Concept for a Safe Oil Train

[Overmod]

Euclid

The point of my proposal is to prevent or reduce the pileup effect of a derailment at speed.  Such pileups have the potential to produce an enormous crushing force that is capable of rupturing tank cars even if they are strengthened to the newly planned standards of crashworthiness.

One of the interesting things to me about the way this thread has been developing is that much of the 'opposition' to the plan involves the exclusive use of very heavy/strong drawbars (without much discussion of the draft gear arrangement between them and a presumably-armored center sill).  To the full extent that the braking system mods can keep a derailed car or cars  -straight' while decelerating, a more normal combination of drawbar, and 'antitorque' provision will fully accomplish the job.  Meanwhile, if the braking fails to work as necessary, just about any drawbar that has meaningful draft-gear attachment will not prevent a tendency for car ends to misalign up to accordioning...the 'real' issue being more that throwing all the torque and eccentric loading on what amounts to no more than two knuckles and pins is more likely to produce separation than whatever is in the draft gear at either end of one of the drawbars.

I don't see anything about maintaining, say, a ten-car set of dedicated unit-train tanks than would be required for something like a FuelFoiler, and it's not 'that' difficult to devise means of handling and servicing road failures of the kinds that have been hinted at.  On the other hand, you will NOT want 60-car blocks of oil cars semipermanently joined -- this is not a coal train, and isn't run or serviced as one.  What I might concentrate on, at this point, is what's involved in making the drawbars field-demountable without compromising their strength -- and providing means to disconnect the electrical and air, etc. in the field under inclement conditions, etc. -- and designing a proper torque-resistant kind of coupling (it can be proprietary in this intended service) that has the strength expected of a 'drawbar' while being easily mated and separated by normal procedures.

[Euclid]

I don’t know that the solid, semi-permanent drawbars need to be stronger than usual.  The weakest point may be the draft gear, as some have suggested.  But can’t we just omit the draft gear?  Normally, you have coupler slack and draft gear slack.  Coupler slack is needed to provide enough lost motion to free the pin to drop.  Draft gear slack is only needed to cushion coupler slack as it responds to braking propagation and power application. 

So, eliminate the couplers and there is no need for draft gear; and also no more slack.  And it goes perfectly with the ECP brakes which apply simultaneously rather than propagating through the train over time. 

ECP brakes and no slack is a quantum leap in train handling performance potential.         

[Norm48327]

BaltACD

Good idea Balt.

I'll bring the butter and salt.

This thread is going nowhere but in Bucky's usual circles.

[schlimm]

Norm48327

This thread is going nowhere but in Bucky's usual circles.

I don't think that is true in this case.  You, dhusman, Dave Klepper, Overmod and others have made a lot of interesting and useful posts well beyond Euclid/Bucky's.

[daveklepper]

Overmod

daveklepper

With magnetic track brakes, it goes into friction between the magnetic shoe and the rail.   Rail wear? Sure.

I thought at least some of this was eddy-current braking.  ISTR some of the streetcar brakes working this way, with nonfrictional contact being touted as an advantage.  Surely that makes better sense than mutual magnetic attraction with a friction acting at right angles?

And have you actually calculated the amount of magnetic current, and the path for the lines of force, necassary to provide orthogonal clamping force of the designed magnitude when the only available material for the magnet to attract is... a rail section? What's saturation for a length of rail?

When I run some preliminary numbers and compare them against the mass of loaded freightcars, I do NOT see 'automobile-like' stopping distances.   Can you provide some sample physics, including required currents (and friction material on your track shoe or sled) for substantiation?

I repeat (perhaps erroneously) that I expect some leverage transfer of car weight onto the frictions before a system of track brakes actually produces short stopping distances -- this would use 'electromagnetism' more for the deployment and release control than to exert the force to induce friction.  Not saying that would be a good idea as a default emergency-brake option...

1.  It is a combination of eddy current and firction.  Both produce heat.  But the rails are obviously better heat sinks than even 400 freightcar wheels.

And you are again correct, that is intended iin an oil train, FOR EMERGENCY USE ONLY

But the point is, this is all existing technology getting a new application.   The derailment detector is just repurposed slip control apparatus.  Magnetic track brakes have been around since 1912, maybe earlier.  I am not going to argue whether couplers or drawbars are better.  For a test train, I would like to see couplers so the coupler technology can be perffected and demonstrated.   I think the train should be tested in a variety of lanes and lengths of consists.  Some with distributed power.   Because of electrical control, even normal braking will be faster, and the uniformity of cars will help with electric brake control in train handling.   But it should also, at the engineer's discrsion, allow for normal airbrake operaton, but with the graduated release feature taken from passenger equipment, to allow slack control when desired.  In all of this, I would give the engineer as many options for brake and power control as possible.

Magentic track brakes were used on both the Cincinnati and Lake Erie "Red Devils" and on the Indiana High Speeds, and the latter ran in three and occasionally four car trains.  This is admittadly a far cry from 100 cars.   That is why I stress that the cars must be identacle.  Possibly a reason for couplers is that train length must exactly fit the load.  We cannot have empties and fulls in the same train, either all full or all empty.  (As experience is developed, this problem can also be solved with two settings for each car.)

It is not the weight of the car that causes the friction, but the magnetic attraction.  Stopping like an automobile may be an exaggeration considering the greater weight of a loaded tankcar compared with an interurban or light rail car or streetcar.  But it will certainly be a very great improvement over disk or tread brakes alone.

[Norm48327]

schlimm

Norm48327

This thread is going nowhere but in Bucky's usual circles.

I don't think that is true in this case.  You, dhusman, Dave Klepper, Overmod and others have made a lot of interesting and useful posts well beyond Euclid/Bucky's.

schlim,

I think rather than trying to stop a derail once it starts it's better to prevent it to begin with. That was the emphasis of the article in the March issue. The company developing this equipment in cooperation with BNSF is aiming to catch any and all defects prior to an incident so the offending car can be set out before trouble begins. I think that is a much better and less costly approach than redesigning a train.

[daveklepper]

Is not that exactly what I am trying to do with the repurposed slip control used as a derailment detector and emergency magnetic track brakes to suppliment the usual tread brakes?

[Euclid]

Norm48327

I think rather than trying to stop a derail once it starts it's better to prevent it to begin with.  I think that is a much better and less costly approach than redesigning a train.

Adding a lot of sensors does lead to what might be called a redesigned train.  It takes sensors, cabling, connectors, processing, and a program of detection and response. 

But more to the point, I am not proposing the control of derailments to the exclusion of preventing them.  It does not have to be one or the other.  Indeed, I have mentioned several times that the electronic cable that is fundamental with ECP brakes also lends itself to transmitting data from sensors not directly related to ECP.  Those extra sensors can prevent derailments by detecting trouble before it causes a derailment. 

But I don’t think it is realistic to expect to prevent all derailments by the use of sensors.  Derailments can be caused by train or track.  Sensors can be applied to both.  Here is a document that details the prevention of freight train derailments by the use of sensors and ECP brakes:  http://www.era.europa.eu/Document-Register/Documents/DNV%20Study%20-%20Final%20A4%20Report%20-%2020110419%20-%20Public.pdf

So I see nothing wrong with using sensors to prevent derailments.  But what I am proposing here is to mitigate the destructive nature of derailments that are bound to occur despite sensors.  I believe this idea of controlling derailments has barely, if ever, been explored before. 

[Norm48327]

"I believe this idea of controlling derailments has barely, if ever, been explored before."

Probably because the laws of physics have never been repealed.

[Euclid]

When it comes to controlling a train derailment, you need a tool that is big and powerful.  The best tool is a train. 

[Anonymous]

Norm48327

BaltACD

Good idea Balt.

I'll bring the butter and salt.

This thread is going nowhere but in Bucky's usual circles.

Why did Bucky change his name?

[dehusman]

daveklepper

It is not the weight of the car that causes the friction, but the magnetic attraction.  Stopping like an automobile may be an exaggeration considering the greater weight of a loaded tankcar compared with an interurban or light rail car or streetcar.  But it will certainly be a very great improvement over disk or tread brakes alone.

 

Do you have any information on how the track structure will stand up to absorbing the force of stopping 100 286k tank cars in 200 ft?  Will the ballast and tie structure be able to handle that amount of loading? Will it shove the rails out of alignment in a curve?  The conventional system puts a lot of the energy into heat either burning the brake shoes or burning flat spots in the wheels.  Even at that the rail at the bottom of a hill can be shoved out of alignment by heavy braking.  With magnetic systems  there will be a lot more force transferred to the rails (and thus ties and ballast).  Stopping an oil train is several orders of magnitude higher than stopping a trolley car.  Will a stop by a magnetic system load up forces in the rails that may result in a subsequent train derailing if it stops in the same area?

[dehusman]

Euclid

 I believe this idea of controlling derailments has barely, if ever, been explored before. 

 

If you ignore air brakes, shelf couplers, guard rails in curves and on bridges, dragging equipment detectors, runaway tracks, etc. you are correct.

[Euclid]

Murray

Why did Bucky change his name?

Euclid is the name; father of geometry is the game.

[Anonymous]

Euclid

Murray

Why did Bucky change his name?

Euclid is the name; father of geometry is the game.

Thanks for the clarification Bucky.

[Euclid]

dehusman

Euclid

I believe this idea of controlling derailments has barely, if ever, been explored before. 

If you ignore air brakes, shelf couplers, guard rails in curves and on bridges, dragging equipment detectors, runaway tracks, etc. you are correct.

The best examples are guard rails and shelf couplers.  I don't think runaway tracks count.  Air brakes stop the train, and it is good for a derailing train to stop.  But they also often cause derailments or make a derlailment worse.

But I don't think anybody has proposed the effect of controlling that I am suggesting.  Maybe it has been proposed.  If so, I would be interested in seeing the proposal.  It might be an objective that has been targeted for passenger trains. 

[tree68]

Euclid

But I don't think anybody has proposed the effect of controlling that I am suggesting. 

There may be a reason for that...

[Norm48327]

Euclid

Murray

Why did Bucky change his name?

Euclid is the name; father of geometry is the game.

For this discussion perhaps you should consider "Newton" and the laws of physics.

[Norm48327]

Murray

Norm48327

BaltACD

Good idea Balt.

I'll bring the butter and salt.

This thread is going nowhere but in Bucky's usual circles.

Why did Bucky change his name?

Long, drawn out story behind that.

[Euclid]

Dave Klepper,

As I mentioned earlier, I wanted to ask you for some clarification on your idea using magnetic brakes.  If I recall correctly, you mentioned that your idea will prevent derailments as opposed to only mitigating their destructive effects as I am proposing.  However, I can’t locate the post in which you said that, so maybe I am confusing it with someone else’s point. However, if you did say that, how will your idea prevent derailments?

 

daveklepper

But I am assuming that the whole train acts as a unit, and under those circumstances, stoping the entire train behind the point of accident is certainly by far the safest procedure.  For the portion of the train forward of the accident (assuming there is break-in-two at the accident), the override button will provide the engineer to judge this situation and do what is necessary to minimze any damage first to people and then to property.  

 

In this quotation of one of your earlier comments, what do you mean when you refer to stopping the entire train behind the point of the accident?

 

[Anonymous]

I can only imagine.......

[Anonymous]

tree68

Euclid

But I don't think anybody has proposed the effect of controlling that I am suggesting. 

There may be a reason for that...

Indeed......

[RRKen]

Bucky

As I have mentioned, onboard detectors or sensors will be a key component of the oil train concept that I am proposing. 

When you have sold one of  your trains, let us know how it works.  Good luck selling it.

[Murphy Siding]

Euclid

Murray

Why did Bucky change his name?

Euclid is the name; father of geometry is the game.

  Pay no attention to the man behind the curtain.

[daveklepper]

Stopping all the cars as a unit behind any car that has derailed is what I mean, and as quickly as possible. Do you understand that?  With all cars identacle, electric control of brakes, and magnetic track brakes supplementing the usual tread brakes.

I do NOT agree with Euclid's idea of trying to guess what the derailed car is going to do and modifying the braking accordingly.  This is an impossible task for any computer system.  The handling of the train forward of the break-in-two and derailment should be left up to the engineer. with an override button that can release the brakes if he so chooses.

[Murphy Siding]

dehusman

daveklepper

It is not the weight of the car that causes the friction, but the magnetic attraction.  Stopping like an automobile may be an exaggeration considering the greater weight of a loaded tankcar compared with an interurban or light rail car or streetcar.  But it will certainly be a very great improvement over disk or tread brakes alone.

 

Do you have any information on how the track structure will stand up to absorbing the force of stopping 100 286k tank cars in 200 ft?  Will the ballast and tie structure be able to handle that amount of loading? Will it shove the rails out of alignment in a curve?  The conventional system puts a lot of the energy into heat either burning the brake shoes or burning flat spots in the wheels.  Even at that the rail at the bottom of a hill can be shoved out of alignment by heavy braking.  With magnetic systems  there will be a lot more force transferred to the rails (and thus ties and ballast).  Stopping an oil train is several orders of magnitude higher than stopping a trolley car.  Will a stop by a magnetic system load up forces in the rails that may result in a subsequent train derailing if it stops in the same area?

  Kinda reminds me of the saying that falling off a tall cliff won't hurt you.  It's the sudden stop at the bottom that kills you.

     If you have 100 tank cars screeching to a halt, isn't that enough transfer of forces to simply peel up the track...and derail the train?

[Euclid]

daveklepper

Stopping all the cars as a unit behind any car that has derailed is what I mean, and as quickly as possible. Do you understand that?  With all cars identacle, electric control of brakes, and magnetic track brakes supplementing the usual tread brakes.

So, with the system you are proposing, all of the cars in the train would have magnetic track brakes, and some form of electrically controlled air brakes.  Then when a derailment is sensed, all of the track brakes behind the point of derailment apply automatically to a maximum application force; and all of the electrically controlled air brakes remain released.  And all of the magnetic track brakes ahead of the derailment remain released. 

Then, as you say, the engineer manually controls the electrically controlled air brakes only on the cars ahead of the point of derailment.  I don’t know anything about magnetic track brakes, but I assume they are a supplementary brake system for quick emergency stops, and are either fully applied or fully released. 

In any case, I wonder about this:  Say a sensor detects a wheel on the ties about the middle of a 100 car train, and it applies the magnetic brakes fully on the 50 cars behind the point of derailment.  If the stopping force is anything like what you say it is, this would be like a conventional emergency application on steroids. 

When this happens, the first 50 cars will be running with brakes fully released.  With 50 cars running out against 50 cars that are suddenly thrown into super braking, won’t that conflict of forces instantly pull a drawbar or break a knuckle somewhere close to the dividing line between the free running cars and the cars that are in full magnetic track brake application?  I don’t see how the engineer could possibly react fast enough to avoid that outcome even if he instantly made a full emergency application on the leading 50 cars.

And I also wonder about this:  When a derailment is sensed, I assume that the electrically controlled air brake system will be automatically cut out behind the point of derailment.  It cannot remain active throughout the derailing cars because as they derail, they may part and put the entire electrically controlled air brake system into emergency, and thus prevent the engineer from exercising any manual control over that system on the 50 cars ahead of the derailment. 

And also:  When the conflict of force occurs between the 50 cars with magnetic track brakes applied and the 50 cars with all brakes released; and it breaks the train in two near that dividing line, that beak may not be in the 50 cars behind the derailment.  It may be very well be close to the derailment, as expected, but in the cars right ahead of the derailment.    When that happens, the engineer will again face an automatic emergency application of the first 50 cars, and thus lose manual control of braking them.  This would happen despite the fact that the automatic response system has cut off the electrically controlled air brake system behind the point of derailment. 

Do you agree with my conclusions? 

[tree68]

So it appears that we're all in agreement that differential braking won't work.

Semi-permanent drawbars won't make a difference because they still have to have enough play for the train to go through turnouts and around fairly sharp curves.  The play necessary to do so will allow a car to skew if the forces pushing on it from behind are great enough - and probably enough to force it off the rails and into an accordian if the cars ahead of it have stopped fairly abruptly.

Most, if not all, hazmat tank cars already have shelf couplers, which have the same effect as the suggested drawbars.  Thing is, the only thing keeping the trucks on the cars is gravity.  

I'd say the whole idea is "busted," as they say on Mythbusters. 

[zugmann]

Why don't we like slack or knuckles?  Sometimes I wonder how many broken knuckles would have turned into a major derailment (or car torn in half) if it wasn't for the broken knuckle?

[Overmod]

tree68

So it appears that we're all in agreement that differential braking won't work.

Speak for yourself.  I see no showstopping reason why differential braking, EITHER with ECP or with some type of electromagnetic track brake, is not possible; I think there are good reasons to investigate its utility... particularly considering the relative lack of other approaches that will deal with an evolving 'derailment event.'

If we accept the argument about 'play' in the drawbars leading to skew and then accordioning, then the immediate sensible conclusion that follows is that "the cars ahead of it must NOT stop fairly abruptly" ... which again is something involving controlled braking.

The shelf couplers only have the effect of preventing vertical disengagement of the knuckles (and exposure of a 'naked' knuckle to punch through an adjacent tank).  This still leaves relatively small components to take loads like -- well, like the skew loads that were just mentioned; I'd be surprised if a pair of knuckle pins close to the effective center of the 'bar' being bent would survive better than whatever arrangement prevails at the outer ends of a drawbar connection.  (I also wonder whether the prismatic contour of a type F makes it a better 'penetrator' in off-axis contact... but that's probably a minor point.)

I repeat that I think it was something of a mistake for Euclid to defend the heavy drawbar-and-underframe idea so protractedly early on, as it detracted from the better argument that his solution depends highly on differential braking to keep the derailed cars from skewing.  The 'other stuff' is basically just the Miller coupler arguments with more mass and longer consists involved (neither of which is trivial, of course, but the basic idea of keeping the train together and unskewed at all costs is still inherent.)

Problem is that you can't predict how any particular event is going to evolve.  In the Lac Megantic wreck, I don't see how any system of 'keeping cars together' would have helped with the outcome, Dave's brake would either have prevented it early or been of no practical use, and an ECP brake not equipped with remote modulation would have been of little or no help either. The anecdotal Miller coupler results all ippeared to involve situations where the connected consist could run off the track in a fairly straight line until it stopped, with the avoidance of telescoping being the primary 'end'.  Derailing a long heavy consist, on an excessively sharp curve near the middle of town, with both ends now being pulled toward the point of derailment, isn't going to end well no matter how the cars are connected together...

On the other hand, wrecks that involve casual contact between derailed cars and adjacent consists may be very well addressed by a system that holds the cars in line and prevents them from torquing over to the point their structure is 'compromised' (charming expression!) by collision with something else.  Now, I confess that if I were the mad inventor, I'd be thinking about shields at the top of the car that would protect against oblique impact at the upper ends, and 'self-restore' the car to upright position as far as possible after contact, as there is likely no arrangement of drawbars capable of going around curves that will be assured of holding the car fully upright in a derailment event... a topic that has not so far been given much play.