SSR resignalling

[D7666]

With some of Chiltern’s DMUs reaching end-of-life in the foreseeable future, will whatever new replacement rolling stock that Chiltern gets in the future have to have multiple signalling systems, including CBTC, or will they just remain tripcock-based?

Any future Chiltern main line stock will almost certainly have an ETCS solution for Met line operation; ETCS and CBTC are not dissimilar when it comes to underlying signalling principles - it is the track to train communications path and the bespoke software applications that are different. A very simple view is all that is needed is for new Chiltern stock be ETCS fitted but with additional CBTC antennae and the conversion done by train borne processors, or, for Met line trackside CBTC kit to transmit ETCS messages, or for the central 4LM CBTC processors to communicate with the central NR ETCS processors. This type of integration of dual ETCS / CBTC equipped trains is already in place on 345s that have (Siemens) CBTC for the Eliz line core and ETCS for Heathrow Airport - it is not unknown territory. Hitachi/Thales CBTC is simply another variation, not a blocking point. IIMU a longer term aspiration, at least a concept not plan, for LU Ealing / Richmond / Wimbledon is some form of integration with ETCS when NR do eventually get around to it.

 

[bramling]

With some of Chiltern’s DMUs reaching end-of-life in the foreseeable future, will whatever new replacement rolling stock that Chiltern gets in the future have to have multiple signalling systems, including CBTC, or will they just remain tripcock-based?

There’s probably no real need for CBTC. What might become an issue is if it becomes impracticable to fit tripcocks to Chiltern stock, or if LU wish to take the opportunity to ditch the trackside trainstops. Presumably the CBTC system could achieve the same level of protection with TPWS fitted, though there’s the question of who pays.

However, I agree with the view that it will probably be left in the “too difficult” box until such time as it needs to be tackled, and probably end up with an ETCS-based solution, probably at considerable cost getting all the various stakeholders to talk to each other.

 

[zwk500]

and probably end up with an ETCS-based solution, probably at considerable cost getting all the various stakeholders to talk to each other.

Not impossible that it could be an ETCS Level 1 applied to the signals already overlaid on the CBTC interlocking. It's LU infrastructure so NR's preference for L2 wouldn't be an issue, and the Chiltern units wouldn't need any kit they're not already getting. And it potentially makes the transitions to NR easier as well because it'd just be L1 to NTC/L2 (as appropriate).

1 issue with this would be if the Eurobalises and Euroloops would be non-interfering with the CBTC beacons and v.v., and whether LU would want to get certified for installing and maintaining ETCS which they otherwise wouldn't be using.

 

[penfold22]

Any replacement trains in the future will still be required to be tripcock fitted. The introduction of CBTC north of Harrow does not affect Chiltern trains in terms of general operation - the drivers run to colour light signals and have to have operational tripcock equipment. The current and any subsequent fleets are not and will not be CBTC fitted.

It has taken this long to get CBTC and there are no plans in the long foreseeable future for any additional systems to work alongside it (eg ETCS etc.)

 

[A60stock]

Is there some sort of TSR or something just outside Wembley Park southbound? All fast trains I have been on have either stopped in the platform or performed a sharp break just after passing through

 

[penfold22]

Unfortunately we have several TSRs across the line, but nothing in that area that would cause that. Given you mention running on the southbound fast I would presume they have been trains in the morning peak? The number of trains operated southbound especially between 0730-0900 can still cause some congestion even with CBTC in charge. Occasionally, the system does get a little too enthusiastic and in its efforts to recover time and get trains as close together as designed, it can catch itself out and as a result certain trains may be slowed at the last minute. Nothing unsafe in what it is doing - in fact the reverse, in that it is being sometimes over-cautious.

 

[A60stock]

Yes, morning peak.

That makes sense, I thought maybe the driver applied the break as a precaution or something!

 

[Snow1964]

An update in TfL Board Programmes and Investment Committee papers

5.2 Once SMA9 goes live, there will be just two SMAs (SMA13 and 14 covering the northern branches of the Metropolitan line) left to go live to complete the full roll out of the new signalling system. These remaining Metropolitan line areas are however complex and present unique challenges as they require novel system functionality and significant operational readiness training to integrate the new signalling system with manually driven Chiltern trains and the Piccadilly line

5.4 During 2027 software development and system testing for the remaining SMAs will be undertaken alongside work to optimise the existing software with SMA13 commissioning then following in the first half of 2028 (rather than the second half of 2027 as reported to the Committee in July 2026). Our forecast completion date for the final SMA (SMA14 Rayners Lane to Uxbridge) is expected to be in the second half of 2028. Programme close-out would follow in 2029 (once required software optimisations have been completed).

https://board.tfl.gov.uk/documents/s27207/pic-20261007-item09-part%201-4LM.pdf

 

[Edvid]

I read a Modern Railways article on this project when I was a teen. Since then I've more than doubled in years and it's still not finished

Fingers crossed that commissioning of SMA9 (and Chiltern Railways formations on part of that stretch afterward) stays on track for early November.

 

[A60stock]

I really cannot wait to see the S8s (and hoping they do), thrash up and down the fast lines between Harrow and Wembley.

On which note, the sluggish journey between Wembley and Kilburn has still not been sorted!

 

[swt_passenger]

What’s the eventual solution to SMA11 being abandoned? Include in a future Piccadilly Line signalling scheme?

 

[bramling]

What’s the eventual solution to SMA11 being abandoned? Include in a future Piccadilly Line signalling scheme?

It’s essentially been kicked into the long grass, but the vague idea is to look at the whole thing combined.

Piccadilly resignalling is a big problem, as there’s now the problem of who supplies is, and in particular whether compatibility with the Seltrac system means that one supplier has an inherent advantage and can essentially name their price, which clearly isn’t an ideal position for LU to be in. Unfortunately this is another long-term consequence of the whole SSR resignalling debacle.

It’s already the case that the current supplier is very much “you know we are the only game in town, sorry about all the bugs with our system, throw us a list of all the issues over to Canada and we may or may not have a look at them”.

 

[swt_passenger]

It’s essentially been kicked into the long grass, but the vague idea is to look at the whole thing combined.

Piccadilly resignalling is a big problem, as there’s now the problem of who supplies is, and in particular whether compatibility with the Seltrac system means that one supplier has an inherent advantage and can essentially name their price, which clearly isn’t an ideal position for LU to be in. Unfortunately this is another long-term consequence of the whole SSR resignalling debacle.

It’s already the case that the current supplier is very much “you know we are the only game in town, sorry about all the bugs with our system, throw us a list of all the issues over to Canada and we may or may not have a look at them”.

Interesting, if I’d included a guess that’s probably very close to what I was thinking. Am I correct in thinking the final legs of the District to Wimbledon and Putney were never included though?

 

[OscarH]

Fingers crossed that commissioning of SMA9 (and Chiltern Railways formations on part of that stretch afterward) stays on track for early November.

Delayed again I'm afraid

 

[bluegoblin7]

It’s essentially been kicked into the long grass, but the vague idea is to look at the whole thing combined.

Piccadilly resignalling is a big problem, as there’s now the problem of who supplies is, and in particular whether compatibility with the Seltrac system means that one supplier has an inherent advantage and can essentially name their price, which clearly isn’t an ideal position for LU to be in. Unfortunately this is another long-term consequence of the whole SSR resignalling debacle.

It’s already the case that the current supplier is very much “you know we are the only game in town, sorry about all the bugs with our system, throw us a list of all the issues over to Canada and we may or may not have a look at them”.

Without drifting the thread too much, I am aware of some novel ideas to solve this problem, which are both viable and could deliver something far more suited to LUL’s operations than a bodged people mover.

 

[Edvid]

Delayed again I'm afraid

View attachment 212434

Sigh...

Good thing I didn't label 2027 as another SMA gap year!

 

[bluegoblin7]

Interesting, if I’d included a guess that’s probably very close to what I was thinking. Am I correct in thinking the final legs of the District to Wimbledon and Putney were never included though?

East Putney to Wimbledon and Stamford Brook to Richmond/Ealing Broadway were originally in scope through as SMAs 10 (Barons Court-Richmond/Chiswick Park), 11 (Chiswick Park-Ealing) and 12 (West Brompton-Wimbledon).

Obviously 11 has been abandoned entirely and a reduced 10/12 is now in operation.

 

[Recessio]

It’s essentially been kicked into the long grass, but the vague idea is to look at the whole thing combined.

Piccadilly resignalling is a big problem, as there’s now the problem of who supplies is, and in particular whether compatibility with the Seltrac system means that one supplier has an inherent advantage and can essentially name their price, which clearly isn’t an ideal position for LU to be in. Unfortunately this is another long-term consequence of the whole SSR resignalling debacle.

It’s already the case that the current supplier is very much “you know we are the only game in town, sorry about all the bugs with our system, throw us a list of all the issues over to Canada and we may or may not have a look at them”.

Is Seltrac a proprietary system, and if so does that mean it's not really (or easily) possible for a different supplier to build a system that is compatible then?

 

[Metrider]

Delayed again I'm afraid

View attachment 212434

Is this likely to be pushed back to next year now? After what happened to SMA11 I just hope it goes ahead at all.

 

[Edvid]

After what happened to SMA11 I just hope it goes ahead at all.

AIUI most of the SMA9 prep is done and the Met legacy signalling is very old (far beyond life-expired by modern standards), so from a layman's POV I still think it'll be commissioned at some point.

 

[bluegoblin7]

There’s no danger of 9 being cancelled - 11 was for somewhat specific reasons around inter compatibility and other upgrades that negated some of the benefits.

That said, I wouldn’t necessarily call the existing signalling “far beyond life expired”. I don’t deny that the Met line cabins are creaking (and I probably know this better than anyone else on this forum ) but the beauty of the mechanical, legacy signalling is that it can pretty much be refurbished and keep running forever. The biggest issues are with wire degradation rather than anything else, and there would need to be some significant interventions there if the programme slips much more.

The main reason for 4LM is around capacity increases and having one single system, frankly.

 

[bramling]

Is Seltrac a proprietary system, and if so does that mean it's not really (or easily) possible for a different supplier to build a system that is compatible then?

I’d say it’s more that LU has become a notoriously difficult client. It may well be less of an issue now, but what happened in the 2010s was the Infracos folded, much of what experience that did exist was lost (especially on the contractual side of things, less so in terms of raw engineering). The direct result of this was ditching the DTG-R system, the disastrous flirtation with City Flo, and ending up in the situation where Thales was essentially the only viable supplier.

This isn’t the only consequence of all this. If we look at the Jubilee and Northern lines, the system as it now is feels somewhat unfinished. It, give or take, meets the spec, but there remain a lot of issues, and properly addressing these seems to have pretty much evaporated. The Victoria Line seems to have managed to avoid this, feels like there was a much healthier supplier/client relationship on that project.

== Doublepost prevention - post automatically merged: 30 Sep 2026 ==

There’s no danger of 9 being cancelled - 11 was for somewhat specific reasons around inter compatibility and other upgrades that negated some of the benefits.

That said, I wouldn’t necessarily call the existing signalling “far beyond life expired”. I don’t deny that the Met line cabins are creaking (and I probably know this better than anyone else on this forum ) but the beauty of the mechanical, legacy signalling is that it can pretty much be refurbished and keep running forever. The biggest issues are with wire degradation rather than anything else, and there would need to be some significant interventions there if the programme slips much more.

The main reason for 4LM is around capacity increases and having one single system, frankly.

ISTR descoping the north end of the Met Line was rejected for exactly the reason that the cabling was completely life-expired (and that was a few years ago), and it wasn’t worthwhile to have to spend out on life-extending all that.

I’m less sure where the Uxbridge branch fits in to this, as the signalling west of West Harrow is much more recent. Probably standardisation was the main reason here, otherwise it would be a nuisance to have to revisit this in the future.

 

[A60stock]

Do we have an estimation for when SMA9 will go live? Typical, what was the reason for the delay?

 

[GFE]

Do we have an estimation for when SMA9 will go live? Typical, what was the reason for the delay?

Something might be asked/said in the update/questions at next weeks programmes & investment panel advisory/scrutiny meeting.
The report to that meeting already said there were a number of underlying issues on the current live CBTC system which are planned to be addressed before going beyond SMA9 commissioning

== Doublepost prevention - post automatically merged: 2 Oct 2026 ==

Must admit their case is quite well made and they seem informed.

I’ve never really understood why this section seems so problematic. 1980s signalling isn’t excessively old, and it’s a fairly simple section of line.

Looks like the Wimbledon branch campaign are exploring various avenues to build their case/pressure for redirecting the savings from descoping the "CBTC overlay works" into other reliability/service improvement works. Perhaps the money has already been diverted to the issues referenced in my previous post which will benefit a wider clientele.

FOI request detail

FOI request detail

tfl.gov.uk

 

[Backroom_boy]

Looks like the Wimbledon branch campaign are exploring various avenues to build their case/pressure for redirecting the savings from descoping the "CBTC overlay works" into other reliability/service improvement works. Perhaps the money has already been diverted to the issues referenced in my previous post which will benefit a wider clientele.

Almost certainly any money saved will have been spent on the overruns on the still in scope sections

 

[bramling]

Something might be asked/said in the update/questions at next weeks programmes & investment panel advisory/scrutiny meeting.
The report to that meeting already said there were a number of underlying issues on the current live CBTC system which are planned to be addressed before going beyond SMA9 commissioning

== Doublepost prevention - post automatically merged: 2 Oct 2026 ==


Looks like the Wimbledon branch campaign are exploring various avenues to build their case/pressure for redirecting the savings from descoping the "CBTC overlay works" into other reliability/service improvement works. Perhaps the money has already been diverted to the issues referenced in my previous post which will benefit a wider clientele.

FOI request detail

FOI request detail

tfl.gov.uk

I have some sympathy for their viewpoint. However it will just be swept under the carpet like the bottom end of the Wimbledon branch tends to have been for very many years. For whatever reason it’s always ended up in the “too difficult” pile.

It’s very easy for LU to just chuck the service round at Parsons Green which means it doesn’t wreck the rest of the District Line too badly, and just regard any issue as down to Network Rail.

Same with Richmond, just divert to Ealing Broadway and forget about it. Been like this for years.

 

[penfold22]

Do we have an estimation for when SMA9 will go live? Typical, what was the reason for the delay?

No further estimate - though it will be 2027 at the earliest (not difficult to assume given we are in the closing weeks of 2026).
When you weigh up the closure it requires - both early on the Saturday to the H&C, Circle, District and Jubilee - but also the entire weekend on the Met Line, and then look over other events which surround the line. For example, anything at Wembley Stadium will be a big no for a closure. Other already scheduled engineering work may not make it possible to have this on top as well.

Without putting my neck on the line too much, let's just say some equipment was found not to work as it should, and may mean a not insignificant software update to fix it.
"Luckily" this issue was found in recent days and before the go-live date (which was, of course, originally meant to have been in August, which was a date some weeks before this issue was discovered...)