Fire and Wind

Alexandria, probably in the first century of our era. An engineer named Hero writes a book of machines worked by air, water and steam, and one entry, number fifty, begins with an instruction:

Press play and watch the centuries run: the cities lit, in order, where the ideas of this prologue were written down or built, with the empires that paid for them on the timeline beneath. Drag the timeline to stop at any year.

Alexandria under the Ptolemies was where Greek engineering met the wealth of Egypt. Its first great mechanic, Ctesibius (285–222 BCE), wrote the first known treatises on compressed air and its use in pumps; both Vitruvius and Hero drew on his work.

The Roman architect Vitruvius (c. 80 – c. 15 BCE) already knew the device by name, though not as a spinning machine — as a vessel that blows:

The first machines to do real work with the wind were built in Persia. The historical record does not support any windmill before the ninth century CE, and the earliest recorded working design, invented some time around 700–900, was Persian: the panemone, sails of reed matting or cloth on a vertical axle, inside a wall pierced with slits. It was first built to raise water and later adapted to grind grain (panemone windmill). The sources disagree on the number of sails — four to eight in one account, six to twelve in another.

By the ninth century such mills were turning in Khorasan, across eastern Iran and western Afghanistan; from there they spread across the Middle East and Central Asia, and later to Europe, China and India. A legend credits their invention to Abu Luʾluʾa Fīrūz, a Persian slave in the time of the caliph ʿUmar who was said to have built windmills — a story, not a record.

The country of the mills is the east of Iran, where a north wind blows almost all year: in Sistan they call it the wind of 120 days, and it can reach 100 km/h. Iran’s nomination of its windmills — asbād— to UNESCO’s tentative list says the first of them seem to have originated in Sistan; at Nashtifan, near Khaf, a row of them still stands (UNESCO tentative list, 2017).

In ninth-century Baghdad the caliph al-Maʾmūn gathered scholars and manuscripts at the House of Wisdom. Among them were three brothers, the Banū Mūsā — the sons of Mūsā ibn Shākir, an astronomer at the caliph’s court — raised and educated there by al-Maʾmūn’s order after their father’s death. They had Greek works acquired from Byzantium translated, measured the length of a degree of latitude to find the size of the Earth, and wrote some twenty books, of which three survive.

The most famous, the Book of Ingenious Devices of 850, describes about a hundred machines. Some were inspired by Hero and by Philo of Byzantium; many were their own. What sets them apart is a fascination with machines that control themselves — “self-operating valves, timing devices, delay systems” — which took them, in one historian’s words, “well beyond anything achieved by Hero or Philo” (Book of Ingenious Devices).

In 1206 an engineer of the Jazira, the land between the upper Tigris and Euphrates, finished a book for his patron at the Artuqid court in Mardin. Badīʿ al-Zamān al-Jazarī had served there as chief engineer, as his father had before him. The Book of Knowledge of Ingenious Mechanical Devices describes fifty machines, with instructions for building them: water clocks, among them the famous elephant clock, automata, fountains, and machines for raising water.

Among their parts is the camshaft, a shaft carrying cams, which al-Jazarī used in his automata, clocks and water-raising machines (al-Jazarī). Claims that he invented the crank do not survive scrutiny: the crank and connecting rod appear in a Roman sawmill at Hierapolis in the third century. What survives is the systematic way he put such parts together, and wrote down how.

Under the Mamluk sultans a Syrian chemist and engineer, Ḥasan al-Rammāḥ, studied gunpowders and explosives. He devised several new types of gunpowder, a new fuse and two kinds of lighter, and he sketched weapons — among them one he called “an egg which moves itself and burns”: two sheet-metal pans fastened together and filled with naphtha, metal filings and saltpetre, pushed across the water by a large rocket and kept on course by a small rudder.

It is often called the first torpedo. Historians of technology classify it instead as a rocket-propelled surface weapon: it was meant to skim across the water, and there is no evidence it was ever used in battle (Ḥasan al-Rammāḥ). Its engine, though, is the one this course is about. A rocket carries everything it throws out; a jet engine breathes in the air it throws back. Both are pushed by the same reaction.

Taqī al-Dīn was born in Damascus in 1526, studied there and in Cairo, and served as a judge in Palestine, Damascus and Cairo before he moved to Istanbul in 1570 and became the Sultan’s chief astronomer the next year. In 1551, in a book whose title translates as The Sublime Methods of Spiritual Machines, he described a spit that turned itself: steam is directed onto the vanes of a wheel, and the wheel turns the axle with the meat on it (Taqī al-Dīn).

Where Hero’s ball was pushed by the steam it threw out, Taqī al-Dīn’s wheel was pushed by the steam that struck it: an impulse turbine. The same book describes four water-raising machines, the last a six-cylinder pump whose pistons are worked by cams and trip-hammers from a paddle wheel. Under Sultan Murad III he built the observatory of Istanbul, and from it observed the Great Comet of 1577.

Hot air had been turning wheels for centuries. Chinese lanterns spun by the rising air of their own flame were shown at lantern fairs as early as the Northern Song dynasty, and around 1500 Leonardo da Vinci drew a smoke jack: a little turbine in the chimney, turned by the hot air of the fire, driving the roasting spit through gears and a chain (gas-turbine engine).

In 1629 Giovanni Branca published in Rome Le machine, sixty-three engravings of machines, one of them a wheel turned by a jet of steam — like Taqī al-Dīn’s, an impulse turbine. The historian Alex Keller found that Branca’s machines looked like armchair inventions, seldom if ever built (Giovanni Branca).

Then, in 1791, an English coal viewer from the collieries around Nuneaton, John Barber, took out a patent that contained the whole idea of a gas turbine: a chain-driven compressor, a combustion chamber, and a turbine driven by the hot gas (John Barber). Nothing practical came of it. The reason is the subject of the figure below, and of Lesson 4-1: the turbine has to pay for its own compressor before it can give anything away. In Berlin in 1904 a gas turbine built by Franz Stolze, on a design he had first filed in 1873, ran too inefficiently even to keep itself turning. A year earlier, in Kristiania, the Norwegian Ægidius Elling had built the first gas turbine that produced more power than it consumed.

Maxime Guillaume had patented a jet engine in 1921; it could not be built with the technology of the day. In January 1930 a young Royal Air Force officer trained at Cranwell, Frank Whittle, patented the turbojet. Five years later he could not afford the £5 renewal fee, and the patent lapsed. With his small company, Power Jets, and the turbine makers British Thomson-Houston, he built an experimental engine at Rugby; on 12 April 1937 the Whittle Unit ran for the first time (Frank Whittle).

In Germany a physics student at Göttingen, Hans von Ohain, had come to the same idea: “My interest in jet propulsion began in the fall of 1933,” he wrote, “when I was in my seventh semester at Göttingen University.” Working as assistant to the physicist Robert Pohl, he patented a jet engine of his own — in 1935 or 1936; the sources differ. In February 1936 Pohl wrote to the aircraft builder Ernst Heinkel, and on 27 August 1939 von Ohain’s engine flew in the Heinkel He 178 from the company’s airfield at Marienehe, near Rostock, for six minutes: the first flight of a jet aircraft (He 178). Whittle’s engine had run earlier; von Ohain’s flew first. Whittle’s own first flight came on 15 May 1941 — that day opens Lesson 3-1.

None of these machines was a jet engine. Between them they held every idea one needs: a jet that pushes back on what throws it; wind turned into torque on a shaft; impulse and reaction blading; machines that regulate themselves; and the three parts — compress, burn, expand — that must be made good enough for the turbine to come out ahead.

Hero’s ball and a GE9X obey the same law. Each throws gas backwards, and each is pushed forwards by exactly as much as it throws. On Hero’s ball the push was enough to turn a toy. On a GE9X it is about 470 kN — enough to hold up 48 tonnes — and all of it has to reach the wing through a handful of metal parts between the engine and its pylon.

In 2024, after a test flight from Kona, Hawaii, mechanics found one of those parts cracked. That is where Lesson 1-1 begins: with the force itself — where it comes from, how to calculate it, and where it goes.