Positive Train Control Myths and Facts

I have found that I have less "pre-actions" and I can run harder and faster when I know whats going to be happening, meaning when there are waysides and automatics as well as cab signals, or somekind of visual reference. I know where, when and how my cab signals are going to change, because I can see where the automatic is, and I can adjust my running accordingly - I can "pound" the signal, or i can "pussyfoot" it, or somewhere inbetween. But either way, I know what its going to do, so im not surprised. And like i said above, I don't like surprises.

Maybe PTC could be developed with a visual reference of some sorts. Say for example - 1000 feet from a stop signal, a colored sign or pole, and a rule that states you must be no faster than 15mph if coming up to a stop signal. At 500 feet from the signal, another different colored reference, which requires the enginere to be at 10mph or less. So and so on. Also say there is a different colored pole or sign that is say 75 feet from the signal, and trains MUST stop no closer than that point for the stop signal. Passing that point, will automatically trigger a penalty or emergency brake application. If permission past the signal is required, punching in the code the dispatcher gives will make the system happy and hten the train could be moved again. If the train is going more than 15mph and is 1000 feet from the stop signal, the eng has say 5 seconds to get brakes on and suppression, just like what happens now with cab signals. If he doesn't, the train goes into penalty and stops.

Something like this might be easier than a "floating" system that has all these variables and braking curves, and trying to account for all those variables which seem impossible to account for in a computer software. It could be done in 10mph tiers, down to stop signals or speed restrictions. A radio wave could be transmitted out when a signal is at a stop signal or a speed restriction is in placed, and a receiver on board the train also shows which "colored block" (say orange for 500 ft and 10mph) and thereofre speed allowed, just like a cab signal system, and would correspond with the wayside colored poles or signs

Just a few thoughts.....

Don't look at an engineers "pre-actions" as a hinderance or a negative. Use them in a positive manner.

OK, I'm going to try to keep this nice, because I think my ugly self-righteous side reared its ugly head. I will say it again - you need to have a lot of training and knowledge to be an engineer. That includes knowledge of the physical characteristics. However, in cab-no-wayside territory where there are code drops, you can't possibly know where the code drop actually is. So, in order to run your train the way that you yourself said that you run, you would need to basically decide by feel (which comes from the experience of running over the territory many times) where you need to be on the brake in order to not run into the drop.

Knowing where your code will drop, that depends. In no wayside territory, if you are following another train, it's guesswork. As an Engineer, you are familiar with the operation of the signal system, and use it to your advantage. Impedance boxes, opposing signals etc are some identifiers where you will probably hit a code.
However, there are some instances where an Engineer will know where he will get coded down. An example is approaching an interlocking. Since you have no waysides, obviously this will only work if you have a straight line of vision to the home signal some distance, and it's aspect is visible. If you are familiar with the railroad, you will know where each code drop will occur. For example, if you have a stop signal at XYZ interlocking, you will know you will receive a code drop at the next impedance box, another just past that signal hut by the overgrade bridge, etc. If you have an approach, you wont get coded down until the crossing after that impedance box. If the homes at approach medium, the Engineer has another series of points he knows from experience. The beauty of it is, you will be at that speed exactly where the code hits you.

Well, you've probably only worked in a few places at most. I've worked in a variety, both as an employee and as an operations consultant where engineers are supervised by non-engineers. Of course, for training purposes, recertification purposes, and for *some* efficiency testing purposes, you need an engineer. And in those cases, supervisors who are also engineers are used.

Well, I am glad that is not the case with my RR.

I have not worked at PATH as a supervisor. I have, however, done extensive work for them as an operations consultant. Just so we're clear, none of the things that I've said here have been intended to be offensive to you or any other engineer. But, to be just as clear, the things that I've said have also been well established, including the pre-action phenomenon. They are as real to those of us who work in managing and designing rail transportation as your brake handle is to you.

In designing PTC systems for passenger rail applications, pre-action is one of many myriad things that have to be considered - there are other human factors considerations, debates about the lowest possible rail/wheel adhesion, debate about enforcement of bumper blocks, debate about were to apply for MTEAs (main track exclusion authorities), and debates about interoperability.

The one thing that is clear through all of it is that PTC, like most safety systems, is not primarily designed to enhance capacity. The best we can hope for is minimal effect. At worst, PTC will severely affect capacity and throughput. In high density operations, that will have a profound effect on the railroad's ability to deliver its current level of service.

No offense taken. I took what you meant by "preaction" another way. This is obviously going to occur when RR's take away the visual aid of automatic signals. Personally, I do not agree with anything this mandate will bring about, but bumper blocks? So am I going to have to punch in a code when I have to pull within inches of the block to clear the foul on the other end? The same can be said regarding stop signals. Sounds like complete nonsense to me.
Everyone wants to live in a bubble regarding high profile transportation, but wont give any thought to the highway safety statistics.

justalurker66 wrote:So, is the goal (and base requirement) of PTC that the train cannot ever break a rule? The brakes will automatically apply if an engineer/operator fails to apply them (perhaps even throwing the train into emergency)? All speeds will be absolutely enforced (line speeds, permanent restrictions and temporary restrictions)?

Yes, that's part of the requirement. Right now, the use of emergency brakes is not being discussed (at least in the projects I'm familiar with), only service braking. Allowing consideration of emergency braking would shorten up the "enforcement" point.

justalurker66 wrote:What does the system need to know? CP and intermediate signals are obvious. I'm assuming the absolute position of every mainline switch needs to be tied in. The status of every crossing gate? (No, I'm not suggesting vehicle on track detectors, just an alert if there is a malfunction detected. Then again, are vehicle on track detectors required?)

Mainline switches are definitely "in", even in dark territory. Crossing warning protection is in, but at what level I'm unsure right now. Still wading through the final rule.

justalurker66 wrote:I understand that cab signals are not a requirement of the system ... I assume that in non-cab signaled systems engineers will know where the restrictions are primarily based on signals. Will there also be the ability to add in restrictions at any point on the line for work zones/etc.? Is this a requirement of PTC or just a feature many will implement?

Speed restrictions will still be listed in the special instructions, and temporary speed restrictions will still be posted in a TSRB, Daily Operating Bulletin, Track Warrant, or some other type of notice/order. It's still the crew's responsibility to know where restrictions are and comply; the in-cab displays will be a guideline. Temporary work zones will also be protected. That's an interesting one - PDAs for EICs to give permission through work limits. We'll see how that one develops over the next few years.

justalurker66 wrote:"PTC required by 2015" needs definition. If "SafeAuto" designed a PTC system, what would they need to do?

Well, railroads have to comply with the new FRA rule that just came out this past week. At a minimum (there may be some added things - still reading), the system has to be able to stop a non-compliant train before it strikes another vehicle, before it gets into a position where it could be struck by another vehicle (past the fouling point at a turnout, for example), before it runs over a switch lined improperly, and before it enters a work zone without permission. It will also enforce civil and temporary speed restrictions.

If you have the time and patience, the final rule is posted on the FRA website. The truly relevant part is the last 100 pages or so, which detail the rule.

Jtgshu wrote:I have found that I have less "pre-actions" and I can run harder and faster when I know whats going to be happening, meaning when there are waysides and automatics as well as cab signals, or somekind of visual reference. I know where, when and how my cab signals are going to change, because I can see where the automatic is, and I can adjust my running accordingly - I can "pound" the signal, or i can "pussyfoot" it, or somewhere inbetween. But either way, I know what its going to do, so im not surprised. And like i said above, I don't like surprises.

That's one of the advantages of cab with wayside, as I've heard it from engineers. On SEPTA, the engineers *hated* the rule 562 installation on the main line because they lost their previous visual cues.

Jtgshu wrote:Maybe PTC could be developed with a visual reference of some sorts. Say for example - 1000 feet from a stop signal, a colored sign or pole, and a rule that states you must be no faster than 15mph if coming up to a stop signal. At 500 feet from the signal, another different colored reference, which requires the enginere to be at 10mph or less. So and so on. Also say there is a different colored pole or sign that is say 75 feet from the signal, and trains MUST stop no closer than that point for the stop signal. Passing that point, will automatically trigger a penalty or emergency brake application. If permission past the signal is required, punching in the code the dispatcher gives will make the system happy and hten the train could be moved again. If the train is going more than 15mph and is 1000 feet from the stop signal, the eng has say 5 seconds to get brakes on and suppression, just like what happens now with cab signals. If he doesn't, the train goes into penalty and stops.

PTC will have visual indications in the cab when you've entered a "braking profile". The ones I've seen so far give you current speed, target speed, distance to target, time to target, and time to enforcement, so there is some "help". There is debate as to how much help to give - there is a strong feeling from Transportation folks that you don't want to get engineers dependent on the in-cab display,because when the inevitable failure occurs, that crutch is taken away. Also, a system that tells you what to do when, where, and how will make it nearly impossible to efficiency test properly.

Jtgshu wrote:Don't look at an engineers "pre-actions" as a hinderance or a negative. Use them in a positive manner.

Oh, I don't! It is what it is. Engineers are simply doing what comes naturally (and it's not bad behavior, it's "enhanced" good behavior), and it's the job of the experts to try to harness that and make it work with PTC without killing capacity.

RogerOverOutRR wrote:Knowing where your code will drop, that depends. In no wayside territory, if you are following another train, it's guesswork. As an Engineer, you are familiar with the operation of the signal system, and use it to your advantage. Impedance boxes, opposing signals etc are some identifiers where you will probably hit a code.
However, there are some instances where an Engineer will know where he will get coded down. An example is approaching an interlocking. Since you have no waysides, obviously this will only work if you have a straight line of vision to the home signal some distance, and it's aspect is visible. If you are familiar with the railroad, you will know where each code drop will occur. For example, if you have a stop signal at XYZ interlocking, you will know you will receive a code drop at the next impedance box, another just past that signal hut by the overgrade bridge, etc. If you have an approach, you wont get coded down until the crossing after that impedance box. If the homes at approach medium, the Engineer has another series of points he knows from experience. The beauty of it is, you will be at that speed exactly where the code hits you.

Most assuredly. And when you get on the brake before the code drops, that's "pre-action". You've now described the phenomenon almost perfectly about three times now. So, I'm failing to see where it is that we disagree.

RogerOverOutRR wrote:Well, I am glad that is not the case with my RR.

I'm very happy for you that you're happy in your employment. That's critical.

RogerOverOutRR wrote:No offense taken. I took what you meant by "preaction" another way. This is obviously going to occur when RR's take away the visual aid of automatic signals. Personally, I do not agree with anything this mandate will bring about, but bumper blocks? So am I going to have to punch in a code when I have to pull within inches of the block to clear the foul on the other end? The same can be said regarding stop signals. Sounds like complete nonsense to me.

No, not exactly. The way I've seen it described is making the block a "zero" target - i.e. the functional equivalent of a Stop signal to the PTC system. So what PTC will do is to supervise your braking profile coming up to a bumper block. I don't know where I am yet on this, because I'm afraid railroads will lose some usable track space if trains don't "snug up" to the bumpers. Also - in terminal areas with MAS under 20 mph, MTEAs will cover the area, meaning you don't have to have a PTC system in place.

RogerOverOutRR wrote:Everyone wants to live in a bubble regarding high profile transportation, but wont give any thought to the highway safety statistics.

Agreed. Believe me, while this is a windfall for some engineering firms and equipment suppliers, I think as an industry this is a huge financial and operational hit. It remains to be seen whether or not enough operational advantages accrue to offset the substantial costs. I personally think the FRA's assessment that these systems will pay for themselves within 20 to 25 years is wildly optimistic.

Moderator's Note:

After some discussion with Otto, we're shipping this topic off one more time, to General Discussion: Locomotives, Rolling Stock, and Equipment. Apologies to Col. Perkowski for the imposition, but PTC is definitely a wider topic than just commuter rail, and has at least as much to do with equipment as it does operations.

Thanks for keeping it civil during your stay.

-Andy

While I never had the displeasure of running on NJT's Pascack Valley LIne when the SES was being "tested" (for 10 years...I repeat again, a miserable failure, and since REMOVED) from what I have been told, the "target speed" was one of the things that was hated the most. It required a very slow, constant braking without giving much leeway. A train isn't very good for slow, constant braking. Of course, while different kinds of trains handle differently and require different styles of running, like a 6 car commuter train vs. a 100 car coal train, there are still some constants. With one ofthem being little brake pipe reductions can and do lead to sticking brakes.

So then you are required to take out more air, (my RR likes at least 12psi taken out just to prevent sticking brakes), you are going to slow down too fast. But you have to have some kind of brakes on for the train to comply with the slowly decreasing "target speed". there are two options - Taking out say 12psi and slowing down to 5mph below the "target speed" then releasing, then going for more brake again when your speed and the target speed meet up again, and doing that until you 1) either stop, or 2) meet the speed restriction. A real good way to "run out of air" pretty quickly. Another option is to power brake and keep some power on with the brakes set and control your speed that way. Very good at controlling your speed exactly, but in these days of expensive diesel fuel, the RR's aren't gong to be happy, if it is even allowed by Special Instructions or the Equipment itself.

Thats why I think a step down method with wayside visual cues is something that shold be considered, or a "target speed" that doesn't decrease by 1mph - or if it does, not enforce it until say the speed is off by lets say 10 or 15 percent, and beep or something when it gets to that point, and gives a few seconds to get some brakes on. That would allow the engineer to control the train speed in a safe way, still let him be in control, and hopefully not have the "crutch factor" for when there is a failure. While of course, the whole time bringing the train down to a stop.

Andy,

Thanks! Got it.

Jtgshu wrote:While I never had the displeasure of running on NJT's Pascack Valley LIne when the SES was being "tested" (for 10 years...I repeat again, a miserable failure, and since REMOVED) from what I have been told, the "target speed" was one of the things that was hated the most. It required a very slow, constant braking without giving much leeway. A train isn't very good for slow, constant braking. Of course, while different kinds of trains handle differently and require different styles of running, like a 6 car commuter train vs. a 100 car coal train, there are still some constants. With one ofthem being little brake pipe reductions can and do lead to sticking brakes.

So then you are required to take out more air, (my RR likes at least 12psi taken out just to prevent sticking brakes), you are going to slow down too fast. But you have to have some kind of brakes on for the train to comply with the slowly decreasing "target speed". there are two options - Taking out say 12psi and slowing down to 5mph below the "target speed" then releasing, then going for more brake again when your speed and the target speed meet up again, and doing that until you 1) either stop, or 2) meet the speed restriction. A real good way to "run out of air" pretty quickly. Another option is to power brake and keep some power on with the brakes set and control your speed that way. Very good at controlling your speed exactly, but in these days of expensive diesel fuel, the RR's aren't gong to be happy, if it is even allowed by Special Instructions or the Equipment itself.

Thats why I think a step down method with wayside visual cues is something that shold be considered, or a "target speed" that doesn't decrease by 1mph - or if it does, not enforce it until say the speed is off by lets say 10 or 15 percent, and beep or something when it gets to that point, and gives a few seconds to get some brakes on. That would allow the engineer to control the train speed in a safe way, still let him be in control, and hopefully not have the "crutch factor" for when there is a failure. While of course, the whole time bringing the train down to a stop.

The ASES system was definitely a failure, and from what I've heard in the "after actions" from people who were involved, most of the problems were avoidable, but NJT's insistence on 'going it alone' was a big problem. Amtrak, at that point, had ACSES up and working for at least a portion of their territory, and while the bugs were still being worked out, it was undoubtedly working better than NJT's ASES. ASES could have taken some lessons from ACSES, but didn't.

The engineers do not have a list of the timer settings. They get used to the time-outs, and they run accordingly. And sometimes get tripped by the stop arm.

Some sets of timed signals in the New York City and PATH subways are labeled with the speed limit. I don't know if the signal designers obey the signs, or if they are accurate. http://www.ntsb.gov/publictn/1996/RAR9604.pdf That link about a Washington Metro accident at Shady Grove, Md. on 1/6/96 tells about how the Automatic Train Operation can be adjusted during bad rail weather in different sections of the line, and what happens when a train losses a "bad track adhesion speed and brake command" and defaults to Full Speed Ahead!
I have ridden many Market-Frankford Subway-Elevated trains in Philadelphia, Pa., and have seen and felt motormen approaching timed stop signals and automatic stop trips with the brakes released. Some motormen even apply power approaching the last in a series of timed signals before it changes from Stop! In Philadelphia's system (and others), there are usually 2 stop signals and 1 entire block behind every preceding train.
From about 1950 to 1980, some electric freight locomotives used by PRR - PC -CR had speed control that would engage if a signal more restrictive than clear was passed faster than the permitted speed. The accident report for Chase, Md. tells of how the ConRail Diesel locomotives in that crash didn't have the older system, and how the NTSB though it could have lessened the impact if ConRail had used the existing system, which Amtrak trains on the same tracks still used, on its replacement Diesel locomotives. The Interstate Commerce Commission report on the Newark Bay drawbridge wreck of 1958 claims that Even Central Railroad Company of New Jersey Diesels had Speed Control as well as Cab Signals, for use on the New York & Long Branch RR, on which Pennsylvania Railroad trains also ran. Did the PRR K-4 steam locos have speed control between about 1950 and 1957?
Accident reports for the hill near Horse Shoe Curve, Pa. mention that PRR used regular signals with cab signals to enforce a 30 m.p.h. speed limit downhill. They didn't do that job when the last car of an uphill sleeping car train became uncoupled with defective air brakes and a rusted stiff hand brake staff about 1943.
The difference in efficiency between manual and automatic operation on PATCo seems to be that the operator will keep the train just under the cab signal speed limit, while the Automatic Train Operation will use the Automatic Train Control to slow the train after it exceeds the cab signal speed limit. For example, going downhill, the operator will "ride" the brakes to keep a steady speed about 1 m.p.h. under the limit. The ATO will apply the brakes after the train exceeds the limit, but it will still go a bit faster before the brakes are effective. Then the ATO will fully release the brakes when the train slows below the limit. It behave in a somewhat similar fashion when accelerating. The ATO only has to obey its programmed instructions, it doesn't have to be more conservative to avoid any chance of being disciplined. Railroad engineers seem to be even more fearful of heavy discipline if they go an inch past the insulated joint at a Stop Signal.

Don't know why the moderators keep trying to bury this thread in forums nobody goes to, but I wanted to try to steer things back on track with a study on the effect of youth bicycle helmet laws in the 21 states that have them vs those that don't. Long story short the safety laws caused a dramatic decrease in youth riding bicycles. This isn't a direct PTC analogy here, but just an example of unintended consequences that come with new safety mandates. It will also be interesting to see how risk compensation will factor in in that PTC gives engineers a false sense of security and will run right up to the enforcement point, even if for whatever reason, that point is insufficient to stop the train. Wherein before they would have been more careful to avoid a 241 violation, engineers might end up emboldened by the technology and end up causing more accidents as a result.

jb9152 wrote:If you have the time and patience, the final rule is posted on the FRA website. The truly relevant part is the last 100 pages or so, which detail the rule.

Thank you for your answers. I am aware of the final rule but started reading it at the wrong end (the beginning, who would have thunk it!). Hopefully the actual system required won't be too high of a burden. Safe operation of trains is irrelevant if one can no longer afford to run trains.

Jersey_Mike wrote:It will also be interesting to see how risk compensation will factor in in that PTC gives engineers a false sense of security and will run right up to the enforcement point, even if for whatever reason, that point is insufficient to stop the train. Wherein before they would have been more careful to avoid a 241 violation, engineers might end up emboldened by the technology and end up causing more accidents as a result.

From what I understand reading this thread, the opposite is more likely to be a problem. It is more likely that the system won't allow the trains to run to the best of the engineer's abilities. The extra "safety margin" added by PTC will place the enforcement point far enough away ("safer") than the current level of operation that engineers who run the old way will be hitting enforcement points and will have to be more timid it the operation of their trains.

BTW: The hot potato treatment of this thread reminds me of the core issue of "what is PTC"? Is it something that Amtrak needs to deal with? Yes. PTC being required on (nearly) all passenger rail. But it is also required on hazmat carrying rail and not all passenger lines are Amtrak. It is more than "equipment" but as long as we have a place to discuss the finer details hopefully everyone will find the thread. (Thanks for finding a place for it, mods.)

How will the railroad's hearing examiner treat the case of an engineer who applies the brakes at the proper pressure at or before the proper place, and still doesn't slow down or stop soon enough?

From what I understand reading this thread, the opposite is more likely to be a problem. It is more likely that the system won't allow the trains to run to the best of the engineer's abilities. The extra "safety margin" added by PTC will place the enforcement point far enough away ("safer") than the current level of operation that engineers who run the old way will be hitting enforcement points and will have to be more timid it the operation of their trains.

Freight railroads simply will not install any PTC system that kills their operations. Will there be some capacity reduction? Probably, but the industry won't stand for their main lines getting tied in knots like after the UP/SP merger. As PTC gets installed on freight lines we'll see a few accidents caused by engineers running to the PTC curve instead of their own knowledge. They'll probably be a flap on if the PTC should be made more conservative, but if the choice is between a functional rail network or PTC, PTC will quickly get downgraded to preventing long hanging fruit type crashes. All the railroads have to do is simply starve some power plants of coal and blame PTC.

Passenger trains are less of a problem because their braking curves are easy to calculate. The Amtrak ACSES box doesn't even have any sort of speed curve information. You just see a restriction pop up and u have to begin braking w/in some number of seconds to avoid a penalty application. I don't notice any difference performance wise between the ACSES zone on Amtrak between Wilmington and Perryville and the non ACSES areas.