Elizabeth Line / Crossrail Updates.

[St. Simon]

Hi,

 

A note to add about CBTC and ETCS on the central core.

 

The central core is only operated by CBTC because a mature ETCS Level 3 solution is not available, once ETCS Level 3 has become a matured product, then Crossrail are legally bound to change the CBTC for ETCS Level 3.

 

Simon

[phil-b259]

It is not a technical issue at all, it is because it is not European. After Brexit we may well be able to install TPWS, but that will be too late for this line.

 

TPWS is an INFERIOR SYSTEM to ATP!

 

Let me remind you why TPWS was invented, it was only developed to address one single risk - namely that posed by a driver who failed to stop at a red signal by slamming on the brakes. It cannot cope with high speeds effectively and is a pretty crude system.

 

ATP provides contentious protection for every meter/ yard of track as it is constantly monitoring the trains speed regardless of whether there is a signal at that position.

 

The simple issue BR devised system is now obsolete (i.e. its proving very hard to source the electronics, etc) and the choice of its replacement has NOTHING to do with the EU in the slightest. It is simply the case that the only 'off the shelf' system which provides the same (or better) levels of protection is ECTS.

 

Dropping the level of safety protection is never acceptable.

 

For a traditionally signalled line, TPWS is better than nothing, but its far safer if an ATP system is installed, so going back to TPWS is not acceptable.

Edited September 2, 2018 by Andy Y
Political reference removed

[phil-b259]

As a matter of interest, why is adopting TPWS for the Heathrow Airport line and stock such a big no-no ? The route is neither long nor complex, and similar trains elsewhere have TPWS, so technical considerations should not be an issue, and TPWS is well-proven in both service and effectiveness.

Quite simply because TPWS is a significant downgrade in safety terms compared to an ATP system.

 

Please remember that in tunnels collisions between trains can be far more catastrophic than in the open (think emergency evacuation / attendance by emergency services difficult, ventilation issues if fire / smoke is present etc).

 

The Heathrow branch is entirely in tunnels and although these are far better designed to cope with incidents than say Victorian ones, they still represent a heightened risk.

 

Now I can hear you saying 'well we don't demand the same protection said Victorian tunnels' - and you are correct, we don't. However there is a well established principle that while it may not be cost effective to retrospectively fit new systems to old structures (unless its being done as part of a long planned upgrade) we don't cut corners on new stuff.

 

Its why the ORR have made it pretty much impossible for anyone to install exposed conductor rails to power trains (but are content to leave the large amounts already fitted in the LUL / the ex SR / Merseyside areas alone) or why there is a defaco ban on new level crossings being created (the East-West rail line reinstatement project has to have all 101 crossings between Bicester and Bletchley replaced by bridges or closed) but those that currently exist can be kept if safety systems are kept up to date (13 crossings between Bletchly and Bedford can be kept as part of the same East-West project).

 

The other factor is that the current HEX stock doesn't have TPWS anyway!

 

As I said earlier, because the BR ATP is fitted from Paddington right the way through to Heathrow there has never been any provision made for TPWS on the HEX units. Yes TPWS could be fitted - but that does not get away from the fact that you are reducing the level of safety compared to an ATP system.

Edited September 1, 2018 by phil-b259

[caradoc]

You make very valid and interesting points phil-b259, but on the 99% of Network Rail lines which do not have ATP, TPWS is deemed perfectly adequate, even for newly constructed lines (eg the Borders Railway), and to the best of my knowledge there are no plans, and certainly no finance, to install ATP, or indeed any other kind of upgraded signalling system, on any existing routes, including the many miles passed for 125mph.

And while TPWS is not normally installed to prevent rear-end collisions, there is no technical reason to prevent it if deemed necessary in what might deemed higher-risk locations, such as tunnels.

[St. Simon]

You make very valid and interesting points phil-b259, but on the 99% of Network Rail lines which do not have ATP, TPWS is deemed perfectly adequate, even for newly constructed lines (eg the Borders Railway), and to the best of my knowledge there are no plans, and certainly no finance, to install ATP, or indeed any other kind of upgraded signalling system, on any existing routes, including the many miles passed for 125mph.

And while TPWS is not normally installed to prevent rear-end collisions, there is no technical reason to prevent it if deemed necessary in what might deemed higher-risk locations, such as tunnels.

TPWS was / is only deemed adequate because ETCS (an ATP system) will be adopted for the network in the future, that was the legal thinking behind the regulation when it was introduced.

 

Network Rails / Government policy is that ETCS WILL be installed on the whole network eventually, and the finance is being put aside for it. You can argue about the finances and timescales, but that is the policy and a sensible policy at that.

 

There has been, to the best of my knowledge, one case where TPWS has widely been installed to protect against a rear-end collision, and that is the additional TPWS equipment installed on the lines between Heathrow and Paddington to enable the Class 345s some additional protection in the absence of ETCS. TPWS was only considered adequate for this because of the temporary nature of its installation in this case.

 

If we could, we would installed ATP everywhere, the issue being the vast expense of fitting the existing trains with the system (the GW HSTs were particularly difficult as each power car was effectively hand built and therefore a pain to integrate with the ATP system). However new trains are being built with ETCS fitted and older stuff, such as the Electrostar, Desiros and Turbostars, has been built with the space provision for the Euro computer, so it's getting easier and easier.

 

Simon

Edited September 3, 2018 by St. Simon

[Mike Storey]

Warnings over probable delays, despite official "spokespeople" saying otherwise, existed as far back as February.

 

https://www.bbc.co.uk/news/uk-england-london-42921305

[phil-b259]

You make very valid and interesting points phil-b259, but on the 99% of Network Rail lines which do not have ATP, TPWS is deemed perfectly adequate, even for newly constructed lines (eg the Borders Railway), and to the best of my knowledge there are no plans, and certainly no finance, to install ATP, or indeed any other kind of upgraded signalling system, on any existing routes, including the many miles passed for 125mph.

And while TPWS is not normally installed to prevent rear-end collisions, there is no technical reason to prevent it if deemed necessary in what might deemed higher-risk locations, such as tunnels.

 

Plans are well developed to remove ALL lineside signalling on the ECML south of Peterborough within a decade and replace it with ECTS (and that includes fitting the ECTS system to the likes of class 66 FOC locos*). Due to budget pressures the target date has slipped but the intention is still to avoid having to replace the 1970s era conventional signalling which is now life expired on a like for like basis.

 

*Network Rail / the DfT are funding the 'first in class' fitment project (literally equipping the first loco / EMU / DMU of any class required to work with ECTS with the appropriate equipment for TOCs / FOCs to copy on the rest) as we speak - although as Simon has indicated this is not a simple task for traction that doesn't have passive provision built in.

 

Similarly due to obsolescence issues with the BR ATP installed on the GWML, ECTS (albut retaining lineside signals initially) is due to go in on the GWML when electrification has been completed so as to allow the maximum permissible speed of the 800s to go up to 140mph. Again we are talking within the next decade here.

 

By contrast plans to resignal the Edinburgh area are not on the agenda yet (hardly surprising given the kit installed by BR in the late 1980s for electrification - and of couse other modernisation schemes since then) still has plenty of life left in it so adding an isolated ECTS installation for the Borders line was not viable. This is rather different from the GWML where ECTS is happening in the near future.

 

 

Note:- The expected lifespan of electric / electronic railway signalling equipment is around 40 / 50 years - installing something that will become redundant in 10 years is not worth the expense unless there is a immediate H&S issue that needs addressing.

Edited September 2, 2018 by phil-b259

[Platform 1]

 

...remove ALL lineside signalling on the ECML...

 

The way things are going, there won't be much left to model...

[Zomboid]

One thing I noticed as part of the TfL-isation of the GWML slow lines is that "London Paddington" is now referred to at places such as Burnham and Taplow by its TfL name of simply "Paddington".

 

Just an observation...

[jim.snowdon]

Plans are well developed to remove ALL lineside signalling on the ECML south of Peterborough within a decade and replace it with ECTS (and that includes fitting the ECTS system to the likes of class 66 FOC locos*). Due to budget pressures the target date has slipped but the intention is still to avoid having to replace the 1970s era conventional signalling which is now life expired on a like for like basis.

 

*Network Rail / the DfT are funding the 'first in class' fitment project (literally equipping the first loco / EMU / DMU of any class required to work with ECTS with the appropriate equipment for TOCs / FOCs to copy on the rest) as we speak - although as Simon has indicated this is not a simple task for traction that doesn't have passive provision built in.

 

Similarly due to obsolescence issues with the BR ATP installed on the GWML, ECTS (albut retaining lineside signals initially) is due to go in on the GWML when electrification has been completed so as to allow the maximum permissible speed of the 800s to go up to 140mph. Again we are talking within the next decade here.

 

By contrast plans to resignal the Edinburgh area are not on the agenda yet (hardly surprising given the kit installed by BR in the late 1980s for electrification - and of couse other modernisation schemes since then) still has plenty of life left in it so adding an isolated ECTS installation for the Borders line was not viable. This is rather different from the GWML where ECTS is happening in the near future.

 

 

Note:- The expected lifespan of electric / electronic railway signalling equipment is around 40 / 50 years - installing something that will become redundant in 10 years is not worth the expense unless there is a immediate H&S issue that needs addressing.

And in the last 40 years, we have gone from Windows 3 to Windows 10, and several generations of computer based interlocking. As railway signalling technology has progressed from the mechanical age, through the electric age and into the electronic age, the working life of the equipment has got progressively shorter. At the same time, the manufacturers have a greater vested interest in not only protecting access to their systems, making alterations more difficult and expensive (this problem already exists with the proprietary SCADA systems), but also in building in obsolescence.

 

Jim

[mdvle]

And in the last 40 years, we have gone from Windows 3 to Windows 10,

It's actually worse than that.  Windows 1.0 was only 32 years ago, MS-DOS only takes you to 37 years...

[jim.snowdon]

I'm pretty certain that when BR developed Solid State Interlocking (SSI), a key principle was that it was open access software, so that it could be adopted by any manufacturer and, more importantly, the site specific data modified by BR as and when track and signalling changes were required. Now, we seem to have moved to an era where making any change to the physical railway can only be undertaken, for a suitably fat fee, by the manufacturer of each particular system, and woe betide you if you try and use anyone else to try doing it.

Exactly the same situation has bedevilled the SCADA systems used for electrical control, and resulted in manufacturer specific systems for such functions as selective door operation, with systems tied to particular classes of rolling stock. It's slowly getting back to the days of the early railways, where you couldn't guarantee to be able to couple different company's stock together.

 

Jim

[Ron Ron Ron]

London Reconnections recently published their own interpretation of what the Croosrail service pattern may look like.

This differs from Geoff Marshall's guesswork shown here earlier in post #664 on page 27 of this thread.

 

 

 

Edited September 3, 2018 by Ron Ron Ron

[62613]

Haven't seen a schematic as good as that since I left work!

[62613]

What does it all mean, though? Are we saying that there's no direct services between the GW and the GE sides?

[Ron Ron Ron]

What does it all mean, though? Are we saying that there's no direct services between the GW and the GE sides?

 

 

These two interpretations (from LR & GM) are basically guesswork, based on the bits and pieces of detail made available.

There are still lots of questions to be answered though, so nobody outside of the people involved in the timetable planning, is able to give definitive answers.

 

 

 

.

[phil-b259]

What does it all mean, though? Are we saying that there's no direct services between the GW and the GE sides?

 

YES thats exactly what we are saying - do keep at the back, this has been the case for well over 5 years now

 

All trains from the GEML will go no further than Paddington low level (well Westbourne Park turnback siding to be exact) until the HS2 interchange station at Old Oak opens, when they will transfer to terminating there.

 

In all seriousness the reasons for no through services are:-

 

(1) Canary Wharf group stumping up an extremely large contribution to construction costs on the proviso that it got the majority of Heathrow trains so bankers didn't have to change.

(2) Concerns about disruption on one main line GEML / GWML  being imported to the other*

(3) The problem of meshing different service intervals - trains on the GWML are based around / flighted to fit round 15 minute intervals - the Shenfield line has traditionally based around a 10minute service interval.

 

* There are already big problems where ECML, MML and various Southern / SE services are routinely delayed to ensure Thameslink trains hit their slot through the core on time - and thats before they get to 24TPH as planned!

[Ron Ron Ron]

TfL Rail Class 345's at Ealing Broadway.

Paddington to Hayes & Harlington (terminators).

 

 

[The Stationmaster]

Not in part.

A sub-fleet of 387's will be modified, repainted and refitted to take over the whole 4 tph HEX service.

GWR are going to be operating HEX on behalf of Heathrow Airport (HAL).

It's still going to be HEX, owned and managed by HAL, on HAL's open access licence.

 

The other 387's will be operating Thames Valley services, except for the relief line local services between Reading/Heathrow and London, which will go to TfL's Elizabeth Line.

They should be serving Oxford too, but we know what a massive c**k up has happened over that.

 

 

.

 

Although that 'London Reconnections' link suggested something different for east of Reading semi-fasts (really 'limited stop stopping trains) but that in turn does not agree with the last I heard from GWR (which was some time ago) although i know that various local rail user/commuter groups had been pressing for a retained off peak GWR/successor 387 worked service off-peak.   But there will still be peak period 387s calling at Twyford and Maidenhead so maybe that is what 'London reconnections' has got into a knot with?

[jonny777]

One thing I noticed as part of the TfL-isation of the GWML slow lines is that "London Paddington" is now referred to at places such as Burnham and Taplow by its TfL name of simply "Paddington".

 

Just an observation...

 

 

 

No no no - naughty  

 

You know the "slow" lines on the GWR are actually the relief lines.  Slow lines are such a northern invention. 

[Zomboid]

"Slow lines" is the terminology on the LSWR route, and is therefore the correct terminology (or at least not a Northern term). As Duck used to say, there's two ways to do things, the GWR way and the right way...

Edited September 3, 2018 by Zomboid

[jonny777]

Ah, but did the LSWR just copy the GNR and/or LNWR practice? 

[Zomboid]

If they did, they were simply validating it as the correct nomenclature...

[Mike Storey]

And in the last 40 years, we have gone from Windows 3 to Windows 10, and several generations of computer based interlocking. As railway signalling technology has progressed from the mechanical age, through the electric age and into the electronic age, the working life of the equipment has got progressively shorter. At the same time, the manufacturers have a greater vested interest in not only protecting access to their systems, making alterations more difficult and expensive (this problem already exists with the proprietary SCADA systems), but also in building in obsolescence.

 

Jim

 

Not sure where you are coming from with this Jim, as I agree with almost none of it.

 

I only agree with the issue over manufacturers' IP. But this goes back at least to SSI. I have pulled out many of my few remaining hairs trying to get Westinghouse 1980's SSI altered for various projects, only to find that the "open architecture" claim was bogus, in today's market. We simply had to wait until Westinghouse (and whatever they kept changing into) got around to it, and at whatever price they deemed appropriate. Only the, almost empty, threat of barred future projects would very occasionally, moderate their charges.

[jim.snowdon]

Mike,

 

The underlying point is that computer software is only as good as the support for the operating system on which it runs and the supply of hardware components. There are many software packages out there that were designed to run on, for example, Windows XP and cannot be migrated to current versions of Windows now that XP is no longer supported. On top of that, there comes a point where manufacturers would rather you bought a new system than attempt to keep an old one going. They are, after all, in business to sell product, not provide the customer with a product that will last decades without replacement.

 

Jim