On the Problem of High Speed Trains

Guest post by Dr Geoff Hudson

Everyone likes high speed trains. The Shinkansen (aka “Trunk Route”) trains in Japan are famous. Smooth, speedy, punctual. The problem is the track costs a fortune – $30,000,000 per kilometre is typical. Mag-Lev (Magnetic Levitation) and Hyperloop (Evacuated tube) systems are even more expensive. The problem for high speed rail is that the rails must be in exactly the right position, and must stay in that position from day to day. This is often referred to as good or perfect track alignment. If the tracks move, the train is bumped and possibly derailed. Laying tracks on sleepers on gravel (aka ballast) on the ground means that freight trains can jiggle the tracks around and changes in the moisture content of the ground below the ballast can move the track. Replacing the sleepers and ballast with immovable concrete is the major part of the cost of high speed rail.

Of course the rails and sleepers are not all you need. Signalling systems to avoid train collisions must be included and can cost around $2,000,000 per kilometre. High speed trains in Europe and Japan have overhead electric power at another $2,000,000 per kilometre, and similar costs can be spent removing level crossings.

There have been dozens of studies into high speed rail in Australia. They have all adopted the view the European and Japanese trains are the only available technologies. All have been rejected mainly because of the high cost of laying the rail. There have even been studies showing that the long rail line lengths and low population in Australia condemn high speed rail here. It is just not economic.

Our politicians say that innovation is what we need to supply employment and economic growth. We badly need high speed trains, but seem to make no attempt to create a technology which escapes the basic rolling stock design of the last century. The major innovation which is relevant to the use of existing rail tracks for high speed transport in the last 100 years is the tilting mechanism created by the British Advanced Passenger Train (aka APT) about 50 years ago and then sold to Italian train developers. A Swedish tilt train was tried by NSW a couple of decades ago, but failed because the train alignment was inadequate.

The only escape from this financial trap that I can see is to develop a new train which can run at high speed on our existing tracks. It might have to slow down for bends and stations, tunnels and passing other trains, but if it could do 250 km/hr, you could commute from many regional towns. Why is no-one studying this? It is a peculiarly Australian need, yet we persist with the basic bogie design of the last 100 years. We still use sloped wheels to allow the outer wheel to cover more ground than the inner wheel on a turn, even though that fails with flange squealing so often. We still bolt the carriage down to the bogie with the all the restraint needed at a station, for the entire journey. It is as if there is a religion that says that only the Europeans and Japanese can develop high speed trains and we must not change the basic suspension of trains. The rails that must remain as they are, not the rolling stock.

I have done some sums and I am convinced that a new vehicle can be designed to travel safely and comfortably at 250 km/hr on many parts of our existing rail network. The longest straight stretch of rail in the world (478 km), on the Nullarbor plains is only one of the appropriate targets. Will anyone help me build this vehicle?

A documented and more detailed technical account can be downloaded here.

24 Replies to “On the Problem of High Speed Trains”

  1. ‘High Speed Trains’ are an obsession with certain politicians who think they will be voted for if they promise to spend billions to reduce travel times on suburban lines by five to ten minutes.

    1. He is talking about high speed trains between major cities, specifically Melbourne/Sydney/Brisbane. Ideas for such an Australian GTV have been around for years and always abandoned because of the cost of the track. Imagine travelling from Spencer Street to Sydney Central in under two hours. Less than the time taken by air when the trip to Tullamarine is taken into account and certainly a lot more pleasant.

  2. I believe in this idea, that trains probably could be made to run much faster on existing tracks here in Australia, starting by dropping burdensome prevailing misconceptions.
    My first job was in the Swedish paper industry, for their Newsprint Research Lab, aimed at reducing newsprint breakage in printing machines, a big problem in those days of poor quality newsprint, unlike today. My argument soon was that it was a printing machine problem, that a printing machine should be able to print well onto poor quality paper without breaking it.
    I then moved onto printing quality, decided that paper elasticity must be a key factor and invented a device for measuring it and proved that there was indeed a strong correlation between its measured paper elasticities and print quality.
    I quit after that, their still not paying me as a professional researcher after 2 years. I went to IBM.

  3. More thought required, not only for speed but to promote trains over trucks and alleviate our roads. This particularly relevant here in Tasmania where our state government don’t give a dam about trains and continue bowing and scraping to the trucking magnates.

    1. There are many challenges. I was limiting myself to just one in the hope of making some progress with it.

      1. This is the key 3 emboldened parts of the article that needs to be addressed :
        “We still use sloped wheels to allow the outer wheel to cover more ground than the inner wheel on a turn, even though that fails with flange squealing so often. [
        ]We still bolt the carriage down to the bogie with the all the restraint needed at a station, for the entire journey. [
        ]It is as if there is a religion . . .”
        It was only the 2nd point that I was directly responding to, with agreement that we should NOT be so doing IMPLICIT in my AI-ed Incat Aluminium ferry Ground Effect aircraft takeoff idea .
        But I was also agreeing with the 1st point, again implicitly, with the idea that AI controlled 90% lift and rail following could be used to replace sloped wheels, whether or not they were ever necessary.
        It could be that lift is not necessary. Just better suspension, lighter carriages and AI could be sufficient.
        Where this thought train is pointing is obviously towards where we are headed with cars, suggesting that :

        All that is needed on our tracks is existing and developing AI-ed driver-less road vehicle technologies.

        That is lighter, smarter, faster, more perceptive trains and carriages.

  4. Hi Geoff,
    If instead of designing the system to provide rapid transit between large population centres, it could be rolled out piecemeal with the objective of creating satellite towns within high speed commuting distance. The developer would be permitted to capitalise on the increased value of real estate to amortise the cost of the project?

  5. I have always thought that the Hyperloop should NOT be Evacuated , that the vehicle should actually fly through normal pressure air using Ground effect (aerodynamics) GTEs and flaps.
    https://en.wikipedia.org › wiki › Ground_effect_(aerodynamics)
    In fixed-wing aircraft, ground effect is the increased lift and decreased aerodynamic drag that an aircraft’s wings generate when they are close to a fixed surface.
    CCTV and other sensors and AI would keep the vehicle away from not much bigger cylindrical sides,
    a MUCH SIMPLER PROBLEM than looking out for cows on freeways.
    This idea is relevant to Geoff’s idea , a rail track being similar to a Hyperloop tube in simplicity.
    I propose a similar use of Ground effect (aerodynamics) GTEs and flaps for Geoff’s train.
    It would thus look more like an aeroplane than my Ground effect Hyperloop vehicle.

    1. Hi Peter. As I understand it the reduced pressure is to reduce air friction. My problem with new tubes for transport is the cost of the tubes.

      1. Yes. Hence my jumping onto your idea, that existing, under-utilized railways be put to better use. No tube costs there !?!

  6. Or this, with elastic links to parallel, wide apart 2 lots of train tracks, achievable railway reserves being so wide, the whole thing being designed around Ground Effect , as explained in my last two posts : https://www.google.com/url?sa=i&source=images&cd=&ved=2ahUKEwiG4ZWx3IPmAhX37HMBHdf2BsQQjRx6BAgBEAQ&url=https%3A%2F%2Fwww.incat.com.au%2Ffast-passage-for-incats-second-112-metre-japanese-catamaran%2F&psig=AOvVaw0Rmx_M_kkXKnazCnhZe4wo&ust=1574714923638027

  7. Note how with takeoff in an aeroplane, the wheels take less load the faster they have to turn because of lift. A lucky complementarity allowing lightweight wheels.
    A Ground Effect train would be similarly lucky. Its AI would have to make sure, of course, that it never takes off, keeps say ~10% of its weight on its wheels.
    I had an exhilarating experience of this sort of thing on a helicopter in 1969, almost certainly driven by a highly respected Vietnam veteran, he was getting away with so much !
    I had flown into SFO and had to get to San Jose as Silicon Valley was getting started there.
    My boss had told me to simply take a taxi, but I noticed a helicopter service, fortunately, cheaper than a taxi. I got onto it on the N side of SFO fortunately, because the 1st leg of the trip was a whirlwind very fast tour around the edge of the tarmac to the S end of SFO at about 10% weight followed by 2G up and down into each of the airports between SFO and San Jose : Palo Alto and so on.
    How this pilot was tolerated told me that good Wild Western values were still strong in California. This experience turned me on to the idea of staying there, trying to get a transfer there beyond my Business Trip.
    The Altamont Rock Festival a few months later, how it went so bad, sent me the other way in my thinking, made me homesick about Sweden . . . ultimately Australia.

  8. ~1.2G due to super-fast landings and take-offs, NOT the “2G” of my last post!
    None of the above is going to happen . . .
    Our being so peripheral . . . and so long into a Peace now dominated, Economically, by a China into fast trains. https://www.chinadiscovery.com/china-maps/high-speed-railway-map.html
    and wanting to sell them . . .
    The only time our leaders become technophilic is during wars, technology being so important to winning wars.
    In Peace time it is everybody else who wins.
    In Australia.
    Chinese might become interested in my Ground Effect train idea. That’s the only way it might ever be developed.
    Least Risk that way, to growing Money, that is.
    Who’d want to R&D it here ?!?
    Nobody !?!

  9. The ultimate ground effect vehicle is a hovercraft. And people have tried to make hover trains. The extra note on my original examines this.

    1. I was suggesting that the hovercraft idea had NOT been taken far enough, particularly along railway lines and in relation to AI micro-control of lift produced by fast forward motion rather than fans.

          1. When using LIFT to greatly reduce weight, do away with wheels altogether, use the steel sprung skids that aircraft sometimes use in lieu of front wheels. Cheap and NO Hunting oscillation . . .

            1. Not as silly as it might seem, this “steel sprung skids . . . in lieu of front wheels” idea. Not only “Cheap and NO Hunting oscillation”. It also enables a quick takeoff following the GTEs’ getting up to takeoff thrust and overcoming their large static friction
              What I am envisioning here is say a Skytrain looking like the above-pictured Incat catamaran with its central body thus sitting on the railway with GTEs overhead.
              Its wave-piercing sides would straddle the railway reserve to produce Ground Effect.
              Instead of wave piercing floats though, it would have Worst-Case skids similar to the wing tip wheels on B-52s, to correct any abnormal tilt,
              normally NOT happening because of AI control.
              AI control would have several stages, all of them controlling ailerons, “the hinged flight control surfaces usually forming part of the trailing edge of each wing of fixed-wing aircraft”.
              The first stage would also take the Skytrain up to 10% weight, by somehow following the rail lines.
              The 2nd stage would take this further, to say 1% weight, by also following, and minimizing, say the screeching sound of the “steel sprung skids . . . in lieu of front wheels”.

          2. While The Plan of my big AI-ed rail vehicle, would look rectangular as for the Incat ferry in the above photo,
            its Front Elevation would look more like a B-52’s in the following video, its main skids and cabin being very central above the railway tracks.
            Its Graceful Degradation side skids would look like the wing tip wheels on this B-52 https://www.youtube.com/watch?v=oZm6IhYNmfw

  10. Further to this earlier :

    “All that is needed on our tracks is existing and developing AI-ed driver-less road vehicle technologies”.

    The “lighter, smarter, faster, more perceptive trains and carriages” would be cars, running on rails instead of roads, that automatically join into trains to minimise drag while decreasing congestion, THE SAME AS WOULD BE HAPPENING ON OUR ROADS.
    The single rail problem created by thus putting cars onto them has two solutions :
    1.) Building an extra rail to allow 2-way traffic ;
    2.) Time-Sharing existing single rails. Say
    a) city-bound traffic in the mornings, forcing country-bound morning traffic onto roads,
    b) country-bound traffic in the afternoons, forcing city-bound afternoon traffic onto roads.

Comments are closed.