Search This Blog

Thursday, September 29, 2016

PTC and Stub End Terminals

Despite what some on the news media are saying, the PTC mandate would have not stopped the recent overrun at Hoboken Terminal, nor was it ever intended to.  The law, as drafted by congress, specified 4 things that PTC should protect against. 
  • Train to train collisions
  • Speed in excess of engineering limits
  • Workers in designated work limits
  • Open switches
Just like Stop Signal violations are not included in this list, stopping short of the end of track, or any other obstruction, is also not included on the list.  In fact, the FRA PTC regulations specifically exempt "terminal areas" from the PTC requirement, as long as the system enforced a 20mph (Restricted speed) limit in said terminal areas.  So as I said, the law, as written, and the approved PTC plans would not include terminals such as Hoboken.

Remember that Hoboken Terminal is already cab signal territory and all the trains are ATC equipped.  Despite the news reports that the train was moving at a "high rate of speed", from the condition of the cars it was likely only traveling at 10-15mph, well in accordance with the rules and the permitted speed of the track.  If PTC had been present to enforce timetable speeds, no action would have been taken.  If the end of track had been set as a positive stop point, yes, that would have stopped the train....and every other train that ever used the station, about 100-200 feet short of the end of the track because that is the safety margin that PTC systems work with.  That means every terminal station track in the country would instantly lose two car lengths.  Already overcrowded trains would need to be shorter and the net result would be more drivers dying in road accidents.

Bumper block at Hoboken dating from 1907 when a lot of rolling stock was still made of wood.

Setting aside the costs of adding PTC to complex terminal interlockings, even if you did something clever like step the timetable speed down to like 5mph, at that point you start running into issues with either slowing trains to unacceptable levels (remember the 100-200 foot buffer zones still apply) or not slowing trains enough enough to prevent a similar accident (ie safety theatre).  Why not?  Because even at 5-10mph, a modern train would blow right throw Hoboken terminal's 1907 vintage bumper posts shown above.  That's right, in this case I am pinning the blame for this accident's drama on bumper blocks that act more like a ramp to launch railcars into the roof than any sort of stopping mechanism.

Here at La Salle St station in Chicago we see a more modern setup with standard bumpers positioned in front of a deceleration zone and finally backed up by a thick concrete wall if those prove insufficient.  Other stations, especially those with high level platforms, have sufficient defenses to stop a train traveling at even the full 20mph.

Trains are not elevators.  We can't stop every accident and if we try we will only throw more and more money at ever smaller gains.  Money that could be spent on new lines and new services that can take people off the roads where over 30,000 Americans die each year.

Thursday, September 22, 2016

PRR Main Line Re-Signaling and Other News

Well I can fully confirm that the much anticipated resignaling project on the former PRR Main Line (currently NS Pittsburgh Line) has begun.  Perplexingly it has started on one of the line's newest sections located at the eastern end between CP-CANNON (exclusive) and (at least) CP-LEWIS.  As was foretold by regulatory filings NS is adopting a Rule 562 configuration by eliminating the wayside intermediate signals in favor of the already present cab signals.  The remaining wayside signals at interlockings will be equipped with 'C' boards for Rule 280a "Clear to Next Interlocking"

Replacement signaling will not spoil the view before the position lights are removed
This will see the end of a great deal of bi-directional PRR PL mast signals and a 3 track, 6-signal tubular gantry just east of CP-HAWSTONE.  As of this time the pneumatic switches are still in place at CP-PORT and CP-MIFFLIN and no new signaling has appeared from CP-CANNON eastward.  Furthermore, no replacements for the PL dwarf signals have appeared at CP-MIFFLIN. Both Conrail and NS era replacement color light signals are also being replaced at CP-PORT and CP-MIFFLIN.

Up north, CSX is finally reaching the end of its multi-year NY Central Main Line re-signaling project, with small target searchlight automatics being removed in the Batavia region where Engine 999 performed her 112mph record run.

I also just heard that the B&O CPL signals at VINE and LOCUST interlockings have also just been retired, so it looks like I wasn't quick enough getting followup shots with those.

 Elsewhere on CSX, the famous modern-mounted SCL elephant ears around Hamlet, NC continue to be whittled away.

 In a bit of interesting news, on the former Conrail Chicago Line, NS is continuing to apply Conrail blue interlocking nameplates, however they have added the 'CP' designation to the signs, which Conrail omitted, preferring just the name or milepost.

Finally, in another bit of Conrail related news, it looks like SEPTA is on its way to replacing the last Conrail era signals on its West Trenton line at CP-TRENT.  Most of this 90's vintage signaling bit the dust when SEPTA decided to separate itself from the CSX line rather than try to handle the joint running any longer.

Well that's all I have.  Remember to get your buts out there and take some pictures before, like in the case of VINE and LOCUST, it is too late.  I'm specifically referring to the PRR Main Line here.  Her out there!

Tuesday, September 13, 2016

Conflicting Route Conundrum

So the following photo drew a number of questions on both its own page and also a few of the signaling forums.  As one can see it appears that two conflicting routes have been lined over the same switch at the end of a passing siding, something that appears to indicate some sort of failure in the interlocking, or some photographic trick like multiple exposures.  The photo was taken on a Norfolk Southern line in Georgia and it's location is one reason that people are so confused.

Before I give away the answer I'll first state that the signaling logic is functioning as intended and there was been no photographic trickery.  Still, Clear and Diverging Clear appearing next to eachother in this context should be contradictory.  Here's I'm going to post another photo from a different part of the country that is a pretty big hint about what's going on here.

This was taken on the former ATSF Raton pass line where the last operational semaphores in North America are located. These semaphores are constantly being photographed, but nobody seems to be confused as to why adjacent signals can both display clear like this.  Well, there are two reasons.  The first is that there is something about Diverging Clear sitting next to a Clear at the end of a siding that just seems plain wrong.  The other reason is because the sort of signaling where this circumstance occurs is much more common out west than it is in the east.

This is your last hint before I give away the answer/
 If you haven't figured it out yet the answer to the puzzle is that the signals in both these pictures do not protect an interlocking and are not part of a CTC setup.  They are part of an ABS-TWC (aka NS Rule 271) arrangement, that also likely an example of Automatic Permissive Block .  The signals operate automatically based on occupancy of the line ahead and, more importantly for this post, trains exiting the siding do so over a spring switch so both routes through the control point are valid.

Surprise!  It's a spring switch and its also not an interlocking.
ABS-TWC / Rule 271 can actually take a couple of forms.  The first involves signals placed only on the main track so trains on the main and the siding will encounter the same signal and act accordingly.  The other places signals on both the siding and the main.  Since they are operating automatically and they 'protect" a trailing spring switch, both signals will default to a Clear indication.  Out west if there is only one possible route from a signal then the railroads don't bother with a lower head since they are happy to have route signaling handle the switch speed through the timetable.  However in the east it is standard practice to provide reduced speed movements with a  Diverging or Reduced Speed Clear signal indication.

Same situation, one less signal.
 This arrangement is not always an example of APB because APB involves an element of traffic control with absolute headblock signals that prevent trains from entering a line segment where an opposing movement is under way.  Typically the presence of absolute signals at siding exits implies APB, but this is not always the case and NS Rule 271 operation requires Track Warrants for traffic control.  In theory an APB line can operate under Rule 261 with the Conrail Southern Tier Line being one example

Well, what's enough of me rambling on.  Like I said this arrangement is far more common out west and there you don't see the Diverging aspect.  This is why it is absolutely critical to have a good understanding of how traffic control applies to various methods of signaling.  Realizing the line was running under Rule 271 with manual traffic control, it is clear how two trains could never take both routes simultaneously.

Wednesday, September 7, 2016

Letting the Air Out

Well I just completed a round trip on the NEC, paying a bit more attention to some of the signaling changes than I normally do and I have the sad duty to report that both DOCK and MIDWAY interlockings have been completely stripped of their pneumatic point machines.  This isn't any surprise as both were in the midst of years long re-signaling efforts, but at least in the case of DOCK it is a severe disappointment as it was a very high density interlocking where Amtrak has been known to preserve pneumatics. 

MIDWAY will be missed because it was a fine example of a pneumatic 4-track complete crossover, with many of the turnouts having two A-5 machines.  This leaves only HOLMES and OVERBROOK as "proper' PRR 4-track air plants. 

DOCK on the other hand needs no introduction as it is a sprawling interlocking covering the Newark Penn Station complex.  As recent as 10 years ago Amtrak had still been installing brand new A-5 point machines during a turnout renewal project.  As with the Jamaica terminal, Penn Station and 30th St, I was hoping that DOCK would win a similar exemption.

At least we had them for longer than other interlockings like MORRIS and LANDLITH, which were seemingly converted overnight.

Wednesday, August 31, 2016

Additional News Items

I was so pumped with LENOX tower escaping to live another couple years that I forgot to post the rest of the news items that I had accumulated.  Sorry for the lack of hard content lately, just been too busy to get started on my next big project or finish up some of my remaining old ones. 

I first wanted to remind everyone that CSX is in the process of ruining the last great bastion of NY Central signaling in the Buffalo-Depew area, so if you happen to be nearby, please go and check it out before it gone forever.

I wanted to call out NS for replacing 1990's vintage Conrail signaling when Union Pacific is keeping similarly aged Southern Pacific target signals in place and going so far as to re-use the existing relay case as well.

I wanted to confirm that NS has also finally replaced the Southern vintage signals in the Charlottesville, VA area.  While the old signals definitely had more character to them, the replacements will be able to display a R/R/Y Restricting indication in addition to Y/G Approach Diverging.

One case of addition by subtraction is the removal of NS's experiment with wind-up, wind-down ladderless signals.  These two soul-less examples are near the Danville, VA Amtrak station.  Just goes to show that things could be worse.

I'm going to keep the spotlight on NS here, this time on their Southern main line in North Carolina.  If you were just going to double track things anyway, why dd you bother removing the Southern masts just to replace them again?

I also have news on the Reading Line re-signaling project, specifically that it will stretch to at least CP-LAUREL, nixing the Conrail signals there and at CP-BLANDON.

Well I think that clears the docket.  Have a great Labour Day weekend!

Wednesday, August 24, 2016

LENOX Tower Gets Reprieve, PRR Main Line :-\

So I just heard that the State of Illinois did not get a TIGER grant earmarked for the replacement of the LENOX tower complex between Alton and Granite City on the Amtrak Lincoln Corridor.  Without the government funding, UP has no plans to decommission the tower and its GRS Model 2 interlocking machine.  I guess one benefit to corporate greed is that when they feel the state will pony up for capitol improvements, the neglect (aka "historic preservation") sets in.  Let's hope for further austerity measures going forward.

LENOX in 2005
On the other side of the coin it looks like the long rumored re-signaling effort is starting up on the PRR Main Line between Duncannon and Lewistown.  This will be a Rule 562 conversion so we won't be forced to deal with Darth Vader intermediates and the absolute signals will be sporting 'C' boards.

Friday, August 19, 2016

Better Know a Signaling System: BNSF Combined Signals

In 1996 the Burlington Northern Railway merged with the Atchison, Topeka and Santa Fe.  Typically when two railroads are involved in a mega-merger the signaling systems are so incompatible that the surviving railroad needs to employ multiple signal aspect systems, or the systems are similar enough that they can be merged without much change.  However when looking at the famous BNSF signal chart we see an almost bizarre mix of route and speed signaling, almost as if two different systems that should have been kept separate were merged.  The shocking truth is that prior to the merger, both BN and ATSF used very similar bizarre signaling systems and the merger was nearly seamless.

Like I said, what stands out is the mix of route and speed signaling.  At home signals it's all route with Diverging Clear and Diverging Approach.  At the distant however we see Approach Medium and Approach Limited, traditionally speed signaling aspects.  This is why it feels like one road was speed signaled and one was route signaled since it would be easy enough to just remove the  speed signal aspects at interlockings and just rely on the timetable while applying speeds to the distant signals. However if one goes back and looks at the original ATSF and BN rules, we can see what actually happened.

BN Chart is a PDF so click the link.
Both these nominally route signaled western roads included speed signaling aspect Approach Medium and Approach Restricting while the ATSF additionally included Approach Limited .   During the unification process BN's Approach Diverging was simply folded into Approach Medium so that the rule now reads "Approach next signal at Medium speed (40mph) AND prepared to proceed on diverging route".  In practice the Y/Y is still used for diverging routes 40mph and below while *Y* tends to be used for 4-block signaling, however this was not always the case on legacy installations, which allowed for two lamp distant masts. 

Approach Medium (Diverging)

Approach Medium (to Stop)
The other signal that stands out is the use of Y/G for Advance Approach.  Now with the Y/Y Advance Approach popular in the east being used for Approach Diverging, it does make some sense to use Y/G, but why was *Y* labeled Approach Medium and then used in the role of of Advance Approach as seen in the preceding video?  Well on BN this remains a bit of a misery, but from the ATSF side of the family tree we got what became the final BNSF rule that states pass next signal at 50mph AND be prepared to advance on diverging route.  It's basically filling the role of Approach Limited, or in Union Pacific parlance, Approach Clear 50. As BNSF re-signals it appears to be common practice for distant signals to display Y/G Advance Approach for Diverging Clear in the 30-50mph range and Y/Y Approach Medium for Approach Diverging (or the extra block length gained with an interlocking allows for a more permissive 4-block signaling). However this could just be BNSF installing signal lamps that aren't used.

BNSF has been investing in 6 lamp distant signals that can display Y/G and Y/Y.
Now I know you must be looking at the signal chart and noticing that there is indeed an Approach Limited using Y/*G*.  Well again, read the rule, not the name.  It's basically the same as Advance Approach, only the speed is raised to 60mph for high speed turnouts.  On UP this is more appropriately called "Approach Clear 60".  

To make things even more confusing BNSF maintains a distinction between R/*Y* and R/Y/Y.  Whereas R/*Y* is typically used for back-to-back diverging movements, on BNSF it is labeled Diverging Approach Medium and is clearly intended for short signaling distances as there is no mention of proceeding on a diverging route and the speed is limited to 35mph instead of 40.  R/Y/Y is the more straightforward Diverging Approach Diverging, but appears on neither of the predecessor systems so as one might guess R/*Y* is still used for back to back diverging movements where BN traditionalists still hold sway.  I have to assume the R/*Y* Diverging Approach indications on the ATSF were converted to plain R/Y.

Diverging to stop at the second signal over Diverging Approach Medium
While BNSF allows for both Lunar white and *R* Restricting, the *R* is clearly the preferred choice nowadays.  Like UP, BNSF only employs Y/L for Approach Restricting without the obvious Y/*R* option.  However unlike UP there seems no desire to distinguish Yard routes with Lunar and occupied block routes with *R* as I have not encountered any 4-lamp signal heads. (For those who don't know, Approach Restricting provides advanced notice of being routed into a yard or non-signaled siding).

So there are the interesting parts of the BNSF signal aspect system.  Hopefully I'll get some feedback regarding how some of the overlapping indications are utilized out in the field and also some more of the historical context I'm not aware of.  Specifically I was unable to locate either a CBQ or GN signal aspect chart to trace to origin of the odd speed signaling components further.