Sydney Metro - Tallawong to Bankstown

[moa999]

it is not just as simple as replacing the rolling stock, witness the years of weekend and Christmas shutdowns needed to convert the Bankstown line.

Quite a few things contributing to the shutdowns.
Most notably because it's a new type you need to meet accessibility standards.
So that's adding the requirements for platform leveling, straightening and gap barriers and track levelling.
And then in addition they are replacing signalling (for automated trains) and PSDd (for safety and automated operations)

I'd actually think if you introduced a new DD set to the Sydney system tomorrow (and not more Waratah Bs or further derivations that can be added under grandfather provisions) then some of those might also apply.

And suspect that will play into decisions on Tangara replacements in coming years.

[tonyp]

I'm as puzzled by that as I am about the amount of money being spent to run more than 20 trains per hour through the city underground, so I've gone back to a post of mine from 2007, where I noted that in the 1938 timetable the number of trains departing Wynyard in the 60 minute period from 5pm was as follows:

• Up Shore to Down Western and Northern lines: 26

• Up West and North to Down Shore: 21

• Wynyard low level dead end to Bankstown and Local West: 27

• St James dead end to Illawarra and Kingsgrove 25.

Those 27 trains per hour being turned back at Wynyard Low Level dead end shunt equated to an average frequency of 2.2 minutes. Each train had only 4 minutes to shunt and reverse, which must have taken some quick work!

How did we do it then?

What had changed by the 1960's?

Talent. In the 1930s, Sydney's public transport, including that brilliant tram system that moved phenomenal numbers of people very efficiently (considerably more than the railways at that time), was run by a generation of Australian-born specialists whose careers coincided with an equally brilliant period of technical and engineering education, emerging from the then-young Sydney University and Sydney Technical College. Born in the late 19th century, their careers peaked in the first half of the 20th century and most of them had gone by post WW2. They were talented, pragmatic and hands-on people and of course the visionary Bradfield was one of them (though originally educated in Queensland). It was the period when NSW public transport was at its brilliant best.

Notably also, there were almost no imported British managers. This came after WW2 and the general decline in talent was certainly evident in the move from the efficient tram system to the bumbling DGT bus operation. The railways still had (and have) a lot of talent, but a systemic change crept in, one in which checks and procedures became more burdensome, a practical consequence being a cautious slowing down of how everything ran. I think these limitations have subsequently been well recognised and led directly to the solution of the automated rapid transit system to recover the lost achievements of past years. The metro is your single deck train system of decades ago but even better.

[Campbelltown busboy]

This metro and all future metros will be a good thing for Sydney's public transport system as a whole in the long term. It would take years to convert the entire suburban network to a metro system then there is the issue of main line intercity regional and freight trains

[Linto63]

Public transport went from being profitable pre WW2 to requiring large subsidies. Combined with the arrival of affordable cars and thus an increased spend on roads, it dropped down government’s list of priorities.

[Swift]

Don't know why we cling to this outdated notion that poms know best when their transport system isn't exactly held up as a beacon to the world.
It's like the railways are still stuck in a 1950s mother England attitude.
Get rid if them!! Replace them with people from any developed non English speaking country.and watch it thrive.

[Fleet Lists]

We are now getting totally off topic- this thread is supposed be about Tallawong to Bankstown Metro.

[Swift]

This metro and all future metros will be a good thing for Sydney's public transport system as a whole in the long term. It would take years to convert the entire suburban network to a metro system then there is the issue of main line intercity regional and freight trains

Who says we need to convert it? Just stop ordering deckers all the time.

[boronia]

Those 27 trains per hour being turned back at Wynyard Low Level dead end shunt equated to an average frequency of 2.2 minutes. Each train had only 4 minutes to shunt and reverse, which must have taken some quick work!

How did we do it then?

What had changed by the 1960's?

A system of "relay drivers" was used to shunt the trains through the cross over. A driver at each end of the train allowed almost instant reversal.

[tonyp]

A system of "relay drivers" was used to shunt the trains through the cross over. A driver at each end of the train allowed almost instant reversal.

I don't think that this would occur to them nowadays. The metro has inbuilt relay drivers.

[tonyp]

a 21st century metro with trains with better acceleration and deacceleration characteristics will always outrun a train on the classic network. There would be a big problem if it didn't.

The Fremantle line in Perth was opened in 1881. The very similar Bankstown line in Sydney was opened between 1895 and 1909. Both lines have not changed in profile since then, apart from modern signalling. The Fremantle line is operated by single deck stock as a rapid transit service, the Bankstown line is operated by double deck stock as a standard suburban service, soon to be converted to a single deck rapid transit operation.

To compare both lines within the same parameters, the distance between Perth and Mosman Park is 13.5 km with 11 intermediate stops. The distance from Sydenham to Bankstown is 13.4 km with 9 intermediate stops. The comparison is based on services stopping at every stop.

Sydenham to Bankstown is presently traversed in 25 minutes at an average speed of 32 km/h.
Sydenham to Bankstown will be traversed by the metro in 21 minutes at an average speed of 38 km/h.
Perth to Mosman Park is traversed in 21 minutes at an average speed of 38 km/h (including two more stops than the Bankstown line).

Remember, both lines are legacy lines over a century old. Single deck trains are snappier than double deck trains, regardless of whether the line is Victorian or space age.

It's often argued that the double deckers can be revved up with new signalling. Let's see how that goes on the completely up to date Paris RER A, the poster child of double deck advocates.

Line A from Boissy-St-Leger to Cergy-Le-Haut is 63 km with 24 intermediate stops. The services take 80 minutes at an average speed of 47 km/h.

Sydney Metro from Tallawong to Canterbury is also 63 km and has 23 intermediate stops. The journey will take 68 minutes at an average speed of 58 km/h.

Whichever way you look, ancient or modern, single deck trains deliver faster urban journey times than double deck trains and can do so without cheating by skipping stops. This also solves the injustice of people who use stops that are regularly skipped having a poorer, less frequent service than others, not to mention delivering more trains, thus more capacity along a line. A single deck rapid transit service is the only satisfactory method for urban commuter rail. Double deck trains may be suitable for journeys beyond the urban areas but not within.

[Swift]

They need to suspend ordering suburban double deck stock and gradually replace most of the dd fleet with sd and only maintain a minority as dd. We need to get with the times instead of defaulting to the late 60s for purchasing policies.

[Linto63]

Other than being bored and wanting to pick a fight, I have no idea why you want to go over this yet again. Nobody disputes that single deck trains with superior power to weight ratios and their ability to load and unload are quicker. If speed was the only factor, yes you would go for single deck trains. But it isn’t, there are other factors where double decks are superior, e.g. more seats.

Anybody can all pick and choose statistics to come to a conclusion, but without factoring in different underlying circumstances, they prove little.

We can hypothesise that Sydney should only have a single deck rapid transit service, but the reality is that double deck trains will still be the nucleus of the fleet long after we are all gone.

[Glen]

What was the typical (faster) running time used on the Bankstown line before the great slow down of twenty years ago?

We need to keep remembering that we deliberately timetable trains with unnecessary and extraordinarily slow running speeds and long dwell times in Sydney.

[Aurora]

The July 2004 timetable provided for an average 23 minutes between Sydenham and Bankstown stopping all stations with 2 min intervals between most stations.

[Glen]

I did a lot of running time research almost 20 years ago when CityRail was (then) planning to slow down the train timetables, supposedly because of 'safety' (post-Waterfall) but in truth because they were desperate to improve On Time Running.

Here is an extract of that information for the Bankstown line. I had measured Bankstown - Central all stations or vice versa. Bear in mind that can be impacted by the amount of recovery time they place at Bankstown or Central.

Some of you may notice the influence of newer double-deck stock having an impact on off-peak running times in 1992.

TIME- Down Up
TABLE minutes minutes
YEAR

1938 31 31
1949 31 31
1952 32 33
1957 32 33
1958 33 33
1960 33 34
1962 33 33
1973 33 32
1980 31 32
1992 29 30
2000 31 32
2005 36 37
2022 38 35

[tonyp]

there are other factors where double decks are superior, e.g. more seats.

And even that is subjected to a dishonest train by train comparison by double deck supporters when the legitimate comparison is seats per hour where the difference becomes negligible.

Presently there are typically four trains an hour on the Bankstown line with the lucky stations that don't get skipped having six trains an hour. That's 3,600 seats per hour for all stations, or 5,400 if you're at a stopping station. The metro will have an opening service frequency of 15 trains per hour in peak and 6 trains per hour off-peak, that's 5,670 seats per hour in peaks and 2,268 seats per hour off peak - at every station. The latter figure is lower, but of course trains are usually far from full off-peak, so getting a seat is rarely a problem. The ultimate potential capacity of the metro is 2 minute headways with eight-car trains, that's 15,120 seats per hour.

That's more realistic than saying that a metro train has only 378 seats in a six car set (or 504 in an eight car) compared to 900 in a double deck eight car set. That doesn't give anything remotely like a true, honest picture.

I did a lot of running time research almost 20 years ago when CityRail was (then) planning to slow down the train timetables, supposedly because of 'safety' (post-Waterfall) but in truth because they were desperate to improve On Time Running.

Here is an extract of that information for the Bankstown line. I had measured Bankstown - Central all stations or vice versa. Bear in mind that can be impacted by the amount of recovery time they place at Bankstown or Central.

It would be interesting to see those figures for only the 13.4 km section between Sydenham and Bankstown because that would more truly reflect the running time along the Bankstown line. The suburban trains have a disadvantage in being subjected to possible slowdowns and delays between Sydenham and Central, compared to the metro which will have an uninhibited run between Sydenham and Central. I do try to give the suburban services the maximum advantage in these comparisons as it distorts the picture to compare a free running metro against a section of suburban line that may be subjected to delay and disruption.

I did some time ago compare historical figures over that range of dates for Sydenham-Bankstown (though I no longer have them with me) and, while there were certainly faster services in the past, I recall that the metro will still outperform them, just by a smaller margin. It's not surprising when the comparison is between high-performance single deck stock with three doors per side per car and lower performance single deck stock with two doors, or double deck cars with all their internal issues and only two doors.

[Glen]

Sydenham to Bankstown is presently traversed in 25 minutes at an average speed of 32 km/h.
Sydenham to Bankstown will be traversed by the metro in 21 minutes at an average speed of 38 km/h.

I did a lot of running time research almost 20 years ago when CityRail was (then) planning to slow down the train timetables, supposedly because of 'safety' (post-Waterfall) but in truth because they were desperate to improve On Time Running.

Here is an extract of that information for the Bankstown line. I had measured Bankstown - Central all stations or vice versa. Bear in mind that can be impacted by the amount of recovery time they place at Bankstown or Central.

It would be interesting to see those figures for only the 13.4 km section between Sydenham and Bankstown because that would more truly reflect the running time along the Bankstown line.

I did some time ago compare historical figures over that range of dates for Sydenham-Bankstown (though I no longer have them with me) and, while there were certainly faster services in the past, I recall that the metro will still outperform them, just by a smaller margin. It's not surprising when the comparison is between high-performance single deck stock with three doors per side per car and lower performance single deck stock with two doors, or double deck cars with all their internal issues and only two doors.

Here you go:

Off-peak running times between Bankstown and Sydenham & vice versa

TIMETABLE YEAR

1938
Down - 22 mins
Up - 22 mins

1992
Down - 21 mins
Up - 21 mins

BTW, in 1938 the peak hour running time was also 22 mins.

[tonyp]

Thanks Glen. Yes, that's pretty much what I recall finding. For a single deck train, there's no reason that a peak period run should take any longer than an off-peak one. If there is a difference, it's an indication that the trains aren't designed to process the turnover of large crowds. Of course there were less people in 1938. How would you explain 1992 and was it found to be unsustainable? I found that the previous efforts in the 1980s simply resulted in late running trains and a general breakdown.

[Linto63]

And even that is subjected to a dishonest train by train comparison by double deck supporters when the legitimate comparison is seats per hour where the difference becomes negligible.

Presently there are typically four trains an hour on the Bankstown line with the lucky stations that don't get skipped having six trains an hour. That's 3,600 seats per hour for all stations, or 5,400 if you're at a stopping station. The metro will have an opening service frequency of 15 trains per hour in peak and 6 trains per hour off-peak, that's 5,670 seats per hour in peaks and 2,268 seats per hour off peak - at every station. The latter figure is lower, but of course trains are usually far from full off-peak, so getting a seat is rarely a problem. The ultimate potential capacity of the metro is 2 minute headways with eight-car trains, that's 15,120 seats per hour.

An apples and oranges comparison. The Bankstown line has long been capable of accommodating 15tph between Bankstown and Redfern, ever since the T8 services were diverted via the Airport 20 years ago. The cork in the bottle is that it has to share the City Circle with T2 and T8 services. If it had a dedicated line into the CBD as the M1 will, 15 tph could easily be accommodated, which would be 13,500 seats per hour. Or put another way, if the M1 had to share the City Circle with the other services, it would get nowhere 15tph.

That's more realistic than saying that a metro train has only 378 seats in a six car set (or 504 in an eight car) compared to 900 in a double deck eight car set. That doesn't give anything remotely like a true, honest picture.

In the context of this discussion, the point was made that double decks can carry more seated passengers. While you seemingly don't consider whether passengers sit or stand an issue, others do. Your suggestion that a dishonest picture is being portrayed is without foundation.

How would you explain 1992 and was it found to be unsustainable?

Timetables are set so that the slowest stock can operate any service, 1992 was the year after the last red rattlers were withdrawn, hence a reduction in journey times taking advantage of the better performance of the double deckers.

[tonyp]

An apples and oranges comparison. The Bankstown line has long been capable of accommodating 15tph between Bankstown and Redfern, ever since the T8 services were diverted via the Airport 20 years ago. The cork in the bottle is that it has to share the City Circle with T2 and T8 services. If it had a dedicated line into the CBD as the M1 will, 15 tph could easily be accommodated, which would be 13,500 seats per hour. Or put another way, if the M1 had to share the City Circle with the other services, it would get nowhere 15tph.

In the context of this discussion, the point was made that double decks can carry more seated passengers. While you seemingly don't consider whether passengers sit or stand an issue, others do. Your suggestion that a dishonest picture is being portrayed is without foundation.Timetables are set so that the slowest stock can operate any service, 1992 was the year after the last red rattlers were withdrawn, hence a reduction in journey times taking advantage of the better performance of the double deckers.

The context of this discussion is seated passengers. If we talk total capacity, the metro smokes the suburban services.

So let's talk just apples and assume that the Bankstown line moves 15 tph which is 13,500 seated passengers per hour. The ultimate seated capacity of the metro at its full operational potential is 15,120 seated passengers per hour. So there are not "more seats", as you state, in a double deck service on the Bankstown line, there are slightly less. There are more seats in a double deck train, but ultimately not more seats in a double deck train service, if you get that critical distinction. I rest my case that it's a dishonest picture to compare seating capacity on a train by train basis.

How long did those 1992 running times last? For many years, or did they progressively dissolve in a growing heap of trains not being able to keep up with the timetable?

[Swift]

I caught a suburban train after a long time between Epping and Meadowbank today. It absolutely dawdles most of the way. I was just shaking my head at it.

[Linto63]

The ultimate seated capacity of the metro at its full operational potential is 15,120 seated passengers per hour. So there are not "more seats", as you state, in a double deck service on the Bankstown line, there are slightly less. There are more seats in a double deck train, but ultimately not more seats in a double deck train service, if you get that critical distinction. I rest my case that it's a dishonest picture to compare seating capacity on a train by train basis.

In order to keep things on a like-for-like basis, it the metro can at full operational potential seat 15,120, then double deck trains operating at three minute intervals, as they do at times on some lines, would have 18,000 seats. Suburban smokes metro. Again your suggestion of a dishonest picture is flawed.

[boronia]

[Glen]

Not sure how we got here but anyway.......

Apologies if this has been published before, a comparison of DD vs SD by the ABC in 2014:

https://www.abc.net.au/news/2014-04-11/ ... %2Ddeckers.

[Merc1107]

The article puts forth the hypothesis that less people will use the Metro because of lesser seating capacity. How's that going?