Impact of platform staffing arrangements on performance of the 'Castlefield Corridor'

[yorksrob]

Agreed ...but nevertheless the chord has achieved exactly what it was intended to do. As regards the biggest white elephant I think you could find a great many more in the military fields from TSR2 onwards.

And don't forget all those I.T. systems that never worked.

The talk of signalling issues is instructive. Were 16/17 to be built, it would be an opportunity to review the signalling.

 

[notlob.divad]

Not necessarily all that common - Hamburg, Berlin, Brussels, Cologne (as examples) 'disperse' over multiple stations.

When done properly, it helps operationally by spreading boarding/alighting passengers across stations so its not all concntrated at one (with effects on dwell times).

You missunderstood what I was saying. I was not saying that everyone has to go to one station. I was saying that once you reach a station in the city you want to go to, you use the local transportation systems, S-Bahn, U-Bahn, Trams, buses etc to get around the city, interchange between stations, or even dare I say it, get to the cities airport.

A simpler layout (less pointwork/platforms) if anything might make what's left that bit faster and actually pump things through that bit quicker.

I quite agree. Given the two platform limitation at both Deansgate and Piccadily, I feel there would be benefits rationalising the whole layout to plain line all the way through from Slade Lane to Castlefield Junction. Less things to go wrong = Less chances of delays.

 

[edwin_m]

A major part of the problem is signalling that's still based on Victorian operating practices. Those who drive on roads are not subject to such block systems. If there's nothing in front we can drive up to the speed limit (and some beyond!) and should take account of corners, gradients other traffic and weather. There are lots of accidents, mostly minor, but there are millions of vehicles making countless accident free journeys every day. In heavy traffic they'll close up. At faster speeds on motorways they should be further apart.

Clearly trains can't run like that. However, one train creeping along 100-200 yards behind another at 5-10 mph should be possible in congested sections like this. How far can this be done now, and how much more could be achieved? Surely modern technology should be able to permit safe closing up?

I compare this congestion bottleneck with Totley Tunnel where there's a block of 4 miles between Totley Tunnel East and Grindleford (as laid down in 1893), another bottleneck particularly impeding that Liverpool - Norwich service!

Trouble is signalling systems take decades to change and each advance is overtaken before it can be expensively introduced across a fraction of the network!

A digital signaling system was being pushed by Grayling at one point. There are moving block (or very short block) signaling systems that achieve what you suggest, including several lines of the Underground, and Automatic Train Operation operational on Thameslink though not yet routinely used. However in a low-speed area such as this the closing-up signals achieve nearly the same result so the benefit of advanced signalling may be limited - the critical issue is how long the train spends in the platform not how long between it leaving and the next one arriving.

At Oxford Road there are also signalling overlap issues which mean a longer train can't even arrive in one of the platforms while one is leaving the other platform in the same direction. This is why the planned re-modelling extended the four-tracking beyond the platform ends. And to pick up another point, closing Oxford Road or having fewer trains stop would worsen the overcrowding at Piccadilly.

 

[Fawkes Cat]

Ok. I'm still looking for quick fixes - things that could be done without actually changing the infrastructure. So please could anyone advise

- do customers on Castlefield corridor services end up in any number in the Deansgate area or anywhere where they could sensibly connect from Deansgate/Castlefield tram stop? What I am thinking here is that all trains pass through Deansgate - so would the time lost in scheduling them to call there be repaid by reducing dwell times at Oxford Road and Piccadilly?
- are the 'spare' platforms at Oxford Road (1 and 3 I think) capable of taking terminating trains from the west, and then taking departures back west again? I get that for practical purposes they are limited to four car trains, but can they be used when there are through trains in platforms 2 or 4? If they can, that would in the short term go some way to meeting the lack of west-facing terminating platforms in Manchester.

 

[Bletchleyite]

Platform 1 used to be used to terminate the Warrington Central/Irlam stopper - I don't know if anything has been changed that would now prevent that?

 

[The Planner]

And don't forget all those I.T. systems that never worked.

The talk of signalling issues is instructive. Were 16/17 to be built, it would be an opportunity to review the signalling.

That is likely one of the issues that is bubbling under, making changes to the corridor is likely to trigger a wholesale resignalling of the area, that is a once in a control period job and certainly wouldnt be a CP6 item. Picc box is 40 years old in 2028 I think, so its falling into the right timescales at the end of CP7.

 

[LowLevel]

Platform 1 used to be used to terminate the Warrington Central/Irlam stopper - I don't know if anything has been changed that would now prevent that?

There's no step free access to platform 1 so it's used as little as possible

 

[notlob.divad]

Platform 1 used to be used to terminate the Warrington Central/Irlam stopper - I don't know if anything has been changed that would now prevent that?

It would have to run 'wrong road' all the way from Deansgate East when arriving. So I am not sure with the current setup this would help much.

Reversing from Platform 3, also has the same problem as departure from the existing bay platform. as highlighted in the report. You need a gap in trains in both directions. (I have just seen it happen on Tracksy). 2073 departed Platform 3 for Liverpool Lime Street.

 

[Bletchleyite]

There's no step free access to platform 1 so it's used as little as possible

Though I'm sure installing something would be cheaper than rebuilding the entire station?

I guess the advantage of a 2 platform through station with a central bay is that reversing doesn't obstruct any line (and if you open doors on both sides it's cross-platform to and from Picc, too). But would the listed aspect of the station prevent such a heavy rebuild?

 

[Ianno87]

Though I'm sure installing something would be cheaper than rebuilding the entire station?

I guess the advantage of a 2 platform through station with a central bay is that reversing doesn't obstruct any line (and if you open doors on both sides it's cross-platform to and from Picc, too). But would the listed aspect of the station prevent such a heavy rebuild?

Not just the lack of step free access - it's quite a narrow platform too and wholly unsuitable for a busy city centre station, really. No point installing a lift when the platform is pretty much unfit for purpose and could do with a proper re-build anyway.

Re-build with 3 through platforms not 4, and make what's left wider (and the layout faster).

 

[Ianno87]

So why, then, are 6x26m trains an issue here, as it's not about stacking them up? Once they've cleared the signal you can signal another in. The length of the train is surely relatively unimportant here.

At a low speed, the extra time that the additional train length takes to clear the platform from a stand is quite a high proportion of the platform re-occupation times. All the seconds here and there add up.

 

[Bletchleyite]

At a low speed, the extra time that the additional train length takes to clear the platform from a stand is quite a high proportion of the platform re-occupation times. All the seconds here and there add up.

So, back to what others were saying...given that this is a very low-speed line, could permissive (line of sight) working on the entire stretch be feasible if it was simplified down so there were no points between Castlefield Junction and the south end of 13/14 and only single-direction working on each line? You could properly stack units up then.

Or does line of sight require track-braking?

 

[jfollows]

could permissive (line of sight) working on the entire stretch be feasible if it was simplified down so there were no points between Castlefield Junction and the south end of 13/14 and only single-direction working on each line?

I'm no expert and will be happy to be proved wrong, but there was a huge clamp-down to prevent permissive working of passenger trains in the past, 1980s perhaps, and I can't see that being reversed. I remember Stockport station, arriving in platform 2 (86+stock) at the same time a local service was leaving the same platform, so it clearly saved time. I just get the feeling that it'd be a big no-no today.

 

[Ianno87]

So, back to what others were saying...given that this is a very low-speed line, could permissive (line of sight) working on the entire stretch be feasible if it was simplified down so there were no points between Castlefield Junction and the south end of 13/14 and only single-direction working on each line? You could properly stack units up then.

Or does line of sight require track-braking?

The closest you'd get would be Thameslink-core-esque signalling enabling frequent signal sections so trains could get close together; about as close to 'line of sight' as would credibly be allowed.

Difference is that, unlike Castlefield, Thameslink doesn't take freight trains in normal operation.

I think ETCS Level 2 (without any conventional lights on sticks) might be the ultimate solution; freights could proceed on much longer Movement Authorities through the area to account for their braking needs; whereas passenger multiple units would only need relatively short Movement Authorities between relatively closely spaced marker boards so could 'close up'. But the amount of stock that would need to be ETCS-ready to cover this section would be huge!

 

[edwin_m]

Though I'm sure installing something would be cheaper than rebuilding the entire station?

I guess the advantage of a 2 platform through station with a central bay is that reversing doesn't obstruct any line (and if you open doors on both sides it's cross-platform to and from Picc, too). But would the listed aspect of the station prevent such a heavy rebuild?

Perhaps something like this:

• Platform 1: Abandoned
• Platform 2: Continues as main westbound platform
  
• Platform 3: Becomes central turnback platform, either totally (with buffer stop) or mainly for trains from the east turning back in that direction including current CLC stoppers. This means that all westbound passengers would go to the same island for either 2 or 3
• Platform 4: Continues as main eastbound platform
• Platform 5: Abandoned

Removal of P1 and P4 probably allow simplification of station throats and lengthening of remaining platforms, but I don't know how much. Removal of P1 may allow some widening of the P2/3 island. This can probably all be done without changing the wooden canopies which I guess are the main reason for listing.

So, back to what others were saying...given that this is a very low-speed line, could permissive (line of sight) working on the entire stretch be feasible if it was simplified down so there were no points between Castlefield Junction and the south end of 13/14 and only single-direction working on each line? You could properly stack units up then.

Or does line of sight require track-braking?

I'm no expert and will be happy to be proved wrong, but there was a huge clamp-down to prevent permissive working of passenger trains in the past, 1980s perhaps, and I can't see that being reversed. I remember Stockport station, arriving in platform 2 (86+stock) at the same time a local service was leaving the same platform, so it clearly saved time. I just get the feeling that it'd be a big no-no today.

I think it's very unlikely to be allowed without it. I don't believe there's anywhere moving passengers trains follow each other on line of sight, only moving trains approaching stopped trains. The risk is that the train in front may stop suddenly, for example due to inadvertent DSD release or various other systems "failing safe". Without brake lights that's probably quite hard to discern from directly behind, especially in the dark. Train brakes also take several seconds to apply properly and are less powerful than on trams, so there would be a significant risk of slamming into the train in front.

Actually the benefit of stacking up trains is questionable. If there is disruption this just means they end up nose to tail instead of spread along the line, so if it turns out the line is blocked indefinitely then they would all have to be reversed out rather than most of them proceeding to a station under normal working arrangements. There's not much benefit for scheduled working either, as with the slow speed and close signals each train can already be pretty close to the one in front. ​

 

[Bletchleyite]

I think ETCS Level 2 (without any conventional lights on sticks) might be the ultimate solution; freights could proceed on much longer Movement Authorities through the area to account for their braking needs; whereas passenger multiple units would only need relatively short Movement Authorities between relatively closely spaced marker boards so could 'close up'. But the amount of stock that would need to be ETCS-ready to cover this section would be huge!

Yes, this is probably somewhere ETCS would give a benefit - or as you say short sections where a freight would occupy more than one - not a rural branch line in Wales!

Much as there seems to be some resistance to a MancLink, that doesn't mean some concepts of it wouldn't work.

 

[Bletchleyite]

Perhaps something like this:

• Platform 1: Abandoned
• Platform 2: Continues as main westbound platform
  
• Platform 3: Becomes central turnback platform, either totally (with buffer stop) or mainly for trains from the east turning back in that direction including current CLC stoppers. This means that all westbound passengers would go to the same island for either 2 or 3
• Platform 4: Continues as main eastbound platform
• Platform 5: Abandoned

Removal of P1 and P4 probably allow simplification of station throats and lengthening of remaining platforms, but I don't know how much. Removal of P1 may allow some widening of the P2/3 island. This can probably all be done without changing the wooden canopies which I guess are the main reason for listing.

Yeah. The MKC platform 2 or 5 layout (or Tring "bay") is probably worth looking at as a concept.

The listing is a nuisance, as the most useful layout would probably be to flatten the lot including all buildings and have two long island platforms with a single track between them (which could be used for terminating trains in both directions, but would primarily be used from the west) and doors released on both sides, making the enforced change there painless as it'd always be cross-platform. That said, I don't know if the viaduct would support that.

 

[edwin_m]

Yeah. The MKC platform 2 or 5 layout (or Tring "bay") is probably worth looking at as a concept.

The listing is a nuisance, as the most useful layout would probably be to flatten the lot including all buildings and have two long island platforms with a single track between them (which could be used for terminating trains in both directions, but would primarily be used from the west) and doors released on both sides, making the enforced change there painless as it'd always be cross-platform. That said, I don't know if the viaduct would support that.

Short of that, it would be helpful if the P2/3 island could be widened enough to have an access at the eastern end down onto Oxford Street. This might be needed anyway in a rebuilt station for evacuation purposes, and would reduce pressure on the footbridge as many people heading for westbound trains would find it a quicker route. Maybe something similar at the other end too - but it all depends on the viaduct structcure.

 

[Tomnick]

I quite agree. Given the two platform limitation at both Deansgate and Piccadily, I feel there would be benefits rationalising the whole layout to plain line all the way through from Slade Lane to Castlefield Junction. Less things to go wrong = Less chances of delays.

It'd represent an enormous loss of flexibility south of Piccadilly though - at present, the layout has lots of options to avoid conflicts, and I think it's inevitable that there'll be some trains from the slow lines into the main shed still. As an example, Up trains from the Castlefield corridor towards Stockport tend to be booked slow line to Slade Lane in my experience, but if that's going to result in a conflict with Down Fast trains (because one or the other is late), a decent signalman will turn the Up train out fast line either straight from the end of platform 13, or at Ardwick Junction, if there's a margin there instead. It happens reasonably frequently.

I'm no expert and will be happy to be proved wrong, but there was a huge clamp-down to prevent permissive working of passenger trains in the past, 1980s perhaps, and I can't see that being reversed. I remember Stockport station, arriving in platform 2 (86+stock) at the same time a local service was leaving the same platform, so it clearly saved time. I just get the feeling that it'd be a big no-no today.

Agreed, there's a definite preference against permissive working nowadays unless it's absolutely necessary. The risk is considered too great - see the recent-ish collision at Plymouth. Having two trains moving simultaneously in the same signal section is a definite no-no, as much as there's a clear capacity benefit as you describe ("Huddersfield Control", which prevents it in the interlocking, has been provided in signalling installations for at least two or three decades now). The tight curves into both Piccadilly and Oxford Road - you can't see very much at all entering Picc on the Up, especially if the platform's crowded - just make it worse.

 

[Ken H]

suppose I am travelling from newcastle to the palace theatre. I could go to Victoria, go and get a tram to St Peters Sq and wak
Or I could await a train round to Oxford Road and get off there.

If I did the tram solution I would have to get an extra ticket for the tram. but nipping round to Oxford road would be included in my train ticket.

Maybe allowing BR tickets to Manc be used for the tram in Zone 1 would relive the problem a bit?

 

[Senex]

The listing is a nuisance, as the most useful layout would probably be to flatten the lot including all buildings and have two long island platforms with a single track between them (which could be used for terminating trains in both directions, but would primarily be used from the west) and doors released on both sides, making the enforced change there painless as it'd always be cross-platform. That said, I don't know if the viaduct would support that.

So might this be one of those very rare cases where the listing really should be cancelled on the grounds that this is an instance where there is total incompatibility between retaining the 1960s listed elements and having a station fully fit for purpose? Then flatten the whole lot and provide four decently-aligned through platforms with all connections at line speed. (Compare the Berlin Stadtbahn, where line-speed on a viciously-curvy railway is 60 km/h and all connections into the loops at Friedrichstraße, Hbf/Lehrter Bf, and Zoo are for the same speed — the Hamburg line has a similar 60 km/h maximum, but I don't remember if any of the intermediate stations have four main-line platforms.)

 

[Bletchleyite]

So might this be one of those very rare cases where the listing really should be cancelled on the grounds that this is an instance where there is total incompatibility between retaining the 1960s listed elements and having a station fully fit for purpose? Then flatten the whole lot and provide four decently-aligned through platforms with all connections at line speed. (Compare the Berlin Stadtbahn, where line-speed on a viciously-curvy railway is 60 km/h and all connections into the loops at Friedrichstraße, Hbf/Lehrter Bf, and Zoo are for the same speed — the Hamburg line has a similar 60 km/h maximum, but I don't remember if any of the intermediate stations have four main-line platforms.)

No, Dammtor is one island for IC/ICE/RE and one for S-Bahn.

 

[Killingworth]

So might this be one of those very rare cases where the listing really should be cancelled on the grounds that this is an instance where there is total incompatibility between retaining the 1960s listed elements and having a station fully fit for purpose? Then flatten the whole lot and provide four decently-aligned through platforms with all connections at line speed. (Compare the Berlin Stadtbahn, where line-speed on a viciously-curvy railway is 60 km/h and all connections into the loops at Friedrichstraße, Hbf/Lehrter Bf, and Zoo are for the same speed — the Hamburg line has a similar 60 km/h maximum, but I don't remember if any of the intermediate stations have four main-line platforms.)

Once we get into total reconstruction we face really serious delays, and for decade. Given the prevarication over much lesser schemes there seems almost a zero chance of that. We'll continue patching things up for ever.

I'm not an advocate of bombing, but the aftermath of wartime leaves much greater scope for radical redesign of all aspects in city centres. It also adds urgency. Peacetime guarantees eternal overlapping inquiries, focus groups, special interests and ever changing budgets.

 

[Bletchleyite]

Once we get into total reconstruction we face really serious delays, and for decade.

Given that you can build an office building in 2 months, how should rebuilding a station be that difficult?

 

[notlob.divad]

It'd represent an enormous loss of flexibility south of Piccadilly though - at present, the layout has lots of options to avoid conflicts, and I think it's inevitable that there'll be some trains from the slow lines into the main shed still. As an example, Up trains from the Castlefield corridor towards Stockport tend to be booked slow line to Slade Lane in my experience, but if that's going to result in a conflict with Down Fast trains (because one or the other is late), a decent signalman will turn the Up train out fast line either straight from the end of platform 13, or at Ardwick Junction, if there's a margin there instead. It happens reasonably frequently.

Sorry, got my junctions mixed up. I didn't mean Slade Lane, I meant the Piccadilly throat whats that Fairfield Street, of Temperance Street somethign like that.

By the way, what is the unused rusty track between Hoyle Lane and the A635. I was looking to see if you could turn the Mayfield siding into a central reversing rather than off to one side, and noted the unsed track down the middle of the formation.

 

[LNW-GW Joint]

Having read the NR report, it seems a good analysis of all the operational problems of the Castlefield corridor.
What I don't quite understand is how we have got to this position, given all the work over decades with the Castlefield network, much of it under the Northern Hub moniker.
How is it we are considering remodelling Irwell St Jn, which is only 2 years old, to deliver parallel running?
Were these problems not forseen when the Ordsall Chord was given the go-ahead?
If a centre reversing platform at Oxford Road is so obvious as an improvement, why isn't it in the current plans?
As the two CLC terminating trains seem to cause huge conflict, why not extend them out of the corridor completely?
The HS2/NPR dimension is not mentioned at all (possible underground route Pic-Vic-Miles Platting).

The NR view seems not to match that of the TOCs' plans (train lengths and types) which have been current since the 2015 franchise awards.
Longer trains (eg 6-car 323/195) seem to further screw up capacity at Victoria and the Airport as well as Oxford Road.
We seem to be a long way from a plan to resolve the issues.
Even the digital railway doesn't seem to be the answer.

There's a small typo in some of the wider diagrams - going straight on at Heald Green doesn't take you to Stoke-on-Trent.​

 

[jfollows]

By the way, what is the unused rusty track between Hoyle Lane and the A635. I was looking to see if you could turn the Mayfield siding into a central reversing rather than off to one side, and noted the unsed track down the middle of the formation.

Hoyle Street, I think (or, rather, http://www.streetmap.co.uk/map.srf?...Manchester,+M_12&searchp=ids.srf&mapp=map.srf thinks), and are you talking about the Midland Engine Siding? Very heavily used when locomotives ran at the front of trains. Rather a long time ago though!

Engines released by an outbound working followed the train out until the signal at the end of the platform, then went into the MES, and subsequently reversed onto a newly-arrived rake of stock.

Gives me an excuse to upload my track diagram from, I think, late 1970s when London Road box was still in use before platforms 13 & 14 were extended. A lot of things remain unchanged since then! Note that there was an East Engine Siding also at the time.

 

[Tomnick]

I’d agree that it’s the engine siding that you’re referring to. Not sure whether it has any booked use, but it’s handy for shunts between platforms in the main shed, particularly during disruption. It’s inaccessible from platforms 13/14 though. I’m sure something could be done though - slewing the Down Slow into the alignment of Mayfield Loop to free up room for a centre turnback might be preferable, or even do both to give you room for two sidings in the middle.

I doubt there’d be much to be achieved by removing the various connections between the station throat at Picc and Castlefield Junction. Most aren’t used frequently (and therefore are unlikely to fail in normal operation!), but are useful to work around problems when it’s all going wrong.

 

[Eccles1983]

The engine siding is only usable for station reversals - you cannot exit it in the up direction.

Unless you rip the lot up then I can't see the attraction in a centre siding.

The long and short of it is that without ripping it up or ripping up most of the buildings around it then it will remain as broken as it is - the only way of fixing it is not a uniformed stock, it's a removal of services.

 

[Ianno87]

Given that you can build an office building in 2 months, how should rebuilding a station be that difficult?

**Resists from getting crayons out to plan out how a phased re-construction might work**