Crossrail - Construction updates and progress towards opening (now expected 24 May 2022)

[43096]

Yes, it would seem so.

The data published by Roger Ford in his monthly Modern Railways column show a jump after the Christmas/New Year software changes. The Miles per Technical Incident doubled between Period 10 (commencing 12 December 2021) and 11 (commencing 9 January 2021). One would expect the trains reliability figures to also be affected by changes in the signalling software so the two things have to be considered together although the TIN figures relate to the train only. An extract from Mr Ford's figures is in the table below (MAA - Moving Annual Average).

Period	Starting	No. of units	No. of TINs	Fleet miles	MTIN	MAA MTIN
9	14 Nov	53	180	245,368	1,363	2,033
10	12 Dec	53	160	248,149	1,551	2,110
11	9 Jan	56	87	267,333	3,073	2,297
12	6 Feb	58	67	253,766	3,788	2,441

Whether these figures will improve sufficiently over the next couple of months to enable TfL to offer a reliable service when Crossrail opens is outside my knowledge. Historically a 'reliable' train has MTIN values of 50,000 and above.

But it has also been noted that once reliability gets above around 10-15,000 MTIN there's very little impact on the TOC's PPM scores from train faults.

 

[Roast Veg]

Though I will say it’s incredibly hard to develop software for hardware and interfaces that dont yet exist.

Especially if your hardware and software teams are on separate continents.

 

[matt_world2004]

A quick look at real time trains for the crossrail core this week shows the trains are running pretty reliably except when it comes to transitioning between the core and the east and west ( for example travelling to old oak common and back ) where there are often considerable delays

 

[ijmad]

I'm not a railway person but I am a software engineer and as an observer, it seems like the one of the biggest missteps with Crossrail has been the amount of custom software development. These trains that need to deal with 3 different signalling systems and the transitions between them, and the interfaces and safety interlocks between those systems.

Crossrail is not the first project running heavy rail trains through a deep tunnel under a city using a CBTC system. Ultimately, might it have been more effective to resignal the whole route end to end using an off the shelf, battle tested ETCS implementation with Level 2 ATO in the core tunnels?

I appreciate this would mean sacrificing some interworking potential but it seems like Crossrail is very close to having the the GWML relief lines to itself anyway, shares the Heathrow tunnels with HeX which is doing ETCS, and has its own pair of tracks east of Stratford.

RER A has been running 24 heavy rail trains per hour in each direction since 1989.

Maybe a little too much 'not invented here' thinking, or people at the top not realising the difficulties of bespoke development and interfacing multiple systems at the ground level.

 

[theking]

Ultimately, might it have been more effective to resignal the whole route end to end using an off the shelf, battle tested ETCS implementation with Level 2 ATO in the core tunnels?

Believe some rumours of the reason cbtc was used so was that TFL had exclusive use of the tunnels between Stratford and Paddington.

Altho when crossrail was supposed to open was ETCS as proven as it is now?

On another note part suspended because of lack of control staff on the west.

 

[swt_passenger]

I'm not a railway person but I am a software engineer and as an observer, it seems like the one of the biggest missteps with Crossrail has been the amount of custom software development. These trains that need to deal with 3 different signalling systems and the transitions between them, and the interfaces and safety interlocks between those systems.

Crossrail is not the first project running heavy rail trains through a deep tunnel under a city using a CBTC system. Ultimately, might it have been more effective to resignal the whole route end to end using an off the shelf, battle tested ETCS implementation with Level 2 ATO in the core tunnels?

I appreciate this would mean sacrificing some interworking potential but it seems like Crossrail is very close to having the the GWML relief lines to itself anyway, shares the Heathrow tunnels with HeX which is doing ETCS, and has its own pair of tracks east of Stratford.

RER A has been running 24 heavy rail trains per hour in each direction since 1989.

Maybe a little too much 'not invented here' thinking, or people at the top not realising the difficulties of bespoke development and interfacing multiple systems at the ground level.

IIRC, because this has been discussed many times before, TfL basically persuaded themselves that ETCS level 2 and ATO wouldn’t be ready in time. My personal view is they wanted intentionally to be different to mainline rail, to keep their new route independent.

 

[matt_world2004]

Believe some rumours of the reason cbtc was used so was that TFL had exclusive use of the tunnels between Stratford and Paddington.

Altho when crossrail was supposed to open was ETCS as proven as it is now?

On another note part suspended because of lack of control staff on the west.

TfL would have had exclusive routes to the tunnels due to the platform edge doors anyway.

They had to do an open access operators consultation on the technical standards for open access operators to access the tunnels last year. So it is not impossible for open access operators to access the infrastructure. Providing they meet the technical requirements. Although the utility would be rather limited and operators were required to ensure technical interoperability with tfl Ticketing.

Types of freight was highly limited during the consultation for track access in the crossrail core, and there was a limit of electric traction only too.

Crossrail Central Operating Section

These documents show how train service operators might apply for access to the Crossrail Central Operation Section

tfl.gov.uk

 

[Mojo]

On another note part suspended because of lack of control staff on the west.

That was missing a signaller for the Heathrow section - not part of the TfL infrastructure.

Believe some rumours of the reason cbtc was used so was that TFL had exclusive use of the tunnels between Stratford and Paddington.

Altho when crossrail was supposed to open was ETCS as proven as it is now?

It was asserted at the time that ECTS was not suitable for a rapid transit operation.

 

[coppercapped]

Wow, it has a long way to go! getting above 50,000 would require a 21 times improvement.

Right, quite a hill to climb!

However various older articles in Modern Railways suggest that once the MTIN has passed 15,000 then train failures no longer have a significant effect on the TOCs' Public Performance Measure.

How this rule of thumb will transfer from the normal TOC's average track usage to an intensive underground network I really don't know. But my gut feeling tells me that there ought not to be more than a couple of TINs per day for the whole fleet as a three minute (or longer) delay in a service with potentially a two minute headway can easily throw out the whole service[1].

This, I am sure, is the background to Mark Wild's statements that reliability is not where he would like it to be. I would suggest that for the start of the service through the core tunnels with five minute headways he is probably looking for no more than three TINs per day and preferably lower. Two per day would translate to 56 for the whole fleet per four week period giving a value for MTIN of just under 5,000. Judging by the figures quoted in my earlier post this looks to be achievable as it would require a reliability growth of a third between the beginning of March and the opening of the Paddington - Abbey Wood section. If this is towards the end of May then an improvement of MTIN of around 10% per month might just about do it - and this section has no signalling changeovers so is a bit 'simpler'. The Easter software update will, I am sure, have helped in this respect but as yet no data are publicly available.

This analysis is, of course, simplified, as some TINs will occur between Paddington and Reading and Liverpool Street and Shenfield but the type of failure occurring on these stretches may not be applicable to the Paddington - Abbey Wood section.

Depending on the scale of the improvements gained through the Easter software updates the opening may be sooner rather than later; but the later the opening, the more reliable the service is likely to be. A difficult call!

[1] Of course the other side of the coin is the length of time that is needed to recover from each incident. If it takes five minutes (say) to reboot the train's computers then the service will be disrupted for at least ten minutes (3 mins for the TIN, plus a couple of minutes for fault finding and 'phoning a friend' and five minutes to go through the reboot process). For an intensively used underground railway this length of delay is, I suggest, unacceptable.

 

[Bald Rick]

Ultimately, might it have been more effective to resignal the whole route end to end using an off the shelf, battle tested ETCS implementation with Level 2 ATO in the core tunnels?

Altho when crossrail was supposed to open was ETCS as proven as it is now?

IIRC, because this has been discussed many times before, TfL basically persuaded themselves that ETCS level 2 and ATO wouldn’t be ready in time. My personal view is they wanted intentionally to be different to mainline rail, to keep their new route independent.

In the past I have spoken to those involved in the decision. At the time it was taken, ETCS was not proven in use for intensive operation, and the ATO ‘add on’ was little more than a concept. The cycle* time for ETCS was (and still is) rather longer than CBTC. CBTC was proven and working. Also, ETCS did not - indeed still does not - have the facility to interface with Platform Screen Doors, and similarly there is no Autoreverse facility in ETCS. Both required for Crossrail.

it is fair to say that with 6/6 hindsight there would have been a different decision, albeit it would have required the Crossrail project to be the Sponsor of the two ‘add ons’ mentioned above, as well as possibly other issues.

* the cycle time is the duration between updates by Radio from the control centre to the train. Metro services need then very frequently (<5 sec). ETCS doesn’t do this, although in theory it could.

 

[dm1]

* the cycle time is the duration between updates by Radio from the control centre to the train. Metro services need then very frequently (<5 sec). ETCS doesn’t do this, although in theory it could.

It could, but in the ETCS 'world' probably only practical once the railway version of 5G is established.

A cycle time that is too short can have issues if a transmission is disrupted for any reason. If the train expects a transmission to update its movement authority and does not receive it, then the train will automatically trigger emergency braking, in order to stay within the previous movement authority. That would not have a positive effect on reliability.

Of course any CBTC system does face this problem, but proprietary systems can apply various solutions and work entirely differently in order to deal with it, be that through software or through other transmission methods (WiFi, induction loops, tags), whereas any changes to ETCS need to go through a whole series of committees and other beaurocratic steps before they can be implemented.

Going entirely off topic for a moment, I wonder if that's the reason why the standard ETCS DMI looks so horrible from an aesthetic point of view - by definition designed by committee.

Nevertheless it all boils down to the fact that programming and designing software for something as safety-critical as railway signalling is just extremely hard. ETCS has numerous quirks, as do the various flavours of CBTC that follow from this fact.

As I understand it, the signalling in the central section is also integrated with the ventillation systems too. I'm not sure whether ETCS is designed to deal with this either.

 

[Domh245]

* the cycle time is the duration between updates by Radio from the control centre to the train. Metro services need then very frequently (<5 sec). ETCS doesn’t do this, although in theory it could.

The cycle* time for ETCS was (and still is) rather longer than CBTC.

How does the Thameslink core cope in this case? On the surface they would seem to be similar with a 24tph design

 

[Bald Rick]

How does the Thameslink core cope in this case? On the surface they would seem to be similar with a 24tph design

I was hoping that question wouldn’t come up, as I’m not sure of the answer!

it might be something to do with the TL core is shorter (the part with 24tph), and has a lower linespeed, than Crossrail; there are therefore fewer messages being processed each cycle. But my slightly befuddled Saturday evening brain can’t work the logic.

 

[kevin_roche]

How does the Thameslink core cope in this case? On the surface they would seem to be similar with a 24tph design

There is still a concern that the GSM-R system might not cope if most London signalling is converted to ETCS. On the GWML Airport Junction to Paddington will eventually be ETCS all the way. Originally Waterloo was planned to be done early on but has now been delayed. The Trainguard MT (CBTC) system used on Crossrail is a moving block system and reports train position and gets a new movement authority every 400ms while ETCS L2 is a fixed block system and will report position and receive movement authorities much less frequently (seconds to minutes). The Trainguard MT system on Crossrail is actually supposed to be a 30tph capable system.

 

[Gerald Fiennes]

But the fundamental, IMHO, remains the complete inabilty of rail organisations to talk to each other and find common solutions rather than 'my way is best and everyone else has to fall into line'. Some truly appalling examples of this out there at present. The 'complex' fares and fares systems environment between national rail and TfL is just one. DfT effectively refused to 'knock heads together' in the way that was really needed....

 

[Domeyhead]

There was, by all accounts, a sizeable crowd at Battersea Power Station early on its opening day, even though that was very early in the morning. A soft unannounced opening is much more sensible.

I passed through Paddington mainline station yesterday and walked along Eastbourne Terrace. Behind the railings around the Crossrail station, I noted staff on the gateline even though no trains were running.

Frankly, if they are confident of the line opening by the end of June, with the trains already running, I don't really get why they don't announce Sunday 26 June (or even Thursday 30 June) now as the opening date and then have a soft opening mid morning a few days earlier where they just decide to open the entrances. There really isn't any need to have this speculation about an earlier opening when they can just go for the latest date possible. Is the issue that they might not make the end of June?

Having worked (and managed) quite a few projects in my time I know exactly why it is fatal to plan for the latest date possible! You can almost smell the complacency when the pressure is taken off.

 

[Snow1964]

A friend of mine was passing through Tottenham Court Road yesterday and he said all the direction signs to Elizabeth line have been uncovered (not personally verified by me). But could have happened anytime in last few weeks

 

[ijmad]

Thanks all for your thoughts about ETCS and whether it was a viable option when the Crossrail project began.

I do still wonder if there were other more 'established' systems that could have been deployed throughout the line. TfL spent a lot of time looking at how to overlay/underlay Thales CBTC and coloured signals for the 4LM project.

Might there have been an opportunity for the 345s to use CBTC throughout the line, with a coloured signal overlay/underlay for other traffic where necessary? Would this have vastly decreased the complexity of the train software and perhaps led to improved reliability more quickly?

Water under the bridge I suppose, although I think there should be an inquiry in to what happened in the lead up to 2018 at some point.

 

[Cowley]

probably best we leave the ETCS speculation there now before we end up going too far off topic.

Thanks

 

[traji00]

A friend of mine was passing through Tottenham Court Road yesterday and he said all the direction signs to Elizabeth line have been uncovered (not personally verified by me). But could have happened anytime in last few weeks

Indeed so. Here is an image I took on the 20th April, from inside a Northern Line train.

 

[Mojo]

When I came off the Central line there are still a small number of signs still covered up which in the dirt of the covering sticker you can read Crossrail.

 

[AM9]

When I came off the Central line there are still a small number of signs still covered up which in the dirt of the covering sticker you can read Crossrail.

I would guess that they will stay covered until they are replace with Lizzy ones or removed altogether. They don't want to dilute the brand name.

 

[BahrainLad]

Noticed line diagrams by the PEDs on the Jubilee line at LBG have Elizabeth line on them now. Inclusion of the word “line” looks odd when set against the others.

 

[pdeaves]

Noticed line diagrams by the PEDs on the Jubilee line at LBG have Elizabeth line on them now. Inclusion of the word “line” looks odd when set against the others.

That's a strange picture, like the overlays are 3D. Are they attached with, say, some thick double sided tape or something to give the shadow? (Applies to Stratford and Canary Wharf, just visible at the top).

 

[DC1989]

I heard the opening date will be the 17th May now

 

[Basil Jet]

Mayor Khan was elected on a promise to give the Overground Lines (back their) individual names, but this hasn't happened. Since most of the enamel signs in London are being changed in connection with the opening of the Liz, a great opportunity to rename the Overground lines at minimal expense appears to have been missed.

 

[ijmad]

Most of the enamel signs are having full size decals stuck over them as far as I had heard

 

[Mojo]

That's a strange picture, like the overlays are 3D. Are they attached with, say, some thick double sided tape or something to give the shadow? (Applies to Stratford and Canary Wharf, just visible at the top).

On the Jubilee line PEDs the line diagrams are stuck on the inside of the glass, whereas the overlay sticker is stuck on the other side of the glass, hence it looks 3D in the picture, but it’s just the difference between the thickness of the glass, where the lighting from the platform is creating a shadow behind the new sticker.

 

[pdeaves]

On the Jubilee line PEDs the line diagrams are stuck on the inside of the glass, whereas the overlay sticker is stuck on the other side of the glass, hence it looks 3D in the picture, but it’s just the difference between the thickness of the glass, where the lighting from the platform is creating a shadow behind the new sticker.

Ah, inside/outside, thanks. I didn't think they made special 'shadow' stickers!

 

[Wolfie]

Mayor Khan was elected on a promise to give the Overground Lines (back their) individual names, but this hasn't happened. Since most of the enamel signs in London are being changed in connection with the opening of the Liz, a great opportunity to rename the Overground lines at minimal expense appears to have been missed.

Your first sentence is massively over-egged in that the way that it is written it suggests that was the primary reason that Londoners voted for Khan. In reality most Londoners don't care in the slightest about what individual transport lines are called.
While l personally agree with you that an opportunity has possibly been missed l suspect that TfL's priority has been finally delivering the long-delayed Crossrail.
Oh, and new enamel signs are not being bought, updates are being done by means of stickers.