In 1698, the English mechanical designer Thomas Savery invented a steam pumping appliance that drew water directly from a well by a vacuum, then sent it up to a higher level by steam pressure. The appliance was also proposed for draining mines, but limited pumping height made this impracticable. It also consumed an inordinate amount of fuel.

thumb|The model Newcomen engine upon which Watt experimented The solution to draining deep mines was found by Thomas Newcomen who developed an „atmospheric“ engine working only on the vacuum principle. It employed a cylinder containing a moveable piston connected by a chain to one end of a rocking beam that worked a mechanical lift pump from its opposite end. The top side of the power cylinder was open to the atmosphere, steam being introduced at top stroke to the underside of the piston then water sprayed in, condensing the steam and creating a vacuum; thus atmospheric pressure acting on the upper side of the piston drove it down.

It was both powerful and useful and for the first time water could be raised from a depth of over 150 feet. The first example from 1711 was able to replace a team of 500 horses that had been used to pump out the mine. In the next fifty years only a few small changes were made to the basic engine, seventy-five examples of which were at mines in Britain, France, Holland, Sweden and Russia.

The system brought great practical benefits, but still at the price of very high coal consumption as the water jet into the cylinder cooled the walls at each stroke; this meant that when the next charge of steam was introduced it would continue condensing until the cylinder approached working temperature once again.

A Scottish instrument maker, James Watt, was given the job in 1763 of repairing a model Newcomen engine for the University of Glasgow, and noted how inefficient it was. In 1765, while wandering across Glasgow Green he conceived the idea of a separate condensing chamber for the steam engine. Watt's idea was to separate the condensation system from the cylinder, injecting the cooling water spray in a second cylinder, connected to the main one. When the piston had reached the top of the cylinder, the inlet valve was closed and the valve controlling the passage to the condenser was opened. External atmospheric pressure would then push the piston towards the condenser. Thus the condenser could be kept cold and under less than atmospheric pressure, while the cylinder remained hot.

Watt also realised that the new operating cycle might increase engine speed and the power produced. Suppose low pressure steam could now be substituted for atmospheric pressure? If the top of the cylinder was closed off, the steam could act upon the piston during the power stroke; the low steam pressure would not be sufficient to move it in normal circumstances, but it could if acting upon a vacuum.

thumb|left|300px|The major components of a Watt pumping engine. This led to the fully developed version of 1775 that actually went into production <ref>Hulse David K (1999): „The early development of the steam engine“; TEE Publishing, Leamington Spa, U.K., ISBN, 85761 107 1 p. 127 et seq.</ref>. There was no spray, the condenser being immersed in a water tank and at each stroke the warm condensate was drawn off and sent up to a hot well by a vacuum pump which also helped to evacuate the steam from under the power cylinder. The still-warm condensate was recycled as feedwater for the boiler.

The separate condenser showed dramatic potential for improvements on the Newcomen engine but Watt was still discouraged by seemingly insurmountable problems before a marketable engine could be perfected. It was only after entering into partnership with Matthew Boulton that such became reality. Watt told Boulton about his ideas on improving the engine, and Boulton, an avid entrepreneur, agreed to fund development of a test engine at Soho, near Birmingham. At last Watt had access to facilities and the practical experience of craftsmen who were soon able to get the first engine working. As fully developed, it used about 75% less fuel than a similar Newcomen one. In order to be able to recoup substantial development costs, Boulton and Watt licensed the idea to existing Newcomen engine owners, taking a share of the cost of fuel they saved.

Driving the engines by the pressure differential between low-pressure steam and a partial vacuum raised the possibility of reciprocating engine development<ref>Hulse David K (2001): „The development of rotary motion by the steam power“; TEE Publishing, Leamington Spa, U.K., ISBN, 1 85761 119 5 : p 58 et seq.</ref>. An arrangement of valves could admit steam to either end, or connect either end with the condenser. Consequently, the direction of the power stroke might be reversed. The resulting double action gave a very even movement to the beam and made possible the development of rotative engines. Furthermore, the linkage to the beam had, until then, been by means of a chain, which meant that power could only be applied in one direction, by pulling; it had to be made possible for the piston to also push the beam whilst keeping the piston rod vertical; this Watt achieved by developing his parallel motion.

In order to avoid patent rights already claimed by another party, on the use of the crank, he adopted the epicyclic sun and planet gear system suggested by an employee William Murdoch, only later reverting once the patent rights had expired to the more familiar crank seen on most engines today.

Because factory machinery needed to operate at a constant speed, Watt adapted the centrifugal governor (earlier used to automatically control the speed of windmills) linked to a steam regulator valve.

These improvements allowed the steam engine to be used to replace water wheels, thereby freeing British industry from geographical constraints and becoming one of the main drivers in the industrial revolution.

Watt was also concerned with fundamental research on the functioning of the steam engine. His most notable measuring device, still in use today is the Watt indicator incorporating a manometer to measure steam pressure within the cylinder according to the position of the piston; this enabled a diagram to be produced representing the action of the steam throughout the cycle.

The oldest working engine in the world is the Smethwick Engine, brought into service in May 1779 and now at Thinktank in Birmingham (formerly at the now defunct Museum of Science and Industry, Birmingham). The oldest still in its original engine house and still capable of doing the job for which it was installed is the 1812 Boulton and Watt engine at the Crofton Pumping Station. This was used to pump water for the Kennet and Avon Canal; on certain weekends throughout the year the modern pumps are switched off and the two steam engines at Crofton still perform this function. The oldest rotative steam engine (the third rotative engine ever built) is located in the Powerhouse Museum in Sydney, Australia.

The Henry Ford Museum in Dearborn, Michigan houses a Watt rotative engine manufactured in 1788 by Charles Summerfield. This is a full-scale working replica of a Boulton-Watt engine. The American industrialist Henry Ford moved the engine to Dearborn around 1930.

*Carnot cycle *Heat engine *Thermodynamics *James Watt *Newcomens dampmaskine

<References/>

*[http://www.egr.msu.edu/~lira/supp/steam/wattengine.htm Watt atmospheric engine] - Michigan State University, Chemical Engineering *[http://www.newcomen.com/excerpts/watt.htm Watt's 'perfect engine'] - excerpts from Transactions of the Newcomen Society. *[http://www.powerhousemuseum.com/collection/database/?irn=7177 Boulton and Watt Steam Engine at the Powerhouse Museum, Sydney]