ERTMS - let's get on with it

[DerekC]

As a newbie I am a bit nervous about this, but here goes...

ERTMS has been around since the late 1990s. So far all we in GB have done is fit Cambrian. Although there are lots of strategies and plans (and even more discarded old ones) we haven't achieved anything else. ERTMS Level 2 (plus ATO) on Thameslink now looks irreversibly committed and likely to deliver in the next 3/4 years, which is great. GWML is still havering - something has to be done to run Crossrail trains to Heathrow, but IEP is dual fitted (BR-ATP and ERTMS) so the pressure for the rest of the route is uncertain. ERTMS Level 2 on ECML keeps getting launched -- and delayed. ERTMS Level 3 would be great, but doesn't exist yet. There is a danger that it all get put off - nothing pleases governments more than a bit of strategic procrastination. But ERTMS Level 2 (as planned for GWML and ECML - and Thameslink) has cost and capacity benefits, so let's get on with it.

 

[Tio Terry]

As a newbie I am a bit nervous about this, but here goes...

ERTMS has been around since the late 1990s. So far all we in GB have done is fit Cambrian. Although there are lots of strategies and plans (and even more discarded old ones) we haven't achieved anything else. ERTMS Level 2 (plus ATO) on Thameslink now looks irreversibly committed and likely to deliver in the next 3/4 years, which is great. GWML is still havering - something has to be done to run Crossrail trains to Heathrow, but IEP is dual fitted (BR-ATP and ERTMS) so the pressure for the rest of the route is uncertain. ERTMS Level 2 on ECML keeps getting launched -- and delayed. ERTMS Level 3 would be great, but doesn't exist yet. There is a danger that it all get put off - nothing pleases governments more than a bit of strategic procrastination. But ERTMS Level 2 (as planned for GWML and ECML - and Thameslink) has cost and capacity benefits, so let's get on with it.

ERTMS 2 is planned to be installed between Airport Junction and Heathrow next year.

 

[The Planner]

Still want to see proof of capacity benefits, level 2 is still effectively block sections with marker boards, so unless you can make them all shorter than an equivalent sighting distance, green to red and overlap section then I don't see where anything substantial is coming from headway wise. The Cambrian doesn't offer anything capacity wise from RETB as the split sections can't be utilised properly at places such as Westbury and on the approaches to Aberystwyth.

 

[scott118]

isn't there to be a trial of this, in East Anglia ?

 

[jopsuk]

Widespread ERTMS has more potential application on the continent, removing one of the many complications with cross border services by eventually bringing everyone to the same system- for the same reason it's likely that more and more OHL systems will convert 25kV 50Hz AC.

 

[Gareth Marston]

Still want to see proof of capacity benefits, level 2 is still effectively block sections with marker boards, so unless you can make them all shorter than an equivalent sighting distance, green to red and overlap section then I don't see where anything substantial is coming from headway wise. The Cambrian doesn't offer anything capacity wise from RETB as the split sections can't be utilised properly at places such as Westbury and on the approaches to Aberystwyth.

There's no or very insignificant gains on the Cambrian the same will be true of any single track with passing places line the capacity is still governed by the location and quantity of crossing places. On the approaches to London termini I believe that most trains are almost within sight of each other especially south of the Thames and then the overall capacity is governed by termini platform capacity. ETCS will not provide any magic capacity bullet as in too many locations the physical limits are already reached. You might be able to whizz along certain sections with better capacity than now but they'll be a constraint not too far down the track.

 

[Bald Rick]

ETCS is working in the TL core, testing going well I understand.

The capacity benefits of Level 2 ETCS are, indeed, rather limited. However when you then add in Automatic Train Operation (which is a simple overlay to ETCS) then you do get capacity benefit. This is because you can tighten all the margins and head ways as each train will take exactly the same time in clearing junctions / section times etc.

 

[Gareth Marston]

ETCS is working in the TL core, testing going well I understand.

The capacity benefits of Level 2 ETCS are, indeed, rather limited. However when you then add in Automatic Train Operation (which is a simple overlay to ETCS) then you do get capacity benefit. This is because you can tighten all the margins and head ways as each train will take exactly the same time in clearing junctions / section times etc.

Where talking about margins maybe 5 to 10 % in some locations and those termini platforms are still looming up ahead except for Thameslink and Crossrail.

 

[The Planner]

Agree with ATO but apart from intensive core sections where everything needs to do the same thing even that is a bit limited at the mo.

 

[306024]

isn't there to be a trial of this, in East Anglia ?

The East Suffolk Line was mooted for ERTMS trials some years back as a replacement for RETB, but the Cambrian went ahead, and in the end the novel idea of signals on posts replaced RETB on the East Suffolk.

 

[WatcherZero]

70% of currently planned signalling renewals are ERTMS.

 

[The Planner]

Really? Starting from what date?

 

[asylumxl]

70% of currently planned signalling renewals are ERTMS.

ERTMS or with ERTMS compatibility?

 

[SpacePhoenix]

Which routes in SWT territory would see the best improvements/benefits from being switched over to ERTMS?

 

[LNW-GW Joint]

ERTMS/ETCS had legs when it was seen as the only way to get ATP on to the wider network, prompted by the serious accidents around 2000.
Once TPWS (and TPWS+) eliminated most of the accident risk, the business case was not as great.
We also seem very poor at just getting on with it, the cost of fitting rolling stock being seen as prohibitive.
The test section is near Stevenage on the Hertford Loop, where different manufacturers' kit is being proved for use on the wider network.
The GWML case is all about replacing the current obsolete ATP.
The ECML case is about getting 140mph running and eliminating "lights on sticks".
The roll-out is subject to Peter Hendy's review. Given the shortage of signal engineers it would be no surprise if the programme slipped to the right.

 

[Philip Phlopp]

ERTMS/ETCS had legs when it was seen as the only way to get ATP on to the wider network, prompted by the serious accidents around 2000.
Once TPWS (and TPWS+) eliminated most of the accident risk, the business case was not as great.
We also seem very poor at just getting on with it, the cost of fitting rolling stock being seen as prohibitive.
The test section is near Stevenage on the Hertford Loop, where different manufacturers' kit is being proved for use on the wider network.
The GWML case is all about replacing the current obsolete ATP.
The ECML case is about getting 140mph running and eliminating "lights on sticks".
The roll-out is subject to Peter Hendy's review. Given the shortage of signal engineers it would be no surprise if the programme slipped to the right.

There's also the bi-di argument on both ECML and GWML - less use at the moment on the ECML for reasons you all already know, but more bi-di signalling on the GWML will be useful with electrification.

 

[Phil.]

Still want to see proof of capacity benefits, level 2 is still effectively block sections with marker boards, so unless you can make them all shorter than an equivalent sighting distance, green to red and overlap section then I don't see where anything substantial is coming from headway wise. The Cambrian doesn't offer anything capacity wise from RETB as the split sections can't be utilised properly at places such as Westbury and on the approaches to Aberystwyth.

I believe that ERTMS is a European driven thing so it has to be done. There are benefits but I don't think that they truly outweigh the costs.

 

[jopsuk]

I believe that ERTMS is a European driven thing so it has to be done. There are benefits but I don't think that they truly outweigh the costs.

I seriously doubt that ERTMS is required to be installed to the main GB network by European directive. Interoperability requirements as I understand them apply to dedicated High Speed lines and access to associated stations for International Open Access. Due to the unique nature of the main GB Network and complications caused by our border control requirements through running from the European High Speed network is unlikely. I would think that if Network Rail were subject to such a directive, they would be able to apply for a derogation if they wished and would almost certainly be awarded it.

Network Rail have a bee in their bonnet about delivering a "digital railway" of which ERTMS is a part.

 

[Philip Phlopp]

I seriously doubt that ERTMS is required to be installed to the main GB network by European directive. Interoperability requirements as I understand them apply to dedicated High Speed lines and access to associated stations for International Open Access. Due to the unique nature of the main GB Network and complications caused by our border control requirements through running from the European High Speed network is unlikely. I would think that if Network Rail were subject to such a directive, they would be able to apply for a derogation if they wished and would almost certainly be awarded it.

Network Rail have a bee in their bonnet about delivering a "digital railway" of which ERTMS is a part.

I'm convinced someone at Network Rail HQ has been sold magic beans and will shortly begin a problem of work to build a beanstalk too.

The digital railway, with a small number of exceptions, is a fairy story concocted by suppliers to sell more tat that won't work properly, which won't remotely begin to do what it promises to do, and which will cause so many headaches in service nobody will notice it's a catastrophic waste of money.

ERTMS isn't looking as brilliant as it once did either - we're doing all this testing on the Hertford Loop because we can't trust suppliers to follow the specification and make things which work correctly.

 

[NSEFAN]

It's always the way. The spec is good but the implementation is of variable quality and reliability.

 

[TheNewNo2]

Just on a tangent - how should one pronounce ERTMS? Is it "Artemis", E-R-T-M-S, "ur-tem-ess" or something else? There's a lot to be said for acronyms like TBTC or xkcd...

 

[craigybagel]

I'm not sure what the official pronunciation is but throughout ATW it's known as ur-tims.

 

[MarkyT]

Personally I like the Swiss approach. Despite not being an EU member, their railway administration has spearheaded the ERTMS effort, being involved at the heart of developing the specs and rolling out early pilot schemes in Switzerland itself. This is clearly strategically very important to them as they are at the very heart of the rail network, the very crossroads of Europe especially with their new Alptransit routes (Gotthard and Loetchberg base tunnels). In addition to this international effort, their approach to ETCS on their own domestic network is very illuminating. Instead of the very complex conversion to level 2 as UK has committed to, they have developed a variant of Level 1 called limited supervision (LS). This, overlaid on existing signalling or incorporated with conventional resignalling, just provides an equivalent of what their previous classic train protection systems used to do, namely a simple advance warning, over-speed and train-stop functionality rather like our UK AWS/TPWS, using active ETCS balises and loops. Using this method SBB plan to have their entire network ETCS fitted by 2017. They will continue to use and specify Level 2 for new high speed, Alptransit and some other very busy main lines where the higher level offers some additional performance and capacity benefit, but for the vast majority of the network Level 1-LS is being and will be used. Their master stroke is to have the LS mode incorporated in the ETCS specifications for Europe-wide application. Hence any international unit turning up on Swiss metals will have access to the entire network, using either Level 2 where fitted or the LS mode.

There are two very good reasons why the Swiss baulked at the complexity of Level 2 or even 'normal' Level 1 system wide:

1. Older traction units with their simple coarse power and braking controls are very difficult to adapt to the very tight control paradigms of traditional ATP systems, of which 'conventional' ETCS is a development. These often require very small frequent adjustments to power and brake and very accurately monitored instrumentation. LS allows the simple concepts of traction interlocks and simpler brake interventions used for the older stock to be reused as they are in the historic protection systems, without significant expensive alterations to the systems in older units with only limited remaining lives.

2. LS doesn't require the highly accurate route characteristics data that onboard computers for higher levels need for continuously calculating their permitted speed within their movement authorities, and equally importantly, the management of creating, changing and dissemination all that data in a suitably controlled way is avoided.

The result of this in Switzerland is that the entire (standard gauge!) network will be accessible to all modern traction units fitted with full ETCS, whilst within their mixed domestic fleet, older units and locos impractical to convert to the higher levels will still be able to run on the vast majority of the network, with the exception of the very busy Level 2 lines (where their performance is probably unsuitable anyway).

Both French and German railways have been very supportive of SBB pursuing the LS concept, including cooperating with its proposal for Europe-wide adoption in the specs. I believe they also see widespread possible application of this technique in their own countries. UK has been strangely silent by contrast, NR always taking the view that UK will not consider Level 1 anywhere as that seems not to be able to deliver the trackside equipment simplifications over colour lights that level 2 initially seemed to promise.

 

[jopsuk]

Switzerland is bascially an oversized junction sat in the middle of the European network. Removing barriers to international transit is smart business.

GB is, to the European network, a load of sidings at the end of a branchline, with gauging issues.

 

[MarkyT]

Switzerland is bascially an oversized junction sat in the middle of the European network. Removing barriers to international transit is smart business.

GB is, to the European network, a load of sidings at the end of a branchline, with gauging issues.

Ha ha! Excellent summary!

Mind you I still think Level 1-LS would be a good incremental policy for UK, providing a better safer upgrade path from TPWS on secondary routes (the majority of the network) alongside full Level 2 on the busiest trunk sections.

 

[jopsuk]

You're not wrong, it's just the business case has to be entirely internally justified.

 

[Tio Terry]

I seriously doubt that ERTMS is required to be installed to the main GB network by European directive. Interoperability requirements as I understand them apply to dedicated High Speed lines and access to associated stations for International Open Access. Due to the unique nature of the main GB Network and complications caused by our border control requirements through running from the European High Speed network is unlikely. I would think that if Network Rail were subject to such a directive, they would be able to apply for a derogation if they wished and would almost certainly be awarded it.

Network Rail have a bee in their bonnet about delivering a "digital railway" of which ERTMS is a part.

No, that's not the case. Every project has to be assessed against each TSI to decide if it can comply. For example, the new Rochester Station has to comply with the Infrastructure TSI because it's new build and the Persons of Reduced Mobility TSI. NR and the ORR have agreed a list of Signalling led projects that are considered to be "Class B" which will utilise lights on sticks but even then the interlockings used must be able to meet the requirements of ERTMS and nothing the project does can prevent the installation of ERTMS. The main routes have to have a plan for the introduction of ERTMS - some of which are a long time in the future - but there still has to be a plan.

Anything that's new build or classed as a major upgrade - Birmingham New Street for example - has to meet the requirements of the TSI's and NNTR's where appropriate.

As far as I am aware, nobody has got any derogations against TSI's.

 

[Railsigns]

Just on a tangent - how should one pronounce ERTMS? Is it "Artemis", E-R-T-M-S, "ur-tem-ess" or something else? There's a lot to be said for acronyms like TBTC or xkcd...

I've only ever heard it pronounced E-R-T-M-S; that's after attending countless lectures, conventions and technical visits where various boffins were discussing it at length.

 

[hulabaloo]

Just an opinion but there seems to be a lack of motivation to "get on with it" as franchises have their own agendas. Perhaps a nationalised railway would have pushed this on as a solitary force?

 

[Phil.]

Let's take some routes as an example.
Victoria to Brighton.
Liverpool Street to Cambridge.
Birmingham to Bristol.
Leeds to Hull.
There are no worthwhile track improvements that can be made to speed these routes up from the existing line speeds. The only way would be faster accelerating and braking trains. What advantage does ERTMS have over AWS and TRWS equipped conventional 3 or 4 aspect signals coupled with the mark one eyeball?