E.R.T.M.S.

[Edwin_m]

They were talking about keeping signals north of Peterborough because of the number of trains that only use this section and could avoid having ETCS fitted (including the EMT 158 fleet).  Either this has changed or the linked article is wrong in saying there would be no more overlays.  

[ess1uk]

52 minutes ago, Edwin_m said:

They were talking about keeping signals north of Peterborough because of the number of trains that only use this section and could avoid having ETCS fitted (including the EMT 158 fleet).  Either this has changed or the linked article is wrong in saying there would be no more overlays.  

At present the Doncaster PSB takes over at Stoke Tunnel (south of Grantham) so that is the cut off north of Peterborough 

[Edwin_m]

8 hours ago, ess1uk said:

At present the Doncaster PSB takes over at Stoke Tunnel (south of Grantham) so that is the cut off north of Peterborough 

Yes indeed, and I believe this is also the northern limit of the ECML ERTMS fitment so far agreed.  But between Stoke and Peterborough, as well as the EMR service I mentioned, there is also some freight that turns off towards Ely - and there is even more that goes via Spalding.    

[ess1uk]

15 hours ago, Edwin_m said:

Yes indeed, and I believe this is also the northern limit of the ECML ERTMS fitment so far agreed.  But between Stoke and Peterborough, as well as the EMR service I mentioned, there is also some freight that turns off towards Ely - and there is even more that goes via Spalding.    

Also Cross country to Stamford and Leicester turns off at Helpston so would need trains fitted to run with ETCS around the Peterborough area

[DY444]

Hmm so there are potentially going to be sections aside from the first trial section around Welwyn with conventional signals.  Remind me again what the actual financial benefits of this are supposed to be. 

 

The indisputable gains are ATP for fitted trains, "free" bang road operation for fitted trains and a theoretical capacity increase for fitted trains which may or may not be realised in actual operation.  You have a potential saving from not renewing signals, gantries and brackets on sections with no signals.  You still have most of the other costs such as interlockings, point machines, loc cases, cabling etc and the additional enormous cost of fitting cabs with the equipment.  I wondered how long it would be before the signals away idea started to be compromised by the cost of the latter. 

 

The other issue is that with no signals on some sections what we're going to end up with is sections of railway that cannot be used by a proportion of the train fleet.  So for example it may no longer be possible to divert the Scottish sleepers via the ECML without an almighty faff involving locomotive changes or piloting using fitted locomotives.  That is not consistent with reduced costs. 

 

There are perfectly valid reasons for doing this but I'm just not seeing the widely parroted cost reduction being one.   I suspect any cost reductions there are will be to NR but the overall costs to the industry will actually be higher than they would have been if this wasn't done. 

Edited May 13 by DY444

[david.hill64]

1 hour ago, DY444 said:

There are perfectly valid reasons for doing this but I'm just not seeing the widely parroted cost reduction being one.   I suspect any cost reductions there are will be to NR but the overall costs to the industry will actually be higher than they would have been if this wasn't done. 

Unless the rules have changed, this will be covered by the 'Network Change' clauses in the access agreements, the costs of which will be covered by Network Rail.

 

Under EU law, the relevant section of which may or may not have been repealed by the UK, it is mandatory to use ETCS when resignalling an existing route or opening a new route on the mainline railway, which constitutes a valid reason for doing it if still mandatory! I think that ETCS level 2 with back up axle counters will not save much money. I am not aware of any mainline railway that has been brave enough to do away with secondary detection, and only one MRT (Bangkok Skytrain) that has. If someone knows different, I would like to hear.

[The Stationmaster]

8 hours ago, david.hill64 said:

Unless the rules have changed, this will be covered by the 'Network Change' clauses in the access agreements, the costs of which will be covered by Network Rail.

 

Under EU law, the relevant section of which may or may not have been repealed by the UK, it is mandatory to use ETCS when resignalling an existing route or opening a new route on the mainline railway, which constitutes a valid reason for doing it if still mandatory! I think that ETCS level 2 with back up axle counters will not save much money. I am not aware of any mainline railway that has been brave enough to do away with secondary detection, and only one MRT (Bangkok Skytrain) that has. If someone knows different, I would like to hear.

Unless things a have changed (they might have now the legal trade are involved) Network Change requires the agreement of train operators although the idea that such agreement 'can be bought' may well fit with various perceptions of the procedure

 

Can see teh big advamtahe of ATP to, i presume, a greater and more effective level than TPWS (although it s in any case pretty good) but I simply don't buty the idea that it ai utomatically increase capacity - that can only happen with a common fleet of trains running precisely the same service pattern and even then would probably need moving block to produce any real capacity benefit.  Freight trains on the ECML aren't going to run any differently with ETCS or whatever and Class 8XX trains will still perform very differently from the units operating local stopping, or semi-fast, services.  you might occasionallhy be able to squeeze headways but altering margins won't be very easy uness trains speeds etc change.

 

Incidentally has ETCS been sufficiently tested in UK conditions to assume consistent braking distances on British freight trains?  The original ATP tests on the WR worked beautiully with HSTs stopping very consistently on the predicted braking curve.  But the freight tests were a very different matter and I don't think any two tests nmanaged to produce the same result and none of them matched the computed braking curves.

[DY444]

On 13/05/2024 at 08:51, david.hill64 said:

Unless the rules have changed, this will be covered by the 'Network Change' clauses in the access agreements, the costs of which will be covered by Network Rail.

 

Under EU law, the relevant section of which may or may not have been repealed by the UK, it is mandatory to use ETCS when resignalling an existing route or opening a new route on the mainline railway, which constitutes a valid reason for doing it if still mandatory! I think that ETCS level 2 with back up axle counters will not save much money. I am not aware of any mainline railway that has been brave enough to do away with secondary detection, and only one MRT (Bangkok Skytrain) that has. If someone knows different, I would like to hear.

 

That law clearly isn't worth the paper it's written on because several resignalling schemes have been carried out in recent years where lights on sticks have been replaced by new lights on sticks or semaphores have been replaced with lights on sticks.  Sounds like the sort of law NR or whoever can cite to justify ETCS when it wants to but which can be totally ignored otherwise. 

 

Whatever the merits of it may be, the financial case doesn't add up at all.  As has been said you save the cost of signals, their structures, cabling, and control modules albeit in what appears to be a slowly diminishing number of situations due to the belated realisation that fitting cabs is expensive.  Presumably LED signals need little in the way of maintenance relative to those using conventional lamps so little will be saved there or else why has NR spent a fortune replacing just about every signal head in the country?   Points will still fail, axle counters will still get confused, lineside modules will still have nervous breakdowns, cables will still get eaten by rodents or nicked.  The argument that fitting the fleet is a one off cost doesn't stack up either, for example the 700s have just had to have a very expensive ETCS upgrade to make them compatible with what's being installed on the ECML.  I expect that to become a regular feature and ongoing cost.

 

Another hidden cost is diminished fleet inter-changeability.  I've already mentioned the likely compromising of diversion options and things like GN borrowing a couple of GX 387/2s to cover a short term lack of availability isn't going to be an option any more.  I know GN are getting the 379s which may make that specific example a moot point but in general moving stock around in future will be more difficult and costly.  Even when stock is fitted we may get a 700 type situation where one route is on version 97 and the stock is on version 65 or whatever.  We're already seeing stock with potentially useful life being scrapped, the requirement for ETCS on some routes and the cost of fitment or upgrade will just make that more likely.

Edited May 15 by DY444

[icn]

On 13/05/2024 at 08:34, DY444 said:

Hmm so there are potentially going to be sections aside from the first trial section around Welwyn with conventional signals.  Remind me again what the actual financial benefits of this are supposed to be. 

 

The indisputable gains are ATP for fitted trains, "free" bang road operation for fitted trains and a theoretical capacity increase for fitted trains which may or may not be realised in actual operation.  You have a potential saving from not renewing signals, gantries and brackets on sections with no signals.  You still have most of the other costs such as interlockings, point machines, loc cases, cabling etc and the additional enormous cost of fitting cabs with the equipment.  I wondered how long it would be before the signals away idea started to be compromised by the cost of the latter. 

 

The other issue is that with no signals on some sections what we're going to end up with is sections of railway that cannot be used by a proportion of the train fleet.  So for example it may no longer be possible to divert the Scottish sleepers via the ECML without an almighty faff involving locomotive changes or piloting using fitted locomotives.  That is not consistent with reduced costs. 

 

There are perfectly valid reasons for doing this but I'm just not seeing the widely parroted cost reduction being one.   I suspect any cost reductions there are will be to NR but the overall costs to the industry will actually be higher than they would have been if this wasn't done. 

Your post assumes that conventional signalling and existing equipment (AWS/TPWS/etc) is free. It's not. The transition certainly leads to higher costs in the short-term, but once you make the switch you have reduced costs: for example new trains in Europe are already all ETCS compatible - therefore, new trains for Switzerland are now cheaper because they don't have to bother with adapting and installing legacy signalling equipment in them (in fact they're not allowed to, new trains must be certified with ETCS for Switzerland). Moreover you don't have to pay inflated prices for a safety system that is only used in one country, instead you have a bigger market, more suppliers, and presumably lower prices.

 

Once could argue that the UK needs different trains and won't benefit from European standardisation due to gauging anyway - but that's also not relevant for signalling discussions: the underlying software and electric systems are going to be reused between contintental and British trains.

 

Focusing on signal removal is missing the forest for the trees.

[The Stationmaster]

2 hours ago, icn said:

Your post assumes that conventional signalling and existing equipment (AWS/TPWS/etc) is free. It's not. The transition certainly leads to higher costs in the short-term, but once you make the switch you have reduced costs: for example new trains in Europe are already all ETCS compatible - therefore, new trains for Switzerland are now cheaper because they don't have to bother with adapting and installing legacy signalling equipment in them (in fact they're not allowed to, new trains must be certified with ETCS for Switzerland). Moreover you don't have to pay inflated prices for a safety system that is only used in one country, instead you have a bigger market, more suppliers, and presumably lower prices.

 

Once could argue that the UK needs different trains and won't benefit from European standardisation due to gauging anyway - but that's also not relevant for signalling discussions: the underlying software and electric systems are going to be reused between contintental and British trains.

 

Focusing on signal removal is missing the forest for the trees.

So if I read that correctly all signalling in Switzerland has now been converted to ETCS and there is no need to fit to trains anything required to handle earlier signalling systems.  That was quick!!

 

And while the story about new trains makes sense what about umpteen various older trains which require installation of ETCS kit?    That's a different design and fitment task for each type of train requiring lots of work and means, quite likely, trying to shoehorn the equipment into very limited space because the trains still have to be able to operate in areas not yet converted to ETCS.

 

Changing to any new system involving safety certified electronics isn't quite on the same level as the GWR converting miles of railway to a different gauge over a weekend. (and even then authorities were subsequently sought for overspends on signalling alterations which had not originally been thought necessary). I'm not against improvements, or wider commonality of equipment, especially where it genuinely improves on existing safety systems.  But some of things being claimed for ETCS fitment on NR sit rather oddly with the nature of traffic, and unavoidable variations in train performance on the routes where it is being installed.  Simple fact - fitting ETCS does not suddenly give every sort of train using a route exactly the same performance;  the normal laws of physics still apply. (as a past signal engineer friend and colleague used to say).