

Supposedly it's switched by SMS Rail Services, which is based just a little further down that line, at the Pureland Industrial Complex - see: http://www.smsrail.com/locations-pa-nj-sms-rail.html  SMS is notable for running a large fleet of Baldwin diesels - see: http://en.wikipedia.org/wiki/SMS_Rail_Lines 

Further down yet is a former B.F. Goodrich plant, at which I spent many hours.  The double-curved crossing frog at these coordinates was not designed or originally installed by me, but I was responsible for its upgrading and replacement about 30 years ago, and as of 18 months ago it was still going strong: N 39 45' 48" W 75 25' 11"

FInally, there's the Deepwater power plant here, at which I've also done some trackwork: N 39 41' 42" W 75 29' 4"

What I'm really curious about is whether there's an alternate rail route further inland, so that all those big industries are not stranded/ cut-off from rail service.  I'm not real familiar with the current configuration and connections in South Jersey, but it appears that there isn't - see: http://www.sjrail.com/maps/maps.html and http://www.sjrail.com/maps/SJ_Map.jpg   

- Paul North.   

[Falcon48]

oltmannd

ValleyX

I disagree about the dispatcher, Carl.  Dispatchers give permission by stop signals all the time without knowing the cause of the circuit being down or failing to clear.  That's what restricted speed is all about.  

Not knowing, I would think that there are some employee timetable instructions that cover the operation of the bridge and what to do if a signal is not received.  

I have to say that I obtained permission by more than one stop signal that no one knew what the cause was, sometimes it would be something we would find, like a broken rail, but frequently, it was nothing we found, and I don't think they were all operating tests.  The bridge, I admit, is a special circumstance, but I'm guessing the dispatcher followed procedure.  I hope the conductor did.

Up until several years ago, this bridge had a bridge tender.  Conrail automated it.   This particular bridge is  a weird one.  It is a swing bridge, but not like most.  It is hinged at one end an the other end swings away like a gate when it's open.  There is a large A frame over the hinge end with cables down to the open end of the movable piece to bear its weight when open.

The crew has said they couldn't get the signal to come in, but the conductor had taken a look at it and it was okay.  I wonder if the locking bars were in place but not locked in and the span "giggled"open as the train passed.

There are some good pictures of the bridge "before" here: http://www.flickr.com/photos/kevinhooa/2950926527/in/photostream/

  I have just enough experience with moveable bridges to be dangerous (and I'm not a bridge engineer either).  But I had exactly the same reaction you had - the bridge may not have been locked in place, and moved out of alignment as the train crossed it which caused the derailment.  That would also explain the red signal, as the signal would not clear unless everything was properly in place.  Of course, this is all speculation.  We'll just have to see what NTSB finds