S5C1: Trains & Pipelines

>>> Click here to access this episode of the Syllab Podcast on Spotify <<<

a) The steam engine and the locomotive

The advent of the industrial age during the second half of the 18^th century in the Western world was marked by the mechanization of equipment, itself made possible by the harvesting of various types of energies during the following decades. These were described shortly in S1 Section 6.c before the notion of energy transfer and conservation was explained in the subsequent S1 Section 6.d. Later within this first series, we spent some time understanding the concept of mechanical energy and the ability to do work, this was in S1 Section 10.d and it underpinned one of the key inventions which ushered in the Industrial Revolution, namely the steam engine and its association with the iconic steam locomotive.

In regions increasingly relying on the extraction of raw materials such as coal and iron oxide located at some distance from the centres of consumption, the development of an efficient way to move large tonnage unsurprisingly proved to be transformational. And that which could transport goods could also move passengers; only a different, more comfortable type of carriage was required. I expect pretty much anybody would have seen a modern electric train and even though there would have been some weight savings between now and then using new materials, we are still within the same order of magnitude. It may thus seem quite incredible a steam engine in a locomotive could pull such a weight. How does this work?

Something we need to acknowledge is the need for trains to run quite fast from point A to point B, though this requirement is much less of a concern for freight compared to passenger services, but they are not required to rush out of the station therefore elevated speeds can be attained over time through the continuous application of a force as long as this propulsion force is not offset by an equivalent drag and friction – as per Newton’s second law of motion (refer to S1 Section 6.d). From this perspective, one of the main purposes of rails is not merely to guide a train, it is also to reduce friction. We will take a look at railway tracks in S5 Section 7.c.

This ability to stretch the timeframe of application alleviates the instantaneous force requirement, which is produced by concentrating pressure against a piston, typically around 1.5MPa (about 218 psi) and much higher in the high-pressure versions. The first part of the process is the boiling of water into high-temperature steam and with this phase transition comes an expansion in volume to which further kinetic energy is injected via heat transfer. This water vapour is then channelled into a chamber with a piston on one side and the pressure resulting from the volume expansion and kinetic energy of the gas causes the piston to move. That’s essentially it. There only remains for the spatial translation of the piston to be transmitted through a rigid metal rod and then a cog or crank so it induces the rotation of one or several wheels. By using a valve system to conduct the steam on either side of the piston in an alternating fashion, one side at a time, there is no lost mechanical phase and this to-and-fro is matched by the chugging noise accompanying the exhausting of used steam, at its most intense when a train is pulling out of the station and maximum force is required.

The amount of force applied to the piston is a function of the displacement so the volume of air contained in the cylinder is often used as a proxy for power – this should remind you of a car engine size such as 2.0 litres or that of a motorbike such as 500cc (or 0.5 litre), though the size of the boiler is clearly an important factor as well. Indeed, pistons generally come by pair for each set of drive wheels and there may be several sets of those wheels. Some of the most powerful steam locomotives ever used in freight had tractive power of over 600kN and power outputs of up to 5,000kW, or about 30 to 40 standard modern mid-sized cars.

b) Electric locomotives and high-speed trains

The success of the steam engine was however always going to be short-lived on account of both its noxious fumes, the result of incomplete combustion entailing the presence of particles, and its low thermal efficiency of 6-8%. Thermal efficiency is the ratio of the heat input over the work output. And producing heat has a cost and practical implications. In comparison, the ratio for diesel locomotives developed later on was 30-40% and it reaches 90% for electric locomotives, the technology used today in most versions of rail transport. We will skip the diesel engine for the time being, but only because it will be covered during the next chapter on cars and trucks, in S5 Section 2.b. Nor will we explore the hybrid versions such as diesel-electric because what we are interested in is electricity as prime mover, not as a transmission system between the engine and the traction mechanism.

Thus, electric locomotives are powered by electricity which can be stored in the form of batteries or it can be transmitted from an external source, by way of overhead power lines or a conducting “live rail” (and of course, a combination of both can be used). Not only is using electricity much cleaner (this says nothing about the production of electricity which occurs elsewhere) and the conversion into work more efficient, but this efficiency is increased by the ability to recover some kinetic energy through friction, in particular with regenerative braking. I include a link to the Wikipedia entry for this concept and technology, also used in cars and trucks, at the end of this chapter.

Whereas a steam engine extracts chemical energy from fuel, then converts this into heat (thermal energy) to eventually yield mechanical energy, there is only one conversion involved in an electric motor, from electrical to mechanical. Depending on the gearing, the mechanical force produced can be linear or rotary – what we call torque, a concept we introduced in S1 Section 8.d when discussing angular momentum and magnetic moment. I invite you to read or, hopefully, re-read this section because the magnetic field is the phenomenon at the core of the device.

In an electric motor, electric current is channelled to a static structure holding field magnets called a stator. The flow of electric charges creates a magnetic field exerting torque on a rotor, which unlike the stator can move, and it is this movement which constitutes the conversion into the mechanical energy leveraged to drive the wheels. No piston moving, no explosion as in a combustion engine and no exhaust noises, this makes electrical engines ideal for trains operating in urban environment although the infrastructure that goes with transmitting electrical power is somewhat cumbersome, and not visually very attractive in the case of overhead lines.

When coupling more powerful electric locomotives with high-speed rail infrastructure, we end up with trains starting much faster (their full torque is available at near zero revolution per minute) and that can also go much faster much longer since there is no refuelling required. How fast? At the time of writing, the maximum non-operating speed for trains moving on metal rails is 575km/h, set by the French TGV in 2007, and for maglev (short for “magnetic levitation”) it is 603 km/h set in Japan in 2015 – we will cover maglev as part of the section on railways track mentioned earlier. In terms of operating speed, the Shanghai Maglev linking the main airport with the city centre sprints at 430km/h and high-speed lines in Germany, Japan, China and France all operate at top speeds of between 300 and 350km/h.

c) Urban rail transit

Originally, trains were developed to move freight or to transport passengers between cities rather than within them. However it soon became evident lighter versions could be used to provide public transportation within dense urban areas. Obviously, the infrastructure implies fixed routes and schedule, thus it cannot provide the same flexibility as individual means of transport, but it turns out to be cheaper on a per-trip-per-person basis and doesn’t have the same congesting effect as (non-autonomous) car usage. Arguably, walking or cycling is still the best option; however it is not always a practical one depending on weather, topography, dress-code and the distances involved. The three main forms of urban rail transit are the metro, the tram (called streetcars in the USA), and monorails. We will look at each one in turn, in this order.

Metro, or technically “mass rapid transit”, has the highest throughput and is probably the fastest method, though perhaps not the most enjoyable one. Its drawbacks are the flip side of its advantages and are the result of grade separation. Indeed, the reason it is so efficient is that it is designed and built to operate on a horizontal plane which does not conflict with the road network at ground level. In most cases, this means the digging, railing and cabling of underground tunnels with expensive stations and a complex array of escalators and access points to the surface where everyday life takes place. At the periphery, or if the nature of the ground prevents stable tunnelling, metros can also run on elevated lines. Typically, the trains are guided by two metallic rails with electrical power delivered through a third rail and the lines can be designed over time with extensions, bifurcations, and circle lines linking several other straighter lines. The regular interspersing of stations and non-sharing of rails between lines and directions of service allows for high frequency while ensuring safety and a fully automated system.

The decongestion impact of metro lines is well proven, however the capital investment required means it isn’t always the optimal option for urban rail transit. In this respect, the tram or “light rail” version of metros often proves popular because it does not require overheads or underground tracks and the investment outlays this implies. Nevertheless, this means there is no grade separation and trams operate on the same horizontal plane as cars, bicycles and pedestrians. This does not necessarily translate into the absence of spatial segregation and, whereas the tracks sometimes coincide with a road lane, whenever possible the ground-level network is redesigned to accommodate dedicated tram lanes with priority given to these public services at intersections through a signalling system. Historically, this did not make for pretty cityscape since power was transmitted via overhead lines and the tram cars would have trolley poles sticking out diagonally over their top but modern versions tend to use sub-surface live rails supplied with direct current instead. For safety reasons, this rail can only be accessed physically when the tram vehicle is above it. There are competing technologies in this respect and, if you wish to know more about them, you can refer to the Wikipedia entry for ground-level power supply, a hyperlink is enclosed in the last section. Interestingly, in cities with significant slopes, trams sporting rubber tires rather than rolling on metal rails can be used in conjunction with a central guiding rail.

Lastly, lighter than the metro and also operating on grade-separated infrastructure, is the monorail. It is mostly laid on overhead tracks rather than underground, it has a smaller infrastructure and visual footprint, but it is slower with only a fraction of the capacity of a full-blown heavy-rail metro. The single beam acts as a guiding rail and, being physically segregated from pedestrians, the delivery of power through a live rail embedded within the track carries less safety issues than for tram lines.

d) Rail freight

If the key rationale for mass rapid transit is the cost of trip per passenger and relieving the saturated road network, for the movement of goods it is the price of transport per tonne per kilometre. In this respect, maritime transport is hard to beat, but many destinations do not have seas or rivers close by and the lay of the land may not always make digging out canals practical or even possible. Railway freight thus takes the second spot thanks to the low friction of rail Vs rubber tyres on asphalted roads and can be operated as an independent trip or as one link within an intermodal journey. The ability to chain several locomotives and dozens of freight wagons without creating a traffic mess or raising safety issues also helps bringing down manpower requirements and therefore operating costs.

Even though wagon length and width as well as track gauge are subjected to standardization within specific geographic markets, this doesn’t prevent a variation in the shape and properties of the freight wagons. Accordingly, these can be adapted to best suit the goods being transported, whether it is for bulk raw materials, liquids in tanks (including flammable ones), boxcars for stowing cattle and the fabled free riders of movies, as well as flatcars with decks that can accommodate containers, and auto carriers to transport cars.

The same way passengers transfer from one train or plane to another in a station or airport and containers are being transhipped in some of the largest ports in the world rather than always following direct point-to-point routes, freights wagons can hitch a ride up to a classification yard and the cargo of incoming trains will be split into series of wagons and then re-assembled into new cargoes pulled by the same or other sets of locomotives to another yard or their final destination. This is only an option and, when it is more economical to do so or time is a constraint, then point-to-point routing through “unit trains” provides a direct alternative.

A key factor in the popularity of train usage when it comes to freight is what we call the network effect. Indeed, it is not sufficient for one party to have access to a nearby railroad, the other party in a logistic leg also needs to have access, and the more parties have access to the network, the more optionality and the more usage. Accordingly, regions with significant railway density will see a higher rate of rail freight compared to road trucking or barges on rivers and canals; to be sure this density should not be thought of on the basis of a geographic dimension only but also with respect to production and processing centres. This explains why the USA, Europe and China, in this order, have the highest penetration rate – more than 20% of the inland freight traffic pie (on a ton-kilometre basis) in the case of the first two regions.

There can also be limiting factors, often a historical legacy, such as difference in gauge size (the distance separating the two rails) requiring trans-shipment, which is both a scheduling hassle and an additional handling cost. For example, the rail gauge in France is 1.435m and in Spain it is 1.668m. We will revert to the topic of gauge later in this series, in S5 Section 7.c on railway tracks.

e) Pipeline technologies

Pushing the concept of freight train to the extreme, what is analogous to an uninterrupted flow of tank cars? It is liquids on rail, which we know as pipeline. Pipelines are cylinder-shaped sections made of plastic or metal linked to each other that can transport fluids, be they in liquid or gas forms. Unlike trains, the technology isn’t really used to transport passengers – or at least not yet, depending on whether the speculative hyperloop proves to be economically sensical and technically feasible to provide rides in a comfortable manner. The main commodities making their way through pipes include crude oil and petroleum products, natural gas liquids, raw materials in slurry form, water of course, ammonia, as well as more esoteric products such as milk or beer (those last two are very localized examples over short distances).

Similarly to freight train lines, point-to-point is not required and a hub system comes into play with feeders lines aggregating at tanks farms, from where the relevant fluids are injected in batches into long trunk lines, and as the cargo nears its point of delivery, it will be transferred into another hub with tank farms and then introduced into smaller distribution lines until its final destination, or a depot near it, from where it can be transported by tanker trucks.

Like rail, this requires significant investment upfront depending on distance, topography and requirements to lease land or right of way. Think for example of the laying of submarine pipelines; it requires special vessels carrying huge tonnage of pipe sections that are welded together, one piece at a time, starting from the shore and laid down onto the seafloor as the ship moves away offshore along the planned pipe route. Of course, the entire pipe need not be laid in one single journey, a second ship can take over and the first one return to embark another load, and so forth.

The positive trade-off of this very significant capital expenditure is low operating cost on a cubic-metre-per-ton basis. Those costs are incurred in relation with maintenance and the energy expenditure involved in pumping the fluids at regular intervals, on average every 50 to 100kms, so as to maintain pressure and keep the fluid moving in the correct direction, including uphill when necessary. The idea of mechanical pumps is to induce movement in a specific direction, generally by creating a pressure gradient or by applying mechanical force to physically displace the fluid (this works best with viscous fluids). For example, in the case of a centrifugal pump, an electrically-powered rotor will accelerate the movement of the fluid and this increased kinetic energy will then be translated into potential energy manifested in the form of increased pressure.

In order to separate batches, either because they consist of different products or they are fed in the pipeline system by different parties, hardware can be inserted to seal off the products and avoid commingling. The nickname for those is “pigs” due to the squeal-like noise they make while traveling, and they are also used to clean the pipes from the inside. Talking about maintenance, by introducing valves at regular intervals, it is possible to interrupt the flow within the pipe and empty sections to proceed with repairs as and when need be. In case of leaks, these are most often detected through the drop in pressure they occasion, yet locating them is the more challenging part, in particular in underground pipes. There are various instruments and techniques used including sensors mounted on pigs, the injection of tracer chemicals combined with surface detector and ground-penetrating radar which can detect changes in soil density.

f) Geopolitical flows

Pipelines are so efficient and such an integral part of trade flows, in particular for energy-related commodities, that with them comes increased mutual reliance between customers and producers. This works well on the surface, until it doesn’t, as evidenced by the energy crisis Europe experienced when stopping the flow of piped Russian gas. This means, there is a geopolitical aspect coming into the equation that needs to be considered upfront since this infrastructure and energetic dependence can be weaponized by either side, regardless of the collateral damage on neighbouring economies, or for that matter their own.

In addition, pipelines cannot be abstracted away from the territories they traverse, meaning it is not merely about the producer and consumer but also about what and who separates them. “Who” refers to different countries that may be politically unstable or simply unfriendly, not to say hostile. And this also captures the potential for terrorist attacks that may or may not benefit from tacit governmental support. Such considerations are central in deciding which routes to lay pipes on, as evidenced by the Baku–Tbilisi–Ceyhan pipeline bringing oil from Azerbaijan, Kazakhstan and Turkmenistan from the western shore of the Caspian Sea all the way to the Mediterranean Sea without transiting through Russia or Iran, after having had to be carried across the Caspian to begin with, in the case of the Kazakh and Turkmen oil.

As for “what”, this refers to the ecosystems exposed to disruption from construction and potential product leakage. For instance, a planned extension to the Keystone pipeline system dubbed “Keystone XL” would have linked oil-producing areas in Alberta (Canada) to the existing trunk line north of Cushing on the way to the refineries located along the Gulf of Mexico in Texas (USA) by way of the Bakken formation across Montana and North Dakota, another producing area. However, due to multiple environmental concerns, including the crossing of a large wetland ecosystem, the risk of spills into a massive aquifer as well as broader concerns relating to the role of fossil fuels in global warming and climate change, the project was eventually cancelled (at least at the time of writing) – and this despite the intense financial lobbying behind and not-so-behind the scenes.

Nord Stream is another interesting example of geopolitically sensitive pipeline but this is not a political essay so I include the link to the Wikipedia entry at the end of this chapter for those interested.

g) Trivia – Rack railways & funiculars

Pipes can go up rather steep inclines and traditional metallic rails can’t, or rather the rails can but not the train, and using rubber tyres would not always solve the issue and would require the building and maintenance of a road, a different prospect in mountainous terrain. Friction alone will not do. Instead, toothed rack rails are used, which allows for a pinion to push against the teeth of the rail without the risk of sliding. Effectively, the slope is being negated, or even inverted, over a few crucial centimetres and, in modern versions, electrical power is delivered though overhead lines.

This system is quite impressive in terms of the gradients it can defy, up to 48% for the Pilatus Railway in Switzerland which ascends 1,629m over a short 4.6km; this works out to a 35% average gradient.

Rack railways should not be mistaken with ground funiculars. Those consist of a pair of carriages attached to a cable forming a loop and they service the uphill and downhill stations in turn, in opposite directions. The idea is simple: the potential energy of the descending carriage is transferred to the ascending one. All that is required is to compensate for the potential imbalance in passenger weight and the cable friction. Generally, the traction does not take place in the carriages but in an engine room located within the upper station.

When moving at ground level, funiculars are practical though not really impressive. Not so with the aerial counterpart called cable cars, some of which can carry several dozen persons, and their skis, across large chasms and up vertigo-inducing mountain faces.