4 Tracking Welwyn Viaduct: could it happen?

[Pigeon]

Here's an idea for partial widening, with four tracks through Welwyn North station itself but only one additional new track north and south thereof through new single bore tunnels and over a new viaduct respectively. This concept removes the platforms from the fast lines and allows a stopping arrival at the same time as a fast is passing in the same direction. It also allows a Welwyn North stopper to depart on yellow very close behind a fast that has just passed the station.
http://www.townend.me/files/welwynviaduct.pdf

Interesting, but I don't get it... once you commit to making a new viaduct and tunnels at all, that's the big difficult thing called for, and whether they're single or double track doesn't make much difference, so you might as well go the whole hog while you're at it.

 

[MarkyT]

Out of interest and probably a question for somebody like @Bald Rick - what or is there a limit about having points in close proximity to something like Welwyn viaduct where there is a large drop if something happened?

That's why I put overrun spurs, leading to friction bufferstops or other arresting measures, at my trailing connections. It also means a stopping train approaching a red at the platform starter can have it's overlap set into the spur in parallel with an express movement taking place in the same direction, rushing past on the adjoining fast line. It's over 200m from the south end of the platform to the viaduct so enough for a standard overlap along the spur. The distance from the north end of the down platform to the tunnel mouth is somewhat shorter, but still capable of accomodating a clear overlap spur for a reasonable approach speed. The extended slow lines (particularly on the up when not in use for a stopping train) could also be useful for regulating freights while a non-stopping outer suburban train following on the corresponding slow can overtake via the retained slow to fast crossovers at Digswell and Woolmer Green.

 

[MarkyT]

Interesting, but I don't get it... once you commit to making a new viaduct and tunnels at all, that's the big difficult thing called for, and whether they're single or double track doesn't make much difference, so you might as well go the whole hog while you're at it.

It avoids realigning the fasts I suggest, which can continue running largely unaffected throughout construction. A new single track viaduct would have to carry only half the load of a double, possibly quite significant for such a tall structure. Say the new viaduct was to carry both down lines then the worst case longitudinal braking forces of two trains emergency stopping in the same direction simultaneously would be massive. That design choice could also be an issue for the existing viaduct if the two lines remaining on it were both for the same direction. A double track bore tunnel using modern methods is probably not that much more expensive than a single I concur, but if two new single track bores were constructed either side of the existing then that could be considerably more expensive. Either way Welwyn North station would need complete reconstruction.

 

[Pigeon]

OK, I understand your point about realignment and tunnels. I don't agree about braking forces on the viaduct though. I haven't tried to estimate any actual numbers, but I am pretty sure that they would be so small in comparison to the weight of the structure itself that the thrust lines would not move far enough from vertical to be a problem.

 

[Indigo Soup]

Surely you build it modern, with slim pillars in line with the old pillars?
I am not having this untouchable listed nonesense - presumably if such delicate opinions existed back then the original would never have got built! The juxtaposition of Victorian and modern would add to the scene....

Frankly, if the Victorians had the same problem, they'd have ripped down the 'heritage' viaduct and built a new one in its' place to carry four tracks and designed to suit current aesthetic tastes.

 

[DynamicSpirit]

Frankly, if the Victorians had the same problem, they'd have ripped down the 'heritage' viaduct and built a new one in its' place to carry four tracks and designed to suit current aesthetic tastes.

That would've been a little unfortunate for passengers wanting to cross the viaduct during the 10 years or so between starting to rip down the old one and completion of the new one!

I'm going to hazard a guess that, for that reason, the Victorians might not have been so blase about ripping down the structure as you think!

 

[DynamicSpirit]

Out of interest and probably a question for somebody like @Bald Rick - what or is there a limit about having points in close proximity to something like Welwyn viaduct where there is a large drop if something happened?

I'm (obviously) not Bald Rick, but I wonder if there's a bit of a clue in Google Maps: Look at where the 4 lines merge into 2 lines South of the viaduct. For Southbound trains (the ones heading away from the viaduct), the points where the tracks diverge are barely 100m from where the viaduct ends: Practically on the viaduct by railway standards! But for the Northbound tracks, the points are much further South - fully 500m from the viaduct. I can't see any obvious obstacles preventing the 4 tracks from continuing a bit nearer the viaduct, so I'd hypothesize that they are that far South for exactly the safety reason you are thinking of.

Of course, if the whole thing was 4-tracked throughout, this would be a non-issue, as no need for any points!

 

[Meerkat]

Isn’t there a speed restriction in the tunnels to the north of the viaduct that could be got rid of if trains weren’t crossing in there?

Re that photo of the Swiss viaduct....
I have driven round the Alps several times and think the big long viaducts are works of art as well as engineering excellence. I come back and stare at my GB atlas, metaphorical crayons in hand.......

 

[ess]

Could the existing viaduct be modified to add an additional track on each side?

 

[adamedwards]

If Welwyn North loops work, how about simpler layout like this: A fast line thorugh the car park to the west on a well canted curve and making the northbound platform a reversible reversing loop. Trains could start from Welwyn North in the peak heading south, removing the problem of capacity taken as they slow from the north. In the evening peak trains would terminate, again requiring only the time taken to stop, not restart. At othertimes the line would function as a normal through line.

 

[Ianno87]

I'm (obviously) not Bald Rick, but I wonder if there's a bit of a clue in Google Maps: Look at where the 4 lines merge into 2 lines South of the viaduct. For Southbound trains (the ones heading away from the viaduct), the points where the tracks diverge are barely 100m from where the viaduct ends: Practically on the viaduct by railway standards! But for the Northbound tracks, the points are much further South - fully 500m from the viaduct. I can't see any obvious obstacles preventing the 4 tracks from continuing a bit nearer the viaduct, so I'd hypothesize that they are that far South for exactly the safety reason you are thinking of.

Of course, if the whole thing was 4-tracked throughout, this would be a non-issue, as no need for any points!

That's mainly because on the Down (Northbound) there is no operational advantage to having the Down Slow continue to tie in parallel with the Up Slow turnout.

Instead, the Down Slow merges with the Down Fast as soon as it is a signal overlap length clear of the protecting signal. So if train passes on Down Fast, the points can swing for a Down Slow line train merging behind sooner.

On the Up, it *is* advantageous to get Up Slow trains clear of the Up Fast as soon as is physically possible to give greens to a following fast train.

 

[Indigo Soup]

That would've been a little unfortunate for passengers wanting to cross the viaduct during the 10 years or so between starting to rip down the old one and completion of the new one!

I'm going to hazard a guess that, for that reason, the Victorians might not have been so blase about ripping down the structure as you think!

To be fair, they'd probably just build a second parallel one. But they wouldn't get wound up about the fact that the first one was quite nice.

 

[Bald Rick]

Here's an idea for partial widening, with four tracks through Welwyn North station itself but only one additional new track north and south thereof through new single bore tunnels and over a new viaduct respectively. This concept removes the platforms from the fast lines and allows a stopping arrival at the same time as a fast is passing in the same direction. It also allows a Welwyn North stopper to depart on yellow very close behind a fast that has just passed the station.
http://www.townend.me/files/welwynviaduct.pdf

This is a great idea, which never occurred to me nor, much more importantly, to the authors of the report into various options for quadrupling all the way between Stevenage & Welwyn Garden City.
That report must have been at least 10/12 years ago & it concluded they were all too expensive. From memory it did not mention the viaduct listing as a problem.

The other virtue of MarkyT's excellent scheme is that it will not be a barrier to full quadrupling. Indeed it provides quadrupling through the critical Welwyn North station.

Interesting, but I don't get it... once you commit to making a new viaduct and tunnels at all, that's the big difficult thing called for, and whether they're single or double track doesn't make much difference, so you might as well go the whole hog while you're at it.

Hmm, all the hassle and effort of the preparation of building a twin track viaduct and twin track tunnel, and about 80% of the cost, only to do rather less than half a job in terms of point output.

 

[Meerkat]

The victorians just extended them sideways didn’t they, not worrying too much about matching brickwork.

 

[HowardGWR]

Hmm, all the hassle and effort of the preparation, for building a twin track viaduct and twin track tunnel, and about 80% of the cost, only to do rather less than half a job.

Actually three tracks as in MarkyT's suggestion are promoted elsewhere (e.g. Lea Valley) as a cheap solution where the problem is chiefly a peak one. I assumed that was Mark's assumption for his approach.

 

[HowardGWR]

The victorians just extended them sideways didn’t they, not worrying too much about matching brickwork.

See Brunel's Maidenhead Bridge and Wharncliffe Viaduct.

 

[Ianno87]

Actually three tracks as in MarkyT's suggestion are promoted elsewhere (e.g. Lea Valley) as a cheap solution where the problem is chiefly a peak one. I assumed that was Mark's assumption for his approach.

Lea Valley 3 track won't be a tidal flow.

It's a conventional two track railway with a long siding built alongside for an extra train to shuttle back and forth (in basic terms).

3 track railways of significant length seldom work for trying to operate a reasonably intensive service in both directions.

 

[MarkyT]

It would be possible to put short platform loops in at Welwyn North alone, but a stopping train would probably have to dwell for at least five minutes to allow a fast to pass, and it's approach to the red platform starter would necessarily be fairly slow due to defensive driving policy, which would slow down clearance of the fast for the following express. At least my extended slows allow the fast and slow train approaches to occur at the same time and also permit a stopper to depart very closely behind a fast. In the up direction, the departing stopper only has to go around half a mile or so before it can be fully clear of the fast again on the slow line the other side of the viaduct for the next fast to pass. In the down the distance to recommencement of four tracks is rather further at nearly two miles through the tunnels. Perhaps quad tracking through the tunnels and station and leaving the viaduct as double would be another possibility. Tidal flow is not a factor in this case. There's simply too much traffic at almost all times to consider bidirectional working on the centre track of the 3 line sections, and a number of additional crossovers would be required which would need very long turnouts indeed to maintain speed for the expresses.

 

[Bald Rick]

Actually three tracks as in MarkyT's suggestion are promoted elsewhere (e.g. Lea Valley) as a cheap solution where the problem is chiefly a peak one. I assumed that was Mark's assumption for his approach.

The Lea Valley Reversible is built on land long since in railway ownership, and doesn’t need any tunnels, viaducts or major consents.

The cost of building a twin track tunnel of the length we are talking about at Welwyn is not that much more than that of a single track. It is effectively the cost of disposing of the extra spoil (about twice as much) and more concrete in the lining. All the other costs are about the same.

Similarly, a twin track viaduct doesn’t cost twice that of a single track version.

Anyway it’s all irrelevant, as there’s minimal benefit and it’s not going to happen.

 

[yorksrob]

That would've been a little unfortunate for passengers wanting to cross the viaduct during the 10 years or so between starting to rip down the old one and completion of the new one!

I'm going to hazard a guess that, for that reason, the Victorians might not have been so blase about ripping down the structure as you think!

They would have more likely just widened the viaduct with a new adjoining one alongside. Happenned in numerous instances where lines were doubled etc.

 

[Ken H]

They would have more likely just widened the viaduct with a new adjoining one alongside. Happenned in numerous instances where lines were doubled etc.

could they not put new reinforced concrete columns down the middle of the existing piers, then put a wide deck supported by those piers? Bung in a 'fake' parapet similar what is there now.

The bricks you see are cladding added in the 1930's to remedy structural deficiencies.

 

[Meerkat]

How would the trains run whilst you are doing this?

 

[Bevan Price]

Sure it's possible; we can do anything with enough time and resources. Should we do it? Probably not, or at least not until post-HS2 ECML traffic levels are known.

The government has proposed building a considerable amount of new housing in a roughly linear belt between Cambridge & Hitchin. If that is ever completed, some of that housing may be near the ECML somewhere north of Hitchin -- and increase the need for outer suburban services to/from Kings Cross. That may well more than cancel out any demand reduction due to HS2b.

 

[yorksrob]

could they not put new reinforced concrete columns down the middle of the existing piers, then put a wide deck supported by those piers? Bung in a 'fake' parapet similar what is there now.

The bricks you see are cladding added in the 1930's to remedy structural deficiencies.

Well, the Victorians were only just getting used to using concrete, (concrete viaducts didn't come about until later on Cannington on the Lyme Regis branch and Glenfinnan being early examples), but nowadays I suppose there are issues as to whether you would damage the integrity of the pier by digging into it - also whether the overhang could be suitably supported (a civil engineer would be better to comment).

But it seems easier to build a modern viaduct alongside it and put a brick skin around it to match.

 

[brad465]

If capacity is currently full south of the viaduct, would capacity improve on the re-modelling of Kings Cross in 2020? If the running lines are full all the way to Welwyn then maybe not, but I would imagine smoother and faster train movements in+out of the terminal and more lines on approach to the station would free up space where congestion around the terminus of a railway influences the capacity of the mainline a long way out

 

[DynamicSpirit]

Just been trying to figure out what the pressure on the lines is. If I've understood the timetables correctly, the weekday off-peak frequency (after 9 Dec) is:

Slow lines:
4tph Welwyn Garden City-Moorgate
(+4tph via Hertford between Alexandra Palace and Finsbury Park)
2tph Cambridge-Kings Cross (semi-fast South of Welwyn Garden City; these are the ones that call at Welwyn North)

Fast lines:
1tph Cambridge-Brighton
2tph Ely/Kings Lynn-Kings Cross
2tph Peterborough-Horsham
5tph LNER services
irregular open access services

Can anyone confirm if that's correct, or if I've missed anything? And is there anything else that should be running that currently isn't because of the Thameslink timetable issues?

 

[Bald Rick]

Just been trying to figure out what the pressure on the lines is. If I've understood the timetables correctly, the weekday off-peak frequency (after 9 Dec) is:

Slow lines:
4tph Welwyn Garden City-Moorgate
(+4tph via Hertford between Alexandra Palace and Finsbury Park)
2tph Cambridge-Kings Cross (semi-fast South of Welwyn Garden City; these are the ones that call at Welwyn North)

Fast lines:
1tph Cambridge-Brighton
2tph Ely/Kings Lynn-Kings Cross
2tph Peterborough-Horsham
5tph LNER services
irregular open access services

Can anyone confirm if that's correct, or if I've missed anything? And is there anything else that should be running that currently isn't because of the Thameslink timetable issues?

It’s the 2020 spec you want to use. Which may or may not come into service in 2020.

That includes 2tph Cambridge - Brighton, and 8tph long distance high speed (Franchised and Open access combined).

Then add in the peak Peterborough extras, and the 2tph WGC - Maidstones, and extras to the Hertford loop.

 

[717001]

Plus freight trains

 

[Bald Rick]

Plus freight trains

Not via Welwyn in the peak.

 

[59CosG95]

Not via Welwyn in the peak.

Many of the freight trains that do use the ECML (and there are very few) tend to start north of Woolmer Green anyhow (e.g. Langley Redlands Stone, whatever Biggleswade produces etc.)
I've only ever seen the ECML south of Peterborough used for long-distance container trains (as an example) at weekends when other major corridors (e.g. the WCML) are shut.