Welwyn Viaduct section doubling - How?

[AlanFry1]

There's a simple solution.. start to do some route maintenance work to the bridge and then, whoops, blow it up by accident!

Judi Drench, The Cast of Spooks and the man from Sky1's Spy are all outside your house as we speak and i'm not too fond on sabotage as it's a very beautiful landmark but surely widening it shouldn't p[beep] many people off as it's beinifitical for hundreds of thousands of people across the country.

 

[Class377/5]

There are no plans for a bay platform at Stevenage. None.

It's not just East Coast passengers that want to travel via the loop, but passengers from other FCC stations. Again, I insist that all up trains should run from the same side of the station.

I wouldn't say that. There are plans but just not concrete ones.

Isn't the Hertford Loop being used now/soon to test new signalling? Upgrade that and create some passing loops and it's surely worth doing whatever happens along the ECML? Trains closer together, able to run faster etc..

In fact, is there any other station (or in between two stations) that could allow for passing loops to allow an additional overtake for what might eventually become a more regular all-station service (whether part of LO or not)?

The plan is for the testing to take place with bi directional working on the other line. Interestingly the new class 700's should see testing on the line once they start doing training from Hornsey depot post 2015/6.

 

[Aictos]

I'd expect there's a fair few people along the Hertford Loop that might want to get to Stevenage for onward trips along the ECML and to Cambridge etc.

When I've needed to go from Hatfield to Hertford, I've gone via Stevenage. I've often considered it might be quicker (given the poor connection times at Stevenage) to go via London though.

Exactly, imagine if Hatfield only had a hourly service northbound in the off peak Monday to Friday and at weekends, you get to the station to find the hourly service has been cancelled so you're now looking at a 2 hour wait especially if you're just missed the last train.

Would that be attractive to you, no I wouldn't say it would be but increase the service to a half hourly service and it's not so bad.

I would say it would be a safe bet that if services were half hourly north of Hertford, then passenger growth would show this to be a good idea, it's only because demand is suppressed that the demand isn't obvious.

At weekends, it is very possible to extend the hourly Hertford terminator though to Stevenage without delaying any services - for example you have the xx:47 and there's nothing using platform 4 until xx:12 so you could get a train to terminate at xx:50 and depart at xx:00 which is just 10 mins and it still be on time by the time it's supposed to leave Hertford.

Granted this removes any service recovery by starting a service back from Hertford BUT it makes much better use of the driver and of the rolling stock to provide a improved service.

In fact, I've done a SimSig timetable which shows this is viable to do and can be done

 

[jon0844]

Hatfield certainly relies on you going to Stevenage, or Hitchin, and changing for some of the day - where trains alternate between PBO and CBG, and you take these and swap later on.

Otherwise it's a very long wait or the need to go in to London and back out (which your ticket wouldn't allow unless there's some funky routeing allowed - as per the London to Cambridge thread).

 

[D365]

ECML will get ETCS up to Doncaster from 2019 IIRC. This should, coupled with Hertford Loop Loops (excuse the puns!) allow greater capacity along this critical section through Hertfordshire.

 

[MarkyT]

ECML will get ETCS up to Doncaster from 2019 IIRC. This should . . . . allow greater capacity along this critical section through Hertfordshire.

Although a great supporter of more modern control systems, I am yet to be convinced that ETCS will allow following fast trains to pass magically through trains stopped in Welwyn North platforms, which is the key capacity limiting factor on the double track section.

 

[D365]

Although a great supporter of more modern control systems, I am yet to be convinced that ETCS will allow following fast trains to pass magically through trains stopped in Welwyn North platforms, which is the key capacity limiting factor on the double track section.

Fair enough, there's only so much that can be done without treating the root of the problem, in this case the two-track sections.

 

[Bald Rick]

Although a great supporter of more modern control systems, I am yet to be convinced that ETCS will allow following fast trains to pass magically through trains stopped in Welwyn North platforms, which is the key capacity limiting factor on the double track section.

Correct. But it will let slower and stopping trains that follow a fast train onto the two line section a little bit sooner than they can do now, saving a half minute here and there. May be worth a path an hour.

 

[Fred26]

Exactly, imagine if Hatfield only had a hourly service northbound in the off peak Monday to Friday and at weekends, you get to the station to find the hourly service has been cancelled so you're now looking at a 2 hour wait especially if you're just missed the last train.

Would that be attractive to you, no I wouldn't say it would be but increase the service to a half hourly service and it's not so bad.

I would say it would be a safe bet that if services were half hourly north of Hertford, then passenger growth would show this to be a good idea, it's only because demand is suppressed that the demand isn't obvious.

At weekends, it is very possible to extend the hourly Hertford terminator though to Stevenage without delaying any services - for example you have the xx:47 and there's nothing using platform 4 until xx:12 so you could get a train to terminate at xx:50 and depart at xx:00 which is just 10 mins and it still be on time by the time it's supposed to leave Hertford.

Granted this removes any service recovery by starting a service back from Hertford BUT it makes much better use of the driver and of the rolling stock to provide a improved service.

In fact, I've done a SimSig timetable which shows this is viable to do and can be done

I think two trains an hour from Letchworth - Moorgate via Hertford calling all stations should run, but perhaps only once the Hitchin flyover opens.
If work to improve capacity happens at Gorden Hill, then the Letchworth - Moorgate should run all to Gordon Hill then fast to Finsbury Park and all stops to Moorgate, with four an hour calling all stops from Gordon Hill - Moorgate.

 

[HSTEd]

Is there any particular reason Welwyn North can't be rebuilt so that the platforms are on loops?

 

[Tomnick]

Probably not, but I think that'd make the problem far worse. Each stopping train would then eat into more capacity overall.

 

[D365]

Is there any particular reason Welwyn North can't be rebuilt so that the platforms are on loops?

It could, but honestly, there's not enough space without dozing the existing station and taking away from the car park. That's the simple reason.

 

[LE Greys]

It could, but honestly, there's not enough space without dozing the existing station and taking away from the car park. That's the simple reason.

However, it is part of an option for doing the tunnels, but not the viaduct. This would again only be a partial solution, but would reduce the scale of the problem.

 

[MarkyT]

Probably not, but I think that'd make the problem far worse. Each stopping train would then eat into more capacity overall.

Unlikeley to reduce overall capacity, but trains stopping at Welwyn North would need to dwell longer to allow a fast train to pass, which would impact on journey times and stock utilisation for the stopping service; at peak times perhaps 2 fasts could overtake, but that could require stops of up to 5 minutes.

 

[Tomnick]

Taking a headway of three minutes, for example, a stopper would take up just over 1.5 paths if it took two minutes longer between Digswell and Woolmer Green. If it went inside, it'd take up at least two paths (one either side of Welwyn North), and probably a bit more if the entry and outlet from the loops were quite restrictive (especially with approach controlled signals going in). I'd still suggest, without knowing the exact timings and headways involved here, that there'd be an overall loss of paths!

As far as the increase in journey time is concerned, I reckon you'd do well to get one fast past with a five minute dwell in the loop. Probably three minutes (again depending on the headway you can get away with) between the stopper arriving to the first fast passing, then probably another two (on paper at least) until the stopper can depart. I'm sure you could tighten that up, especially with thoughts of more advanced control systems that are floating around (and, I'm sure, a long way off yet!), but at the cost of some breathing space if one train or the other manages to lose a half minute or two somewhere.

 

[MarkyT]

Taking a headway of three minutes, for example, a stopper would take up just over 1.5 paths if it took two minutes longer between Digswell and Woolmer Green. If it went inside, it'd take up at least two paths (one either side of Welwyn North), and probably a bit more if the entry and outlet from the loops were quite restrictive (especially with approach controlled signals going in). I'd still suggest, without knowing the exact timings and headways involved here, that there'd be an overall loss of paths!

As far as the increase in journey time is concerned, I reckon you'd do well to get one fast past with a five minute dwell in the loop. Probably three minutes (again depending on the headway you can get away with) between the stopper arriving to the first fast passing, then probably another two (on paper at least) until the stopper can depart. I'm sure you could tighten that up, especially with thoughts of more advanced control systems that are floating around (and, I'm sure, a long way off yet!), but at the cost of some breathing space if one train or the other manages to lose a half minute or two somewhere.

From http://www.opentraintimes.com . . .

In the down direction typically

An EC fast passing Welwyn GC passes Woolmer Green Junction 2 minutes later.

A FCC stopper departs Welwyn GC, then passers Woolmer Green Junction 5 minutes later.

So 3 minutes difference in running time and assuming paths are nominally 3 minutes apart at Woolmer Green to maintain clear signals. That means a following Express must pass WGC at least 6 minutes after previous all stations stopper departs WGC.

If there was a loop at Welwyn North and assuming an unconstrained run in with a clear overlap, the stopper would be clear of the main line some 2 1/2 minutes after leaving WGC.

With no passing time at North assume 1 minute non-stop running time for the express, so the difference now is 1 1/2 minutes and the fast can pass WGC 4 1/2 minutes after the stopper.

On departure the stopper can leave on a yellow immediately the fast has passed the next signal; as long as the next signals have good visibility, or are enhanced with repeaters etc, they should be seen to clear up quickly allowing acceleration behind the express as it races away ahead. ERTMS would no doubt improve all of this this marginally.

Difficult to justify I agree with the construction costs, extra complexity of junctions etc and increased journey time for the stoppers; as long as there are only 2 stoppers an hour there's plenty of room for fasts for the remainder of the 1/2 hour cycle.

 

[Tomnick]

Apologies for the slow response! I acknowledge that neither of us are convinced that (short) loops are justified in this situation, but it's suggested so often that I thought I'd put something (attached) together to illustrate my point. Based on the timings given (two minutes for a fast, five for a stopper) and an assumption that a three minute headway is required, but a stopper can follow a fast onto the main at a two minute interval*. Unless I've missed something, the sequence of three fasts and a stopper takes 1.5' longer if the stopper's recessed at Welwyn North, thus representing a slight loss of capacity overall. Although the first fast following the stopper can pass Welwyn G.C. 4.5' behind the latter, there's also the effect of the path occupied by the stopper as it leaves the loop to consider.

* - even if reduced to just one minute (which I'd suggest would be an absolute minimum), it'd still take half a minute longer.

 

[Lee_Again]

I think what the last few posts have done is prove that any solution other than the four track solution is futile. Loops, improved signalling and clever timetabling have virtually no possitive impact. And when you factor ECS and freight paths even less so.

So we're back to four tracking. Unless a new (and likely long) by-pass route is built it's going to be expensive tunnels and viaducts. And the by-pass would be even more expensive with other likely tunnels needed.

I could get 5 paths extra to Stevenage with only modest investment. What other ideas give extra paths?

My idea is to send 4 trains around the loop and join together some OA operators at King's Cross and split them further North.

I guess it might be possible to join/split trains at Stevenage/Hitchin. A 4 coach train from Peterborough joins an 8 coach from Cambridge. That's a path saved. Do that twice an hour for 2 paths an hour.

So we've got 7 new paths with no changes at all on the Welwyn infrastructure.

I'm not saying my ideas are great or indeed others are not but I do think that once we've identified the number of paths needed (as somebody said in an earlier post) a solution can be found.

 

[tbtc]

I could get 5 paths extra to Stevenage with only modest investment. What other ideas give extra paths?

My idea is to send 4 trains around the loop and join together some OA operators at King's Cross and split them further North.

I guess it might be possible to join/split trains at Stevenage/Hitchin. A 4 coach train from Peterborough joins an 8 coach from Cambridge. That's a path saved. Do that twice an hour for 2 paths an hour.

So we've got 7 new paths with no changes at all on the Welwyn infrastructure

Whilst five coach Open Access operators clearly aren't optimal use of ECML paths, joining/splitting them at Doncaster wouldn't make a huge difference to the number of Welwyn paths as they aren't that frequent.

Sending four trains an hour via the Hertford loop? Which four? The East Coast services to Leeds/ Newcastle/ Edinburgh? The Cambridge "fasts"? Would all four paths an hour cope with the three stoppers on that route? Presumably at least one would clash with a 313 somewhere? Is it worth slowing them down that much?

Joining the Peterborough/ Cambridge semi-fast services sounds like a good way of combining resources (although it would mean some would only be four coaches north of Stevenage), but at the moment there's a half hourly service stopping at Welwyn, Potters Bar etc - would the combined service still stop at these stations (in which case it would be twelve minutes slower than the ones that run non-stop from Stevenage to Finsbury Park)?

I've been tinkering with the idea of replacing the current "all stops" Moorgate services with some skip stopping, but the terminus at Moorgate couldn't cope with the increased number of trains that you'd need to stick down there to make such a thing work. There are no easy answers.

 

[Aictos]

Easy solution would be to blow up the viaduct and rebuilt it as a 4 track viaduct or built a second viaduct to the side of the existing one.

 

[jon0844]

Ideally, even if it is listed, the best idea would be to remove it and build a new one in keeping with the design, except that would likely be very costly and - erm - cause quite a bit of disruption for some time!

 

[ChiefPlanner]

From memory - you can timetable 17 tph over Welwyn. Remember , looking ahead - any HS2 will potentially reduce log term demand for Inter - City type trains over the viaduct.

(options for 3 /4 tracking south of Peterborough is flavour of the month)

 

[tbtc]

Can anyone confirm what the top speed is over Welwyn? Is there much 100+ mph running in the area?

 

[MarkyT]

Can anyone confirm what the top speed is over Welwyn? Is there much 100+ mph running in the area?

From Sectional Appendix:

http://www.networkrail.co.uk/browse...x/london north eastern sectional appenidx.pdf

See module LN2 page 19 . . . .

115 both fast lines from South of Potters Bar to Woolmer Green then 125 beyond, EXCEPT up fast line limited to 105 through both Welwyn North and South tunnels.

Please note this is not from the live updated system so may not include any late changes.

 

[tbtc]

From Sectional Appendix:

http://www.networkrail.co.uk/browse...x/london north eastern sectional appenidx.pdf

See module LN2 page 19 . . . .

115 both fast lines from South of Potters Bar to Woolmer Green then 125 beyond, EXCEPT up fast line limited to 105 through both Welwyn North and South tunnels.

Please note this is not from the live updated system so may not include any late changes.

Thanks

 

[MichaelAMW]

From Sectional Appendix:

http://www.networkrail.co.uk/browse...x/london north eastern sectional appenidx.pdf

See module LN2 page 19 . . . .

115 both fast lines from South of Potters Bar to Woolmer Green then 125 beyond, EXCEPT up fast line limited to 105 through both Welwyn North and South tunnels.

Please note this is not from the live updated system so may not include any late changes.

Would I be right in saying that one way to increase the number of paths per hour would be to lower the line speed? Has that been considered or is the consequent increase in journey times more than could be accepted by the various interested parties?

 

[MarkyT]

Would I be right in saying that one way to increase the number of paths per hour would be to lower the line speed? Has that been considered or is the consequent increase in journey times more than could be accepted by the various interested parties?

If the signals remained where they are, a reduction in speed would only result in a LOSS of capacity, as each train would take longer to get through each fixed length block. The capacity could only be increased thereby whilst also reducing the signal spacing to suit that new lower speed, resulting in more signals overall. Positioning of the existing signals through the area was no doubt particularly difficult however, trying to maintain consistent spacing despite the tunnels limiting sighting, and together with the viaduct and station, restricting the places where it is safe to stop a train at a danger signal. A wholesale renewal of the system through that area with more, closer spaced signals would be a very complex and costly exercise given those layout constraints, and lowering the linespeed I believe would be sufficiently unpopular amongst operators (particularly the longer distance ones) to make it unviable.

It is to be hoped that ERTMS will one day provide a partial solution as the spacing of the fixed block markers in such a system is no longer linked in the same way to worst case train braking distance; In busy areas you can add more discrete blocks to allow slower trains to 'close up' without sacrificing maximum speed for the fastest schedules.

 

[Lee_Again]

Sending four trains an hour via the Hertford loop? Which four? The East Coast services to Leeds/ Newcastle/ Edinburgh? The Cambridge "fasts"? Would all four paths an hour cope with the three stoppers on that route? Presumably at least one would clash with a 313 somewhere? Is it worth slowing them down that much?

I would send the two non-stop Cambridge and the semi fast Finsbury Pk>Stevenage services. I would increase the Hertford stoppers to every 15 mins with each waiting at Gordon Hill (on new loops) for them to pass.

The increase in journey time would be about 5 mins. Obviousy extra power, new loops and some linespeed improvements would be needed but relatively modest. And that gives 4 new paths across through Welwyn.

At present there are three OA destinations (Hull, Bradford and Sunderland). Each is after extra paths and if they agreed to join together (and split somewhere north of Welwyn) they could each have 1tph. It would also allow EC to introduce the 3tph Leeds services and other destinations.

 

[tbtc]

I would send the two non-stop Cambridge and the semi fast Finsbury Pk>Stevenage services

Okay, but if you divert the half hourly semi fast Finsbury Park - Stevenage services non-stop via Hertford then you can't also "save" two paths over Welwyn an hour by tagging them on to the other half hourly Finsbury Park - Stevenage services (which stop at Potters Bar, Knebworth etc), so the seven paths an hour over Welwyn don't add up.
--- old post above --- --- new post below ---

If the signals remained where they are, a reduction in speed would only result in a LOSS of capacity, as each train would take longer to get through each fixed length block

If we are just talking about one block then yes.

But at the moment the speed difference between a 125/180/225 and a 317/321/365 mean that the former will catch up the latter fairly fast over a distance of several "blocks"

 

[MichaelAMW]

If the signals remained where they are, a reduction in speed would only result in a LOSS of capacity, as each train would take longer to get through each fixed length block. The capacity could only be increased thereby whilst also reducing the signal spacing to suit that new lower speed, resulting in more signals overall. Positioning of the existing signals through the area was no doubt particularly difficult however, trying to maintain consistent spacing despite the tunnels limiting sighting, and together with the viaduct and station, restricting the places where it is safe to stop a train at a danger signal. A wholesale renewal of the system through that area with more, closer spaced signals would be a very complex and costly exercise given those layout constraints, and lowering the linespeed I believe would be sufficiently unpopular amongst operators (particularly the longer distance ones) to make it unviable.

Well, indeed, I took the resignalling to be taken as read, as the way to pack the trains a bit closer together. I appreciate what you are saying about the general awkwardness of the locations along there - but I would have thought it's still peanuts compared with major civil engineering. But, as you also say, the Inter City operators probably wouldn't much like the idea - particularly since they wouldn't reap all of the benefits of the extra paths, even if the time penalty were worth it.