Prospects for Proper Light Rail in Melbourne

I had actually started this post with the intention of discussing tram priority measures in Melbourne’s narrower streets. I was thinking about things like full time clearways, proper traffic light priority, blanket right turn bans etc., and whilst some of these might be sensible policies (especially traffic light priority and some right turn bans), I’d been approaching it from exactly the wrong way. Rather than trying to make street track as fast as reserve track, perhaps we should first aim to get the reserve track working like proper light rail. Under such a plan, we can – for want of a better term – get the low hanging fruit on the tram network and provide the biggest speed and capacity improvements at the lowest cost.

It’s predominantly the Dandenong Rd. tracks that have piqued my interest, and the idea of upgrading them has been covered extensively by Riccardo in the past. I didn’t initially realise the potential of the area, but with a bit of work, it could function as serious light rail. To do so, new high capacity vehicles would be required; as would platform stops at less frequent intervals, near instant light priority, basic signalling to prevent bunching and a high service standard (frequency of every 4 minutes of better, service at a high speed). The aerial photo below shows how substantial the reservation is.

But ultimately, a proper light rail network should be segregated from not only cars, but conventional street trams as well. For example, there’s a very compelling case for high standard light rail for Dandenong Rd, but that same case doesn’t really apply for Wattletree Rd. or Hawthorn Rd. Likewise, St. Kilda Rd. is ripe for a high capacity, segregated light rail upgrade, but High St., Malvern Rd. etc are not. Consequently, the street track sections need to run as shuttles and terminate at the light rail trunks. Yarra Trams have been advocating something along these lines for a while now – and good on them for doing so. But if a one size fits all approach is taken, the network is only as good as its weakest part – and in the case of Melbourne’s trams, that’s very worrying.

Some possible harebrained schemes to split up the tram system…

Following through with this logic dictates that Melbourne’s tram network would really become two distinct systems in terms of the service standard offered. One system would follow the current street tram set-up (although with ThinkTram and associated improvements), focussing predominantly on shorter trips. The other system would be based on high frequency, high capacity, high speed light rail. I’ve drawn up a speculative scheme based on this framework for the St. Kilda Rd. and Dandenong Rd. areas, shown below (note the map I made was basically drawn over the excellent Melbourne tram map on Wikipedia – hats off to the creator, John Shadbolt).

As you can see, I’ve connected Dandenong Rd. to the St. Kilda light rail via Fitzroy St. The Fitzroy St. connection would cost money to upgrade – probably substantial amounts at St. Kilda Junction (especially if flyovers are included), which is not purely in keeping with the ‘upgrade the easiest bits first’ philosophy I set out at the start of this post. However, these costs could well be offset by the benefits of the creation of a high standard light rail line. Also note that basically all the street trackage is running as feeder lines into the light rail/heavy rail trunks.

This arrangement largely mirrors my already stated plans for connecting up Plenty Rd., Nicholson St. and St. Georges Rd. as a single light rail line. For clarity, this is shown again below – perhaps the northern and southern light rail lines could be through-routed.

Speed and capacity of decent light rail

I’d argue there are three main things that separate Melbourne’s current trams from modern light rail – frequency, speed and capacity. The frequency issue has already been discussed – a 4 minute headway (or 15tph) seems about right to me, although St. Kilda Rd. really needs double that. It’s capacity and speed that warrant more thought.

In terms of capacity, the choice of vehicle is very important. Obviously running Z3s at a 4 minute headway isn’t going to to much for capacity – you need proper light rail vehicles. I don’t know enough to recommend a specific vehicle – just buy something off the shelf. However, Budapest’s version of the Combino Plus did attract my attention. At 54m long, it is the biggest passenger tram in the world, with a passenger capacity of 352. Run at a 4 minute headway, they could move 5280 people per direction per hour. For comparison, that’s about what Paul Mees claims the Dandenong line is handling at peak hour now.

Increasing speed is the trickiest part of the plan. It’s obviously easiest to do in reserve track areas, but even then, instant light priority is required to keep services moving smoothly. This could be likened to level crossings on the rail system – trains get absolute priority. Whilst this may not be easy to do in St. Kilda Rd. (given the sheer number of trams and large number of intersecting streets – some of which themselves have trams), it should be no substantial problem on Dandenong Rd. Fewer stops and the use of vehicles with lots of doors would be helpful too.

But even if you can get average speeds along Dandenong Rd. and the St. Kilda light rail up high enough, it’s all going to come crashing down at Whiteman St. Currently in peak hour, travel time on the 96 between Spring St. and Whiteman St. (some 2.8km) is timetabled at 20 minutes. The MMTB had a plan back in the 50s to underground the city tram lines – and let’s be frank, it was insane. Undergrounding traditional tram lines is crazy, it’s pricey and you lose many of the short trips that make trams so useful. But for light rail/premetro, the equation is perhaps somewhat different. I honestly don’t know whether it would work or not – and I’m prepared to be convinced either way.

Advertisements

18 Responses

  1. Dandenong Road works reasonably well as it is. The peak service is inadequate, but off peak, weekdays day is ok, with a six minute service. Most of the peak service problems are uneven headways due to outside of Dandenong Road influences.

    The traffic lights have had quick ‘t’ phases for a long time, and these have recently being upgraded to work almost every time. In the past, they never worked counter peak flow. They certainly could be better by being instantly responsive and repeat if the tram does not move.

    The non super stop west bound before Chapel Street is a nuisance stop, but since it was put there for school children and there is an overhead walking bridge, it won’t be removed in a hurry.

    The running times are not excessive, often quite tight.

    The improvements to where cars cross, with tram activated warning signs have helped, but not fully solved the problem of cars delaying trams. A bit more could be done there.

    The biggest problem for trams is parked cars, reverse parking into spaces and ‘u’ turns. Clearways on major roads being extended, especially the 3pm to 7pm will help a lot. Clearways in Glenferrie Road, Burke Road and Balaclava Rd/Carlisle Street would help a lot too.

    So much could be done with what is there already. Thought provoking post though.

  2. One way of getting Trams/Light rail vehicles from Dandenong Rd (and Queens Way as the continuation of Dandenong road is called between Chappel St and the junction) would be along Queens Rd to connect with the Kings Way tram line. Grade seperation could even be achieved if deired. If desired High St could be connected via Lorne St.

    Dandenong Rd-Fitsroy St direct has not been possible since they took the trams out of Wellington St when they did the junction and got rid of the direct link (I read somewhere that the original plans for the junction had the trams staying in Wellington St).

  3. I’ll back your plan Phin.

    It needs to be understood that we are not talking about an upgraded tram service, but essentially a pre-metro.

    And the reason we are talking about it is because the Caulfield-SY rail corridor is “FULL” according to Rod Eddington and the tramway corridor is not lifting its share.

    I have to disagree, respectfully with Andrew, as I said it is not about whether the tramway service is satisfactory, but whether a “pre-metro” service is there, which it isn’t.

    Thinking of tram and rail as different in this city arises because of gauge but also because of the historic rivalry between the MMTB and VR which many rail and PT fans have taken as a given, rather than challenging it.

    And there is nothing wrong with full-blown level crossings on the new system – Glenelg is a working example of them, as is the existing St K and Port Melbourne lines.

    As for synchronising the lights at Orrong, Kooyong etc – well as I see it Phin’s system DOES allow you to set a timetable, something effectively missing from the current tramway system where you could not possibly expect a tram to adhere to timings with road traffic intereference.

    If the lights were set to a 2 minute frequency and lo and behold, the light rail vehicle appeared on the 2,4,6 or 8 minute headway, loaded passengers during the light phase and was ready to go when it ended, I don’t see a problem. You could put boom gates in as well, more as a visual cue than an actual means of stopping the traffic.

    I also expect a 5 metre trench could be put under each of these roads, just enough for a LRV with its pantos squeezed down, with approach ramps no worse than the current Dandenong Rd to St Kilda Rd ramp, but even using more respectable grades, 75 metres either side of the road would give you 1:15, perfectly acceptable for a high-torque fully motored LRV.

    You could even make the grade sep area the ‘station’ (but no need to do that, the existing platform styles would be OK and the tram could use momentum to traverse the dips.

  4. Have a read of the specs for Citadis Dualis on the Alsthom website – a Stadtbahn-able vehicle that would certainly be low enough in height for the grade seps I am talking about

    http://www.transport.alstom.com/home/elibrary/technical/products/32031.EN.php?languageId=EN&dir=/home/elibrary/technical/products/&idRubriqueCourante=13957

    I would be going for a very large number of modules, so that a single driver, with cameras and a single long corridor, can have visual right down the tram. The vehicle would then be able to take 300 or more people at a time, run from Malvern Station area (where people are currently parking their cars on the side) and relieve pressure on the Caulfield line, as well as being an effective trunk route for the Hawthron Rd, Glenferrie Rd, Wattletree Rd and even the Caulfield Station tram routes, which would shuttle of it using much smaller vehicles

  5. Like Tom said, I would probably link the Dandenong Rd section to Kings Way, but probably differently.

    The Albert Park section of Kings Way is not going to be an effective tram route, because of the heavy car traffic and the 5 lanes. I would run the Dandy Rd trams up St Kilda Rd till Domain, then route them via 55 if you can properly LR Park street.

    If there is a way to go through the CBD with out any major traffic crossings (probably easier than it sounds because the 55 doesn’t share tracks with other trams) and connect with a decent city station (ie. not Flagstaff because it isn’t open on weekends), find a way to get the other trams off Flemo Rd, run it through Royal Park, along the old rail line (Upfield extinct until Jewell) and terminate at Jewell, where the Airport line comes out of the tunnel.

    Not tricky at all!
    Calembeena

  6. That gives you a total of 4 LR routes – not bad considering that there were (in reality) none before
    Calembeena

  7. What happens at the city end (beyond the Whiteman Street bit). As you suggested, Whiteman Street is a problem and the city end could well make your improvements essentially useless if it causes congestion issues.

    I wonder about the cost and need compared to a proper metro. If the metro is built, you do not need anything more than a surface tram (not premetro) for surface movements along the St Kila Rd corridor and probably even along Dandenong Road, depending on route.

    What is the cost of this proposal and its life cycle> How long would investing money in this option + the opportunity cost of making pre-metro infrastructure redundant set back construction of a metro? You are looking at a 30-50 year life cycle for this sort of infrastructure so it is a big decision.

    I should make it clear that I am not against premetro and other non-standard applications of rail-based PT…BUT…this is one of the few areas in Melbourne where you could realistically expect densities sufficient for a full-blown metro.

    The premetro might be better suited to, say, Doncaster from Victoria Park out to Shoppingtown or The Pines. This corridor would be unlikely to get the density (thanks Eastern Freeway!) to support a metro and, well, conventional heavy rail has its limits in terms of bang for buck.

  8. Interesting comment Dr Waddles

    I’ll give it some thought.

    I sometimes wonder if we are in ‘endgame’ and whether you spend $4b or $8b or whatever it doesn’t seem to matter any more.

    While the science (or the economics I should say) of transport policy is supposedly better than it has ever been, the newpaper headlines are shrieking louder than they ever have on the ‘demand’ side, and politicians on the ‘supply’ side are coming up with more and more hysterical proposals in response.

    As in the Sydney case, separating the sense from the dross among these very large numbers is a huge task.

    So maybe it doesn’t matter if Sir Rod asks for $7b, he could have asked for $3b or $15b. How long is a piece of string?

    Sorry, not trying to sound cynical, far from it, just failing to see any science behind transport costings of late. Hence Phin’s blog posts have been a useful reality check.

    In terms of a convention metro to Caulfield, bring it on, but only if there is no interaction with the legacy system – I see Sir Rod’s proposal as doing nothing to improve reliability on the Dandenong corridor where it needs to improve most, and all these pax will be waiting at Domain Rd underground for the usual overdue and overcrowded Comengs and Siemens that we have now, because some nong has rammed his Landcruiser into a train at Grange Rd.

  9. Many thanks for the comments all.

    Although we’re looking at decent offpeak daytime tram frequencies (6 min) on Dandenong Rd already, as Riccardo mentions, this plan is more about building a premetro to set it apart from the tram system in terms of speed and capacity. With plenty of almost rail quality reservation, it seems a shame not to use it as such.

    Ricc, the Citadis Dualis looks interesting – can hold up to 292 passengers on the longest (52m) version and could jump onto the Dandenong line (gauge issues aside of course). Ultimately, there’s plenty of decent off the shelf stuff from Europe these days, and we should just go with what’s best.

    The metro vs premetro cost problem is certainly a legitimate one. Whilst we could reasonably expect a premetro tunnel to cost less (steeper grades, more basic signalling, smaller stations etc.), it’s still pricey to do tunnelling and if you get into it you want the maximum benefit. Personally I’d much prefer to see a real metro. However, if Sir Rod’s tunnel goes in as planned, the metro possibilities for the corridor are substantially hampered, and perhaps converting tram lines to premetro standard is the next best thing…

  10. Oh and I forgot to mention, I’ve been giving some thought to the premetro Idea mentioned by Drwaddles – hopefully I’ll have a post about it up soon.

  11. Excellent idea, Phin, solving the problem of the mass of trams that goes up and down Swanston St. has always been in the back of my mind. Your pre-metro idea has merit.

    However, I’d like to find out your plans for the lines once they hit the CBD. If they continue on as they are, I fear that the 4 minute headways are going to cause mass congestion when the CBD bottleneck hits, and who knows how far down the line that will translate to.

  12. Not a bad idea.

    I would suggest making the St Kilda and Port Melbourne lines Pre-Metros with tunnels under the CBD.

    It would be good to see some integration between heavy and light rail in Melbourne even if dual gauging is necessary.

    As for level crossings isn’t one of the Glenelg level crossings in Adelaide being replaced by an overbridge?

    It may be controversial but I wonder about having light rail to the outer suburbs as extensions to existing tramlines. Could we have a line to Doncaster? built this way?

    Also could light rail be a way to create express tramlines? For instance consider the length and journey time of route 75. Maybe a shorter existing route could run part of the way eg City to Hawthorne with trams running to Burwood or Vermont south running underground say via a tunnel between Hawthorne and the City creating a quicker journey to the outer suburbs? Maybe Hawthorne isn’t the best change point.

    Also is the planned metro in Dublin, Ireland a pre-metro?

  13. Has Yarra Trams been reading this blog? : )

    Herald Sun article.

  14. Although it’s a bit old now, I came across this response to the Herald Sun Story from the Yarra Trams CEO.
    http://www.yarratrams.com.au/desktopdefault.aspx/tabid-39/44_read-1593/

  15. light rail is ok but the st kilda port melbourne area which was heavy rail should have been that way had a proud history of it.make it underground from flinders street and have a station underground for the casino passengers instead of trams going around to bourke street.

  16. Regarding your scheme to split up the tram system into light rail and tram in St Kilda Rd. I know of this having been tried in Europe, it was unpopular and quickly dropped – people didn’t like to change trams, for reasons such as; losing their seat, changing in the rain, have to pick up all the shopping, wake the baby etc, etc.
    Works well on paper but not when real people are involved.
    You need bigger trams, have a look at the double decker ‘Autonomous Tram’.

  17. Glad my map could be of some use :). Great blog, I’ll read it all when I get time.
    John

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: