Issues at Paddington

[andypurk]

I wonder if anyone has any insight into the issues which I've been seeing with trains from the Elizabeth line core at Paddington. For example, this evening my train from Liverpool Street was delayed by a failed train at Paddington. We left Liverpool Street more or less on time and then sat in the tunnel between there and Farringdon for about 30 mins before losing a few more minutes to Paddington. At Paddington the train then stopped short (at least half a coach length) and we had to await permission to draw forward to the correct spot on the platform (which took at least another five minutes). Presumably this was for the driver to correctly dock the train manually. Two trains with problems in an hour seems most unusual.

Train was 9T88 due from Liverpool Street at 18:52 and heading for Heathrow Terminal 5 and annoyingly it looks like cumulative delays took me over the maximum journey time for a Tottenham Hale to Harrow & Wealdstone via Zone 1 (diversion via Stratford [the reason for the trip] and issues on the Bakerloo / DC line as well).

This has not been the first occasion when I've been delayed in the core heading westbound due to issues at Paddington and there does seem to be a bit of a weak point here when there are problems.

[rincew1nd]

Is Paddington a CBTC boundary? The system change seems to be causing issues out east.

[100andthirty]

It is my understanding that the change boundary is just past where through trains go beyond the reversing sidings to the west of Paddington.

[spsmiler]

A delay over 30 minutes caused by the railway should invoke a 'delay repay' payment.

I'm not sure of the exact amount but you might get a 50% refund, or you might even get 100% refund. Certainly you should not be left out of pocket because you exceeded a journey time limit.

btw, it might happen automatically, if not you will need to apply for the refund - and correction to your card for exceeding the journey time limit.

[spsmiler]

Dec 3, 2022 23:12:15 GMT rincew1nd said:

Is Paddington a CBTC boundary? The system change seems to be causing issues out east.

There is another signalling boundary at the start of the Heathrow Airport route, as this uses ETCS (or is it ECTS?)

I've not heard of any issues here. Maybe this is because trains are not 'falling over' during this changeover.

What I also do not know is whether this is another automated driving system with ATP etc (ie: a fully fledged alternative to CBTC) or its just a train protection system similar (in theme) to that used on the rest of the Great Western and Great Eastern sections of the Elizabeth line

[andypurk]

Dec 4, 2022 12:50:17 GMT spsmiler said:

A delay over 30 minutes caused by the railway should invoke a 'delay repay' payment.

I'm not sure of the exact amount but you might get a 50% refund, or you might even get 100% refund. Certainly you should not be left out of pocket because you exceeded a journey time limit.

btw, it might happen automatically, if not you will need to apply for the refund - and correction to your card for exceeding the journey time limit.

Thanks. Yes, I know this, just waiting for the 48 hours in case of automated refund, before making a claim for the journey time limit (2 x £6.00 vs £3.40) and/or Delay-Repay. The two legs are already showing with nearly two hours Liverpool Street (TfL) to Harrow & Wealdstone as opposed to the 40-50 minutes it should actually have taken.

As an aside, Delay-Repay is always an interesting one when there are delays on the DC line, as London Overground and the Bakerloo have different criteria but are likely to suffer from related delays.

[andypurk]

Dec 3, 2022 23:12:15 GMT rincew1nd said:

Is Paddington a CBTC boundary? The system change seems to be causing issues out east.

What looks to be the train causing the initial delay did seem to have an issue at Whitechapel being there for 16 minutes after arriving from the Shenfield direction and then loosing time between there and Paddington and taking another ~20 mins to leave the platform for Old Oak Depot (it looks to be one of the trains booked to depot not to Westbourne Park CS).

I'm still quite surprised that the train I was on undershot the stop location so badly at Paddington and that it took more than five minutes to resolve. It seems to have kept running to the end of service though, finally getting back on time leaving Reading at 22:52!

[tut]

Dec 4, 2022 12:57:58 GMT spsmiler said:

Dec 3, 2022 23:12:15 GMT rincew1nd said:

Is Paddington a CBTC boundary? The system change seems to be causing issues out east.

There is another signalling boundary at the start of the Heathrow Airport route, as this uses ETCS (or is it ECTS?)

I've not heard of any issues here. Maybe this is because trains are not 'falling over' during this changeover.

What I also do not know is whether this is another automated driving system with ATP etc (ie: a fully fledged alternative to CBTC) or its just a train protection system similar (in theme) to that used on the rest of the Great Western and Great Eastern sections of the Elizabeth line

It's ETCS: European Train Control System. It's the signalling part of the ERTMS (European Rail Traffic Management System) package.

Lol. Almost without fail there will be at least one, if not two, sizeable delays attributed to 'ETCS reset' at Heathrow. This is of course extremely funny and always brightens my day.

ETCS is a cab-signalling system that can be superimposed on the existing signalling system (level 1) or can replace lineside signalling (levels 2 and above (I understand level 2 is fixed block, level 3 is moving block)). It incorporates ATP. It very obviously facilitates ATO, but ATO is not, so far as I understand, an essential feature of any level of the system.

[steeevooo]

Dec 3, 2022 22:06:48 GMT andypurk said:

At Paddington the train then stopped short (at least half a coach length) and we had to await permission to draw forward to the correct spot on the platform (which took at least another five minutes). Presumably this was for the driver to correctly dock the train manually. 

I've experienced this a few times at Paddington now - train pulls in slowly and stops, doesn't align with doors (sometimes this could be a half coach length, other times seemingly substantially more), waits, announcement made, train edges forward slowly and stops, doesn't align with doors, repeat accordingly until doors aligned. Sometimes a 2 minute job, sometimes a 5 minute job. Would also be interested to understand exactly why this occurs.

[sunnyday]

To stop too many trains being in each ventilation section, on occasion a train may stop slightly short of the correct stopping point in a station. 

As it is so close to where the train should stop, once movement authority is received, it is in protected manual so the driver moves it manually and the doors should then release. 

I am not sure why the position to stop too many trains entering the next ventilation section wasn't beyond the platform stopping point, rather than a few metres before it!

[spsmiler]

Dec 4, 2022 15:17:25 GMT tut said:

It's ETCS: European Train Control System. It's the signalling part of the ERTMS (European Rail Traffic Management System) package.

Lol. Almost without fail there will be at least one, if not two, sizeable delays attributed to 'ETCS reset' at Heathrow. This is of course extremely funny and always brightens my day.

ETCS is a cab-signalling system that can be superimposed on the existing signalling system (level 1) or can replace lineside signalling (levels 2 and above (I understand level 2 is fixed block, level 3 is moving block)). It incorporates ATP. It very obviously facilitates ATO, but ATO is not, so far as I understand, an essential feature of any level of the system.

Ah, thanks. 

So many different functions, no wonder why I became confused.

Of course it would have been better to use the ETCS etc level 3 version on the Paddington - Abbey Wood section, as then the trains would not have needed the CBTC. But ETCS was not ready for use - whilst CBTC is a known product (perhaps not perfect but still proven safe and commercially viable)

[d7666]

Dec 5, 2022 15:26:22 GMT spsmiler said:

whilst CBTC is a known product (perhaps not perfect but still proven safe and commercially viable)

CBTC as a stand alone term means a technology, not a product.

XR uses Siemens CBTC, which is not the same "product" as the Thales CBTC used on LU surface lines.

I *think* XR uses Siemens Trainguard MT for ATO, Siemens Vicos for ATC, and Siemens Trainguard some-other-name for ATP and the whole lot makes for their CBTC; LU is all Thales Seltrac S40.

The underlying principles are obviously very similar, but differ greatly in detail; each has a lot of proprietry stuff to make it all work, and these proprietary differences make each suppliers CBTC unique.

Unlike ETCS, which is not only an international standard, NetworkRail has an integration centre (at Hitchin, named ENIF) dedicated to ensuring each maker's ETCS products are fully compatible, there is no such interoperability between the CBTC on XR and CBTC of LU.

[joshua]

Will the XR core become ETCS one day, since the GWML is going over to ETCS?

[d7666]

Not aware of any plan to do such a thing.

Nor aware of any reason to do so.

[100andthirty]

Elizabeth Line was designed to conform to the various so-called interoperability specifications. This was a requirement for all "main line" in the UK. Main line meant any railway not designated by the DfT as exempt. London Underground, for example, is exempt. The signalling requirement for new interoperable railways is ETCS, but Crossrail obtained a derogation because ETCS didn't, at the time, have the capability to do all that Crossrail required. A condition was that Crossrail/TfL produce, at intervals, an assessment that Interoperability is stil possible and that ETCS could be added. I'm sure this makes perfect sense in the world of compliance and paperwork.

[revupminster]

February Modern Railways has an article about Luminate Traffic Management that controls GWR and Anglia routes will be extended to take over the Elizabeth Line. It should allow delays due to failures be mitigated. A train sat down at Paddington might be difficult to get round.

[principlesdesigner]

I *think* XR uses Siemens Trainguard MT for ATP & ATO, Siemens Vicos for ATC, and Siemens Trainguard some-other-name for ATP and the whole lot makes for their CBTC; LU is all Thales Seltrac S40.

[spsmiler]

Jan 29, 2023 15:47:20 GMT 100andthirty said:

Elizabeth Line was designed to conform to the various so-called interoperability specifications. This was a requirement for all "main line" in the UK. Main line meant any railway not designated by the DfT as exempt. London Underground, for example, is exempt. The signalling requirement for new interoperable railways is ETCS, but Crossrail obtained a derogation because ETCS didn't, at the time, have the capability to do all that Crossrail required. A condition was that Crossrail/TfL produce, at intervals, an assessment that Interoperability is stil possible and that ETCS could be added. I'm sure this makes perfect sense in the world of compliance and paperwork.

Perhaps more relevant to this message on the "New tube for London stock" thread, I wonder whether the Bakerloo line will be declared exempt when it gets new trains ... after all, a significant part of the route (including a depot) includes use of Network Rail metals.

[Chris M]

As far as I am aware, it will continue to be exempt unless the secretary of state makes an active decision otherwise. The current exemption reads:

London Underground (LU) vehicles and heritage vehicles, and other structural subsystems provided predominantly for the provision of London Underground services or provided predominantly to ensure the compatibility of other services using LU infrastructure

www.gov.uk/government/publications/exclusions-from-the-railways-interoperability-regulations-2011/approved-list-of-exclusions-from-the-scope-of-the-railways-interoperability-regulations-2011