Ten years ago I was provided with a amazing PDF document from the New York City Transit Authority that contained diagrams of every rail line, every rail yard and every interlocking on the Subway, Staten Island Railroad and even some railroad shotlines. The interlocking diagrams themselves had details from switch and signal numbers, details regarding the control panels and interlocking machines and even the date of details of the contracts that the signaling equipment was installed under.
Since then the NYCTA has been pushing hard to eliminate the last bastions of single interlocking towers with CTBC projects ongoing on the Flushing and Queens Boulevard Lines. As my document aged I attempted to get an updated copy, but was told that the NYCTA was annoyed that the first consolidated document had ever been created and that the information had been effectivly split up between all the various line managers. I don't know if that was true or not, the NYCTA culture thrives on control of information (often using "terrorism" as a justification), but while searching for some photos of the old E 180th St tower to fix a dead link I came upon a 2019 version of the same document.
Because nobody had metaphorically slipped me the new version "under the table" with a promise of confidentiality, I am just going to link both documents here so you all can see for yourself the extent of signaling heritage that has been lost on the NYCTA over the last ten years. Enjoy.
NYCTA Schematic Track and Interlocking Diagrams 2009
NYCTA Schematic Track and Interlocking Diagrams 2019
Sunday, March 29, 2020
Friday, March 20, 2020
DART - One System, Two Methods
Dallas Area Rapid Transit is one of the new generation of North American light rail systems, as large in scope as many heavy or commuter rail networks, but built to a much lower price point. DART was built in two waves, the initial Blue and Red lines built between 1996 and 2002 and then the Orange and Green lines built roughly between 2010 and 2012. Despite being part of a unified transportation system, DART uses two completely different signaling methods.
Signaling on the Red and Blue lines consists of pretty standard ABS CTC with wayside signaling. Signals consists of single direction stub masts placed back to back. An ATS capability is provided via some sort of IIATS or Westcab transponder to enforce signals displaying Stop, however track speeds are not enforced.
Interlocking signals use a NYC Subway style system with the upper head indicating block state and the lower head route. Thus G/Y is diverging clear and Y/G is approach straight. Also note the use of traditional track circuits with impedance bonds.
Of course for the second phase Orange and Green lines, DART decided to switch gears and adopt a cab signal system without fixed wayside signals. Cab signals are of the jointless audio frequency type and signals are only provided at interlockings with the ability to display Stop (R), Restricting (R/Y) and lighted directional arrows for everything else.
These directional signals were also used in the Phase 1 street running portion downtown, so it was natural to reuse the system in cab signal territory to reduce the need for re-training.
Unfortunately my photo of the cab signal display did not turn out, but there is not a lot of complexity. The operator sees the vehicle speed in red LED and a target speed in orange LED. The speed changes as soon as a new cab signal code is received with an ATO style brake application being instantly applied, as opposed to a penalty application after a grace period. This system enforced both track and block speeds with speed control in 5 or 10 mph increments between 10 and 60mph.
It was interesting to see these two generations of systems running side by side with Orange line trains in particular starting under one system and finishing under the other. It was also interesting to see how much train operation was being impacted by the speed enforcement with operators having to brake about 5mph past the speed target and then re-accelerate. This created a very jerky form of operating where as the lines without speed enforcement saw much smoother operation.
Although many light rail systems are adopting or converting to heavy handed forms of train control, there are still many systems built since 1990 that have retained lower cost, human-centric signaling systems that will likely be the last such examples as heavy and commuter rail lines become increasingly "supervised".
EDIT: DART actually has a THIRD method of operation that I discuss in a followup post.
Of course for the second phase Orange and Green lines, DART decided to switch gears and adopt a cab signal system without fixed wayside signals. Cab signals are of the jointless audio frequency type and signals are only provided at interlockings with the ability to display Stop (R), Restricting (R/Y) and lighted directional arrows for everything else.
These directional signals were also used in the Phase 1 street running portion downtown, so it was natural to reuse the system in cab signal territory to reduce the need for re-training.
Unfortunately my photo of the cab signal display did not turn out, but there is not a lot of complexity. The operator sees the vehicle speed in red LED and a target speed in orange LED. The speed changes as soon as a new cab signal code is received with an ATO style brake application being instantly applied, as opposed to a penalty application after a grace period. This system enforced both track and block speeds with speed control in 5 or 10 mph increments between 10 and 60mph.
It was interesting to see these two generations of systems running side by side with Orange line trains in particular starting under one system and finishing under the other. It was also interesting to see how much train operation was being impacted by the speed enforcement with operators having to brake about 5mph past the speed target and then re-accelerate. This created a very jerky form of operating where as the lines without speed enforcement saw much smoother operation.
Although many light rail systems are adopting or converting to heavy handed forms of train control, there are still many systems built since 1990 that have retained lower cost, human-centric signaling systems that will likely be the last such examples as heavy and commuter rail lines become increasingly "supervised".
EDIT: DART actually has a THIRD method of operation that I discuss in a followup post.
Thursday, March 12, 2020
Thinking About the Price of Safety
I have been a pretty vocal critical about the value of PTC when it comes to general rail safety. It's the sort of thing that sounds great in practice, but looking at the number of lives lost in PTC preventable rail accidents the costs are seriously hard to justify. Anyway, in this day of weighing the cost of hundreds of thousands of lives lost due to a pandemic against the severe cost of an economic depression, the folks at 538 took a crack at trying to figure out how much the government should spend to save a life. It provides some additional metrics beyond "lives" or even Quality Adjusted Life Years that get to people's own subjective valuation of risk.
Sunday, March 1, 2020
Right Hand Rule - How De-Regulation Ruined US Signaling
Why the US railroads seem to go from having a pathological need to mount their signals to the right hand side of the tracks they governed, like this...
Or this...
Or even this...
To, seemingly overnight, just doing this...
Well in 1939 most locomotives provided the engineer with visibility like this...
So in 1939 the ICC decided to legislate the obvious and required that all railroad signals be situated to the right hand side of the track they governed. Of course the usual system of waivers and grandfathering applied so that left hand running lines such as the CNW didn't have to rip out its fireman-side signals, but all of that created barriers to cost cutting, even after the 1960's rolled around and all of the main line steam locomotives had been retired.
By the 1980's, chopped short hoods on the F end had been the norm for the better part of two decades with only a few out-layers like the Southern, still holding on to the practice. In 1985 the ICC successor, FRA decided that the Right Hand Rule was obsolete and railroads could put signals on whichever side of the track they wanted.
The effect was immediate with back-to-back masts replacing split masts automatics and end-of-siding signal bump-outs becoming some sort of vestigial organ. Bracket masts, the go-to low cost option for bi-directional automatics on two track main lines were hit particularly hard along with the two track automatic signal bridge. Multi-track cantilever signals have hung on, but only in areas of restricted visibility like curves.
Now in Europe many of the right (or left) hand replacement requirements are either still in force or at least still practiced by the state owned railways and as such there is much more diversity in the signal mounting systems, typically in the form of signal bridges. It just goes to show how important the regulatory environment can be for something as esoteric as signal diversity.
Or this...
Or even this...
To, seemingly overnight, just doing this...
Well in 1939 most locomotives provided the engineer with visibility like this...
So in 1939 the ICC decided to legislate the obvious and required that all railroad signals be situated to the right hand side of the track they governed. Of course the usual system of waivers and grandfathering applied so that left hand running lines such as the CNW didn't have to rip out its fireman-side signals, but all of that created barriers to cost cutting, even after the 1960's rolled around and all of the main line steam locomotives had been retired.
By the 1980's, chopped short hoods on the F end had been the norm for the better part of two decades with only a few out-layers like the Southern, still holding on to the practice. In 1985 the ICC successor, FRA decided that the Right Hand Rule was obsolete and railroads could put signals on whichever side of the track they wanted.
The effect was immediate with back-to-back masts replacing split masts automatics and end-of-siding signal bump-outs becoming some sort of vestigial organ. Bracket masts, the go-to low cost option for bi-directional automatics on two track main lines were hit particularly hard along with the two track automatic signal bridge. Multi-track cantilever signals have hung on, but only in areas of restricted visibility like curves.
Now in Europe many of the right (or left) hand replacement requirements are either still in force or at least still practiced by the state owned railways and as such there is much more diversity in the signal mounting systems, typically in the form of signal bridges. It just goes to show how important the regulatory environment can be for something as esoteric as signal diversity.
Sunday, February 23, 2020
NS Altavista District: A PL Haven...For Now
The NS Altavista district is a low grad coal route between Roanoke and Abiline, VA and I just saw some current photos showing that it has somehow avoided the mass re-signaling efforts that hit the N&W main line via Lynchburg. I'm not sure how similar it is to the Shenandoah District, which had its classic signaling refreshed (but not replaced) around the year 2000, but from what I can see most of the interlockings are still equipped with N&W style color PL's.
Still, nothing lasts forever and ALTAVISTA interlocking in Altavista, VA has the Darth Vaders up and ready to take over. Other interlockings seem to be OK, but if you live anywhere nearby, put this whole route on your must see list. The Altavista district is somewhat isolated and low traffic so it doesn't attract a lot of photography. It definitely needs some attention from the railfan community seeing the general lack of good signaling options.
Still, nothing lasts forever and ALTAVISTA interlocking in Altavista, VA has the Darth Vaders up and ready to take over. Other interlockings seem to be OK, but if you live anywhere nearby, put this whole route on your must see list. The Altavista district is somewhat isolated and low traffic so it doesn't attract a lot of photography. It definitely needs some attention from the railfan community seeing the general lack of good signaling options.
Sunday, February 16, 2020
Stop vs Proceed - Telling the Difference
Where railroad signals are involved there are two schools of thought. Should I stay or should I go? Usually this is pretty clear cut, but where Red signals are involved sometimes one should kind of do both. Back in the day in North America, there were long stretches of railroad that were literally out of all communication. A minor failure of some kind could leave a train stranded with no means of proceeding for perhaps hours. Therefore signals not protecting against a conflicting movement or other crash inducing hazard were defined as being permissive in that trains could pass them at slow speed and proceed until a more favorable signal indication was passed. Of course with three standards colors (R,Y,G) and signal heads only capable of displaying 2 or 3 aspects, railroads needed to get creative in how they indicated which signals could be passed, and which could not.
The first solution was to use the color and shape of a semaphore blade to define the type of signal and the rules involved. Absolute signals were painted red with a squared off end while distant signals were given a fishtail end and painted yellow. When permissive automatics came on the scene the fishtail was inverted into a point with the yellow background and black band remaining.
Of course semaphore signals are only disable during conditions of daylight. For the nighttime a marker light was applied below the main head. Also popular on position light signals, marker lights were the common way to define an automatic in the position/semaphore systems. Although there were the occasional exception of using a marker to denote an absolute signal (I'm looking at you N&W), more often the markers were used to make absolute signals that could be switched to act like a permissive signal.
Alright, you have your marker, but what if you also have absolute signals with more than one head. It's possible to mistake the marker for a lower absolute head in the darkness, even if the marker was smaller. One solution would be to use a yellow or lunar marker, but back in the day railroads still wanted a distinction between Stop and Proceed and a rolling Restricting. The solution was to offset the signal heads, using the relative position to convey the permissive meaning.
At this point someone realized that maybe they should just tell the crews that a signal was permissive. In the United States a number of railroads, typically in the South, adopted the (P) board for "Permissive".
Canadian railroads, on the other hand, have gone with a [R] Board for Restricting.
Of course the easiest/cheapest solution was to simply use the milepost number plate, useful in defining the particular the signal location, to also indicate a permissive status. Now milepost number plates have been around since the dawn of the automatic signal era with Hall Banjo, so why the need for markers and offset heads? Well reflector technology was pretty primitive up until the 1960's, either consisting of white paint with a bit of glass or glass cats eye jewels. Film plastic reflectors was what gave the railroads assurance that the number plate would be distinguishable at sufficient distance at night (at least when they still cared about such things).
Of course one outlier from the whole number plate system is Canada where offset heads or [R] boards are used instead. In fact, virtually all signals, absolute and permissive, come with a number plate denoting either the milepost or signal (lever) designation.
I'm sure for many of you this was pretty basic information, but these pages tend to come up in Google searches so its nice to have a comprehensive guide to the North American way of distinguishing permissive signals from absolute signals. The same distinctions also apply overseas, but there most countries have more unified sets of signal rules.
The first solution was to use the color and shape of a semaphore blade to define the type of signal and the rules involved. Absolute signals were painted red with a squared off end while distant signals were given a fishtail end and painted yellow. When permissive automatics came on the scene the fishtail was inverted into a point with the yellow background and black band remaining.
Of course semaphore signals are only disable during conditions of daylight. For the nighttime a marker light was applied below the main head. Also popular on position light signals, marker lights were the common way to define an automatic in the position/semaphore systems. Although there were the occasional exception of using a marker to denote an absolute signal (I'm looking at you N&W), more often the markers were used to make absolute signals that could be switched to act like a permissive signal.
Alright, you have your marker, but what if you also have absolute signals with more than one head. It's possible to mistake the marker for a lower absolute head in the darkness, even if the marker was smaller. One solution would be to use a yellow or lunar marker, but back in the day railroads still wanted a distinction between Stop and Proceed and a rolling Restricting. The solution was to offset the signal heads, using the relative position to convey the permissive meaning.
At this point someone realized that maybe they should just tell the crews that a signal was permissive. In the United States a number of railroads, typically in the South, adopted the (P) board for "Permissive".
Canadian railroads, on the other hand, have gone with a [R] Board for Restricting.
Of course the easiest/cheapest solution was to simply use the milepost number plate, useful in defining the particular the signal location, to also indicate a permissive status. Now milepost number plates have been around since the dawn of the automatic signal era with Hall Banjo, so why the need for markers and offset heads? Well reflector technology was pretty primitive up until the 1960's, either consisting of white paint with a bit of glass or glass cats eye jewels. Film plastic reflectors was what gave the railroads assurance that the number plate would be distinguishable at sufficient distance at night (at least when they still cared about such things).
Of course one outlier from the whole number plate system is Canada where offset heads or [R] boards are used instead. In fact, virtually all signals, absolute and permissive, come with a number plate denoting either the milepost or signal (lever) designation.
Sunday, February 9, 2020
Off Brand Modular Signals
Usually when a railroad has gone all in for modular color light signals they place an order for a GRS D type, a Safetran CLS-20 or whatever the heck US&S is selling, but sometimes your railroad might be looking at a weak 3rd quarter or is funded by the public sector. In those cases you might want to go with one of the little guys because they are probably willing to sell that $10,000 aluminum box for $7,000. In this case your major minor players are L&W, Harmon, something I think might be related to Harris Technologies and whatever is ripping off the CLS-20.
First up, L&W Industries is based in rural Missouri and made some inroads with Union Pacific or its predecessors in the 1980's before being generally rejected for the Safetran. It's current claim to fame is the LED searchlight still being bought by Amtrak for various terminal projects. Noticeable by the L&W logo I have found these in the wild from Virginia to Texas.
If you are looking to save some money by purchasing your signaling equipment from Harmon Industries, located near Kansas City, you will have to look elsewhere because they were purchased by GE in 2000. However some of their signals can be encountered from time to time, specifically on the Amtrak Shore Line and Atlantic City Line.
If you see a logo that sort of looks like an H with a swoop, I think this may have been something related to Harris Technologies that also got swept up in a GE joint venture in 2001. These also appeared on Amtrak and other commuter systems in the northeast and Caltrain.
Finally we have this blatant CLS-20 knockoff that I encountered on the low budget Buckingham Branch that have the same clam-shell look, but no glass lens or even that many LED modules. I'm not sure, but UP might also be investing in these bargain bin clunkers.
Well those are all the ones I can dig up at this time. If any of you are aware of others please let me know in the comments!
First up, L&W Industries is based in rural Missouri and made some inroads with Union Pacific or its predecessors in the 1980's before being generally rejected for the Safetran. It's current claim to fame is the LED searchlight still being bought by Amtrak for various terminal projects. Noticeable by the L&W logo I have found these in the wild from Virginia to Texas.
If you are looking to save some money by purchasing your signaling equipment from Harmon Industries, located near Kansas City, you will have to look elsewhere because they were purchased by GE in 2000. However some of their signals can be encountered from time to time, specifically on the Amtrak Shore Line and Atlantic City Line.
Subscribe to:
Posts (Atom)