Longer Trains Cause More Derailments

[BaltACD]

Euclid

blue streak 1

I have had a problem of it just being length of trains.  All buff and draft forces are due to how many couplers are on a train.  Each coupler allows for some slack. Also any slack action from cushioned cars.  Now what about a 15 - 20k IM train made up of 3 and 5 packer well cars.?  It might have as few couplers as maybe a 5000 - 8000 regular manifest train. 

It would be nice if someone created a graphic computer model that demonstrates the actions of so-called “in-train forces” and how they act in a long freight train as it travels over varying track geometry at varying speeds.

 

...

It has already been done.  I was given a opportunity by one of the Senior Road Foremen of Engines about 30 years ago to operate one of CSX's Engineer Training Simulators.  Simulator could be programmed, on demand, to run any Subdivision on the property and any train that existed on the CSX Car & Train Database could be loaded - with the engine consist that operated that train IRL.

When making throttle and/or brake inputs there was a graphic representation of the various intrain forces that were being generated with the territory being negotiated.  At the time, CSX's maximum train length on most territories was 9000 feet as this was well before the implementation of PSR principles.

[Euclid]

blue streak 1

I have had a problem of it just being length of trains.  All buff and draft forces are due to how many couplers are on a train.  Each coupler allows for some slack. Also any slack action from cushioned cars.  Now what about a 15 - 20k IM train made up of 3 and 5 packer well cars.?  It might have as few couplers as maybe a 5000 - 8000 regular manifest train.

 

It would be nice if someone created a graphic computer model that demonstrates the actions of so-called “in-train forces” and how they act in a long freight train as it travels over varying track geometry at varying speeds.

 

There is a lot of talk that the advent of much longer freight trains comes with the danger of more high speed derailments.  They say this is obvious because the longer trains are heavier and thus more likely to derail.  But why should that be just a forgone conclusion?  Railroad labor advocacy claims this to be a fact as though it should be obvious.  Apparently, so does the FRA. 

 

I would conclude that slack is not the primary creator of buff and draft forces.  The primary creator is locomotive tractive effort and braking.  There would be buff and draft in a train even if it had solid drawbars and no slack at all. 

 

But what slack does allow is the buff and draft forces being able to change one joint at a time, and thus accumulate their force so that when it is finally arrested against a hard stop, it contains the kinetic energy of a long string of cars, rather than of just one car.  And if those cars happened to have slack bunched, they will collide with all of their kinetic energy being released at once.

 

So say you have a solid wall across the track and rolled one car into it at 5 mph. The wall and car would absorb the impact from one car moving 5 mph.  But if you have 50 cars rolling at 5 mph, and if their slack is bunched and the first car hits the wall, the wall and first car would absorb the impact from 50 cars moving at 5 mph.  How much damage would that do to the first car?  It would probably cause a derailment starting with the first car.

 

However, in most cases, you will not have a condition where the effect of a solid, stationary wall exists because the train is moving and once it stops, all slack force issues cease.  All of the conditions for derailment-causing excess forces will depend on the following factors: 

 

Slack positions

Distribution of loads and empties

Train speed

Track curves and grades.

Distribution of power and braking

 

 

Since proper Distribution of power and braking will manage the other conditions of slack positions, distribution of loads and empties, train speed, and track curves and grades; the Distribution of power and braking can be increased to naturally match an increase in train length.   Under that assumption, trains length can be increased indefinitely without any added risk making derailments more likely due to increasing in-train forces. 

 

So the only question I see is whether Distribution of power and braking be adequately established and controlled. 

[Overmod]

blue streak 1

I have had a problem of it just being length of trains.

The physics of train length and handling aren't part of the actual discussion, any more than actual roller-bearing safety through increased carman inspections is.  That we see so much reference to not blocking crossings and keeping more people mandated for various operations... carefully engineered to involve "safety"... gives it away.

Now, mandating a limit in train length between DPU units... that would make sense.  So would regulation of train makeup, which so many railroads have demonstrated lethal incompetence in arranging.  As you note, long consists of well cars, even baretable, have far less 'unsafe' potential than uncoordinated long-travel cushion cars.  (One of the fun things advanced ECP makes possible, although lost on the Feinberg contingent, is that individual braking rates can be modulated during an application to control likely intercar and interblock dynamics.)

[blue streak 1]

I have had a problem of it just being length of trains.  All buff and draft forces are due to how many couplers are on a train.  Each coupler allows for some slack. Also any slack action from cushioned cars.  Now what about a 15 - 20k IM train made up of 3 and 5 packer well cars.?  It might have as few couplers as maybe a 5000 - 8000 regular manifest train.

[Euclid]

When I asked the question in the thread title, these are the terms and details of that question:

 

Compare an example of two trains:

 

TRAIN #1 is 150 cars long with distributed power optimally arranged throughout the train.  It has a mixed consist, which is made up of empties and loads, which are distributed throughout the train according to commonly accepted good principles.

 

TRAIN #2 is 250 cars long with distributed power optimally arranged throughout the train.  It has a mixed consist, which is made up of empties and loads which are distributed throughout the train according to commonly accepted good principles.  The additional 100 cars of this train are similar in weight and distribution to the cars in TRAIN #1.

 

Both trains are tested on the same sample of average track with frequent curves and a variety of grades.  Ambient conditions are average and identical for both test trains.  Operating speeds are identical for both trains.  Both trains are operated by the same engineer who is well qualified.

 

QUESTION:  What is the comparative likelihood of each train derailing due to excessive in-train buff and draft forces that cannot be adequately controlled and limited by the train handling of the engineer?

 

The purpose is to compare these two trains to determine if the buff and draft forces of each train are identical, or if those forces are higher in the 250-car train compared to the 150-car train.    

 

The ultimate purpose of the comparison is to determine whether the longer train has a higher probability of derailing due to having higher buff or draft forces. 

 

Another way to look at the question:  As the train length increases, so too does the distributed power and dynamic braking.  It may be that the power and braking continues to be capable of avoiding excess buff and draft forces as the train length increases.  Or it may be that the ability of power and braking reaches a plateau where there is no longer sufficient control to limit buff and draft forces enough to prevent derailments. 

 

If it is found that the longer train does have a higher probability of derailing due to these higher forces, that would lead to inquiry as to what can be done to reduce those forces with the 250-car train to make them the same as the with the 150-car train.  This could look at the distribution of power and its dynamic braking, engineer operating techniques, and computer control by a program of the train optimum train make-up, route geometry, weather conditions, and other variables. 

 

The point of this explanation is just to show the objective, but not to actually design the testing for the effects of the in-train forces.  The tests imply a need for measuring the in-train forces and the time and location of their occurrence.  This would seem to require using actual revenue trains with strain gages on their couplers.  The logistics of that seem to be impossible.   

 

The FRA is asking railroads to upgrade crew training to keep up with increasing train lengths.  This would require research to discover what is needed in the upgraded crew training. 

[dpeltier]

Euclid

Here is some news about this question.  Does anybody actually know the answer?

 

Short answer? No. The best available evidence I can find is a paper written by my good friend Darwin Schafer back when we were grad-school office mates:

https://railtec.illinois.edu/wp/wp-content/uploads/pdf-archive/TRBDarwin---The-Relationship-Between-Train-Length-and-Accident-Causes-and-Rates.pdf

It notes that the traditional model is to divide accident causes into train-mile-related causes and car-mile-related causes, and if you believe in that model then increasing train lengths will always increase the derailment rate per train-mile and decrease the derailment rate per car-mile (i.e. decrease the derailment rate overall). The paper observes in passing that certain categories - notably train-handling - don't always fit neatly into either category, and warns that you can't conclude from this study that longer trains reduce derailments over all (especially for trains > 150 cars, which were not well represented in this 2007 study). It's also worth noting that several "train-mile-related" causes such as passing red signals or exceeding speed restrictions have been reduced in importance by the widespread use of PTC, so theoretically that dimishes the safety advantage of longer trains.

That said - there certainly is not any study out there to suggest that longer trains result in MORE derailments. The labor leaders making this claim are spouting pure BS with nothing to back them up.

FRA thinks that some recent derailments may be due to train makeup procedures, which are kind of indirectly related to train length. But even if that's the case, there's reason to doubt that this small number of accidents would overcome the safety benefit of having fewer (i.e. longer) trains out and about and getting into trouble.

And if you step back and use a different safety metric - lives lost, rather than number of derailments - there is absolutely no question in my mind that longer trains save lives. That is because the vast majority of all lives lost in railroad accidents occur in grade-crossing and trespasser collisions, which aren't affected at all by train length. Longer trains means fewer trains means fewer deaths, period.

Dan

[dpeltier]

SD70Dude

An excerpt from the following report:

"In 2005, the TSB conducted a safety issues investigation involving an extensive analysis of train derailments and their relationship to bulk tonnage traffic.  Loaded high-capacity rail cars in unit trains pose special problems to main lines where weak track conditions (ties, ballast, and subgrade) may be common. A unit train consist is usually uniform; that is, all cars are of the same design and loading, with the car trucks and car bodies responding more or less as one unit. Therefore, each rail car on the train responds to track irregularities in the same manner as the previous car, leading to cumulative impacts at irregularities that the train encounters in the track structure. Trains with numerous rail cars of the same design and with high load capacity provide the track little or no opportunity for elastic recovery during their passage. As a result, high-capacity unit trains can hasten permanent and usually non-uniform track deformation."

https://www.tsb.gc.ca/eng/rapports-reports/rail/2020/r20w0102/r20w0102.html

Long, heavy trains will have greater in-train and train-track forces than shorter trains.

The 2005 study itself is here:

https://www.tsb.gc.ca/eng/rapports-reports/rail/etudes-studies/siiR0501/siir0501.html

You have quoted the 2020 report accurately, but the 2020 report refers misleadingly to the 2005 study. The 2005 study does not purport to show that the statement is true, it just sets forth the statement itself, without references or support. It is not a conclusion of the 2005 study, it is a hypothesis that is then used to justify the conclusions of that same study, and if that sounds messed up... it gives you a pretty good idea of how well that 2005 study was put together. It's junk. But that doesn't stop a careless person from citing it in a future report.

Dan

 

[SD70Dude]

An excerpt from the following report:

"In 2005, the TSB conducted a safety issues investigation involving an extensive analysis of train derailments and their relationship to bulk tonnage traffic.  Loaded high-capacity rail cars in unit trains pose special problems to main lines where weak track conditions (ties, ballast, and subgrade) may be common. A unit train consist is usually uniform; that is, all cars are of the same design and loading, with the car trucks and car bodies responding more or less as one unit. Therefore, each rail car on the train responds to track irregularities in the same manner as the previous car, leading to cumulative impacts at irregularities that the train encounters in the track structure. Trains with numerous rail cars of the same design and with high load capacity provide the track little or no opportunity for elastic recovery during their passage. As a result, high-capacity unit trains can hasten permanent and usually non-uniform track deformation."

https://www.tsb.gc.ca/eng/rapports-reports/rail/2020/r20w0102/r20w0102.html

Long, heavy trains will have greater in-train and train-track forces than shorter trains.

[Fred M Cain]

Well, aside for the serious issue of slack action on super long trains, there is also the mathematical law of averages.

If there is a chance, however small, that a defective car can cause a derailment in a 50-car train, those chances double when you go to 100 cars.  Then, they double AGAIN when you go to 200 cars.  I think the NS derailment in Springfield, OH, involved a train with 200+ cars.

My hunch is that there must be a way to make long trains operationally safe.  We just need to find a way to "push the right buttons" so to speak.

[Euclid]

Here is some news about this question.  Does anybody actually know the answer?  Railroad labor advocacy seems to frequently cite this increase of danger with monster trains.  However, that view is understandable because everyone agrees that monster trains reduce crew costs, and thus cause a loss of jobs.   The media obviously jumped onto that bandwagon in response to the East Palestine wreck, which was then affirmed by the Springfield, OH wreck.   

 

This latest news suggests that engineers may need more training to know how to operate the longer trains. 

 

 

https://fortune.com/2023/04/28/railroad-derailment-long-trains-federal-regulators-problem-contributor/

 

 

The feds are warning railroads that their love of long trains is leading to horrible accidents and derailments—but they’re not doing anything about it yet