E.R.T.M.S.

[DY444]

2 hours ago, Edwin_m said:

Two problems with the Heathrow branch.  Firstly the 332 units didn't have TPWS (as they were only ever operated on GW-ATP routes) and secondly it was deemed that GW-ATP could only be replaced by something with similar levels of safety.  

 

I don't know why they need an interim TPWS stage for Moorgate, but at least TPWS is designed to be interfaced into existing signalling with minimum disruption.  

 

The 332 issue seems to me to be irrelevant.  They didn't have TPWS or ETCS so had to go either way. 

 

Deeming "that GW-ATP could only be replaced by something with similar levels of safety" may be the official explanation but doesn't make much sense to me given that the 387s weren't going to be using it on the main line as far as Airport Junction. 

 

I don't have a problem with what they did at Heathrow in going straight to ETCS but Moorgate going via TPWS just looks like money being wasted on the face of it.  They know ETCS works.  They know the 707 works on ETCS so just get on with it.  A single traction type on a short branch with the ability to turn back at Drayton Park if the computer says no.  If they don't have any confidence in that being relatively straight forward without two steps of intermediate belt and braces then things really are bad.

Edited January 24, 2022 by DY444

[phil-b259]

On 24/01/2022 at 09:45, DY444 said:

 

The 332 issue seems to me to be irrelevant.  They didn't have TPWS or ETCS so had to go either way. 

 

Deeming "that GW-ATP could only be replaced by something with similar levels of safety" may be the official explanation but doesn't make much sense to me given that the 387s weren't going to be using it on the main line as far as Airport Junction. 

 

I don't have a problem with what they did at Heathrow in going straight to ETCS but Moorgate going via TPWS just looks like money being wasted on the face of it.  They know ETCS works.  They know the 707 works on ETCS so just get on with it.  A single traction type on a short branch with the ability to turn back at Drayton Park if the computer says no.  If they don't have any confidence in that being relatively straight forward without two steps of intermediate belt and braces then things really are bad.

 

 

The Moorgate branch previously (when the 313s were in use) used LU style tripcocks and as such it lacked TPWS

 

I believe TPWS was fitted to replace said Tripcocks because (1) its an obsolete technology with moving parts and as such is far more expensive / difficult to maintain than an electronic system like TPWS and (2) the new 717 Siemens units do not have the ability to have tripcocks fitted - but TPWS is mandatory for all units on the national rail network so represents no extra cost.

 

As for ECTS - please remember that this is a COMPUTER BASED system that is still, in railway signalling terms, a very new technology - and as with all such computing based systems the technology is constantly evolving. What you don't want to do is rush in and apply whats currently available 'off the shelf' (bearing in mind the 40 plus life expectancy of signalling systems - just ask yourself who is still running WIndows 3.1 at home these days) only to find that the system which is fitted to the ECML mainline proper is different thus requiring expensive re-working of the Moorgate installation.

 

So the choice with the Moorgate branch was effectively keep the ageing and unreliable 313s on for a further 4-5 years (assuming HM Treasury is still willing to fund ECTS rollout) till the ECML ECTS design / specifications were finalised or get new trains thus bringing big improvements to passengers but fit TPWS (which is actually pretty cheap due it it being a stable and extensively used bit of kit) to signals on the branch as an interim measure.

Edited May 15, 2022 by phil-b259

[ess1uk]

12 hours ago, jim.snowdon said:

The latest, ie most recent, stages have been about preparing the line for ETCS, which is still some time away, and replacing all the trainstops with TPWS. That change goes live in May 2022. Early in 2023 the line should go over to ETCS overlaid on the existing fixed signalling, before eventually going over to full ETCS with no fixed signals.

 

 

i heard

Commencing in mid-April 2022 there will be a 6-month proving period of the ETCS system. At this stage, there will be co-located ETCS block marker boards and colour lamp signals to cater for both ETCS Driver Training and legacy train control operations.

There will be a one-year period of driver training, which is expected to conclude in October 2023.

In October 2023 the colour lamp signals will be de-commissioned ready for a programme of recoveries, at which point the ETCS block markerboards will permanently replace the colour lamp signals.

[jim.snowdon]

13 hours ago, DY444 said:

How bizarre.  If NR had gone GW ATP -> TPWS -> ETCS (eventually) on the Heathrow branch it would have been so much easier initially (no need for 387s to have ETCS from the get go, 7 car 345s could go to Heathrow, de-risk Crossrail completion due to one less signalling system to get working initially etc etc).  But no they went GW ATP -> ECTS.  Yet on Moorgate they are going to the trouble and, importantly, cost of going Train Stops ->  TPWS -> ETCS eventually despite 717s having been successfully tested on ETCS.

The critical element is that, for whatever reason, it has been deemed necessary to remove the trainstops, along with the trainstop tester and its associated part in the Finsbury Park interlocking, before the line can be equipped with operational ETCS. That, at least for the interim, requires the installation of an equivalent system, ie TPWS. At least, when it becomes redundant, the TPWS equipment can be recycled into other places of the network. That said, as and when (and if) the whole of the southern end of the ECML goes over to ETCS, there will potentially be an awful lot of redundant equipment.

 

It does, of course, remain to be seen if large scale implementation of ETCS actually brings any worthwhile increases in network capacity, given the thoroughly mixed traffic nature of the railway.

 

[w124bob]

I was informed before retirement 5 years ago that ERTMS had little to do with real capacity increases (a bonus) but was really all about cost savings. I came into contact a few times a year with someone who was directly responsible for writing the new rules, although they worked for a major TOC this was a sideline as a consultant to NR. It was clear back then that the costs in training for TOCS was going to be considerable, only a portion was going to be subsidised by NR. Their prime motivation was the ultimate removal of expensive to maintain lineside signalling equipment and the creation of what was termed "a sterile trackside" environment. The first part was the roll out and reliability of GSMR (cost savings to TOCS, increased DOO full passenger operation)with full onboard train to signaller communication. I remember being told that, as a depot, we would encounter this within five years on the Reading/Didcot corridor. One of the reasons the new signalling remained 3 aspect in places? 

The sterile trackside environment was talked about both in terms of risk reduction but also an acknowledgement that drivers would be concentrated on a screen and the observation and control of matching actual speed to required speed. Particularly with regard to signal markers requiring a stop or major speed reduction, a complete change of in the way we drove. The capacity increase was deemed to come from the ability for the computers to squeeze up rolling blocks by speed matching of trains, easy to do on TGV style routes, but on mixed traffic routes, who knows?

Edited January 25, 2022 by w124bob

[The Stationmaster]

Reading - Didcot remained 3 aspect for the very simple reason that it met the required headway spec although I beliebve (and hope). that the underbraked signal spacing for 125 mph near Goring has been corrected.

 

As far as GSMR is concerned all teh Class 373 Eurostar c variants were equipped with receivers and printer ports to handle GSMR printed messages one specific reason being that the Driver would always receive a printed message as authority to pass a signal at danger in the UK.  During the late 1990s Eurostar were repeatedly told that the necessary GSM  upgrade (=GSMR capability) would be carried out 'within a couple of years's  - same message from Railtrack every year.

 

Drivers, including British Drivers working Eurostars since it began, have been driving in accordanace with Instructions placed in the cab for many years - I even have a photo of the method in the cab of a Polish tank engine which I took back in the 1970s.  There is simple a book on the desk in front of the Driver who has to turnover the pages as the train goes on its way.  Using screens instead of pieces of paper is in reality an updating of that method plus it saves masses of clerical time and printing of new pages for daily variations such as temporary speed restrictions where there is not a speed related cab signalling system.

 

This system allows for massive savings in road learning costs.  For example when it was planned to run a Eurostar service for skiers to Bourg St Maurice  SNCF decided that to avoid Class 373 traction training costs it would be worked throughout from Lille by a Lille Driver - they were given three days to learn the road (386 miles although they already knew the LGV part of it).  No problem at all using the printed livre ligne and turning over the pages as they went - and no continuous cab signalling once they were off the LGV.

 

And despite having to look down at the livre ligne the  SPAD rate on SNCF is miniscule although that is helped by the manner in which they carry out defensive driving. (Which wouldn't, and doesn't, work too well in Britain with a Driver putting in the brake the instant he catches sight of a double yellow - so braking well before braking distance.)

[ess1uk]

8 hours ago, The Stationmaster said:

As far as GSMR is concerned all the Class 373 Eurostar c variants were equipped with receivers and printer ports to handle GSMR printed messages one specific reason being that the Driver would always receive a printed message as authority to pass a signal at danger in the UK.  During the late 1990s Eurostar were repeatedly told that the necessary GSM  upgrade (=GSMR capability) would be carried out 'within a couple of years' - same message from Railtrack every year.

that would imply that data could be transmitted and not just voice.

that would mean that GPRS would have been installed.

in reality, GPRS is only just being installed on the NR telecom network

[The Stationmaster]

44 minutes ago, ess1uk said:

that would imply that data could be transmitted and not just voice.

that would mean that GPRS would have been installed.

in reality, GPRS is only just being installed on the NR telecom network

I don't know what was going to be installed but it was definitely intended to transmit data messages which is why the trains were arranged to have the necessary receiving equipment (I don't know if all of it was actually fitted but they had a. very sophisticated radio system in any case and the agreed message formats definitely existed although they were destined never to be used).

[The Stationmaster]

48 minutes ago, ess1uk said:

that would imply that data could be transmitted and not just voice.

that would mean that GPRS would have been installed.

in reality, GPRS is only just being installed on the NR telecom network

I don't know what was going to be installed but it was definitely intended to transmit data messages which is why the trains were arranged to have the necessary receiving equipment (I don't know if all of it was actually fitted but they had a. very sophisticated radio system in any case and the agreed message formats definitely existed although they were destined never to be used).

[icn]

On 25/01/2022 at 15:40, w124bob said:

The capacity increase was deemed to come from the ability for the computers to squeeze up rolling blocks by speed matching of trains, easy to do on TGV style routes, but on mixed traffic routes, who knows?

Not just. Part of the purported improvements - which start with L2 - come from getting rid of signal sighting issues. And some systems are able to optimise traffic with speed advice (used on the new Lotschberg tunnel). Apparently the SBB has seen capacity improvements of 25% on mixed traffic routes - if this document is to be believed (I was unable to find the original source): https://www.ertms.net/wp-content/uploads/2021/06/10.-Increasing-infrastructure-capacity.pdf . (I don't believe the original test route actually had much freight, but it does have a mix of S-Bahns stopping everywhere, IC/IR's stopping nowhere, and RE's stopping every couple of stops - all on 2 tracks.)

 

 

  

On 24/01/2022 at 11:21, phil-b259 said:

As for ECTS - please remember that this is a COMPUTER BASED system that is still, in railway signalling terms, a very new technology - and as with all such computing based systems the technology is constantly evolving. What you don't want to do is rush in and apply whats currently available 'off the shelf' (bearing in mind the 40 plus life expectancy of signalling systems - just ask yourself who is still running WIndows 3.1 at home these days) only to find that the system which is fitted to the ECML mainline proper is different thus requiring expensive re-working of the Moorgate installation.

Why should or would the Moorgate installation need reworking? ETCS is a standard, and as long as the trains support the right standard they'd be able to run on any ETCS system (yes, they need some testing to be safe, and there's the question of baselines). Of course, there's the question of the transition between signalling regions... but that problem already exists.

 

The various ETCS routes in Switzerland use equipment from different manufacturers, and nevertheless the same trains have no issues using those routes. Even better, the ICE 4 (and perhaps other trains - Railjet for example) use ETCS in multiple countries. Given the UK's minimal experience with ETCS, waiting some time might be sensible - but system incompatibility can't be the issue given that it's been solved years ago.

[Edwin_m]

On four-track sections such as the ECML, colour light signals on the fast and slow lines are placed alongside each other to reduce the risk of drivers "reading across" to a signal intended for the other tracks.  This means that the same spacing has to be used for both, which may not be optimal for capacity.  By removing the signals, ETCS Level 2 also removes that constraint and this is a additional, fairly small, capacity benefit.  

 

In fact neither ERTMS Level nor the future Level 3 offers a huge capacity benefit, because on a real railway the achievable capacity depends on what happens at stations, and ERTMS doesn't make much difference to that.   

[DY444]

14 hours ago, Edwin_m said:

On four-track sections such as the ECML, colour light signals on the fast and slow lines are placed alongside each other to reduce the risk of drivers "reading across" to a signal intended for the other tracks.  This means that the same spacing has to be used for both, which may not be optimal for capacity.  By removing the signals, ETCS Level 2 also removes that constraint and this is a additional, fairly small, capacity benefit.  

 

In fact neither ERTMS Level nor the future Level 3 offers a huge capacity benefit, because on a real railway the achievable capacity depends on what happens at stations, and ERTMS doesn't make much difference to that.   

 

That signal pairing may be largely universal now but it wasn't always the case.  The WR didn't do it in its 1960s signalling schemes particularly between Paddington and Reading.  On that section each line type was effectively signalled separately with a shorter signal spacing on the reliefs in some places.  Some of that survived until the electrification/modernisation (eg on the up relief approaching Ruscombe and approaching Slough).  The GEML had a similar arrangement with (lots) more signals on the Up and Dn Electric.   Yes the lines were paired by use rather than by direction as on the ECML but reliance was placed on route knowledge to know which line a given signal applied to.  

 

Edited February 6, 2022 by DY444

[The Stationmaster]

On 06/02/2022 at 08:23, DY444 said:

 

That signal pairing may be largely universal now but it wasn't always the case.  The WR didn't do it in its 1960s signalling schemes particularly between Paddington and Reading.  On that section each line type was effectively signalled separately with a shorter signal spacing on the reliefs in some places.  Some of that survived until the electrification/modernisation (eg on the up relief approaching Ruscombe and approaching Slough).  The GEML had a similar arrangement with (lots) more signals on the Up and Dn Electric.   Yes the lines were paired by use rather than by direction as on the ECML but reliance was placed on route knowledge to know which line a given signal applied to.  

 

Not strictly correct on the WR.  Generally the signals for the Mains and Reliefs were opposite each other e.g Sonning Sidings - Twyford West (aka Lands End running junction) and in places they were on the same gantries (e.g. Ruscombe -Maidenhead West).  But in the vicinity of some stations east of Reading the Relief Line signals closed up with additional signals such one in the Twyford station - Ruscombe section, which you mentioned, on the Up Relief.  The only difference at Slough in the original resignalling was an additional signal on the Down Relief approaching the station otherwise all signals were opposite each other or as near opposite as siting/sighting allowed.    There was also an additional signal at Sonning Sidings on the Down Relief for access to Kennet Bridge loop otherwise all Main and relief line signals between Kennet bridge and Twyford were originally opposite each other but later some on the Main were removed.

 

But over the years there were changes and alterations - in some cases to improve sighting (such as UM 31 at Twyford being moved back west of the station which led to another signal on the Up Main being removed so the Up Direction signals were no longer opposite each.

[St. Simon]

Hi,

 

Just to add some further information which I don't believe has been touched on. There's a couple of additional, although quite small, capacity increases that can be unlocked by using E.T.C.S. Level 2 even in an overlay other than the whole signal spacing constraints that are usually the primary benefit of E.T.C.S. 

 

These have to do with the way that E.T.C.S. interfaces with the interlocking. First of all is the elimination of approach control / release for trains that are running under full supervision, as E.T.C.S. provides Automatic Train Protection, there's no need to slow the train on approach to a junction just to ensure it is under control. As MAR (and MAY) controls can lead to excessive slowing for divergent trains, suppressing the controls for trains that are under full supervision (which the Radio Block Centre can tell the interlocking), then a few seconds can be clawed back. This is a feature that we've just implemented on a lot of MAR routes between Paddington and Heathrow.

 

The other feature is that of release of Overlap Locking. As with approach controls, where a train is under full supervision and can prove via the RBC that it is at a stand, the interlocking can release the overlap locking much quicker than in a non-E.T.C.S. world, allowing routes to be set early, again increasing capacity (particularly useful at somewhere like Reading). However, I'm not sure whether this is implemented anywhere (primarily being a Schemes Designer, I rarely see interlocking data design).

 

There are lots of other benefits other than these that are rarely given as advantages for E.T.C.S, but outside of the scope of discussions on unlocking capacity.

 

I know that a lot of people have said that E.T.C.S. does not bring huge capacity increase for mixed traffic lines, but it does provide lots of little increases in capacity, which might just be enough. I'm a huge advocate in E.T.C.S. as I see it has the potential to be very powerful, although I will admit that it will take time and some false starts to get there.

 

Simon

[DY444]

14 hours ago, The Stationmaster said:

Not strictly correct on the WR.  Generally the signals for the Mains and Reliefs were opposite each other e.g Sonning Sidings - Twyford West (aka Lands End running junction) and in places they were on the same gantries (e.g. Ruscombe -Maidenhead West).  But in the vicinity of some stations east of Reading the Relief Line signals closed up with additional signals such one in the Twyford station - Ruscombe section, which you mentioned, on the Up Relief.  The only difference at Slough in the original resignalling was an additional signal on the Down Relief approaching the station otherwise all signals were opposite each other or as near opposite as siting/sighting allowed.    There was also an additional signal at Sonning Sidings on the Down Relief for access to Kennet Bridge loop otherwise all Main and relief line signals between Kennet bridge and Twyford were originally opposite each other but later some on the Main were removed.

 

But over the years there were changes and alterations - in some cases to improve sighting (such as UM 31 at Twyford being moved back west of the station which led to another signal on the Up Main being removed so the Up Direction signals were no longer opposite each.

 

Re Slough West on the up.  I agree that originally there was one signal on each of the UR and UM and these were paired.  However this was changed subsequently when the layout was remodelled and it ended up with one signal on the UM (S19) and three on the UR (S117, S119 and S121).  I was thinking specifically of S121 when I made the comment.

[The Stationmaster]

4 hours ago, DY444 said:

 

Re Slough West on the up.  I agree that originally there was one signal on each of the UR and UM and these were paired.  However this was changed subsequently when the layout was remodelled and it ended up with one signal on the UM (S19) and three on the UR (S117, S119 and S121).  I was thinking specifically of S121 when I made the comment.

The Up Main signal spacing approaching Slough was something of a pig after signals were taken out.  If you got a double yellow on the auto at Burnham (UM20A?) when running at 90mph all you needed to do was shut the controller and watch very carefully for the next one just round the curve (UM 20B) because it  was effectively the real braking point on sighting it.  Taking out S21 unevenly increased the length of a signal section which meant that you would get double yellows where on the original spacings there would have been greens - but underbraked for 125 mph of course.    

 

Overall the changes made over the years west of Slough were really something of a dog's breakfast but for very good reasons, rather like the changes on the Up Main at Twyford. where UM 31 was moved back in rear of the station in order to improve sighting (thank you Jack Hancock - the Divisional Traction Supt at that time who had the signal moved) which increased the distance between it and the protecting signal for the running junction at Ruscombe and by taking out another signal in consequence also increased the distance from the next signal in rear at Lands End/Twyford West.  this created another situation where trains running at a consistent distant apart on greens found themselves encountering a double because of the longer signal sections.  Yet again - when they arrived - it was ideal for HSTs but not so clever for something running at 90mph.

 

[ess1uk]

On 22/05/2021 at 09:33, St. Simon said:

 

For starters, NO units are fitted with ERTMS, they are fitted with the on-board elements of ETCS. 

 

Off the top of my head:

 

• Class 158 - TfW Cambrian Units only
• Class 195
• Class 196
• Class 221**
• Class 313 - Okay, only one unit
• Class 331
• Class 345
• Class 374 - I presume they have ETCS fitted and use it via a STM?
• Class 390**
• Class 397
• Class 387/1A
• Class 700
• Class 701*
• Class 707
• Class 710
• Class 717
• Class 720 
• Class 730
• Class 745
• Class 755
• Class 800*
• Class 801*
• Class 802*

 

The ones in bold are units that have it fitted and are using it in an ETCS Level. The ones with * are ones that are using the Packet 44 element of the ETCS Data for ASDO / APCO / ISDD / ABDO systems. The ones with ** have an EVC to read the Packet 44 TASS Data for Tilting, but don't have the full ETCS DMI.

 

The rest have it fitted (I believe as they are all new build units, although I'm happy to corrected) and the ETCS DMI is used to provide the speedometer etc (expect for the GWR Class 80x's as they use the ATP Speedometer).

 

Simon 

And a 180 being used at Old Dalby RIDC

[ess1uk]

Class 43 HST power cars being fitted at Old Dalby 

[St. Simon]

On 08/05/2022 at 22:05, ess1uk said:

Class 43 HST power cars being fitted at Old Dalby 

Hi,

 

Not just fitted now, today saw one of the Network Rail’s Class 43s operate at 75mph in ETCS Level 2 at RIDC Melton. I

 

The Cab does look a little odd with the Thales screen, but I think it is a great development.

 

Simon

[big jim]

12 hours ago, St. Simon said:

Hi,

 

Not just fitted now, today saw one of the Network Rail’s Class 43s operate at 75mph in ETCS Level 2 at RIDC Melton. I

 

The Cab does look a little odd with the Thales screen, but I think it is a great development.

 

Simon

might get to see a HST down the Cambrian soon then?

 

I was looking in the cab of a 197 the other week in crewe and they have the screen in, I was curious as to how it would look compared to a 196, it fits in quite well but I suppose with a new unit it can be specced at the planning stage to fit things like that rather than retro fit

[corneliuslundie]

Don't forget the four class 97s used at Shrewsbury by NR - and for excursions and apparently now for freight too (logs starting next week).

Are any other locos fitted? If not the 97s will be a bit stretched at times.

Jonathan

[iands]

1 hour ago, big jim said:

might get to see a HST down the Cambrian soon then?

 

If I understand correctly Jim, the current fitment/versions of ETCS are not compatible with the Cambrian system, and vice versa, so probably not.

[big jim]

38 minutes ago, iands said:

If I understand correctly Jim, the current fitment/versions of ETCS are not compatible with the Cambrian system, and vice versa, so probably not.

 

be interesting to see what they will do with the 197s then, I’d assume they have the current ‘Cambrian’ system fitted to the units now (or will be fitted when the dedicated units are built) but they would also need the current version fitting for when it gets rolled out nationwide unless they rip the Cambrian system out and replace it with new 

 

personally I’d sooner see the back of ertms/ects as i feel it’s a backward step when it comes to flexibility especially on the freight side with things like run rounds, coupling, possessions, top and tail working etc unless the new system is a complete rewrite of what I already know, I’m going to take a lot of convincing to change my opinion of what I think is an awful set up!

 

 

[icn]

5 minutes ago, big jim said:

 

personally I’d sooner see the back of ertms/ects as i feel it’s a backward step when it comes to flexibility especially on the freight side with things like run rounds, coupling, possessions, top and tail working etc unless the new system is a complete rewrite of what I already know, I’m going to take a lot of convincing to change my opinion of what I think is an awful set up!

 

 

Run rounds, coupling, and top and tail are all happening regularly on the ETCS systems in Switzerland. Not sure about how possessions are handled. The convenience probably depends on both the signalling software and the interface in the cab as opposed to ETCS itself?

 

I'm curious as to why the UK has incompatibility issues mind you, Switzerland's installations partly predate the Cambrian one, and there at trains from 3-4 countries and yet more manufacturers running on the oldest regular installation. My brief research suggests the Cambrian runs baseline 2.30d, whereas CAF is building new systems with baseline 3 (although that doesn't mean the 197s will be baseline 3) - but in theory the trains are backward compatible and should be installed with the most recent version available. In Switzerland there are plenty of trains with both baselines, the pure ETCS lines are using the older baseline but the rest of the network has also received an overlay with the newer baseline allowing for ETCS-only trains if they have baseline 3 (trains with baseline 2 need to keep their conventional safety equipment).

[big jim]

Regards run rounds etc it’s not so much they don’t happen here, just not very often given how often freight goes down the Cambrian but it’s time consuming having to input your details with every change end, you can’t set back more than 2m, even just moving a loco within a yard requires you to be in shunt mode etc 

 

I’m hoping Simon will come on and say that the various systems are in fact compatible, if not it will add to my dislike of the system as the whole ‘selling point’ was that it was a standard Europe wide system that could be fitted to anything