—Preventing a Derailment from Becoming a Train Wreck—
What I am proposing here is a bold and robust concept for a stronger and smarter derailment-resistant train that uses the power of the locomotive and smart braking of the train to control derailments if they do happen.
This concept relies mostly on functional redesign, monitoring the train for trouble to prevent derailments, and mitigating the effects of derailments; as opposed to making tank cars more crashworthy. While both approaches cost money, making cars more crashworthy leaves less capacity for cargo, and thus drives up the cost of shipping.
While it is true that trains have been hauling hazardous materials since the beginning, I don’t think that justifies complacency in believing that nothing is new about the hazard. What is new is crude oil that is as explosive and flammable as gasoline, and that crude is being handled in long unit trains as the preferred method of transport in lieu of pipelines. That is all new. Also new is the domestic boom in this new type of crude that will lead to unprecedented oil traffic levels on railroads. And the final thing that is new is a reduction of sympathy for oil over the last fifty years. That last item does not make oil shipping more dangerous, but it leads to a more strident call to action to reduce the danger. Fifty years ago when we all used oil without questioning its value in making our lives easier, we would tolerate the risk of fire and explosions, and the smell of refineries. One other thing that is new since fifty years ago is Lac Megantic actually demonstrating wha
First since I don’t know you can you describe your experience with train wrecks. It sounds like you might have some background or at least your well read.
Here’s the killer to your plan, The “unbreakable” drawbar means that you must also have an unbreakable car. While I’m sure this is possible what you will end up with is a car that weighs 80-90 tons EMPTY. Drawbars and knuckles are designed to break before the car does.
Most tank cars don’t have center sills anymore, they have stub sills welded to the tank body.
Randy’s points seem valid to me, but there is one other: Your concept would work for fixed unit-trains but not for single-car railroading, and some crude and other petroleum products do move in single and short consist moves. Possiblly tankcar construction with center sills and tanks that disengage from the sills miight make for addtional safety or perhaps actually reduce safety. But such heavy equipment would exceed axle loadings of the existing track structure and bridges. New track and bridges would be required. Probably cheaper to construct pipelines!
Randy’s points are well articulated and thought out. I’m not a professional railroader so can’t speak from experience, just an outsider looking in.
Here’s the thing: the stuff HAS to be moved, one way or another. It’s GOING to be moved, one way or another. The railroads have a choice, they can take the bull by the horns and come up with as safe a way to move crude as possible and hang the expense, or they can wait for the heavy hand of government bureaucracy to come up with who-knows-what at even more expense, or they can lose the business to pipelines. They’re going to have to do SOMETHING.
A dedicated unit train built to Randy’s specifications makes perfect sense to me. Actually, Randy’s couplers sound a lot like the “tite-lock” couplers passenger cars have been equipped with for years.
Sounds heavy, too slow to bring online as a replacement and expensive.
My guess (absolutely no inside knowledge) is that until the newer tank cars come online (they will be kept in dedicated unit trains, not mixed with the 111’s) the existing cars will be run with some/many of the special precautions mentioned by Dave K and others. And the filling of the cars with Bakken will be more carefully monitored. The additional costs will be allowed to be passed on to the shippers. No one will be fully satisfied, but the railroads will come out of it best.
Part of this is little different in principle from the old Miller arrangement: combination of inseparable couplers with some positive means of providing anticlimbing in multiple axes.
A much more ‘sensible’ method of doing this than with drawbars might be a combination of a modified type F coupler (as already mandatory on tank cars) with a rotary mechanism adapted from hoppers (provided with some sort of progressive limitation on rotation). The modification to the F coupler would be to provide some sort of positive locking if the mutual vertical displacement of the couplers exceeded a certain number of inches. The purpose of this change would be to keep the couplers from separating, and perhaps removing any out-of-axis loads from the knuckle mechanism during an incident (or relieving stress from the knuckles or ins themselves, and thereby keep car ends from contacting past the tank-end shields, or telescoping.
The idea of positive, and perhaps locking, buffer arrangements might be useful here, especially if the buffer ‘faces’ constitute arcs of a circle around the rotary drawhead axis.
Once that is done, you can start providing armor, leak suppression, internal full-height baffling, etc. to prevent the liquid from spilling to the greatest possible extent, managing potentially explosive vapor expulsion, etc. – none of which I see addressed in the original proposal.
As Randy noted, the integrity of the cars becomes a critical factor – one which is not particularly likely to be assurable without making them so heavy that their own mass becomes a more destructive aspect than the fact of a derailment.
More to the point, however: the principal issue of current concern does not appear to be oil-train derailments so much as it is other trains derailing INTO oil-train movements. No amount of fancy antiseparation arrangement helps when the cars are buckled or punctured from the side, or shock force on the order of that produced by mutual
My idea was to put a CO2 or Nitrogen blanket in the cars after loading to purge the oxygen from the explosive mixture in the tops of the cars. It seems that at least some of the cars blew up from the inside out when sparks entered the tank body.
I actually mentioned this to some pretty important folks in the government… we’ll see, I sincerely wish I had an answer. I enjoy low gas and heating oil prices (not low enough for sure) as I’m sure everyone does, I fear the impact on the economy could be devastating if this is not resolved and soon. High gas adn heating oil prices would certainly affect me , my family and my friends adversely.
The heavier inert gasses displace oxygen and is heavier than air. There should still be a layer of inert gas between the combustible and the oxygen at least for a while. I’m not talking about eliminating fires or explosions, I’m talking about delaying them at least so people have a few precious seconds to run away.
Think about it … how long does an inert gas blanket really have to last providing there are no external sources of ignition ? Till the cars come to a rest right?
In Megantic , death was immediate or nearly so… if they had 1 to 5 minutes before an explosion to get away many lives would have been preserved.
I’m actually not sold on F couplers on tank cars other than TIH cars. In my experience a minor derailment can be turned into a major derailment because of one car dragging the next one off the track and so forth.
I copied some of this from a previous thread that got lost in a dick measuring contest.
Some interesting ideas. If I were to critique it would say that you are looking for a technical solution whereas the problems may be organizational in nature. Your solutions likely wouldn’t have changed the outcome much in the Quebec derailment I think (it was going too fast around a curve), or a runaway on a mountain pass like Cajon, but may make a difference in a relatively low speed derailment like the one on BNSF. It WOULD have worked in the ethanol crash on CN in Illinois I think. The problems in the Quebec crash were mostly organizational in nature, with MM&A trying to get by as cheaply as possible on things like crew costs, track maintenance, locomotive maintenance, field supervision (didn’t even check a unit that was on fire!), facilities/ yards to tie the trains down, etc.
I don’t expect this proposal to require excessively heavy tank cars. The point of the solid drawbars is to keep the cars coupled during a derailment as they are placed under tension by selective braking and power application to prevent them from jackknifing and piling up. That is a two-part objective. The larger part of that objective is accomplished by eliminating the knuckles and related features that can break and disengage from twisting and bending as well as pure pulling stress. There will be a lot of that twisting and bending coupler stress because the cars will be running on the ground, tearing up track and plowing ballast, so what is needed is coupler integrity that goes beyond the abilities of tightlock or shelf couplers.
The second part of the objective is the capability of the drawbars to handle the tensile pull of the selective braking and power application from the moment the derailment begins to when the train stops moving. I don’t know whether that would subject couplers to a higher than usual tensile pull. If it does, it would require heavier drawbars and possibly more weight added to reinforce the tank. When I mentioned center sills, I am not necessarily suggesting full, independent center sills with the tank sitting on top of them. I am only referring to the structural features integrated with the tank that reinforces the pulling line of the car. It may only be a thickening of steel along the bottom. They simply replicate the purpose of a center sill.
What’s interesting in this whole thing are the parallels to arguments about passenger trains. We all know the schools of thoughts regarding safe operations of passenger trains:
Make the cars lighter so they can go faster and implement track and signal engineering to avoid crashes
-or-
Make the cars heavier, tougher (and therefore slower) so that if they crash, the occupants survive.
I think the “best” solution (because face it, there are no perfect solutions in life - esp. since I believe in the idea of equivalent exchange) will be a combination of both.
No real information in this post, but i think the similarity in passenger trains vs oil trains is interesting. Especially since there are safeguards when operating pax trains in freight territory (absolute blocks in some places, for example).
One thing to consider in these 90 or so permanently connected drawbared trains, what happens of you have a hotbox enroute, a defect on one car that will prevent the entire train from moving. Normally if you get a hotbox, its 4mph car until it is set out.
Many of the suggestions by the Original Poster, Randy, and others echo those of John G. Kneiling (Trains’s “Professional Iconoclast”) as set forth in his 1969 Kalmbach Publishing Co. book Integral Train Systems, and several columns and articles in the 1965 - 1975 time frame about the railroads should attempt to compete to get the oil move itself, instead of just “How Do We Route The Pipe?” (the title of one such column).