"A Practical Hero" Feature Article, February 2004

a practical hero

Or, how an obscure New York mechanic got a steam-powered toy to drive sawmills.

by Frederic A.
Lyman

Perhaps in the first century of the current era, Hero of Alexandria described in his Pneumatica a rotary machine driven by steam. A hollow sphere was connected to a cauldron by a tube that also served as the axle. Jets of steam from two nozzles attached to the sphere at opposite ends of its diameter, perpendicular to the axle, caused it to rotate at a high speed. Historians of ancient technology regard the Hero turbine as a toy, since it probably couldn't have produced enough power except for the most trivial applications. It might have been powerful enough to turn a roasting spit.

Hero's turbine remained a curiosity for more than a thousand years. In the early 19th century, several inventors, including the steam engine builders James Watt and Richard Trevithick in England and Oliver Evans in America, experimented with steam reaction turbines of the Hero type, but without much success. It was in the 1830s that an obscure and now largely forgotten mechanic, William Avery, designed and built a Hero turbine that could manage significant, useful work. It powered several gristmills and sawmills in New York State, and even drove a locomotive.

Avery might have read about Hero's machine, but more likely, he'd seen or heard of a water turbine invented in England about 1740 by Robert Barker. This turbine, called Barker's mill, was used in Europe and America. It worked by action of water flowing down through an axial pipe, proceeding radially outward through two arms, and finally exiting tangentially from holes on opposite sides of the arms.

William Avery brought back the steam in the 1830s with his engine.

In 1747, Andreas Segner of the University of Göttingen described a water turbine similar to Barker's mill, but with six jets instead of two. Segner's turbine drew the attention of the Swiss mathematician Leonhard Euler, who analyzed the flow of an ideal fluid in the rotor passages and determined the torque and power that such a machine could produce.

Euler's theory (1754-56) had little influence on the technology of water turbines in the next hundred years, however. Euler's son, Johann Albrecht, built an improved version of Segner's turbine for experimental purposes, and several other versions of the Barker-Segner-Euler turbine were used to provide power for factories and mills.

James Watt apparently tried to get useful work by running some version of Barker's mill on steam, but without much success. Oliver Evans, who had built one of the first successful high-pressure steam engines, believed Watt failed because he used low-pressure steam and cooled the rotor, causing steam to condense in the arms. Evans wrote that he had tested a 3-foot-long rotary tube at a steam pressure of 56 pounds per square inch, and it turned at speeds of about 700 to 1,000 revolutions per minute.

According to Evans, "It exerted more than the power of two men, and would answer to turn lathes, grindstones, &c., when fuel is cheap." He added, "I have specified and explained it in the patent office," by which he apparently meant the one in Philadelphia, since this was in 1784 and the U.S. Patent Office was not established until 1790. No record of the patent has been found.

'Continually Contriving'

William Avery was born in Herkimer, N.Y., in 1793 and in the early 1800s moved west with his parents to settle on a farm in the town of Pompey in Onondaga County. As a young man, he "was continually contriving water mills," and his skills as a mechanic and millwright were in demand throughout central New York. In 1824, he invented a machine for making wire harness for looms, and his nephew later wrote that thereafter "hardly a year passed without a patent being granted to him."

Ambrose Foster and William Avery were granted a patent on Sept. 28, 1831, for "their improvement in the Steam Engine, commonly called the Reacting Engine."

According to the patent, "What we claim as our invention is, simply, the giving the oblate, or flat, form to the revolving arms, so that in proportion to their capacity, they shall experience much less resistance from the air than that to which they have been heretofore subjected, thereby obtaining a greatly increased power."

Barker's mill improved on the idea of Hero's steam-driven device in the 18th century and used water instead of steam.

Instead of making the arms "in the form of round tubes, which has been heretofore done," Foster and Avery made their cross-section a profile consisting of two circular arcs of the same large radius. They thought that this shape, with its sharp edges, would give "the least possible resistance," but they noted that tubes of elliptical, or oval, cross-section would reduce the resistance almost as much. Although such aerodynamic intuition might seem quite remarkable for mechanics in 1830, it is a somewhat dubious basis for a patent.

The patent was duly noted in the March 1832 issue of The Journal of the Franklin Institute. The editor, Thomas P. Jones, an M.D. who wrote pithy and often caustic comments on the patents, was in this case quite forbearing. He mostly quoted the inventors' claims.

Jones concluded his review of the patent notice with the following statement: "We have been induced to give more room to the notice of this engine than we should otherwise have done in consequence of the information contained in a letter recently received from one of the patentees, that it had been in continued use ever since the issuing of the patent, in a factory owned by him, and had fully justified all his expectations."

A letter dated May 12, 1834, from the Fulton Foundry in Syracuse and printed in a local newspaper announced that it "is under the charge of Mr. William Avery, who is well known as an experienced Millwright and Engineer," and that "the Rotary Engine invented by Mr. Avery ... has been in operation for the last 18 months in their shop."

From Hard Time to Hardware

The owner of the foundry was Elam Lynds, former warden of Auburn Prison who also built Sing-Sing using prisoners as laborers. By 1830, Lynds had retired from prison administration and taken over a hardware store in the rapidly growing village of Syracuse on the Erie Canal. (Alexis de
Tocqueville, who came to Syracuse in 1831 to interview Lynds about prison conditions in America, recorded in his journal an account of their meeting that makes interesting reading even today.)

From 1835 to 1837, several notices about "Avery's Engine" appeared in local newspapers and in mechanics' magazines with wider circulation. Foster's name was seldom mentioned. These sources said that the engine not only provided power for all sorts of machinery in Lynds's shop, but also for several sawmills in central New York. The salt industry in the nearby town of Salina required considerable amounts of firewood for salt boilers and lumber for solar drying sheds, and lumber was also needed for construction in the burgeoning city of Syracuse. There was need for steam-powered mills in the flat northern reaches of Onondaga County.

The high-speed Avery engine was well-matched to sawing lumber. There was no need to gear it up to match the saw speed, as was necessary with the low-speed reciprocating steam engines of the time. (Oliver Evans's high-pressure engine drove some sawmills in Pennsylvania.)

In April 1835, an Avery engine at a sawmill was demonstrated to a group of interested citizens, 13 of whom signed a letter to a local newspaper declaring that it is "much preferable to any other Steam Engine now in use for Milling Purposes." In June 1836, the same newspaper noted that Avery and E. Lynds and Son had "more orders than they will be able to fill for the summer."

The October 1835 issue of the New York Mechanics' Magazine gave more detailed information about "Avery's Rotary Engine," as well as perspective drawings of a "two-horse power engine, built for this office, to drive a printing machine." It also reported that a locomotive engine of this type had been "run for a short time, last spring, on the Newark railroad."

Hero's steam-driven device, which dates back to antiquity, was novel, but hardly a powerhouse.

This article elicited a letter to the editor of the Journal of the Franklin Institute signed "Fair Play," contesting the validity of Avery's patent. The letter, published in the March 1836 issue, suggested that Avery's idea was not novel.

The Journal's editor published Foster and Avery's 1831 patent specification and drawings in full in the April 1836 issue, "in order that the nature and amount of the part claimed may be fully understood."

In early 1837, the last year of its existence, Mechanics' Magazine printed several letters about the use of Avery's engine in gristmills and sawmills. One, from Avery himself, described a gristmill in Cayuga County driven by an engine with a rotor 12 feet in diameter, which made about 1,000 rpm at a steam pressure of 120 psi. Four men from Ithaca wrote a letter about a local sawmill driven by an Avery engine "estimated at 20 horsepower." This letter showed a diagram of the sawmill, which had two reciprocating vertical saws connected to opposite ends of a rocking beam driven by a connecting rod from a drum belted to the engine.

Two other testimonials were printed. One was from the owners and millers at the gristmill and one was from Henry Seymour, one of the commissioners of the Erie Canal about "a sawmill in Cicero, which is propelled by one of Avery's Rotary Engines."

The Vanishing Turbine

It's too bad that we have only newspaper and magazine accounts of Avery's turbine. They rely too heavily on the claims of the inventor, manufacturer, and their friends and customers.

If Avery's turbine was as good as claimed, why did it disappear so soon afterward?

Avery's engine powered mills and even a printing press.

Sir Charles Parsons, the British engineer who in 1884 patented what was to become the first commercially successful steam reaction turbine, gave the Rede Lecture at Cambridge University in 1911. The printed lecture has a photograph of the rotor of Avery's turbine, which was 5 feet in diameter and had a tip speed of 880 feet per second. In Parsons's view, "These wheels were inefficient, and it is not so obvious that an economical engine could be made on this principle." In 1882, Dr. Gustav de Laval had tried to use a Hero-type reaction turbine to drive his high-speed cream separator, but he rejected it and instead developed the impulse turbine that bears his name.

The Avery engine probably had other problems: noise, vibration, the difficulty in sealing the rotary coupling, and the problem of speed regulation. These problems would have been difficult to solve with 1830s technology.

A short retrospective on the Avery engine in The Scientific American of Nov. 19, 1864, described two other problems. The cast rotor of one Avery engine had flown apart, and a piece of it had gone "up through two or three floors with a force equal to that of a cannon shot." The sides and edges of the rotor arms tended to become furrowed and jagged after long use. The article noted that, nevertheless, an Avery engine drove a sawmill in New York City for 20 years, and its proprietors later regretted replacing it with a reciprocating engine.

Avery and his co-filer, Ambrose Foster, provided these sketches of their "Reacting Engine" with their patent application, which was granted in 1831.

The Panic of 1837 hastened the demise of both the engine and its inventor. Avery moved to Chicago that year looking for work and soon contracted with the state to cut the Illinois and Michigan Canal through rock at the summit of its path between Lake Michigan and the Illinois River.

He died in November 1840 and was buried at Rockport, Ill. Elam Lynds & Son went bankrupt in 1842.

The only memorial to William Avery around Syracuse is a New York State historical marker on Route 92 near his family's farm in Oran. It claims that the steamboat he built and launched there in 1823 became the first on the Erie Canal. Avery also built the machinery for the first steamboat on Lake Ontario, as well as for several other lake steamers.

William Avery apparently left no descendants, but one of his nephews, John E. Sweet, gained prominence as a mechanical engineer and served as the third president of ASME (1884-85). In a 1910 talk on the "History of Industrial Syracuse" delivered at the city's Technology Club, Sweet referred to Avery as "the first great inventor of this city."

Sweet estimated that 50 to 75 Avery engines were built and used to run cotton gins and equipment in sawmills and woodworking shops. He also said that the rotor of one of these engines was in the museum of ASME in New York City. The present author would be grateful to learn what happened to it when the museum's collection was dispersed.

Frederic A. Lyman is Professor Emeritus of Mechanical Engineering at Syracuse University in New York. He is also a member of the Society for the History of Technology.