Newcomen atmospheric engine

• 1. Introduction

  ›

  In 1712, Thomas Newcomen built a machine that had never existed before: a device that turned steam into useful mechanical work. He erected it at the Conygree Coalworks in Bloomfield Road, Tipton, in the Black Country, and it pumped water out of a flooded coal mine. That single engine, clanking and hissing at around twelve cycles per minute, set off a chain of events that would reshape how human beings thought about power itself.

  Newcomen was not a scientist. He was an ironmonger, a metal merchant who dealt in the practical stuff of forge and foundry. He came to the problem of flooding mines not as a theorist but as someone who knew what materials could bear what loads. That background turns out to have been exactly the right preparation for what he was about to build.

  The engine he produced was surprisingly intricate for something built on a deceptively simple idea. It used atmospheric pressure, not steam pressure, to do its work. Steam filled a cylinder, cold water was sprayed in, the steam collapsed into a near-vacuum, and the weight of the air above drove a piston downward. That piston was attached by chain to a great rocking beam, and the beam drove pump rods deep into the mineshaft below.

  Hundreds of these engines were constructed across Britain and Europe during the eighteenth century. James Watt would later improve on them, and Watt's name would become the one history remembered. But the story of how steam first went to work in the world belongs to Newcomen, and to Tipton in 1712.

• 2. What Came Before

  ›

  Around 1600, a cluster of experimenters across Europe began playing with steam in ways that were clever but largely impractical. They built devices that worked like oversized coffee percolators: a sealed container of water was heated until steam pressure forced liquid up through a submerged pipe and out of a nozzle at the top. The principle was real. The usefulness was limited.

  In 1606, the Spaniard Jerónimo de Ayanz y Beaumont went further. He demonstrated a steam-powered water pump before witnesses and was granted a patent for it. His pump was put to practical use draining the flooded mines of Guadalcanal, Spain, which makes it one of the earliest documented cases of steam doing genuine industrial work.

  Edward Somerset, the 2nd Marquess of Worcester, published a book in 1662 outlining several inventions he had been developing. One was a steam pump that alternated between vacuum and steam pressure across two containers, allowing it to deliver water continuously to a raised header tank. The design had conceptual sophistication, but Somerset never scaled it into anything that worked at a mine.

  Thomas Savery took the next step in 1698, patenting a device he called the Miner's Friend. It had no moving parts and could not transmit power to any external mechanism, so calling it an engine is generous. A practical ceiling also haunted it: a vacuum alone could only lift water roughly 30 feet, and adding steam pressure gave perhaps another 40 feet. For a deep mine, that was nowhere near enough. When a boiler at Wednesbury exploded, perhaps in 1705, the dangers of pressurized operation became grimly clear. Parliament had extended Savery's patent by 21 years, meaning it would not expire until 1733, a legal fact that would later force Newcomen into an awkward arrangement.

• 3. The Papin Connection

  ›

  Denis Papin published a paper in 1690 in the journal Acta eruditorum in Leipzig, titled in French as a method for cheaply obtaining considerable forces. The idea came to him while working with Robert Boyle at the Royal Society in London. Papin described pouring a small quantity of water into a vertical cylinder, inserting a piston on a rod, driving steam upward to raise the piston, then removing the heat so the steam condensed and created a vacuum beneath. When a spring catch was released, the pressure differential between the atmosphere above and the vacuum below pulled the piston sharply down. The force generated was enough to lift a 60-pound weight.

  The experiment was elegant and real. What Papin had not solved was how to make it repeat automatically at regular intervals. A demonstration is one thing; a working engine is another. Several of his papers were put before the Royal Society between 1707 and 1712, including a description of this 1690 atmospheric device.

  Newcomen took that experiment and made it continuous. He added a boiler capable of supplying a steady stream of steam, a system for condensing that steam reliably, and a method for removing the condensed water afterward. Where Papin had a single dramatic stroke, Newcomen had a cycle. One historian, David Wootton, has argued in his work The Invention of Science that Newcomen's specific source may have been a Papin brochure on a bone-softening digester, not the 1690 atmospheric paper itself. The precise path of influence remains uncertain. What is certain is that Newcomen's engine worked and Papin's never left the laboratory.

• 4. How the Engine Worked

  ›

  The boiler at the base of a Newcomen engine produced steam at extremely low pressure, no more than 1 to 2 pounds per square inch. Early versions were made of copper with a lead dome; later ones were assembled from small riveted iron plates. The steam rose into a cylinder above, open to the atmosphere at the top. Early cylinders were cast brass, but cast iron proved cheaper and more effective. The first Newcomen engine to use cast iron was installed at Hawarden in Wales between 1714 and 1715. Foundries including Coalbrookdale and the Carron ironworks supplied and bored these iron cylinders.

  A piston sat inside the cylinder. Its underside was bolted with wood to insulate it from the steam below, which it would otherwise prematurely condense. To seal the gap between piston and cylinder wall, early engines used a leather ring; because the bore was finished by hand and not perfectly round, a layer of water was kept on top of the piston at all times. Later soft hemp rope packing replaced the leather, weighted down with iron.

  To start a power stroke, the operator opened the regulator valve to let steam fill the space below the piston. The valve was then closed, and the water injection valve was briefly snapped open, sending a cold spray into the cylinder. The steam collapsed into a vacuum. Atmospheric pressure above the piston then drove it down, the power stroke, which pulled one end of the great rocking beam down into the engine house and lifted the pump gear on the opposite end outside. The cycle repeated around twelve times per minute.

  A persistent early problem was that air dissolved in the water came out with the steam during boiling and could not be condensed. This air accumulated until the engine became what operators called wind logged and stopped working. The solution was a release valve near the bottom of the cylinder called a snifting clack or snifter valve, which briefly opened when steam was first admitted and purged the non-condensable gas. Its name came from the noise it made, which someone at the time compared to a man snifting with a cold.

• 5. Automation and Scale

  ›

  Early Newcomen engines required a plug man, a worker whose entire job was to open and close the valves by hand in precise and repetitive sequence. The timing had to be exact, which was exhausting and error-prone. The solution was a plug tree, a beam suspended vertically alongside the cylinder that used tappets and escapement mechanisms to open and close valves automatically as the main beam reached certain positions. It was, in effect, the first form of valve gear.

  A legend attached itself to this innovation. According to the story, in 1713 a cock boy named Humphrey Potter, who was employed to work the valves on one engine, grew bored and rigged cords and catches so the beam would do his job for him. This tale was widely repeated. The evidence, however, suggests the plug tree was established practice before 1715 and is clearly visible in the earliest known images of Newcomen engines, drawn by Henry Beighton in 1717 and Thomas Barney in 1719. Potter's cleverness may have been real; whether he invented the idea is another matter.

  Even the engine's restart after running low on steam was automated. A buoy rose and fell in a vertical standpipe attached to the boiler. When steam pressure dropped, the buoy fell and operated a weighted lever called the scoggen, which blocked the water injection valve until enough steam had built back up.

  By 1725 the engine was in common use in collieries across Britain. Its reach extended beyond coal. The first Newcomen engine in France was built at Passy in 1726 to pump water from the Seine to Paris. At Coalbrookdale, a horse-powered pump that had returned water to the pool above the Old Blast Furnace since 1735 was replaced by a Newcomen engine in 1742-3. Richard Arkwright used one to supply additional power to his cotton mill.

• 6. Watt's Intervention

  ›

  The central inefficiency in Newcomen's design was thermal: after the cold water spray condensed the steam and created the vacuum, the cylinder walls were chilled. When steam was admitted for the next stroke, a portion of it immediately condensed on those cold walls rather than filling the cylinder. Fuel was being burned simply to reheat iron. Efficiency improved as engines grew larger, because the ratio of surface area to volume shifted in the engine's favor. Inside a colliery, where waste coal was freely available as fuel, this waste was acceptable. Elsewhere it was costly.

  James Watt encountered the problem in 1769 when Glasgow University asked him to repair a small model of a Newcomen engine. The model's small size exaggerated the heat-loss problem. His solution was to move condensation out of the main cylinder entirely, into a separate exterior condenser unit connected by a pipe. Opening a valve evacuated the lower part of the cylinder without ever cooling the cylinder walls. Fuel consumption dropped dramatically. The design also allowed a double-acting cylinder, with power strokes in both directions, increasing output without a proportional increase in size.

  Watt could not manufacture his improved engines until 1774, when John Wilkinson developed an accurate cylinder-boring method precise enough for Watt's tighter tolerances. In 1776 Watt wrote to John Smeaton that Wilkinson had improved the art of boring cylinders so that a 72-inch cylinder would not deviate from true by more than the thickness of a slim sixpence in the worst part.

  Watt's patent on the separate condenser ran until 1800. During those years his vigorous enforcement of patent rights actually kept many Newcomen engines in operation, because operators chose to pay no royalties at all rather than accept Boulton and Watt's terms. When the patents finally expired in 1800, the rush to adopt the separate condenser was immediate, and most remaining Newcomen engines were adapted or replaced.

• 7. Engines That Survived

  ›

  The Newcomen Memorial Engine stands today in Dartmouth, Newcomen's home town, having been moved there in 1963 by the Newcomen Society. It is believed to date from 1725, when it was first installed at the Griff Colliery near Coventry.

  An engine installed at a colliery in Ashton-under-Lyne around 1760, known locally as Fairbottom Bobs, is now preserved at the Henry Ford Museum in Dearborn, Michigan.

  The only Newcomen-style engine still in its original location is at what is now the Elsecar Heritage Centre, near Barnsley in South Yorkshire. It ran commercially from 1795 until 1923, making it likely the last of its type in active use. Conservation work on the engine, its original shaft, and the engine house was completed in autumn 2014.

  An unusual example is on display at Summerlee, Museum of Scottish Industrial Life. Originally used at Farme Colliery, it was employed for winding rather than water pumping, and had been operating for nearly a century when it was examined in place in 1902. A second static example is in the National Museum of Scotland, formerly at Caprington Colliery at Kilmarnock, and another is in the Science Museum in London.

  In 1986 a full-scale operational replica of the 1712 engine was completed at the Black Country Living Museum in Dudley, believed to stand a couple of miles from the site of the original. Its cylinder is more than 2 metres long and 52 centimetres in diameter. It is the only full-size working replica in existence, and visitors can watch it repeat, stroke by stroke, the same atmospheric cycle Thomas Newcomen demonstrated a few miles away in 1712.