Article: Debate rages over Beaudesert rail service

[colinw]

I'm with you on that. We sure seem keen to give up and say "too hard, can't fit any more in" at levels of service well below what other cities manage, then cite $billions of extra infrastructure as required to even get past pathetic half hourly train frequencies.

The stumbling block definitely seems to be the Merivale bridge, specifically about 1 hour in peak where the bridge is running at 20 tph.  Merivale also comes into play in a big way if you desire to go to quarter hourly offpeak services on the South Brisbane group of lines.

Inbound the worst case currently is the hour from 8 AM where there are departures from South Brisbane towards Roma Street at roughly 3 minute intervals for the entire hour: 8.02, 8.04, 8.06, 8.09, 8.12, 8.15, 8.18, 8.21, 8.24, 8.28, 8.32, 8.34, 8.37, 8.40, 8.43, 8.47, 8.47, 8.49, 8.53, 8.58.  Clearly no room to put a PEAK Browns Plains service through South Brisbane, particularly in the hour from 8 AM, and in fact little or no room to even add any more Cleveland/Beenleigh/Gold Coast trains.

At other times of day you could comfortably fit in a half hourly Browns Plains service, e.g. in the middle of the day there is a mere 6 tph over the bridge: (12.01, 12.19, 12.24, 12.31, 12.49, 12.54).  It is easy to see where a Browns Plains service could fit through here:

* = hypothetical half hourly off-peak Browns Plains service :-
(12.01, 12.10*, 12.19, 12.24, 12.31, 12.40*, 12.49, 12.54).

The problem is that half hourly offpeak really isn't good enough.

Once you get to quarter hourly offpeak on each of Browns Plains, Beenleigh/Kuraby, Gold Coast and Cleveland/Manly, you're at full time 20 tph across Merivale, which clearly is not going to be robust.  If Merivale remains the sole southside route in use, then 4tph offpeak is out of the question, although maybe 20 minutes / 3tph is achievable?

What then to do about Browns Plains?

The long term, and best solution is CRR.  That will create enough additional capacity to cater for Browns Plains (even if it goes via South Brisbane, replacing Beenleigh & Gold Coast services now routed via CRR).

If we assume CRR will go ahead, and be open by 2016, then I think it best to simply stage Browns Plains AFTER 2016 and route it via South Brisbane utilising the spare Merivale Bridge capacity freed by putting Beenleigh & Gold Coast down CRR, or else route it via CRR and put one of the others via South Brisbane.

But what if CRR doesn't go ahead?

In the interim, a Browns Plains service could fit in off peak at current CityTrain frequencies, as I have shown above, BUT NOT AT PEAK or at the desired level of 4tph off peak.  

For peak services, we would be faced with a number of less than optimal possibilities or some "off the wall" suggestions :-

1.  Route via Tennyson as suggested above.  This will suffer from conflicting moves at the Yeerongpilly & Sherwood junctions, and at Milton to Roma St, but might still be doable with some junction & signalling improvements.
2. Add a terminating road & platform at Salisbury (or some point inbound from Salisbury - Moorooka?).  Force a change of trains there.
3. Find space to terminate the service at Park Road - a terminating road to a 5th platform.  Force change there, to train or the busway.  There's room to put in a "dock" platform at Park Road here.  Have a good look at the layout of Park Road, and note the big gaps between the disused dual gauge platform 4 & platform 3, and between the road leading to platform 3 and the original pair of tracks that lead to platforms 1 & 2. Clearly room for one or two terminating platforms.
4. Off-the-wall suggestion - up over the freight flyover at Dutton Park & terminate at a 3rd platform at Buranda with onward busway & Cleveland line connections each way. Will obviously block out formerly dedicated freight capacity.
5.  Via Tennyson & terminate at Corinda with a change.
6.  Build a 4th platform at South Brisbane where the standard gauge goes behind the station.  Terminate there.

The counter-peak XPT departure along the dual gauge is a nuisance that will have to be dealt with as well.

Of all these, the ones I'm most in favour of are either routing via Tennyson & Sherwood, or building a terminating dock platform at Park Road.

[ozbob]

Some good ideas there Colin as well as the reality of looming issues.   

[#Metro]

Inbound the worst case currently is the hour from 8 AM where there are departures from South Brisbane towards Roma Street at roughly 3 minute intervals for the entire hour: 8.02, 8.04, 8.06, 8.09, 8.12, 8.15, 8.18, 8.21, 8.24, 8.28, 8.32, 8.34, 8.37, 8.40, 8.43, 8.47, 8.47, 8.49, 8.53, 8.58. Clearly no room to put a PEAK Browns Plains service through South Brisbane, particularly in the hour from 8 AM, and in fact little or no room to even add any more Cleveland/Beenleigh/Gold Coast trains.

Well maybe the Merivale Bridge isn't at capacity in 2016, maybe it is actually at capacity right now? 
I don't see how any more peak trains can fit in there to be honest. However, I did read something in the 2031 plan or
supporting documents that a "cheap" signal upgrade could increase trains/hour to 24- so an extra 4 trains.
A more "expensive" upgrade would increase that to 30- so an extra 10 trains through there.  Any comments about this?

It would be very interesting to know what a signal upgrade could actually do as the paragraph I read was very general, and this generalisation may not apply to specific parts of the network. Signal upgrade, even an expensive one would be far far cheaper and quicker than $8 billion CRR and could be done as an emergency stop-gap measure. That chokepoint is holding back service levels and improvements to rail service and frequency along that whole branch.

But what if CRR doesn't go ahead?

UNTHINKABLE!!!

[ozbob]

http://english.peopledaily.com.cn/mediafile/201001/05/P201001052033151445725051.jpg

[ozbob]

Clearly more trains via Tennyson is an option.  This will buy a few more services.  The more I think it about, the more I am leaning to a double stack twin tunnel  4 track.  You are only going to get one crack at this.  

Melbourne got it right  City Loop, Melbourne


http://upload.wikimedia.org/wikipedia/commons/thumb/0/0d/City-loop-exit-spencer-street.jpg/800px-City-loop-exit-spencer-street.jpg

Comeng train exiting the 'Caufield Loop' tunnel at Southern Cross

Note the length of transition.

Similarly ..


http://upload.wikimedia.org/wikipedia/commons/thumb/3/30/City_Loop_exit_Jolimont_Yard.jpg/220px-City_Loop_exit_Jolimont_Yard.jpg

City Loop exit towards Richmond, in the former Jolimont Yards.

[colinw]

Well maybe the Merivale Bridge isn't at capacity in 2016, maybe it is actually at capacity right now?  

In practical terms it is, at least for that one hour of the day.  I haven't (yet) looked at the afternoon peak to see if the same situation exists outbound.

I don't see how any more peak trains can fit in there to be honest. However, I did read something in the 2031 plan or
supporting documents that a "cheap" signal upgrade could increase trains/hour to 24- so an extra 4 trains.
A more "expensive" upgrade would increase that to 30- so an extra 10 trains through there.  Any comments about this?

A signalling upgrade would certainly help, however if we found ourselves running at 2.5 or 2 minute headways across Merivale for extended periods it would not be robust due to the nature of the services feeding into the bottleneck.  One delayed train or other problem and the knock on delays would be horrendous.

The major benefit of signalling for 30 tph / 2 minute headways is where you have a "metro like" railway with but a single stopping pattern and utterly consistent services.  That does not apply at Merivale, where you're trying to jam three lines (or four with Browns Plains & Flagstone) into one bridge, and hit tight headways with trains that have started their journeys anything up to 90km away.

But what if CRR doesn't go ahead?

UNTHINKABLE!!!

Agreed, but in this political climate, let's just say - I have fears. I can't really see Oakeshott or Windsor grasping the significance of CRR and how critical it is (just look at the City vs. Country arguments in the feedback to yesterday's Courier-Mail article on the subject).  To their minds, it is irrelevant to their state, their constituencies and their rural interests whether Brisbane gets a single dollar of Federal funding for railways.  The Greens, on the other hand ...

[colinw]

Clearly more trains via Tennyson is an option.  This will buy a few more services.  The more I think it about, the more I am leaning to a double stack twin tunnel  4 track.  You are only going to get one crack at this.  

Melbourne got it right  City Loop, Melbourne


http://upload.wikimedia.org/wikipedia/commons/thumb/0/0d/City-loop-exit-spencer-street.jpg/800px-City-loop-exit-spencer-street.jpg

Comeng train exiting the 'Caufield Loop' tunnel at Southern Cross

Note the length of transition.

Who says the CRR portals can't be towards Dutton Park from Fairfield, with Fairfield station staying where it is?  With electric trains we should be able to use 1 in 40 or steeper gradients freely.

[ozbob]

Indeed,  seems to be a bit of over-engineering going on ....

I suppose the looming predicament (some might say it has already arrived ...) is rather a sad indictment on the lack of forward planning by a number of governments over the last 30 years.  The present mob is at least trying to do something, something for which I am grateful.

[#Metro]

Clearly more trains via Tennyson is an option.  This will buy a few more services.  The more I think it about, the more I am leaning to a double stack twin tunnel  4 track.  You are only going to get one crack at this.

For the CRR tunnel? I would also agree. One crack, 4 track! (Or three tracks with one being reversible if 4 can't be done).
1st new station in Brisbane CBD in over 100 years, so once it is done it is done.

I'd be disappointed if it were one track in, one track out for $8 billion. IIRC, Melbourne's loop is 4 tracks.
If we assume trains every 5 minutes from both Cleveland and Browns Plains going over Merivale Bridge, post-CRR, this would be 24 trains/hour in peak?

[ozbob]

Yes, CRR, 4 track future proofs it to some extent, also opens up other possibilities down the track, eg. Gabba a junction for a line over to the west etc.

[somebody]

Who says the CRR portals can't be towards Dutton Park from Fairfield, with Fairfield station staying where it is?  With electric trains we should be able to use 1 in 40 or steeper gradients freely.

Hear here.  The ICRCS had a 1 in 50 limit gradient!

We sure seem keen to give up and say "too hard, can't fit any more in" at levels of service well below what other cities manage, then cite $billions of extra infrastructure as required to even get past pathetic half hourly train frequencies.

And a classic example is the suggestion that we can't get a 15 minute frequency at Shorncliffe without duplication.  Cronulla had better than a 20 minute peak frequency with a single dual length platform at Cronulla, single platform at Woolooware and single track all the way to the next station.

The stumbling block definitely seems to be the Merivale bridge, specifically about 1 hour in peak where the bridge is running at 20 tph.  Merivale also comes into play in a big way if you desire to go to quarter hourly offpeak services on the South Brisbane group of lines.

Inbound the worst case currently is the hour from 8 AM where there are departures from South Brisbane towards Roma Street at roughly 3 minute intervals for the entire hour: 8.02, 8.04, 8.06, 8.09, 8.12, 8.15, 8.18, 8.21, 8.24, 8.28, 8.32, 8.34, 8.37, 8.40, 8.43, 8.47, 8.47, 8.49, 8.53, 8.58.  Clearly no room to put a PEAK Browns Plains service through South Brisbane, particularly in the hour from 8 AM, and in fact little or no room to even add any more Cleveland/Beenleigh/Gold Coast trains.

The bifurcation at South Bank-Central makes this capacity limitation better than 20tph though.

Once you get to quarter hourly offpeak on each of Browns Plains, Beenleigh/Kuraby, Gold Coast and Cleveland/Manly, you're at full time 20 tph across Merivale, which clearly is not going to be robust.  If Merivale remains the sole southside route in use, then 4tph offpeak is out of the question, although maybe 20 minutes / 3tph is achievable?

What then to do about Browns Plains?

I have trouble seeing Browns Plains opening before CRR1.  Kippa-Ring would be something which is completely doable without it.

Yes, CRR, 4 track future proofs it to some extent, also opens up other possibilities down the track, eg. Gabba a junction for a line over to the west etc.

Are there much cost savings to having a 4 track tunnel?  I'm pretty sure the ICRCS considered this possibility, but rejected it.

[ozbob]

Are there much cost savings to having a 4 track tunnel?  I'm pretty sure the ICRCS considered this possibility, but rejected it.

Yes I know .. but as history shows it would be good to do it now. Clearly it would cost more but would build in future capacity and redundancy.   Brisbane will pass Melbourne in population down the track.  By having the southern portal in a sensible location it will be cheaper to build.

[somebody]

Are there much cost savings to having a 4 track tunnel?  I'm pretty sure the ICRCS considered this possibility, but rejected it.

Yes I know .. but as history shows it would be good to do it now. Clearly it would cost more but would build in future capacity and redundancy.   Brisbane will pass Melbourne in population down the track.  By having the southern portal in a sensible location it will be cheaper to build.

How do you mean?

[ozbob]

Posts from this point moved to the CRR rail discussion thread as it was really just on that.

--> http://railbotforum.org/mbs/index.php?topic=2034.msg33613#msg33613