On this page
Most of the electricity you use comes from a turbine spun by steam, gas or water and coupled to a generator. The engineer who made the steam turbine practical was Charles Parsons. His key idea was simple to state and hard to build: do not let the steam expand all at once. This post explains why, with numbers, and what students can learn from how he proved it.
Who was Charles Parsons?
Britannica's biography gives his dates as 13 June 1854, London, to 11 February 1931, Kingston, Jamaica. The Science Museum Group's record notes that he began studying applied mechanics and mathematics at St John's College, Cambridge, in 1873. After joining the Armstrong engineering works in Newcastle in 1877 and a spell at Kitson and Co. in Leeds, he became a junior partner at Clarke, Chapman and Co. of Gateshead in 1884. That same year the first Parsons turbo-dynamo was built: a turbine driving an electrical generator. In 1889 he founded C. A. Parsons and Co.
Why did the turbine need many stages?
A piston engine pushes on a piston with steam pressure. A turbine works differently. Steam expands through nozzles or fixed blades, speeds up, and its kinetic energy turns a wheel of moving blades. Britannica describes Parsons' 1884 design as using several stages in series, each limiting the expansion so that the most kinetic energy could be taken out without the blades overspeeding.
To see why, use energy conservation. If steam loses an amount of energy Δh per kilogram as it expands, and all of it becomes kinetic energy, then ½v² = Δh, so v = √(2Δh).
Worked example (sample numbers). These are round figures for a classroom model, not data from Parsons' machines. Suppose the steam can give up 800 kJ per kilogram in total.
One stage: v = √(2 × 800,000) = √1,600,000 ≈ 1,265 m/s.
Sixteen equal stages, each taking 50 kJ/kg: v = √(2 × 50,000) = √100,000 ≈ 316 m/s.
Ratio: 1,265 ÷ 316 = 4, which is √16.
Splitting the drop into n equal stages cuts the steam speed in each stage by a factor of √n. For the blades to take up most of that kinetic energy, they have to move at a good fraction of the steam's speed, so a single-stage wheel would have to spin dangerously fast. Many stages let each wheel run at a speed the metal can stand.
How did Turbinia prove the idea at sea?
The Science Museum Group record says that in 1894 Parsons patented propelling a vessel by a steam turbine. His experimental boat Turbinia followed. According to Discovery Museum in Newcastle, where Turbinia is displayed, it was the first ship powered by steam turbines, and in 1897 it raced through the fleet at Queen Victoria's Diamond Jubilee naval review at about 34 knots, faster than any vessel present. Britannica gives the speed as 34½ knots. A knot is one nautical mile, 1.852 km, per hour, so 34 knots is about 63 km/h.
Britannica adds that his turbine was fitted with a condenser for power stations in 1891, and that he later invented reduction gearing between turbine and propeller, because a propeller works poorly at the high speeds a turbine prefers.
How is this linked to what you study?
The turbine is where heat becomes rotation; the generator is where rotation becomes electricity, by electromagnetic induction. If you are revising the second half, our guide to AC generators and transformers works through the emf equations. For why no heat engine can turn all its heat into work, see our post on Sadi Carnot and heat engines.
His honours, as listed by Britannica and the Science Museum Group, include Fellowship of the Royal Society (1898), the Rumford Medal (1902), a knighthood (1911) and the Order of Merit (1927).
What can a student take from Parsons' way of working?
Find the limit first. Parsons' design started from a physical limit, blade speed, and worked backwards.
Prove it where people will see it. Turbinia was a demonstration as much as a ship.
Keep improving the system, not just the part. Condensers and gearing came after the turbine itself.
The square root in v = √(2Δh) is the kind of step that is easier to grasp when you can change a number and watch the result. The Fermi Simulation Lab has interactive simulations for ideas that are easier to see than read.
Next step: rework the example for 25 stages and check that the stage speed falls by a factor of 5.
