Rumours are not always true and in the case of making a TBM to a wider profile I have even less faith in rumours.

In 10 yrs time when the city and sourthen side is open I suspect the public opinion will towards more and not less.

As mentioned many times its not DD vs SD its current Sydney trains cost model versus what is available on market now off the shelf.

Regardless of whether the rumours are true or not, the extra cost of boring tunnels to the wider profile (if any) would be miniscule in comparison to the overall project cost.  What is the problem in making a TBM to a wider profile?  I don't get your point.  It certainly wasn't a problem with the Epping to Chatswood Rail Link.  It just demonstrates how small minded the current bunch of transport planners are, without any vision for the future.  In the past, stretching back to Bradfield's era, there was an appreciation of making provision for future contingencies.  We wouldn't otherwise have a maze of unused underground tunnels in the CBD or provision for future amplification of the suburban rail network.  

It remains to be seen whether the Rail Safety Regulator will approve the emergency evacuation procedure in the narrower tunnels, particularly in the long 6km stretch between Epping and Cherrybrook.  It's certainly not world's best practice (unlike the ECRL). Whichever way you look at it, boring the tunnels to the narrower profile, is still an act of bastardry.

Putting aside the DD v SD debate, I understand your contention that an off the shelf automatic rapid transit option is more cost effective than the current Sydney Trains cost model.  But you are comparing chalk and cheese.  What I, and I'm sure many other critics of the rapid transit proposal object to, is transposing a high density inner urban transport model to a low density outer suburban rail line, such as the North West Rail Link.  It's entirely inappropriate.  If the Rapid Transit went only as far as Epping, then it might have had more widespread support.  But out into The Hills and the North West is a different scenario. That's suburban rail territory, which demands faster express services to the CBD, which can't be provided by an all stations Rapid Transit link.  The other major objection is trying to convert parts of the existing Sydney Trains network to rapid transit and in the process compromising the overall network efficiency. I'm sure a new separate inner urban rapid transit system would be supported.

Name me any major city legacy suburban rail system that has successfully, let alone attempted, to convert it to a pseudo automatic rapid transit system. I don't think there are many, if any.

You can look no further than London's Crossrail project, which is a mirror image of Sydney's existing network as well as the Paris RER, to see how an outer suburban rail system can provide a high frequency service through the city centre.  It is not fully automatic, but it does have enhanced signalling such as European Train Control System (ETCS) and Communication Based Train Control (CBTC) which allows for higher frequency of services of 24tph with the potential for up to 32tph.  It's not pretending to be a rapid transit service, and is complementary to the London Underground.

Regardless of whether the rumours are true or not, the extra cost of boring tunnels to the wider profile (if any) would be miniscule in comparison to the overall project cost. What is the problem in making a TBM to a wider profile? I don't get your point. It certainly wasn't a problem with the Epping to Chatswood Rail Link. It just demonstrates how small minded the current bunch of transport planners are, without any vision for the future. In the past, stretching back to Bradfield's era, there was an appreciation of making provision for future contingencies. We wouldn't otherwise have a maze of unused underground tunnels in the CBD or provision for future amplification of the suburban rail network.

It remains to be seen whether the Rail Safety Regulator will approve the emergency evacuation procedure in the narrower tunnels, particularly in the long 6km stretch between Epping and Cherrybrook. It's certainly not world's best practice (unlike the ECRL). Whichever way you look at it, boring the tunnels to the narrower profile, is still an act of bastardry.

Putting aside the DD v SD debate, I understand your contention that an off the shelf automatic rapid transit option is more cost effective than the current Sydney Trains cost model. But you are comparing chalk and cheese. What I, and I'm sure many other critics of the rapid transit proposal object to, is transposing a high density inner urban transport model to a low density outer suburban rail line, such as the North West Rail Link. It's entirely inappropriate. If the Rapid Transit went only as far as Epping, then it might have had more widespread support. But out into The Hills and the North West is a different scenario. That's suburban rail territory, which demands faster express services to the CBD, which can't be provided by an all stations Rapid Transit link. The other major objection is trying to convert parts of the existing Sydney Trains network to rapid transit and in the process compromising the overall network efficiency. I'm sure a new separate inner urban rapid transit system would be supported.

Name me any major city legacy suburban rail system that has successfully, let alone attempted, to convert it to a pseudo automatic rapid transit system. I don't think there are many, if any.

You can look no further than London's Crossrail project, which is a mirror image of Sydney's existing network as well as the Paris RER, to see how an outer suburban rail system can provide a high frequency service through the city centre. It is not fully automatic, but it does have enhanced signalling such as European Train Control System (ETCS) and Communication Based Train Control (CBTC) which allows for higher frequency of services of 24tph with the potential for up to 32tph. It's not pretending to be a rapid transit service, and is complementary to the London Underground.

If the proposed tunnel width is too small for safety reasons then it needs to be corrected and this is a completely separate issue. The fact that it got this far potentially means the issues have been addressed or addressable.

The ECRL was built and costed to the specification requested at the time. Again completely separate issue.

Numerous legacy lines have been converted from HR to LR or some other RT system including some using rubber tyres. No, I don't know any that have been converted to automatic systems, but does it matter if it was or wasn't?

The SWRL and extension to the airport is likely to be the last major greenfield HR line project in Sydney using the current profile and technology. Time to move forward and find ways to reduce the longterm operating/capital costs. If the whole network could be grade separated and removed from long haul services, convert the whole thing, but its not practical, so use whats suitable. The more capital and operating cost you remove the more money there is for expansion. I believe a current or previous state transport minister said that the current cost and funding model for Sydney's rail network limited expansion to one line per decade at best.

Bradfield got us to 21st century based on largely 20th century technology, time to use 21st century vision and technology.

I caught the Paris RER a number of times, also watched trains with the boys a number of times, some in peak and I don't recall 24t/hr to a station in the Latin Qtr. Which line are you refering to?

Vancouver's train network would closely replicate that Sydney and the NWRL into the city and southern side (note it was mostly built over a former HR freight line). I've used this prior to the airport line expansion and its far from a bad system. Interestingly both Singapore and Vancouver expanded their networks (auto) with non-compatible technology and its nothing to do with the tunnel diameter. I believe there are also some conflicts with many of Paris's lines and same for London.


But anyway, so what they make the tunnel that little bit bigger, then what? The trains, stations and operating conditions will be incompatible for mixing and as the next gen network is progressively expanded, trains and stations will be designed for that standard. Like Singapore, Paris, Vancouver. The new network being automatic will have lower operating costs than the older network and be more flexible in train frequency as its able to response to changes within less than 1hr live.

So do you really believe that in the future the govt of the day would ever be temped to convert the new network back and under what conditions and why couldn't the automated technology achieve this?
ALP was going to build a completely separate automated inner NWRL following Victoria Rd to Epping? Had that actually happened, further into the future which technology do you think they would have used to continue along the outer NWRL project?

Basically no matter which govt would have built a NW something, none would have used Cityrail technology for numerous cost reasons now or in the future and no govt is likely to support further major expansion of the current technology network unless like the SWRL its not practical.

I did say its not about DD vs SD, its about the current Cityrail manually operated custom made stock versus off the shelf automatic stock. The driver and guard add hundreds $ per hour to the operating cost of each train. This includes the need to supply and maintain a two drivers cabs per 8 car set.

You did!

...a whole new route could be built from Redfern to Chatswood with the NWRL built to DD standard and achieve a similar physical route outcome, but this completely ignores the opportunity to shift to a lower operating cost model transport medium...

There is nothing stopping you from building the basic infrastructure (which excludes station fit-out) to be compatible with the minimum standards of the existing network (the decision to do that makes absolutely zip difference to operating cost) and then running "off-the-shelf" automated single deck stock.

You keep conflating "automated operation" with structure gauge.

(Again, again, again, there is nothing stopping you from configuring DD stock as a driverless automated system. You could use that as the trigger to start automating the rest of the system. As probably implied by others - if the cost of the legacy CityRail network is problematic (no argument from me!), then perhaps it would be a good idea to actually deal with that! It is going to be multiple decades before you have a "new" network that has anything like the scope of the current legacy system - because this new infrastructure is still gob-smackingly expensive in terms of capital cost. Are we seriously going to sit around for several decades ignoring the cost of the current system? Are we seriously going to sit around for several decades ignoring relatively cheap capacity expansion options in the face of very expensive alternative infrastructure solutions? That is asininely ridiculous.)

No one builds large scale infrastructure on the basis that in 30-50years time we may change our mind? At $8.3B they have probably already gone through to try and remove ever $1 possible including choice of technology. To build the tunnel that extra width, even if its just $10m could be argued that its a blatant waste of taxpayers money when they have no need for it.

Don't be absurd. Large scale infrastructure like that is always built with an appropriate provision for future flexibility, well and truly considering possibilities over three to five decades, where what's appropriate is based on the present cost of providing that flexibility and the likelihood of it being required, discounted with time.

Given the concepts floating around for where that line will go it is pretty clear that operation as a suburban railway at some stage in the foreseeable future is plausible. (Given the nature of where and what they are building, it is in fact probable that it will be operating with the characteristics of a suburban railway from day one - look at the spacing between some of the stations!)

You wouldn't blink at spending an extra $10 million to buy DD flexibility as an option. If someone told me it was going to cost me an extra $500 million on top of the current project cost (which is way over the top given unit costs for tunnelling) , I'd still jump for it.

You did!



There is nothing stopping you from building the basic infrastructure (which excludes station fit-out) to be compatible with the minimum standards of the existing network (the decision to do that makes absolutely zip difference to operating cost) and then running "off-the-shelf" automated single deck stock.

You keep conflating "automated operation" with structure gauge.

(Again, again, again, there is nothing stopping you from configuring DD stock as a driverless automated system. You could use that as the trigger to start automating the rest of the system. As probably implied by others - if the cost of the legacy CityRail network is problematic (no argument from me!), then perhaps it would be a good idea to actually deal with that! It is going to be multiple decades before you have a "new" network that has anything like the scope of the current legacy system - because this new infrastructure is still gob-smackingly expensive in terms of capital cost. Are we seriously going to sit around for several decades ignoring the cost of the current system? Are we seriously going to sit around for several decades ignoring relatively cheap capacity expansion options in the face of very expensive alternative infrastructure solutions? That is asininely ridiculous.)



Don't be absurd. Large scale infrastructure like that is always built with an appropriate provision for future flexibility, well and truly considering possibilities over three to five decades, where what's appropriate is based on the present cost of providing that flexibility and the likelihood of it being required, discounted with time.

Given the concepts floating around for where that line will go it is pretty clear that operation as a suburban railway at some stage in the foreseeable future is plausible. (Given the nature of where and what they are building, it is in fact probable that it will be operating with the characteristics of a suburban railway from day one - look at the spacing between some of the stations!)

You wouldn't blink at spending an extra $10 million to buy DD flexibility as an option. If someone told me it was going to cost me an extra $500 million on top of the current project cost (which is way over the top given unit costs for tunnelling) , I'd still jump for it.

No I focused on Cityrail manually operated technology and platform versus something Alstom, Serco etc might have in their design department.

Of course you could build everything to comply with the current profile, but then you have immediately moved away from off-the-shelf. Legacy is starts creeping in on your brand new network.

Despite the option to run Automated DD stock, who does it? There are numerous reasons why they don't.

From a safety point of view, imagine in an emergency being forced to move a large group of frail people or those in wheel chairs along a DD for evacuation. Maybe not an issue in the past, but legally it may be an issue in the future?

I agree that the legacy network costs need to be resolved, but this is an entire project on its own that will itself take a decade or more. The untangle was just the tip of the ice-berg. So the end result you want is a half arsed improved new line and eventually 2nd network still held back by 100 year old design limitations. The line is being built now, what you do now on the NWRL sets the standard for the next 100 years. Start with the latest technology, not a yet to be confirmed and implemented half baked improvement project.

Looking at the number of new Greenfield automated lines built in last 10 years. Name any that have been designed with such potential scope that they could run DD's? While I agree automated DD and SD are essentially the same, no one is doing it. Why? Dwell time, Car length, limited tight corners under cities.

The NWRL will ultimately be connected to a network that will involve more suburban style operation as well as high frequency/high volume city operation. As Sydney's has a number satellite CBD's such as Chatswood and Epping will also grow to be same. high volume stations won't be limited to the CBD. My understanding the NWRL will eventually connect to Parramatta, another CBD in its own right.

Comparing with Paris, the longest line is roughly 55min from one end to the other. it has 32 stations from one end to the other, the NWRL to city would have less and travel with an overall higher speed. So as fare as the commuter is concerned the trip is similar.

....

Look at the end of the day what I see the govt is doing simply drawing a line in the sand and saying that's it for Cityrail because its high operating and high capital cost. We are going to look around the world and do what others are doing.
What are they doing?
- Automated trains
- 3-4 door cars
- Moving away from fixed timetable to frequency timetabling
- Independent line end to end operation rather than mixing branches
- In many cases including automated services building new lines separate from existing to get the latest technology
- Moving to as much as possible off-the shelf vendor who at times becomes the operator.
- Increased capacity is achieved through frequency, not making the trains taller. its more popular and easier to manage.

There is no defined limit on where and how long a "Metro" style line should be built. By the time the NWRL is completed to the city, it won't be the longest and its not now. Many Metro lines have been extended in last 10 years, Singapore, Paris, numerous in Germany as cities grow.

The suburban style networks, like Cityrail continue to operate on legacy lines focusing on moving people from further away more efficiently and trying to improve capacity. Legacy lines have been replaced with RT systems, ie tram, LR and with ECRL automated rail. eventually Bankstown will follow and if we think this will be the only line in the world this will happen I think we are kidding ourselves. Unions here tried to talk down the benefits and may have blocked other proposals in Europe for obvious reasons. Hence part of reason Greenfields are automated.

Its all a mute point, its happening and its happening now. No more talk. The nay-sayers bitched about the GC LR and now its running they are quiet and enjoying the service. The same is happening on the Sydney SE LR and on NWRL. Once open, again will probably see the call for more of the same, not less.

Of course you could build everything to comply with the current profile, but then you have immediately moved away from off-the-shelf. Legacy is starts creeping in on your brand new network.

Absolute twaddle.  The design decision around tunnel bore is just a minimum standard for clearance.  Making that standard more permissive doesn't compromise your ability to buy anything off the shelf - it increases choice.

Or what are you buying off the shelf?  Space?  Air?  

Despite the option to run Automated DD stock, who does it? There are numerous reasons why they don't.
From a safety point of view, imagine in an emergency being forced to move a large group of frail people or those in wheel chairs along a DD for evacuation. Maybe not an issue in the past, but legally it may be an issue in the future?

Or more likely it won't be.  I don't see what difference automation makes here.

I agree that the legacy network costs need to be resolved, but this is an entire project on its own that will itself take a decade or more. The untangle was just the tip of the ice-berg. So the end result you want is a half arsed improved new line and eventually 2nd network still held back by 100 year old design limitations. The line is being built now, what you do now on the NWRL sets the standard for the next 100 years. Start with the latest technology, not a yet to be confirmed and implemented half baked improvement project.

What "design limitations"?  Increasing the tunnel bore does not stop you running your single deck automated trains with multiple doors!

But I just can't wait for these 90 second headway trains to turn up at the NWRL terminus, meeting feeder buses running that will probably run on a 10 minute headway at best...

Absolute twaddle. The design decision around tunnel bore is just a minimum standard for clearance. Making that standard more permissive doesn't compromise your ability to buy anything off the shelf - it increases choice.

Or what are you buying off the shelf? Space? Air?


Or more likely it won't be. I don't see what difference automation makes here.


What "design limitations"? Increasing the tunnel bore does not stop you running your single deck automated trains with multiple doors!

But I just can't wait for these 90 second headway trains to turn up at the NWRL terminus, meeting feeder buses running that will probably run on a 10 minute headway at best...

Unfortunately/fortunately there is alot more to design and cost than ramming a TBM down into the ground. I work on large scale Capital projects and it always amazes me how much these so called "small" changes away from something that the designs have on the shelf can cost.

On one project a building facility was at last minute changed from one vendor to another. The building had two fixed excavator arms and dust collection with a heavy rail vehicle to move the piece of kit in and out for cleaning. $100k in doing the civil changes to the drawings to change slight differences in pilling as they expected the first vendor to get the job based on previous project.

Railway projects are moving to TK (turnkey) package, like alot of other things to reduce risk, complexity in managing different vendors etc. If vendor A says to run his train he needs a tunnel of this size, then pretty much this happens. Heavy industry is no different. Having worked for companies Rio Tinto, I have zero doubt they would make that tunnel 1cm wider than what was required because the argument is we are running trains of this diameter so that's all we need. With each new project I am seeing increasing focus on only spending money on scope with the "wish lists" and "what if's" are getting less air time.


The govt is running trains of X requirements on NWRL because they don't want a Cityrail type train, manual or auto to run on it.
The govt has chosen what is basically a common standard approach for new Greenfield rail projects. Lowest Risk = Lowest Cost.
The trains will last 40 years.
The trains will be automated.
The trains will be replaced in the future with a train of similar profile
Additional stock will be purchased in future for expansion of both line capacity and route km as they are added thus overlapping the replacement.
The trains and the tunnel design eventually run through the city, satellite CBD's and other suburban areas.
The trains will be not compatible to mix with Cityrail stock. Is the O/H on the ECRL going to be retained or operate at same voltage/AC/DC for a starter?
No vendor I could find on the www offers automated double deck trains
If the cost of building and running the NWRL is lower than Cityrail per pax, then more (where suitable) will be converted to automated services in future, not less.
The platforms will have platform doors, therefore future trains will need to match this profile unless you shut the whole system down for months while a change is done.
This mitigates the ability to run stock with DD configuration unless you do a complete conversion like what will happen to ECRL. As the city grows the ability to do this becomes more difficult.

Once the decision was made to move away from the Cityrail stock profile, the option of running Cityrail stock simply no longer exists, especially if the NWRL is extended beyond the city. The larger the automated network gets, the even more remote of conversion to Cityrail stock profile.

We all know full well there is no way in hell either party will allow manually operated trains on the NWRL, not now and certainly not in the future.

Was there an outcry that the initial Sydney LR project precluded the possibility of ever running Cityrail stock on that line ever again?

While I highly doubt in our lifespans we will see 90sec at Rouse Hill, the fact is the line will be connected to the city and that sub 2.5min frequencies are reasonable to expect within 15 - 20 years especially if a line is built towards Manly and expanded into the Hurtsville, Bankstown and Inner West. What the bus frequency is in 20 years, who cares? I'm sure it will match the need.

Your last paragraph also killed your argument on the future need for DD to enable more capacity and hence the need for a larger tunnel.

If someone told me it was going to cost me an extra $500 million on top of the current project cost (which is way over the top given unit costs for tunnelling) , I'd still jump for it.

Assume DD is required in 30 years, that $500 is now worth $2.2B at 5% interest. Double that 7.5%.

I would hope for the people of NSW the NSW treasurer doesn't run the states long-term finances on such extravagances.

Basically in 30 years with the money you have saved by not making the tunnel wider using your numbers, you could build a 2nd line to the area away from the current line's foot print and dramatically increase the foot print of those served by rail and further reducing road congestion.

Assume DD is required in 30 years, that $500 is now worth $2.2B at 5% interest. Double that 7.5%.

I would hope for the people of NSW the NSW treasurer doesn't run the states long-term finances on such extravagances.

Basically in 30 years with the money you have saved by not making the tunnel wider using your numbers, you could build a 2nd line to the area away from the current line's foot print and dramatically increase the foot print of those served by rail and further reducing road congestion.

$500 million today is worth... $500 million today!  Today it buys you what $500 million buys you today!  You have future value mixed up with present value.  Your last paragraph doesn't make sense.  If you could build a second line to the area for $500 million, then the NWRL would cost about $500 million.

In 30 years time if you decide you need to put in new DD capable tunnels, then you are probably looking at needing to spend around $18.2 billion dollars in 2044 money - assuming a current (2014 $) fixed infrastructure capital cost of $7.5 billion and a capital cost inflation rate of 3% (which is probably conservative, given changes in things like land reserves, land costs, population/environmental receptor sensitivity increasing with density, etc.).

As you perhaps meant to say, your $500 million in 30 years time, at 5% is worth $2.16 billion.

So if the chance of wanting to run DD stock in 30 years or so is more than 11.9% (2.16/18.0) ... the decision to build the larger tunnels is the right thing to do.

~12% chance in 30 years?  For a line that is already been pencilled in to go much longer distances, already being planned to interconnect with the existing network in two more locations, has potential of bifurcating and heading further to the north west, is likely to run into line constraints in the inner part of the network long, long, long before it hits them on the part now under construction?  Seems like a pretty good bet to me.

TfNSW's internal costing for larger tunnels was leaked... it was only $200 million... so I'm not being particularly aggressive with my estimates.  The approach above is simplistic, but it tells the story.

It isn't extravagance, it is basic engineering project evaluation.

So if the chance of wanting to run DD stock in 30 years or so is more than 11.9% (2.16/18.0) ... the decision to build the larger tunnels is the right thing to do.

~12% chance in 30 years? For a line that is already been pencilled in to go much longer distances, already being planned to interconnect with the existing network in two more locations, has potential of bifurcating and heading further to the north west, is likely to run into line constraints in the inner part of the network long, long, long before it hits them on the part now under construction? Seems like a pretty good bet to me.

TfNSW's internal costing for larger tunnels was leaked... it was only $200 million... so I'm not being particularly aggressive with my estimates. The approach above is simplistic, but it tells the story.

It isn't extravagance, it is basic engineering project evaluation.

So let me get this right. You want to spend $200m of NSW taxpayers money which if financed at say 5% would employ another +100 odd teachers, nurses, police etc for 30 years to make a tunnel wider.

Knowing full well that the extra capacity is unlikely to be needed to in at least 30yrs or going by your previous comment maybe never? You also know its unlikely that mixing train technology including DD and SD, regardless of traction, manual driven or not on the same line is becoming more difficult as these projects move to lump sum wholistic turn key packages with one vendor and previous reports stated mixing SD and DD in Sydney is the worst of both worlds.

Basic engineering and project evaluation is looking at what the customer wants, what the real need is and finding the most cost effective solution through vendor evaluation, matching budgets and CAPEX vs OPEX etc. Extravagance or gold plated engineering is adding cost to a project for capacity that is not requested or likely to be utilised or cannot be utiliised in the design life or intended project life of the plant. I am not sure what the design life of the tunnels will be, probably 50+ years. But the train technology will be 30+yrs.

I think we can simply agree to disagree,

"Comparing with Paris, the longest line is roughly 55min from one end to the other. it has 32 stations from one end to the other, the NWRL to city would have less and travel with an overall higher speed. So as fare as the commuter is concerned the trip is similar."

Line 13 (the longest) goes from the north of Paris (St Denis) to the south (Chatillon-Montrouge). It would be very unusual for any person to travel the full length on a regular basis. That would be like someone commuting from Hornsby to Sutherland, possible but not common. My regular experience of using the Paris metro is you board, find no available seat but know that within a couple of stops one will become available as passengers are boarding and alighting in large numbers at each stop especailly the many transfer stops. The NWRL is more to be compared with the DD RER in Paris.

So let me get this right. You want to spend $200m of NSW taxpayers money which if financed at say 5% would employ another +100 odd teachers, nurses, police etc for 30 years to make a tunnel wider.

Knowing full well that the extra capacity is unlikely to be needed to in at least 30yrs or going by your previous comment maybe never? You also know its unlikely that mixing train technology including DD and SD, regardless of traction, manual driven or not on the same line is becoming more difficult as these projects move to lump sum wholistic turn key packages with one vendor and previous reports stated mixing SD and DD in Sydney is the worst of both worlds.

Basic engineering and project evaluation is looking at what the customer wants, what the real need is and finding the most cost effective solution through vendor evaluation, matching budgets and CAPEX vs OPEX etc. Extravagance or gold plated engineering is adding cost to a project for capacity that is not requested or likely to be utilised or cannot be utiliised in the design life or intended project life of the plant. I am not sure what the design life of the tunnels will be, probably 50+ years. But the train technology will be 30+yrs.

I think it prudent to spend that amount to retain the option in future. This is because the likelihood of needing that option is high enough, the cost of providing that option in future (if the additional spend now isn't made) is high enough and the time frame for the option being required is short enough, that, on balance, spending the money now saves you in the long run.

If you spend money now for a future benefit (which is pretty what happens for any capital project), then that spend now comes with an opportunity cost (the "teachers, nurses, police" you allude to). But if that opportunity cost is outweighed by the benefits, appropriately discounted for time and risk, then it is still worth incurring that cost!

I don't "know full well that the "extra capacity" (!) is unlikely to be needed"... you are asserting there almost the opposite to what I think. What I think is, based on circumstances of the line and its future plans, it is plausible (i.e the chance expressed as a probability is material - not 100%, but more than 10%) that within the next thirty years we may want to run higher capacity sets down that segment of the line, perhaps associated with a limit on service frequency due to constraints elsewhere or due to a desire to use that segment of the line for sets coming from future extensions and future network interconnections.

(There is relevant precedence for this. In 1964 an existing railway began to introduce double deck cars on services that had been previously single deck only, on multiple lines that in many cases the underground infrastructure for which had only been completed between eight and thirty two years previous.)

The proponents for this line also think that we may want to run higher capacity sets in future - platform length accommodates ten cars while only seven car sets are being ordered. Station voids and stations in general are expensive and their cost increases with size - I wouldn't be at all surprised to find that accommodating ten car sets over seven has a similar cost to the wider tunnels! They don't need this capacity when the line opens. Is this gold plating? Think of the nurses and teachers and policemen we are missing out on, just because of the possibility that we might need higher capacity per set in future! How extravagant!!

Quantitative assessment of the costs and benefits of options, accounting for likelihoods and opportunity costs, is "looking at what the customer wants, what the real need is and finding the most cost effective solution... and CAPEX vs OPEX" (vendor evaluation and matching budgets are aspects of project execution and estimation, not assessment). It is not gold plating. It is not extravagant. In many cases people's desire to retain flexibility in a long life asset isn't justified when you look at a quantitative assessment, but in this case, supported by the little simplistic assessment in the previous post, I think a $200 million spend is comfortably carried. In risked present cost terms it delivers a lower cost option to the customer.

I think running fully automated DD trains in 30 years on the same infrastructure as single (if still required) would be a doddle. Issues with mixed service are more likely to be due to aspects such as platform configuration. But that's only an issue if you have your mixed services stopping at the same platforms. Given there's a contiguous six kilometre section of the new line without any platforms at all (biggest distance between stations in the suburban area? Not so very "metro" is it... a line with characteristics selected for frequent starting and stopping of train sets, that doesn't have much need for trains sets to frequently start and stop!), I think that can be put aside.

(I think if you wanted to run fully automated double decker trains today on a new system it would be a doddle. The subsystems that support automation are mature off-the-shelf components that are completely independent from the structure of the unit and relatively easy to integrate into any platform... the reason you don't see many fully automated double deck systems has everything to do with the cost/benefit of automation on lines where double deck services are typically used and nothing to do with the vehicle profile.)

The design life of the trains isn't relevant to this. I am talking about the fixed infrastructure, which probably has a design life of 80 years or more.

I think it prudent to spend that amount to retain the option in future. This is because the likelihood of needing that option is high enough, the cost of providing that option in future (if the additional spend now isn't made) is high enough and the time frame for the option being required is short enough, that, on balance, spending the money now saves you in the long run.

If you spend money now for a future benefit (which is pretty what happens for any capital project), then that spend now comes with an opportunity cost (the "teachers, nurses, police" you allude to). But if that opportunity cost is outweighed by the benefits, appropriately discounted for time and risk, then it is still worth incurring that cost!

I don't "know full well that the "extra capacity" (!) is unlikely to be needed"... you are asserting there almost the opposite to what I think. What I think is, based on circumstances of the line and its future plans, it is plausible (i.e the chance expressed as a probability is material - not 100%, but more than 10%) that within the next thirty years we may want to run higher capacity sets down that segment of the line, perhaps associated with a limit on service frequency due to constraints elsewhere or due to a desire to use that segment of the line for sets coming from future extensions and future network interconnections.

(There is relevant precedence for this. In 1964 an existing railway began to introduce double deck cars on services that had been previously single deck only, on multiple lines that in many cases the underground infrastructure for which had only been completed between eight and thirty two years previous.)

The proponents for this line also think that we may want to run higher capacity sets in future - platform length accommodates ten cars while only seven car sets are being ordered. Station voids and stations in general are expensive and their cost increases with size - I wouldn't be at all surprised to find that accommodating ten car sets over seven has a similar cost to the wider tunnels! They don't need this capacity when the line opens. Is this gold plating? Think of the nurses and teachers and policemen we are missing out on, just because of the possibility that we might need higher capacity per set in future! How extravagant!!

Quantitative assessment of the costs and benefits of options, accounting for likelihoods and opportunity costs, is "looking at what the customer wants, what the real need is and finding the most cost effective solution... and CAPEX vs OPEX" (vendor evaluation and matching budgets are aspects of project execution and estimation, not assessment). It is not gold plating. It is not extravagant. In many cases people's desire to retain flexibility in a long life asset isn't justified when you look at a quantitative assessment, but in this case, supported by the little simplistic assessment in the previous post, I think a $200 million spend is comfortably carried. In risked present cost terms it delivers a lower cost option to the customer.

I think running fully automated DD trains in 30 years on the same infrastructure as single (if still required) would be a doddle. Issues with mixed service are more likely to be due to aspects such as platform configuration. But that's only an issue if you have your mixed services stopping at the same platforms. Given there's a contiguous six kilometre section of the new line without any platforms at all (biggest distance between stations in the suburban area? Not so very "metro" is it... a line with characteristics selected for frequent starting and stopping of train sets, that doesn't have much need for trains sets to frequently start and stop!), I think that can be put aside.

(I think if you wanted to run fully automated double decker trains today on a new system it would be a doddle. The subsystems that support automation are mature off-the-shelf components that are completely independent from the structure of the unit and relatively easy to integrate into any platform... the reason you don't see many fully automated double deck systems has everything to do with the cost/benefit of automation on lines where double deck services are typically used and nothing to do with the vehicle profile.)

The design life of the trains isn't relevant to this. I am talking about the fixed infrastructure, which probably has a design life of 80 years or more.

Not sure why there is a 6km gap after Epping, but there are "gaps" in most lines for one reason or another, but looking at a map it would appear this area is partially covered to the nth by the Nth Main which curves out west around Beecroft. So at a guess its avoiding costly duplication for a project that is already one of the more expensive things ever built in one go for NSW. As for other stations spacings, once you remove that 6km, the station spacing are not far off Singapore/Vancouver and more in line with modern approach of fewer stations moving more people to reduce CAPEX and OPEX and speed up trains.

Singapore, higher population density than Sydney, a far higher usage of PT than Sydney, survives on SD trains, with high % of door space per car for faster unloading/loading.
- The longest (oldest) line is 52km, average speed along the route is 62km/hr max 80 with a station average 1.5km apart. It is to be extended further 4 stations, 11km.
- The N-S line average station spacing 1.7km apart, average speed of 57km/hr over its 57km route length
- NE line (auto), 20km long, 16 stations 1.25km apart, average speed 36km/hr

Sydney NWRL, low to medium density housing and ECRL medium density light industry, university, shopping etc Greenfield section, 2.9km apart, roughly same if you include the ECRL section as well. Average speed at a quick guess over 65km/hr


The line is being built technology separate from Cityrail to make the platform leap from manual to auto operation. This saves money longterm. At the same time it enables sub 90sec frequency which covers the capacity issues, should it be needed. Its unlikely the control system will be 90sec from day one and it won't have the trains anyway, but it will be up gradable as required to faster frequency.

The longer platform than required from day one is a common strategy employed by some projects including the scope for the Brisbane CRR/UBAT or what ever is the most recent name. Its relatively cheap to add to the capital and can be utilized fairly easily in the future as you simply install the required platform finishes and buy the additional cars for the trains.

Going from SD to DD with platform doors is potentially far more problematic, especially if the SD trains are used with more than two doors and typical shorter body length that I have noticed on many RT systems. Mixing the two would be very complex and hence unlikely to be done. And unless you are at 90sec frequency on SD, why would need DD?

Sydney's previous move from SD to DD over 30 years was basically based on the same car body length and door design. Hence the mixing on +3min frequency was not an issue. You cannot mix a Sydney stock with a typical RT type carriage that has 3-4 doors over often shorter car bodies and expect to move the same bodies on a 3min frequency as Sydney DD stock.

No argument that the NWRL is getting on the long side for what other cities use for RT type stock with lower capacity seating. But there is no black and white of what technology runs where and others are moving towards NWRL design, not away from it. Including Paris, Singapore, Dubai and Vancouver to name a few (all of which I have used). ie they are approaching 1hr travel time from terminus to city. As stated before, the state govts own trip survey shows that about 30-50% (hard to tell as its bubbles on a map) of target users will not go past Chatswood.

Any concern for number of people standing for long is a moot point when that happens now. Hell it happened when I last commuted in Sydney 20 years ago. Standing for 50min was standard and I know from recent casual trips it hasn't improved. At least on the suburban services if you traveled from say Hornsby you got a seat in the morning, but in the after noon you may have to stand, however as the train unloaded around half way you could claim as seat for the remainder of the trip. If you are a mid line commuter then likely you stood in AM and in PM you may or may not get a seat, but for 30min or so, standing not so bad.

The main down side in this project is that the connection to the city is in phase 2, which will be at least 5 years behind the NWRL if they wait for the NWRL to open before announcing the project is approved.

Not sure why there is a 6km gap after Epping, but there are "gaps" in most lines for one reason or another, but looking at a map it would appear this area is partially covered to the nth by the Nth Main which curves out west around Beecroft. So at a guess its avoiding costly duplication for a project that is already one of the more expensive things ever built in one go for NSW. As for other stations spacings, once you remove that 6km, the station spacing are not far off Singapore/Vancouver and more in line with modern approach of fewer stations moving more people to reduce CAPEX and OPEX and speed up trains.

Singapore, higher population density than Sydney, a far higher usage of PT than Sydney, survives on SD trains, with high % of door space per car for faster unloading/loading.
- The longest (oldest) line is 52km, average speed along the route is 62km/hr max 80 with a station average 1.5km apart. It is to be extended further 4 stations, 11km.
- The N-S line average station spacing 1.7km apart, average speed of 57km/hr over its 57km route length
- NE line (auto), 20km long, 16 stations 1.25km apart, average speed 36km/hr

Sydney NWRL, low to medium density housing and ECRL medium density light industry, university, shopping etc Greenfield section, 2.9km apart, roughly same if you include the ECRL section as well. Average speed at a quick guess over 65km/hr


The line is being built technology separate from Cityrail to make the platform leap from manual to auto operation. This saves money longterm. At the same time it enables sub 90sec frequency which covers the capacity issues, should it be needed. Its unlikely the control system will be 90sec from day one and it won't have the trains anyway, but it will be up gradable as required to faster frequency.

The longer platform than required from day one is a common strategy employed by some projects including the scope for the Brisbane CRR/UBAT or what ever is the most recent name. Its relatively cheap to add to the capital and can be utilized fairly easily in the future as you simply install the required platform finishes and buy the additional cars for the trains.

Going from SD to DD with platform doors is potentially far more problematic, especially if the SD trains are used with more than two doors and typical shorter body length that I have noticed on many RT systems. Mixing the two would be very complex and hence unlikely to be done. And unless you are at 90sec frequency on SD, why would need DD?

Sydney's previous move from SD to DD over 30 years was basically based on the same car body length and door design. Hence the mixing on +3min frequency was not an issue. You cannot mix a Sydney stock with a typical RT type carriage that has 3-4 doors over often shorter car bodies and expect to move the same bodies on a 3min frequency as Sydney DD stock.

No argument that the NWRL is getting on the long side for what other cities use for RT type stock with lower capacity seating. But there is no black and white of what technology runs where and others are moving towards NWRL design, not away from it. Including Paris, Singapore, Dubai and Vancouver to name a few (all of which I have used). ie they are approaching 1hr travel time from terminus to city. As stated before, the state govts own trip survey shows that about 30-50% (hard to tell as its bubbles on a map) of target users will not go past Chatswood.

Any concern for number of people standing for long is a moot point when that happens now. Hell it happened when I last commuted in Sydney 20 years ago. Standing for 50min was standard and I know from recent casual trips it hasn't improved. At least on the suburban services if you traveled from say Hornsby you got a seat in the morning, but in the after noon you may have to stand, however as the train unloaded around half way you could claim as seat for the remainder of the trip. If you are a mid line commuter then likely you stood in AM and in PM you may or may not get a seat, but for 30min or so, standing not so bad.

The main down side in this project is that the connection to the city is in phase 2, which will be at least 5 years behind the NWRL if they wait for the NWRL to open before announcing the project is approved.

While I acknowledge your expertise in infrastructure planning, I don't think you have a grasp of the detail in respect of the North West Rail Link.  As has been mentioned over and over again, it is not what would be typically considered to be appropriate for a high density rapid transit system.  It is a "suburban" low density catchment.  For example, the 6km section between Epping and Cherrybrook without any intermediate stations was a result of the decision to cancel the original proposal to connect the NWRL with the Northern Line between Cheltenham and Beecroft.  This would have enabled the flexibility of routing services to the CBD via either the Northern Line via Strathfield or the ECRL.  The adopted proposal restricts the operation of the NWRL with a steeper alignment under Devlins Creek.  The preferable location of an intermediate station at West Pennant Hills (Thompsons Corner) is not feasible because it would be too deep.  If the original plan had been implemented, then this station would have been viable because of a more gradual alignment.

The difference with other rapid transit systems such as Singapore (and other systems in Europe) where the total length of lines extends either side of the city centre, is that the NWRL extends 40 odd kms from the city centre in one direction.  This is not typical for a rapid transit system.

You should come down from your ivory tower and accept that a legacy suburban rail network can't be converted to a rapid transit system without compromising the existing network.  The cost would be prohibitive.  We have to work with what we've got and make it operate as efficiently as possible.  It doesn't preclude having a separate rapid transit system servicing the inner city suburbs.

While I acknowledge your expertise in infrastructure planning, I don't think you have a grasp of the detail in respect of the North West Rail Link. As has been mentioned over and over again, it is not what would be typically considered to be appropriate for a high density rapid transit system. It is a "suburban" low density catchment. For example, the 6km section between Epping and Cherrybrook without any intermediate stations was a result of the decision to cancel the original proposal to connect the NWRL with the Northern Line between Cheltenham and Beecroft. This would have enabled the flexibility of routing services to the CBD via either the Northern Line via Strathfield or the ECRL. The adopted proposal restricts the operation of the NWRL with a steeper alignment under Devlins Creek. The preferable location of an intermediate station at West Pennant Hills (Thompsons Corner) is not feasible because it would be too deep. If the original plan had been implemented, then this station would have been viable because of a more gradual alignment.

The difference with other rapid transit systems such as Singapore (and other systems in Europe) where the total length of lines extends either side of the city centre, is that the NWRL extends 40 odd kms from the city centre in one direction. This is not typical for a rapid transit system.

You should come down from your ivory tower and accept that a legacy suburban rail network can't be converted to a rapid transit system without compromising the existing network. The cost would be prohibitive. We have to work with what we've got and make it operate as efficiently as possible. It doesn't preclude having a separate rapid transit system servicing the inner city suburbs.

No ivory tower here, just while I wouldn't personally go down the path they are going. I don't see a reason why the RT technology isn't the path should follow to reduce the longterm CAPEX and OPEX.

Is there any definition for what RT should be used for?
Is there any defined length of RT?
Is there any defined urban density?
Is there any defined time travel?

No usually based on cost and its just what we are used to seeing in some places, usually old cities with existing infrastructure. Newer cities and/or systems don't build 300-400km of PT rail infrastructure over night, it usually takes decades. However in places that they are doing this, they are going RT technology. As automated RT comes down in price i suspect we will see more.

Singapore's E-W line from thewestern branch to city will approach that of the NWRL and certainly take longer due to more stops and slower speed trains.

Dubai's Red line will exceed the NWRL from Downtown to the new airport, Expo 2020 site and urban/industrial area

Abu Dhabi will probably exceed the NWRL length when complete as the city centre is at the sea end of a long island with mainland development away on opposite end.

The reason's traditional suburban has been built over RT is
- Mix with freight services/long haul pax
- Cost of automation vs cost of manual driven and signally system including platform doors
- More exposed to public domain ie LX

The option of flexibility is pretty much going in Sydney during peak hour. That's what the untangle project was all about. Yes off-peak you have choices, but not in peak.

You state a "legacy suburban rail network can't be converted to a rapid transit system without compromising the existing network". Rubbish. Melbourne managed ok, twice! Others probably have too.


I'm sure if they wanted to go via West Pennant Hills that was still possible. The choice of technology in this case isn't forcing them to not follow the Nth Main.

As I said, this is not what I personally would have done as far as where to lay tracks, but I understand the reasoning by the govt and if anyone thinks that either side if NSW politics was ever going to build the NWRL or any other line where practical (SWRL is not) on anything other than automated technology then they are not paying attention. Its been done for a reason and the reasons are clear. Cheaper short and longterm. Time for that technological "platform" change.

No ivory tower here, just while I wouldn't personally go down the path they are going. I don't see a reason why the RT technology isn't the path should follow to reduce the longterm CAPEX and OPEX.

Is there any definition for what RT should be used for?
Is there any defined length of RT?
Is there any defined urban density?
Is there any defined time travel?

No usually based on cost and its just what we are used to seeing in some places, usually old cities with existing infrastructure. Newer cities and/or systems don't build 300-400km of PT rail infrastructure over night, it usually takes decades. However in places that they are doing this, they are going RT technology. As automated RT comes down in price i suspect we will see more.

Singapore's E-W line from thewestern branch to city will approach that of the NWRL and certainly take longer due to more stops and slower speed trains.

Dubai's Red line will exceed the NWRL from Downtown to the new airport, Expo 2020 site and urban/industrial area

Abu Dhabi will probably exceed the NWRL length when complete as the city centre is at the sea end of a long island with mainland development away on opposite end.

The reason's traditional suburban has been built over RT is
- Mix with freight services/long haul pax
- Cost of automation vs cost of manual driven and signally system including platform doors
- More exposed to public domain ie LX

The option of flexibility is pretty much going in Sydney during peak hour. That's what the untangle project was all about. Yes off-peak you have choices, but not in peak.

You state a "legacy suburban rail network can't be converted to a rapid transit system without compromising the existing network". Rubbish. Melbourne managed ok, twice! Others probably have too.


I'm sure if they wanted to go via West Pennant Hills that was still possible. The choice of technology in this case isn't forcing them to not follow the Nth Main.

As I said, this is not what I personally would have done as far as where to lay tracks, but I understand the reasoning by the govt and if anyone thinks that either side if NSW politics was ever going to build the NWRL or any other line where practical (SWRL is not) on anything other than automated technology then they are not paying attention. Its been done for a reason and the reasons are clear. Cheaper short and longterm. Time for that technological "platform" change.

I don't give a stuff what they're doing in Singapore, Dubai or any other Asian city.  It's totally irrelevant.  You can't make comparisons with a city like Sydney which has an extensive, long established Suburban/CBD rail network.  There has been over a century and a half of investment in rail infrastructure in Sydney and you can't just ignore it by switching to so called "new generation" train technology.  It may well suit new "greenfield" urban environments, but where there is already established rail infrastructure on the scale that is current in Sydney, you don't throw the baby out with the bathwater.  This is a valuable resource and rather than compromising it by confiscating part of the network to accommodate "new technology", the emphasis should be on incremental upgrading to increase its overall efficiency. It's not exactly rocket science, as has already been demonstrated in London, Paris, Berlin and Zurich as well as I'm sure other European cities.  You don't seem to acknowledge that a modernised legacy rail network has some place in the overall scheme of things.  Your suggestion that an eventual fully automated rapid transit system in Sydney will become the dominant rail network is pure fantasy.

With regard to your comments about Melbourne, it just shows how out of touch you are. It's not a rapid transit system, despite its designation as a "Metro".  It's still a legacy suburban rail network, rebranded by its franchise operator.  


From my reading of your posts, you don't seem to be fully cognizant of the current operational constraints on the Sydney Trains network and community attitudes in particular towards the NWRL Rapid Transit proposal and its future extension across the harbour.  Well let me tell you I live in the area and there is no support whatsoever.  The original NWRL proposal taken to the last election by the then Opposition Coalition provided for the direct link between Epping and Cherrybrook integrated with the existing ECRL and CityRail DD network.  When this morphed into the privatised rapid transit proposal with smaller diameter tunnels, it also suddenly adopted a steeper alignment between Epping and Cherrybrook, unsuitable for DD trains.  Coincidence?   Regardless of the merits of the automated single deck rapid transit, dubious in my opinion as they may be, if the public doesn't like it, then in a democratic society they have every right to vent their spleen on the government at the next election.  Engineers in their ivory towers don't always get it right.

I don't give a stuff what they're doing in Singapore, Dubai or any other Asian city. It's totally irrelevant. You can't make comparisons with a city like Sydney which has an extensive, long established Suburban/CBD rail network. There has been over a century and a half of investment in rail infrastructure in Sydney and you can't just ignore it by switching to so called "new generation" train technology. It may well suit new "greenfield" urban environments, but where there is already established rail infrastructure on the scale that is current in Sydney, you don't throw the baby out with the bathwater. This is a valuable resource and rather than compromising it by confiscating part of the network to accommodate "new technology", the emphasis should be on incremental upgrading to increase its overall efficiency. It's not exactly rocket science, as has already been demonstrated in London, Paris, Berlin and Zurich as well as I'm sure other European cities. You don't seem to acknowledge that a modernised legacy rail network has some place in the overall scheme of things. Your suggestion that an eventual fully automated rapid transit system in Sydney will become the dominant rail network is pure fantasy.

With regard to your comments about Melbourne, it just shows how out of touch you are. It's not a rapid transit system, despite its designation as a "Metro". It's still a legacy suburban rail network, rebranded by its franchise operator.


From my reading of your posts, you don't seem to be fully cognizant of the current operational constraints on the Sydney Trains network and community attitudes in particular towards the NWRL Rapid Transit proposal and its future extension across the harbour. Well let me tell you I live in the area and there is no support whatsoever. The original NWRL proposal taken to the last election by the then Opposition Coalition provided for the direct link between Epping and Cherrybrook integrated with the existing ECRL and CityRail DD network. When this morphed into the privatised rapid transit proposal with smaller diameter tunnels, it also suddenly adopted a steeper alignment between Epping and Cherrybrook, unsuitable for DD trains. Coincidence? Regardless of the merits of the automated single deck rapid transit, dubious in my opinion as they may be, if the public doesn't like it, then in a democratic society they have every right to vent their spleen on the government at the next election. Engineers in their ivory towers don't always get it right.

Dubai is also not in Asia.

I actually do give a stuff what others do because it opens your eyes to the options around you and hardly irrelevant. But you obviously care about some places that suit your needs.

So lets start with Paris
- Some lines are tyre technology (5 lines), others steel, so its not a common network
- Lines 14 and now recently Line 1 were converted to Automation and as such their old rollingstock removed and are NOT compatible with other lines any more. Expect more lines to follow
- There are other more subtle differences that prevents fast response of rains running one one line using another. Some is simply train length. They do not mix train lengths on the same line.
- The lines are all operates nominally as separate lines with rolling stock nominally captured by that line even where technology differences do not occur. .
- Interesting comment I found, the Paris Metro was deliberately built with narrow tunnels to prevent mainline trains running on them.
- It is planned some Paris Metro lines will in future have distances that are approaching with NWRL
- Transit times for the longer Paris metro lines to city is similar to that of NWRL
- Automation was introduced as a way to deal with congestion, why because you can run closer spacing!
- "The network reached saturation after World War II, with new trains to allow higher traffic, but further improvements have been limited by the design of the network and in particular the short distances between stations"
"The low speed virtually precludes feasible service to farther suburbs, which are serviced by the RER." So the so called Metro standard that the NWRL does not comply with (according to some) isn't such a great bench mark for what defines a Metro after all?

With reference to other networks
"In contrast to many other historical metro systems (such as New York, Madrid, London, and Boston), all lines (in Paris) have tunnels and operate trains with the same dimensions"

Moscow
- "Long distances between stations have the positive effect of a high cruising speed of 41.7 kilometres per hour (25.9 mph)." Note Paris Metro is down to below 25km/hr.
- Moscow Metro operates as separate systems because of differences in train length limits. Yes its the same stock.


London/ manyGerman cities I have been to
- At least two different technologies


New York
- Different fleets due to curves and widths to suit different routes


- Melbourne, The Heavy Rail St Kilda and Port Melbourne Lines were replaced with Lightrail (RT). It handed 20 years ago, nothing to do with Branding! Out of touch, hardly I caught a tram from Port Melbourne a few weeks after conversion. and St Kilda a few weeks later.
- Adelaide, The current tram line is actually a former heavy rail line
- Sydney, The current LR is a former heavy rail line


Ok, lets go back to the start.
What is the No.1 target of the NSW govt and previous in providing a line to some where in NW (former inner NW project or current NWRL project). AUTOMATION. Why? Because Legacy systems have legacy's including higher operating costs.


Ok, so you got that point, that want to make it automated because its cheaper, more efficient and basically more flexible.


for the govt, what is one of the main issues for the past and present state govts in running Cityrail. COST COST and more COST. its costly to build and costly to operate. Its very high subsidy limits on how much of a network we can have. Its technology is costly to build and we keep having rolling-stock procurement issues that other systems generally don't have. Its also not an issue limited to Sydney, read Qld website to find out about very high fares to try and limit the subsidy to 75%. NSW is around 65%. Melbourne, Adelaide and Perth are not much better if at all. DOO, privatisation, increased ridership etc isn't cutting the subsidy costs to the level they expected.


So next, can you mix automated with non automated? Probably, but its not done, too complex, too much risk, you destroy the benefits of automation for the costs of manual. So the line is going to be seperate. Once you have made that separation, you can do what you like after as there is far less back compatibility that is required. So steeper hills, etc.


You claim there are efficiency gains to be had in Cityrail. So why are they so hard to gain. How much are we talking 1%, 5%? What is that offered by NWRL Automation, 10%, 20%....????


Yes I look at some of these European cities and others including Singapore and Vancouver and I see the same theme, Automation of existing (where practical) and often new greenfield projects! Some cities with existing automated ones like Vancouver changing from one technology to the next for the new projects because they get a better deal and because most Metro/suburban networks are moving away from spaghetti operation to line based. The move to off-the-shelf holistic lumpsum suppliers. Lower risk with less interfaces = lower costs.


I have not proposed conversion of all Sydney to Automation, the obvious ones are NWRL to new route via City, conversion of Bankstown/Huntsville and inner west to Olympic park. I would also propose the Nth beaches would follow the same path if a line was built and the Epping to Chatswood line via Carlingford. This is about 25 years worth of projects. The traffic from these suburban branches will also most likely saturate the city section so consider it full. These routes will also take pressure off the current system in the right areas to allow more services from outer suburban and interurban areas to access the city.


"The original NWRL proposal taken to the last election by the then Opposition Coalition provided for the direct link between Epping and Cherrybrook integrated with the existing ECRL and CityRail DD network. When this morphed into the privatised rapid transit proposal with smaller diameter tunnels, it also suddenly adopted a steeper alignment between Epping and Cherrybrook, unsuitable for DD trains. Coincidence? "



How is it a coincidence that tunnels are now too steep or the tunnels too narrow following decision to go off the shelf? As I stated above and many many times before. Once the decision was made to go Auto, there is zero reason, ZERO to back the line backwards compatible in as many ways as you want. The 1500VDC O/H is being kept, probably because its cheaper, but its hardly a deal breaker or major cost. If you cannot run a train up there because its too steep (there was/are issues with this on ECRL), then tunnel width is irrelevant. If a different technology means tunnels can be steeper and hence shorter, then we are saving money?


"Regardless of the merits of the automated single deck rapid transit, dubious in my opinion as they may be, if the public doesn't like it, then in a democratic society they have every right to vent their spleen on the government at the next election. Engineers in their ivory towers don't always get it right."

What merits of automation do you find dubious? The world has been automating things since the start of the industrial revolution and not stopped. If you get behind a automated network, the benefits scream out. Live response to passenger loading, faster boarder as people realise there an't no one watching the door when it closes.


Its highly unlikely the govt will loose because of a railway line. The people asked for a railway, after years of inaction by previous govt they got one and they are getting one faster than most other lines that have been built in recent years in anywhere of Australia from conception ton completion. They will also get one of the most modern services available at the time. The studies show that the people in this area who use PT or likely to use PT travel to Epping where they can catch a suburban or alight from the line at or prior to Chatswood. 1/3 would travel onwards towards the city. Its their traffic movements, not the govts. The Govt has also continued to focus on the city extension so the Chatswood terminus is as short as possible, but still likely to be 5 years. What the NWRL users will get is the ability to watch out the front and back of the trains and this is EXTREMELY popular in places its available (from what I have seen and where I personally stand with my kids each time I ride a Auto train).  


Most engineer's don't work in ivory towers and they deal with real issues including unrealistic budgets and expectations by clients and usually travel visit to benchmarks and other options. This includes the so called divine right to have a single train passage from home to work that many Sydney commuters have become very used to.

- It is planned some Paris Metro lines will in future have distances that are approaching with NWRL
- Transit times for the longer Paris metro lines to city is similar to that of NWRL

I have pointed out on another thread. The 3 longest metro lines (7, 8 & 13) are between 20 and 25 km long but travel from one side of Paris through the city to the other. While possible, it is unlikely many people would travel the full length on a regular basis. It would be like commuting from Hornsby to Sutherland. The longer commutes are with RER and double deck trains. In my regular visits to Paris,  I find I rarely stand for more than a few very close stations as people are getting on and off all the way. very different to most Sydney travel.

Yes, with plans to extend. Including the 100km ring line.

Also its not distance, its time thats more important.  So with nearly 40 stations for nearly 20km, this results in an average speed of barely 20km/hr.  The close spacings of the statations is ond of the nextworks biggest hang ups.

The NWRL will eventually connect three branchs, Richmond, Nth Main, NSL. Travel time between each will be less than 20min. At Epping, there will be a significant volume exchange and expect similar at Chatswood. My experience in peak in Sydney that those joining a peak train from around Chatswood and Epping will most likely be standing. I used to stand from Woy Woy to Hornsby, time wise this is similar to total distance on NWRL. And its very unkikely to not get a seat for full distance. So the seat issue is a falicy.

Yes, with plans to extend. Including the 100km ring line.

Also its not distance, its time thats more important.  So with nearly 40 stations for nearly 20km, this results in an average speed of barely 20km/hr.  The close spacings of the statations is ond of the nextworks biggest hang ups.

The NWRL will eventually connect three branchs, Richmond, Nth Main, NSL. Travel time between each will be less than 20min. At Epping, there will be a significant volume exchange and expect similar at Chatswood. My experience in peak in Sydney that those joining a peak train from around Chatswood and Epping will most likely be standing. I used to stand from Woy Woy to Hornsby, time wise this is similar to total distance on NWRL. And its very unkikely to not get a seat for full distance. So the seat issue is a falicy.

Yes, with plans to extend. Including the 100km ring line.

Also its not distance, its time thats more important.  So with nearly 40 sations for nearly 20km, this results in an average speed of barely 20km/hr.  The close spacings of the statations is ond of the nextworks biggest hang

The NWRL will eventually connect three branchs, Richmond, Nth Main, NSL. Travel time between each will be less than 20min. At Epping, there will be a significant volume exchange and expect similar at Chatswood. My experience in peak in Sydney that those joining a peak train from around Chatswood and Epping will most likely be standing. I used to stand from Woy Woy to Hornsby, time wise this is similar to total distance on NWRL. And its very unkikely to not get a seat for full distance. So the seat issue is a falicy.

Sydney Orbital Network 2.0: Railway edition. Only $99999999.9*

*Does not include trains.

On a more serious note. the ring line is a great idea, if it joins the 4 main lines out of Sydney and complements the existing network allowing commuters to bypass the city. Removing the amount of people transferring in the city to other lines should benefit all commuters.

Problem on start up it stops a modetate hike from Richmond line. However I think Scholfields station needs to be rebuilt a few hundred metres south to make room for a four platform station, hence the govt probably drew a line in the sand based on cost.

Dubai is also not in Asia.

I actually do give a stuff what others do because it opens your eyes to the options around you and hardly irrelevant. But you obviously care about some places that suit your needs.

So lets start with Paris
- Some lines are tyre technology (5 lines), others steel, so its not a common network
- Lines 14 and now recently Line 1 were converted to Automation and as such their old rollingstock removed and are NOT compatible with other lines any more. Expect more lines to follow
- There are other more subtle differences that prevents fast response of rains running one one line using another. Some is simply train length. They do not mix train lengths on the same line.
- The lines are all operates nominally as separate lines with rolling stock nominally captured by that line even where technology differences do not occur. .
- Interesting comment I found, the Paris Metro was deliberately built with narrow tunnels to prevent mainline trains running on them.
- It is planned some Paris Metro lines will in future have distances that are approaching with NWRL
- Transit times for the longer Paris metro lines to city is similar to that of NWRL
- Automation was introduced as a way to deal with congestion, why because you can run closer spacing!
- "The network reached saturation after World War II, with new trains to allow higher traffic, but further improvements have been limited by the design of the network and in particular the short distances between stations"
"The low speed virtually precludes feasible service to farther suburbs, which are serviced by the RER." So the so called Metro standard that the NWRL does not comply with (according to some) isn't such a great bench mark for what defines a Metro after all?

With reference to other networks
"In contrast to many other historical metro systems (such as New York, Madrid, London, and Boston), all lines (in Paris) have tunnels and operate trains with the same dimensions"

Moscow
- "Long distances between stations have the positive effect of a high cruising speed of 41.7 kilometres per hour (25.9 mph)." Note Paris Metro is down to below 25km/hr.
- Moscow Metro operates as separate systems because of differences in train length limits. Yes its the same stock.


London/ manyGerman cities I have been to
- At least two different technologies


New York
- Different fleets due to curves and widths to suit different routes


- Melbourne, The Heavy Rail St Kilda and Port Melbourne Lines were replaced with Lightrail (RT). It handed 20 years ago, nothing to do with Branding! Out of touch, hardly I caught a tram from Port Melbourne a few weeks after conversion. and St Kilda a few weeks later.
- Adelaide, The current tram line is actually a former heavy rail line
- Sydney, The current LR is a former heavy rail line


Ok, lets go back to the start.
What is the No.1 target of the NSW govt and previous in providing a line to some where in NW (former inner NW project or current NWRL project). AUTOMATION. Why? Because Legacy systems have legacy's including higher operating costs.


Ok, so you got that point, that want to make it automated because its cheaper, more efficient and basically more flexible.


for the govt, what is one of the main issues for the past and present state govts in running Cityrail. COST COST and more COST. its costly to build and costly to operate. Its very high subsidy limits on how much of a network we can have. Its technology is costly to build and we keep having rolling-stock procurement issues that other systems generally don't have. Its also not an issue limited to Sydney, read Qld website to find out about very high fares to try and limit the subsidy to 75%. NSW is around 65%. Melbourne, Adelaide and Perth are not much better if at all. DOO, privatisation, increased ridership etc isn't cutting the subsidy costs to the level they expected.


So next, can you mix automated with non automated? Probably, but its not done, too complex, too much risk, you destroy the benefits of automation for the costs of manual. So the line is going to be seperate. Once you have made that separation, you can do what you like after as there is far less back compatibility that is required. So steeper hills, etc.


You claim there are efficiency gains to be had in Cityrail. So why are they so hard to gain. How much are we talking 1%, 5%? What is that offered by NWRL Automation, 10%, 20%....????


Yes I look at some of these European cities and others including Singapore and Vancouver and I see the same theme, Automation of existing (where practical) and often new greenfield projects! Some cities with existing automated ones like Vancouver changing from one technology to the next for the new projects because they get a better deal and because most Metro/suburban networks are moving away from spaghetti operation to line based. The move to off-the-shelf holistic lumpsum suppliers. Lower risk with less interfaces = lower costs.


I have not proposed conversion of all Sydney to Automation, the obvious ones are NWRL to new route via City, conversion of Bankstown/Huntsville and inner west to Olympic park. I would also propose the Nth beaches would follow the same path if a line was built and the Epping to Chatswood line via Carlingford. This is about 25 years worth of projects. The traffic from these suburban branches will also most likely saturate the city section so consider it full. These routes will also take pressure off the current system in the right areas to allow more services from outer suburban and interurban areas to access the city.


"The original NWRL proposal taken to the last election by the then Opposition Coalition provided for the direct link between Epping and Cherrybrook integrated with the existing ECRL and CityRail DD network. When this morphed into the privatised rapid transit proposal with smaller diameter tunnels, it also suddenly adopted a steeper alignment between Epping and Cherrybrook, unsuitable for DD trains. Coincidence? "



How is it a coincidence that tunnels are now too steep or the tunnels too narrow following decision to go off the shelf? As I stated above and many many times before. Once the decision was made to go Auto, there is zero reason, ZERO to back the line backwards compatible in as many ways as you want. The 1500VDC O/H is being kept, probably because its cheaper, but its hardly a deal breaker or major cost. If you cannot run a train up there because its too steep (there was/are issues with this on ECRL), then tunnel width is irrelevant. If a different technology means tunnels can be steeper and hence shorter, then we are saving money?


"Regardless of the merits of the automated single deck rapid transit, dubious in my opinion as they may be, if the public doesn't like it, then in a democratic society they have every right to vent their spleen on the government at the next election. Engineers in their ivory towers don't always get it right."

What merits of automation do you find dubious? The world has been automating things since the start of the industrial revolution and not stopped. If you get behind a automated network, the benefits scream out. Live response to passenger loading, faster boarder as people realise there an't no one watching the door when it closes.


Its highly unlikely the govt will loose because of a railway line. The people asked for a railway, after years of inaction by previous govt they got one and they are getting one faster than most other lines that have been built in recent years in anywhere of Australia from conception ton completion. They will also get one of the most modern services available at the time. The studies show that the people in this area who use PT or likely to use PT travel to Epping where they can catch a suburban or alight from the line at or prior to Chatswood. 1/3 would travel onwards towards the city. Its their traffic movements, not the govts. The Govt has also continued to focus on the city extension so the Chatswood terminus is as short as possible, but still likely to be 5 years. What the NWRL users will get is the ability to watch out the front and back of the trains and this is EXTREMELY popular in places its available (from what I have seen and where I personally stand with my kids each time I ride a Auto train).


Most engineer's don't work in ivory towers and they deal with real issues including unrealistic budgets and expectations by clients and usually travel visit to benchmarks and other options. This includes the so called divine right to have a single train passage from home to work that many Sydney commuters have become very used to.

Err RRT_Rules

Dubai is in Asia, It is in the Middle East which is a part of Asia! Dubai is not in South East Asia.

The Heavy Rail St Kilda and Port Melbourne branches were replaced by a Tram, because the Government at the time did not want to pour money into repairing the Sandridge Bridge crossing over the Yarra, hence they built a short section of tram track along Whiteman St in South Melbourne to link the former Heavy Rail alignment with the Tram Services along Clarendon Street.

The Point I think that most posters are trying to make is that The Sydney Rail Network is a commuter rail network in the main and it makes no sense to separate system and to spend billions converting the existing system. Automated systems that go any distance will still need a member of staff on that train for safety should auto operation fail. To my knowledge the only systems that do not have staff on them are the people mover systems which are in the main very short systems, correct me if I am wrong.

Michael

Michael,
I'll pay the asian comment, my flag poster had Dubai in Africa and Cyprus in Asia, clearly wrong.

I wont pay the attendent on auto trains except very short. WIKI ix a good reference. Also why does distance come into play? If a train of 400 breaks down, does it matter how far it was travelling?

I beliieve most auto trains are commuter lines, apart from the likes of airport shuttles.

I looked up Barcelona last night. 45km planned extension on auto lines.

Vancouver, 68km of automated lines, longest line 42km, average speed 45km/hr, a city very similar in profile to Sydney as the geography forces the smaller population to grow in a longer narrrower arc and they are building more with another 11km branch to open in 2 years with more on the books including extdnsions to the existing lines. Designd with a headway capacity of 75sec.

Michael,
I'll pay the asian comment, my flag poster had Dubai in Africa and Cyprus in Asia, clearly wrong.

I wont pay the attendent on auto trains except very short. WIKI ix a good reference. Also why does distance come into play? If a train of 400 breaks down, does it matter how far it was travelling?

I beliieve most auto trains are commuter lines, apart from the likes of airport shuttles.

I looked up Barcelona last night. 45km planned extension on auto lines.

Vancouver, 68km of automated lines, longest line 42km, average speed 45km/hr, a city very similar in profile to Sydney as the geography forces the smaller population to grow in a longer narrrower arc and they are building more with another 11km branch to open in 2 years with more on the books including extdnsions to the existing lines. Designd with a headway capacity of 75sec.

Hi RTT_Rules

I stand corrected, I think most automated lines that go Underground, Victoria Line, DLR have an attendant should anything go wrong. I am not being Party Political here. Gladys Berejiklian is a Fantastic administrator. She has rescued the Waratah Train contract and is going to complete both the North West and South West Rail links. A refreshing change from years of ALP inaction. And you can send her down to Victoria, she is much better than Mulder !!!

They may be commuter lines but they are probably stand alone,self contained lines much in the way that London Underground operated, not sharing track with any other line. Sydney's Rail system is not configured in that way but is rather a network of lines, for e.g. South West Trains or Chiltern Trains in the UK. Completely new lines purpose built metros within a 25km of Sydney's CBD are suitable for Metro. A Railway 25 to 50 km out from Sydney is better served by a commuter rail system. Moreover converting the rest of the City Rail network many of them sharing tracks with express services, NSW Train Services and Freight Services is difficult to accomplish

Michael