Oil Train

[Euclid]

Posted by zugmann on Monday, March 30, 2015 10:27AM   [QUOTE]                     

Euclid [said]

So your improved car is working extra hard and wearing out brake shoes faster, but with little net benefit to the other guy’s train, and no benefit to you.

zugmann said:

It's a lot easier, quicker, and cheaper to replace worn brake shoes than to replace wheels. Sliding wheels = flat spots.

-----------------------------------------------------------------

What I am talking about does not amount to a choice between those two options.  It is a choice between replacing worn out brake shoes or not needing to.  As I said, the point of adding the load detector is not to prevent empty car wheel slide. 

[zugmann]

Euclid

  As I said, the point of adding the load detector is not to prevent empty car wheel slide. 

 

Source?

[Euclid]

zugmann

 

Euclid

  As I said, the point of adding the load detector is not to prevent empty car wheel slide. 

 Source?

 

It is here:

 

http://www.railway-technical.com/brake2.shtml#EmptyLoadSensors

[zugmann]

Here's Wabtec's patent:

 

https://patents.justia.com/patent/20100283316

 

Brake equipment for railway freight cars typically employs dual capacity empty/load equipment which adjusts the brake application force according to the empty or loaded conditions of the freight car. In such dual capacity empty/load equipment, a two-setting control is provided where normal brake pressure is realized under full load conditions and a reduced or modulated brake pressure is realized under an empty load condition. In contrast, single capacity brake equipment, which produces a brake application force independent from the load condition of the car loading, is susceptible to wheel lock and sliding wheels due to the same brake force being applied to an empty car as a loaded car. Sliding wheels undesirably cause flat spots on the wheels as well as decreased brake performance. By modulating the brake pressure under empty load conditions using dual capacity empty/load equipment, the occurrence of sliding wheels is reduced or eliminated

[zugmann]

Euclid

What I am talking about does not amount to a choice between those two options. It is a choice between replacing worn out brake shoes or not needing to. As I said, the point of adding the load detector is not to prevent empty car wheel slide.

From the catalogs and patents I read, the load detector doesn't increase braking force.  It is a device for reducing it for empites.

[tree68]

Euclid

The only way this load sensing/enhanced braking makes sense to me is if all railroads agreed to install this on all of their cars.

Since the railroads no longer own the majority of the cars running on their lines, this point is really moot.

You're sounding a little schizophrenic here.  In one post you note that your concept of how "adjustable braking" works agrees with what Al Krug wrote - braking force is added to loaded cars.

But then you write

2)    Add a load sensor that causes the pneumatics of the brake system to reduce the brake force when the car is empty.    

 While this is more or less correct, it reverses the role of the load sensor.  If the sensor Al describes fails to actuate, braking will remain at "stock" levels.  This would  result in lesser braking effort, but would minimize the possibility of sliding wheels.  If the sensor you describe fails to actuate, the car will operated at increased braking levels at all times, increasing the probability of sliding wheels.  You're making this a lot more complicated than it really is.  

[Euclid]

Larry and zugmann,

I understand what you are saying, and this has been a point of confusion through most of the previous page.  The following explanation is as clear as I can explain my understanding:

My understanding of load sensors has been this:   They are intended to provide greater braking power for loads. This goal is accomplished by two modifications to a freight car as follows:

1)    Modify the pneumatics of the brake system to increase the maximum brake force to as high as possible when the car is loaded.

 

 2)    Add a load sensor that causes the pneumatics of the brake system to reduce the brake force when the car is empty.    

So adding a load sensor is not fundamentally intended to prevent wheel slide on empties, although this point can be confusing because in item #2, load sensor does do that in order to achieve the basic goal of item #1. 

So the purpose of the load sensor itself is to reduce the empty car brake force in order to prevent wheel slide.  But that is not the basic, underlying reason for adding a load sensor to a freight car.  The basic reason is to get more braking power on the loads.

 

 

 

[Euclid]

From the Al Krug reference on load sensors:

“The first step [which correlates to my item #1 above] was to put larger reservoirs on the cars so that the traditional 2.5 to1 ratio of reservoir volume to brake cylinder volume was greater. This will result in higher brake cylinder pressures for any given brake pipe reduction. The problem is that higher pressure will slide the wheels of an empty car. So a pressure limiting valve is attached to the brake cylinder which will vent any excessive pressure to the atmosphere thus limiting braking effort. The exhaust of this limiting valve is open to atmosphere on an empty car allowing it to vent excess pressure. On a loaded car it is closed off so it cannot vent any pressure from the brake cylinder thus taking advantage of the higher pressure which results in higher braking effort.”

 

Clearly the intent is to provide higher braking effort on the loads, although it does also entail reducing the pressure for empties.  But if you did not install this system and raise the braking force as it requires, there would be no need to protect the wheels from sliding on empties. 

[zugmann]

Euclid

. But that is not the basic, underlying reason for adding a load sensor to a freight car. The basic reason is to get more braking power on the loads.

But the load sensor has no way of doing that.  All it can do is exhaust air from the cylinder.  It can't create more air and therefore, more pressure and braking effort. It can only reduce.

 

 

[Euclid]

tree68

You're sounding a little schizophrenic here.  In one post you note that your concept of how "adjustable braking" works agrees with what Al Krug wrote - braking force is added to loaded cars.

But then you write

 

2)    Add a load sensor that causes the pneumatics of the brake system to reduce the brake force when the car is empty.    

 

 While this is more or less correct, it reverses the role of the load sensor.  If the sensor Al describes fails to actuate, braking will remain at "stock" levels.  This would  result in lesser braking effort, but would minimize the possibility of sliding wheels.  If the sensor you describe fails to actuate, the car will operated at increased braking levels at all times, increasing the probability of sliding wheels.  You're making this a lot more complicated than it really is.  

You are leaving out my item #1, which is, the pneumatic modification to raise the brake force for the car by changing to a different reservoir. 

You are right that if the load sensor fails to actuate, braking will remains at stock levels, as you say.  But that stock level has been increased with the change of the reservoir.  Then from that increased level, the load sensor reduces it for empty cars. 

The basic point is to increase load brake pressure, and this is done by changing the car reservoir.  That change then requires the load sensor to protect empties from wheel slide.

Therefore if a load sensor were to fail on an empty car, that higher stock level of braking force would not be reduced for that car, and the wheels most definitely slide. 

[Deggesty]

Euclid

Larry and zugmann,

I understand what you are saying, and this has been a point of confusion through most of the previous page.  The following explanation is as clear as I can explain my understanding:

My understanding of load sensors has been this:   They are intended to provide greater braking power for loads. This goal is accomplished by two modifications to a freight car as follows:

1)    Modify the pneumatics of the brake system to increase the maximum brake force to as high as possible when the car is loaded.

 

 2)    Add a load sensor that causes the pneumatics of the brake system to reduce the brake force when the car is empty.    

So adding a load sensor is not fundamentally intended to prevent wheel slide on empties, although this point can be confusing because in item #2, load sensor does do that in order to achieve the basic goal of item #1. 

So the purpose of the load sensor itself is to reduce the empty car brake force in order to prevent wheel slide.  But that is not the basic, underlying reason for adding a load sensor to a freight car.  The basic reason is to get more braking power on the loads.

 

 

 

 

Quoting from your source, the intent of the load sensor is to  keep wheels on empty cars from sliding:

Empty/Load Sensors

Traction between the wheels and the rail is directly proportional to the weight on the wheels. The amount of traction determines the amount of braking that can be applied without sliding the wheels. Sliding wheels develop flat spots within a few feet. Train cars have a large weight difference between the loaded condition and the empty condition, especially modern coal hoppers and grain cars.

The maximum braking effort of a car must be designed so that when in emergency (when the highest brake cylinder pressure is obtained) the EMPTY car will not slide its wheels. Unfortunately this means a heavily loaded car is under braked even in emergency. A way was needed to allow higher brake cylinder pressures on loaded cars than on empty cars.

(Color added for emphasis by J. Degges) 

[Wizlish]

zugmann

Here's Wabtec's patent:

 

https://patents.justia.com/patent/20100283316

 

Brake equipment for railway freight cars typically employs dual capacity empty/load equipment which adjusts the brake application force according to the empty or loaded conditions of the freight car. In such dual capacity empty/load equipment, a two-setting control is provided where normal brake pressure is realized under full load conditions and a reduced or modulated brake pressure is realized under an empty load condition. In contrast, single capacity brake equipment, which produces a brake application force independent from the load condition of the car loading, is susceptible to wheel lock and sliding wheels due to the same brake force being applied to an empty car as a loaded car. Sliding wheels undesirably cause flat spots on the wheels as well as decreased brake performance. By modulating the brake pressure under empty load conditions using dual capacity empty/load equipment, the occurrence of sliding wheels is reduced or eliminated

That was the way I remembered it working.

I am beginning to think there are two distinct commercially-available systems involved here, one of which increases braking effort for heavier loads, and one which modulates brake pressure for light loads/empties.

It would be theoretically possible to use a single 'arm' sensor and truck sideframe plate for both these systems, assuming a reasonable degree of spring travel between full load and empty position.  That would introduce at least two potential common modes of failure (breakage/bending of the arm, and the previously-mentioned loss of the 'sensor' plate from the sideframe) but it would allow two separate adjustment setpoints for the 'augment' and 'decrement' functions, and at least the possibility of proportional modulation within the 'range of motion' of the arm corresponding to each function's range.

Whether such a thing is cost-effective for many types of car in common interchange service is far from certain.  But we are not discussing common interchange cars, or even regular train service, in this thread. 

What I'd like to redirect the question slightly toward is: Does prevention of wheelslide contribute in any way -- particularly with regard to reduction of either the likelihood or severity of derailments -- toward the safety of unit oil consists?

[Euclid]

What I am saying does not conflict with the above quotes posted by Wizlish and Deggesty.  Here again is the quote from Al Krug’s explanation that I am relying on.  I have broken it up and added my emphasis in red bold.  Al describes two steps to a process that leads to the objective of achieving higher brake force for loads.  

His first paragraph describes the problem that loaded cars are normally under-braked, and what follows is a way to improve their braking.  It is only this objective of increasing brake force on loads that creates a wheel slide problem for empties.  Otherwise, if this objective were not being pursued, there would be no wheel slide problem for empties that needs a remedy.     

 

 

The maximum braking effort of a car must be designed so that when in emergency (when the highest brake cylinder pressure is obtained) the EMPTY car will not slide its wheels. Unfortunately this means a heavily loaded car is under braked even in emergency. A way was needed to allow higher brake cylinder pressures on loaded cars than on empty cars.

The first step was to put larger reservoirs on the cars so that the traditional 2.5 to1 ratio of reservoir volume to brake cylinder volume was greater. This will result in higher brake cylinder pressures for any given brake pipe reduction.

The problem is that higher pressure will slide the wheels of an empty car.

So [as the second step] a pressure limiting valve is attached to the brake cylinder which will vent any excessive pressure to the atmosphere thus limiting braking effort. The exhaust of this limiting valve is open to atmosphere on an empty car allowing it to vent excess pressure. On a loaded car it is closed off so it cannot vent any pressure from the brake cylinder thus taking advantage of the higher pressure which results in higher braking effort.

[Euclid]

Wizlish

What I'd like to redirect the question slightly toward is: Does prevention of wheelslide contribute in any way -- particularly with regard to reduction of either the likelihood or severity of derailments -- toward the safety of unit oil consists?

The answer to that question is no.  But preventing wheel slide is not the fundamental point in applications where it is used.  The fundamental point is permanently raising the brake force for loaded cars by modifying the pneumatic system.  This then requires the load sensor to lower that force for empties. 

The fundamental point of raising the brake force on loads would be of great benefit in reducing the severity of derailments of oil trains.

If this were applied to oil trains in conjunction with ECP brakes, still more benefit would be gained by not requiring load sensors.  This would eliminate all of the problems with the installation and maintenance of load sensors.  It would also eliminate the empty-car wheel slide problems that will occur if a load sensor fails.

With the control of ECP brakes, the high brake force for loads and the low force for empties would be switched for the entire train at once with one master switch.

[zugmann]

Euclid

With the control of ECP brakes, the high brake force for loads and the low force for empties would be switched for the entire train at once with one master switch.

 

With load/MT sensors, we don't need that switch.  It's already being done.  Plus, even the empty unit trains have buffers that may be loads.

[zugmann]

Wizlish

What I'd like to redirect the question slightly toward is: Does prevention of wheelslide contribute in any way -- particularly with regard to reduction of either the likelihood or severity of derailments -- toward the safety of unit oil consists?

 

Sliding wheels lead to flat spots.  Flat spots can lead to broken rails.  Broken rails lead to derailments.  So yeah, preventing wheel slide is pretty important to safety.

[Euclid]

zugmann

 

Euclid

With the control of ECP brakes, the high brake force for loads and the low force for empties would be switched for the entire train at once with one master switch.

With load/MT sensors, we don't need that switch.  It's already being done.  Plus, even the empty unit trains have buffers that may be loads.

 

When you mention the load/MT sensors, how common are these today?  My impression was that load sensors were very rare in the total rolling stock fleet. 

[zugmann]

Euclid

When you mention the load/MT sensors, how common are these today? My impression was that load sensors were very rare in the total rolling stock fleet.

 

Never counted them, but I see them a bunch.  Mostly on covered hoppers.

[Euclid]

So are these a box-like device welded onto a railcar frame with a mechanical arm that contacts the top of a truck frame?

[edblysard]

They bolt on, in a bracket.

See them on tanks, some covered hoppers in plastic service, grain hoppers, and covered steel coil cars...took me a minute to figure out why, then realized the empty was a really light car, the loaded coil cars are really heavy.

And to add to all this, there are a bunch of chlorine tank cars out there that "call  home" with their location, powered by a small solar panel on the top of the tank. 

[Paul_D_North_Jr]

Also some ballast hopper cars (i.e., Herzog's), which have small solar panels to enable them to respond to radio or GPS programmed instructions, etc. to dump their loads in designated locations.  See:

http://hrsi.com/services/plus-train/ 

http://hrsi.com/services/solar-ballast-car-automation/

- Paul North. 

[BaltACD]

With many of the new build cars over the past decade being built to handle the 286K max load and having empty weights less than 60k, the load/empty difference in braking power is effectively mandated just by the sheer 226K difference between loaded and empty states.

[Euclid]

BaltACD

With many of the new build cars over the past decade being built to handle the 286K max load and having empty weights less than 60k, the load/empty difference in braking power is effectively mandated just by the sheer 226K difference between loaded and empty states.

I understand your point.  Here is my question:  Considering the new 1232 tank cars used in oil trains; are these cars routinely manufactured with enhanced load braking and load sensors to reduce braking for empties?  Are these features a routine part of all new 286K cars?

[Euclid]

I am looking at the proposal by TSB of Canada for the new TC-140 tank car to replace the current 1232 cars that are found to breach too easily.  The new TC-140 design includes stronger construction, fitting protection, thermal jacket, etc.  The proposal also mandates ECP brakes for this improved tank car, and describes them thusly:

“Electronically Controlled Pneumatic Brakes - This feature significantly reduces train forces during emergency braking, reducing the likelihood of potential derailments and providing better and smoother braking abilities during train operation.”

 

As I understand it, ECP brakes have been requested by USDOT, but nothing has been mandated yet in current practice or proposal for a future tank car.  I wonder if ECP brakes will be required in the new USDOT tank car rules due in May.  I sense that ECP brakes could create a lot of friction between the railroads and the regulators.     

https://www.tc.gc.ca/eng/tdg/clear-modifications-menu-1193.html

[Wizlish]

One difficulty with 'mandating' ECP  braking is that there is no objective standard for how it would be provided.  That might result in windfall profits for whatever manufacturer provides the solution that is politically chosen ... or incompatible ECP systems, with risk for various types of failure if different manufacturers or 'users' choose different variants for their particular unit operations.

I do think that some form of mandated ECP should be incorporated on oil trains, )although it has been pointed out that a somewhat limited number of observed 'disaster' oil train accidents would have been prevented if ECP braking had been available).  If for no other reason than it provides a market for enough units to cost-down the technology, and provide real-world experience to improve its design and give experience with how to maintain it.

In my opinion, any ECP system on oil trains needs a realtime method of determining actual car weight and proportioning braking ratio; the object of the overall exercise being to reduce both the response TIME and the effective braking DISTANCE without causing problems that can lead to derailments ... flatted wheels being one of the most significant sources of those problems.  That could be done with a calibrated arm-and-plate system (that both augments braking effort in loaded situations and modulates decreases when running light).  It could also involve relatively cheap and simple sensors since so much of the signal-conditioning, detection, and processing work can be handled by properly-designed components in, or added inexpensively to, the ECP system.

I also think that adding active sensing of buff, draft, and draft-gear extension to such a system adds substantial capability compared to its cost -- again, first on dedicated high-volatile crude-oil trains, with the costing-down and experience making the technology increasingly attractive for other services.

 

(Euclid: did you ever resolve how your system was going to differentiate stringlining situations from ordinary derailments?)

[Euclid]

Wizlish

(Euclid: did you ever resolve how your system was going to differentiate stringlining situations from ordinary derailments?)

I am not sure I understand your question.  Could you explain it a little more?

I have quite a few thoughts about combining ECP brakes with sensors for the purpose of preventing derailments and for controlling them after they occur.

[Wizlish]

As I understood the system, it detected derailed car(s), probably via a combination of vibration sensor and load cells, and then acted to apply brakes 'differentially' (with less ahead of the car and more behind it, in order to pull it into line with the track and help keep it coupled).

If the derailment were to occur on a sharp curve, the use of this approach might lead to stringlining across the inside radius of the curve; if there is an active track there, or any obstructions, the use of 'too much' increase of braking behind the derailed cars may add to the problem of collision rather than improving things.

I originally thought that yaw sensors on the trucks might give an indication of what the cars adjacent to a derailed one are negotiating, with the secondary benefit that an excessive or disproportionate amount of yaw, or rate of change, would signal a derailment or other problem.  The question then becomes what happens if the yaw sensors themselves fail, or produce ambiguous or complex results.  I don't think measuring relative angle between cars at the draft gear (to the required precision the required number of times per minute or second) is practical except in unusual circumstances.

[Euclid]

Wizlish

As I understood the system, it detected derailed car(s), probably via a combination of vibration sensor and load cells, and then acted to apply brakes 'differentially' (with less ahead of the car and more behind it, in order to pull it into line with the track and help keep it coupled).

 

 

Okay, I see what you are asking about.  Providing less braking force on the cars ahead of the derailment than on the cars behind the derailment is exactly what I was proposing earlier.  It requires derailment sensors and ECP brakes.  The instant a derailment happens; the sensors detect it and signal the ECP brakes to develop this differential braking on either side of the point of derailment. 

The amount of braking difference is something that would have to be determined.  I suggested that the system would make that decision based on the location of the train on the line, and take into account a wide variety of track, wayside, and roadbed characteristics.  But that alone would be a really big deal to develop.  Not only would the system have to know where the train is, but also, it would have to decide how to react to all of the various characteristics.  So, I am thinking that the system might be developed initially to just provide a modest braking differential that would be adequate for most situations without causing unusual risk in features such as track curves.

Some derailments begin and persist without parting the train for a considerable distance.  During this time (in conventional brake systems), there is no brake reaction caused by the derailment.  So this presents an opportunity to prevent the derailment from progressing to a destructive pileup that will breach tank cars.  Therefore, it would be possible detect a derailment and get the train stopped before a compressive jackknife, accordion pileup begins. 

The derailment sensors of the system I am describing will detect the derailment the instant it happens, even if the train does not part.  The sensors will signal the ECP brakes to begin the differential braking, applying some tension though the derailed car, to help prevent the derailment from becoming more perturbed and thus causing the train to part.  Another point of the differential braking tension will be to provide an assurance against the sudden onset of compression from normal slack run-in of routine braking that happens to be occurring at the time of the derailment, or from variations in track profile under the train.

It is possible that such a derailment could progress to involving several cars that all stay coupled together and generally in line until the train gets stopped by the automatic system.  If this happened on a curve, there would indeed be a stringline force developed that may perturb the derailed cars enough that a parting occurs.  But, in any case, this system would not be intended to prevent every destructive pileup.  Just preventing some of them will be good enough.  Just delaying the onset of a pileup until the train has lost considerable amount of its speed will be of great benefit.

[oltmannd]

ECP braking just replaces the "brakes on" signal, that now runs at nearly the speed of sound down the trainline, with one that operates at near speed of light.

All the other fancy stuff that could be done, like load/empty variable braking, car health monitoring, ride monitoring, bad journal bearing detection, are not part of basic ECP.

ECP provides smoother braking because brake actuation can be faster (no need for damping that keeps pneumatic control valves stable) and will apply simultaneously on every car.

You get faster recharge because the trainline just provides one fucntion - fill the reservoirs - and can be run "wide open".

You get much shorter braking distances at low speed, but not very much improvment at higher speeds.

One thing I don't know - how will an ECP train know to go into emergency at a break-in-two?  Loss of comm?

[oltmannd]

Euclid

It requires derailment sensors

What is this and how would it function?  How could it tell a derailment from truck hunting, rough crossing, slack action, etc?