In November 2017, I learned of a builtin order showing that Union Pacific was shutting down the Automatic Cab Signals (ACS) and ITCS that had been installed on the Lincoln Corridor as part of the multi-billion dollar 110mph upgrade. (You can see how they operate in this 2012(!) video).
Although it seemed to be part of the general ETMS PTC rollout and the elimination of the more unique ITCS PTC system I was puzzled that operation of ACS and PTC were to be exclusive, the former only to be cut in if the latter failed.
13.1.4 PTC/ACS Operations:
The Automatic Cab Signal (ACS) system on the lead unit must be cut out
upon successful initialization of the Positive Train Control (PTC) system and
prior to initiating movement. If the PTC system disengages, is cut out under
authority of the train dispatcher, or otherwise fails en-route while leading
engine is within PTC/ACS territory, the train must be stopped. After stopping,
the ACS system on the lead unit must be cut in prior to any subsequent
movement. If the ACS departure test cannot be performed while on energized
track, a departure test must be conducted in accordance with Rule 13.1.5 at
the train's next forward location where such a test can be performed. If unable
to cut in ACS system on the lead unit, the train must comply with Rule 13.3.3.
Of course this flies in the face of NS's integration of cab signals and PTC, which allowed them to reduce the reliance on a real time digital radio link.I was about to do a post on this situation and how ETMS and ACSES/CSS may be incompatible, but further research indicated that such incompatibility did not exist and I left the post in draft form.
Well a comment to my previous post on the elimination of IIATS on UP's METRA commuter lines in Chicago indicated that the Joliet Sub rule regarding ACS and PTC had become a System Special Instruction also covering ATS and CNW ATC.
Item 10-B Positive Train Control (PTC) Operations
8. PTC, ACS, ATC and ATS Operations
PTC must be the system utilized by the engineer. Upon successful initialization of PTC all subsequent systems (ACS, ATC, and ATS) must be Cut Out. If at any time PTC disengages, is Cut Out under the Train Dispatcher's authority or otherwise fails, the train must be stopped and the secondary system cut back in prior to any further movement.
This covers a HUGE amount of cab signaled territory stretching from Chicago through Iowa (CNW ATC), Iowa through Wyoming and the Portland Sub in the Columbia River valley. While the elimination of the clunky 2-aspect CNW ATC system was expected due to its uniqueness, ACS is a standard, 4 aspect, pulse code cab signal system generally compatible with what is used by NS, CSX and the Northeast passenger roads. As I said before, NS has actually expanded its CSS territory in response to PTC as it solves many problems with radio coverage. It is also used by Demver's new RTD commuter rail to support a reduced aspect signaling system.
As I pointed out with my NS example, the motivation behind this move is puzzling. UP went as far as to get FRA permission to operate ACS and PTC trains in mixed company, so there is no technical safety or technical reason that these systems cannot get along. (BTW, the document is a great primer on how both ACS and PTC are displayed in cab). In addition to solving much of the issue with a reliable and secure radio path, the FRA PTC regulations also give additional leeway to trains operating after a PTC failure where ATC (as could be enabled by ACS) is still working.
In theory it could mean an additional test or form, if such a test was not rolled into the existing PTC test. The ACS antennas behind the pilot are also a potential source of damage that can take a locomotive out of service. I asked around on some forums and the desire to allow foreign leaders on run-through freight was also brought up. There are also two significant technical issues that may be playing a role.
The first is that ACS does not map to wayside aspects in the same way PRR/CR/NS CSS due to the lack of strong speed signaling. For example, Approach Diverging triggers Approach in the Cab while Diverging Clear, Approach Clear 50 and Approach Clear 60 all trigger Clear in the cab. This cannot really be used to inform a speed based ATC function. Furthermore, UP never installed cab signal cut points in advance of wayside locations, so the cab signal never drops to Restricting in advance of a Stop or Restricted Proceed signal. All of this might require intermediate PTC data links to be maintained, or at the very least require specialty programming to deal with the difference between CSS and ACS practice.
On the other hand ACS has at least two advantages that PTC lacks and both are rooted in using the rails to deliver the cab signal codes. The first is that a train in a Restricted block can take advantage of a ACS upgrade after running its own length (granted not a huge deal with monster freight trains) and the second is that PTC is blind to any sudden circuit shunt in an already occupied block, whereas ACS will immediately fail safe. The second is likely to result in deadly accidents on an infrequent yet regular basis.
Hiawatha Sub Aspect Change Point near Baileyville, KS
Perhaps the ultimate answer to this question will be what happens on the Hiawatha Sub in Kansas, where UP installed its version of Rule 562 (Cab Signals Without Wayside Signals) on about 60 miles on single track main line. Google shows PTC antennas in place at what they term Aspect Change Points, which are spaced at twice the frequency of normal wayside signal locations, clearly to ensure that any occupied block has 1.5 miles of Restricting indication behind it. If ACS are retained on this line, it would show that PTC and ACS can function together and that UP (or the FRA) still wants some kind of signal indication to be presented to the crew. If the System rules apply on the Hiawatha Sub, then we will are likely seeing the first instance of a rail line signaled solely through the PTC system, something railroads have notably gone out of their way to avoid.
As of 2019 UP had already applied to abandon the ACS so only time will tell if the FRA will grant this and the other requests that will follow. At this point safety panic might be working in favor of the concept of a secure, reliable in track signaling system that has a lot of what is necessary to allow for 90+ mph operation without costly and time consuming certification processes. It might be worthwhile in trying to engage in the regulatory process to prevent such a short sighted move.
The UP has already applied to discontinue the ACS in the Portland area, filing FRA-2019-0043. My understanding is that they will do the same for all ACS and ATC lines as soon as they are permitted to from a regulatory standpoint. Except for the Hiawatha Sub, their other cab signal territory has full wayside signaling now (including former C&NW ATC territory). I think this is where the difference comes in with NS, which is using cab signals in lieu of permissive/intermediate wayside signals. The cab signals on the UP now are essentially a legacy method to enforce wayside signals, and since PTC does the same thing, the cab signals are no longer needed.
ReplyDeleteThanks for the correction. I guess we'll see what the FRA says.
DeleteThis is just bonkers from a technology standpoint because of the natural fail safety of coded track circuits. If you enter a block on a Restricted proceed, you cannot increase speed until passing the next wayside because the PTC datagrams can't detect a rail break. Moreover, ACS provides instant warning of sudden occupancy in the same block, something PTC cannot detect.
I thought they might want to keep the ACS too, but as someone recently reminded me, the cab signals are essentially a third system now that costs money to maintain, equip locomotives, run ACS tests, and arrange to have ACS-equipped power in the lead. I'm not very knowledgeable about the specific operation of PTC systems and didn't think about PTC not detecting sudden occupancy or broken rail. However, the risk is probably considered negligible like on the thousands of miles of mainline trackage that has never had cab signals, and with PTC enforcing upcoming signal aspects, it essentially replaces the function of cab signals overlaid onto standard wayside systems like the UP's anyway. The BNSF has also applied to drop ACS on commuter trains between Chicago and Aurora -- in dense traffic territory with tight timings (though they don't run them like they used to). If there was some potential or even perceived loss of safety, I would think the FRA would reject such petitions. The PTC systems must have enough functionality to cover this.
ReplyDeleteI recall from the early days of PTC development that the plan all along was for PTC to replace ATS, ACS, and ATC. Sorry I don't have a link to support this though.
ReplyDeleteNow Amtrak decided to go with ACSES as it's "PTC" system. But in reality ACSES is sort of a hybrid between the PTC the freight RRs chose and ACS/ATC. ACS/ATC is an integral subpart of the ACSES system. Sort of like a super enhanced ATC. Pax railroads that routinely share parts of Amtrak's NEC all decided to adopt ACSES as their PTC. MN, LIRR, SEPTA, NJT, etc.
I'm guessing here that since NS locos have to install ACSES if they want to run over the NEC, which requires ACS/ATC as a subsystem, it decided to keep ACS on other parts of its system because why not, they'll need it to run over the NEC anyway.
But UP? Their locos will never run over the NEC, so ACS/ATC is truly redundant (although admittedly a decent backup); but it looks like they want to remove them, just like BNSF is seeking for the Metra corridor. Probably the value of the backup doesn't exceed the hopefully rare cost of when PTC fails.
BTW Jersey Mike, if you have any interest in foreign systems check out the French Crocodile system! Similar to our ATS, it was invented by the Chemin de Fer du Nord (North Railway) of France in 1872 and the tests went so well that by 1880 they installed it on all of their locos and double tracked lines. And it's still in use today. Check out the youtube TGV cab rides and you'll see it on all of the non high speed sections of the lines such as around the terminals.
ReplyDeleteWikipedia has a good article in English on it. https://en.wikipedia.org/wiki/Crocodile_(train_protection_system)
And at the bottom of the article check out the external link for more details on the web archive
https://web.archive.org/web/20060304145854/http://www.du.edu/~jcalvert/railway/croco.htm#Croc
And I'll leave it to you and your readers to find out why they call it the Crocodile!