Sections of normal rail lines are joined buy plates with a gap between the sections; this gives the familiar clackerty–clack sound we all know. This gap between the sections allows for expansion plus the many curves also allow slight movement of the rails due to heat changes.
The new high speed rail sections on the other hand are all welded together and laid as straight as practically possible with few curves.
Why don’t the rails buckle and twist out of shape with temperature changes?
Every thing expands and contracts with temperature change. Bridges do it even concrete ones, as do buildings. The outer skin of Concorde AFAIR expands about nine inches in supersonic flight. So stands to reason hundreds of miles of rail line would expand a considerable amount.
Continuously welded track (as it is called) is not actually completely continuous. It’s just that the expansion joints are further apart – typically a couple of miles (3-4km), vs. 39 feet (10m) on traditional track.
Plus, the rails are stretched and fixed under tension, which means that as they “expand” they actually become “less stretched” instead. (I believe they are actually stretched so that the “zero point” is somewhere within the expected temperature range, so they will still be under compression some of the time.)
In addition to expansion joints, there are two other ways to cope with heat expansion. The rails can be laid on concrete ties. The added mass makes expansion more difficult. The most common thing that is done, however, is to pre-stress the rails before they are welded together. This is simply done by laying the tracks down during the summer. If a track is laid when the ambient temperature of the rails is around 90deg F, they should be able to survive in temperatures up to 120deg F.
If you get it wrong, this happens (second picture).
Ok I understand that so the tracks are laid in the summer so what happens in the winter when the temperatures drop to below freezing. I imagine contraction would be a problem to also.
Some of this has been mentioned above, but to clarify:
The US is a railfan’s Mecca in a way, because it holds on to older short-length rails in most parts, so it’s ear candy for train buffs. Most industrialised countries these days use continuously welded rail (CWR).
CWR can vary in length from quite short pieces up to hundreds of miles long. The reason it doesn’t expand in the heat is simply because it can’t - the brute force method is used in laying the stuff…
Old-fashioned rail is hammered into wooden sleepers using dog spikes. In the heat, the spikes simply pop and the rail turns to spaghetti. CWR is generally clamped onto extremely heavy concrete sleepers (US: ties) using specialised clips (Pandrol is a company that makes a lot of them) that look a bit like oversized paperclips. These typically exert a pressure of around two tons pressing the rail against the sleeper. The sleepers in turn are laid in a generous ballasting of metal chips. So when the rail wants to expand, the expansion forces are large, but the manmade impediments to these forces are larger still, and the rail simply can’t move the huge weights of concrete in the ballast, so it just sits there. Rails are generally laid at the midrange of temperature (extreme cold can be nasty too, making rails prone to breakage). In cold laying conditions, the rails are artificially heated first.
It’s estimated that if the Adelaide - Darwin (north-south Australian trans-continental) railway were allowed to expand as it wished in the summer, the rails would expand four kilometres off the northern coast!
CWR can basically be laid in indefinite lengths. It generally isn’t though, but that is for reasons other than heat - you need to incorporate turnouts (switches) and also insulated rail joints for signalling track circuitry (if you look closely on the departure side of a grade crossing, you’ll see these).
Here’s another way of dealing with it. The Sydney Harbour Bridge is basically a huge lump of steel. It expands and contracts to all hell day and night, and has been doing this for 75 years straight. The bridge itself rests on huge hinges at the base to allow for this, but what about the railway that crosses it? Check out this photo of the expansion joints:
Credit to user 42101 of Railpage Australia for the photo.
Contraction is a problem, but a small one. This is for two reasons.
Small amounts of contraction are no big deal, the rail + rail road ties are strong enough to cope.
In the rare case where the track does break, it’s detected straight away (at least in the US). Since the tracks are continuously welded, they are a good electrical conductor. If the rail breaks, there is a change in resistance which is continuously monitored. Time to send out the repair train. Measuring a track for buckling is a lot more difficult.
Indeed. And one of the best aspects of this is what happens to the very next train that is bearing down on that section of track, which demonstrates an aspect of the failsafe nature of railway signaling. Well before the repair team arrives - in fact instantly and without human intervention - the break in the circuit will fool the signalling system into thinking the section is occupied by a train (a real train would show up in the same way by shorting the circuit through its steel wheels and axles). So the signals down the line are automatically thrown to danger.
Surprisingly though, track can be pretty damned rough and still allow trains to get through. Check out this quaint WW2 US Military training film showing exactly how difficult it is to derail a train intentionally.
Yeah. The tracks across the bridge are only for low speed conventional suburban rail, but the joints are there (three of them - one at each end of the brisge and one in the centre) because of the unique problems posed by the location (the expansion is the bridge’s rather than the rails’, and wooden sleepers are still used here as they perform better on the steel base).
But you’re right - such angled joints are frequently used in high speed rail applications in “normal” areas other than bridges. These are a result of railways’ safety obsession being up there with aviation’s, ie an expansion joint might not be strictly necessary for CWR, but hey, it can’t hurt, right?
There are disadvantages though - on the Sydney Harbour Bridge, for example, a train is not permitted to travel backwards over these joints unless they are manually clipped by a track worker and the train travels dead slow. Normally this is not an issue for high speed trackage, as it tends to be dual track, but you can’t easily have these things on single track lines.
Bit of a hijack from the original OP, but in the movie I wonder why they didn’t try blowing track on a curve? Seems to me it should be easier to derail a train engine when it is going around a curve than when it is going in a straight line.
