The Overland - How can we fix it?
The Overland is a long-distance passenger train which has run between Melbourne and Adelaide in various forms since 1887. This once highly prestigious train has been seriously neglected over the last few decades, and today is undeniably a shadow of its former self. In the 70s and 80s, The Overland was a night train, operating in both directions between Adelaide and Melbourne every night, with each train typically consisting of at least 10 cars, including a Club Car and a couple of vans. Today the train runs in daylight, and makes just two return trips per week – a frequency so low it's almost unusable – and the usual consist is now typically around 6 cars.
There has been a lot of very passionate debate over this train in recent years and it tends to polarise railway enthusiasts. The topic of marginal country passenger services is always controversial - I'm looking at you Mildura with arguments about cost vs social benefit kicking off just about every time the topic comes up. Plenty of people have written and talked about possible solutions for fixing the Overland, including here on YouTube – but I think I have a slightly different take to most.
This video isn't going to be about massively expensive ideal-world scenarios like high speed rail - instead I'm going to focus on some relatively inexpensive and simple changes which could be implemented in a reasonable timeframe, that I think could help rescue the Overland from what is currently a very uncertain future. I'm also going to address what I think are some common misconceptions about this train, and long distance trains in general. I'm going to break this down broadly into four chapters, these being: -The frequency of the train -The slow running time and indirect route -The private operation of the train and resulting identity crisis -And the rollingstock used to provide the service Before I go any further I want to make one thing very clear – I'm going to bring up all the Overland's failings in order so suggest solutions for them, but I still think it's a great train to travel on. It's a wonderful experience as a passenger, and the staff all seem very professional and passionate about what they do, so if you have a chance to travel on it, I highly recommend it. So here we go with Chapter one – Frequency! Over the last 20 odd years the frequency of the train has been incrementally reduced, arguably because of reduced demand.
The initial decline in ridership was probably inevitable as domestic air travel became more affordable however long distance trains still have a role to play and offer many advantages over air travel. The Melbourne to Sydney XPT is a great example of a similar distance train still attracting plenty of passengers in the modern world despite competition from the air, and I see no reason why the Overland shouldn't also be able to fill a similar niche role. I'm going to make a few comparisons to the XPT service, so for reference: the XPT runs on a 12 hour frequency – so that's a day train and a night train in both directions, every day. Yes.. I know Sydney's a lot bigger than Adelaide, but it's still a good comparison by most other measures. And Melbourne is growing very, very rapidly, something that will increase demand on both interstate routes in years to come.
On an established passenger route, frequency and ridership have a bit of a chicken/egg relationship, and it seems the 'powers that be' keep smashing the Overland's eggs, then wondering where all the chickens have gone. The highest priority for fixing the Overland, has to be increasing the frequency. The current timetable of two return trips per week is near useless from a passenger point of view and it's worth remembering that most people won't catch the train in either direction if they can't catch it in BOTH directions, so if the days of operation don't line up with your travel plans, the train is immediately off the table.
I would suggest returning the Overland to a 24 hour frequency, running in both directions either every day, or every night. This huge increase in service would only require one additional train to operate, and would create the kind of timetable that would actually attract patronage. Now the day or night question is a really interesting one. There are lots of advantages to running a night train, and some of them might not be immediately obvious if you've never caught one before. It's important to think of a night train not only as a form of transport, but also as a form of accommodation. When executed properly, it should be like going to bed in one city, then conveniently waking up the next morning in another.
If you were going to have to stay the night somewhere anyway, this effectively eliminates travel time from the equation. There's another reason why timekeeping becomes less important when running through the night, and that's because the journey actually becomes less convenient if it isn't long enough for a full nights sleep. Ideally, you want to leave your departure city some time after dinner, have a few drinks in the onboard bar, head off to bed, then conveniently wake up as you near your destination around breakfast time.
Getting there any earlier would be just annoying, so there's really no need for the train to rush. A good overseas example is the Caledonian Sleeper in the UK, which runs overnight from London to various points in Scotland and vice versa. It runs to a leisurely timetable through the night, taking about an hour and half longer between London and Edinburgh than your average day train. Where possible, the train actually remains in the platform for some time after arrival at the final destinations, allowing passengers more time to wake up and vacate their compartments. For all these reasons I think a return to overnight operation is really worth considering for the Overland, however it does make things less convenient for anyone travelling between the intermediate stations – for instance a trip from Stawell to Horsham would only be possible in the middle of the night.
To me that's just a good argument for supplementing the night train with a day service which is more focussed on serving the intermediate destinations. At the very least this could take the form of a train serving the relatively well populated western Victoria part of the route, say between Ararat and Horsham, but you could also make an argument for doing it over the entire line. So, Chapter two – Private operation The Overland was privatised in 1997, and that's really where a lot of the problems began.
The contract was snapped up by Great Southern Railway, who also took over operation of the iconic Ghan and Indian Pacific trains. The Overland was an awkward fit into this group, being at the time very much a public transport service rather than a tourist operation. GSR's expertise was in tourism, not transport and they really wanted to operate all three trains more like cruise ships – ie travelling luxury hotels. This worked fairly well for the Indian Pacific and the Ghan, which both have high tourist value, but has never really worked for the Overland. While the trip between Melbourne and Adelaide is perfectly pleasant, it just doesn't have the spectacular and diverse scenery – or the epic scale - of the other two routes.
In 2019, GSR was renamed Journey Beyond Rail Expeditions, awkwardly emphasising that their interest lies predominantly BEYOND the rail experience. I honestly think the attempt to appeal to tourists above all else has seriously harmed the Overland's reputation as a serious passenger service. For instance, the Journey Beyond website is all flashy imagery and inspirational language, but practical things like fares and timetables seem almost an afterthought.
On the main page for The Overland – sorry, the Overland EXPERIENCE - the most prominent feature is this rather grandiose quote: “A day trip in a league of its own. Ride the rails of romance on your next interstate adventure with this day-long train voyage aboard The Overland." They also weirdly seem quite proud about the very indirect journey, frequently referencing the 828km length of the route – which I will elaborate on shortly. In fact the cafe car is named Cafe 828.
I genuinely wonder how many people might consider using the Overland as transport, but be quickly put off by what looks very much like an operation aimed at the high end tourist market. The reality is that once you get onboard, it's barely touristy at all, and feels very much like any other long distance train – the only exception being the unusually large number of staff, and the occasional tourist-esque announcements they make. It seems clear to me that Journey Beyond just aren't the right operator for the Overland if it's going to return to a serious transport role in the future.
The obvious solution here is to hand the operation over to Victoria's government owned regional transport operator, V/line. They know how to run a proper service, are experienced at running long-ish distances and providing clear information, and in fact.. they already run a bus service to Adelaide, which must be.. a real adventure. If we take a look at the XPT again, there is really no attempt to advertise the Melbourne to Sydney route for tourism, it's shown as just a solid public transport service, and it works like one. Remember that effective public transport is useful for tourists anyway, without being specifically aimed at them.
Chapter three – Why does it take so long, and.. how much does that actually matter? The Overland is timetabled about ten and a half hours to travel the 828 rail kilometres between Melbourne and Adelaide. Now the first problem is immediately obvious - the road distance is significantly shorter, at about 725km.
So why is the railway 100km longer than the road? Well both Victoria and South Australia built the majority of their state railway networks to 1600mm (5'3”) Broad Gauge, so of course the interstate line connecting the two capitals was built to that gauge. The Overland originally ran on Broad Gauge, taking the most direct route out of Melbourne, via Bacchus Marsh and Ballarat to Ararat. In 1995, the line was converted to Standard Gauge, however rather than converting the direct but hilly line between Ararat and Melbourne, a different route was chosen, with the Standard Gauge diverging to the south from Ararat, then running through Maroona, Cressy and Gheringhap to North Geelong, then running parallel to the Broad Gauge Geelong line as far as Newport, then taking an excruciatingly slow and round-about route along the goods lines into the city.
The flatter grades of this route were a benefit for goods trains, but the longer distance makes for a terrible passenger route. Meanwhile at the other end of the trip, the line takes a slow winding route through the Adelaide hills between Murray Bridge and Adelaide, while the freeway takes a very steep but direct route straight into the city. There really isn't any easy fix for the Adelaide Hills problem because it's dictated by geography, but for the Victorian section, there is actually a really simple an inexpensive workaround. When the main western line was gauge converted in 1995, the line between Ararat and Ballarat was abandoned, while the Ballarat to Melbourne section remained an important broad gauge main line. Then in 2004, the line to Ararat was reopened, being served by several daily passenger trains from Melbourne.
At Ararat, broad gauge trains use a dock platform at the Up end of the station, while the standard gauge line runs through the main platform. This provides a perfect cross-platform connection, however.. it seems nobody thought about this when timetabling either service. If the Overland made a well timed connection to a broad gauge train at Ararat, then you've immediately shaved around an hour off the journey time – without any new infrastructure. If the broad gauge train could be given a nice express pattern, even better. The standard gauge train could still continue to Geelong as quite a few passengers do actually get on there, and by the time you get there you might as well run into Melbourne and use the existing servicing facilities there, though it might be hard to make an argument for carrying passengers on that section if you're also encouraging them to take the direct route.
At the moment you can make the change at Ararat if you feel like it, but the timetables don't line up at all, so any time saved is absorbed into the wait time. While making people change trains is hardly an ideal option, it is a very simple way to make a significant time saving, and over such a long journey, one quick change is hardly a major inconvenience. Something else that has been talked about a lot recently, is the possible use of gauge convertible trains, a technology which is already in use in a handful of places in Europe.
Basically this would involve having a gauge changer at Ararat, allowing a single train to complete the direct journey. A lot of people are viewing this as a silver-bullet solution, however I can't say I'm entirely sold on it. For one thing, it's a proprietary technology produced by essentially only two companies – Talgo and CAF - meaning all future rollingstock purchases would have to come from those same companies. This can be problematic if a company stops supporting the technology, and also prevents a competitive tender process taking place. Introducing gauge convertible trains could also reduce the incentive for future standardisation, which would be more beneficial long term.
That said, I think it is worth having a discussion about this option, and it will be interesting to see what happens with this in the future. The other factor in running time is of course, speed - and I'm going to cover that in a moment when I talk about rollingstock. Speaking about timing more generally though: We all know the train isn't the quickest possible way to travel between Melbourne and Adelaide. Flying is obviously the fastest way, however that's the case over most long distance routes world wide – so that's hardly a local factor. I've noticed a lot of people are very quick to loudly comment on how they can drive to Adelaide quicker than catching the train – but while technically possible, that's actually a very misleading statement.
Right now Google maps tells me it will take about eight and a half hours to drive between the Melbourne and Adelaide CBDs. Yes, that's two hours quicker than the train.. but that's without stopping! If I add in a couple of reasonable rest breaks – as I should being a responsible driver - then I'm already up there with the train's running time.
Then there's another factor – not everyone actually wants to sit in a car for eight and a half hours, especially if you're the only driver. Speaking for myself personally here - while I've certainly clocked up those sorts of miles in a day once or twice – it's not really a good experience. I've driven to Adelaide a few times in the past, and always divided it into two days, and the large number of motels around Nhill and Bordertown tells me I'm not the only one! I suspect that the kind of people who are willing to do the drive in one day probably aren't a key part of the target market for rail travel anyway. So while improvements to the running time would obviously be great, I would argue the current timing is already quite competitive with road transport. The Melbourne to Sydney XPT also takes about 2 hours longer than a non-stop drive – and again, plenty of people use it. Before we move on, there's one more thing about timing – while I said the Overland takes about 10 and a half hours, there's currently one very odd exception.
The Thursday eastbound Overland mysteriously takes 50 minutes longer than the Sunday one. Looking at the ARTC Master Train Plan I can't see any good excuse for this, but it appears to result from poorly timed crosses with a handful of westbound freight trains. If this is indeed the only reason, then it could be solved by making some minor changes to the timetable, and giving the Overland priority – like you would expect for any other passenger train. I actually asked Journey Beyond if they could shed any light on the 50 minute discrepancy, and they replied “We might have some freight train occupying our train line on Thursdays, which causes a traffic and consequently a delay in our schedule for the Overland.”
So ah.. thanks for that. And so onto Chapter 4 – Rollingstock Currently the train is comprised of a single set of passenger cars built in the 1950s and 1960s, hauled by an NR class locomotive hired from freight operator Pacific National. When talking about the current service, one of the first things that usually comes up is the age of the existing rollingstock. It's true that some of the current passenger cars are approaching 70 years old and there's a good argument for a replacement fleet, however I don't see that as a significant factor in the train's decline. The cars have been refurbished several times over the years, are very nice to travel on, and certainly don't feel old from a passenger's perspective. It's important to remember that unpowered passenger carriages are mechanically very simple, and there are very few ways they can fail in such a way as to prevent the train from actually opertating.
The NR class locomotives were built in the mid 90s, and are still the main workhorses of the country's interstate freight network – so attributing any reliability issues to the train's age is really not accurate. Despite that, there are certainly benefits of a replacement fleet going forward, and it's obviously an essential part of increasing frequency, so.. what should the replacement be? By far the most common suggestion in this debate is that the train should be replaced by VLocity DMUs, as currently used extensively by V/line on the Victorian regional network.
Politicians and the public alike love VLocities, probably because they look fast and are built locally. However I would question which particular aspect of the VLocity makes it jump out as a candidate for the Overland. I suspect most people are going to say it's because they're fast, so let's talk about speed for a moment.
The VLocity fleet has a top operating speed of 160km/h, compared with 115km/h for most Victorian passenger trains including the current Overland. By world standards, 160km/h isn't really all that fast – in fact the British were doing it a century ago in the steam era. Any brand new long distance train ordered in the 21st century will be capable of 160, so it's not something specific to the VLocity. But getting a train that can run at that speed is the easy part – the bigger challenge is raising the maximum permitted speed of the actual line.
Quality of track needs to be up to scratch (in many areas it isn't, especially in Victoria), and changes need to be made to signalling and level crossing circuits to allow the line speed to be increased. In fact probably the biggest expense involved is upgrading level crossing protection: To allow a line speed of 160km/h, ALL level crossings have to be fitted with a minimum of lights and bells, whereas at the moment there are very frequent passive level crossings along the entire route protected only by signs. As a current example of this, while the XPT is rated for 160km/h, it only runs at 130km/h in Victoria – because that's the maximum allowed line speed. So increasing the top speed of the train to 160 is certainly achievable and should absolutely be done, but it would require major infrastructure spending, and is not as simple as just buying a pointy train. Ok so back to the VLocity..
what exactly are they, and what are they good for? Well they're a good medium distance train, well suited to Victoria's busy regional commuter routes for which they were designed. They have no first class, seats that are fine for two hours but not ten, not a lot of luggage space, and.. mostly, no catering facilities. At the time I'm writing this, V/line has just introduced a new long distance version of the VLocity which includes a buffet for the run to Albury, however I'm still not convinced they're the best choice for long distance travel – and Adelaide is more than double the distance of Albury anyway.
All the existing examples run in fixed three car units, often run in pairs as a six car set. There is no way for passengers or staff to move between the two units, and that's pretty inconvenient on routes with long gaps between stations. VLocities are a Diesel Multiple Unit made up of cars which all carry passengers, and are all powered.
This means that no matter where you sit, you're never far away from a running diesel engine under the floor. For me, the sound and vibration makes for a fairly uncomfortable experience when travelling on the existing longer routes such as Ararat and Echuca. Compare this to the XPT or QR's Diesel Tilt Train, which have dedicated power cars at either end, and unpowered passenger cars in the middle. To me this is a much better model for long distance operation, giving better passenger comfort.
It can also provide better fleet availability, as power cars can be easily interchanged for maintenance without having to take an actual passenger car out of service. I think something 'similar' to the Diesel Tilt Train could actually be a good fit for the Overland route. It is a good example of a fairly recently built, very small order train specifically for long distance use – two sets were built at EDI Maryborough in 2003, with one additional set being ordered in 2010. To be clear I'm just using this as an rough example, as it's not quite right for 'down south' the Tilt Trains are designed for Queensland's narrow gauge system, and the actual tilting mechanism wouldn't be particularly useful down here, except perhaps for saving a few minutes through the Adelaide Hills. Another good example of a high quality small order long distance train, are Kiwirail's AK class carriages.
They were built in 2011 and specifically designed for use on New Zealand's three long-distance passenger trains, including the Northern Explorer between Auckland and Wellington, which has a similar 'part tourism part transport' character to the Overland. Just 17 cars were built, and they're extremely well designed for the job. But at the end of the day, picking an existing type of train now is kind of the wrong way to look at it the best approach with acquiring a new fleet is to come up with specifications, put out a tender, and see what the different manufacturers come up with as a solution. If the best thing on offer turns out to be a VLocity, then by all means go with that – but I think we can do better. Now we're almost at the end, but there's one more thing to cover, and I'm going to call it the 'Keswick Conundrum.' Now I could have talked about this in the timekeeping chapter, or even the privatisation chapter, but this is a problem that kind of spans multiple issues, and is really emblematic of what has gone wrong with the Overland.
So Adelaide railway station is located on the north side of North Terrace, right next to the busiest part of Adelaide's CBD, and well positioned to make connections with local busses, trams and suburban trains. The Overland.. doesn't go there any more. In the mid 1980s, all long distance and country passenger trains were booted out of Adelaide station, instead departing from a large purpose built station to the west of the city, known as Keswick Terminal. In 2008 it was renamed Adelaide Parklands Terminal, but for the purposes of this video I'm just going to call it Keswick – as it's shorter and less pretentious. Now the problem with Keswick is obvious – it's over 2km as the crow flies from central Adelaide, which is where most passengers actually want to go. So, how do you cover the last part of the journey into town? Well if you glance quickly at the map, you might be forgiven for thinking there's an easy solution – Keswick is right next to a busy suburban railway corridor, carrying the Belair, Seaford and Flinders lines...
however, there isn't a station there. Instead you will need to walk about 450m to the nearest bus stop, or 750m to Adelaide Showground station. Neither of those options offer a particularly high frequency to the city, with services as much as 30 minutes apart. You could also do what I did last time, and just walk the entire 3km or so into town – which for a healthy adult will only take about 30 minutes anyway – potentially quicker than waiting for the bus.
My guess it that most people complete the journey by taxi or ride share, which is pretty pathetic for an interstate passenger service. The whole area is also really poorly designed for pedestrians – if you jump in on streetview, you will see all the footpaths in the area are narrow, poorly maintained and right next to the road. It's clear that the station was designed with private road transport in mind, and basically nothing else. There are several possible solutions for the Keswick problem, but the best one is: Get the train back into Adelaide station. This could be expensive, but would provide by far the best result.
All trackage currently leading into to the station is Broad Gauge, so you would need to find a way of providing Standard Gauge access. This actually wouldn't be all that hard – you could make a branch from the existing standard gauge alignment here, join the Up Belair and Down Gawler lines here, then run on dual gauge trackage into one of the central platforms of Adelaide station. If using the existing rollingstock, a second loco would probably need to come in to haul the train back out, or you could do what they do in numerous places around the world and have a shunter drive the train from the rear car using a radio control unit. If you really didn't want to interact with the suburban lines at all, you could jump the northern lines on a flyover, and build a new dedicated platform over here somewhere. A potentially simpler, but also significantly worse option, would be to reinstate some basic platforms on the suburban lines adjacent to Keswick. Connect them up with a footbridge, and then timetable suburban trains to stop there before the Overland's departure and after its arrival.
One advantage of this option is that it could also provide a connection with the Indian Pacific and the Ghan, which wouldn't easily work with the Adelaide station option because of those trains' significant length. The last option I can think of is extremely easy, and could be implemented immediately with no changes to infrastructure. And that is, just meet the train with a bunch of busses, and provide a timely – and ideally free – connection straight into the city.
Seriously, why don't they do that already? Whatever happens in the future, the Overland really just needs one big thing, and that is government interest and support. With that, any of these fixes I have suggested would be possible, and we could get back to having a proper passenger service between the Victorian and South Australian capitals. If you've made it this far into the video, congratulations – I know it's been a long one.
Like I said at the start, if you get a chance to travel on the Overland, do it. If you want to learn more about the train's history, go get yourself a copy of John Wilson's excellent book, 'The Overland: A Social History.' And if you want real improvements for this service, write to your MP. Thanks for watching!
2022-06-10 12:11
