ERTMS Operation

You are here: Home > Forum > Miscellaneous > The real thing (signalling) > ERTMS Operation

Page 1 of 2

ERTMS Operation 26/07/2016 at 11:23 #83795
Gwasanaethau
Avatar
509 posts
Hi all!

I was wondering if anyone had any links to any examples of how ERTMS (level 1 or 2) worked. By this, I mean examples of how the system responds when a train runs over a balise (level 1) or when a movement authority is given/revoked; how and when this happens, and so on. I’ve delved into the world of Wikipedia and YouTube, but I have only been able to find very abstract information (eg “An MA is extended based on the position of the train in front and a safe speed profile is calculated.” (or words to that effect)), or obvious (and annoying!) marketing ‘propaganda’ (eg “ERTMS will revolutionise train signalling and is forecasted to provide an increase in capacity.”).

Does anybody know where I can get more detailed information similar to that available for standard UK signalling (something similar to Clive’s excellent descriptions and examples on his website, for example)?

Cheers!

Mark (Gwasnaethau)

Log in to reply
ERTMS Operation 26/07/2016 at 11:42 #83796
headshot119
Avatar
4869 posts
Hi Gareth,

My knowledge of ERTMS only extends to what has been done on the Cambrian route, and the research I did into the system when developing the Shrewsbury simulation. I don't claim to be an expert on ERTMS but have a general understanding of how it works.

ERTMS Level 1 (Without Infill) - This is an overlay over an existing signalling system, so lineside signals are still present. The movement authority is transmitted to the train through the balise at the lineside. So the movement authority can only update as the train passes over a balise.

ERTMS Level 1 (With Infill) - Still an overlay over an existing signalling system, so you still have lineside signals. The movement authority is either transmitted to the train through a continuous loop placed along the track (similar to how the underground works on one of the ATO lines), through extra balises placed (Without infill the balises are placed only at signals; think cheap!), or through radio (aka GSMR). This allows the train to recieve a new movement authority sooner, and thus accelerate quicker if the new limit allows it to.

ERTMS Level 2 - Lineside signals are a thing of the past in theory; however block markers are placed in various physical locations to give a method of degraded working. The position of the train is determined by the balise units at the track side, they are placed at fairly regular intervals and are calibrated so the ERTMS system knows there location (a real pain if a tamper takes one out apparently). Movement authoritys are transmitted to the train via radio (GSMR).

Movement authoritys on Level 2 only exist between block markers, it is NOT a moving block (That's what level 3 will be). So let's have a working example (Level 2); note I've completely made all the IDs up.

MH1000 ---- MH10002 ---- MH1004 ---- MH1006 ---- MH1008.

If I've given your train an MA from MH1000 to MH1008, and the train is just passing MH1000, and I shorten the authority down so you only have it as far as MH1006 the train will automatically recalculate a safe braking curve, and will do nothing if it's within that braking curve.

With the same example, but I shorten the MA so you only have it as far as MH1002 the train will calculate a new safe braking curve, will realise it needs to stop fairly quickly and will brake the train fairly quickly.

With the same example, but I completely cancel the movement authority the train will go straight into emergency braking.

"Passengers for New Lane, should be seated in the rear coach of the train " - Opinions are my own and not those of my employer
Log in to reply
ERTMS Operation 26/07/2016 at 13:10 #83797
clive
Avatar
2736 posts
A few years ago a departure from Aberyswyth almost hit a petrol tanker on a level crossing. If you can find the RAIB accident report, there was a fair amount of detail in it.
Log in to reply
ERTMS Operation 26/07/2016 at 14:21 #83798
TUT
Avatar
505 posts
" said:
A few years ago a departure from Aberyswyth almost hit a petrol tanker on a level crossing. If you can find the RAIB accident report, there was a fair amount of detail in it.
This one:

https://assets.publishing.service.gov.uk/media/547c8fdded915d4c0d000171/R112012_120627_Llanbadarn.pdf

?

Log in to reply
ERTMS Operation 26/07/2016 at 17:45 #83799
Steamer
Avatar
3913 posts
" said:
The position of the train is determined by the balise units at the track side, they are placed at fairly regular intervals and are calibrated so the ERTMS system knows there location (a real pain if a tamper takes one out apparently). Movement authoritys are transmitted to the train via radio (GSMR).
So does that mean that there are no track circuits or axle counters on a Level 2 route? That must be tricky over pointwork.

"Don't stress/ relax/ let life roll off your backs./ Except for death and paying taxes/ everything in life.../ is only for now." (Avenue Q)
Last edited: 26/07/2016 at 17:46 by Steamer
Log in to reply
ERTMS Operation 26/07/2016 at 17:59 #83801
headshot119
Avatar
4869 posts
" said:
" said:
The position of the train is determined by the balise units at the track side, they are placed at fairly regular intervals and are calibrated so the ERTMS system knows there location (a real pain if a tamper takes one out apparently). Movement authoritys are transmitted to the train via radio (GSMR).
So does that mean that there are no track circuits or axle counters on a Level 2 route? That must be tricky over pointwork.
As I understand it the spec doesn't require track circuits or axle counters for a level 2 route. Though the Cambrian is axle countered throughout to aid with train detection, and to provide degraded working methods. The trains still use the balises to determine where they are, and to calculate braking curves.

"Passengers for New Lane, should be seated in the rear coach of the train " - Opinions are my own and not those of my employer
Last edited: 26/07/2016 at 18:02 by headshot119
Log in to reply
The following user said thank you: Steamer
ERTMS Operation 26/07/2016 at 18:25 #83802
Jan
Avatar
889 posts
The usual definition of Level 2 I've encountered is still based on classic train detection via track circuits and axle counters. If the interlocking starts relying on the train determining and transmitting its own position (and also guaranteeing that it hasn't lost a wagon or two along the way) you've already reached Level 3.
Two million people attempt to use Birmingham's magnificent rail network every year, with just over a million of them managing to get further than Smethwick.
Log in to reply
The following user said thank you: Steamer
ERTMS Operationt 26/07/2016 at 19:13 #83803
mallard1938
Avatar
1 posts
Level 2 in operation on the Cambrian

http://www.opentraintimes.com/maps/signalling/cambrian

Driver interface, so-called "heads-down" style driving. As everything is in front of you on the desk rather than on the lineside.

https://upload.wikimedia.org/wikipedia/commons/thumb/c/c5/ETCS_driver_machine_interface.svg/718px-ETCS_driver_machine_interface.svg.png

Level 2 in testing on Thameslink, the new class 700 trains have an automatic train operating (ATO) module fitted so they can drive themselves. Along with traffic management software, it will enable 24 trains per hour through the core.

http://www.opentraintimes.com/maps/signalling/tlk_core

ERTMS Level 3 is a way off, it does away with traditional train detection and used on train equipment and radio.

http://www.trl.co.uk/media/773946/cpr798_-_ertms_level_3_risks_and_benefits_to_uk_railways.pdf

Log in to reply
ERTMS Operationt 27/07/2016 at 11:49 #83808
Jersey_Mike
Avatar
250 posts
I'm still waiting for someone to show how this is more cost effective and/or more reliable than proven technology like coded track circuits. The people who thought wireless technology would be cheap and reliable are the children of those who said we'd all be wearing jetpacks.
Log in to reply
ERTMS Operation 27/07/2016 at 15:33 #83815
Finger
Avatar
220 posts
" said:
ERTMS Level 1 (With Infill) - Still an overlay over an existing signalling system, so you still have lineside signals. The movement authority is either transmitted to the train through a continuous loop placed along the track (similar to how the underground works on one of the ATO lines), through extra balises placed (Without infill the balises are placed only at signals; think cheap!), or through radio (aka GSMR).

Citation needed. I don't think sending MA over radio is possible without Level 2.

" said:
This allows the train to recieve a new movement authority sooner, and thus accelerate quicker if the new limit allows it to.

Not only that, it allows the train to run faster by advising the train of the new MA well in advance (that's how they use it in Slovakia).

" said:
ERTMS Level 2 - Lineside signals are a thing of the past in theory; however block markers are placed in various physical locations to give a method of degraded working

Block markers are not a method of degraded working, they are the main signals governing traffic in full supervision! Passing a block marker without a MA, even if the engine allows you to, is PSAD!

Also, ETCS L2 doesn't preclude lineside signals. They may be there eg. for unfitted trains, maybe giving them more adverse signals.

" said:
If I've given your train an MA from MH1000 to MH1008, and the train is just passing MH1000, and I shorten the authority down so you only have it as far as MH1006 the train will automatically recalculate a safe braking curve, and will do nothing if it's within that braking curve.

What you describe is an ACOA. While it is certainly possible, describing operation in terms of ACOAs and PSADs is severely misleading. Also, it might not be really successful, because the train may be in a radio hole.

" said:
As I understand it the spec doesn't require track circuits or axle counters for a level 2 route.

Yes they do! There must be some automatic vacancy detection and interlocking for ETCS L2 to work. OTOH, ETCS L1 could be quite easily deployed on a line without track circuits, just with signallers observing tail-lamps.

Log in to reply
ERTMS Operation 27/07/2016 at 15:37 #83816
taffy
Avatar
28 posts
Just to add to some of well explain responses.

Level 2 can also have overlay, in fact the initial design for the Great Western Main Line will have Level 2 with signaling to allow for mixed traffic operation.

Log in to reply
ERTMS Operation 27/07/2016 at 15:46 #83818
Finger
Avatar
220 posts
" said:
I'm still waiting for someone to show how this is more cost effective and/or more reliable than proven technology like coded track circuits. The people who thought wireless technology would be cheap and reliable are the children of those who said we'd all be wearing jetpacks.

While running into a real risk of replying to a flamebait, I have to remind you that while there are many successful ATPs based on other technologies (loops, balises, radio), I don't know of any ATP with speed supervision for mixed traffic mainline based on coded circuits. So "proven technology"? I'd say proven wrong.

Last edited: 27/07/2016 at 15:47 by Finger
Log in to reply
ERTMS Operation 27/07/2016 at 16:18 #83819
Jersey_Mike
Avatar
250 posts
" said:
" said:
I'm still waiting for someone to show how this is more cost effective and/or more reliable than proven technology like coded track circuits. The people who thought wireless technology would be cheap and reliable are the children of those who said we'd all be wearing jetpacks.

While running into a real risk of replying to a flamebait, I have to remind you that while there are many successful ATPs based on other technologies (loops, balises, radio), I don't know of any ATP with speed supervision for mixed traffic mainline based on coded circuits. So "proven technology"? I'd say proven wrong.
TVM-430 would be one, but I was specifically referring to the reliance of ETRMS and similar systems of wireless data links for continuous wayside to train communication.

Log in to reply
ERTMS Operation 27/07/2016 at 16:24 #83820
headshot119
Avatar
4869 posts
" said:
" said:
ERTMS Level 1 (With Infill) - Still an overlay over an existing signalling system, so you still have lineside signals. The movement authority is either transmitted to the train through a continuous loop placed along the track (similar to how the underground works on one of the ATO lines), through extra balises placed (Without infill the balises are placed only at signals; think cheap!), or through radio (aka GSMR).

Citation needed. I don't think sending MA over radio is possible without Level 2.
The ERTMS level 1 spec quotes Eurobalise + infill(euroloop, radio, or extra balises). I took that to mean the MAs can be sent via radiom but perhaps that isn't the case.


" said:
" said:
This allows the train to recieve a new movement authority sooner, and thus accelerate quicker if the new limit allows it to.

Not only that, it allows the train to run faster by advising the train of the new MA well in advance (that's how they use it in Slovakia).
Indeed and that's what I was trying to convey in the original post.

" said:
" said:
ERTMS Level 2 - Lineside signals are a thing of the past in theory; however block markers are placed in various physical locations to give a method of degraded working

Block markers are not a method of degraded working, they are the main signals governing traffic in full supervision! Passing a block marker without a MA, even if the engine allows you to, is PSAD!

Also, ETCS L2 doesn't preclude lineside signals. They may be there eg. for unfitted trains, maybe giving them more adverse signals.
The level 2 spec doesn't require physical signs to denote block markers as I understand it, but on the other hand it doesn't preclude it. In the Cambrian implementation they've placed block markers at strategic locations, but not at all locations where a block marker exists in the system, this was done specifically to allow degrading working if a train couldn't get a movement authority as I understand it.

" said:
" said:
If I've given your train an MA from MH1000 to MH1008, and the train is just passing MH1000, and I shorten the authority down so you only have it as far as MH1006 the train will automatically recalculate a safe braking curve, and will do nothing if it's within that braking curve.

What you describe is an ACOA. While it is certainly possible, describing operation in terms of ACOAs and PSADs is severely misleading. Also, it might not be really successful, because the train may be in a radio hole.
In the opening post I thought Gareth was asking specifically about what would happen if you revoked an MA, and what the train did when it passed over the balise. Rereading the OP I wonder if Gareth actually meant how was the MA revoked when the train had actually "used it up" for want of a better term.

" said:
" said:
As I understand it the spec doesn't require track circuits or axle counters for a level 2 route.

Yes they do! There must be some automatic vacancy detection and interlocking for ETCS L2 to work. OTOH, ETCS L1 could be quite easily deployed on a line without track circuits, just with signallers observing tail-lamps.
As I understand it Level 2 can be implemented using the eurobalise units as a pseudo axle counter units. Though it was decided this wasn't adequate for a UK implementation.


I don't claim to be an expert, and a lot of this is based off the research I did when I developed the Shrewsbury simulation, as well as talking to various staff who work with the system. What is also to important to remember is as a country we've not gone for the cheapest option to implement ERTMS, we've chosen to have the "optional extras" so to speak.

"Passengers for New Lane, should be seated in the rear coach of the train " - Opinions are my own and not those of my employer
Last edited: 27/07/2016 at 16:36 by headshot119
Log in to reply
ERTMS Operation 27/07/2016 at 23:01 #83823
Finger
Avatar
220 posts
Ah, apparently it is possible to announce the next signal aspects in ETCS L1 by radio (the next two aspects in specification 3.0.0). No wonder ETCS is so complex.

Quote:
The level 2 spec doesn't require physical signs to denote block markers as I understand it, but on the other hand it doesn't preclude it.

I believe it does require it, at least indirectly. How else would the driver know where to stop? Other than route knowledge or looking specifically for balises, he has to look for trackside markers. The DMI won't show him because of the inaccuracy of the odometry.

Of course, there can be trackside markers placed specifically to accommodate degraded modes, eg. to allow a SR->FS transition earlier in case of a failure.

To add something that may answer Gareth's question, this video might help, although it's ETCS L1 and in Polish. It shows quite well how does the train behave with ERTMS.

Log in to reply
ERTMS Operation 29/07/2016 at 19:11 #83845
Gwasanaethau
Avatar
509 posts
Thanks for all your replies! I didn't think I'd be opening such a can of worms when I started the thread! Sorry about that!

Incidentally, it was actually that RAIB report that got me interested in all of this.

Thanks again,

Mark (Gwasanaethau).

Log in to reply
ERTMS Operation 12/02/2017 at 03:50 #93049
Puro
Avatar
18 posts
Jersey_Mike in post 83819 said:
" said:
" said:
I'm still waiting for someone to show how this is more cost effective and/or more reliable than proven technology like coded track circuits. The people who thought wireless technology would be cheap and reliable are the children of those who said we'd all be wearing jetpacks.

While running into a real risk of replying to a flamebait, I have to remind you that while there are many successful ATPs based on other technologies (loops, balises, radio), I don't know of any ATP with speed supervision for mixed traffic mainline based on coded circuits. So "proven technology"? I'd say proven wrong.
TVM-430 would be one, but I was specifically referring to the reliance of ETRMS and similar systems of wireless data links for continuous wayside to train communication.

TVM430 is highly dependent of KVB for complex layouts

Log in to reply
ERTMS Operation 11/08/2017 at 12:10 #100975
Jriver
Avatar
49 posts
Hi Gareth

You can find a whole raft of technical information on the ERA's website (or whatever they are calling themselves this week).

Specifically, the SRS Subset 26 Chapters 3, 4, 5 and 6 are quite informative.

Bear in mind also that the Cambrian was based on baseline 2.3 - things have moved on significantly since then, we are now up to Baseline 3.6 (which apparently wont be tinkered with now 'for at least 10 years' so that suppliers can develop their products without fearing another update).

Log in to reply
ERTMS Operation 13/03/2018 at 02:52 #106721
jamieP
Avatar
6 posts
taffy in post 83816 said:
Just to add to some of well explain responses.

Level 2 can also have overlay, in fact the initial design for the Great Western Main Line will have Level 2 with signaling to allow for mixed traffic operation.
How does that work? Would drivers of trains running under level 2 just ignore the fixed signals?

Log in to reply
ERTMS Operation 13/03/2018 at 04:01 #106722
GeoffM
Avatar
6274 posts
jamieP in post 106721 said:
taffy in post 83816 said:
Just to add to some of well explain responses.

Level 2 can also have overlay, in fact the initial design for the Great Western Main Line will have Level 2 with signaling to allow for mixed traffic operation.
How does that work? Would drivers of trains running under level 2 just ignore the fixed signals?
Talking generally (not ETCS) any fixed signals would show a favourable aspect to the driver, or show a specific aspect meaning "cab signalling effective". Unfitted/faulty trains would generally be the only train permitted between fixed colour light signals which drastically reduces the number of trains per hour depending on section length. IIRC the Thameslink core has fixed signals at typical UK spacing for the line speed, plus intermediate marker boards for fitted trains.

SimSig Boss
Log in to reply
ERTMS Operation 13/03/2018 at 20:53 #106738
Muzer
Avatar
718 posts
Jersey_Mike in post 83819 said:
" said:
" said:
I'm still waiting for someone to show how this is more cost effective and/or more reliable than proven technology like coded track circuits. The people who thought wireless technology would be cheap and reliable are the children of those who said we'd all be wearing jetpacks.

While running into a real risk of replying to a flamebait, I have to remind you that while there are many successful ATPs based on other technologies (loops, balises, radio), I don't know of any ATP with speed supervision for mixed traffic mainline based on coded circuits. So "proven technology"? I'd say proven wrong.
TVM-430 would be one, but I was specifically referring to the reliance of ETRMS and similar systems of wireless data links for continuous wayside to train communication.
Do existing ERTMS installations actually rely on long-distance wireless communication beyond the induction loops used for short-distance communication between balises/loops and trains? I didn't think so, but I could be wrong.

Log in to reply
ERTMS Operation 13/03/2018 at 21:17 #106739
Jan
Avatar
889 posts
Level 2 does and there are already a number of routes using it.
Two million people attempt to use Birmingham's magnificent rail network every year, with just over a million of them managing to get further than Smethwick.
Log in to reply
ERTMS Operation 14/03/2018 at 07:11 #106748
Chromatix
Avatar
190 posts
Here in Finland, all passenger lines are fitted with an ATP system which appears to be equivalent to ERTMS Level 1 - much like BR-ATP was, back in the 1980s. It appears to work very well, and was sufficiently cost-effective to roll out in a very sparsely-populated country. I assume that train detection is still conventional, ie. based on track circuits, and that is backed up by the ability of the Helsinki signalman to detect and respond appropriately to an unbraked runaway rake some years ago, directing it into a clear platform with upgraded buffers compared to its original destination.

As far as I can tell, the move to cab-signalling allowed removing the numeric displays from the otherwise German-style lineside signals, which otherwise remain in place and are still (in some places such as Helsinki itself) controlled by 1970s equipment (which I assume is basically an NX panel). At Helsinki there are 19 platforms feeding a ten-track main line, of which two tracks lead directly to Pasila depot, and four each of the remainder lead to the western and northern main lines, and there are upwards of 50 movements across the station throat during peak hour. Until recently there was also a motor-rail terminal adjacent to the station, but this has now been moved to Pasila.

So the lineside signals are now capable of displaying eight indications:

Stop
Proceed for Shunting (from ground signals only)
Slow, Expect Stop
Slow, Expect Slow
Slow, Expect Clear
Clear, Expect Stop
Clear, Expect Slow
Clear, Expect Clear

(There are also individual Home and Distant heads, which can be used where signal spacings are very long, and simplified "block signals" which are like UK three-aspect signals; Yellow is treated as "Clear, Expect Stop".)

Previously, these could all be combined with numeric indicators indicating exactly what speed "Slow" meant, on both the immediate and distant heads. This proved to be error-prone, since the default 35kph limit for a signalled divergence was indicated by the *absence* of a numeral, and misreading such a signal led to a serious derailment with fatal consequences.

Now, if driving on lineside signals, "Clear" always means 80kph and "Slow" always means 35kph, unless the line speed itself is lower. Higher speeds are authorised only through cab signals, and authorities are communicated to trains through pairs of Eurobalises, which are provided both at and in rear of lineside signals and speed restrictions. The ATP system computes braking curves and enforces the authorised speed limits, whether imposed by permanent way or signal.

Drivers are now trained to drive on the cab signals, without even looking at the lineside signals unless the cab-signalling display indicates that they should. This became an issue in an incident where the balises had been taken offline due to engineering work, but the driver simply followed the 80kph default protection which ATP enforces in such cases. It was found that the cab display didn't indicate sufficiently clearly that lineside signals took precedence in that situation, and I assume that modifications have been made accordingly.

Log in to reply
The following user said thank you: nroberts
ERTMS Operation 11/04/2018 at 08:39 #107453
Giantray
Avatar
330 posts
GeoffM in post 106722 said:
jamieP in post 106721 said:
taffy in post 83816 said:
Just to add to some of well explain responses.

Level 2 can also have overlay, in fact the initial design for the Great Western Main Line will have Level 2 with signaling to allow for mixed traffic operation.
How does that work? Would drivers of trains running under level 2 just ignore the fixed signals?
Talking generally (not ETCS) any fixed signals would show a favourable aspect to the driver, or show a specific aspect meaning "cab signalling effective". Unfitted/faulty trains would generally be the only train permitted between fixed colour light signals which drastically reduces the number of trains per hour depending on section length. IIRC the Thameslink core has fixed signals at typical UK spacing for the line speed, plus intermediate marker boards for fitted trains.
Geoff is absolutely right about the Thameslink core. Through the Core there is level two also with convential signals. There are block markers between signals too. Unless things have changed, trains through the Core will either be in
ERTMS+ATO or using convential signalling. If someone could confirm this, the TOC is refusing to allow/train its drivers to drive under ERTMS.

A trains under ATO will run much closer together, you will notice a difference in the speed at which a train in ATO approaches a platform. The technical side of it loses me, but Class 700 units communicate with the ERTMS system using GSM-R. Anyone who has travelled on a Class 700 will notice a loading indicator in the coaches for passenger use. A clever piece of technology which is a by-product for the passenger. The loading is measured by weight, which in turn is required as information in calculating acceleration and braking rates when a train is in ATO.

Under ERTMS/ATO, signals and block markers indicate sections, by providng more sections enables more trains to run closer together. Because conventional signalling is running alongside, drivers ignore those signals when in ERTMS/ATO. As there are block markers between conventional signals it is entirely possible for a conventional signal to show a yellow aspect with a train in the "conventional" signal section ahead if it is past the next blcok marker However if a following train is not in ERTMS/ATO mode, the system will recognise this and the signal will be at Danger.

As far as the signaller is concerned, they just carry on signalling trains as normal, but also have an additional bit of kit that monitors trains to let the signaller know the state of trains, whether in Convential, ERTMS or ERTMS+ATO mode. Obviously Rules and Regs are slightly different, but follow the basic rules of signalling.

Am I right in saying there are no physical overlaps since the system just alters the braking curve of a train to ensure it will never fail the path of another train?

Retired Professional Railwayman (1981-2023); Pway & S&T (1981-88); Former Signalman/Signaller/ Signalling Trainer (1989-2023) [AB, TCB, Mechanical, NX, WestCad, Hitachi SARS]; Railway Historian (esp.SER, LCDR); Member of The Permanent Way Institution..
Log in to reply
The following user said thank you: BarryM
ERTMS Operation 11/04/2018 at 10:03 #107455
Jan
Avatar
889 posts
Giantray in post 107453 said:
Am I right in saying there are no physical overlaps since the system just alters the braking curve of a train to ensure it will never fail the path of another train?

Generally speaking overlaps are beneficial even with ETCS, as the emergency braking curve will be calculated to the end of the overlap. So having an overlap available means you can potentially start braking later.

Two million people attempt to use Birmingham's magnificent rail network every year, with just over a million of them managing to get further than Smethwick.
Log in to reply