The North-West Rail Link

[somebody]

One other comment here: The state libs will go to the election in March promising to deliver this as well as the NWRL, AIUI.  I personally think that the NWRL is just crazy talk, but it may well happen.

[colinw]

You think the promise is crazy talk, or the idea of the NWRL is crazy talk?

[somebody]

You think the promise is crazy talk, or the idea of the NWRL is crazy talk?

The idea of the NWRL, without significant upgrades to CBD capacity is crazy talk.  It's also likely to be far less convenient than the existing bus services for most of the affected passengers.

If the ECRL was better (for a start, bridge over Lane Cove River), then I could support an NWRL which would feed into it more easily.

[colinw]

Have you used the buses that come out of the North West?  I use them every time I stay in Sydney and frankly it is no great surprise to me that people out that way choose to drive.  About the best of the options is to catch the bus to Pennant Hills, but it makes for a very long & slow trip to the city compared to the same distance even on the somewhat tortuous Beenleigh line.

A rail line, even from Castle Hill (let alone Norwest business park or better still linking through to the Richmond line) would make a HUGE difference to the area.  But, as you say, more track capacity will be needed further in.  Either quad track down to Strathfield, or run it via the ECRL but then more tracks across the harbour will be needed.

Maybe it is time to reclaim the east side of the bridge for rail?

[somebody]

Have you used the buses that come out of the North West?  

Yes, I have.  And in particular the 620 and 626 buses which you are probably referring to if you are going to the Boundary Rd end of Franklin Rd.

I think it's a good service, which could be better.  A BUZ timetable would be nice, of course.

[colinw]

Thanks for splitting the thread.

I consider those services to be token public transport to an area that has been abandoned to the car. The 620/621 isn't too bad in peak, but the 626 is dire, particularly in the off peak when I want to use it - hourly service from 9:30AM.

For others reading, here are the timetables we're talking about :-
620: click here.
626: click here.

And as for when I stay with my wife's aunt at Galston near Rowland Village - here's the 638 timetable - click here.  My wife's aunt won't even bother with the local buses, and tends to catch a taxi through Galston Gorge to Hornsby if she wants to access the rail system.

Last time I was in Sydney I used public transport to visit a friend of mine who lives in Wollongong.  Route 638 from Galston Rd & Arcadia Rd to Pennant Hills, thence train to Central and onto an interurban to North Wollongong.  What an adventure (particularly as I was carrying a large styrofoam crate loaded with control-line model planes at the time - got 'em to Wollongong with no damage!).  It left me shaking my head in disbelief that people actually dare to complain about the Gold Coast line & surfside bus connections.  (But I do love the rail journey to Wollongong on a V-set - the Stanwell Park bit is stunning in any weather).

I think my view of bus services is somewhat skewed because I'm used to the BUZ routes & core busway service.  My work is quite close to Eight Mile Plains busway station.

Anyway - interesting discussion.  I'd be interested in hearing your views about what the priorities for Sydney should be.

cheers,
Colin

[somebody]

I would definitely agree with your comments about the 626.  20 years ago, when I think it was still route 91, last outbound journey used to leave Pennant Hills around 7pm.  It's a bit better now in the week, but they've abolished the weekend service.  The establishment of the 620/621 has been a major win, really.  It's now providing at least an hourly service on weekends and during the day.  I guess I just remember days when you virtually couldn't leave or enter Cherrybrook by PT.

For significantly less money than the NWRL, the 610X, 621 and some other routes as well could be given a BUZ service.  I would think that the journey time, waiting time, walking time and cost factors are all in favour of ramped up bus services.

EDIT: The really ridiculous part about the 626 is that it never coordinated with the train.  It was just completely unacceptable, and still is really because it co-ordinates with the Chatswood short terminators heading inbound.

[somebody]

I'd be interested in hearing your views about what the priorities for Sydney should be.

My priorities are:

• abolishing the slow timetable
• far greater focus on off peak, increasing frequency for rail and bus
• greater bus service from the NW to non-CBD locations such as Macquarie Park and North Sydney
• integrated ticketting
• Western Express line
• further rationalising of stopping patterns
• SW Rail Link
• Sorting out the track constraints on the Richmond line
• Sorting out the Carlingford Line or replacing the service with buses

[colinw]

Thanks for that.  I was talking to my sister about the NW rail link last night, and she hit the roof at the suggestion that NW could be deferred indefinitely with bus service improvement.

I'm inclined to agree with you about the short term priorities.  IMHO the main things to do are :-

1.  Speed up the horribly slow timetable
2.  Finish clearways, get the system untangled & sectorised.
3.  Focus on frequency & consistency of service patterns.
4.  Then extend - starting with SW, then NW + new cross harbour capacity.

Not sure what to make of the Carlingford line problem.  If Epping - Parra doesn't go ahead it may be better to just close the darn thing, or tramify it.  I had no idea it was down to hourly service until you posted that - the Sydney equivalent of the Doomben line.

--Col

[somebody]

Good point about the clearways.  That's a necessary plan which needs to be completed.  I guess I was just taking it as a fait accompli.

I was talking to my sister about the NW rail link last night, and she hit the roof at the suggestion that NW could be deferred indefinitely with bus service improvement.

I would wonder if she had any rational reason for thinking that though.  Or was it just an emotional response?

EDIT: I've suggested a solution for the Carlingford line, here: http://railbotforum.org/mbs/index.php?topic=4251.0
The main problem is that it would take two trains to provide better than an hourly frequency with the dwells required.  And the passing opportunities are very restricted.  Maybe a crew swap at both ends would help, but that means 3 crews rather than two.

[colinw]

My sister lives near the proposed Franklin Rd station & works at Norwest business park, & her husband works in the CBD & drives to either Pennant Hills or Epping.  Both of them would have directly benefited from the NWRL, hence the reaction.

Re Carlingford, the shonky decision in 2008 to cancel the Rydalmere crossing loop is looking very stupid now.

[somebody]

Anything that keeps the change at Clyde is shoddy.  And at least the Doomben line has a railbus filling in the gaps in its service, as well as other city bound bus services to fill in the gaps.  The Carlingford Link has little of either.

[colinw]

The Sydney Morning Herald: Rail link key in state's growth

THE north-west rail link is the most worthwhile public transport project the state government should pursue, since it has the greatest economic spinoff, research has found.

Modelling by the Centre for International Economics, on behalf of the Property Council of Australia, used the multi-criteria approach adopted by Infrastructure Australia, the federal body which has refused to fund most projects nominated by the state due to inadequate assessment and planning.

''Our recent survey of Sydney residents showed just 3 per cent felt we have a good road network and only 32 per cent gave a tick to our public transport system,'' the executive director of the Property Council, Glenn Byers, said.

''It's no surprise the north-west rail link is the highest ranked public transport project in testing its economic upside and strategic importance to Sydney's growth.''

...

[somebody]

The Sydney Morning Herald: Rail link key in state's growth

THE north-west rail link is the most worthwhile public transport project the state government should pursue, since it has the greatest economic spinoff, research has found.

Modelling by the Centre for International Economics, on behalf of the Property Council of Australia, used the multi-criteria approach adopted by Infrastructure Australia, the federal body which has refused to fund most projects nominated by the state due to inadequate assessment and planning.

''Our recent survey of Sydney residents showed just 3 per cent felt we have a good road network and only 32 per cent gave a tick to our public transport system,'' the executive director of the Property Council, Glenn Byers, said.

''It's no surprise the north-west rail link is the highest ranked public transport project in testing its economic upside and strategic importance to Sydney's growth.''

...

I think the research is wrong.  Might be arrogant of me, but I do wonder how they reached this conclusion?  I wonder if I can get a copy.

[colinw]

I really don't see how NWRL can proceed without an additional harbour crossing, or much tighter headways on the existing one.  Its the Merivale Bridge problem, but of a larger scale.

[Sunbus610]

A Sustainable Model for Rail Development in Sydney

Supporting documents

> A sustainable model for rail development in Sydney

In the media

> Proposed rail plans unveiled at Council 08 April 2011
> SMH Parramatta high-speed rail hub 07 May 2011

Click HERE (from the Parramatta City Council website) for more on this topic

[colinw]

The Sydney Morning Herald -> All aboard O'Farrell's rail funding express

UNDETERRED by the state's ugly record in teaming up with the private sector for big transport developments, the O'Farrell government is asking financiers how they would like to pay for its signature rail project.

The north-west rail link in Sydney, a 23 kilometre line to service 300,000 people, will cost an estimated $8 billion to build.

But the first step for the government will be to thrash out with banks and finance houses ways of paying for it through a public private partnership.

...

[#Metro]

Brendan Lyon, the chief executive of the lobby group for infrastructure companies, Infrastructure Partnerships Australia, praised the government for looking for the cheapest way to build the line.

Read more: http://www.smh.com.au/nsw/all-aboard-ofarrells-rail-funding-express-20110606-1fphy.html#ixzz1OXkjwrUg

"Public Private Disaster"

Beware the den of concrete worshippers... $8 BILLION!! !  
$8 billion here (CRR), $8 billion there (NW Rail link), $8 billion over there (Melbourne Metro)... don't think there will be much left over for HSR...

[SurfRail]

Brendan Lyon, the chief executive of the lobby group for infrastructure companies, Infrastructure Partnerships Australia, praised the government for looking for the cheapest way to build the line.

Read more: http://www.smh.com.au/nsw/all-aboard-ofarrells-rail-funding-express-20110606-1fphy.html#ixzz1OXkjwrUg

"Public Private Disaster"

Beware the den of concrete worshippers... $8 BILLION!! !  
$8 billion here (CRR), $8 billion there (NW Rail link), $8 billion over there (Melbourne Metro)... don't think there will be much left over for HSR...

I would have thought the cheapest way would be to get the Commonwealth to borrow the funds (they generally can do that for cheaper interest) and on-loan it to NSW for a "small" fee.  Same with us and everybody else.

[#Metro]

I would have thought the cheapest way would be to get the Commonwealth to borrow the funds (they generally can do that for cheaper interest) and on-loan it to NSW for a "small" fee.  Same with us and everybody else.

Interesting...

[colinw]

The Daily Telegraph -> Plans to extend North West rail link

HE North West Rail Link could be extended to join the Richmond line, linking it to a $2 billion industrial, office and residential precinct at Marsden Park, confidential tender documents have revealed.

The state government plans to extend the line from Epping past Rouse Hill to help cope with the population explosion in Sydney's north west.

Transport experts said the proposal would free up capacity on a number of congested lines and make RAAF Base Richmond a more likely second airport option.

The proposal is flagged in confidential Department of Transport tender documents dated June 9.

The tender states: "The service provider will be required to advise on the relevant benefits and impacts, from a transport view, of an extension of NWRL beyond Cudgegong Rd, Rouse Hill, to join the Richmond line, e.g. either at Schofields or Riverstone, or beyond (for example; to Marsden Park)."

Transport Minister Gladys Berejiklian confirmed the alignment could be extended.

"The project team is currently defining the size, scope and cost of the project," she said yesterday.

"It is prudent to ensure any expansion of the rail network will be able to accommodate possible further future expansions, and this will be reflected during the planning process."

The line would likely link up with the planned Sydney Business Park, a $2 billion, 550ha commercial, industrial, bulky goods and residential development at Marsden Park.

The precinct is expected to create 10,000 new jobs and around 1200 new homes and be built around a new Marsden Park town centre.

If the North West Rail Link, currently mapped between Epping and Rouse Hill, links up with the Richmond line it will free up capacity across congested areas of the CityRail network, according to former NSW infrastructure boss David Richmond.

"In the original planning there was always the belief that you could extend the North West Rail Line to the Richmond line, and there is certainly sense in doing so," he said.

Marsden Park locals yesterday welcomed news of an extended rail link. "It would be a good thing to do," local resident Norm Swift said.

"It would probably get a lot of cars off the roads around here, especially Windsor Rd which is just full all the time."

"free up capacity across congested areas of the CityRail network"    Not on the Harbour Bridge & lower north shore it won't!

Frankly, it boggles the mind that this project could be proceeding without doing something about additional capacity across the harbour.  Kind of like building Kippa-Ring without CRR really!

[somebody]

Frankly, it boggles the mind that this project could be proceeding without doing something about additional capacity across the harbour. 

Indeed, but there may be a couple of options, and the first of which would be abolishing Central Coast via Shore.

You also boggle at the impotence of the now departed Labour Government for not saying so more loudly.  The message didn't seem to get across, based on comments from relatives in the NW.

Kind of like building Kippa-Ring without CRR really!

That one can be achieved without destroying everything IMO.  Although it's a bit prickly.

It's not at all clear that the NWRL would not cause the sky to fall in for CityRail.

[colinw]

About the only things I can see working are very suboptimal, namely:

- Run only NWRL trains via the ECRL, and revert Northern line services to via Strathfield.  This will limit NWRL to 4TPH and not much better in peak

or

- Run NWRL trains via Strathfield as an extension of the Epping via Strathfield end of the Northern Line loop.

NWRL is simply going to have to run as a not particularly frequent extension of either the ECRL services or Epping via Strathfield, otherwise the lower north shore is going to congest or the north shore line is going to get shafted.

Like you I have relatives in the North West, and they are gung-ho about this line proceeding without considering how it will fit into the rest of CityRail.

The reality is the true cost of doing NWRL right is probably more like $20 billion, including a new harbour crossing and additional tracks under the city, or else quadding the lower north shore and reverting the former tram lanes on the bridge to rail use (but then what - terminate at the former tram platforms at Wynyard?)

All too hard no doubt, so the libs can promise to deliver the line and then leave CityRail to sort out the mess.  Can't see the North Shore line getting shafted, so my guess is a fairly poor NWRL service tacked on like I say above.

[somebody]

My understanding is that the NWRL will (likely) be built connected to the underground platforms at Epping only, so no via Strathfield running would be possible.  I don't really see this as a limitation though.

There is (as you point out) the option of moving the Upper Northern Line back to via Strathfield.

[SurfRail]

An article by David Hensher in this month's Australian Bus & Coach suggests that 28,500 buses could be bought instead of the NWRL and SWRL and that cost v benefits would still come out in front by a factor of 4.  

No suggestion of who is going to drive them, where they are going to be stowed, where they will fit in the CBD, wages bills, where they are going to come from given the new bus figures in Aus delivered last financial year was at around 1,300-1,400 (transit and long distance combined)...

The man is a bit loopy generally speaking.

[somebody]

Can't see the North Shore line getting shafted, so my guess is a fairly poor NWRL service tacked on like I say above.

Really?  Not sure what other possibilities are at all likely.  Perhaps Upper Northern Line loses 1tph in peak, down to 20 minute frequency.

I guess there is the option of removing 4tph from the Western Line through Town Hall, and increasing Sydney Terminal terminators.  Urgh.

An article by David Hensher in this month's Australian Bus & Coach suggests that 28,500 buses could be bought instead of the NWRL and SWRL and that cost v benefits would still come out in front by a factor of 4. 

No suggestion of who is going to drive them, where they are going to be stowed, where they will fit in the CBD, wages bills, where they are going to come from given the new bus figures in Aus delivered last financial year was at around 1,300-1,400 (transit and long distance combined)...

The man is a bit loopy generally speaking.

While agree that it is a loopy idea to can the SWRL, I think the exact opposite regarding the NWRL.  Can we get rid of the slow timetable first?

[colinw]

Can't see the North Shore line getting shafted, so my guess is a fairly poor NWRL service tacked on like I say above.

Really?  Not sure what other possibilities are at all likely.  Perhaps Upper Northern Line loses 1tph in peak, down to 20 minute frequency.

Because most of the upper north shore line stations fall within or on the boundary of the premier's own electorate (Ku Ring Gai). I think it would be politically unpalatable for the Premier to be seen to reduce rail service to his own electorate.

What a delicious irony. O'Farrell campaigned hard on the NWRL, but implementing one of his key promises may end up shafting the residents of his own electorate.

The Transport Minister on the other hand holds the next electorate south (Willoughby).  Her electorate potentially benefits from having direct rail service from the North West to stations Chatswood & south.

This is going to interesting to watch over the next few years.

[SurfRail]

While agree that it is a loopy idea to can the SWRL, I think the exact opposite regarding the NWRL.

Have to give them credit for scrapping the City Relief Line though.  Would have been like building Cross River Rail from one side of town to Roma Street only - and much like Melbourne Metro 1 to Domain Interchange only...

City Relief line and NWRL together with no additional capacity into the CBD from the North Shore area would have been amusing to try and fix later!

[#Metro]

An article by David Hensher in this month's Australian Bus & Coach suggests that 28,500 buses could be bought instead of the NWRL and SWRL and that cost v benefits would still come out in front by a factor of 4. 

No suggestion of who is going to drive them, where they are going to be stowed, where they will fit in the CBD, wages bills, where they are going to come from given the new bus figures in Aus delivered last financial year was at around 1,300-1,400 (transit and long distance combined)...

The man is a bit loopy generally speaking.

So does this bus idea require the construction of a Class A ROW? Because if it does then you are already looking at 100 million/kilometre and approaching the cost of rail. If you use bus lanes then you still have earthworks and that has a limited capacity around 6000 - 9000 pphd.

[colinw]

Not to mention:

- where do you find enough drivers for 28,500 buses.
- and where do you stable 28,500 buses
- where do you build bus stops to accommodate 28,500 buses when they converge on major employment areas in peak hour

Taking this to the next level, the driver problem can easily be solved by replacing the 28,500 buses with 1,425,000 cars, each of which comes with its own driver.  The charge each of those cars a nice fat toll to drive on the motorway, thereby fattening Government coffers with money to build more roads, so we can have more cars. And, then ...

Hang on, didn't we already try this?

[#Metro]

Not to mention:

- where do you find enough drivers for 28,500 buses.
- and where do you stable 28,500 buses
- where do you build bus stops to accommodate 28,500 buses when they converge on major employment areas in peak hour

Taking this to the next level, the driver problem can easily be solved by replacing the 28,500 buses with 1,425,000 cars, each of which comes with its own driver.  The charge each of those cars a nice fat toll to drive on the motorway, thereby fattening Government coffers with money to build more roads, so we can have more cars. And, then ...

Hang on, didn't we already try this?

I think Hensher is just trying to be dramatic with the numbers. However I am of the view, and it is no secret, that I dislike branching on railways and would rather see buses running to existing train stations over branches, particularly where the capacity is below 6000 pphd. I say this simply because I like FREQUENCY of service plus this way you can get high frequency on both the branches and the main line (all the way to the terminus).

[somebody]

An article by David Hensher in this month's Australian Bus & Coach suggests that 28,500 buses could be bought instead of the NWRL and SWRL and that cost v benefits would still come out in front by a factor of 4. 

No suggestion of who is going to drive them, where they are going to be stowed, where they will fit in the CBD, wages bills, where they are going to come from given the new bus figures in Aus delivered last financial year was at around 1,300-1,400 (transit and long distance combined)...

The man is a bit loopy generally speaking.

So does this bus idea require the construction of a Class A ROW? Because if it does then you are already looking at 100 million/kilometre and approaching the cost of rail. If you use bus lanes then you still have earthworks and that has a limited capacity around 6000 - 9000 pphd.

This (until recently) was near to being present.  There were a few gaps in the M2 bus lane (like Epping-Gore Hill), but a reasonable amount of bus priority was indeed present, and the service was quite good between the Hills and the CBD.  Criminally, this has recently been watered down in the PM, although it is still good in the AM.

I don't know why 28,500 buses was proposed.  Perhaps he could have said: "For 1/10 the cost we could have 2 850 buses".  And that would have made much more sense.  Not sure if he's including drivers, roads depots etc in that.  If not, sounds like something a pollie would say, and he's done himself a disservice.  But he does have a point IMO.  For significantly less investment we (Sydney) could have had a better coverage and frequencies, works in the CBD to fix the conga lines, depots and drivers.  And integrated ticketing, although theoretically Gladys (Transport Minister) has promised that.

[Golliwog]

I think Hensher is just trying to be dramatic with the numbers. However I am of the view, and it is no secret, that I dislike branching on railways and would rather see buses running to existing train stations over branches, particularly where the capacity is below 6000 pphd. I say this simply because I like FREQUENCY of service plus this way you can get high frequency on both the branches and the main line (all the way to the terminus).

I'm sorry, but branches are a fact of life. They will happen, and yes they will split services (thats not a new phenomenon). The trick isn't not to build branches but to make it so it doesn't matter.

[#Metro]

I'm sorry, but branches are a fact of life.

Er. No they are not. They are an artifact of the design.
You can design a system to look like a starfish or where the lines are separated like a subway.

[Golliwog]

Ok then, in most cases in a practical world, they are a fact of life. By your standards we should have 10 tracks through Brisbane's CBD. Eventually it may happen, but building it that way from the get go would have priced it to oblivion.

[#Metro]

Don't you think that is a bit of an extreme projection Golliwog? 10 tracks through the CBD?
Really now?

All I'm saying is that if the pphd on the branch is << 5000 pphd or so then it might be better off as a shuttle or by bus just so the frequency can be kept up.

[Golliwog]

But thats the only way you would have it so there wouldn't be any branches. One pair of tracks for each current branch.

But 5000pphd would require buses running at roughly 100bph. Should the Richlands line be a shuttle until it can get these 5000pphd? Shuttle operation would make using any such line less attractive. If you don't run a through service to begin with, you're hardly likely to get the patronage you're saying should be required to run a through service. It's a Catch 22.

[Gazza]

The trick isn't not to build branches but to make it so it doesn't matter.

This.

Branches are only bad if they result in poor frequency on the branch, or if it's some stupid little stub that cant support a decent service. It's ok to have lines merging so long as the shared track can handle the headways.

Instead of looking at as if you are splitting the frequency, it should be thought of that each branch has a decent base frequency, which gets 'added' together from the point where they converge.

Eg imagine 2 branches with 10 minute frequency, and this becomes 5 minute frequency on the shared section. Is this 'bad'. No, because 10 minute frequency is good, and the shared section is most likely gonna be in the centre of the network, where higher frequency is needed anyway.

that I dislike branching on railways and would rather see buses running to existing train stations over branches, particularly where the capacity is below 6000 pphd.

I can see where you are coming from, but there comes a point where you need to establish a new line haul route like a train because feeder buses become too arduous. A 5-10km long feeder is never going to get popular.

[#Metro]

But thats the only way you would have it so there wouldn't be any branches. One pair of tracks for each current branch.

But 5000pphd would require buses running at roughly 100bph. Should the Richlands line be a shuttle until it can get these 5000pphd? Shuttle operation would make using any such line less attractive. If you don't run a through service to begin with, you're hardly likely to get the patronage you're saying should be required to run a through service. It's a Catch 22.

One wonders how subway operations overseas work...!!!

Cultural Centre does ~ 180 bph
5000 pphd is around a train every 12 minutes... requirements are set at peak capacity... to use the richlands example it is probably best off as rail (as the timetable during peak is a train ~ 12 minutes) but off-peak this drops right off. The alternative would be to design it such that it runs as a shuttle to darra and let all trains just go to and from Ipswich. That way you get high frequency to all stations... if you just look at the branch in isolation you might forget about those other stations on the network that would benefit from higher frequency.... that's my train of thought anyway.

Branches are only bad if they result in poor frequency on the branch, or if it's some stupid little stub that cant support a decent service. It's ok to have lines merging so long as the shared track can handle the headways.

The problem is the starburst effect- trains approaching a single point (the CBD) get in the way of each other and cause conflicts/eat train paths. So as a hypothetical, you might have a doomben train carrying 80% air eating a slot for another service that might have a higher loading...

The same issue appears on the SE Busway- when you max out the slots on the main section, one wonders if making them feeders will increase capacity...

Instead of looking at as if you are splitting the frequency, it should be thought of that each branch has a decent base frequency, which gets 'added' together from the point where they converge.

Eg imagine 2 branches with 10 minute frequency, and this becomes 5 minute frequency on the shared section. Is this 'bad'. No, because 10 minute frequency is good, and the shared section is most likely gonna be in the centre of the network, where higher frequency is needed anyway.

Yes but imagine two branches one with a shuttle. You run the longest branch with 5 minute frequency and the other branch at 5 minute frequency with an interchange and you get high frequency EVERYWHERE.

[Gazza]

Well, if you're going to do that you can also achieve high frequency to everywhere by running one line express to the branching point. Perth does this.

Re the starburst effect, other places in the world sucessfuly run lines that converge reliably (Eg the Paris RER) but I guess we can't use these sort of operations in oz because 'were different' and because I used Paris as an example, and we all know how much you like Paris,.

Lines that never branch is an ideal operational scenario, but we have legacy networks in Australian cities that make this difficult, unless you were to do a new inner city tunnel to correspond with each new suburban line.
Bear in mind a modern railway can have trains running every couple of minutes right down to 90 secs, so you could have say 3,  1. Min frequency lines converging comfortably.