Circles to Go

[londonstuff]

I've always wondered what's meant by flat junctions...

[mrfs42]

A flat junction is a a non-grade separated junction.

Think of Aldgate, there are three junctions there, one at each apex of the triangle - I know junction pedants will claim that there are in fact more, if you take into account the two terminal bays.

As a slightly easier conceptual example, think of Baker Street (upstairs) - where the Met. branches off from the Circle, this is a flat junction. Met trains to/from the City will delay Outer Rail Circles that will be held in platform 5 waiting for the Met. to clear the junction.

If there was a magic wand that could be waved to create space, the Met. heading from the City could diverge 'into' the centre of the Circle, pass underneath the Outer and Inner Rails and debouch onto the surface beyond the former flat junction - thereby removing part of the conflict by dint of vertical (i.e. gradient) separation.

Indeed flat junctions are a cause of delay, but these can be minimised with careful scheduling, design of signal spacing and judicious use of 'draw-ups': like MB16/17 that you can see between the Circle platforms at Baker Street. Of course any signalling system that is based on 'fixed block' cannot cope with every vagary of out-of-course running - but delays can be minimised. The oft-quoted example of the Districts use of multiple home signals on the approach to stations is an apposite example of allowing trains to close up, legimately reducing the spatial separation and keeping trains moving.

With a flat junction, there will always be an element of 'hang time' with on train waiting for another to clear the junction. Flat junctions, with both routes on the same level are a considerably cheaper option that grade-separated junctions.

[Chris M]

A possibly simpler explanation is to use the anaology of road junctions.
Think of a cross roads controlled by traffic lights, this is a flat junction. Traffic going north-south cannot travel across the junction at the same time as traffic going east-west.
Now think of a motorway junction, for example the M4-M25 junction. This is a grade separated junction, meaning that there is vertical separation between north-south and east-west traffic so both can proceed at the same time.

The junctions mentioned by MRFS are basically T junctions, of which there are many millions on the road network. Traffic on the main road has to stop to let traffic from the side road join, and similarly traffic on the side road has to wait for traffic on the main road.
An example of a grade-separated T-junction is where the M11 meets the north circular - there are bridges over the road so that clockwise traffic on the north circ doesn't have to stop and wait for anti-clockwise traffic joining or leaving the M11.
www.cbrd.co.uk/reference/interchanges/directionalt.shtml may help you picture it.

[mrfs42]

I didn't think of the analogy of road junctions - and I'd forgotten the cbrd site.

I suppose the nearest to capturing the ethos of a railway grade separated junction is the Partially Unrolled Cloverleaf, which if you remove the really steep bits in the middle will give you the general flavour of a railway version. Does help if you can visualise in the standard 4 dimensions, though.

[londonstuff]

So the underpass going from High St Ken to Earl's Court is a non-flat junction because it allows trains to connect to another bit of the track without interfering with the ones coming from Gloucester Road?

Whereas where the Circle and District converge at Gloucester Road is a flat junction because one holds up the other?

Is that about right?

[Harsig]

Oct 16, 2008 9:42:02 GMT londonstuff said:

So the underpass going from High St Ken to Earl's Court is a non-flat junction because it allows trains to connect to another bit of the track without interfering with the ones coming from Gloucester Road?

Whereas where the Circle and District converge at Gloucester Road is a flat junction because one holds up the other?

Is that about right?

That's it. It's as simple as that.

[Dmitri]

Oct 16, 2008 9:42:02 GMT londonstuff said:

So the underpass going from High St Ken to Earl's Court is a non-flat junction because it allows trains to connect to another bit of the track without interfering with the ones coming from Gloucester Road?

Whereas where the Circle and District converge at Gloucester Road is a flat junction because one holds up the other?

<pedant>
About right , however physical layout is primary, that is:

1) trains can connect to another bit of the track without interfering with the ones coming from Gloucester Road due to existence of the non-flat junction;

2) Circles and Districts hold up each other at Gloucester Road because there is a flat junction there.
</pedant>

[ribaric]

Oct 16, 2008 9:42:02 GMT londonstuff said:

So the underpass going from High St Ken to Earl's Court is a non-flat junction because it allows trains to connect to another bit of the track without interfering with the ones coming from Gloucester Road?

Whereas where the Circle and District converge at Gloucester Road is a flat junction because one holds up the other?

Is that about right?

[another pedant] The classic 'flying' junction (non-flat) is at North Acton junction on the CL. These junctions remove the need for 'X' type crossovers so anything moving in any direction does not conflict with anything moving in the opposite direction. Trains moving in the same direction are still required to merge over trailing points of course. Woodford junction is a classic of the 'flat' variety [/another pedant]

Best not to get into Camden Town eh.

[cetacean]

Oct 15, 2008 12:13:50 GMT Ben said:

From an engineering perspective it doesnt seem too much a problem; granted the crossovers to gain access to the centre platforms might have to be at Farringdon and Moorgate

I was in the area yesterday (Thameslink diverted to Moorgate) and I can see how you'd build the various connections at the west end on the site of Farringdon sidings. But what happens at the east end? Presumably you bend the inner rail south, diagonally across the whole terminus platform area at Moorgate? Would there be space to do this and still have a straight-ish platform on it?

[metman]

Yes, this is exactly what I was thinking of doing. Some pics:

i100.photobucket.com/albums/m6/Adthelad/DSCN0685.jpg
i100.photobucket.com/albums/m6/Adthelad/DSCN0688.jpg
i100.photobucket.com/albums/m6/Adthelad/DSCN0687.jpg
i100.photobucket.com/albums/m6/Adthelad/DSCN0686.jpg

Barbican has been reduced to three platforms to provide more platform space although Barbican could be left unaltered. The exits at Moorgate would have to be moved to provide space for the new Westbound Circle line platform.

[cetacean]

Interesting. Are you planning to do some tunnelling east of Moorgate? I'd have thought the only reasonable scheme would be to start turning south somewhere near the buffer stops of platform 3, keeping everything within the existing platform area.

[metman]

That's the plan-maybe not obvious on the diagram!

[Ben]

Interesting idea metman, surely though it would be logistically easier to have the junction to the west of Moorgate station, afterall there were once connections there previously between mainline and LT, and it would avoid tunneling. Is there any reason why all 4 platforms cant be used at Barbican? The two outer ones are the same width seemingly, surely the disused one could be brought back into service for the (new) westbound trains?

[metman]

If the junction is west of Moorgate you loose all the benefits of a rebuild. Terminating trains still have to cross the path of Westbound trains.

The Barbican platforms are really too small. A tip out could mean overcrowding. Platform 4 is quite narrow too. Barbican doesn't need to be rebuilt however, its just an option.

Bringing the current platforms 5-6 at Moorgate into action is the really aim of the rebuild.

[ribaric]

Just saw your sketches Metman, this is very close to a study in which I was involved in 2005.

Then, the idea was born from the need to reduce the flat working through Aldgate junction by shorting all or most Mets at Moorgate. A new IR tunnel was planned to run from west of Liverpool Street to the southern-most platform at Moorgate and then continue over the old widened lines to rejoin the existing IR main line immediately east of Farringdon. This would effectively twin the existing IR between Moorgate and Farringdon which would be used almost exclusively by the Met services coming out of Moorgate. This meant that OR Mets terminating at Moorgate would not be in conflict with the IR Circles and H&C services. The provision of a higher speed diverging point set would speed up the approach of Mets into Moorgate and further assist the remaining OR services.

We had a bunch of timetables written which showed immense promise but there are some serious problems.
The cost of the new tunnel is big expense, followed by very difficult track and signal work at Moorgate and Farringdon - the latter would lose it's sidings and create another huge problem for early and late trains. We never did overcome the problem of OR Met passengers (sorry - "customers") wanting to go to Liverpool St. from having to step back. This would probably happen at Barbican where the platform capacity is woeful. The plans remain in the offices at 'Major Upgrades' (my last job) but I don't believe this particular bird will ever fly.

I'm sorry if this is a "Hitler invades Poland" piece of news but I haven't the patience to read the whole of this thread.

[cetacean]

Oct 20, 2008 20:02:50 GMT metman said:

If the junction is west of Moorgate you loose all the benefits of a rebuild. Terminating trains still have to cross the path of Westbound trains.

The Barbican platforms are really too small. A tip out could mean overcrowding. Platform 4 is quite narrow too. Barbican doesn't need to be rebuilt however, its just an option.

Bringing the current platforms 5-6 at Moorgate into action is the really aim of the rebuild.

The scheme I was thinking of would sacrifice all the terminating platforms at Moorgate (two are shortly out of use, two are hardly used) in favour of a connection between the IR from Liverpool Street to the widened lines, passing the southernmost platfrom at Barbican antrejoinging the IR via Farringdon Sidings (which are doomed anyway, aren't they?). Also at Farringdon Sidings there'd be pointwork to allow trains from the west to terminate in the two middle platforms at Barbican.

And the aim would be to extend new-Circle or Wimbleware trains, rather than shorten the Met.

[Deleted]

Wouldn't a solution for Barbican simply to have just two larger platforms and two through tracks - demolishing the other two platforms to make more room - then you'd still get the extra capacity through to Moorgate - you'd just have to have some trains - say met line - not stopping at Barbican.

[Chris M]

Would a layout like Loughton be feasible - two island platforms with a bi-directional centre track accessible from both platforms? This would allow trains to terminate from either direction and able to reverse without tipping out, or to allow two trains in one direction in the station at one time.

[cetacean]

The problem with changing the layout is that the Circle and Thameslink pairs of tracks run in separate tunnels immediately either end of Barbican station. Perhaps the wall between them is demolishable, I don't know.

[metman]

Barbican should not become a terminal station because it does not have the capacity to cope.

I believe there is the remains of old tunnels so anything is possible in the Barbican area!

[Deleted]

You've got the remains of the old Smithfield market siding and Blackfriars connection tunnels to the west of Barbican, but to the east the lines are basically in the lower basement of the Barbican Centre.

There's around 150 yards of the original tunnel left leaving the station eastwards, but beyond that the line was realigned in the 60s. Wartime damage had left the area between Barbican and Moorgate literally razed to the ground.

As part of the redevelopment the lines between the stations were moved southwards, into concrete tunnels which became part of the structural foundations for the Barbican Centre above. I think that anything that involves altering these is going to be impossibly expensive and disruptive.

Any scheme would have to avoid this and involve only works west of Barbican and east of Moorgate to keep to any sort of budget and timescale. Still would be a massive project though, similar to grade-separation at Baker Street.

Perhaps the easiest way would be to close Barbican for a couple of years convert to 3 track 2 island layout shifting the platforms 100 yards west, opening up the tunnels to the west for construction work and re-covering after completion. That would give you enough space for pointwork at the east end, where the stairs are at the moment.

Would still leave the BR platforms at Moorgate unused though!

[Deleted]

i am doing this in my transport management disertation.

the idea i have is this:

Hammersmith- Paddington-Moorgate every 10-15 minutes
Wimbledon- Paddington-Moorgate- Barking every 10-15 minutes

with the improved wimbledon service from what i can make out using an old london tube map the service will be like a diamond with passengers wanting to travel from edgware road to tower hill etc either having to change at earls court or aldgate east.

if crossrail is built however i believe this may take some pressure off the cirle line etc .

[Chris M]

As I understand it, unless you rebuild Moorgate to remove the flat junction (as suggested on this forum several times), it couldn't cope with that level of reversing.

[mrfs42]

Nov 1, 2008 22:16:43 GMT Chris M said:

As I understand it, unless you rebuild Moorgate to remove the flat junction (as suggested on this forum several times), it couldn't cope with that level of reversing.

It could - but not without enhancing delays. Likewise Edgware Road is going to be interesting if ths service alteration is enacted.

[Deleted]

How is going beyond Edgware Road at all viable with the future (and even the existing) Met. line service, especially in the peaks? As far as I understand there'd be quite a few more trains on the Met in the future (I've heard talks of 10 tph on the Uxbridge branch), and I'm guessing some of those would go through to Aldgate. If not you'd run into problems with more trains terminating at Baker Street I suppose.

[metman]

I don't think 10tph will happen on the Uxbridge service now, at least until ATO comes in. There won't be enough trains. Only 58 S8s are expected-the same as the initial A60/62 order!

[Deleted]

On Thursday I got on a 'Circle Line' train from Hammersmith. Is this a regular occurrence? Never come across it before and thought maybe some of the T-Cup already existed or was used for busy periods (Westfield opened that day).

[upfast]

Nov 3, 2008 2:31:48 GMT @cjc said:

On Thursday I got on a 'Circle Line' train from Hammersmith. Is this a regular occurrence? Never come across it before and thought maybe some of the T-Cup already existed or was used for busy periods (Westfield opened that day).

Most likely train that has just come back into service from the depot due to no train operator or being defective. Or was sent to Hammersmith to reverse because it was running late, or to supplement the service on the Hammersmith branch due to crowding or cancellations.

[ruislip]

Nov 2, 2008 22:51:44 GMT metman said:

I don't think 10tph will happen on the Uxbridge service now, at least until ATO comes in. There won't be enough trains. Only 58 S8s are expected-the same as the initial A60/62 order!

If and when that happens, will that affect the Piccadilly service on the Uxbridge line?

[metman]

That, I don't know! With a train every 6 minutes, there leaves little room for the Piccadilly.