Why does everyone love pairing by use, for 4-track railways

[zwk500]

If you have the fasts in the middle you can of course just splay the slow lines and keep the fasts perfectly straight. And no splay is needed at stations which only have slow line platforms (though if you do that you can't serve those stations if the slows are not in use).

Generally 2 islands needs slightly more space than 1 island and 2 side stations, but plenty of paired by direction lines do use islands for this very reason.

 

[WesternBiker]

A slight tangent, but the Metro North line from New York to New Haven has an interesting approach to undertaking engineering work at stations like Pelham on this four-track section (with the platforms on the outer tracks, normally paired by direction). In this video on YouTube, working has been restricted to the northern pair of tracks (which have bidirectional capability).

Trains stopping on the way out from New York have a curious temporary footbridge arrangement (seen at 1:00 in the video), combined with selective door opening, to enable passengers to walk over the pair of tracks not in use. I've never seen this anywhere else. Not sure that would be allowed here...

Off topic, but Pelham is also interesting as the point where trains switch from the overhead to 3rd rail, on the move - it's quite a sight to watch up to 5 pantographs rising in unison (one pantograph for each 2-car unit).

 

[66701GBRF]

That's the issue with closing the middle tracks, but potentially can be addressed by having a line block on an outer track when access is needed - obviously only possible if the timetable allows for the necessary blocks. Or there might be a station where staff can access via the platform.

With the amount of access required during engineering works line blocks are impractical. Renewals would also be a lot harder, if not impossible, under paired by direction unless all lines were blocked.

 

[edwin_m]

But one up and one down can’t be on the same side of the formation with paired by direction.

Sorry, referring to wrong post. Deleted.

 

[Spartacus]

Pairing by use use is great when you have little interaction between fasts & slows and services on one tend to have a different destination to the other.

In your example it was great for the MML for most of it's history, but it's pretty lousy now with GTRs running an intense service, weaving from slows to fasts and back again, with every weave crossing something coming in the opposite direction, and given the frequency they almost always are timed to cross right behind something, or fairly close in front, so that just a small delay usually causes a knock on. Even the turnbacks at St Albans and Luton offer little relief these days as something's usually right behind a terminator.

About the only benefit now is that it's easier to get out of the freight terminals at Radlett, Luton, Limbury Road and Elstow, but I'm no longer convinced it's worth the trade off for a few freight conflicts compared with heaven knows how many passenger ones (and the freights still conflict with two lines instead of 3 or 4). The most difficult would be Radlett, which is perhaps close enough to run round facilities on the Hendons near West Hampstead to make a southbound departure line at Radlett the best option.

The MML really is hamstrung now by the combination of the pairing of lines and the combination of slow, semi fast, fast and express services, many of which weave across to get around the others or out of their way, and it's a bit of a nightmare for regular delays happening seemingly out of nothing, but I can't really see things changing.

 

[WAO]

An interesting case is the track plan between Roby Jn and Huyton Jn shown in:

Northwest Sparks - Huyton and Roby Additional Tracks

All About The Lancashire Electrification Project

www.nw-sparks.co.uk

making a virtue out of necessity.

WAO

 

[zwk500]

An interesting case is the track plan between Roby Jn and Huyton Jn shown in:

Northwest Sparks - Huyton and Roby Additional Tracks

All About The Lancashire Electrification Project

www.nw-sparks.co.uk

making a virtue out of necessity.

WAO

Similar in concept to the Acton Bridge - Hartford LNW Jn section on the WCML, and formerly the Amington Jn - Colwich Jn section. Junction at one end, so slow lines aligned to side of diverging junction to avoid a conflict between fast and slow lines in one direction.

 

[plugwash]

Generally 2 islands needs slightly more space than 1 island and 2 side stations, but plenty of paired by direction lines do use islands for this very reason.

The other nice thing about 2 islands and paired by direction is it means all the trains in a given direction use the same island. This is a lot more conviniant for passengers than dashing about across bridges/subways as the delays/cancellations are announced.

 

[Sad Sprinter]

It depends on what system you're running on. The Brighton line works pretty well with PBU, as there are no semi-fast services which weave from fast to slow. Although arguably paring by direction would work better south of East Croydon.

On the South Western, its got so many junctions on either side PBU won't really make sense.

The South Eastern is on a completely different planet.

 

[zwk500]

It depends on what system you're running on. The Brighton line works pretty well with PBU, as there are no semi-fast services which weave from fast to slow. Although arguably paring by direction would work better south of East Croydon.

PBD isn't really needed south of East Croydon, especially with the terminating trains at Gatwick. A flyover to take the down slow from PBU to PBD at Balcombe tunnel Jn (and 4 tracks through to Keymer jn) would be nice though.

 

[Sad Sprinter]

PBD isn't really needed south of East Croydon, especially with the terminating trains at Gatwick. A flyover to take the down slow from PBU to PBD at Balcombe tunnel Jn (and 4 tracks through to Keymer jn) would be nice though.

I remember a London Reconnections article for some years ago suggested it for one reason or another. I think it was as a way to get East Grinstead trains off the slows and to sort out crossings at Stoats Nest.

 

[AM9]

Pairing by use use is great when you have little interaction between fasts & slows and services on one tend to have a different destination to the other.

In your example it was great for the MML for most of it's history, but it's pretty lousy now with GTRs running an intense service, weaving from slows to fasts and back again, with every weave crossing something coming in the opposite direction, and given the frequency they almost always are timed to cross right behind something, or fairly close in front, so that just a small delay usually causes a knock on. Even the turnbacks at St Albans and Luton offer little relief these days as something's usually right behind a terminator.

About the only benefit now is that it's easier to get out of the freight terminals at Radlett, Luton, Limbury Road and Elstow, but I'm no longer convinced it's worth the trade off for a few freight conflicts compared with heaven knows how many passenger ones (and the freights still conflict with two lines instead of 3 or 4). The most difficult would be Radlett, which is perhaps close enough to run round facilities on the Hendons near West Hampstead to make a southbound departure line at Radlett the best option.

The MML really is hamstrung now by the combination of the pairing of lines and the combination of slow, semi fast, fast and express services, many of which weave across to get around the others or out of their way, and it's a bit of a nightmare for regular delays happening seemingly out of nothing, but I can't really see things changing.

I think the MML works pretty well most of the time, with a resulting service overall no worse than the other 4-track routes with mixed services. Potentially the most disruptive crossover, between Kentish Town and West Hampstead is mitigated by the separation of the fast up to slow up, which is just south of WHP station and the slow down to fast down, being south of the Belsize Tunnel. This gives some leeway trying to maintain a gap across both fast lines for a northbound slow to fast transfer in an area where the 100/125mph line speed aligns with the beginning of ST P approach speeds.
There is also a recovery move by swapping up fasts from slow to fast at Radlett if the Harpenden Junction path is too tight.

 

[zwk500]

I remember a London Reconnections article for some years ago suggested it for one reason or another. I think it was as a way to get East Grinstead trains off the slows and to sort out crossings at Stoats Nest.

It's one of those where you solve some problems but trade it for others. I preferred CARS's solution for EG of building grade-separated 'semi-fast' connections to take trains from the Slows at East Croydon to the Fasts at Selhurst and V.v.
Building the Grade-separation required at South Croydon and Purley for PBD would be a monumental undertaking. I can see the appeal of it - even from south of Balham tbh, as Streatham Jns could be reworked reasonably easily - but to me PBU is cleaner, and the transition would be south of Three Bridges for a PBD through Balcombe, HHE and Burgess Hill. NR did draw up sketches for PBD (with flyover) Keymer Jn to HHE at one point, so other than the tunnels the main question would be what to do at Preston Park.

 

[Sad Sprinter]

It's one of those where you solve some problems but trade it for others. I preferred CARS's solution for EG of building grade-separated 'semi-fast' connections to take trains from the Slows at East Croydon to the Fasts at Selhurst and V.v.
Building the Grade-separation required at South Croydon and Purley for PBD would be a monumental undertaking. I can see the appeal of it - even from south of Balham tbh, as Streatham Jns could be reworked reasonably easily - but to me PBU is cleaner, and the transition would be south of Three Bridges for a PBD through Balcombe, HHE and Burgess Hill. NR did draw up sketches for PBD (with flyover) Keymer Jn to HHE at one point, so other than the tunnels the main question would be what to do at Preston Park.

How would you have a PBD Balham Junction?

 

[edwin_m]

I think the MML works pretty well most of the time, with a resulting service overall no worse than the other 4-track routes with mixed services. Potentially the most disruptive crossover, between Kentish Town and West Hampstead is mitigated by the separation of the fast up to slow up, which is just south of WHP station and the slow down to fast down, being south of the Belsize Tunnel. This gives some leeway trying to maintain a gap across both fast lines for a northbound slow to fast transfer in an area where the 100/125mph line speed aligns with the beginning of ST P approach speeds.
There is also a recovery move by swapping up fasts from slow to fast at Radlett if the Harpenden Junction path is too tight.

PBD would need an extra flyover somewhere in this area to get all northbound Thameslinks over to the west side, which would be very difficult considering the tunnels and other constraints. Probably more flyovers at St Albans, Luton and Bedford to deal with turnbacks (similar to Welwyn and Hitchin on the ECML).

 

[zwk500]

How would you have a PBD Balham Junction?

Balham itself would remain PBU for Balham/Falcon/Battersea Pier jns and Vic approach. I'd take the Up slow over the fast lines between Selhurst and Balham somewhere, if doing that. However as I've mentioned I'd actually rather keep it as PBU to Three bridges, and then PBD south of there. Although a lot depends on future service patterns.

 

[SteelWeelApeal]

There are certainly advantages and disadvantages to both. One big disadvantage of paired by direction happens when it comes to terminating stopping trains. By direction, they have to cross the two fast lines, blocking all four, whereas by use, they don’t need to interact with the fasts at all. Compare the conflicts of the Woking or Basingstoke terminators with those at Didcot, Reading or Maidenhead.

This also happens at the other end, hence all(?) lines are paired by use for the final approach into the London terminus.

 

[mr_jrt]

Unless of course, you pair by direction with the fasts on the outside, a-la the Met/Jubilee, then it's just a central turnback between the lines.

 

[zwk500]

This also happens at the other end, hence all(?) lines are paired by use for the final approach into the London terminus.

Off the top of my head, Windsor lines approaching Waterloo are PBD. All others are either 2/3 track approaches, or PBU (either all the way to the station or with a brief bit of fully reversible working for the final throat).

 

[norbitonflyer]

It depends on what system you're running on. The Brighton line works pretty well with PBU, as there are no semi-fast services which weave from fast to slow. Although arguably paring by direction would work better south of East Croydon.

Pair by use has to be used between East Croydon and gatwick, as the western pair of tracks feed into the Quarry Line which bypasses Merstham, and Redhill (and become the eastern pair of tracks when they meet up again near Horley, as the old and new lines cross over at different levels south of Stoats Nest).

EDITED to correct east/west error

 

[edwin_m]

Off the top of my head, Windsor lines approaching Waterloo are PBD. All others are either 2/3 track approaches, or PBU (either all the way to the station or with a brief bit of fully reversible working for the final throat).

Kings Cross and Waterloo (main lines only, at Wimbledon) have flyovers to change their PBD lines to PBU. London Bridge from Croydon now has, but didn't until the Thameslink project built one a decade or so back (and East London Line provided a separate one at New Cross Gate a few years before).

 

[WAO]

IIRC, The Windsor lines are PBD from Barnes until after Vauxhall. Happily, they've got their own terminal at Waterloo back after it had been rebuilt and borrowed (temporarily!) for Eurostar

The outer roads used to be the fasts although they had junction speed limits at Clapham Jn and Barnes which the slows did not. Now, my reading of the SA is that the inner lines are the fasts and both the down fast and slow are 60mph through Barnes Jn, after much work in a confined space with a road crossing!

 

[Sad Sprinter]

Pair by use has to be used between East Croydon and gatwick, as the eastern pair of tracks feed into the Quarry Line which bypasses Merstham, and Redhill (and become the western pair of tracks when they meet up again near Horley, as the old and new lines cross over at different levels south of Stoats Nest).

Interestingly they were also PBD between Balham and Victoria in the the 1890s.

 

[OscarH]

Pair by use has to be used between East Croydon and gatwick, as the eastern pair of tracks feed into the Quarry Line which bypasses Merstham, and Redhill (and become the western pair of tracks when they meet up again near Horley, as the old and new lines cross over at different levels south of Stoats Nest).

I might be completely wrong, but I think it's the western tracks at Croydon and the eastern tracks at Horley rather than vice versa?

 

[DynamicSpirit]

Usually with pairing by direction the fast lines are in the middle, the Met/Jubilee being a rare exception

I'm guessing the reason that arrangement with slow lines in the middle works there is that the fast Met lines aren't actually that fast - it's only about 5 miles from Finchley Road to Wembley Park and the Met trains don't go at a massive speed between them, so there's not really an issue with the fast lines splaying out to accommodate the centre platforms.

It seems to me that, in terms of path capacity, the most efficient layout for a four-track railway is with the slow lines in the middle and fast lines to the outside. Such a line could theoretically handle 16tph on the fasts and 16tph or more on the slows.

• Minor stations only need one platform island, serving both slow lines. (Though platforms on the fasts at some minor stations would add flexibility in times of disruption.)

It's certainly true that you only need one platform, and I guess that might be more efficient in construction costs, especially if you want facilities on the platform. But a minor downside for passengers is that you ALWAYS have to access the platform via stairs/lift. On the other hand, if the platforms are on the outside of the 4 tracks then very often, at least one platform can be accessed directly from the street. And if there's a suitable road bridge, you might even manage level access to both platforms (albeit with some walk).

By which it has always seemed to me to be crippling itself, with the cabbaging of the ability for trains in the same direction to overtake one another and all the fiddling about at Hanslope to get ready for the Roade divergence. Better to do it by direction, and have a diveunder for the down slow at Roade.
mixture.

Interestingly, If it was considered desirable to change the Southern WCML to paired by direction, I wonder if that would be possible to do that without major rebuilding of any station - since all the Avanti/LNWR stations already have platforms on all lines. Not sure if you'd need some track rebuilding at Euston though to allow access to platforms without too many conflicting moves? There's already a dive-under on the Euston approach so perhaps that's already achievable without that much work?

 

[markymark2000]

I think the best place for an easy convert to PBD would be GEML. Because of where the Crossrail/Liz Line portal is at Stratford, it would only work from Ilford to Shenfield. It would be the easiest to convert by doing S-F-F-S. Centre island platforms could fall into disrepair same as on the South West Main Line and eventually lead to lower maintenance costs as you don't need to maintain an abandoned platform.

I don't think it would work F-S-S-F on the basis that you'd need to rebuild the flyover at Ilford to accommodate it, plus at Shenfield the fast line would be the line that branched off towards Southend. In any case, you would have to come up with a solution for terminating trains and access to depots/stabling yards. Technically Liz Line could run ECS from Shenfield, use the flyunder towards Southend, then a reversing siding or two put in there which would solve that conflict with the Anglia fast trains, but it would cost a fair bit because you'd be paying drivers to do that move rather than them having a break or going straight back out.

My main concern though with PBD is engineering possessions. In a world where we are becoming increasingly concerned over health and safety and given the reduction in working on lines adjacent to those where trains are running, I can't see anyone being happy about having to close all lines for a possession rather than just closing the slow or fasts (as currently happens).

 

[Cricketer8for9]

I might be completely wrong, but I think it's the western tracks at Croydon and the eastern tracks at Horley rather than vice versa?

You’re not wrong. Indeed you are correct.

 

[norbitonflyer]

I might be completely wrong, but I think it's the western tracks at Croydon and the eastern tracks at Horley rather than vice versa?

You are right, of course

 

[zwk500]

Interestingly, If it was considered desirable to change the Southern WCML to paired by direction, I wonder if that would be possible to do that without major rebuilding of any station - since all the Avanti/LNWR stations already have platforms on all lines. Not sure if you'd need some track rebuilding at Euston though to allow access to platforms without too many conflicting moves? There's already a dive-under on the Euston approach so perhaps that's already achievable without that much work?

You'd need a rebuild of most (every?) station(s) as having the fast lines either side of the central island is unlikely to be compatible with the existing line speed profile. Milton Keynes, Bletchley, and Tring would also need completely reworking for the trains that terminate there. Watford Junction as well would likely need something. You'd also need to rejig the connections to Wembley Yard/Willesden reliefs as departing freight and the Southern WLL train would now need to cross all tracks instead of just the Up slow as at present.
Euston itself would likely need a rebuild of the DC Line connections and Slow line - primrose hill connection, as well as the alignments all changing because different tracks would need the flyover, but the final approach down Camden Bank would probably still have the same conceptual layout of PBU Fast-Slow-Fast.

 

[mr_jrt]

I don't think the stations are as problematic as you might think. My thinking was slows on the middle to ease the termination issues, but the point is equally valid for fasts in the middle. Possibly even more so. Either way, one of the existing fast lines remains a fast line, and you only have to upgrade one of the slow lines to achieve your goal, and that can all be done whilst it's still the slow line.

At a glance, all the station approaches actually look pretty straight - it's things like the approaches to the Watford tunnels where the alignment appears problematic. Arguably there a better solution there might be to bore a new tunnel west of the existing ones, and leave the fasts using the current one, rather than having to acquire the land to re-profile the current slow line approaches. Dunno if that's worse or not. IIRC, many years ago there was talk of a 5th line or even 6 tracking out to Hemel or Kings Langley to ease freight congestion, which would have been just as expensive, so perhaps not entirely fanciful.