High-Speed Signals (>79 mph)

mtuandrew wrote: ↑Sun Nov 15, 2020 12:10 pm

David Benton wrote: ↑Sun Nov 15, 2020 4:50 am I think it was a orphan system , they tried to reinvent the wheel, instead of going with a tried and true system like the NEC ones. Same with the Michigan system, one could be forgiven for wondering if they were trying to spend as much money as possible.

Less an orphan, more a system that didn’t win. ITCS, the GE-built system on the Michigan Line, was pretty early and pretty robust. I’m not sure why more roads didn’t adopt it. I can’t currently find more info on the former Illinois system but I think it was called ASES; they’re re-standardizing on Wabtec’s I-ETMS which has become the de facto American freight standard. Theoretically I-ETMS should be good for 110 mph operation, so I’m not sure what’s holding them up.

The NEC’s ACSES II is actually more robust and doesn’t depend on GPS. I think it is a descendant of some of the European systems that include wayside and in-track transponders. It’s good for a system where you don’t mind spending more on physical infrastructure that ties into existing cab and wayside signals.

As a locomotive engineer who uses both systems, I can say say in all honesry the I-ETMS system is way better than the ACSES system. I think with this system I could run a train to Cleveland over parts of CSX that I havent been on in close to 20 yrs. and not have an issue.

The I-ETMS system is more techy. We have a seperate screen with a "map" of the route along with speeds both temp and perm, curvature, grades, signals, crossings (if set up it will blow the crossing sequence at each crossing) and work areas. One needs to sign in to the computer and double check things like tonnage length engine number cars and to be sure the bulletin matches whats on the computer. Its accurate as hell I would guess within 20ft or so.

HOWEVER account it being so accurate and the way the system is set up you cant to quick out of or into speed restrictions it WILL let you know. If you have an ooppsie it WILL download the issue then just about everybody in the world will know about it.

ACSES basically get on and go. It dosent care about train length or tonnage engine number ect ect. Your assigned a letter grade for your train A,B,C,D.
A= High Speed ACELA type trains
B= Regionals less than 125
C= Long distance think LSL
D= work trains
Its all based on top speed and braking curves. You as the engineer are still responsible for train handling and your PCs. Will it take action if it dosent like what you are doing yes it will HOWEVER it wont down load and broadcast your ooppsie to the world. The little black box behind you along with the cameras has it and its saved forever.

WhartonAndNorthern wrote: ↑Wed Nov 11, 2020 7:25 pm I believe they envision that there's room for a future "moving block" PTC system that can one day replace an automatic block system and hopefully provide an "easy" (HA!) upgrade for dark territory.

Actually there's one FRA-approved PTC system capable of doing this, supplied by the former signalling department of GE Transportation (now aquired by Alstom). It's ITCS. You might know that ITCS was a system overlaid on existing ABS signals on Amtrak Michigan Line (and nowhere else in North America). What you might not know is that the same system has been operating on Qinghai-Tibet railway in our country, between Golmud and Lhasa since 2006.

Harsh environment on the Tibetian Plateau inhibits cost-effective maintenance of any wayside signalling equipment other than switch machines. ITCS works here as a standalone "virtual block" system, without a single physical signal, transponder or track circuit. Trains locate themselves by differential GPS and the positions are reported to back office servers. Integrity of trains is secured by the EOT device. The on-board system of ITCS periodically communicates with the EOT device, reporting its status to back office servers. With these information, back office server "shunts" and "clears" "virtual blocks", and calculates moving authorities for all trains.

One minor difference is that instead of the 220MHz radio the standalone ITCS uses GSM-R for communication between wayside and on-board systems . GSM-R is quite common anywhere except North America.

And as the system works without a backup, it was quite problematic during the first few years - once GSM-R or GPS signals were down, the territory becomes "dark" again. With redundant GSM-R / GPS-differential stations added, its reliability has been gradually improved.

In the years to come, it will be replaced by a new, indigenous "virtual block" system interoperable with CTCS (Chinese version of European ETCS/ERTMS ) named CTCS-4, similar to ETCS-3. Reliablity will be further improved by adding a few wayside devices back (e.g. track circuits within the limits of staffed interlockings, transponders (balises) in tunnels where signal is quite weak), and by replacing GPS with the more accurate GNSS.

mtuandrew wrote: ↑Sun Nov 15, 2020 12:10 pm Less an orphan, more a system that didn’t win. ITCS, the GE-built system on the Michigan Line, was pretty early and pretty robust. I’m not sure why more roads didn’t adopt it. I can’t currently find more info on the former Illinois system but I think it was called ASES; they’re re-standardizing on Wabtec’s I-ETMS which has become the de facto American freight standard. Theoretically I-ETMS should be good for 110 mph operation, so I’m not sure what’s holding them up.

That's probably because when overlaid on automatic block signalling, ITCS works a "sillier" way than I-ETMS does, quite like the European ETCS level 2 system. Under I-ETMS, WIUs are set up at existing signal locations to interface with vital signal circuits. I-ETMS on-board system talks peer-to-peer via radio with these WIUs. With ITCS, the WIUs also encode signal indications at physical signal locations. But instead of being sent directly to the on-board, these information are relayed via WLAN to some remote back-office servers called WIU-server, usually located at control points. Then WIU-server communicates with ITCS on-board via 220 MHz radio frequency. Even though signal indications are locally generated and will be locally consumed, under overlay-type ITCS, information goes a much convoluted way, from automatics to CPs, then back to trains approaching these automatics, than I-ETMS does.

ETCS L2 is even "worse" under this sense. Traditionally European railways were in favor of centralized signalling equipment. On a typical French LGV line equipped with TVM-type signalling systems, for example, the transmitters and receivers of UM-71/UM-2000 track circuits are centralized at station signalling rooms - along with interlocking machines - and are connected to electrical joints at individual block points with dozens of miles-long cables. The interlocking machines collect the occupancy status of all track circuits in its territory, determine the cab signal codes to transmit, and send them back via these long cables. The American tradition was to install signal huts at every block point , eliminating as many signal cables as possible. Occupancy of blocks can be relayed from block to block, via rail or a single pair of line wire.

With the introduction of ETCS-L2, the situation is even complicated. Instead of sending signal indications via track codes, ETCS-L2 requires all interlockings with in a region - can be an entire high speed line - to relay occupancy status from train detection mechanism to a large, central "WIU-server", named "radio block center" or RBC, usually located hundreds of miles away, before limits of moving authorities are calculated and transmitted back to trains via GSM-R. So you have train detection generating information from the field, collected by long cables or relayed via communication to interlocking machines miles away, then relayed to RBCs hundreds of miles away, and finally wirelessly back to trains at field.

IMHO these RBC/WIU-server stuff only make sense when you go beyond fixed block signalling. Without wayside train detection, trains must report their own locations and you might need a server to manage them. This is how the current generation of CBTC works, under which they are called "zone controllers" or ZC. CBTC trains directly communicate with ZCs, reporting locations and getting movement authorities, without the above many steps. (The next generation CBTC, however, eliminates the need of ZCs by enabling trains to talk directly to each other). ETCS-L2 strives to be a transition between conventional signalling system (ETCS L0/L1) and CBTC (ETCS L3), and it's no wonder being that "ugly".

That was a master class, hxa. Thank you!

One question: what does “WIU” stand for? You explained the function of a WIU server as a central control point but I’m not sure what the acronym represents.

mtuandrew wrote: ↑Wed Nov 18, 2020 6:30 pm That was a master class, hxa. Thank you!

One question: what does “WIU” stand for? You explained the function of a WIU server as a central control point but I’m not sure what the acronym represents.

WIU stands for wayside interface unit, used to interface between existing wayside signalling circuits and train control systems. ACSES/I-ETMS WIUs are connected to radios, enabling a p2p communication between wayside and on-board. ITCS WIUs relay signal indications to central servers at interlocking control points.

In European ETCS, there's something similar: the lineside encoder unit or LEU. ETCS-L1 needs LEUs to interface between existing signalling and variable balises. (Variable balises are balises/transponders with variable/selectable datagrams)

As for "WIU-server", the name is not found elsewhere except in ITCS. lt collects wayside indications from WIUs and generates movement authorities for trains. Works similar to an ETCS radio block center (RBC). Actually WIU-servers are also known internally as RBCs (though not in compliance with ETCS standard) in China Railways.

electricron wrote: ↑Sat Nov 07, 2020 2:03 pm FRA regulations on maximum speeds have been repeated so many times on so many threads, maybe railroad.net administrators should post them here and place it at the top as the first thread to be read? That way questions like this one we would not have to rehash the existing regulations over and over again.
Keeping it simple as much as possible;
No Classification no passenger trains allowed
Class 1 allows speeds up to 15 mph
Class 2 allows speeds up to 30 mph
Class 3 allows speeds up to 60 mph. No block signaling required.
Class 4 allows speeds up to 79 mph. It requires block signaling using wayside semaphores or lights.
Class 5 allows speeds up to 90 mph. It requires block signaling with in-cab signals.
Class 6 allows speeds up to 110 mph. It requires reinforced gates at all public crossings.
Class 7 allows speeds up to 125 mph. It requires reinforced gates at all (+ private) crossings.
Class 8 allows speeds up to 160 mph. It requires complete grade separations at all crossings.

Illinois was never going to require reinforces crossings at private grade crossings, therefore the train speeds were never going to be faster than 110 mph. Why anyone thinks that the Midwest trains will be going 120-125 mph I have no idea.

Additionally, there are many more details with the track classifications by the FRA, take the time to read the appropriate regulations one night you might need help going to sleep.
I kept the limiting factors as simple as I could earlier in this thread.

This is not completely correct, 110 mph does not require "reinforced" crossing gates. See for example the Empire Corridor which has multiple grade crossings with simple dual gates, none of that quad gate or occupancy detection stuff. I think the only difference at all is that they have "High Speed Train" signs up.

Also it should be noted that the "class" designations you are referring to are FRA track standards, and also include different speeds for freight trains up to Class 5. They are completely separate from signals and grade crossing related standards, not to mention crash resistance and line sharing standards, which is why we have some odd situations like the Acelas topping out at 150mph on Class 8 track. I know you mentioned that there is more to it than you said, but I just thought that in particular might be worth adding.

FRA regulations on maximum speeds have been repeated so many times on so many threads, maybe railroad.net administrators should post them here and place it at the top as the first thread to be read? That way questions like this one we would not have to rehash the existing regulations over and over again.

Off-topic, maybe we can start a small wiki on the site, tied to the board to reuse the logins? I don't know if PHPBB has a tie-in wiki to help prevent spam on it. Something for another topic.

§ 213.347 Automotive or railroad crossings at grade.
(a) There shall be no at-grade (level) highway crossings, public or private, or rail-to-rail crossings at-grade on Class 8 and 9 track.
(b) If train operation is projected at Class 7 speed for a track segment that will include rail-highway grade crossings, the track owner shall submit for FRA's approval a complete description of the proposed warning/barrier system to address the protection of highway traffic and high speed trains. Trains shall not operate at Class 7 speeds over any track segment having highway-rail grade crossings unless: (1) An FRA-approved warning/barrier system exists on that track segment; and (2) All elements of that warning/barrier system are functioning. Class 7 is for passenger train speeds over 110 mph.
You have a valid point.
For Class 6, 90-110 mph speeds, it seems the speed limitation is based more on curves and cants limits; more than 3 inches but less than 5 inches, with vehicle floor roll angle less than 8.6 inches. Along with increase, more frequent, track testing by the track owner.
Never-the-less, Illinois DOT spent a lot of money on safety for quad gates and locking gates at private at grade crossings anyways, which apparantly they did not have to do - but did! Maybe a political consideration over a regulatory, but they have set the precedent of a higher standard for new Class 6 routes. Worse yet, now they are only promising Class 5 speeds, maybe they have set a new precedent for higher standards for new Class 5 routes?
Class 1-5 cants and curves are less than 3 inches with a reduced track testing maintenance frequency, possibly “the” reason why Illinois DOT is just promising Class 5 90 mph maximum speeds now on the Lincoln route after improving all the at grade crossings to Class 7 levels.
Maybe, just maybe Illinois could take advantage of the at grade crossings over build if they ever double track the entire Lincoln route, if that is the issue preventing faster than 90 mph speeds. Who knows what their plans will be?

Amtrak706 wrote: ↑Fri Nov 27, 2020 3:37 pm This is not completely correct, 110 mph does not require "reinforced" crossing gates. See for example the Empire Corridor which has multiple grade crossings with simple dual gates, none of that quad gate or occupancy detection stuff. I think the only difference at all is that they have "High Speed Train" signs up.

Where on the Empire Corridor is the track rated Class 5 or above, that would require anything relating to quad gates or occupancy detection? It's been designated as a potential high speed corridor, but by no means meets the definition and classification to allow anything over 79mph.

ctclark1 wrote: ↑Sun Nov 29, 2020 10:33 am

Amtrak706 wrote: ↑Fri Nov 27, 2020 3:37 pm This is not completely correct, 110 mph does not require "reinforced" crossing gates. See for example the Empire Corridor which has multiple grade crossings with simple dual gates, none of that quad gate or occupancy detection stuff. I think the only difference at all is that they have "High Speed Train" signs up.

Where on the Empire Corridor is the track rated Class 5 or above, that would require anything relating to quad gates or occupancy detection? It's been designated as a potential high speed corridor, but by no means meets the definition and classification to allow anything over 79mph.

MAS for most of the Hudson line between the division post MNRR is 90 to 95. Above MP 126.8 its 110mph. No special gates up until recently there was a private crossing in the 110mph zone with no gates

The Keystone Corridor also has at least one private crossing without gates (the one where a Metroliner cabcar slammed into a tractor a couple years back) but the Empire corridor is the only one I can think of that has normal non-quad crossing gates on public grade crossings over Class 6 track.

Also electriciron I think you are conflating FRA track classification with grade crossing protection. Class 5, 6, etc. only refers to track geometry and maintenance specifications, nothing else. The grade crossing regulations are separate and placed on top of the track class stuff.

I replied earlier, do you know what “you have a valid point” means?

Never-the-less, the last 110 mph allowed new build track test in Illinois had increased gates at all, let me repeat that last word again, all at grade crossings. All the rebuilt tracks between Joliet and Alton, around 220 miles of track, except within Springfield where the trains are slowing down below 79 mph. Another key word in the first sentence of this paragraph was new. I’m sorry, the Keystone and Empire corridors faster than 90 mph speed zones are not new.

What is required by regulation and what is required politically can be different.

Sorry, it was a long day and I forgot I had already said that, haha. My bad.

Railjunkie wrote: ↑Sun Nov 29, 2020 11:15 am

ctclark1 wrote: ↑Sun Nov 29, 2020 10:33 am

Amtrak706 wrote: ↑Fri Nov 27, 2020 3:37 pm This is not completely correct, 110 mph does not require "reinforced" crossing gates. See for example the Empire Corridor which has multiple grade crossings with simple dual gates, none of that quad gate or occupancy detection stuff. I think the only difference at all is that they have "High Speed Train" signs up.

Where on the Empire Corridor is the track rated Class 5 or above, that would require anything relating to quad gates or occupancy detection? It's been designated as a potential high speed corridor, but by no means meets the definition and classification to allow anything over 79mph.

MAS for most of the Hudson line between the division post MNRR is 90 to 95. Above MP 126.8 its 110mph. No special gates up until recently there was a private crossing in the 110mph zone with no gates

To finish this thought, a long night in a loud hotel fuzzed up the brain. There are quad gates above Albany, when Amtrak put the second track in up to SDY all 3 crossings are protected by quads.

Huh, I didn't know that. Do you know where?