An interactive guide
How Railways Work
Two steel rails, a wheel with a flange, and 200 years of engineering. Follow the line from the first wagonways to 600 km/h maglevs, then take the driver's seat and run the signal box yourself.
- 1825
- first public steam railway
- 1,435 mm
- standard gauge
- 25 kV AC
- most common overhead supply
- 603 km/h
- fastest train ever run
Two centuries on rails
A short history of the railway
Pick a year on the line. Each stop is a moment that changed how trains run.
Traction
How a train works
Every locomotive does the same job: turn stored energy into a turning force at the wheels. What changes is where the energy comes from.
Why steel wheels on steel rails
A loaded wheel touches the rail on a patch about the size of a small coin. Steel barely deforms, so very little energy is lost to rolling. That is why one locomotive can pull thousands of tonnes, and why a train uses far less energy per tonne than a lorry.
The price is grip. Steel on steel is slippery, so trains accelerate gently, brake over long distances and avoid steep gradients. Sand dropped in front of the wheels helps when the rail is wet.
How a train steers without a steering wheel
Train wheels are slightly cone-shaped and fixed to a solid axle. Slide the wheelset sideways and one wheel rolls on a bigger radius than the other, which steers it back to the centre. The flange is only the backup.
How a train stops
Air brake
A pipe filled with compressed air at 5 kg/cm² runs the length of the train. The driver lowers the pressure to apply the brakes. If the train breaks in two, the pipe empties and every vehicle brakes by itself. George Westinghouse patented the idea in 1869 and it still protects every train.
Regenerative brake
The traction motors are switched to work as generators. Slowing the train produces electricity, which flows back up the pantograph for other trains to use. Modern electric trains recover a large share of their braking energy this way.
Emergency brake
Dumping all the air at once gives the hardest stop the wheels can manage. Even then, a passenger train at 160 km/h needs more than a kilometre, which is why signals warn the driver well in advance.
Electrification
From power station to wheel
An electric locomotive carries no fuel. It is one end of a circuit that starts at a power station and returns through the rails.
- Power station. Electricity is generated and sent across the country at very high voltage to keep losses low.
- Traction substation. Every 40 to 60 km a substation steps the grid supply down to 25 kV single-phase AC and feeds it to the line.
- Overhead line. A copper contact wire hangs from a catenary cable at a constant height of about 5.5 m. It zig-zags slightly so it wears the pantograph evenly.
- Pantograph. A sprung arm presses a carbon strip against the wire with about 7 kg of force and collects the current, even at 300 km/h.
- Main transformer. Under the floor, 25,000 V is stepped down to roughly 1,000 to 2,000 V for the traction equipment.
- Rectifier and inverter. The supply is turned into DC, then rebuilt as three-phase AC whose frequency and voltage the driver's controller varies. Frequency sets motor speed, so the train accelerates smoothly with no gears.
- Motors and return. The motors turn the axles. The current then leaves through the wheels into the rails and flows back to the substation to complete the circuit.
Overhead AC
25 kV at 50 or 60 Hz is the world standard for main lines and high-speed routes, including almost all of Indian Railways' broad-gauge network. High voltage means thin wires and substations far apart.
Overhead DC
1,500 V and 3,000 V DC came first and are still used in parts of Europe, Japan and on many suburban lines. Mumbai ran on 1,500 V DC from 1925 until its conversion to AC was finished in 2016.
Third rail
Metros and some suburban railways use a live rail beside the track at 600 to 750 V DC, picked up by a sliding shoe. It needs no tall masts, which keeps tunnels small, but it limits speed and power.
Permanent way
The track, and how a train changes it
Track looks simple. It is a layered structure that spreads a 20-tonne axle load into soft ground, holds the rails exactly the right distance apart and lets water drain away.
What the track is made of
- Rails carry and guide the wheels. Modern rail is welded into lengths of many kilometres, so the old clickety-clack is disappearing.
- Sleepers of concrete, wood or steel hold the gauge. Spring clips grip the rail foot.
- Ballast, 30 cm of crushed stone, spreads the load, lets rain drain and can be re-packed when the track settles.
Gauge: the distance between the rails
Trains cannot run from one gauge to another. Where gauges meet, passengers change trains, wagons swap bogies, or special axles slide to a new width.
Changing track: how points work
A train cannot steer, so the track steers it. A set of points (a turnout, or switch) has two tapered blades that slide a few centimetres. Whichever blade touches its stock rail picks the wheels up and carries them onto the new route.
- Stock rails are the fixed outer rails. They never move.
- Switch blades are the two moving rails. They are tied together by a stretcher bar, so when one closes against its stock rail the other opens a gap for wheel flanges.
- Point machine. An electric motor beside the track throws the blades in about three seconds, locks them, and proves to the signal box that they are fully home.
- Crossing (frog). Further on, one rail has to cross another. A V-shaped casting leaves a gap for the flange, and check rails on the opposite side hold the wheelset in line while it passes.
- Locked under a train. The points cannot be moved while a train is on them. Try it: the button is refused while the bogie is on the blades.
Signalling
Keeping trains apart
A train cannot stop within the distance its driver can see. So the line is divided into blocks, and the rule is simple: one train in one block at a time. Signals tell the driver the state of the blocks ahead.
At least two blocks ahead are empty. Run at line speed.
The signal after next is red. Start to bring the speed down.
The next signal is red. Be ready to stop at it.
The block ahead is occupied or not proved clear.
How the track knows a train is there
A track circuit passes a weak current along the rails of each block. A train's steel axles short the two rails together, the relay drops, and the signal behind goes red. If anything fails, the relay also drops, so a fault always shows red. Axle counters do the same job by counting wheels in and out.
Interlocking
At a station, signals and points are tied together so that conflicting moves are impossible. A signal can only clear when every set of points on the route is set, locked and proved, and no other route crosses it. It started as steel bars in a lever frame. Today it is duplicated computers.
Automatic train protection
Systems such as ETCS in Europe, PTC in the USA and Kavach in India watch the driver. Track beacons or radio tell the train its permitted speed and stopping point. If the driver does not brake in time, the train brakes itself. Metros go further with CBTC, where trains report their position by radio and run only a few seconds apart.
Simulator
Drive a train
Run a timetabled service that obeys signals and speed limits, or pick free ride and go as far as you like through tunnels, caves and over bridges. Click inside the simulator to use the keyboard.
amber pipe · red cylinder
Know your cab
Selects forward, neutral or reverse. It can only be moved when the train is stopped with power off. In neutral the train cannot take power.
Eight notches. Each notch asks the motors for more pulling force. Open it gradually: full power from rest can spin the wheels on a real locomotive.
Lowers the brake pipe pressure to apply the brakes on every coach. Brakes take a second or two to build up and to release, so brake early and ease off as you stop.
Empties the brake pipe at once. The train's protection system pulls it for you if you pass a red signal or go well over the speed limit.
Raises the current collector to the overhead wire. No pantograph, no power. On the diesel this button starts the engine, and a steam engine is always in steam.
Sounded when starting, before level crossings (the W/L board) and to warn people near the line.
Open only when the train is stopped. Power is locked out while they are open, exactly as on a real train.
The modern dead-man's handle. If the driver touches no control for 45 seconds an alarm sounds, and if it is not acknowledged the brakes apply.
The amber needle is the brake pipe, 5 kg/cm² when released. The red needle is the brake cylinder pressure, which is the force actually on the wheels.
Signal box
Run the control room
You are the signaller at a three-platform station. Trains arrive from the west. Set a route into a platform, let passengers board, then send each train on its way. The interlocking will refuse anything unsafe and tell you why.
Train describer
How a train comes into a platform
- Offered. The box in rear offers the train. It runs up to the home signal, which stays red until a route is ready.
- Route set. The signaller chooses a platform. The point machines move the blades and prove they are locked.
- Signal clears. Only now can the home signal show proceed. The route is locked so nobody can move the points under the train.
- Slow over the points. A diverging route is taken at reduced speed, typically 15 to 30 km/h.
- Stop at the starter. The train draws along the platform and stops short of the starter signal. Doors open.
- Right away. When the line ahead is clear the starter is cleared, the guard gives the bell, and the train leaves. Signals drop back to red behind it.
Railway technology
Kinds of train and the ideas behind them
Facts and records
Fastest, longest, highest
Figures are as reported up to 2026. Records in railways fall often, so treat the newest ones as a snapshot.
Did you know