Passenger Trains > Is PTC already obsolete?

The Wall Street Journal asks whether Positive Train Control, which is now older than the iPhone, is ready for the scrap pile.  This opinion piece opines the network of transponders, interoperability, bandwidth etc is 1990s technology. Old. Obsolete.

Instead, let Google adapt the self driving protocols for trains, GPS, and embed the radar technology used in smart cars already. The process, according to the author, is relatively simple. The tracks are set in place, the switches are there, the current iPhone or Android is already programmed to do the step by step directions.


How Congress Railroaded the Railroads

From what I've read online about PTC, the problem is far more than the locomotive 'knowing' where itself is and where the end of the train is.  It's all about knowing where the OTHER trains are, what wayside/cab signals indicate, slow orders, tracks out of service, etc.  It's a continuous 2-way communication, not 1-way like GPS.  And while military GPS can locate something within (I'm guessing) 2-3 feet, commercially available GPS is limited to about 10 feet, give or take.  At 10 feet, how does it differentiate between a train on track 1 and track 2?

While PTC technology - at least the concepts of locating oneself and methods for knowing where others are - may be 'older', it isn't some new 'bleeding edge' technology that still has to prove itself 100% reliable.  Reliability is paramount!   It can't be done overnight.  Unlike the NASA 'directive' of the '60s of 'make 5-10 copies(rockets)' and that's the starting point for additional improvements to ultimately go to the moon.  PTC not a 'start here and grow/evolve' plan being implemented over a 10 year period.  It's a one-shot-does-it-all deal.  One cannot go back and redesign and upgrade the installation once it's done.  The 'size' alone and the multitude of operating conditions (mountains, tunnels, under electric overhead, in cities, thunder storms, tornadoes, blizzards) makes it far more difficult to design 'fail safe' than simply travelling through space (once free of the atomosphere). 

No, PTC isn't some small group of technologies that can be assembled like a LEGO building.  It's comparable to the initial implementation of Obama Care online application system.  Thousands of people trying to make hundreds of different insurance company and state software systems 'work together' and get it all done in a compartively short amount of time.  There's a million tasks to be planned, scheduled, and completed that must all work together.  And no one 'general' understands the 'whole story' and can 'direct' the operations.    



Edited 1 time(s). Last edit at 05/20/15 18:46 by BRAtkinson.

I would add PTC must also function on top of existing communication and signal systems, many involving complex routings amongst a maze of tracks in a tight area within interlockings. PTC is also is being designed to prevent incursions into areas where roadway workers are deployed and to fucntion as an added layer of enforced protection where highway grade crossings are malfunctioning. 

PTC are a list of functional requirements, not a specific technology.  Unless trains start following different laws of physics they won't become obsolete.

BRAtkinson Wrote:
-------------------------------------------------------
> At 10 feet, how does it differentiate between a train on track 1 and track 2?

GPS is only part of the solution. Transponders on the track bed will help confirm what track the train is on. Communication with control points will tell the trains whether they were routed to another track at the last crossover. GPS will not be working in a vacuum. It is just one of the many inputs to the system.
 

The complexity of GPS systems boggles my mind. 

I like solutions based on old school mentality to measure resistance to sense occupancy.   I once thought about the application and imagined a system that setup mico-blocks using audio frequency overlay circuits like those used for grade crossing systems.  Modern grade crossing systems can pin point the position of a train and calculate its speed within its circuit.

I imagine having these circuits overlapping, and tracking the train as is moves across the railroad. The track and occupancy information is processed, with train control data modulated and transmitted through the rail to be deciphered by equipment installed in the locomotive.  No radio spectrum is needed, and no GPS issue.  The microblock system would chop an older multi mile block system into blocks of a few thousand feet in length. 

Todd Clark
Canyon Country, CA
Trainorders.com

justalurker66 Wrote:
-------------------------------------------------------
> BRAtkinson Wrote:
> --------------------------------------------------
> -----
> > At 10 feet, how does it differentiate between a
> train on track 1 and track 2?
>
> GPS is only part of the solution. Transponders on
> the track bed will help confirm what track the
> train is on. Communication with control points
> will tell the trains whether they were routed to
> another track at the last crossover. GPS will not
> be working in a vacuum. It is just one of the many
> inputs to the system.
>  

Similar to the "stay in lane" technology that already exists on new cars, and the self driving features of many new nav systems. Even things like variable speed control sensors, which adjust a truck's maximum speed for the highway on which they operate.

I can understand that the freight railroads have resisted something which will give them little benefit.

webmaster Wrote:
-------------------------------------------------------
> The complexity of GPS systems boggles my mind. 
>
> I like solutions based on old school mentality to
> measure resistance to sense occupancy.   I once
> thought about the application and imagined a
> system that setup mico-blocks using audio
> frequency overlay circuits like those used for
> grade crossing systems.  Modern grade crossing
> systems can pin point the position of a train and
> calculate its speed within its circuit.
>
> I imagine having these circuits overlapping, and
> tracking the train as is moves across the
> railroad. The track and occupancy information is
> processed, with train control data modulated and
> transmitted through the rail to be deciphered by
> equipment installed in the locomotive.  No radio
> spectrum is needed, and no GPS issue.  The
> microblock system would chop an older multi mile
> block system into blocks of a few thousand feet in
> length. 

PTC was politically palitable because it was sold as not requiring railroads to replace or substantially modify their existing signaling infrastructure.  That meant no coded track circuits and no transponders in the five foot.  From what I have been told the actual PTC hardware interfaces with the legacy signal system by clipping onto the output wires to the physical bulbs, ie a "cheap" bolt on addition.  The problem is that making a reliable wireless network has been just as costly as making the physical modifications the railroads were trying to avoid.

Your solution is basically LZB, but the achillies heel is that it is dependant on the communications network.  US freight railroads need to be able to function even when the link between train and the back office is unreliable.  That is why the code line is non-vital, interlockings have dual control switches and intermediate signals function almost completely autonimously.

Lackawanna484 Wrote:
>
> Similar to the "stay in lane" technology that
> already exists on new cars, and the self driving
> features of many new nav systems. Even things like
> variable speed control sensors, which adjust a
> truck's maximum speed for the highway on which
> they operate.
>
> I can understand that the freight railroads have
> resisted something which will give them little
> benefit.

I rather suspect that the 'stay in lane' technology as well as crash avoidance technology being placed in cars now and in the future are a combination of optical and radar-like technology.  All fine and good, if you are controlling a 2 ton vehicle and all safety-directed actions can be completed in line-of-sight distances.  Not so with 3,000 or 6,000 ton trains that can rarely stop in line-of-sight distances when travelling any faster than restricting speeds.

As far as freight railroad resistance goes, how'd you like the US Government to REQUIRE YOU to install a $20,000 security and alarm system in your house by the end of the year and not offer you ONE PENNY to pay for it?  Oh, and you live in a practically crime-free ZIP code.  How much benefit are you going to get out of YOUR $20,000?  Give it time...maybe the Department of Homeland Security may do exactly that!

PTC--too many variable to work properly.  I doubt it will ever work as well as human beings making decisions.  They should drop it, period.

BRAtkinson Wrote:
-------------------------------------------------------
(snip)
>
> As far as freight railroad resistance goes, how'd
> you like the US Government to REQUIRE YOU to
> install a $20,000 security and alarm system in
> your house by the end of the year and not offer
> you ONE PENNY to pay for it?  Oh, and you live in
> a practically crime-free ZIP code.  How much
> benefit are you going to get out of YOUR
> $20,000?  Give it time...maybe the Department of
> Homeland Security may do exactly that!


In NJ, the state required any bar which wanted to allow smoking to install a Smoke Eater system. That drew cigarette smoke up, and vented it out of the bar. Many bars installed them.

Five years later, the state banned all smoking indoors, including bars. Except in Atlantic City casinos and areas of race tracks.  Everybody else was stuck with a $15,000 curiousity in the ceiling...They did the same thing with motor vehicle inspection devices in service stations.  Gotcha!

Lackawanna484 Wrote:
-------------------------------------------------------
> In NJ, the state required any bar which wanted to allow smoking to
> install a Smoke Eater system. That drew cigarette smoke up, and
> vented it out of the bar. Many bars installed them.

Allowing smoking was their option. They could have gotten ahead of the curve, banned smoking and saved the money. Or they could have gotten ahead of the "required smoke eater" curve and installed them before they were required (and provide a better environment for their patrons).