Double Decker Trains on QR tracks

[jason]

On reading a train feasibility study recently, it stands to reason the commercial ability to introduce progressively on QR network. Taking some face value comparisions on the two different sets of trains, you have to allow for some differences including carriage width of the two (QR is approximately 2850mm and CityTrain is 3034mm)

Seating Capacity:
144m IMU 101-110 will seat 448 plus standing room ( 2 x 3 car set )
140m OSCAR H set will seat 794 plus standing room ( OSCAR H (a) + 5 x OSCAR (c)  + OSCAR (a) )

Overall Height:
IMU 3900mm
OSCAR 4403mm

Weight for a comparable 140m unit
IMU 164t
OSCAR 344t




Whilst there are a couple of issues in introducing a similar set of carriages to those like the OSCAR, the prelim feasibility study showed that it was tecnically possible, and also commercially feasible.

It should also note that some alterations are required to the current QR metro network:
* Lowering track height at bridges and tunnels. This is similar to the same work practice undertaken at the new Beaudesert Road Overpass
* standardising the overhead wire height to 5000mm ( 860mm lower than the max height for the IMU and other electric locomotives for contact )
* 2 seating configerations, one design for high occupancy suburban lines and one for inter-urban, one for max capacity and one for comfort
* introduce a similar configeration of individual car components instead of "sets" as currently used by QR

[ozbob]

Interesting data Jason, thanks for posting it.

Clearly we need to ramp up capacity significantly.  Sydney is struggling too at the moment (see http://backontrack.org/mbs/index.php?topic=946.msg4158#msg4158 ).

One of the issues with double deckers is the increased dwell times, but that is something that could be worked on.
Most double deckers that I am aware of are on 1435mm gauge.  The 1067 mm gauge here in Brisbane might be a limitation.

This has some interesting stuff too -->  http://en.wikipedia.org/wiki/Bilevel_car#Double_deckers_for_capacity

In Japan they often run 12 car unit trains sub pass on the 1067 mm.  This might be an option too, that is increase some services to 9 cars.

[mufreight]

Resignaling so as to allow shorter headways between trains to as little as 2 minutes which is dependent on station dwell times is a preferable and more ecenomic solution than the cost of modifing the clearances to enable the operation of double deckers, even with the minimum overhead height of 5 meters it would still be a somewhat constrained in terms of height double decker which would be further restricted by the need to relocate traction transformers & control, and air conditioning equipment.

More trains operating on closer frequencies would prove to be a better option both for the railways and the passengers.

Prior to Sydneys latest order for Double Deck rollingstock consideration had been given to  a new generation of single deck rollingstock rather than more double deckers which are now starting to slow the system down in peak because of station dwell times, there is a limit to the number of doors that can be built into a Double Decker and their size, single deck cars could be built with three or four doors for shorter dwell times.

[stephenk]

mufreight makes quite a few good points here.

Improved signalling with a system such as SACEM would be a much more cost effective way or increasing line capacity. Adding more doors and standing room would be a good option for increasing train capacity, and reducing dwell times.  Increasing platform and train length may also be an option for improving train capacity (Japanese commuter trains can be approx 200m long, compared to Brisbane's approx 140m long trains.

Introducing double deck trains would be very difficult on QR infrastructure without making very expensive and difficult infrastructure changes. Also the two door double deck design significantly increases dwell times, and thus does not actually increase a line's passenger carrying capacity compared to single deck trains.

[jason]

Most of the comments doesn't appear to be very thought out.

The 2 min head way is not very practical. As currently experienced on the SE Busway, which has a headway of 30 seconds, any delays has serious effects on all services behind. 2 minutes is not a very practical headway time.A more practical 10min would work

Other rail networks with double decker cars do have longer dwelling times, but statisically the dwell time for loading and unloading per passenger is shorter than single deck trains.

platform height does not need to be adjusted, entry level can be at the current platform height and the internal heights to suit

increasing platform length is a good idea, but costly. In situations where stations are on or near a curve, is there sufficient space to extend the existing platform, or does the station need to be relocated to a more suitable site. The time to build will have to be over a 10 to 15 year time, considering the time to plan, engineer and build the extended platforms, even for an average 3 months per site, still will take years.

Door opening width should be a percentage of total passenger space of the cars, the bigger the space the bigger the doors. I frequently travel on Sydneys trains and the actually have a shorter dwelling time than QR. The problem with QR passengers is that they disembark slower compared to Sydney.

Whilst Cityrail may have considered single deck trains, ultimately Double deckers were more favourable.

QR has to increase the minimium height of its overhead infrasture to allow for double height containers on its freight network.

[mufreight]

To Jason
Sorry to be a party pooper but I realy think that you have this one a bit out of focus, my comments are based upon actual experience on the job and I would suggest that I would like some of what you have been smoking for your pipe dreams.  With regard to Sydney DD trains having driven them I can tell you that in peak they have longer dwell times than the single deck sets here.
In 1960 when I first commenced working for NSWGR peak services on the City Circle could and often did run on less than 2 minute headways, all single deck sets, as one train cleared a platform another approached, this close operation was possible due to the complex signaling system and train trips that activated the brakes if trains got too close.  The system then in use was by comparison with todays technology primitive, but it worked and is still in use today. 
Today with modern technology an automated system could now be worked using driverless trains on headways as close as the station dwell times.
That the new proposed 38km Metro system for Sydney is to be single deck sets with four doors each side operating on three minute headways pretty much tells the tale there.
Your suggestion of raising the overhead for the purpose of enabling the operation of Double stacked containers and DD passenger stock would be great but it is unfortunately simply not practical, every over rail structure would have to be raised, tunnels would need to be raised then the overhead raised to obtain the needed clearances, 6.5m above rail minimum for the overhead and another .5m to provide clearance for the overhead itself making 7m, financially simply not on and as quick look at Central,Brunswuck Street and Ipswich stations shows that the required clearances is not physicaly practical.
Rolling stock design would be sadly compromised as using a high voltage AC power system for traction our air conditioned rolling stock carries quite a lot of equipment all of which has to go somewhere, if a carriage is built with a drop belly between the bogies to accomodate passengers for a DD configuration there is no room under the car and the space above the passenger space at each end of the cars is limited and equipment installed there is more difficult to maintain with some equipment unable to be fitted there anyway.
Any gain simply aint worth the complications.
The provision of a section of underground from Dutton Park to the exhibition line via Gabba,Gardens Point, and Central and a closer signalling system is more cost effective and makes a better solution to the system's capacity problems.
Cheers

[glossyblack]

Dwell times in Sydney at the moment are appalling and really do not allow for close running or passenger comfort,last week when I was in Syd if you sit close to vestibules, you are continually being trodden on by people trying to get off/on. If dd's are to work anywhere, a how to use them public campaign is needed. RE widths Tangara is only 3000 mm according to Cityrail

[stephenk]

The 2 min head way is not very practical. As currently experienced on the SE Busway, which has a headway of 30 seconds, any delays has serious effects on all services behind. 2 minutes is not a very practical headway time.A more practical 10min would work

You cannot compare bus and rail operations in the way you have done, they operate in a very very different manner! 10 minute practical headway - you are having a laugh arn't you? Citytrain already runs 3-4 minute headways on the core section.

Paris RER A and Munich S-bahn are both suburban rail services (like QR Citytrain) and manage 2 minute headways with modern signalling.  There are plenty of Metro systems that can operate headways less than 2 minutes - Moscow, Prague, Kiev, Tokyo, Santiago, St Petersburg, Sao Paulo, and Paris being examples.

[mufreight]

As said previously trains have been running on less than 2 minute headways on the Sydney City circle since the completion of that line, the key factor in this day and age is the station dwell times of the DD sets in use at the present time not the signalling system or the length of the track blocks.
As for busses they can operate on bumper to bumper headways and when one stops can overtake each other pretty well anywhere so the comparisons between bus and rail operations headways is a no brainer something that one would think should be obvious to a highly educated architect.

[jason]

A 10min is not practical but when factoring on time runnning performance standards normally being withing 6 minutes of timetable.

There seems to be a misunderstanding, from my behalf, on where we are talking about for a 2 min and 10 minute headway.

Other forums and what i was talking about a 10 minute headway outside the CBD. if a train was running within "on time" of up to 6 minutes than it won't affect following services, allowing up to additional 4 minutes "lateness" and "early running" as well.

I do note comments by mufreight about 2 minute headway, but this is part of  cityrails major problem as the minimium headway and intertwining networks, not allowing any late services with out affecting following on services.

My comparison of buses to trains was like comparing chalk and cheese. It wasn't suppose to be a direct comparison, but a statement of what might look good on paper, doesn't necessary work.

The busways are designed with a head way of 30 seconds, but as often experience in the Mater Hill / Buranda / Woolloongabba section the 30second headway doesn't work. factoring in lights sequence, bus dwelling time at stations and saturation, buses are often delayed.

Mufrieght. Its actually nice to be able to talk to someone else who has / is working in this industry. I understand where you are coming from. From my point of view i have worked as an independant person on papers for various State Governments and overseas agencies as well from a Business Economics and Town Planning standpoint.

My comments regarding double decker trains dwelling times were based on research papers prepared by others on the number of passengers serviced per hour when factoring in dwelling times and running times. I don't disagree with you that wider doors are required and longer dwelling times are needed. but studies showed that when factoring this, the rail capacity per hour increased.

The federal government is already investigating standardising the height of the overhead cables with the introduction of double stacked containers on the freight network. This was only a comment to expand the standardised height through out the whole network.

With regards to tunnels and bridges, QR is actively replacing a number of its existing pasenger bridges through out its network, at stations and independant locations. It is more cost effective to increase the size of existing tunnels as the "down time" as each tunnel width / height is increased is a lot less than the time taken to construct a new tunnel. I know its not as good as having new platforms and lines, but in conjection and as a wider implementation along with, its a good long term solution.

Where bridges are already in place, it maybe more economical to lower the track height similar to Beaudesert Road Acacia Ridge project. Here QR etc have lowered the track to allow trains to pass under the new road bridge

The same has been done in other situations before, in australia and overseas. From a cost effective, time to implement and value for money it appears that this is a viable option for the medium to long term.

Basically there is no short term solution here, each has there own merrits and i totally agree with upgrading the signal system, but like all solutions its only as effective as the weakest link in the chain. like double decker trains.

One solution which greatly improve brisbane buses was to stop having across town services through the CBD. Often buses on one side of town were delayed which had an effect on on-time running when reached the other side.

If trains during peak times were only point to point, say Roma Street - Ferny Grove, where passengers weren't trying to disembark and get on at the same time. instead you have one way traffic (passengers) only, similar to melboure's city rail loop where all services start at a single station and proceed from there.

[mufreight]

To Jason
On the subject of headways it would seem that you have no practical conception of the purpose of the train controlers function, an example of this is that if on short headways on train because of delays regardless of their cause then that train is run out of course and other movements may be slotted in ahead of it so that they remain on table,  The shorter the headways the easier it is to run a train out of course or off the table.
The key problem with Sydney City rail at the present time is now station dwell times which are made worse by saturated platforms which in itself increases dwell times and in turn restricts the passenger throughput.
On the subject of operating clerances for double stacking yes there is work currently being carried out on the standard gauge network to raise clerances but there is no such work being carried out in relation to the electrified networks, again although being constructed with clearances for double stacking the southern freight line in Sydney is unlikely to see any such movements due to shared tracks with the electrified network.
Double stacking on narrow gauge, stability problems mean restricted speeds and the higher axle loads of double stack operation render double stack container operation in Queensland impractical even without the consideration of operating clearances.  Lowering tracks in tunnels is not always possible because of the proximity of other below rail structures.
More reseach might help but it is all a matter of balance.
Cheers

[stephenk]

A 10min is not practical but when factoring on time runnning performance standards normally being withing 6 minutes of timetable.

There seems to be a misunderstanding, from my behalf, on where we are talking about for a 2 min and 10 minute headway.

Other forums and what i was talking about a 10 minute headway outside the CBD. if a train was running within "on time" of up to 6 minutes than it won't affect following services, allowing up to additional 4 minutes "lateness" and "early running" as well.

I do note comments by mufreight about 2 minute headway, but this is part of  cityrails major problem as the minimium headway and intertwining networks, not allowing any late services with out affecting following on services.

If you are running a 10 minute headway on branches, and the train is running just 1 minute late, it will still delay other trains from other branches where the lines converge. If that train has to make a conflicting movement across another line it will also cause a delay to other services. Thus you cannot make an allowance for trains running 4 minutes as you stated. Trains must run as close as possible to the timetable at all times!

You mentioned running early? That should not occur in a properly run railway. You would get an awful amount of complaints if a train left a station early.

Also, the further apart you schedule trains to increase the operating margin, tends to actually make things worse. The New York Subway is a good example. They had problems running a circa 2min timetable as trains would catch up with the trains in front and slow things down. So they cut out some trains and tried to run approx 2.5min service. Unfortunately, as there were less trains, they became more crowded, and the dwell times increased. The increased dwell time resulting in trains catching up with the trains in front again, and with more operating margin to play with, the train bunching became even worse! Allowing excessive operating margins to make up for delays is a very lazy way of running a railway!!

[jason]

I think this is why there is major misunderstanding between the lay person and professionals