"Digitising” the signalling system

[TheSignalEngineer]

Other problems arise because TVM relies on the wheel-rail interface to send information to the train; a rigorous regime of de-rusting has to be carried out on a (preferably) daily basis. This includes all points and crossovers; it's manageable on a 'closed' system like the one I work on, but much more difficult to apply on a long mainline, or a station with multiple routeing possibilities. If the link twixt train and track  is broken more than momentarily, then the train has an emergency brake application applied automatically, whilst the pantograph also drops.

 

How times change. It used to be mandatory for new apprentices at the CE depot at Gillingham, to be told to go to the stores to ask for the Rouillegelais paint, so that they could paint the rust on to the tracks......

I remember about 50 years ago we had two S&T Probationers cleaning the rail tops with wire brushes after we 'lost' an 08 in the middle of Crewe North Junction. 

[D1059]

I remember about 50 years ago we had two S&T Probationers cleaning the rail tops with wire brushes after we 'lost' an 08 in the middle of Crewe North Junction. 

 

No one thought about getting hold of a giant Peco Track Rubber ? 

Edited May 31, 2018 by D1059

[DY444]

Reports that the ETCS test unit 313121 has gone to Eastleigh for storage so it appears that its role is over at least for the moment.  Various rumours doing the rounds as to why this is ranging from lack of line capacity to allow continued routine use of the Hertford loop test site through to the entire ECTS programme being deferred/scaled back/abandoned.

Edited May 31, 2018 by DY444

[Christopher125]

IIRC it was being used for testing ETCS through the Thameslink core, so has presumably served it's purpose.

[Peter Crawford]

GTR now has ETCS, but it's too little too late. If they can fix up the system, that would be great.

[adanapress]

If you stand back and take a long term back view,   clearly the movement is towards vastly fewer signal 'boxes'   proceeding forwards to just a handful of signalling centres,

operated by a tiny handful of vastly specialised staff.    Now look at the French Air Controllers situation and regular actions,   and one can begin to see a potential future problem.

Dont ever say that the duck never quacked ........

[DY444]

If you stand back and take a long term back view,   clearly the movement is towards vastly fewer signal 'boxes'   proceeding forwards to just a handful of signalling centres,

operated by a tiny handful of vastly specialised staff.    Now look at the French Air Controllers situation and regular actions,   and one can begin to see a potential future problem.

Dont ever say that the duck never quacked ........

 

I don't see that as a real problem.  Rightly or wrongly the trend these days is for managers to be "invited" to perform "strike breaking" roles during disputes.  That, in principle anyway, is much easier to organise in a big signalling centre than in geographically distributed smaller locations. 

 

A much bigger risk is h/w, s/w, power, comms failures or other single building related incidents.  The chaos which ensued when Three Bridges box was evacuated after the fire alarm went off springs to mind.

[St. Simon]

If you stand back and take a long term back view, clearly the movement is towards vastly fewer signal 'boxes' proceeding forwards to just a handful of signalling centres,

operated by a tiny handful of vastly specialised staff.

There will still be lots, possible several hundred signal boxes for a long time to come, and even when those get closed, there will still be quite a few signalling centres available...

 

Fire Alarms, Power Failures, Comms failures etc. can have a huge effect on the railway even when there are several signal boxes, because of the nature of a signalling system, I personally don't see a much bigger risk with such things when more areas are put into a single box.

 

Of course people forget that each workstation is in effect a signal box, it's just shares a building with several others, so each has a separate power supply, a seperate comma system, so effects from problems with one workstation are reduced.

 

Simon

Edited June 6, 2018 by St. Simon

[phil-b259]

There will still be lots, possible several hundred signal boxes for a long time to come, and even when those get closed, there will still be quite a few signalling centres available...

 

Fire Alarms, Power Failures, Comms failures etc. can have a huge effect on the railway even when there are several signal boxes, because of the nature of a signalling system, I personally don't see a much bigger risk with such things when more areas are put into a single box.

 

Of course people forget that each workstation is in effect a signal box, it's just shares a building with several others, so each has a separate power supply, a seperate comma system, so effects from problems with one workstation are reduced.

 

Simon

 

Quite so.

 

Its a point lost on most people (particularly the mandarins in Whitehall) that the signallers working Panel 2 and panel 3 at Three Bridges are no different to the signallers working Blea Moor and Garsdale mechanical boxes. None can work the adjacent panel / box unless they have been specifically trained and are competent t do so* regardless of whether they are 8 feet apart or 8 miles apart.

 

 

*If you do not work a panel / box for 6 months then your competency lapses and retraining is needed.

[ess1uk]

Already happening at Feltham I believe

[ess1uk]

https://www.railengineer.uk/2018/09/03/digital-delusion-a-lesson-from-not-so-long-ago/

Tried before

[david.hill64]

https://www.railengineer.uk/2018/09/03/digital-delusion-a-lesson-from-not-so-long-ago/

Tried before

Should be compulsory reading for anyone who wants to understand uk rail.

[Junctionmad]

Should be compulsory reading for anyone who wants to understand uk rail.

I think the lessons learnt from WCML and moving block are clear ,but they are “of their time “, things move on.

[jim.snowdon]

I have to say I'm sceptical about NR's claim that the digital railway will actually unlock significant amounts of capacity. Network Rail is not a metro system with a single type of rolling stock and a simple track layout. Whether NR's head of Digital Railway fully appreciates this is, at least to me, a moot point, given that his past experience is largely, if not entirely, with metro systems.

 

I cannot help getting the feeling that ERTMS has been seized upon as a means of avoiding facing up to the infrastructure changes that would be necessary.

 

Jim

[Mike Storey]

I have to say I'm sceptical about NR's claim that the digital railway will actually unlock significant amounts of capacity. Network Rail is not a metro system with a single type of rolling stock and a simple track layout. Whether NR's head of Digital Railway fully appreciates this is, at least to me, a moot point, given that his past experience is largely, if not entirely, with metro systems.

 

I cannot help getting the feeling that ERTMS has been seized upon as a means of avoiding facing up to the infrastructure changes that would be necessary.

 

Jim

 

I think we will find that the new CEO of NR will be just as sceptical, given his experience as MD of SW Trains (and of the promise of similar made to the CAA, while he was CEO there, by NATS - that has had considerable problems, although it has enabled significant efficiencies).

 

I worked alongside Andrew when we were both Account Executives in Railtrack. He always had pretty leading edge notions, and most of us told him to shut the bleep up when he went on and on about them at our monthly meetings. But look who did appreciably better than almost all of us (only Robin Gisby, Account Executive for Freight at the time, was the other high-flyer in our group) in their career.....

[St. Simon]

I have to say I'm sceptical about NR's claim that the digital railway will actually unlock significant amounts of capacity. Network Rail is not a metro system with a single type of rolling stock and a simple track layout. Whether NR's head of Digital Railway fully appreciates this is, at least to me, a moot point, given that his past experience is largely, if not entirely, with metro systems.

I cannot help getting the feeling that ERTMS has been seized upon as a means of avoiding facing up to the infrastructure changes that would be necessary.

Jim

Whilst the MD of Digital Railway is originally from LUL, the people that are looking after ETCS are highly experienced and highly knowledgable heavy rail professionals.

 

Simon

Edited September 9, 2018 by St. Simon

[TheSignalEngineer]

I think we will find that the new CEO of NR will be just as sceptical, given his experience as MD of SW Trains (and of the promise of similar made to the CAA, while he was CEO there, by NATS - that has had considerable problems, although it has enabled significant efficiencies).

 

I worked alongside Andrew when we were both Account Executives in Railtrack. He always had pretty leading edge notions, and most of us told him to shut the bleep up when he went on and on about them at our monthly meetings. But look who did appreciably better than almost all of us (only Robin Gisby, Account Executive for Freight at the time, was the other high-flyer in our group) in their career.....

We had a saying in our office that those who could did, those who couldn't got promoted.

[TheSignalEngineer]

I think the lessons learnt from WCML and moving block are clear ,but they are “of their time “, things move on.

One of the big lessons in signalling from West Coast was actually one forgotten from BR days. That was not to include development projects in an infrastructure project with a fixed end date

Whatever the experts tell you about their brilliant new system they will still be trying to get it to work when they have been at it for three times as long as they originally forecast.

[Dunsignalling]

We had a saying in our office that those who could did, those who couldn't got promoted.

Cynical, but proven to be accurate uncomfortably often. Also by no means confined to the rail industry.....

 

Another one I encountered in a previous career segment was that a good way to get on was being able to play golf almost as well as the boss.

 

John

Edited September 9, 2018 by Dunsignalling

[jim.snowdon]

Whilst the MD of Digital Railway is originally from LUL, the people that are looking after ETCS are highly experienced and highly knowledgable heavy rail professionals.

Simon

That, I do not doubt for one moment. In my earlier years in the railway industry, I had a saying - "we, in the boiler room, kept the ship going - where the ship went was up to them on the bridge". The engineers will, ultimately, be able to make ERTMS work; whether it will deliver as much benefit as management want to believe is another matter.

 

Jim

[david.hill64]

That, I do not doubt for one moment. In my earlier years in the railway industry, I had a saying - "we, in the boiler room, kept the ship going - where the ship went was up to them on the bridge". The engineers will, ultimately, be able to make ERTMS work; whether it will deliver as much benefit as management want to believe is another matter.

 

Jim

Jim,

 

I had much the same concerns on capacity, based on the following couple of thoughts.

 

Firstly, in a well designed 4 aspect conventional lineside signalling system, the distance between signals for 100/125mph operation on level track is set at 1020 metres. This gives a braking distance of 2040m for a train from passing the double yellow to stopping with the buffers level with the signal post. This was based on a braking distance of 1740m for a 7%g braked 100mph train or a 9%g braked 125mph high speed train. 300m was a margin included to allow for the inevitable variability in brake performance, WSP intervention etc. In addition we know that the signal engineer would always include an overlap of 200 yards before the collision point of whatever junction the signal was protecting. TPWS makes use of this overlap. Therefore on first thoughts, given that the signalling system is already designed taking into account the braking performance of trains, allowing for a reasonable safety margin there doesn't appear to be any scope for capacity improvement. There is, however, a 'but'. More later.

 

Secondly, it is oft stated that on the south west line from Waterloo, keeping to time is only possible by driving flat out on double yellows. This style would not be possible with the ATP supervision inherent with ETCS, so capacity would be reduced.

 

The 'but' referred to above are these thoughts.

 

Today's trains have better braking performance than that assumed by the design of the signalling system. For a long time EMUs have been nominally braked at 9%g with a 12%g emergency brake available. (Not sure what the actual rates are for IEP). A signalling engineer involved with ETCS implementation in UK said that the resignalling with ETCS would take advantage of the better braking rates and with ATP supervision a rethink on margins/overlaps is possible. Provided this takes account of extreme rail conditions (and ETCS includes a leaf fall mode that will adjust the guaranteed brake rate value) then this is reasonable.

 

The second area to bring benefits is the ability to throw away signal siting rules particularly on multi-tracked areas with different speeds applicable on different lines. For example on a line with adjacent slow and fast tracks, signals on the slow lines will tend for signal sighting reasons to be sited adjacent to the signals on the fast (to reduce the possibility of mis-read). This means that block sections, and hence capacity, is reduced on the slow lines. With no lineside signals (only marker boards) and ATP supervision the need for this restriction has gone.

 

I think the initial assumptions of 30% capacity increase have been overtaken by a more realistic 10%, but that is still worth having.

[TheSignalEngineer]

10% extra capacity may just make the existing timetable work in some places.

 

Because the speed / braking distance is not a linear relationship, tending to increase exponentially especially as speed gets over 90mph, increasing train speeds is not always the answer to capacity. In the early days of Railtrack I reviewed a proposal connected with WCRM to increase the line speed to 100mph between two locations of severely restricted speed. The line was 4-track and had a mix passenger stock and some long heavy freights. When the service and various stock was modelled it was found that most trains were incapable of reaching more than 80 mph before having to brake for the next restriction. With conventional signalling increasing the maximum speed above 90mph actually cut the capacity because of losing a signal section.

[The Stationmaster]

Whilst the MD of Digital Railway is originally from LUL, the people that are looking after ETCS are highly experienced and highly knowledgable heavy rail professionals.

 

Simon

 

By 'professionals' do you mean professional & experienced engineers, or professional and experienced operators, or professional and experienced driving staff/inspectors, or professional and experienced timetable planners?   Unless you have all the disciplines you are not necessarily going to achieve the best results and when all is said and done whatever is designed into the signalling system is really only meeting a particular headway specification for certain types of train because not all trains are equal.  Great stuff coming out of Waterloo where, traction and braking characteristics apar.

 

But go to the GWML or WCML (as examples) where you have a wide mix of train types, considerable speed differentials, and variation in journey characteristics and hence journey times over any section of route which leads to numerous wide margining differentials and talk about 'increasing capacity' owes less to signalling changes (within reason) than it does to total infrastructure capacity and the overall shape of the timetable.  Throw into that mix any train, or particularly a number of trains with very different timetable characteristics and you can end up reducing capacity - whatever signalling system you have.

[Mike Storey]

By 'professionals' do you mean professional & experienced engineers, or professional and experienced operators, or professional and experienced driving staff/inspectors, or professional and experienced timetable planners?   Unless you have all the disciplines you are not necessarily going to achieve the best results and when all is said and done whatever is designed into the signalling system is really only meeting a particular headway specification for certain types of train because not all trains are equal.  Great stuff coming out of Waterloo where, traction and braking characteristics apar.

 

But go to the GWML or WCML (as examples) where you have a wide mix of train types, considerable speed differentials, and variation in journey characteristics and hence journey times over any section of route which leads to numerous wide margining differentials and talk about 'increasing capacity' owes less to signalling changes (within reason) than it does to total infrastructure capacity and the overall shape of the timetable.  Throw into that mix any train, or particularly a number of trains with very different timetable characteristics and you can end up reducing capacity - whatever signalling system you have.

 

Absolutely true of course. But a combination of removing lower speed trains from key parts of main lines (by infrastructure changes/additions), increasing the top speed capability of local trains, and the introduction of ETCS or similar, enhances the improvements delivered by each. The sum is greater than the parts.

 

I see this happening to great effect on the ECML (upgrading of slow lines, reinstatement of slow lines, separation of conflicting moves, improved alternative routes for freight and improved junctions for same, improvements to alternative routes, some higher speed gearing of local units, and now trial operation of ETCS) which has progressed cumulative, minor improvements over many decades (some of the current schemes date back to 1980's BR planning team at York). Similar approaches have been taken on the WCML and to a lesser extent on the MML. I guess there must have been a similar approach on the GWML but it may have been lost for the moment in the fog of GWIP problems.

 

The issue of being able to avoid the capacity-destroying parallel signalling, of fast and slow lines, as described above, is one of the most obvious benefits.

[ess1uk]

https://www.networkrailmediacentre.co.uk/news/delivering-the-first-major-digital-railway-network-rail-procurement-launched-to-find-train-control-partner-for-route-digital-transformation

get your bids in