Is the European Train Control System (ETCS) ever going to be implemented?

[Trestrol]

Moderator note - split from:

Is it the end for GBR??

Article in the Guardian newspaper seems to think the GBR bill might not get through in time for 2024, with the demise of Schapps and the slow progress is it really just another plan to be put in the Bin??

www.railforums.co.uk

I’ll agree with you on that count. ETCS is coming, and lights on sticks will be replaced. But there was never any chance of ambitious politician timescales coming to fruition.

The only people who think ETCS is coming are the people who work in Digital Railway within NR. The rest of us know it will never happen on a big scale in our lifetime. The truth is we will Never afford it system wide, the government won't pay for it. No point in doing it piecemeal do a complete line or not at all. By the time any funding is in place the system to be used will be obsolete. NR has enough problems with obsolescence in systems without main lines done piecemeal so each section uses slightly different systems. It's already been cancelled for Carlisle as its getting new signals instead.

 

[yorksrob]

With regard to ETCS, as long as the trains keep running, no one cares what the signalling system is

 

[Nicholas Lewis]

The only people who think ETCS is coming are the people who work in Digital Railway within NR. The rest of us know it will never happen on a big scale in our lifetime. The truth is we will Never afford it system wide, the government won't pay for it. No point in doing it piecemeal do a complete line or not at all. By the time any funding is in place the system to be used will be obsolete. NR has enough problems with obsolescence in systems without main lines done piecemeal so each section uses slightly different systems. It's already been cancelled for Carlisle as its getting new signals instead.

ECTS works as designed and should be just implemented but as usual on the railways in this country we have start everything from scratch and spend years and half the budget to prove whats already been proven. Quite frankly im staggered Shapps fell for it but then we turned resignalling schemes into massive black holes as well so it probably stacked up against resignalling KX to Peterborough. That said like Network Management Centres it will fade away in due course.

 

[Trestrol]

ECTS works as designed and should be just implemented but as usual on the railways in this country we have start everything from scratch and spend years and half the budget to prove whats already been proven. Quite frankly im staggered Shapps fell for it but then we turned resignalling schemes into massive black holes as well so it probably stacked up against resignalling KX to Peterborough. That said like Network Management Centres it will fade away in due course.

Thing is we muck around with systems used in Europe to make them work in this country. Then we wonder why they don't work as they should. We have plenty of signalling installations that use simple technology that works well. These could be 50-60-70 years old. Most of the equipment can be repair, or stuff that can't can be re-engineered replacements. We have stuff 10-20 years old where spares or components are not obtainable or repairable.

 

[D365]

The only people who think ETCS is coming are the people who work in Digital Railway within NR. The rest of us know it will never happen on a big scale in our lifetime. The truth is we will Never afford it system wide, the government won't pay for it. No point in doing it piecemeal do a complete line or not at all. By the time any funding is in place the system to be used will be obsolete. NR has enough problems with obsolescence in systems without main lines done piecemeal so each section uses slightly different systems. It's already been cancelled for Carlisle as its getting new signals instead.

It is being done ’piecemeal’… you don’t think that the entire ECML will be done at once.

A fundamental point of ETCS is that it won’t become obsolete. If new features are needed, these come in the form of a new baseline i.e. new software.

With regard to ETCS, as long as the trains keep running, no one cares what the signalling system is

Precisely.

Thing is we muck around with systems used in Europe to make them work in this country. Then we wonder why they don't work as they should. We have plenty of signalling installations that use simple technology that works well. These could be 50-60-70 years old. Most of the equipment can be repair, or stuff that can't can be re-engineered replacements. We have stuff 10-20 years old where spares or components are not obtainable or repairable.

ETCS has the benefit that we will be using the exact same equipment as is used in Europe. Albeit we’re still locked into Siemens or Thales…

 

[Trestrol]

You seem to make out software updates are cheap they are not. Also I have to disagree that its only that that needs doing. Look at SSI, 1970s technology still going strong(ish) in the 2020s but is now causing serious hardware problems. Britain's railways can't afford to change its hardware every 10-20years.at the moment 50-60 years if not longer are the norm. Look at the recent shortage of micro chips new technologies suffer because things move on to fast. I am not saying we should go back to mechanical signal boxes but britain's railways need robust reliable signalling systems.

 

[D365]

You seem to make out software updates are cheap they are not. Also I have to disagree that its only that that needs doing. Look at SSI, 1970s technology still going strong(ish) in the 2020s but is now causing serious hardware problems. Britain's railways can't afford to change its hardware every 10-20years.at the moment 50-60 years if not longer are the norm. Look at the recent shortage of micro chips new technologies suffer because things move on to fast. I am not saying we should go back to mechanical signal boxes but britain's railways need robust reliable signalling systems.

Where did I say that software is cheap? The point is that the hardware will not need to be replaced frequently, if at all.

The idea should be that rolling stock retains the same hardware specification throughout its lifetime, and that the lineside infrastructure is minimal, i.e. as close to zero maintenance as it can be.

 

[The Planner]

You seem to make out software updates are cheap they are not. Also I have to disagree that its only that that needs doing. Look at SSI, 1970s technology still going strong(ish) in the 2020s but is now causing serious hardware problems. Britain's railways can't afford to change its hardware every 10-20years.at the moment 50-60 years if not longer are the norm. Look at the recent shortage of micro chips new technologies suffer because things move on to fast. I am not saying we should go back to mechanical signal boxes but britain's railways need robust reliable signalling systems.

It will be ETCS though. The standard design life of most signaling installations has always been 40 odd years, and if 50 can be squeezed out of them then it will. It is also going to be done piecemeal, you will do an area of control at a time. Software updates are expensive as there is no competition, Siemens have effectively sewn up the UK signaling market, irrespective of it being conventional or otherwise.

 

[Nicholas Lewis]

It will be ETCS though. The standard design life of most signaling installations has always been 40 odd years, and if 50 can be squeezed out of them then it will. It is also going to be done piecemeal, you will do an area of control at a time. Software updates are expensive as there is no competition, Siemens have effectively sewn up the UK signaling market, irrespective of it being conventional or otherwise.

We had this problem with the IECCs at privatisation as Delta Rail ended up with all the rights and they used to charge a fortune as well as telling you when they would do it for you not when you wanted it.

 

[eldomtom2]

A fundamental point of ETCS is that it won’t become obsolete. If new features are needed, these come in the form of a new baseline i.e. new software.

I strongly doubt this claim. You cannot completely future-proof hardware.

 

[D365]

I strongly doubt this claim. You cannot completely future-proof hardware.

To be more specific: it would be a costly exercise, but I can see the OBU (onboard unit, i.e. the EVC computer) having some scope for replacement, as these are fitted into a 19” rack. It’s the ancillary hardware - balise reader, radars, cab equipment - that would take a lot of engineering effort to physically replace, as these have to be specifically designed ’into’ the vehicle.

 

[HSTEd]

I very much doubt I will live long enough for ETCS to be the major signalling system on the railway, although I personally believe the railway should do everything possible to accelerate rollout of that and supporting technologies to take full advantage of it.

The potential benefits are huge, but are locked away behind difficult negotiations with freight operators and capital expenditure negotiations with the treasury.

 

[Nym]

What advantages do ETCS have on a mixed traffic railway where 'we' can't even work out how hard a freight train should be able to brake?

 

[The Planner]

What advantages do ETCS have on a mixed traffic railway where 'we' can't even work out how hard a freight train should be able to brake?

No approach control for a start. It might not have a lot of benefit on straight sections, but at junctions and loops it makes a big difference.

 

[RobShipway]

I have worked in IT for almost 30 years and currently work for the NHS on a project to replace 15 - 20 year old apps servers in surgeries and medical centres around Sussex to be on the same domain system. But this is being done on the basis of 3 - 5 surgeries a week.

I presume that the people working on the ETCS project are working in the same manor, with 3 - 5 sections/areas of track have signalling replaced when engineering work can take place on those routes.

 

[Nicholas Lewis]

No approach control for a start. It might not have a lot of benefit on straight sections, but at junctions and loops it makes a big difference.

Presumably the argument is faster clearance time at approach controlled jcns as ETCS will allow braking capability of train to be deployed and not set at the worst braking curve for the the route. This could also be achieved by reduced track circuit release times and additional TPWS OSS loops if all traction had the same braking curves but with mixed traction and mixed braking characteristics not viable.
Personally i would prioritise ARS over everything as many signaller areas have become quite large and its easy to overlook trains and cause untended delays.

 

[D365]

What advantages do ETCS have on a mixed traffic railway where 'we' can't even work out how hard a freight train should be able to brake?

Freight is one thing; I dread to think how much work it is taking to implement emergency braking on Tornado!

 

[Nottingham59]

Won't HS2 have ETCS?

 

[Energy]

Won't HS2 have ETCS?

Yes.

 

[MarkyT]

Freight is one thing; I dread to think how much work it is taking to implement emergency braking on Tornado!

TPWS can already make an emergency braking response on Tornado and other mainline-certified locos. That is perhaps the easier part of the problem. While the loco will have a full EVC implementation with appropriate sensors and screens to calculate and display a continuous safe speed envelope to manually drive within, the interventions on older traction will likely be less subtle. Whereas on a modern train a controlled descent to a particular speed can often be initiated automatically by an ATP system, on steam and early diesels the response to an overspeed at any point is far more likely to be a full emergency brake application to a stand, like TPWS, simply because the traction control system isn't sophisticated enough allow a target speed to be 'dialled in'. Clearly in a SPAD (or exceeding the extent of movement authority) the emergency brake response will be the same. I don't know but I suspect they'll also be putting in an automatic cutoff valve in the steam supply circuit between regulator and cylinders on Tornado to reinforce the emergency stop response. That could also plausibly be triggered by the TPWS on the loco, which will have to remain for non-ETCS lines clearly.

 

[D365]

TPWS can already make an emergency braking response on Tornado and other mainline-certified locos. That is perhaps the easier part of the problem. While the loco will have a full EVC implementation with appropriate sensors and screens to calculate and display a continuous safe speed envelope to manually drive within, the interventions on older traction will likely be less subtle. Whereas on a modern train a controlled descent to a particular speed can often be initiated automatically by an ATP system, on steam and early diesels the response to an overspeed at any point is far more likely to be a full emergency brake application to a stand, like TPWS, simply because the traction control system isn't sophisticated enough allow a target speed to be 'dialled in'.

Traction cut is the first stage of an ETCS intervention, then a full service brake, then emergency. As you've pointed out, it's difficult to implement braking on a steamer or 'heritage' diesel, i.e. anything pre-56. Slamming the brakes is one thing, recovering from an intervention is another.

I wouldn't even know if a 'traction cut' is possible on Tornado!

 

[kevin_roche]

Is there any news on the status of the ETCS overlay project between Paddington and Heathrow Junction? Last I heard the western part had been installed but not turned on.

 

[zwk500]

Is there any news on the status of the ETCS overlay project between Paddington and Heathrow Junction? Last I heard the western part had been installed but not turned on.

I believe, but may be wrong, that ETCS is providing the ATP function for trains fitted with only ETCS, but it's not yet giving out Movement Authorities.

In regards to the original question: Yes, ETCS is already in use on the UK rail network. However it will likely not ever be 100% coverage of the network.

 

[Pete_uk]

(or exceeding the extent of movement authority)

EEMA - I for one welcome this new acronym.

 

[D365]

However it will likely not ever be 100% coverage of the network.

Indeed it won't. Rural lines will retain some form of radio block signalling.

 

[zwk500]

Indeed it won't. Rural lines will retain some form of radio block signalling.

I fully expect some semaphores to still be kicking about when we've moved on to some form of Skynet controlling driverless trains.

 

[Halifaxlad]

Well presumably its been chosen for TRU between Manchester, Leeds and York according to the IRP.

Does anybody know how services such as Bradford - Huddersfield will be able to operate post TRU ?

Im presuming even the slow lines will be under ETCS!

Also will Penistone line services that won't be ETCS be able to reach Huddersfield ? Im not actually sure where there depot is but I suspect its Neville Hill.

 

[zwk500]

Well presumably its been chosen for TRU between Manchester, Leeds and York according to the IRP.

Does anybody know how services such as Bradford - Huddersfield will be able to operate post TRU ?

Im presuming even the slow lines will be under ETCS!

Also will Penistone line services that won't be ETCS be able to reach Huddersfield ? Im not actually sure where there depot is but I suspect its Neville Hill.

That report is quite vague as to what 'partial' and 'full' digital signalling is, but I'd guess at least one of them involves ETCS being the 'brains' of the signalling but lineside signals remaining for unfitted trains (referred to as ETCS L2 Overlay). It's even possible to have a Setup where there are separate ETCS blocks in between signals, so that ETCS-capable trains can run at a shorter headway but non-ETCS trains can still run on the line (Thameslink has this feature). Given the large amount of new stock, ETCS fitment should not be too troubling for a full signals-free ETCS installation should it be required for the service spec.

If the Penistone-line trains can't be fitted with ETCS before the go-live date, there'll be some functionality for legacy systems left in to permit them to operate to Huddersfield. It's worth noting that there's a lot of freight movement in that part of the world and ETCS fitment on freight has, IIRC, been quite slow so far. Although once the Class 66 design is finalised, fitment will presumably speed up somewhat.

 

[D365]

It's worth noting that there's a lot of freight movement in that part of the world and ETCS fitment on freight has, IIRC, been quite slow so far. Although once the Class 66 design is finalised, fitment will presumably speed up somewhat.

First Class 66 has gone for fitment, other classes to follow. Owing to the large number of interfaces, installation designs have to be tailor-made; in some cases on a vehicle-by-vehicle basis.

 

[zwk500]

First Class 66 has gone for fitment, other classes to follow.

Yippee!

Owing to the large number of interfaces, installation designs have to be tailor-made; in some cases on a vehicle-by-vehicle basis.

Oh, for [redacted]....

Hopefully at least some of the designs can be used, even if tweaks have to be made around the edges.