Skyscraper City Forum banner
1 - 20 of 168 Posts

·
Registered
Joined
·
1,114 Posts
I live in Artarmon and certainly one of the original Metro route options was to stay above ground from Chatswood to St Leonards then do the tunnel dive there. I remember getting correspondence about it at the time.



As others have said, I think it was clearly right to do what they have done. It reduces duplication of the existing line, allows them to do a station at Crows Nest, plus probably have a much better work site to launch the TBMs with less disruption to the existing line (not sure what site at St Leonards would have been used). Also while the Artarmon - St Leonards corridor is technically wide enough it would indeed have involved the loss of a lot of established bushland (there was certainly a local campaign to retain that) and other works.



There was also early talk of a station in the Artarmon Industrial area but that gathered little support so I suspect canning that also influenced going into tunnel just after Chatswood.


I've never understood the people who say the track can be quadded all the way to North Sydney. Notwithstanding the above, yes it COULD be done as far as St Leonards. But I just don't see how it would be physically possible after that. Immediately south of the station there are now buildings right up to the rail corridor so it's nowhere near wide enough. You'd have to tunnel or undergo major property acquisition and demolition.
A railway easement is not "established bushland". The only reason it was preserved was to provide for later track amplification. And given it's adjacent to an operating line, it's hardly an oasis of bucolic wilderness. It beggars belief that anyone could be outraged by land being used for the purpose it was originally set aside for.

The Artarmon industrial area station should have been included in the metro plans (it was dumped without any real explanation), or at the very least passive provision should have been given for a future station. That industrial area is inevitably going to be turned into housing at some point. The same goes for the stretch between Waterloo and Sydenham, which should have had an intermediate station somewhere near Euston Rd (or passive provision for one).

The extra platforms at North Sydney could still be used for their originally intended purpose: a line up to the Northern Beaches, which would use the eastern Harbour Bridge lanes (which will be freed up with the metro taking a lot of buses/cars off the bridge).

Does anyone know if it would be possible to go from the old platforms 1/2 of Wynyard to the Metro West (Sussex St) alignment? A metro is capable of sharp-ish turns and much higher gradients than heavy rail (up to 4.5%), so it could loop around the existing tracks. But are there building basements or other structures in the way?
 

·
Registered
Joined
·
1,114 Posts
The reason is because Greater Sydney Commission has recommended that industrial land should be safeguarded against residential development:





https://www.greater.sydney/portal/n...y/jobs-and-skills-city/retaining-and-managing

https://www.smh.com.au/national/nsw...reater-sydney-commission-20181024-p50boo.html
Yes, but it's a waste to have low-density industrial land adjacent to TWO railway lines 10 minutes from the CBD. Much like Pyrmont, Green Square, the Bays Precinct, etc., it's an obvious candidate for conversion into medium to high-density residential areas, and eventually that's what will happen. The industrial land can be moved to somewhere in the same region away from high-capacity transport infrastructure.
 

·
Registered
Joined
·
1,114 Posts
So I was having a bit of a think about converting further Sydney Trains lines to metro after the Bankstown conversion. One thing that stuck out was the statistic that the expected 24 million annual riders on the NW metro (a conservative prediction) would cover 60% of operating costs. So full cost recovery would require 40 million riders.

As a rule of thumb, this would mean that a metro can achieve cost recovery with:
1.11 million annual riders per km of line (36km), or
3.07 million annual riders per station (13)

In reality it's a function of both, as operating costs are partly a factor of pure line length (and train frequency, but I'm assuming this remains a constant across different metro lines) and partly a factor of the number of stations, which have fixed maintenance and staffing costs (and slow down the trains that serve them).

So how would, for instance, T4 stack up as a metro line? Let's assume that Hurstville-Sutherland is quadruplicated (this would be the main capital expense, along with new signalling and platform works to make them metro compatible), and a metro service is run all stations from Bondi Junction-Cronulla, with a parallel, lower-frequency express service from Central on the Illawarra line for the stations south of Sutherland. Obviously Cronulla would be a long distance to serve with all-stations trains, but with better acceleration and dwell times than the double-decker stock a metro could come close to matching travel times for existing limited stops services.

As a metro, the line would have 31 stations (including Erskineville and St. Peters, which T4 will take over after the Bankstown conversion) over 41.6km.

To achieve full cost recovery, then, would require something between 46.2 million annual riders (41.6km x 1.11 million) and 95.2 million riders (31 x 3.07 million), the discrepancy coming from the much more closely spaced stations than on the NW metro (1.39km instead of 3km). If we split the difference, assuming that 50% of operating costs derive from line length and 50% from the number of stations, then the break-even figure is 70.7 million annual riders.

T4 currently had 67.9 million riders in 2017-18, which includes Sutherland to Waterfall, but excludes Erskineville and St. Peters (which between them have significantly more patronage than the stations south of Sutherland). Even assuming some of these passengers migrate to express services to Central, this means the line would already come close to breaking even on operating costs, and at current rates of growth would become profitable within a few years.

Given that Sydney Trains current farebox recovery rate is around 25-30%, this would be an extraordinary turnaround. Other lines could well achieve similar results, but calculating this is rather more difficult than with the T4.
 

·
Registered
Joined
·
1,114 Posts
Longer term the best candidates for Metro conversion would be:

-T4 as above
-T8 through City Circle to T2
-T9 Hornsby to Hornsby via Stratfield

Obviously these would all be isolated with Western Line and Newcastle/CC trains all terminating at Central.

What kind of capacity constraints are there with the layout of Central Terminus for it to have all Western Line, South Coast, South West, Newcastle and Central Coast trains terminating?
Yes, that's basically what I think should be done, except I would have the T8/T2 go from East Hills (with quadruplication from Revesby), through the City Circle, then to Cabramatta via Sefton (with sextuplication from Homebush to Lidcombe), and then have an express metro line going from Hornsby via the North Shore line to branches to St. Marys and Glenfield via Granville.

The problem with Sydney Terminal is that the intercity services all merge onto one pair of tracks around Redfern, which severely limits its capacity. This could be fixed by slewing the tracks at Redfern, since Platform 9-10 are virtually unused, and then fully separating Western/Northern services (via Strathfield) from Southern/South Coast services (via Sydenham), which would give you capacity for two whole lines of terminating trains.
 

·
Registered
Joined
·
1,114 Posts
Converting legacy lines to fully automated GoA4 might be a bit too hard. They would need to close the whole length Bondi Junction to Cronulla to install a completely new signalling and control system. It may be better to start converting to single deck GoA2 in the short term.
You're right, it may be a bit more difficult to go to full automation all at once. Even moving to single-deck GoA2 would present massive savings on operational costs however.

That said, T4 would probably be one of the most painless lines to shut down for an extended period of time. Between Central and Sutherland (pending quadruplication from Hurstville), you could shut down one pair of tracks, while maintaining all stations service on the other pair. Then bustitution would only be needed for Cronulla-Sutherland and Bondi Junction-City (express intercity services would be substantially slowed down, of course).
 

·
Registered
Joined
·
1,114 Posts
Most junctions are grade seperated these days. It's less about crossing paths and more about the total number of paths. There's a couple of places where you can chuck and overpass or quad a line, but the impact will be marginal in the scheme of things.

2 tracks from Schofields
4 tracks from St Mary's (should be Penrith)
2 tracks from Hornsby via Strathfield, should be 4
2 tracks from Leppington
2 tracks from Bankstown (+ Lidcombe via Bankstown)
4 tracks from Revesby
4 tracks from Hurstville (should be Sutherland)
(+ Olympic Park, Carlingford)

All merge into

2 tracks to Hornsby via Chatswood
4 tracks via the City Circle
2 tracks to Bondi Junction
2 (4?) tracks to Sydney Terminal - should be more - 6?
(soon 2 tracks to Rouse Hill)

We need balance!
There should be a Clearways 2 programme some time in the near future. I can think of the following capacity improvements:

- quadding Hurstville-Sutherland
- quadding Revesby-East Hills
- finishing the quadruplication of Strathfield-Epping, plus triplicating Thornleigh-Hornsby
- sextuplicating Homebush-Lidcombe
- duplicating Schofields-Richmond
- reworking the Redfern junction to provide two pairs of tracks to Sydney terminal
- fully grade separating the following junctions: Wolli Creek, Sutherland, Harris Park, Sefton, Cabramatta. Strathfield and Seven Hills might also need to be reworked.
- adding extra platforms and turn-backs here and there (Hornsby, Cabramatta, Emu Plains)

Then the network could fully separated into a set of stand-alone metros, complemented by "mainline" rail for long-distance suburban and intercity trains into Sydney Terminal, plus freight.
 

·
Registered
Joined
·
1,114 Posts
Remember the future transport strategy plan vision thing had a new Metro from Kogarah via Sydney Airport. This would relieve both T4 and T8.

But also, converting the underground stations on both T4 and T8 will be a chore. Most are already operating above capacity - I'd say Martin Pl is the only one that isn't. They all need more escalators and lifts. $$$

Also they all have curved underground platforms. Airport line is really bad, Martin Pl is really bad, the rest moderately curved.
How much are curved platforms a show-stopper for fully-automated metro? Gap-fillers can't be used everywhere, as they are complex and have reliability issues. But surely you could have PSDs in addition to platform gaps.

Disability access would be an issue but if the centre of the platforms are gap-free, maybe it would be acceptable to have curved gaps at the ends, as long as this is clearly sign-posted (and consistent at each station, so disabled passengers don't get stranded on a train).

I note for instance that Line 1 of the Paris metro (which was converted to GoA4 recently) has PSDs on curved platforms, e.g. at Bastille. Of course there is barely any disability access in Paris at all.

As for station capacity, this would be vastly improved if every service has a set stopping pattern, with trains coming every 2-3 minutes in peak. You can sweep the platforms clear with every train, rather than having passengers waiting around for "their" train.
 

·
Registered
Joined
·
1,114 Posts
what's the point of quadding Revesby to East Hills?
Mainly to fully separate Campbelltown express services from the local trains (which would run as metro), and speed them up, so that they would only need to stop Central-Wolli Creek (via the Illawarra express tracks, new platforms here would be needed)-East Hills-Holsworthy-Glenfield, then all to Macarthur, instead of fiddling around stopping at Revesby and Panania.

There's plenty of space in the alignment for an extra pair of tracks, so it would be very cheap, and East Hills is the natural transfer point between express and local services.
 

·
Registered
Joined
·
1,114 Posts
I'd like to see them use one of the spare platforms at St Leonards to create a new turnback if possible - particularly now that Chatswood isn't possible.
I don't really see the point, at least not in regular service. The North Shore line should just be run as a high-frequency all-stops metro service from Hornsby to Wynyard.

Passengers from north of Chatswood could change to the current metro for a slightly faster service to the city. Passengers from north of Hornsby would go via the Northern line, with a change to frequent trains at Hornsby if they want to go to Chatswood/St. Leonards/North Sydney.
 

·
Registered
Joined
·
1,114 Posts
I personally question if this proposed Kogarah metro will do enough for both lines and if it will pass the BCR, it will be a very expensive build. I think that money will be spent elsewhere, not to mention the resources.
I'm also highly dubious about the Kogarah branch off of the south-east metro. There's no real demand along that alignment, and it seems like a line on a map to try to cram more people into the south-east corridor to balance out the heavily-trafficked western corridor. Branched metro lines should also be done as sparingly as possible.

The Botany Bay area should be served, but I think Ron Christie's old idea of a metro line going from the city to Miranda via Sydney Uni, Newtown, Sydenham, the Airport and then the F6 reservation makes a lot more sense. Much of it could be done cut-and-cover to save costs if necessary.
 

·
Registered
Joined
·
1,114 Posts
Uhh, why? Just because it historically used to be the terminus of the line?

There's not much there aside from a pub and a couple of shops, and it's one of the quietest stations on the line which isn't that surprising given it's catchment area is so hemmed in.

Heck if you were building the line from scratch now you probably wouldn't even bother putting in a station there :nuts:

Revesby is much more useful as an interchange because well, there's actually a lot more there, and it's easily accessible by road from the Bankstown area to the north.

Seriously, if you want to make the trip to Campbelltown 2 minutes faster then just trim some of the fat in the timetable :eek:hno:
It's the logical transfer point because it's where the line goes from continuous suburbia to a mostly empty military reserve, before hitting Glenfield and residential areas again, and it's the point at which the stations go from being every 1.5km or so to being much more widely-spaced apart. So it's where a metro should end and a long-distance suburban line should start making all stops.

A station doesn't have to have high patronage by itself to function well as an interchange station.

Quadruplicating all the way to Glenfield is not justified, again, because East Hills is where metro service logically ends. Any further is going that bit too far for all stations, and mostly going through empty space. The expense of constructing a bridge only adds to the matter.
 

·
Registered
Joined
·
1,114 Posts
The main issue off-peak, and the one that can be most easily fixed, is that the frequency is crap on most corridors. Particularly if you need to change trains, you could be stuck waiting up to 15 minutes twice, even for a short journey.

So the easiest fix to deal with congestion is:
- upgrade all off-peak train frequencies to at least every 10min, and every 5min for inner corridors. Also, have rigidly consistent stopping patterns, and only have separate express and all-stations stopping patterns where there is the track capacity for it.
- as Khaul has suggested, supplement the rail network with a high-frequency grid of rapid bus services along main roads (again, with frequencies every 5-10min are a must, like the existing B-Line). Candidates for this are Victoria Rd, the A3 from Mac Park-Hurstville, Dee Why-Chatswood via Warringah Rd, City-Epping via Epping Rd, City-Watson's Bay via New South Head Rd. These could later be turned into metro lines if they prove popular.
- some corridors with continuous high-density development (or the potential for it) are better suited for light rail, such as Parramatta Rd, Oxford St, King St/Princes Highway, Anzac Pde to La Perouse.

Problem solved!
 

·
Registered
Joined
·
1,114 Posts
How does this crap keep on getting media.

It's an articulated electric bus with wheel covers.
The Chinese version isn't even driverless.
Is it bi-directional? That at least would make it a bit more tram-like, in that it could terminate at the end of a road and then reverse direction.

But yes, a bit of wank, really.
 

·
Registered
Joined
·
1,114 Posts
I'd also have all T1 Western services terminate at Central terminus so T9 would basically be Hornsby to Hornsby.
This looks tempting on paper, but the reality is that there is a huge imbalance in demand between the North Shore line, which has basically metro-level patronage, and the Northern line via Epping, which can barely justify 4tph outside of peak.

Also the Northern line is a mess, with a mix of local, intercity and freight trains, and track that goes from quadruple, to double, triple and back again, whereas the North Shore line is double track all the way and pretty easy to isolate as a local-stops only line.
 

·
Registered
Joined
·
1,114 Posts
I agree with Gimble here. The Northern line and North Shore line could be folded into a single sector without great difficulty.

The service on the Northern line is inadequate as it stands and this probably retards demand (you only have to look at other parts of the network that get 4 trains off peak to see this). Also overcrowding is an issue on the Northern line, especially due to the rise of high density near certain stations - notably Rhodes.

Basically the Northern line needs a turn up and go service. Now while I agree that there is a disparity in demand between the North Shore line and the Northern line, where this actually matters is in peak and during peak the trains from the Northern line will find their way onto the North Shore line, only with a delay and will be headed northbound.

The trains that are "surplus" are the ones coming from the North Shore line in peak and they arrive at Strathfield a bit after peak. The easiest solution to this problem is to construct a new turn-back west of Strathfield or simply divert trains into stabling. There is a space just west of Strathfield where this can be done cheaply and almost trivially easily.

I'm not bothered by the thought of sending 6-8 trains north of Epping and providing the upper Northern line (Beecroft etc) with a decent frequency. You can in any case turn back trains at Epping.

What Gimble suggests has a lot of merit. Particularly with regards to terminating all trains from the west. This has a bunch of knock on benefits for reliability and speed.

As for the other issues. The intercity trains disappear from the Northern line with the implementation of HSR and the Northern line already has enough extra track to cope with freight - with the possible exception of Thornleigh to Hornsby.
Between Central and Chatswood during peak you basically need as many trains on the North Shore line (in both directions) as you can physically fit. It's a waste to then have these either carrying air on the Northern line or turning back at Strathfield, while at the same time having full Western line trains terminate at Central, disgorging passengers who for the most part want to keep going to Town Hall, etc.

The simple fixit that I would be partial to is have half the North Shore trains go to the Northern line, and the other half take the slow tracks on the Western line to St. Marys, with Penrith/Richmond expresses terminating at Central.

Or even better, half to St. Marys and half to Glenfield via Granville, with Cumberland trains going from St. Marys to Glenfield in the gaps. Then you have full frequency service on all parts of the line.

Northern line trains could then run at a moderate frequency into Central. With the NW metro and the western metro coming on line, the Northern line between Strathfield and Hornsby will probably end up playing the role of a circumferential line anyway, with large numbers of passengers changing at Hornsby, Epping and North Strathfield to take more direct services into the CBD.
 

·
Registered
Joined
·
1,114 Posts
Think of the North Shore line in the AM peak. You not only need trains going into the CBD from the north, but also a full slate of trains heading northwards from Central to Chatswood, to make full use of the line for commuters to Town Hall, Wynyard, North Sydney, St. Leonards and Chatswood. Reverse the situation in the PM peak.

So where do these trains heading to Chatswood come from? There's not enough demand or track capacity from the Northern line alone for ca. 20tph. So you have a choice, either through-run Western line trains (the common-sense solution), or do as you suggest and terminate all Western line trains at Sydney Terminal, then have empty trains starting at Strathfield and running through, where the majority of Western line passengers will then transfer onto. Interchanging is not bad per se, but you're creating an unnecessary interchange point where there's no need for it.

This is why combining the Northern and North Shore lines would be counter-productive.

The best solution I can see is the following:

Local tracks: Homebush starters running all-stations to the City Circle, then out to East Hills via Airport. This is suitable for being metrofied, and perhaps also six-tracking can be extended to Lidcombe, with the vestiges of the T3 incorporated into this line.

Suburban tracks: Glenfield/St. Marys to the North Shore line, terminating at Hornsby. Cumberland trains from St. Marys to Glenfield would fill in the headway gaps on the branches. This could be turned into a single-decker "metro express" service or kept as double-decker. 10tph for each service in peak, and higher if signalling is improved and/or metro stock introduced.

Main tracks: Penrith, Richmond and Berowra via Epping expresses, as well as Blue Mountains and Newcastle interurbans, running into Sydney Terminal. This would admittedly be a complex service and stretch the capacity of the line a lot. At present headways each service could have 4tph. But tweaking to the junctions at Redfern and Strathfield could see 30tph on the trunk and 6tph on each service (there would be no issues with dwell-time inhibiting frequency as the trains would all be terminating at Central). For capacity purposes you could also experiment with platform lengthening, since this would avoid underground lines where such extensions would be excessively expensive.

Can the Northern line be served with 4tph suburban trains + 4tph interurbans, all terminating at Central (scaling up with time to 6tph 12-car trains)? At present, no, not really. But once the C&SW and Western metros open its nature as a train line will change radically, as large numbers of passengers will likely transfer at Epping and North Strathfield for more convenient metro services into the CBD, and it will change from a commuter trunk line into more of a circumferential feeder line.
 

·
Registered
Joined
·
1,114 Posts
You're also creating far too much complexity with some trains peeling off onto the mains and into Central terminus and some remaining on the suburbans. In other words preserving the problems in the present setup.
Nope, the mains would be reserved for trains terminating at Central. The suburban tracks reserved for St. Marys/Glenfield trains heading onto the North Shore. No crossovers necessary at all.

Yes, the mainline sector would have a lot of branching, but because it would all be terminating at Central this would actually be quite manageable. No reverse branching, no branching at the other end, no crossing over flat junctions, etc.

And then the other sectors would all be stand-alone lines (potentially metro-converted), with little to no branching.

It's not how I would design a system from scratch but it's the best solution I can think of for maximising capacity out of the current network.
 

·
Registered
Joined
·
1,114 Posts
Agreed. I don't buy into any solution that has limited frequencies on the minor stations on the Northern line. It needs a metro-like service. And one of the best ways to provide this is to simply couple the Northern line to the North Shore line and deal with the few trains that are "surplus" (coming off the North Shore line in peak).
The Northern line is actually a very roundabout way to get from the stations north of Strathfield to the CBD, particularly its northern sections (not to mention North Sydney). If you look at a map it runs in a very tangential direction to the city, as opposed to pretty much all the other lines in the network.

As I've said twice already, once the metros through Epping and North Strathfield come online, these will offer more convenient CBD access and will probably capture a lot of the Northern line's patronage, and it will be relegated largely to the role a circumferential feeder line. Hence why 4-6tph will probably be enough even in the long run.

That's not too mention that it will be long time before you can untangle intercity and freight from all-stations services. Total separation would require six-tracking the whole line. Some conflicts with freight are probably unavoidable, which is okay with 4-6tph, but completely unmanageable with 12tph+ (i.e. metro frequencies).
 
1 - 20 of 168 Posts
Top