Wonders of World Engineering

 © Wonders of World Engineering 2014-20    contents  |  site map  | contact

Although steam had long been recognized as a potential source of power it was not until the seventeenth and eighteenth centuries that such men as Savery, Newcomen and Watt evolved the principles of the engines now in universal use

THE power of steam is one of the most important and widely used motive forces known to man. The first practical attempts to produce a workable steam engine date from the end of the seventeenth century, but ages ago Mediterranean philosophers discovered something of the properties of the power of steam. For example, Hero, a Greek citizen of Alexandria, about A.D. 50 invented an elementary steam turbine.

Lacking contrary evidence, we can claim this as having been the first machine to derive motion from steam power, though the principle of the piston was known to Archimedes in the third century B.C. Hero’s apparatus was simple. He built a closed pot boiler, resembling a covered-in cauldron resting upon legs. From the top he set up two uprights, each ending with a right-angle bend pointing inwards.

THE FIRST STEAM TURBINE is attributed to Hero of Alexandria

THE FIRST STEAM TURBINE is represented by a model in the Science Museum, South Kensington. The discovery of the principle is attributed to Hero of Alexandria, who is believed to have lived in the first century A.D. A closed pot boiler containing water was heated by an open fire. Steam rose into the copper sphere and the force of its exit from the bent pipes in the sphere caused the sphere to rotate rapidly round the horizontal axis. Hero’s engine appears to have been purely experimental.

Between the uprights and revolving freely on the axis formed by them was a hollow copper sphere. One of the uprights also was hollow, and formed a steam pipe communicating with the inside of the sphere. The sphere had two bent pipes leading out of its periphery. Hero found that the force of steam issuing in streams from these two bent pipes caused the ball to rotate rapidly. He raised steam by half filling the boiler with water and lighting a fire beneath it, between the legs of the tripod.

Hero’s engine seems to have been simply an experimental steam toy, probably designed to placate jealous priests who distrusted the philosopher’s experiments, but a modification of it appears to have been used for turning spits in monasteries during the Middle Ages. The Romans improved on Hero’s pot boiler by introducing a central flue, an internal firebox and cross water tubes. This greatly increased the total heating surface, and few modifications would have been needed to convert the Roman boiler into an efficient steam generator.

The Romans, however, were not interested in mechanical engineering, and the water tube boiler known to them was put to no more advanced use than heating water for domestic purposes. Such boilers have been dug out of the ruins of Pompeii, which shows they were in use at least as early as A.D. 79, when that city was destroyed by an eruption of Vesuvius.

The Middle Ages, a dark period of history, were in the main barren of general culture, civilizing influences and scientific experiment. But in the seventeenth century real progress was made. In 1629, or thereabouts, an Italian named Giovanni Branca invented a steam turbine. His principle was that of a steam jet playing on the vanes of a wheel, the steam being generated in a closed boiler. Little practical use was made of this invention. The steam lost its force through condensation almost immediately it had left the jet. Had Branca realized the value of enclosing his vane-wheel he would have gone a long way towards inventing the modern steam turbine.

Various experiments were made in the middle of the seventeenth century, notably by the Marquess of Worcester (1656) and by Sir Samuel Morland a few years later, both of whom claimed to be able to raise water by means of steam. Probably the first application of the cylinder and piston was made in 1678-80 by a Dutch inventor named Christiaan Huygens. The strange thing about Huygens is that he invented, not a steam engine but an internal combustion motor; for to raise his piston he exploded a charge of gunpowder under it. The introduction of the cylinder and piston, however, marked the beginnings of success in the design of steam engines. For a while the principle of the steam cylinder and piston was to remain dormant. Denis Papin was a Frenchman who settled in England in 1675 He was certain he could produce something more practical than the “fire-water-works” of the Marquess of Worcester, the principle of which might have been likened to that of an over-filled kettle boiling over. Papin decided that he could use the power exerted by the atmosphere in driving down a piston under which a vacuum had been formed by condensing steam. Although Papin died in obscure circumstances, his experiments were to pave the way for more significant inventions later on, and he invented a form of safety valve still often used.

The tragic history of Papin is followed almost at once by the more successful stories of two Englishmen. The first of these was Thomas Savery, born at Shilston in Devon, about 1650. Savery was educated as a military engineer, holding the rank of captain when he entered the field of mechanical experiment.

Savery had been a keen student of Papin’s lectures, and was greatly struck with the idea of using atmospheric pressure and the creation of a vacuum from the condensation of steam. He made a number of practical experiments culminating in a patent which he took out on July 25, 1698, for a steam engine designed to pump water out of flooded mines. This service was always one that attracted the early steam engineers. The country was full of rich and promising collieries which had become flooded and thus derelict long before they had been worked out.

Savery described his engine in a pamphlet, entitled The Miner’s Friend, which he published in 1702. Savery’s apparatus was not an engine as we understand the term: though it produced a motive force, it had no moving parts actuated by its own power.

An Early Pumping Engine

There were two boilers, one large and one small, the small boiler acting as a feed-water heater and feed injector to the other. Each had its separate furnace underneath. Beside these boilers were two closed vessels called receivers, connected by stopcocks to a long pipe. The lower end of this pipe was taken down into the water to be pumped, and the upper end acted as a delivery pipe.

Savery found that by admitting steam into one of the receivers, and then cooling it by pouring cold water on the outside, he could cause a certain volume of water to be sucked up the long pipe into the receiver When steam had been admitted to the receiver the water was forced out and discharged through the delivery pipe.

He built the receivers in duplicate so that the action of the machine should be continuous, one forming a vacuum and sucking up the water while the other was driving out the water and being recharged with steam. The pipes were fitted with valves which automatically opened and shut under the movement of the water.

This ingenious machine worked well in model form and on a small scale, but it proved too slow and cumbersome for application to the sterner requirements of a badly flooded mine. One of Savery’s first experimental engines, erected in a Lambeth potter’s workshop before the patent was taken out in 1698, forced water through the delivery pipe at a pressure sufficient to blow the tiles off the roof overhead.


DOUBLE-ACTING ROTATIVE BEAM ENGINE installed in Boulton and Watt’s factory near Birmingham in 1788. The cylinder had a bore of 18¾ in. and a stroke of 4 feet. The timber beam actuated, by a cast-iron connecting rod, the “sun and planet” gearing. A cast-iron flywheel with a diameter of 15 ft. 7 in. and having 296 mortice teeth, drove two pinions from whose shafts the motion was communicated by belts or gearing to forty-three metal-working machines Missing portions of the flywheel have been replaced in wood.

Small examples of Savery’s engine found their way into large mansions and country houses at the beginning of the eighteenth century, for the purpose of lifting water from adjacent wells or ponds to cisterns situated at higher levels. Savery installed such an engine at Campden House, Kensington, round about 1712. It cost £50 to build and raised water to a height of 58 feet. This example was of the simplest description Only a single receiver was fitted, thus giving an intermittent flow, and the boiler also was single, lacking the smaller boiler for heating feed-water. The water was introduced to the boiler through a vertical funnel with a stopcock attached, so it was not possible to maintain a continuous pressure, the boiler having to be recharged with cold water at intervals.

Thomas Savery’s great contemporary was Thomas Newcomen, (1663-1729), who goes down to history as the first man to have made a successful application of steam to a piston and cylinder.

The period of his first experiments coincided with those of Savery, and he too based them on the teachings of Denis Papin. He took the idea of a cylinder and piston from an air-pump which had been invented by a German experimenter in physics. By himself, however, Newcomen was unable to apply Papin’s steam-vacuum-atmospheric principle, as Savery had already taken out a patent for it. The two men decided to go into partnership together. This partnership produced one of those great friendships which stand out, here and there, in the history of modern invention.

The principle of Newcomen’s engine was as follows. He placed his cylinder vertically on the top of a low-pressure boiler, with the admission port between the two. The upper side of the piston inside the cylinder was connected by a chain to one end of a rocker or walking beam, mounted on a fulcrum at the top of the apparatus. The other end of the beam was coupled to the water pump intended to communicate with the flooded workings. Newcomen admitted the steam to his cylinder below the piston, and then made it condense, forming a vacuum, just as Savery had done with the receiver of his first machine.

On the formation of the vacuum the pressure of the atmosphere pushed the piston down to the bottom of the cylinder, dragging down one end of the beam and raising the other, which pulled up the piston of the water pump. Steam was then once again admitted to the cylinder, and the pump rod, which had a heavy weight attached to it, raised the piston to the top of the cylinder to start the cycle again. Both steam cylinder and pump were single-acting, that is, no work was done on one of the two strokes forming the cycle.

Steam did no active work, but merely provided a method of creating a vacuum in the cylinder, thus enabling the natural atmospheric pressure to press down the piston. Newcomen used in his boiler a low steam pressure, little more than that of the atmosphere. These early steam pumps were known as “atmospheric engines”. Their speed, as might be expected, was low indeed, the cylinder performing not more than four to five strokes in a minute. A means for increasing this speed and the corresponding output of the pump was discovered by an accident.

One day, one of Newcomen’s engines was observed to be working at more than twice its usual speed, making some nine strokes a minute. Investigation showed that the leaking packing of the piston was allowing the cold condensing water to penetrate the interior of the cylinder itself, causing a much more rapid rate of condensation. After that a regular practice was made of injecting cold water into the cylinders of atmospheric steam engines, instead of spraying it over the outside, with results that were satisfactory when judged by contemporary standards.

First Automatic Valve Gear

Even so, the working of a Newcomen engine was exceedingly slow, so slow that the opening and shutting of the valves controlling the passage of steam and water to the cylinder was accomplished by hand. There is a story to the effect that on one of the Newcomen engines a young boy named Humphrey Potter was detailed for this job. It could scarcely be described as exciting work and young Potter was of a roving disposition. So that he could have some time for his own amusement, he improvised an arrangement of strings and catches whereby the valves were worked automatically by the beam overhead.

Although Master Potter had evolved this arrangement for his own convenience, it had far-reaching results, for the engine was now working itself while he went fishing. Sooner or later, the engineman discovered what he was doing, and two things happened to Potter without his intending either. He lost his position as engine boy, for there was nothing left for him to do, and he went down to history as the inventor of automatic valve gear. Such is the generally accepted story, although recent researches have somewhat discredited it.

HIGH-PRESSURE ENGINE AND BOILER made by Richard Trevithick in 1811

HIGH-PRESSURE ENGINE AND BOILER made by Richard Trevithick in 1811. This is said to have been the first steam engine to drive a thrashing machine. The boiler is 7 feet long and has a diameter of 4 ft. 6 in. The long connecting rod, attached directly to the piston, passes up to a crank on the shaft of the 10-fcet flywheel. The cylinder has a diameter of 9-£ in. and a stroke of 26 in.

The first regular installation of a Newcomen pumping engine took place at Dudley Castle, Worcestershire, in 1712, but there are records of one which worked in a tin mine at Huel Vor, Cornwall, between 1710 and 1714. It is probable that this one was an early experiment. An outcome of the partnership of Savery and Newcomen was that each got the credit for the other’s inventions, and in more than one old account Newcomen’s invention has been described as Savery’s. Newcomen engines enjoyed a long spell of popularity. Many were built for use in England, France and America during the middle years of the eighteenth century, and continued to appear for years after the death of their inventor in 1729. Savery had died some time before, in 1715. An account published in 1746 described a Newcomen steam-atmospheric engine which with “a cylinder of 47 inches [diameter] at Ludgvamley works in the parish of Ludgvam [near Penzance, Cornwall] making about fifteen strokes a minute, usually drew through pit barrels of fifteen inches diameter, from a pump thirty fathoms deep, about an hogshead at each stroke, that is, fifteen hogsheads of water in each minute.”

The Tyneside collieries were a great field for the use of Newcomen engines. In 1769 there were fifty-seven such machines at work in that area alone, some of them being of considerable size. One at Walker Colliery had a cylinder of 74 in. diameter and 10 ft. 6 in. stroke, supplied with steam from four boilers. Various improvements in the design of Newcomen engines took place during the latter part of the eighteenth century.

The original valve motion was improved by an engineer named Henry Beighton, who devised a much simpler form of gear in which a single rod descending from the beam between the piston chain and the fulcrum actuated levers for opening and shutting the valves. Then the great John Smeaton turned his attention to engines and effected a number of improvements in Newcomen’s design. Smeaton was responsible for the large engines built towards the end of the century, his greatest production of this kind being a pumping engine erected at Chacewater, in Cornwall, in 1775.

Smeaton’s Chacewater engine had a cylinder with a diameter of 72 in. and a stroke of 10 ft. 6 in. The beam was formed out of twenty baulks of timber bolted together. This machine was claimed to exert 150 horse-power and to be capable of lifting 48,330 gallons of water in one hour.

Watt’s Condenser

In the operation of these large atmospheric engines considerable trouble was experienced because of the position of the cylinder on the top of the boiler. The motion entailed a heavy strain on the boiler and furnace, and frequently, when the piston made its upstroke, it tended to carry the whole of the cylinder and boiler with it. To combat this, later designers sometimes placed the cylinder at the side of the boiler instead of on top.

The thermal efficiency of an atmospheric engine was poor. It consumed an enormous quantity of coal to produce a low pressure. Smeaton realized that this could be remedied only by exposing as much heating surface to the fire as possible. He produced a boiler in which the fire, before being discharged up the chimney, passed through a series of “channels” with water between. Smeaton’s boiler, though a considerable improvement on its predecessors, was difficult to keep in repair, and after a while it was superseded.

Meanwhile experimenters were trying to evolve an engine which could give a rotary as opposed to an up-and-down motion. In 1758 Keane Fitzgerald devised a system of cog wheels operated by a rack on the weighted rod from the beam. One cog wheel took the motion of the upstroke and another the motion of the downstroke, each wheel alternately remaining stationary. Nobody appears, however, to have put the invention satisfactorily in practice, and it was by other and simpler means that a rotative motion was later derived from the rocking beam.

James Watt was born at Greenock in 1735. At the age of eighteen he went to London as an assistant to an instrument maker, but ill health made him return to Glasgow as a mechanic and optician. He was of an ingenious turn of mind and was encouraged by Dr. Robinson, by whose influence he became the official maker of mathematical instruments to the Glasgow University. His first real attempt at producing a steam engine took place in 1761, when he fitted a small syringe to the top of a Papin boiler and raised the piston by the direct power of steam alone. In this experiment lay the idea of the high-pressure engine of to-day. Watt, however, was chary of using high-pressure steam and remained so throughout his life. He made no further progress on the lines of the syringe engine and two years elapsed before he gave any further serious attention to the possibilities of steam.

THE CENTRIFUGAL GOVERNOR appears on this cabinet steam engine

THE CENTRIFUGAL GOVERNOR appears on this cabinet steam engine which was once the property of James Watt. Mounted on a vertical axis are two weights which, as the speed of the engine increases, swing out by centrifugal force. As the speed further increases, the weights swing farther out until they cause a collar to lift and close down the throttle valve.

Then, in about 1765, he was asked to repair a small model of a Newcomen engine belonging to the University. The model worked badly, and even with the fire blown up to a white heat could make only a few strokes. It occurred to Watt that the condensing water injected into the cylinder cooled it down to such an extent that much of the steam was wasted before it was filled for the second stroke.

He pondered over this problem for long and then suddenly realized that the correct solution was to condense the steam in a separate chamber, after it had passed through the cylinder. Thus was born the idea of the separate condenser, which was the first milestone of Watt’s road to fame. The steam in the cylinder could now be rapidly condensed without the cylinder itself becoming too cold.

Then he resolved to produce a means for keeping the temperature of the cylinder absolutely constant. This he did by surrounding it with a steam jacket. Events followed one another quickly now. His next step was to use the steam, instead of the pressure of the atmosphere, to force the piston down, the cylinder being closed in at the top and the piston rod passing out through a “stuffing box”, making it completely steamtight. From this it was but a short step to applying steam alternately, at either end, its admission and release being continuously effected by the valve gear. Thus Watt produced the double-acting engine, in which steam alone was the motive power, independent of any atmospheric pressure acting on a piston above a vacuum.

“Sun and Planet” Motion

James Watt was not the father of steam power, as is all too generally believed, but he was the first man to start the steam engine along the lines on which it ultimately developed and he reduced the coal consumption of the pumping engine by 75 per cent. Later he set out to obtain the long-sought-after rotative motion, by which the engine could be applied to purposes other than the pumping of water, such as driving machinery.

Such a machine was, by that time, much needed. In and after 1770 came the inventions of Hargreaves and Arkwright, which were to revolutionize the textile industry. These machines needed an efficient motive power and steam was the obvious one. Watt had patented his engine in 1769, a year before Hargreaves produced the spinning jenny. It was later, however, in 1781-82, that Watt patented his double-acting engine and the means for imparting a rotative motion from the reciprocating masses.

To obtain a rotary motion Watt attached the farther end of the beam to a crank working in connexion with a flywheel. Unfortunately, another inventor had anticipated Watt in the application of the crank to the steam engine, and he was unable, until this patent expired, to apply it to engines for commercial purposes. So he set to work to evolve an alternative. He tried several, but that which really came to the rescue was the gear known as the “sun and planet” motion.

In this type of engine the end of the crankshaft was fitted with a large toothed wheel which engaged with a smaller wheel mounted on the end of the connecting rod, the two wheels being kept in mesh by a connecting link. The large central wheel was the “sun” and the smaller was the “planet” revolving round it. This motion was invented by one of Watt’s assistants,

William Murdock. Watt made many other improvements to the steam engine, including the governor-controlled throttle valve to regulate the flow of steam to the cylinder. The governor consists of a vertical axis rotated by means of a transmission system working off the engine itself.


DRAWING OF SAVERY’S PUMPING ENGINE from his pamphlet, The Miner’s Friend, published in 1702. A high-pressure steam boiler, fed from a smaller one, supplied steam alternately to two receivers. When the steam had expelled the air from a receiver cold water was sprayed on the outside. This condensed the steam and formed a partial vacuum into which water rose from the suction pipe. Steam was again admitted to the receiver and forced the pumped water out by a delivery pipe, the suction pipe being closed by a non-return valve.

Mounted on the axis are two or more weights, which fly out through the agency of centrifugal force as the axis revolves. The faster the revolutions, the farther the weights swing out, lifting up a circular collar sliding on the rod; the collar actuates a rod connected with the throttle valve. Thus when the engine tends to race, the governor automatically lessens the supply of steam. For a long time Watt laboured under great difficulties. He had no private means and had to resort to less important but more remunerative work in the intervals between his experiments with steam. It was not until he met Matthew Boulton that he was able to set up in business as the great producer of industrial steam engines. Then Watt’s engines found their way to every part of the country where industrialism needed them, and the names of Boulton and Watt became household words. The old Newcomen engines vanished and Watt’s invention came to lay the foundations of the industrial revolution. In the ten years from 1782 to 1792, the total trade carried on by Great Britain was doubled.

An important step in the early evolution of the steam engine was taken by that brilliant but erratic genius Richard Trevithick, who developed the high-pressure steam engine. Trevithick was born near Camborne, Cornwall, in 1771, and as he grew up had ample opportunity for watching the working of the Boulton and Watt engines used in the Cornish mines.

His first duties were concerned with mining engineering, but he was fascinated by the possibilities of the steam engine, and set himself to improve on Watt’s work. Just as Watt had been convinced that it was possible to produce something better than the steam atmospheric engine, so did Trevithick consider that a much higher degree of efficiency could be obtained by using steam at a high pressure. A typical Watt engine was an enormous structure. By using high-pressure steam, Trevithick was sure that he could produce a much less cumbersome apparatus, which was at the same time far more economical to work.

Trevithick Champions High Pressure

He received little encouragement, however. Watt himself was openly hostile. He seemed to view high pressure with alarm. The beginning of the nineteenth century saw Trevithick making experiments with stationary and locomotive steam engines. His locomotive was the first true locomotive in the world. Trevithick’s boiler contained a furnace with a return flue, giving what was then a considerable heating surface. He used a single cylinder, with a flywheel on the driving shaft to steady the running. The pressure was sometimes as high as 145 lb. per sq. in., a fact which filled Watt with horror. Trevithick at first used cast-iron plates for his boilers; this was courting disaster in the manufacturing conditions then existing. Yet when the inevitable accident came, neglect on the part of those in charge of the engine was a largely contributory cause.

It took place in 1803, and befell a Trevithick engine at work at Greenwich. The youth who was supposed to watch the engine felt that he needed a little rest from his labour and went off to fish for eels in Plumstead Marshes, leaving a labourer in charge. The labourer knew nothing about engines, and when he saw that this one was racing, he stopped it, leaving the boiler to go on generating steam without an outlet. The boiler exploded, killing three people and so injuring a fourth that he died later.

Boulton and Watt immediately endeavoured to spike Trevithick’s guns by getting the use of high-pressure boilers prohibited by law. The fatal boiler had been 6 feet in diameter and built up of cast-iron plates 1 in. thick. A Government inquiry was held on the accident, but, unusually enough for those days, the inquiring officers acted with scrupulous fairness.

They ascribed the primary cause of the accident to neglect on the part of the engine boy, and issued a report defining the strength of materials that might or might not be used in the construction of high-pressure boilers. Trevithick accepted their ruling gladly enough, and after that used only strong wrought iron for his boilers, also providing them with efficient safety valves to guard against any dangerous excess of pressure.

The work of Trevithick marks a turning point in the story of the steam engine, and its evolution. His introduction of the high-pressure steam engine indicated a new path which, followed up by later engineers, led to the development of the modern steam engine in its reciprocating form. The story of the return to the rotary engines sketched out by Hero and Branca is a recent one, in which the name of the Hon. Sir Charles Parsons stands out beyond all others.

EARLY STEAM ENGINE built by Boulton and Watt

EARLY STEAM ENGINE built by Boulton and Watt in 1777 for a pumping station connected with the Birmingham Canal Navigations. The engine is single-acting, the cylinder having a diameter of 32 in. and a stroke of 8 feet. It was in regular use until 1892 and was worked occasionally until 1898, when it was removed and preserved.

[From part 20, published 13 July 1937]

You can

“James Watt”,

“Machinery of the Cotton Mill” and

“Origin of the Locomotive”

on this website.

You can read about

“The Story of the Locomotive”

in Railway Wonders of the World

Origin of the Steam Engine