News   May 03, 2024
 973     1 
News   May 03, 2024
 596     0 
News   May 03, 2024
 287     0 

New GO Train Control+Signalling (PTC, CBTC, ETC) -- Safety & Subway-Like Frequency

It's on this basis that the original SelTrac signal system was designed, built and operated. If the Dutch have manged to port this to a heavy rail application, this is quite a big deal.

Dan
The original Seltrac system is based on LZB, which has been in use on mainlines in Austria, Germany & Spain since the 70s, and some metro systems in the German-speaking world. You can get 30tph with EMUs with LZB or ETCS L2 with very short blocks (as is done on the Munich S-bahn core section). The main benefit of using virtual sub-sections (proposed with the ETCS Hybrid L3 specification) versus fixed blocks is that you reduce the amount of track-based vacancy detection equipment you need.

Neither the Weston sub nor a fixed block subway system come anywhere near the frequency of movement authority updates of a moving block system.
The Dutch innovation is to use virtual blocks which exist purely within the digital realm (i.e. without any associated track circuits, axle counters or physical signals) as a subdivision of physical blocks which do exist with axle counters. It would be outrageously expensive to install that many physical signals and track circuits, and even if we did, it would be absurd to expect drivers to be able to read a new physical signal every five seconds along the line.

With LZB & ETCS Level 2, the movement authority is also transmitted directly to the cab at regular intervals, so there is no lineside signalling. There are also no minimum block length requirements for either LZB or ETCS L2. The difference with Moving Block is that with LZB & ETCS L2, there is no way to confirm the train's integrity (i.e. that the train hasn't broken apart), so they are completely reliant on track-based vacancy detection.
 
Last edited:
New Metrolinx article provides additional details, such as that they've decided to use Level 2 with a line of sight overlay for CN but not Via. Via will therefore also need to equip their trains with ETCS.

https://www.metrolinx.com/en/discover/world-standard-signalling-system-to-improve-go-train-service
Advanced signalling technology for rail, called the European Train Control System (ETCS), was invented to solve a challenge for train travel across Europe, before becoming a global standard for railways.

Now, it’s coming to Ontario – as the first North American jurisdiction to implement this technology.

A modern signalling system, like ETCS, is a critical success factor for GO Expansion – a program that will deliver faster, more frequent GO and UP Express service throughout much of the Greater Toronto and Hamilton area, from Toronto and Burlington to Oshawa, Markham and Brampton.

Specifically, ETCS Level 2, has been selected by Metrolinx to deliver the increased capacity required to run more trains on the GO network.

What does ETCS “Level 2” mean?
ETCS Level 2 (or ETCS L2) is a radio-based train control system that communicates train speeds, positions, and movements across a rail network.

It achieves this by calculating and exchanging information in real-time between equipment onboard trains and trackside infrastructure, through wireless Long-Term Evolution (LTE) radio technology.

The trackside signalling infrastructure acts as a centralized control system to monitor and manage rail traffic optimally.

This continuous track-to-train radio communication ensures safe and efficient train operation, allowing the advanced signalling system to deliver safe, fast, frequent, and high-capacity train service.

Why did we choose this technology?
“Enhanced Train Control is the future of railway signalling systems, enabling operators to safely increase capacity and decrease journey times,” said Jonita Delaney, Metrolinx vice president of technical systems integration.

“Metrolinx has selected ETCS L2 based on its suitability to operating model and the safety benefits it will bring to passengers, railway workers and level crossings.”

In Europe, with railways using different signalling systems, a train that crosses borders needs different equipment for each country. ETCS was devised to solve this problem with a common system. It’s the core signalling and train control component of the European Rail Traffic Management System (ERTMS) and has spread to several countries around the world – including Australia, China, India, Mexico, Saudi Arabia, South Korea, and Taiwan – because it offers improved safety, performance, and headway times.

In those jurisdictions, all trains and track equipment were fitted with ETCS to enable interoperability among various countries and operators.

In Ontario, GO Trains operate largely on track owned by Metrolinx, so ETCS L2 will be deployed throughout Metrolinx-owned track. Tenant passenger trains that operate on Metrolinx-owned track, such as VIA Rail, will also be equipped with ETCS L2.

Tenant freight trains, such as Canadian National Railway (CN) and Canadian Pacific Railway (CP), that use Metrolinx tracks to service their customers will not be equipped with ETCS.

Instead, conventional line-of-sight signals, which use coloured lights to manage traffic, will be maintained throughout Metrolinx’s territory to:

provide continued support for freight operators
support GO/VIA passenger operations during ETCS deployment phases
serve as backup for ETCS-equipped trains for the long-term, in rare cases of degraded operations
GO Trains operating on tracks owned by third party railroads will operate using existing line-of-sight (colour light) signalling systems and will transition seamlessly into and out of ETCS L2 at territory boundaries.

Headway is the time interval between consecutive trains, and safely reducing it is the key to running more frequent GO service.

Since trains cannot stop as quickly as cars, signalling systems are used to maintain space between them. Railway lines are broken down into “blocks” or zones using train detection systems, for example track circuits or axle counters.

“Metrolinx currently runs a fixed-block signalling system,” said Thomas Casselman, Metrolinx vice president of signalling and communications.

“Lights provide an indication of what the driver is supposed to do.”

The length of the blocks dictates how far apart the trains must be and, on many of the tracks where GO operates, their size is more appropriate for freight traffic.

Freight trains are typically longer and slower than faster, shorter passenger trains. In practice, that means that sometimes passenger trains are at a safe distance apart, but the current system doesn’t reflect that safe status, which results in fewer trains running on the tracks.

Newer, more sophisticated signalling systems, like ETCS, go beyond trackside lights and deliver instructions to each train’s driver, based on implementation of shorter virtual signal blocks in addition to the longer fixed signal blocks.

Because ETCS L2 is an overlay on the conventional signal blocks, freight trains (or other vehicles not fitted with the onboard equipment) can still be safely accommodated on the network at the same time as creating virtual signal blocks being optimised for passenger trains.

“When the location of the train is precisely monitored, trains can safely run closer to each other,” said Keith Ampalavanar, Metrolinx senior manager of signalling and communications at Metrolinx.

That’s where advanced signalling comes in as a key part of the GO Expansion program.

It’s proven technology that reduces headways and delivers both real and virtual signal information directly to the driver on board each train.

“ETCS is an off-the-shelf solution that we’ve already deployed around the world,” said Christophe Wacrenier, head of signalling at ONxpress, the consortium contracted by Metrolinx to deliver the GO Expansion – On-Corridor (OnCorr) Works project.

“It’s been developed over 20 years by all the different companies existing in the market.”

GO Transit will be the first passenger rail system in either Canada or the United States to use ETCS.

“This is a great opportunity to showcase the benefits of this technology,” Wacrenier said.

GO Expansion Advanced Signalling System
Advanced signalling systems, such as the ETCS L2, will be used to deliver more frequent service on key corridors of the GO rail network by making it possible for more trains to run safely on the same track.

By taking a data-driven approach to spacing the trains, systems like ETCS L2 make it possible for more trains to be on any one line at any given time – shortening the blocks and making calculations based on trains’ speed – resulting in more frequent service for customers.

“It’s a system that has a lot more potential for growth,” said Jonathan English, development phase proposal value creator at ONxpress Operations Inc. (OOI).

“It really can handle an enormous amount of train traffic on a pair of tracks, in comparison to the amount that we’ve managed to move in the past.”

Systems with ETCS are “allowing trains to get closer together while still respecting basic safety margins,” said Tyson Moore, a visiting lecturer at the University of Birmingham.

“Safety was one of the fundamental design principles.”

Metrolinx plans to run electric trains, equipped with advanced signalling equipment, as part of the GO Expansion program, which also includes building new track and improving infrastructure throughout the GO network.

This will result in trains departing more often, tripling GO service from 3,500 trips a week in 2019 to more than 10,000 in the future, offering customers better, faster, and more connected service across the region.
 
What about Amtrak is what came to mind to me..

The only Amtrak train affected is the Map,e Leaf, and it will only be impacted east of Burlington West. And if the plan is to retain physical blocks, it won't be held back much.

I do wonder if ML will retain every physical block.... they could take down some or all of the intermediate signals, maybe every second one. Mostly leave the home and approach signals for interlockings.

- Paul
 
The only Amtrak train affected is the Map,e Leaf, and it will only be impacted east of Burlington West. And if the plan is to retain physical blocks, it won't be held back much.

I do wonder if ML will retain every physical block.... they could take down some or all of the intermediate signals, maybe every second one. Mostly leave the home and approach signals for interlockings.

- Paul
Level 2 still has physical blocks defined by track circuits or axle counters so you wouldn't really save much money by having a different set of blocks for the physical signals. They would however need to have a lower track speed for non-ETCS trains since the visual signals need to account for the worst-case train type unlike ETCS which provides a customised movement authority that's tailored for the specific train.
 
What's the process for getting the thread's name changed? Now that Metrolinx has determined what systems they want to use, we could update the title to:

GO Transit: Railway Signalling (ETCS & CROR)

CROR being the Canadian Rail Operating Rules that includes the legacy signal system which will be retained outside of the core network.
 
Level 2 still has physical blocks defined by track circuits or axle counters so you wouldn't really save much money by having a different set of blocks for the physical signals. They would however need to have a lower track speed for non-ETCS trains since the visual signals need to account for the worst-case train type unlike ETCS which provides a customised movement authority that's tailored for the specific train.
The Amtrak-equipped Maple Leaf has a lower speed limit than the rest of the passenger trains on the corridor anyways. There will not be much time lost by having to run at a hypothetical 65mph freight speed versus the 79mph limit it has today.

Dan
 
What is regrettable about Amtrak is that the US left it to host railroads to develop systems and get spectrum allocation etc. when Europe was rolling out ETCS as an interoperability measure, and so PTC equipped Amtrak power is rolling into Canada but is likely to have to operate as if it didn’t have such capability. If the US and Canada (and Mexico) had adopted a single standard and brought governmental force to bear where it would have removed a burden from individual railroads, I would suggest advanced signalling would be much further along.
 
What is regrettable about Amtrak is that the US left it to host railroads to develop systems and get spectrum allocation etc. when Europe was rolling out ETCS as an interoperability measure, and so PTC equipped Amtrak power is rolling into Canada but is likely to have to operate as if it didn’t have such capability. If the US and Canada (and Mexico) had adopted a single standard and brought governmental force to bear where it would have removed a burden from individual railroads, I would suggest advanced signalling would be much further along.
Farther along perhaps, but also inferior. ETCS L2 is far more advanced than PTC in that it is a full control system that enables improved capacity and speeds (fewer unnecessarily restrictive indications) in addition to automatic train protection. PTC is an overlay that adds automatic train protection but otherwise leaves the signalling system relatively unchanged.
 
Last edited:
What is regrettable about Amtrak is that the US left it to host railroads to develop systems and get spectrum allocation etc. when Europe was rolling out ETCS as an interoperability measure, and so PTC equipped Amtrak power is rolling into Canada but is likely to have to operate as if it didn’t have such capability. If the US and Canada (and Mexico) had adopted a single standard and brought governmental force to bear where it would have removed a burden from individual railroads, I would suggest advanced signalling would be much further along.
I doubt it. Freight rail is dominant in N.A. and all three countries have already standardized there freight networks since frieght trains are constantly crossing over each others borders. Compared to Europe, how often do passenger trains cross over N.A. borders?

The signal system that exists in all three countries today is more than adequate for the freight network in N.A.

I don't really see any effort to standadise passenger rail networks across N.A. until we have more trains crossing over each other's borders.
 
Level 2 still has physical blocks defined by track circuits or axle counters so you wouldn't really save much money by having a different set of blocks for the physical signals. They would however need to have a lower track speed for non-ETCS trains since the visual signals need to account for the worst-case train type unlike ETCS which provides a customised movement authority that's tailored for the specific train.
ETCS Level 2 can support shorter block lengths than the existing multi-aspect signalling. Trains in L2 also do not need to read lineside signals. The two aren’t related.

The block layout should be optimized for ETCS L2 to maximize capacity.
 
ETCS Level 2 can support shorter block lengths than the existing multi-aspect signalling. Trains in L2 also do not need to read lineside signals. The two aren’t related.

The block layout should be optimized for ETCS L2 to maximize capacity.
I'm well aware that L2 doesn't require lineside signals. Those are just for the non-equipped trains. My point is that unlike Level 3, you still need physical blocks and thus physical detection for each block regardless of whether you install a signal.

Obviously the block layout will be optimized for L2. That may mean that some blocks are too short (for visual sighting reasons) in which case those would indeed need to be combined for the visual signals. The rationale for longer lineside signal spacing in that case is visibility rather than cost savings.
 
Obviously the block layout will be optimized for L2. That may mean that some blocks are too short (for visual sighting reasons) in which case those would indeed need to be combined for the visual signals. The rationale for longer lineside signal spacing in that case is visibility rather than cost savings.
You can have multiple track vacancy detection sections between lineside signals, so there is no relation. The distance between lineside signals determines the max speed for non-equipped trains, not the length of the track vacancy detection sections (which over open track denote the individual blocks for ETCS L2).
 
Last edited:
You can have multiple track vacancy detection sections between lineside signals, so there is no relation.
Again, for the third time, I never said that you need a lineside signal at each block. I simply said that omitting a lineside signal doesn't make a big difference to the cost of a block because the big expense is the track detection. The cost of maintaing all that detection equipment is commonly the limiting factor that determines the length of blocks in ETCS L2 implementions. That's why the Netherlands is proposing an L2+ system that uses physical blocks with long spacing for trains without train continuity detection (effectively L2) while equpiped trains have access to many virtual blocks within the physical block (effectively L3). That saves a lot of cost in maintenance compared to an L2 system that has very short blocks.
 
Again, for the third time, I never said that you need a lineside signal at each block. I simply said that omitting a lineside signal doesn't make a big difference to the cost of a block because the big expense is the track detection. The cost of maintaing all that detection equipment is commonly the limiting factor that determines the length of blocks in ETCS L2 implementions. That's why the Netherlands is proposing an L2+ system that uses physical blocks with long spacing for trains without train continuity detection (effectively L2) while equpiped trains have access to many virtual blocks within the physical block (effectively L3). That saves a lot of cost in maintenance compared to an L2 system that has very short blocks.

That says to me that there would be a savings in removing intermediate fixed signals so long as they stay within the envelope for track detection....conventional blocks can easily be 5 miles or more in length, so presumably ETCS can accommodate this..
There is a considerable cost of maintaining each intermediate signal and circuitry..... the more of this that can be eliminated, the better the economics.

- Paul
 

Back
Top