First ETCS train operates on the East Coast Mainline

[bramling]

I’ve never heard of redundant signals being reused, except rarities on preserved railways, and some components into spares (ef certain tyoes of signal heads).

To put it in context, a resignalling project with 200 new signals might cost £200m pounds, but the materials cost of the actual signals (post and signal head) will be, at the very most, 1% of that. Driver briefing and training will cost more.

LT certainly used to re-use old signals during resignalling projects. If you look at something like the Bakerloo, resignalled in the late 80s, most of the signal heads are much older than that. Same applies on a lot of the Piccadilly Line.

I’m not sure of the extent to which signals were retained in situ where practicable versus parts going into store, being refurbished, and installed in new schemes.

This generally stopped in the 1990s, the Central Line used new signal heads, although many old ones were retained as part of the interim stage works.

I think I also remember that the rushed interim signalling scheme on the Jubilee Line Extension re-used old train stops.

Presumably a lot removed during the various PPP-era line upgrade schemes has gone in the bin, though amazingly there’s still a few old signal heads left redundant in situ for over a decade now. Quite poor really, as it’s asking for problems if any part of the structure corrodes.

 

[zwk500]

Agreed that if you design the system from the outset as a conventional signalling overlay on an underlying ETCS signalling system, then it will be easier to recover the signals. But the costs will still be significant.

Where there are signals, the ETCS cannot work independently of the signals. The driver must not get conflicting information from the ETCS and the lineside signalling. So either the ETCS movement authorities must be synchronised with what the lineside signals are showing, or the lineside signals must be synchronised to the ETCS (such as by displaying a special "ETCS aspect"). It follows that when you recover the signals, you have not only got the costs of physically recovering the signals themselves, and their associated hardware (object controllers, equipment cases, cabling alterations, etc), but you will have significant software changes to make to the underlying control systems as well. Designing, implementing and testing these kind of alterations doesn't come cheap, and I doubt that the costs would ever be recouped from the savings in maintenance costs. After all, signals don't need much maintaining.

And you need to cost in the disruption to services caused while you make all these alterations.

And they involve the kind of software changes that only the original system supplier will likely be capable of doing. If they are using their limited resources for such recoveries, they won't be able to use them for other resignalling work, further ETCS roll-out, etc.

The transition from Overlay to Signals-Away requires all these changes whether the signals are physically retained or not, doesn't it?

Given one of the justifications for Signals-Away is the reduction in maintenance benefit, I could see a recovery programme of assets in diruptive positions (i.e. 10-foots or on Gantries) after Signals are switched off. But as you note, the disruption of recovering them will mean they likely won't be gone in the same possession they are switched off.

Also all the reasons you mentioned above are why NR are keen to phase ETCS in so that after a 'pathfinder' area the adjacent schemes can go straight from no ETCS to Signals-Away.

I think such recoveries are a nice idea in theory, but I really can't see them ever being justified in practice.

Recovery was justified on the NCL for at least some signals. Not sure if they've left the ones in the tunnel but they did at least one for the publicity shots. However Heathrow branch has notably kept it's signals and GW-ATP despite all trains being ETCS on the branch so I suspect there'll be a case-by-case assessment for recovery.

 

[bahnause]

3.6.0 is the baseline that will be used for the forseeable future.

We should gradually start to abandon the idea, that the end customer has much of a choice in this matter. At the end of the day, it is the industry that dictates the standards for ETCS. The BL 3.6.0 in particular has glaring shortcomings in its specification, which can only be rectified by using BL4. The Stadler DuFour for SBB Cargo is one such example where BL4 will already be used.

 

[St. Paddy]

Recovery was justified on the NCL for at least some signals. Not sure if they've left the ones in the tunnel but they did at least one for the publicity shots. However Heathrow branch has notably kept it's signals and GW-ATP despite all trains being ETCS on the branch so I suspect there'll be a case-by-case assessment for recovery.

All signals were removed in the tunnels on the NCL

 

[zwk500]

All signals were removed in the tunnels on the NCL

Thanks for confirming, thought it was the case but wasn't sure!

 

[Dazza12]

Hi,

We should gradually start to abandon the idea, that the end customer has much of a choice in this matter. At the end of the day, it is the industry that dictates the standards for ETCS. The BL 3.6.0 in particular has glaring shortcomings in its specification, which can only be rectified by using BL4. The Stadler DuFour for SBB Cargo is one such example where BL4 will already be used.

More to the point, the law (separately in both the UK and Europe) requires that baseline 4 is used for new installations (subject to a transition period).

Of course, it is the industry which comes up with the specifications (via UNISIG and the ERA) which are then enshrined in law, but as the specification for ETCS stabilises they will have a harder and harder time convincing the regulators that mandating the latest version is value for money. The next big thing is standardising ATO over ETCS but who knows whether anyone wants to buy it...

 

[Belperpete]

The transition from Overlay to Signals-Away requires all these changes whether the signals are physically retained or not, doesn't it?

Given one of the justifications for Signals-Away is the reduction in maintenance benefit, I could see a recovery programme of assets in diruptive positions (i.e. 10-foots or on Gantries) after Signals are switched off. But as you note, the disruption of recovering them will mean they likely won't be gone in the same possession they are switched off.

Also all the reasons you mentioned above are why NR are keen to phase ETCS in so that after a 'pathfinder' area the adjacent schemes can go straight from no ETCS to Signals-Away.

Recovery was justified on the NCL for at least some signals. Not sure if they've left the ones in the tunnel but they did at least one for the publicity shots. However Heathrow branch has notably kept it's signals and GW-ATP despite all trains being ETCS on the branch so I suspect there'll be a case-by-case assessment for recovery.

If you just switch off the signals, the signaller will get a lot of "signal failed" alarms. Let alone the impact on the signalling controls in the interlocking previously mentioned. The central interlocking needs reprogramming to suit the revised configuration. Which requires skilled data preparers and testers to implement. Resourcing and paying for such work is going to be difficult when there are likely to be much higher priorities.

If you are just going to switch off one or two problematic signals, that will still need the central interlocking reprogrammed. It is likely that the overheads for such a small alteration would far outweigh the costs of the work itself. I suspect that it would probably be cheaper and easier to just renew the affected signals.

A case in point was the old signalbox at Waterloo. At some stage, automatic mid platform signals were provided. However, they were not a success. Because it was easier and cheaper to keep maintaining those signals, than the work required to alter the interlocking to recover them, the signals were simply moved into the relay room! And there those signals stayed, quietly working away hidden in the relay room.

Agreed that the signals were recovered on NCL, but this was done as an inherent part of the initial project. The problem will come on overlay and dual signalled areas that have been up and running for say a decade or so, and a separate project is then required for switching off and recovery of the signals. I think it will be impossible to justify.

It would be interesting to know if the signal recoveries on NCL had a separate budget that had to be justified, or the work was just included in the overall project cost.

I also suspect that recovery of the signals on NCL was done as much for strategic reasons as financial.