In the winter of 1763 the University of Glasgow sent a broken model to its instrument maker for repair. It was a small working Newcomen engine used for teaching, and it would not run properly: a few strokes and it ran out of steam, every time.
James Watt was 27 and had a workshop inside the university because the Glasgow guilds had refused to let him trade in the town, on the grounds that he had not served the required apprenticeship. He fixed the model, and then spent two years working out why the fault was not a fault at all but the design working exactly as designed.
The answer he arrived at is the single most consequential piece of engineering of the eighteenth century, and it is embarrassingly simple once stated.
What the Newcomen Engine Did
Thomas Newcomen’s engine, first working in 1712, pumped water out of mines and did it for fifty years without serious modification. Steam at barely above atmospheric pressure filled a large cylinder under a piston. Cold water was then sprayed into that cylinder, the steam condensed, and the vacuum left behind let atmospheric pressure push the piston down. The beam rocked, the pump rod lifted, and the cycle began again.
It works. It is also, in fuel terms, catastrophic, and the reason is in the sequence. Every stroke required the cylinder to be cooled enough to condense steam and then reheated enough to accept it again. Most of the coal burned went into warming a large mass of iron back up after it had just been deliberately chilled.
Watt measured this instead of estimating it. He worked out how much steam the model actually consumed, found that it was several times more than the volume of the cylinder, and traced the loss to the walls. He was helped by knowing Joseph Black, professor of medicine at Glasgow, whose work on latent heat explained why condensing steam gives up so much energy and why reheating the metal costs so much.
Watt was careful in later life about how much he owed Black, and Black was generous about how much Watt had worked out for himself. The honest summary is that Watt found the problem empirically and Black gave him the vocabulary to describe it.
The Walk on Glasgow Green
The solution came to him on a Sunday afternoon in May 1765, walking near the Glasgow washing house. He described it decades later in terms he never revised: if steam is an elastic body it will rush into a vacuum, so if a connection is opened between the cylinder and an exhausted vessel, the steam will go there, and it can be condensed there without cooling the cylinder.
That is the separate condenser. One extra vessel, kept permanently cold, connected by a valve. The working cylinder can now be kept permanently hot, jacketed with steam and closed at the top so that steam rather than air pushes the piston.
The gain was between three and four times the work from the same coal. In an economy where deep mines were reaching the limit of what they could afford to pump, that was the difference between a colliery being viable and being abandoned.
Eleven Years From Idea to Product
The gap between the insight and a sellable engine is the part of the story that gets skipped, and it is the part that matters to anyone who has tried to build something.
Watt patented the principle in 1769 and could not make it work at scale. The obstacle was that nobody could bore a cylinder round enough. A large cylinder in the 1770s might be out of true by a centimetre, so the piston had to be packed with rope, leather and tallow, and the vacuum leaked away past it. His first backer, John Roebuck, went bankrupt.
Two things rescued it. In 1775 Matthew Boulton, who owned the Soho Manufactory outside Birmingham, took over Roebuck’s share and supplied capital, workshops and commercial judgement. In the same period John Wilkinson patented a boring machine developed for cannon, in which the cutting head is carried on a rigid shaft supported at both ends rather than on the workpiece. It could bore a cylinder of 1.8 metres accurate to about the thickness of a coin, and it made the Watt engine possible.
Boulton and Watt then did something unusual: rather than selling engines outright, they charged a royalty of one third of the fuel saved compared with a Newcomen engine doing the same work, measured by the customer’s own coal bills. The engine paid for itself out of the savings it created, which was the only way of selling it to mine owners who were suspicious of anything new.
Everything After the Condenser
Watt’s later patents matter as much as the first, because they turned a pump into a general source of power.
- Rotative motion. A beam engine goes up and down, and a factory needs a turning shaft. The obvious solution, a crank, had been patented by James Pickard, so Watt’s assistant William Murdoch devised the sun and planet gear to get around it, an elegant piece of spite that also doubles the output shaft speed
- Double acting. Steam admitted alternately to both sides of the piston, so the engine drives on both strokes rather than one
- Parallel motion, a linkage that lets a rigid rod push and pull the end of a rocking beam while staying vertical. Watt considered it his most ingenious invention and said he was prouder of it than of anything else he did
- The centrifugal governor. Two weighted balls spin on a spindle driven by the engine. If the engine speeds up they fly outwards, and the linkage closes the steam valve. It is a self correcting loop with no human in it, and it is the ancestor of automatic control. Maxwell wrote the first mathematical analysis of its stability in 1868
- The indicator diagram, a device that traced pressure against volume inside the cylinder throughout the stroke. The area of the closed loop is the work done per cycle. Boulton and Watt kept it a trade secret for decades, which is why the pressure volume diagram, now the standard tool of thermodynamics, entered the public literature late
He also needed a way to sell an engine to somebody who currently used horses, so he measured what a horse could lift over a working day, rounded generously, and defined the horsepower. The unit was a sales instrument. It outlived the animal by two centuries.
The Patent Argument
Watt’s 1769 patent was extended by a private Act of Parliament in 1775 to run until 1800, giving the partnership a twenty five year monopoly on condensing steam engines in Britain.
Historians still argue about whether this delayed the industrial revolution. Boulton and Watt litigated vigorously, including against Jonathan Hornblower’s compound engine, and Watt was firmly opposed to high pressure steam, which he considered dangerous. He was not wrong about the danger: early boilers exploded and killed people. But high pressure is the route to a compact engine that can move itself, and Richard Trevithick could not develop it seriously in Britain until the patent expired. The locomotive follows within four years of 1800.
The counterargument is that without the patent there would have been no capital and no Soho Manufactory, and the engine would have arrived later or not at all. Both things can be true.
What Watt Did Not Do
He did not invent the steam engine, which was Savery and Newcomen and, before them, a long line of experiments with vacuum and pressure. He did not sit watching a kettle as a boy; that story appears in Victorian children’s books a generation after his death.
He also had no thermodynamics, because there was none to have. The theory that explains why his engine was better, and how much better any engine could possibly be, was written by Sadi Carnot in 1824, five years after Watt died, and it was written by a Frenchman trying to understand why British engines were so much better than French ones. Carnot’s ideal engine is an abstraction of exactly the machine Boulton and Watt were selling, and the discipline of thermodynamics is, historically, an attempt to explain a piece of hardware that already worked.
The unit of power was named the watt in 1882, by the same congress that named the joule after the brewer who established that heat and work are the same currency. Watt would have understood the compliment. He had been charging customers by the coal for a hundred years before anyone proved the exchange rate.
No products in the cart. 