Hot Air Engines.

This gallery is in course of arrangement.
Updated: 22 Dec 2006
Back to Home PageBack to The Museum

There are many websites dedicated solely to hot air engines, and this page does not attempt to supplant them. What it does attempt is to provide some unusual sidelights and some unique images relating to the history of these engines.

There are many kinds of hot-air engine:


Hot air engines use the heating and cooling and expansion and contraction of air to generate power. The hot air engine was once very popular for light duties, largely because of what it did not have; a boiler. It was impossible to have a boiler explosion. Manufacturers were not slow to point this out in their advertisements, also emphasising that skilled supervision was not required. Anyone could stoke a hot-air engine.
Another advantage was that the hot air engine was quiet, as the air stayed inside the engine and there were no gasping or exhaust noises. The engine also required no continuous source of water to keep working.
The great drawback to hot air engines was and is their very limited power-to-weight ratio. The power yielded by heating and expanding air is small compared with that produced by boiling water.

Nowadays the term "hot air engine" is pretty much synonymous with the Stirling cycle in one of its configurations, but history shows that many other approaches, some of them undeniably bizarre, were also tried. Here you can learn about some of them...

According to Knight's American Mechanical Dictionary, 1880, air engines can be put into five classes:

1) Air is compressed, and then heated directly by passing it through a furnace. The hot 'air' then acts against a piston and is discharged to the atmosphere. Since the oxygen is consumed the exhaust cannot be used in a closed cycle.

2) Air is compressed, and then heated indirectly by passing it through pipes heated by a furnace, and then acts against a piston. The exhaust can be returned to make a closed cycle as it is not involved in combustion.

3) Air is held in two reservoirs communicating with each side of a piston. The reservoirs are alternately heated and cooled to apply varying pressures to the piston.

4) The air is mixed with water or steam to reduce abrasion of the moving parts, or in the hope of increasing the power output. See the aero-steam engines. (In preparation)

5) The air or gas exerts its pressure against water which in turn acts against a piston. The rationale behind this indirect method is to keep hot air, cinders etc away from the moving parts. It is also much easier to make a good piston seal for a liquid rather than a gas.

Now this is not a completely satisfactory classification; for example, there is nothing to stop engines in Category 1 or 2 from also being in Category 5. However, it will do for our purposes.


The Stirling cycle.

  • Starting with the air in the hot cylinder, the air is heated and expands. It thus pushes down both pistons and feeds more energy into the flywheel.
  • The hot piston moves upwards, causing the air to move from the hot cylinder to the cold cylinder. As it does so it gives up heat to the regenerator.
  • The air in the cold cylinder cools so less pressure is exerted on the pistons, which move back.
  • The cold piston moves upwards, shifting the air through the regenerator where it warms up, and into the hot cylinder. The cycle repeats.
Crucial to the operation is the 90-degree phase-shift between the pistons, which makes operation hard to visualise. There is a very good animation of an Alpha engine at The regenerator is essential to get reasonable efficiency.

Left: The Stirling hot air engine principle.

This is the Alpha configuration; it has the disadvantage that the sealing arrangements for the hot piston have to be able to cope with hot gas; this would not be a major problem nowadays, but in the early years of hot-air engines it presented serious difficulties. A sizable flywheel is essential.

The whole essence of the hot air engine is that air is heated and cooled as efficiently as possible. The regenerator greatly improves the efficiency. Typically it consisted of a mass of iron wires through which the air passed when the displacer moved it from one end to the other. When the air was displaced from the hot end to the cool end, the regenerator was heated and the air consequently cooled. When the air was moving back to the hot end, it regained the heat temporarily stored in the regenerator.

Stirling engines are classified into three groups, depending on how the power and displacer pistons are connected.

  • The alpha type: there are two pistons in separate cylinders which are connected in series by a heater, regenerator and cooler. This means that the hot piston must have seals, eg piston rings, that can cope with hot gases.

    Left: The Alpha hot air engine configuration.


  • The beta type: A displacer and a power piston in a coaxial cylinder system.
  • The gamma type: A displacer and a power piston in separate cylinders. The Rev Stirling's orginal engine was of this type. Note that the displacer is merely moving the air from place to place- it is not compressing it.

Left: Beta and Gamma hot air engine configurations.

These two configurations use a displacer piston to move the operating gas from the heating zone to the cooling zone and back.

Ringbom engines are an ingenious variation on the Stirling system, invented by Ossian Ringbom in 1905. There is no mechanical link between the power piston and the displacer piston; instead, the displacer "piston-rod" is of a sufficiently large diameter so that changes in pressure inside the engine cause the displacer to be moved between the top and bottom of the cylinder at the appropriate times. In his book "Ringbom Stirling Engines" Dr Senft provides a mathematical analysis of the functioning of the Ringbom engine, and describes the "overdriven" mode of operation, in which stable running is assured by proper selection of engine dimensions.
While this is unquestionably a clever idea, most commercial Stirling engines used mechanical linkages to move the displacer.

Left: A Ringbom hot air engine


The compression type of hot-air engine, which is associated with the Rider company

Left: A Rider hot air engine




Wood Wood 1759.

SIR GEORGE CAYLEY (1773 - 1857)

The English inventor Sir George Cayley, Baronet, is known to have devised air engines around 1807. Letter describing. Sir George Cayley was a remarkable man, probably best known for devising and demonstrating the first man-carrying glider. also the 1837 patent details the application of his air engine to a road carriage.

ROBERT STIRLING (1790–1878) Robert Stirling was a clergyman who invented the hot air engine as we know it; his greatest contribution was the invention of the regenerator. This was a crucial innovation that improved efficiency and specific power output to useful levels.
Stirling came from a family of engineers, but became a minister of the Church of Scotland in 1816. According to some accounts he became concerned about the danger presented by steam boilers, which at the time were prone to explode because of the poor quality of the wrought iron boiler plating available at the time, and decided to improve the design of an existing air engine to provide a safer alternative to steam. Within a year Stirling had invented the regenerator, which he called the "Heat Economiser". This altered the air engine from a curiosity to a practical prime mover. He obtained a patent for the Economiser, and an air engine incorporating it, in 1817. Stirling's engine could not explode because of its low internal pressures, and even if the hot air did suddenly escape it could not cause the horrific scalding that made steam escapes so frequently fatal. In 1818 he built the first practical version of his engine, which was used to pump water from a quarry.
Robert Stirling was assisted by his brother James, who realised that the output of the engine could be greatly increased if it was operated at a high internal pressure, making the changes in force on the power piston greater.

Left: The Stirling hot air engine of 18xx.


The first practical hot air engine was built in 18?? by the Reverend Robert Stirling in 1816, though previous attempts had been made by people like Sir George Cayley around 1807. In 1816 he registered a patent for the an engine and the "Economiser".


John Ericsson (1803–1889) was born in Värmland, Sweden. He is best known for entering his steam locomotive "Novelty" in the famous 1829 Rainhill Trials, and for designing the pioneering ironclad USS Monitor in 1861, but he also spent much effort on what he called Caloric Engines. This is just another name for a hot-air engine.

In 1826 he arrived in England with a working model engine which he called his Flame Engine. This worked with Swedish birch wood as the fuel but burnt out when fired with English coal. The image below is believed to show the gas-generating section of this engine, but this is so far not confirmed.

Above: Part of the first Ericsson caloric engine.

From left to right, this shows the air-pump, the closed furnace and the "boiler". The actual engine is not shown.

In 1833 London Caloric Engine.
Ericsson settled in New York where he built, between 1840 to 1850, eight experimental engines using wire gauze regenerators. These engines worked on an open cycle with external heating and using two pistons of unequal diameters.
In 1851 Ericsson and his financial backers built the Caloric Ship Ericsson, a 260-foot paddle ship. The Caloric engine had four cylinders, each no less than 14 feet in diameter with a 6 foot stroke, but the power output was only BHP. The ship was not a success, and unfortunately sank in a storm off New York. It was raised and then fitted with steam engines.
Ericsson was not disheartened by the failure of the caloric ship, and went on to patent a number of improvements to his engines during the years 1855-1858. These experiments cumulatated with his improved Caloric Engine, an open cycle machine using a power piston and a supply piston, fitted with valves. This engine proved an immediate success with over 3000 being sold with in three years. This machine was sold in sizes of 8 inches to 32 inch cylinder diameters. A Caloric engine of this type was installed at the Pilot Island Light Station to power the fog signal in 1854.
Ericsson became interested in Solar power. Finding that his small caloric engine was not suitable on account of the valves he developed, around 1872, a displacer type (or Stirling) engine to work with a parabolic reflector, the intention being to use it in sunny climates to power irrigation pumps. The solar power application did not take off, but his business backers persuaded him to patent the design in 1880, as a pumping engine heated by coal, wood or gas. The engine was first built by the Delameter Iron Works and later by the Rider-Ericsson Engine Co. in sizes from 5 to 12 inch cylinder diameters. This was the last air engine developed by John Ericsson, who turned to other interests like warships.

Above: Part of the first Ericsson caloric engine.

From left to right, this shows the air-pump, the closed furnace and the "boiler". The actual engine is not shown.


The Philips company of Holland made important advances in Stirling cycle technology. Their development of the Stirling hot gas engine started in 1938, driven by a desire for a versatile electricity generator that would allow increased sales of Philips electronic products in areas with no reliable electricity supply. After the end of World War 2, the first prototypes were presented. One innovation was the rhomboid drive to the power and displacer pistons; another the roll-sock seal.

Left: A Philips Stirling generator set: 1950s.

The engine was fuelled with ; output was 200W.

The market for an electrical generator was reduced by the introduction of the transistor, which made battery-powered radios much more practical. Today, when power is required in remote areas, diesel engines are almost universally used to drive generators.
In 1975 a horizontal 4-cylinder engine was built giving 115 hp at 3200 rpm. The working fluid was helium at the high pressure of 140 bar (2030 psi) and the engine achieved 40% efficiency. Stirling research ceased in the late 1970s. The only enduring commercial product was the Philips Stirling cryocooler, a reversed Stirling engine that produces liquid air.

From Knight: Ericsson Stillman Roper Baldwin Messer Willcox Laubereau Peters Bickford Shearer Kritzer

Resonantly Coupled alpha-Stirling Cooler. Issued on September 29, 1998 as U.S. Patent

Hot-air engines were once widely used for generating small amounts of power. They pumped water, turned fans, animated shop-window displays, drove dental drills and sewing-machines, and even powered gramophones and player-pianos. See The Hot-air Gramophone

Hot air engines typically had long working lives, usually ended by the iron hot end failing. The introduction of modern steels would have cured this, but by the time they were available electric motors had arrived, and the hot air engine lost its market niche.

This device is one for connoisseurs. It is not only a vacuum engine, but a rotary vacuum engine.

There is a large amount of info on the web. Here are a few selected places to look at:
Robert Sier's webpage. Robert Sier is the author of an excellent book on the history of hot-air engines.
A large site on Stirling & vacuum engines, including lots of examples at exhibitions.
A Stirling-powered fan that sits on top of your stove.
The Stirling Engine webring
Example of a Ringbom Stirling engine
Another Ringbom Stirling engine. Beautiful.
A source of model plans and kits
A large collection of links, some of them dead, unfortunately.
The site of Buster Brown of Yuma, Arizona. Has a large picture gallery.
A very nice site; includes some deep theory provided by Dr. Israel Urieli of Ohio University.
Nice pics of Kyko, Ericsson, and Heinrici engines.
A fine site: kits to build various Stirling engines, pictures, simulation programs, etc.


  • No boiler so no potential to explode.
  • Can run on almost any burnable fuel. Combustion is continuous so emissions ca be minimised.
  • No supply of working fluid required. (assuming you're sticking with air)
  • The bearings and seals are usually on the cool side of the engine, improving reliability.
  • Quiet in operation.
  • Low specific power output.
  • Significant warm-up time before starting.
  • Not self-starting.
  • Cannot change speed quickly. Typically, changes in output are achieved by varying the cylinder displacement, (often with a swashplate crankshaft mechanism) by changing the amount of working fluid in use, or by altering the phase angle between piston and displacer.
  • The heat exchangers/regenerators are relatively complex and costly.
  • Advanced Stirling engines use helium as the working fluid, because it approaches the efficiency and power density of hydrogen with less containment problems. (hydrogen will diffuse through solid metal) Helium make-up must be supplied from a gas bottle; this is complicated and expensive.

It is often thought that the great danger with a steam boiler is explosion due to excessive boiler pressure. This is not so. That can be prevented by a simple spring-loaded safety valve, which operates quite automatically- assuming of course it has not been tampered with or incorrectly assembled and is not corroded shut.
The more deadly danger is explosion due to lack of water. If the firebox of a boiler becomes uncovered by a falling water level, it will rise in temperature to the point where the metal loses strength, and the firebox collapses. The effects can be no less fatal and destructive than a boiler explosion provoked by any other means, and automatic safeguards are much harder to provide.

Back to Home PageTop of this  pageBack to The Museum