• Our booking engine at tickets.railforums.co.uk (powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Heathrow Express GWR Contract

Status
Not open for further replies.

hwl

Established Member
Joined
5 Feb 2012
Messages
7,354
But if the 332s are all gone there's nothing to stop them just getting rid of ATP and switching from the 360s to the 345s overnight, isn't there?
If the 345s are still getting upset by GW-ATP equipment installed on the GW(reliefs) in open air then not so easy to do an overnight swap.

Bombardier are being focused by CRL on core signalling issues rather than rolling out solutions for Heathrow (which they apparently already have)
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Muzer

Established Member
Joined
3 Feb 2012
Messages
2,773
I don't think 345s are compatible with the tunnel infrastructure.
The issue as I understood it was that the 345s' ETCS equipment did not like ATP signals. Obviously both ATP and ETCS are currently present in the Heathrow tunnels, but ATP can be switched off once the 332s and 360s are gone.
 

Muzer

Established Member
Joined
3 Feb 2012
Messages
2,773
If the 345s are still getting upset by GW-ATP equipment installed on the GW(reliefs) in open air then not so easy to do an overnight swap.

Bombardier are being focused by CRL on core signalling issues rather than rolling out solutions for Heathrow (which they apparently already have)
345s are running fine on the mainline as they use TPWS there. It is specifically the ETCS equipment that doesn't like the ATP. But good point on that, maybe the interference from the ATP on the mainline is enough to upset the ETCS as the train transitions into the tunnel...
 

Muzer

Established Member
Joined
3 Feb 2012
Messages
2,773
As long as the in cab signalling doesn’t work then further testing is needed until it does. ATP can’t be scrapped until a workable solution is found as there are no other safety systems on the airport branch.
ETCS is (now) fitted on the airport branch. The problem is that 345s don't like both being present at once, as I understand it.
 

cactustwirly

Established Member
Joined
10 Apr 2013
Messages
7,447
Location
UK
Given the 387s won't have GW-ATP will they be able to run on the Mains as today with out causing any issues to other services? (Before ETCS is rolled out on GWML)?

Yes, they already operate over the mains, as well as the non ATP fitted 16xs as well!
 
Joined
19 May 2010
Messages
505
Location
West Drayton
ETCS is (now) fitted on the airport branch. The problem is that 345s don't like both being present at once, as I understand it.
Yeah, as I understand as well. I don't know if ATP can be temporarily switched off to test 345s in engineering hours. How else could they work around the problem?
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
The GW-ATP works on the same carrier frequency as the ETCS replacement system. The balises and track loops of GW-ATP have significantly higher radio signal strength than ETCS, and the 1990s engineers also chose to duplicate all the transponders in the tunnel for reliability. There is some kind of 'waveguide propagation' issue that distributes the GW-ATP signal more widely than expected due to the frequency used and the physical dimensions of the tunnel. The ETCS equipment is already provided in the airport tunnels and is I understand connected to the signalling system and powered up in final configuration. GW-ATP also remains connected to the signalling, also powered up to allow continued operation of cl.332 train. ETCS equipped trains cannot be operated reliably in the tunnels due to interference experienced from the stronger ATP signal. There is no way of easily 'switching off' the GW-ATP equipment widely in the tunnels to allow overnight testing of ETCS trains in the short no train periods available. Large numbers of individual fuses have to be individually hand-removed by signal engineering staff in tunnel equipment cabinets to get a useful section of ATP equipment isolated ready for ETCS testing in the final configuration, and it all has to be manually reconnected and verified again before start of service the following morning. Hence the required testing to get cl.379 and cl.345 approved for running in the tunnels is proving very difficult to orchestrate without extended service outages before commissioning.
 
Last edited:

cactustwirly

Established Member
Joined
10 Apr 2013
Messages
7,447
Location
UK
The GW-ATP works on the same carrier frequency as the ETCS replacement system. The balises and track loops of GW-ATP have significantly higher radio signal strength than ETCS, and the 1990s engineers also chose to duplicate all the transponders in the tunnel for reliability. There is some kind of 'waveguide propagation' issue that distributes the GW-ATP signal more widely than expected due to the frequency used and the physical dimensions of the tunnel. The ETCS equipment is already provided in the airport tunnels and is I understand connected to the signalling system and powered up in final configuration. GW-ATP also remains connected to the signalling, also powered up to allow continued operation of cl.332 train. ETCS equipped trains cannot be operated reliably in the tunnels due to interference experienced from the stronger ATP signal. There is no way of easily 'switching off' the GW-ATP equipment widely in the tunnels to allow overnight testing of ETCS trains in the short no train periods available. Large numbers of individual fuse have to be individually hand-removed by signal engineering staff in tunnel equipment cabinets to get a useful section of ATP equipment isolated ready for ETCS testing in the final configuration, and it all has to be manually reconnected and verified again before start of service the following morning. Hence the required testing to get cl.379 and cl.345 approved for running in the tunnels is proving very difficult to orchestrate without extended service outages before commissioning.

Why was ECTS fitted in the (Heathrow) tunnels instead of TPWS?
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
Why was ECTS fitted in the (Heathrow) tunnels instead of TPWS?
Mainly because it was not considered safe enough for a high frequency tunnel railway. ETCS is also the chosen longer term solution for GWML from Paddington to Heathrow Airport Junctions and has merely been 'deferred' on the open air segment in favour of 'Plan B' which is known as TPWS+. TPWS+ is described as enhanced TPWS, with all previously unfitted plain line signals and speed restrictions newly equipped with appropriate loops to get close to the protection previously provided by GW-ATP. The derailment containment slab between the rails in the tunnels needs expensive holes cut into it to to accommodate additional transponders which under TPWS could not be in the same position as future ETCS equipment if the two systems are ever to be installed in parallel ready for a future changeover. All this extra work would have to take place in the short overnight no train periods.
 
Joined
19 May 2010
Messages
505
Location
West Drayton
The GW-ATP works on the same carrier frequency as the ETCS replacement system. The balises and track loops of GW-ATP have significantly higher radio signal strength than ETCS, and the 1990s engineers also chose to duplicate all the transponders in the tunnel for reliability. There is some kind of 'waveguide propagation' issue that distributes the GW-ATP signal more widely than expected due to the frequency used and the physical dimensions of the tunnel. The ETCS equipment is already provided in the airport tunnels and is I understand connected to the signalling system and powered up in final configuration. GW-ATP also remains connected to the signalling, also powered up to allow continued operation of cl.332 train. ETCS equipped trains cannot be operated reliably in the tunnels due to interference experienced from the stronger ATP signal. There is no way of easily 'switching off' the GW-ATP equipment widely in the tunnels to allow overnight testing of ETCS trains in the short no train periods available. Large numbers of individual fuses have to be individually hand-removed by signal engineering staff in tunnel equipment cabinets to get a useful section of ATP equipment isolated ready for ETCS testing in the final configuration, and it all has to be manually reconnected and verified again before start of service the following morning. Hence the required testing to get cl.379 and cl.345 approved for running in the tunnels is proving very difficult to orchestrate without extended service outages before commissioning.

Mainly because it was not considered safe enough for a high frequency tunnel railway. ETCS is also the chosen longer term solution for GWML from Paddington to Heathrow Airport Junctions and has merely been 'deferred' on the open air segment in favour of 'Plan B' which is known as TPWS+. TPWS+ is described as enhanced TPWS, with all previously unfitted plain line signals and speed restrictions newly equipped with appropriate loops to get close to the protection previously provided by GW-ATP. The derailment containment slab between the rails in the tunnels needs expensive holes cut into it to to accommodate additional transponders which under TPWS could not be in the same position as future ETCS equipment if the two systems are ever to be installed in parallel ready for a future changeover. All this extra work would have to take place in the short overnight no train periods.
Thanks for the detailed information. Confirms my suspicions.
 

Roy Badami

Member
Joined
12 Oct 2014
Messages
108
Well the other way to look at is that ETCS is the future - and given Crossrail trains were already specified with ETCS why wouldnt you fit that rather than TPWS, which is a legacy system.

As to why TPWS (or even AWS, I think?) were never fitted when the airport branch was built (or subsequently), that's easy too: all the trains using the branch were fitted with GW-ATP so legacy protection systems were completely unnecessary.

As to the details of the history of ATP introduction that lead to that state of affairs, I'll leave that to others who are more knowledgeable.
 
Last edited:

Muzer

Established Member
Joined
3 Feb 2012
Messages
2,773
Yes, the original intention (as I understand it) was that GW-ATP and ETCS would be able to operate side-by-side. I believe this was actually tested and found to be the case ---- but for whatever reason, while it might be the case on ideal test trains, it's not the case on 345s. From what I gather it's the 345s themselves that are considered "at fault" for this --- but I don't know enough about the radio frequencies involved to be able to form my own opinion on who is really at fault here. But you'd have thought they'd have had some radio engineers to confirm whether GW-ATP and ETCS would be able to theoretically run side-by-side or not, so I would imagine as a result that the "problem" item here is the 345s being unable to sufficiently discriminate the ETCS from the GW-ATP signals.

From the sounds of it, it seems that they've decided that rather than do the "correct" thing and make the 345s better at discriminating between the two competing signals, it would instead be simply easier to wait until the GW-ATP is gone and then have the 345s running on the ETCS alone. But this would require either a problematic transitionary period where you can have one or the other but not both systems running, or an intermediary period where they set up something like TPWS+ only to abandon it once ETCS is working. Neither is particularly a great option...
 

samuelmorris

Established Member
Joined
18 Jul 2013
Messages
5,121
Location
Brentwood, Essex
Yes, the original intention (as I understand it) was that GW-ATP and ETCS would be able to operate side-by-side. I believe this was actually tested and found to be the case ---- but for whatever reason, while it might be the case on ideal test trains, it's not the case on 345s. From what I gather it's the 345s themselves that are considered "at fault" for this --- but I don't know enough about the radio frequencies involved to be able to form my own opinion on who is really at fault here. But you'd have thought they'd have had some radio engineers to confirm whether GW-ATP and ETCS would be able to theoretically run side-by-side or not, so I would imagine as a result that the "problem" item here is the 345s being unable to sufficiently discriminate the ETCS from the GW-ATP signals.

From the sounds of it, it seems that they've decided that rather than do the "correct" thing and make the 345s better at discriminating between the two competing signals, it would instead be simply easier to wait until the GW-ATP is gone and then have the 345s running on the ETCS alone. But this would require either a problematic transitionary period where you can have one or the other but not both systems running, or an intermediary period where they set up something like TPWS+ only to abandon it once ETCS is working. Neither is particularly a great option...
I imagine since GW-ATP is surplus to requirements in future, they decided against further risky modifications to the 345 software, given that is at the centre of the current Crossrail delay.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
Yes, the original intention (as I understand it) was that GW-ATP and ETCS would be able to operate side-by-side. I believe this was actually tested and found to be the case ---- but for whatever reason, while it might be the case on ideal test trains, it's not the case on 345s. From what I gather it's the 345s themselves that are considered "at fault" for this --- but I don't know enough about the radio frequencies involved to be able to form my own opinion on who is really at fault here. But you'd have thought they'd have had some radio engineers to confirm whether GW-ATP and ETCS would be able to theoretically run side-by-side or not, so I would imagine as a result that the "problem" item here is the 345s being unable to sufficiently discriminate the ETCS from the GW-ATP signals.
From the sounds of it, it seems that they've decided that rather than do the "correct" thing and make the 345s better at discriminating between the two competing signals, it would instead be simply easier to wait until the GW-ATP is gone and then have the 345s running on the ETCS alone. But this would require either a problematic transitionary period where you can have one or the other but not both systems running, or an intermediary period where they set up something like TPWS+ only to abandon it once ETCS is working. Neither is particularly a great option...
I think any train trying to run on ETCS while GW-ATP is still active in the tunnels is going to run into trouble, because the ATP signals are very much stronger than those of ETCS and the ATP balises and track loops broadcast continuously rather than being interrogated only on passing by the train. The frequency used combined with the particular dimensions of the tunnel cause the signal to bounce along the tunnel acting as a waveguide too.
I imagine since GW-ATP is surplus to requirements in future, they decided against further risky modifications to the 345 software, given that is at the centre of the current Crossrail delay.
I agree that is potentially a lot of difficult expensive work for an outcome that is not required once the GW-ATP is decommissioned in the tunnel. I think the project should invest in a temporary remote switching system to allow isolatation and restoration the GW-ATP equipment in the tunnels quickly so testing of 387s and 345s under ETCS only can take place routinely at night before the big switch over when 387s take over the expresses and and 345s start running to the airport.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
It should be noted that the Bombardier cl.387 has exactly the same train control management system on board as the same manufacturer's cl.345, just without the cab control screens already enabled for ETCS. The cabs and systems have been designed carefully with retrofitting in mind however, as have all similar Bombardier trains since the 379s apparently. Inevitably, the challenges faced on the 345s will be shared by the 387s although equally the lessons learned from the Elizabeth trains should be transferable too.
 

gwr4090

Member
Joined
9 Nov 2011
Messages
144
Is the GWR-ATP interference problem with ETCS just in the Heathrow tunnels or is it more general ? I note that the fitting of ETCS to the GW main line from Paddington-Bristol Parkway has apparently been cancelled. It should have been completed by now but Network Rail is totally silent about what is going on. There doesnt seem to be any coherent strategy. The latest Digital Strategy paper merely states that funds have been set aside to investigate the possible life extension of GWR-ATP.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
24,793
Location
Nottingham
I imagine since GW-ATP is surplus to requirements in future, they decided against further risky modifications to the 345 software, given that is at the centre of the current Crossrail delay.
Modifying the software and completed the necessary testing and safety approvals would most likely take longer than the current plan.
 

matt_world2004

Established Member
Joined
5 Nov 2014
Messages
4,504
If the plan is for gw-ATP to be turned off over night and 345s/387s running the next day. Wouldn't there need to be a period where the line is suspended for passenger services so drivers can get rolling stock related route knowledge like stopping positions at each. I can't see how they could switch out the rolling stock overnight without having successful dry runs.
 

hwl

Established Member
Joined
5 Feb 2012
Messages
7,354
I imagine since GW-ATP is surplus to requirements in future, they decided against further risky modifications to the 345 software, given that is at the centre of the current Crossrail delay.
They have modified the software (awaiting testing), just that this is a lower priority change with the focus on getting CBTC in the core working as the highest.
 
Last edited:

hwl

Established Member
Joined
5 Feb 2012
Messages
7,354
Yes, the original intention (as I understand it) was that GW-ATP and ETCS would be able to operate side-by-side. I believe this was actually tested and found to be the case ---- but for whatever reason, while it might be the case on ideal test trains, it's not the case on 345s. From what I gather it's the 345s themselves that are considered "at fault" for this --- but I don't know enough about the radio frequencies involved to be able to form my own opinion on who is really at fault here. But you'd have thought they'd have had some radio engineers to confirm whether GW-ATP and ETCS would be able to theoretically run side-by-side or not, so I would imagine as a result that the "problem" item here is the 345s being unable to sufficiently discriminate the ETCS from the GW-ATP signals.

From the sounds of it, it seems that they've decided that rather than do the "correct" thing and make the 345s better at discriminating between the two competing signals, it would instead be simply easier to wait until the GW-ATP is gone and then have the 345s running on the ETCS alone. But this would require either a problematic transitionary period where you can have one or the other but not both systems running, or an intermediary period where they set up something like TPWS+ only to abandon it once ETCS is working. Neither is particularly a great option...

GW-ATP is based on the Belgian TBL1 system. In 2006 SNCB concluded that TBL1 and ETCS wouldn't play happily with each other and TBL1 needed to be replaced with newer equipment TBL2/2+/3 to enable practical ETCS roll out.

The RF environment in the tunnel is far worse than the open air so it might have been the case that the UK testing was done in open air and everything looked ok...

As MarkyT has said the someone did a very good job convincing Heathrow to double up on all the ATP equipment to improve reliability when they were specing and building the tunnels!

The design of the ATP equipment predates all the EMC rules
 

zn1

Member
Joined
3 Sep 2011
Messages
435
whats that saying - the more you overdo the plumbing, the easier it is to block the drain...emu,loco control and safety systems are too computer dependent, ...the GW ATP system works, if it aint broken dont fix it.
 

Brush 4

Member
Joined
25 Nov 2018
Messages
502
Modern computerised trains, and in other forms of transport do seem over complicated. If they fail there seems little a driver or signalman/woman can do to get things going again. It seems to be either go or stop, nothing in between, no temp fixes to get the train out of the way, or a signal overridden to get the line going asap.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
24,793
Location
Nottingham
whats that saying - the more you overdo the plumbing, the easier it is to block the drain...emu,loco control and safety systems are too computer dependent, ...the GW ATP system works, if it aint broken dont fix it.
Probably very difficult to get spares for it these days, still less extra equipment to extend it.
 

hwl

Established Member
Joined
5 Feb 2012
Messages
7,354
Probably very difficult to get spares for it these days, still less extra equipment to extend it.
Yep, support was massively downgraded after SNCB started actively removing it (2012 - DfT had to twist their arm for IEP on board equipment) and not fitting it to new stock from 2006.
It is a legacy product...
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
Yep, support was massively downgraded after SNCB started actively removing it (2012 - DfT had to twist their arm for IEP on board equipment) and not fitting it to new stock from 2006. It is a legacy product...
Very much so. Also, in the UK application, the equipment was used in a much more comprehensive way than Infrabel/SNCB TBL1, where in combination with ancient 'Crocodile' ramps it provided only simple AWS-like warning and TPWS-like trainstop protection.
https://en.wikipedia.org/wiki/Transmission_balise-locomotive
The GW-ATP, by contrast, is a 'full supervision' system with custom speed curve supervision and additional earlier 'better aspect' update transponders where justified. The UK application will remain the only use of this equipment anywhere in the world after a similar system has been fully removed from part of the Hong Kong MTR, work that is currently underway. Infrabel/SNCB wanted to greatly expand train protection coverage and was faced with a choice of staying with the old tech or changing to a eurobalise-based implementation where an equivalent of the TPWS 'overspeed loop' could also be provided. The choice made makes migration to ETCS very much easier as the same transponders are used and the system can be emulated in the standard onboard ETCS computer and DMI screens of a modern train using the now largely standardised Level 1 'limited supervision' mode, and the standard balise reader hardware is used rather than needing an additional plug-in 'specific transmission module' to read legacy transponders.
 
Last edited:

43096

On Moderation
Joined
23 Nov 2015
Messages
15,164
Isn’t the seating on the 360s the type all the SWT commuters moaned about?
All? You mean the Portsmouth line clowns who couldn’t/wouldn’t understand the explanation given and were stuck in their own selfish little world.
 

jimm

Established Member
Joined
6 Apr 2012
Messages
5,230
Is the GWR-ATP interference problem with ETCS just in the Heathrow tunnels or is it more general ? I note that the fitting of ETCS to the GW main line from Paddington-Bristol Parkway has apparently been cancelled. It should have been completed by now but Network Rail is totally silent about what is going on. There doesnt seem to be any coherent strategy. The latest Digital Strategy paper merely states that funds have been set aside to investigate the possible life extension of GWR-ATP.

Sorry, but it is not correct to say 'it should have been completed by now' - it was decided a long time ago that the GW resignalling would be done in two phases - first with conventional lineside signals, which are what control trains now, to be followed by ETCS at a point in the early 2020s.

There may have been indicative target dates for that but nothing has ever been set in stone when it comes to the timing of ETCS installation.

whats that saying - the more you overdo the plumbing, the easier it is to block the drain...emu,loco control and safety systems are too computer dependent, ...the GW ATP system works, if it aint broken dont fix it.

It works? So why is it the cause of 28 per cent of all recorded technical incidents and 20 per cent of delay minutes on the GWR IET fleet then?

The figures were given in the December issue of Modern Railways.
 
Last edited:
Status
Not open for further replies.

Top