Opportunities Increased Use of Split/Join

[Manutd1999]

Running longer trains helps to lower costs/PAX and make better use of the paths in congested areas. However, this goes against the constant clamour for smaller cities to receive direct links to everywhere. To that end, it's somwhat suprising that the use of split/join moves is quite limited across the network, as this potentially solves both problems.

Off the top of my head:

ECML
There are lots of 5-car sets running down the ECML every day which could be good candidates. The GC services to Bradford and Sunderland are the most obvious. If we ignore the different operators (i.e. assume a post GBR world), then Hull and Middlesborough could be other options. There could easily be scope for 2ph to split/join at Doncaster and head off to the various cities around Yorkshire and the North-East.

Lake District
Northern currently operate a service from Manchester Airport to Barrow. Making this 6-car and splitting at Lancaster, with 3-cars heading to Windemere, could be a good way of providing a regular service to both Lake District branches.

XC
A few different options for XC (assuming they operate double voyagers...), but splitting at Exeter for Paignton/Plymouth, or at Southampton for Bournemouth/Brighton, could be interesting options.

EMR
The Liverpool to Norwich service could be a good candidate to split at Ely, with half reversing to Norwich and half continuing to Cambridge/Stanstead.


I'm sure there are more ideas out there. Where else on the network do you think there may be opportunities?

The big disadvantage is of course performance, as delays to one service can cascade onto the 'joining' service. However, with suitable padding in the schedules, this should be manageable. In addition, most of these schemes would only be deliverable as part of a timetable re-cast, but again that is not a show-stopper.

 

[miklcct]

Splitting and joining block the platform for a long time. This is the reason why Southern no longer splits / joins the Eastbourne service because there isn't enough platform space.

 

[The Planner]

"Suitable padding" that in itself is a capacity eater. If you are splitting and joining close to the start or end of the journey the risk is lower, see WMT a couple of years back to see how it can go wrong.

 

[Topological]

Take XC as an example.

If the aim is capacity in the core, then perhaps something like Manchester to Bournemouth gaining a second unit in Stafford (chosen as it is more likely to have capacity) and Leaving a unit in Reading would help. How many units that saves is not clear, but it could enable more double running through Birmingham. Likewise a unit could attach at Bristol Temple Meads and detach at Derby or Sheffield on the SW-NE axis.

However, then it is easy to see that Manchester to Reading and Bristol to Leeds makes more sense, but suddenly the splitting does not work and there are far fewer units saved to create the doubles in the first place.

Often having suitable connections is sufficient, but that does rely on the main line running well (which has been a problem for Windermere)

 

[David Goddard]

With the now more wider use of five car express trains, the scope for splitting and joining for core route strengthening of our long distance services is more likely. Not just with two destinations but also just for a double set over the core route, which is split at a major stations and the dropped set then making up the next service the other way.
Examples being GW services dropping portions at Plymouth or Cardiff, or LNER's North of Edinburgh trains dropping one off at Edinburgh.
Elsewhere, splits could appear on West Coast with a double 805 splitting into Chester/Blackpool portions at Crewe.

 

[PG]

With the now more wider use of five car express trains, the scope for splitting and joining for core route strengthening of our long distance services is more likely.

IIRC weren't GWR experiencing problems with joining 5 car sets causing them to reduce the number of such moves scheduled?

 

[irish_rail]

IIRC weren't GWR experiencing problems with joining 5 car sets causing them to reduce the number of such moves scheduled?

Yes. Splittng and joining at Plymouth hasn't been terribly successful. Only last week the 0655 was delayed at Plymouth joining up and it caused knock on delays to various services right the way to London. It also incurs quite big staff costs on traction that doesn't have a walk through (and even if it does there are still increased driver costs). Splitting and joining has a place, but not on many routes id argue.

 

[zwk500]

Another big problem with split/join is driver diagrams.

 

[SussexSeagull]

I experience this quite a lot coming back from London to Durrington with the train splitting at Haywards Heath and they should be doing a lot more to point out to people which part of the train they need to be in by such things as putting a sign where the split will happen and having numbers on the outside of carriages.

 

[zwk500]

I experience this quite a lot coming back from London to Durrington with the train splitting at Haywards Heath and they should be doing a lot more to point out to people which part of the train they need to be in by such things as putting a sign where the split will happen and having numbers on the outside of carriages.

A sign on the platform would be helpful, but they do (or at least did) have displays on the outside of each carriage that had the carriage number on.

 

[PG]

Does the Ely/Kings Lynn service from London still split/join at Cambridge now that the overheads aren't limited to 4 coach EMUs? It always seemed a very slick operation.

 

[SussexSeagull]

A sign on the platform would be helpful, but they do (or at least did) have displays on the outside of each carriage that had the carriage number on.

In my experience they tend not to but I have seen it. Also consistently telling people who might have rushed for a train what carriage they are in, although you usually get time to switch at Haywards Heath if you are in the wrong bit.

 

[PGAT]

Splitting and joining block the platform for a long time. This is the reason why Southern no longer splits / joins the Eastbourne service because there isn't enough platform space.

They still do divide and attach at Eastbourne regularly. Not doing so doesn't really save time there, because the train has to reverse regardless.

 

[43074]

Does the Ely/Kings Lynn service from London still split/join at Cambridge now that the overheads aren't limited to 4 coach EMUs? It always seemed a very slick operation.

There are still a handful on weekdays and Saturdays that run as 12-car between Cambridge and London which attach or detach 4 cars at Cambridge, and most Cambridge fast services on Sundays are 12-car between Cambridge and London and 8-car north of there.

 

[Magdalia]

Does the Ely/Kings Lynn service from London still split/join at Cambridge now that the overheads aren't limited to 4 coach EMUs? It always seemed a very slick operation.

No.

The OHL has been cleared for 8 car trains for a long time. Until recently the problem was short platforms.

 

[dk1]

Performance nightmare.

 

[zwk500]

They still do divide and attach at Eastbourne regularly. Not doing so doesn't really save time there, because the train has to reverse regardless.

They divide at Eastbourne but the Eastbourne services no longer divide at Hauwards Heath.

 

[bramling]

Running longer trains helps to lower costs/PAX and make better use of the paths in congested areas. However, this goes against the constant clamour for smaller cities to receive direct links to everywhere. To that end, it's somwhat suprising that the use of split/join moves is quite limited across the network, as this potentially solves both problems.

Off the top of my head:

ECML
There are lots of 5-car sets running down the ECML every day which could be good candidates. The GC services to Bradford and Sunderland are the most obvious. If we ignore the different operators (i.e. assume a post GBR world), then Hull and Middlesborough could be other options. There could easily be scope for 2ph to split/join at Doncaster and head off to the various cities around Yorkshire and the North-East.

Lake District
Northern currently operate a service from Manchester Airport to Barrow. Making this 6-car and splitting at Lancaster, with 3-cars heading to Windemere, could be a good way of providing a regular service to both Lake District branches.

XC
A few different options for XC (assuming they operate double voyagers...), but splitting at Exeter for Paignton/Plymouth, or at Southampton for Bournemouth/Brighton, could be interesting options.

EMR
The Liverpool to Norwich service could be a good candidate to split at Ely, with half reversing to Norwich and half continuing to Cambridge/Stanstead.


I'm sure there are more ideas out there. Where else on the network do you think there may be opportunities?

The big disadvantage is of course performance, as delays to one service can cascade onto the 'joining' service. However, with suitable padding in the schedules, this should be manageable. In addition, most of these schemes would only be deliverable as part of a timetable re-cast, but again that is not a show-stopper.

I’d say no. Too much of a performance risk.

The only situation where it really works are where

* one portion splits off, works to a destination and doesn’t re-attach to anything

or

* very contained operations like Caterham/Tattenham Corner where there’s simply not too much scope for anything to go wrong

 

[Magdalia]

There are still a handful on weekdays and Saturdays that run as 12-car between Cambridge and London which attach or detach 4 cars at Cambridge, and most Cambridge fast services on Sundays are 12-car between Cambridge and London and 8-car north of there.

I forgot the 12 car trains!

 

[Manutd1999]

Splitting and joining block the platform for a long time.

Another big problem with split/join is driver diagrams.

The key might to carefully select where to split/join. Ideally it should be at a large station with plenty of platform capacity and where driver changes are happening anyway. If GBR ever actually happens the opportunities may increase further, as drivers would be less constrained to single TOCs.

one portion splits off, works to a destination and doesn’t re-attach to anything

Agreed this is a more common approach, but it doesn't offer the same benefits in terms of several small destinations 'sharing' a service.

 

[MattRat]

The problem with a plan like this, is that although it may work in theory, in reality it will fall apart very quickly, especially with how shoddily built most new trains are.

 

[irish_rail]

The key might to carefully select where to split/join. Ideally it should be at a large station with plenty of platform capacity and where driver changes are happening anyway. If GBR ever actually happens the opportunities may increase further, as drivers would be less constrained to single TOCs.


Agreed this is a more common approach, but it doesn't offer the same benefits in terms of several small destinations 'sharing' a service.

Yes but the station you describe is Plymouth to a tee. And yet GWR have (wisely in my view) dropped the vast majority of splitting and joining.

 

[DaventryRabbit]

It is always going to be a balancing act
Consider, say, EMR at Nottingham:
If an incoming train from Norwich is, say, 1/2hr late because of problems in East Anglia, EMR Control at Nottm has to decide weather to:
Send out the 2 Car 158 (waiting at Nottm for attatchment) on time towards Liverpool, and incur the wrath of passengers arriving ex Norwich who want to go onward, or
Wait for the incoming and have it running late further north which has implications for the return leg etc.

Do they have it running to time past Nottm but short Formed, causing overcrowding, or correctly formed but significantly late?

Sometimes their call may pay off and sometimes it won't. That is a fact of life

 

[ComUtoR]

Another big problem with split/join is driver diagrams.

In what way ?

 

[zwk500]

In what way ?

The first problem is making compliant diagrams that don't have big empty periods within the constraints of a split/join service, although this is manageable. The second problem is when disruption hits, you have portions stuck in platforms waiting for drivers or guards stuck on delayed inbound workings

 

[ComUtoR]

The first problem is making compliant diagrams that don't have big empty periods within the constraints of a split/join service, although this is manageable. The second problem is when disruption hits, you have portions stuck in platforms waiting for drivers or guards stuck on delayed inbound workings

Gimmie a day. I'm kinda weirded out on your comment. I don't find Driver diagrams to be an issue here but gimmie a day to check and a rethink. Apologies, my comment was a little off the cuff and it's late night and I wasn't expecting an expedited reply. Shift work sucks !

 

[zwk500]

Gimmie a day. I'm kinda weirded out on your comment. I don't find Driver diagrams to be an issue here but gimmie a day to check and a rethink. Apologies, my comment was a little off the cuff and it's late night and I wasn't expecting an expedited reply. Shift work sucks !

No worries, my originaly comment could have been worded better I suppose.

 

[Manutd1999]

Consider, say, EMR at Nottingham:
If an incoming train from Norwich is, say, 1/2hr late because of problems in East Anglia, EMR Control at Nottm has to decide weather to:
Send out the 2 Car 158 (waiting at Nottm for attatchment) on time towards Liverpool, and incur the wrath of passengers arriving ex Norwich who want to go onward, or
Wait for the incoming and have it running late further north which has implications for the return leg etc.

This scenario is made worse by having the split in the middle of the journey.

If, for example, all cars continued to Ely, the performance risk would be reduced. 2-Cars could then reverse to Norwich and the other 2 could either wait at Ely for the next service, or continue to Cambridge/Stanstead.

 

[43074]

This scenario is made worse by having the split in the middle of the journey.

If, for example, all cars continued to Ely, the performance risk would be reduced. 2-Cars could then reverse to Norwich and the other 2 could either wait at Ely for the next service, or continue to Cambridge/Stanstead.

How does moving the split to a location miles away from EMRs unit and traincrew depots help performance? At least at Nottingham most services change crew so there is an option for late running services to be restarted with a fresh crew and unit to stop the delay being carried all the way across the country. Likewise if a unit swap is required (for whatever reason) Nottingham is fairly operationally convenient to EMR.