East Coast Timetable Dec 24

[Bald Rick]

to be clear - I was not seriously advocating closure of Welwyn North.

But it does consume paths, and is another constraint to deal with on the ECML.

 

[Nicholas Lewis]

to be clear - I was not seriously advocating closure of Welwyn North.

But it does consume paths, and is another constraint to deal with on the ECML.

Presumably the ETCS overlay through this section help when its commissioned?

 

[CarrotPie]

Presumably the ETCS overlay through this section help when its commissioned?

ETCS only helps to an extent, because you can only force so many trains through a section of track. Despite the word "overlay", ETCS does not overlay an additional two tracks through Welwyn!

 

[Nicholas Lewis]

ETCS only helps to an extent, because you can only force so many trains through a section of track. Despite the word "overlay", ETCS does not overlay an additional two tracks through Welwyn!

Appreciate that but my observations, albeit from the train window, whenever on an LNER ex Edinburgh is they get checked back in the two track sections following a 9Jxx but as soon as it put back over onto the Up Slow the signals step up to a clear aspect but the driver is rightly driver managing the train speed against the last aspect passed but with ETCS would get earlier advice of the section ahead now being clear. Not saying it really helps with capacity but might allow some pathing time to be relaxed.

 

[Starmill]

Presumably the ETCS overlay through this section help when its commissioned?

It can remove some of the constraints associated with signalling design, such as visibility, yes. That's not a Welwyn area specific thing, obviously.

 

[Bald Rick]

Presumably the ETCS overlay through this section help when its commissioned?

It’s commissioned.

 

[bramling]

Yes, because the only way of achieving that is to close and demolish the shopping centre next to Welwyn Garden City station.


This is an important point. Between Potters Bar and Royston the stopping service is a de facto Hertfordshire Metro with lots of intermediate journeys, especially students going to and from college/university at Hatfield and Hitchin. The Kings Cross-Cambridge service is not just about journeys to and from the end points.

Also worth adding that past timetables have provided more than 4tph for these local journeys. So less than 2tph really isn’t viable. One of the 2tph skipping Welwyn North would be tolerable, but is still going to cause hardship - to buy one path, and only if that path is usable elsewhere.

 

[Railperf]

I might be wrong but it does appear that the layout could be widened to 4 tracks immediately north of the viaduct albeit requiring additional new welwyn north tunnels. With short radius low speed turnout for stopping trains whereas semi fast services could continue switching over at Woolmer Green to slow lines? Has that option ever been considered.

 

[Bald Rick]

I might be wrong but it does appear that the layout could be widened to 4 tracks immediately north of the viaduct albeit requiring additional new welwyn north tunnels. With short radius low speed turnout for stopping trains whereas semi fast services could continue switching over at Woolmer Green to slow lines? Has that option ever been considered.

No, because it leaves a 2 track section and low speed turnouts are not helpful.

 

[Failed Unit]

It Is amazing the disruption that the late running at Welwyn North does cause, in the peak you often see the “2Cxx” held to let faster trains go. Then at Welwyn Garden City the Moorgate gets let out ahead of it. When you couple this we the strange logic of stopping the “2Cxx” at Alexandra Palace, a 3 minute delay at Welwyn North soon turns into a 20 minute delay. (At least it terminates at Kings Cross). The bottle neck doesn’t seem to be as bad Northbound. Maybe because apart from the Thameslink services most of them start at Kings Cross.

I do remember reading an article about making the viaduct bi-directional, but that may be published on 1st April as it made no sense.

 

[Austriantrain]

Also worth adding that past timetables have provided more than 4tph for these local journeys. So less than 2tph really isn’t viable. One of the 2tph skipping Welwyn North would be tolerable, but is still going to cause hardship - to buy one path, and only if that path is usable elsewhere.

Close Welwyn North and build a cable car alongside the railway to WGC.

 

[bramling]

It Is amazing the disruption that the late running at Welwyn North does cause, in the peak you often see the “2Cxx” held to let faster trains go. Then at Welwyn Garden City the Moorgate gets let out ahead of it. When you couple this we the strange logic of stopping the “2Cxx” at Alexandra Palace, a 3 minute delay at Welwyn North soon turns into a 20 minute delay. (At least it terminates at Kings Cross). The bottle neck doesn’t seem to be as bad Northbound. Maybe because apart from the Thameslink services most of them start at Kings Cross.

I do remember reading an article about making the viaduct bi-directional, but that may be published on 1st April as it made no sense.

Yes the Alexandra Palace stop has certainly had a pretty negative effect on reliability, as well as given the perception of slowing the service down. But worse has been its effect on recovering small amounts of late running.

With this stop not seeming particularly well used, in my view it should be removed. Certainly if the 717 service ever returns to 4tph off peak there can be little justification for Alexandra Palace receiving 10tph.

 

[Railperf]

No, because it leaves a 2 track section and low speed turnouts are not helpful.

Low -speed i mean 30-40mph which would virtually be on platform end anyway. The stopper starting at Weleyn GC has a 70mph turnout on the the viaduct, and is unlikely to get up to much speed. With ETCS in operation the driver runs to the speed display and thankfully wont have to worry about restrictive signal aspects. The benefit would be the train clears the two track section quicker as it ends up on a four track section at welwyn north, clearing the route behind. Without it, the stopper sits at Welwyn North and only clears the fast lines at Woolmer Green

 

[Bikeman78]

It Is amazing the disruption that the late running at Welwyn North does cause, in the peak you often see the “2Cxx” held to let faster trains go. Then at Welwyn Garden City the Moorgate gets let out ahead of it. When you couple this we the strange logic of stopping the “2Cxx” at Alexandra Palace, a 3 minute delay at Welwyn North soon turns into a 20 minute delay. (At least it terminates at Kings Cross). The bottle neck doesn’t seem to be as bad Northbound. Maybe because apart from the Thameslink services most of them start at Kings Cross.

I do remember reading an article about making the viaduct bi-directional, but that may be published on 1st April as it made no sense.

Stopping at Alexandra Palace seems to be the killer. The 2Cxx could nip out onto the up fast at Marshmoor or Potters Bar if it ran non stop to Finsbury Park. When the Moorgate trains ran every 20 minutes, one of the semi fasts was booked on the fast in each direction south of Potters Bar.

 

[Failed Unit]

Stopping at Alexandra Palace seems to be the killer. The 2Cxx could nip out onto the up fast at Marshmoor or Potters Bar if it ran non stop to Finsbury Park. When the Moorgate trains ran every 20 minutes, one of the semi fasts was booked on the fast in each direction south of Potters Bar.

I hope they review this and remove the stop, hopefully they can see it is killing their performance. Holding the Moorgate doesn’t really help as it soon cascades delays. They did as you say normally manage to get the train on the fast lines before the Alexandra Palace stop was added to keep both trains reasonably ontime. I have once done Potters Bar - Barnet to overtake on the fasts but that probably added a lot of knock on delay to other trains.

 

[notverydeep]

This could surely be eased a bit by reducing the service at Welwyn North from two trains an hour to one train an hour or fewer. There is a much better service at Welwyn Garden City.

It is true that at present, WGC does have a better service than Welwyn North with the Sevenoaks and Moorgate trains added to the 2CXX Cambridge stopping service. However, prior to the pandemic this was not as clear cut. Welwyn North had an extra two trains per hour (so 4 tph total) that ran non-stop to / from King's Cross at Peak times and as part of the ECML 'ESG' Timetable, it is intended that these would be reinstated.

When I started commuting from WGC, the 2CXX trains were non stop between WGC and Finsbury Park, but since 2015 first Hatfield (to be fair a busier station now even than WGC), then Potters Bar and lately Alexandra Palace have been added turning a 12 minute run to Finsbury Park into a 23 minute trundle and as others have noted, these now have almost no recovery options if even slightly out of path (this morning's 2C15 0756 Cambridge to King's Cross losing 11 minutes late between WGC and Finsbury Park being a case in point).

Unsurprisingly, the 16 minutes non-stop peak service between Welwyn North and King's Cross was well used and also unsurprisingly, in its current absence Welwyn North has had a much poorer recovery from the pandemic (62.4%) than WGC (72.7%) or Hatfield (92.7%) based on the latest ORR data. Never the less, it has about the same annual usage as stations such as Warwick Parkway, Great Malvern, Abergavenny, Kilmarnock or Cwmbran...

 

[Failed Unit]

It is true that at present, WGC does have a better service than Welwyn North with the Sevenoaks and Moorgate trains added to the 2CXX Cambridge stopping service. However, prior to the pandemic this was not as clear cut. Welwyn North had an extra two trains per hour (so 4 tph total) that ran non-stop to / from King's Cross at Peak times and as part of the ECML 'ESG' Timetable, it is intended that these would be reinstated.

When I started commuting from WGC, the 2CXX trains were non stop between WGC and Finsbury Park, but since 2015 first Hatfield (to be fair a busier station now even than WGC), then Potters Bar and lately Alexandra Palace have been added turning a 12 minute run to Finsbury Park into a 23 minute trundle and as others have noted, these now have almost no recovery options if even slightly out of path (this morning's 2C15 0756 Cambridge to King's Cross losing 11 minutes late between WGC and Finsbury Park being a case in point).

Unsurprisingly, the 16 minutes non-stop peak service between Welwyn North and King's Cross was well used and also unsurprisingly, in its current absence Welwyn North has had a much poorer recovery from the pandemic (62.4%) than WGC (72.7%) or Hatfield (92.7%) based on the latest ORR data. Never the less, it has about the same annual usage as stations such as Warwick Parkway, Great Malvern, Abergavenny, Kilmarnock or Cwmbran...

Not sure if this is going off topic, but I do see this everyday. Slight delay and it is behind the Moorgate. Had they held the Moorgate would have caused more chaos as it only has an 8 minute turnaround at Moorgate? Whatever they do with the ECML timetable it needs to be resilient.The digswell viaduct is only one example where there is no wriggle room but I am amazed the timetable was allowed to be constructed that there is no recovery for a train just running 5 minutes late.

The other area where you can lose time significantly is York - Newcastle. I have followed many an XC service with more stops from York - Edinburgh. Where again they can only really overtake at Durham if they don’t hold the slower train at York.

 

[Austriantrain]

I am amazed the timetable was allowed to be constructed that there is no recovery for a train just running 5 minutes late.

How would that work on a two-track line where the average headway between trains needs to be below 5 minutes in order for all of them to fit?

 

[Failed Unit]

How would that work on a two-track line where the average headway between trains needs to be below 5 minutes in order for all of them to fit?

Regulation and a plan b. They may need to hold late running trains making them even later to keep others on time. ie why let a train out of York that stops at Darlington, Durham, Newcastle, Morpeth and Alnmouth ahead of one that stops at Darlington, Newcastle and Berwick-upon-Tweed.

Don’t design a timetable where you have a 4 track railway like the ECML where you can switch between tracks, where a train can’t move from the slow line because of a stop further down the line that is new to the service group.

 

[Bald Rick]

The benefit would be the train clears the two track section quicker

But, you’ve just spent a significant fraction of a billion pounds and there’s still a two track section. Which limits capacity on the line to 20tph, and possibly less due to the margins required for the low speed turnout. (it’s 18 now, assuming 2 stop).

 

[Nicholas Lewis]

But, you’ve just spent a significant fraction of a billion pounds and there’s still a two track section. Which limits capacity on the line to 20tph, and possibly less due to the margins required for the low speed turnout. (it’s 18 now, assuming 2 stop).

If it can handle 18tph including the two stops at Welwyn Nth thats pretty impressive although is that actually timetabled?

 

[Railperf]

But, you’ve just spent a significant fraction of a billion pounds and there’s still a two track section. Which limits capacity on the line to 20tph, and possibly less due to the margins required for the low speed turnout. (it’s 18 now, assuming 2 stop).

How many millions more to build another twin track viaduct to 4 tracks?

 

[Failed Unit]

If it can handle 18tph including the two stops at Welwyn Nth thats pretty impressive although is that actually timetabled?

Most hours you have
4x Thameslink
4x Great Northern
5x LNER
1x Open access (Sometimes 2)
14 services.

Pre Covid this then became
4x Thameslink
8x Great Northern
6x LNER
18 Trains

At the moment the Great Nothern Peak offering is normally 5, so maximum 15.

 

[Railperf]

Most hours you have
4x Thameslink
4x Great Northern
5x LNER
1x Open access (Sometimes 2)
14 services.

Pre Covid this then became
4x Thameslink
8x Great Northern
6x LNER
18 Trains

At the moment the Great Nothern Peak offering is normally 5, so maximum 15.

And some freight paths too

 

[Magdalia]

If it can handle 18tph including the two stops at Welwyn Nth thats pretty impressive although is that actually timetabled?

Only in the peaks between the start of the Thameslink 2018 timetable and the start of covid less than 2 years later.

From 2009 to 2018 the peak service was 16tph and worked very well most of the time.

And some freight paths too

Like the open access operators, freight doesn't get paths in the peak.

Also some freight goes via the Hertford loop.

 

[Failed Unit]

Only in the peaks between the start of the Thameslink 2018 timetable and the start of covid less than 2 years later.

From 2009 to 2018 the peak service was 16tph and worked very well most of the time.


Like the open access operators, freight doesn't get paths in the peak.

Also some freight goes via the Hertford loop.

Looks like most if it goes via the loop during the day, and doesn’t seem to make much difference in journey time compared to goig via Digswell.

I do recall that one of the reasons that the peak services from Welwyn North ran Non-Stop into London was because it cleared the 2 track section quicker. One interesting thing that came of that was that it was quicker to double back to get between London and Welwyn Garden City (with a booked connection)

 

[OxtedL]

I hope they review this and remove the stop, hopefully they can see it is killing their performance. Holding the Moorgate doesn’t really help as it soon cascades delays. They did as you say normally manage to get the train on the fast lines before the Alexandra Palace stop was added to keep both trains reasonably ontime. I have once done Potters Bar - Barnet to overtake on the fasts but that probably added a lot of knock on delay to other trains.

The practice of putting the Cambridge stoppers on the fasts when delayed all but stopped when York ROC took over the signalling, a while before Alexandra Palace stops were added in. Although in recent years it was never that practical - especially in the peaks - due to the ever increasing traffic on the fast lines.

As with most conversations about York ROC and train regulation, I think they simply haven't got enough signalling resource to manage extra overtaking manoeuvres on top of everything else. And I wonder if moving from a panel at Kings Cross to a workstation structure removes some of the ability for signallers to collaborate and learn from each other.

The Ally Pally stop is useful for connecting that part of North London to Hertfordshire (including connections between WGC and Hertford branches) although perhaps not as important in the peaks when there are more Moorgate trains.
The biggest performance problem on the GN is that the peak 8tph timetable WGC to Finsbury Park simply doesn't work very well with the mix of semi-fasts and stoppers that's been specified. The slow lines have no spare capacity at all to absorb any delays.

 

[greyman42]

The other area where you can lose time significantly is York - Newcastle. I have followed many an XC service with more stops from York - Edinburgh. Where again they can only really overtake at Durham if they don’t hold the slower train at York.

I am not sure that XC services delay LNER services between York and Edinburgh. XC services are timed for 55 minutes between York and Newcastle, with stops at Darlington and Durham.

 

[Failed Unit]

I see your point about connection between the branches, as if you did stopper to stopper in the current timetable you are talking a 32 minute connection (as only the slickest people could do it in 2 minutes the current timetable offers - off peak). It is a pity that the overtaking isn’t used, but as you say you need more signallers to do that. I suspect in the AI world the software should be able to see these things in the future and handle it.

I must admit now, if I am at WGC, if I want Finsbury Park I tend to just go for the Moorgate. On odd occasions they will hold the Moorgate at the bottom of the bridge and let the Kings Cross train past but anything more then 5 minutes of delay on the Kings Cross you are fairly certain you are stuck behind the Moorgate to Finsbury Park.

Be interesting how many make the interconnection, passengers from Stevenage and the North will change there. Welwyn / Hatfield - Hertford will use the bus. I don’t see may using it considering changing from the fast is still a 15 minute+ connection. (Which still frequently fails as it is stuck behind the Moorgate service anyway)

Off-Peak things work better. We didn’t have long of the 15 minute Welwyn Garden City / Hertford - Moorgate so hard to comment how reliable it was. It did appear in the timetable consultation along with the Alexendra Palace stop. Hopefully if the 15 minute service ever returns (if they can get enough drivers to resource it) - the stop will be removed from the 2C. Granted you are still stuck behind the 717, but at least you do have a chance of passing it, should a signaller notice.

I am not sure that XC services delay LNER services between York and Edinburgh. XC services are timed for 55 minutes between York and Newcastle, with stops at Darlington and Durham.

From personal experience they do, not as badly as far as Newcastle, but that soon mounts up if the XC stops at Morpeth and / or Alnmouth. Cost me many a connection at Waverley but still within the 10 minutes delay that LNER used to view as acceptable.

 

[Verulamius]

The current impasse seems to arise because the various train and freight operating companies have competing rights which are unable to be timetabled given the current infrastructure on the East Coast Main Line. So I thought it might be useful to go back in history to look at the decision to grant those rights.

In 2016 the Office of Rail and Road approved the current track access rights for the East Coast after detailed analysis of the various competing claims.

In a letter of 24 March 2016 ORR set out their understanding of the various capacity constraints existing at that time and the services that could be possibly delivered following an all parties meeting earlier that month.

Looking at this it would appear that the expectations for infrastructure improvements have not been achieved (such as at Woodwalton).

https://www.orr.gov.uk/sites/default/files/om/ecml-hearing-letters-to-stakeholders-2016-03-24.pdf

Annex
Capacity
1. We recognise the question of what capacity is or will be available is complicated with assumptions about rolling stock choice, service patterns, infrastructure works, power supplies etc. all affecting the answer.
2. We understand VTEC currently uses around 5 off-peak paths/hour and Hull and GC share around 1 path making a total of 6/hour.
3. Our current best view is:
(a) Capacity for up to an additional 0.5 paths/hour out of Kings Cross is probably available now. This reflects, for example, information provided by Network Rail on 6 November showing relatively low use of the Welwyn Viaduct in many off-peak hours (compared to an 18 paths/hour theoretical maximum).
(b) A further additional 1 off-peak path/hour out of Kings Cross should be available from the May 2021 timetable assuming the infrastructure works at Werrington and Woodwalton are completed in line with Network Rails latest draft Enhancement Delivery Plan (EDP).
(c) A further 0.5 paths/hour may be available beyond this point bringing the total capacity up to 8 paths/hour out of Kings Cross but we would want to better understand the risks to connectivity and freight before that final 0.5 capacity was used. For example, would an 8th off-peak LDHS train out of Kings Cross mean reduced calls at Stevenage or reduced capacity for heavier freight that needs to use the Welwyn Viaduct?
4. We have noted Network Rail’s advice that the VTEC and TSGN franchises specify a quantum of services that may not fit over the Welwyn Viaduct in one particular peak hour. Our view is that issue does not need to be - and probably cannot be - settled before we decide the ECML access applications we have.
5. Network Rail’s latest power supply study was discussed with stakeholders on 24 February 2016 in York. We understand Network Rail has identified work such as upgrading feeder stations will be needed around Doncaster and further North to meet any increase in electric load beyond today’s levels.
6. We understand works in the Doncaster area may be delivered around the end of December 2017, but it seems unlikely the other necessary power supply enhancements will be completed much before the end of 2020, and that is subject to Network Rail working out what exactly it should do and securing funding to do it.
7. We consider the Northallerton freight loops listed in Network Rail’s current draft EDP with an indicative completion date of March 2019 are needed to protect freight if extra passenger trains are to run between York and Newcastle.
8. Our view is that no more than 2.5 London to Edinburgh trains per hour could run without unduly impacting freight and local connectivity.
9. Additional capacity on the ECML could be allocated to sub-sets of services taken from up to: 1.0 for an additional VTEC Edinburgh service, 0.5 for VTEC Middlesbrough, 0.5 for FirstGroup Edinburgh, 1.0 for Alliance Edinburgh and 1.0 for Alliance Cleethorpes/WYorks. We recognise these figures are rounded and the services actually proposed would use fewer paths to different degrees.
10. Applicants originally asked for access rights to support additional services starting at various points up to May 2020. None of the dates proposed for full services to start currently appear feasible given the infrastructure issues highlighted.
11. DfT said in its Hendy consultation that its decisions about the ECML connectivity fund might depend on our ECML access decisions. We are exploring with DfT what that statement means. The current position is that we think the value for money concerns raised by DfT are relevant to our duties but we are unclear how much weight we can reasonably give them in our decision-making process in the absence of any information about the assumptions originally made about the value for money of the fund or individual projects, details of how the value for money analysis would change depending on our decisions and how material those changes might be.

In ORR's decisions letter of 12 May 2016 there was an additional comment about the uncertainty of DfT financing. https://www.orr.gov.uk/sites/default/files/om/2016-05-12-ecml-decision-letter.pdf

22. The key uncertainty we considered was around the completion of infrastructure works. In particular, DfT noted it may review the case for the ECML “connectivity fund” works in light of our access decisions.
23. However, DfT did not provide evidence that allowed us to understand the strength of the current business case for the fund or details about how that case could be affected by our decisions. Moreover, any review by DfT would need to take account of how VTEC’s own proposals depended on connectivity fund works and how Network Rail might be liable to compensate operators in the event it could not provide contracted capacity. We have therefore proceeded on the basis of Network Rail’s current enhancement plans.