Francis C. Moon

Joseph Ford Professor of Mechanical Engineering

July 1999

Introduction

I will describe one of the largest remaining 19th century collections of kinematic models in North America and perhaps in the world, namely the Reuleaux models at Cornell University. Collections of Reuleaux models were widely used in Europe, especially in Germany, before the Second World War. Most and maybe all have been lost in the destruction of 1941-45. The Cornell Reuleaux Collection seems to be the last remaining large set of 219 models out of the original 800 that Reuleaux had built in his laboratory in Berlin over a century ago.

In this paper I will review Reuleaux's contributions to kinematics in the context of these models. I will also describe how Cornell University acquired its collection through the vision of its first president, Andrew Dickson White. This acquisition took place amidst a crisis in mechanical engineering education at Cornell which involved ASME's first president Professor Robert H. Thurston of Stevens Institute of Hobocken, New Jersey. I will also describe the current state of the Reuleaux collection at Cornell and a smaller collection of Darmstadt models and a plan for making the models available on the web through construction of a virtual kinematic museum.

As a recently appointed curator of the kinematic models collection at Cornell, I found that I had to learn a lot about the history of kinematics and nineteenth century machines in order to understand the role of the Reuleaux models in teaching kinematics. As a result I realized the shortcomings in my knowledge of kinematics and the history of technology which the reader might discover and I hope will understand.

Franz Reuleaux was born in Eschweiler (near Aachen), Germany on September 30, 1829. He is regarded as the founder of modern kinematics and as one of the forerunners of modern design theory of machines. [A brief biography in English may be found in Gillispie (1970). Another excellent, short biography of Reuleaux has been written by Ferguson as an introduction to the 1963 Dover edition of Reuleaux's 1875 book on kinematics. A separate annotated list of references to Reuleaux in German is provided in the Appendix.] At the time that Cornell's President A.D. White acquired the Collection in 1882, Reuleaux was rector of the Technische Hochschule at Charlottenburg near Berlin. Earlier he was professor of machine design at the Swiss Federal Polytechnic Institute in Zurich where he developed many of his ideas in kinematics. Of his two major books, Der Constructor and Theoretische Kinematik; Grundzüge einer Theorie des Maschinenwesens, the second had a wider impact. It was immediately translated into English by Professor Alexander B.W. Kennedy of Great Britain titled, The Kinematics of Machinery; Outlines of a Theory of Machines (Reuleaux, 1876), after being translated into French and Italian. This pioneering book dealt with both analysis and synthesis of machine kinematics. Kennedy also wrote a book review on Reuleaux's work in the London based journal, Engineering (Kennedy, 1876a). In the earlier sections of his book Reuleaux states that he is interested in how the mind creates new inventions as well as kinematic analysis. Some of his ideas are captured in the quotes in Table 1 below.

Reuleaux was one of the first to use an abstract symbol representation of machines. He is also credited with inventing the idea of a kinematic pair (Kennedy, 1881). Each pair had a different symbol, and each mechanism would be described by a collection of symbols or word. A complete assembly of mechanisms is then a sentence of words in this symbol language.

This abstract methodology for kinematic mechanisms did not propagate in later textbooks in the early 20th century. However, the idea of symbols for machine elements has been renewed in modern design synthesis theory (see e.g., Hartenburg and Denevit, 1964). Reuleaux's contributions to kinematics have recently been recognized in a recent review of modern developments in kinematics (Erdman, 1993). Quotes on Reuleaux from modern practitioners are given in Table 2.

Besides his influence on kinematics, Reuleaux was active in the technological politics of the newly united German Empire. He had an important role in developing German patent legislation and with the founding of the Mannesmann steel works, as well as a member of juries of international exhibitions. Reuleaux was also active in improving the quality of German manufacturing. [See the short biography by Otto Mayr in Gillispie (1970).]

Nineteenth century kinematics flourished because of the need for machine inventors to transmit information and forces (power) from one element in the machine to another. Thus, we see the names of James Watt (1736-1819) and Rankine (1820-1872) among earlier inventors of mechanisms for the new steam and water based machines that revolutionized the 19th century. Also, water and steam power generated circular motions and the need arose to convert these steady circular motions into nonsteady linear and curvilinear motion for machine applications. The challenge to create input-output kinematic devices that could convert circular motion into noncircular, complex, three-dimensional, intermittent motions attracted both practical inventors as well as mathematicians in the nineteenth century. Analysis of mechanisms attracted mathematicians of the likes of Ampere (1775-1836) and Chebyshev (1821-1894) (Ferguson, 1962). Thousands of mechanisms were invented, designed and built in the last century which helped nurture the widespread use and manufacture of machines analogous to the plethora of electronic circuits in the early 20th century and software in the late 20th century.

In the mid 19th century, Franz Reuleaux embarked on a method to codify, analyze and synthesize kinematic mechanisms so that engineers could approach machine design in a rational way. To this end, he created at Berlin a collection of over 800 models of mechanisms based on his book. Reuleaux's models were apparently influenced by a model collection of his former professor at Karlsruhe, F.J. Redtenbacher [see footnote 37 in Ferguson (1977).] At the time (c. 1876) he authorized a German Company, Gustav Voigt, Mechanische Werkstatt, in Berlin, to manufacture these models so that technical schools could use them for teaching engineers about machines. From Reuleaux's letter to A.D. White in 1882, Voigt had apparently worked for Reuleaux or had been a student when Reuleaux taught at the Gewerbe Schule in Berlin.

By 1907 some 368 were available in the Voigt catalog and the Cornell collection was reported to number 266 items (Hartenberg and Denavit, 1964). The present inventory as of December 1998 is 219 models from the Voigt catalog of the Reuleaux kinematic models. The classes of models are listed in Table 3. Because of widespread destruction in Germany in World War II, it is believed that Cornell has the largest remaining collection of Reuleaux models. A curator in mechanical engineering at the American Museum of Industry at the Smithsonian Institution said he did not know of any other large collections in the United States or Canada.

Reuleaux's models go beyond the collection of kinematic pairs as in Christopher Polhem's "mechanical alphabet" models of 1729 now at the Teknista Museet in Stockholm. There are numerous Reuleaux models of complete machines such as eight fully operating clock escapements, and several complex winch machine models. The clock escapements have as many as fifteen moving parts, constructed from over two dozen manufactured machine elements.

The models are clearly designed for teaching kinematics. Some are demountable so that a different link can be fixed. Some are inscribed with letters and numbers on the links corresponding to Reuleaux's books. Many have adjustments built in so the user can find the optimum setting for proper motion.

At the time of Reuleaux's book on theoretical kinematics in 1875, he began to travel abroad, and became a member of juries of world exhibitions. Of particular interest are the world exhibitions in Vienna (1873) and Philadelphia (1876) where Reuleaux served as the German commissioner (Gillispie, 1970).

These are several documents in the Cornell University Library archives that confirm that Cornell acquired the collection in 1882 or thereafter.

Second, the minutes of the Cornell University Board of Trustees, June 14, 1882, "Acknowledges a pledge of $8,000 from the Honorable Hiram Sibley of Rochester to secure the duplicate of the Reuleaux models in the possession of the Imperial Government of Germany." (Hiram Sibley and Ezra Cornell both formed the Western Union Telegraph Company in 1854.)

Concerning the International Exhibition of 1876 at Philadelphia there is evidence in the reports of the U.S. Centennial Commission (Walker, 1877, 1878) that A.D. White, Reuleaux, and Robert H. Thurston were all there at some time during the fair. The Exhibition was open for six months. Over ten million people visited the fair to see over 30,000 exhibitors from all over the world. Reuleaux is listed in the volume on Group XXVIII, (p. 203) Education and Science as a Commissioner-General for the German Empire and having assisted the judging of this group. The official list of judges include Andrew D. White of Ithaca, New York. However, in the list of awards, no mention is made of Reuleaux's models nor is he listed in the overall index of awards in the multi-volume report on the Exhibition.

Robert Thurston is mentioned in Group XXV on Instruments of Precision, Research Equipment, etc. as having exhibited his Machine for Testing Strength of Materials. He is identified as a professor at Stevens Institute of Technology. This machine, or a copy of it, is likely the one now in the strength of materials laboratory of the Department of Theoretical and Applied Mechanics in Thurston Hall at Cornell University. This Group had judges which included Joseph Henry, the discoverer of electromagnetism who was then Secretary of the Smithsonian Institute in Washington, D.C. In Reuleaux's letter to White (see Document 1), he mentions the Smithsonian, but not Henry.

Thus, it is possible that Reuleaux met A.D. White and, perhaps, Thurston in Philadelphia. We have no direct evidence so far of how President White learned of the Reuleaux models though perhaps he saw them in London. There is evidence that 300 of Reuleaux's models were actually on display at the Exhibition of Scientific Apparatus at the South Kensington Museum in London in September 1876 (Kennedy, 1876). In the Cornell Archives on A.D. White, there is also evidence that White had traveled to Europe in the Fall of 1876. Thus, it is possible he saw the Reuleaux models at South Kensington. The South Kensington Museum was the predecessor of the Victoria and Albert Museum built in Kensington around 1900. There is a wonderful little book by Professor A.B.W. Kennedy of University College, London with a 19 page introduction by Robert Thurston (Kennedy 1881). The book title is The Kinematics of Machinery: Two Lectures Relating to Reuleaux Methods. These lectures (88 pages) were given by Kennedy at the Museum. Kennedy describes Reuleaux's theory of kinematic pairs and his symbol representation of complex mechanisms. This small book illustrates the high esteem that Reuleaux was held both in Europe and the U.S. and the relation of his theory to his models. (Kennedy became the President of the Institute of Mechanical Engineers in Great Britain and Thurston was the first president of ASME.) Kennedy mentions the loan to the Museum of the Kinematic Collection of the Gewerbe-Academe in Berlin, designed by Reuleaux. He also mentions a set of models at Dresden as being essentially the same as the Berlin models.

Along with the Voigt catalog of 1907 and the English translation of Reuleaux's book on kinematics by Kennedy, there is a set of 40, 5 x 7 inch cards, with descriptions of various mechanisms in the Cornell Collection. The cards reference Reuleaux's Kinematics of Machinery as well as other books on kinematics in the late 19th and early 20th century. The cards were written around 1940's-1950's by Harvey Roehl then a student in mechanical engineering. There is no other complete description of the 219 models although we hope to assembly one as part of a new Library effort to catalog the colleciton and to create a virtual museum on the web.

The following descriptions of mechanisms are based on the material in Reuleaux's books. The last item is from the 1962 Smithsonian article by Eugene Ferguson on the history of mechanisms and Reuleaux.

• The pump mechanism (F4) known as Ramelli's Rotary Pump (Figure 3) was first described in 1588 by the Italian military engineer, Agostino Ramelli (1531-1610). This mechanism found application in mid twentieth century household refrigerator compressors (Reuleaux, 1876, p. 365).

• The Pappenheim Chamber Wheel (I1) (Figure 4) is the forerunner of modern gear pumps. Versions of this mechanism have been in existence for over 350 years (Reuleaux, 1876, p. 403).

• The Isosceles Slider Crank Train (S2) (Figure 5) was used in an early steam engine in 1816. This mechanism can be used to form ellipses. Its invention has been attributed to Leonardo da Vinci (1452-1519) for machining elliptical surfaces.

• The Crown-Wheel Escapement (X1) (Figure 6) is associated with the Dutch scientist, Christian Huygens who designed the first pendulum clock in 1657. This mechanism allowed the weight of a hanging mass to regularly impart energy to the oscillating pendulum. The Cornell Collection contains eight of these beautiful clock escapement mechanisms. They are the most elaborate devices in the set and can run for a short time by suspended weights. They are the jewels of the Collection.

• The Pin-Wheel Escapement (X4) (Figure 7) was invented by Amant in 1741. It was employed in French made pendulum clocks. It was also used in tower clocks.

• Chebyshev Mechanism (S19) (Figure 8). For a period in the early 19th century, there was a concerted effort to invent a mechanism that would convert rotary motion into a straight-line motion. This quest occupied mathematicians as well as practical mechanikers. The famous Russian mathematician, Professor P.L. Chebyshev (1821-1894), of the University of St. Petersberg invented his own straight-line mechanisms. Such devices were important in metal planing machines and textile manufacturing.

During a sabbatic leave in 1995 at the University of Hannover in Germany, I met Professor Reinhard Braune, a professor of kinematics and computer aided design. He proudly showed me about ten models from the original Reuleaux collection. (These were from Reuleaux's own laboratory, not Voigt's reproductions.) He told me virtually all the collections in Germany were destroyed. However, from recent correspondence with the Deutsche Museum in Munich, it appears that they may have about hundred models, though they could not send an itemized list. In Hannover, they have reproduced several models and displayed them so they can be actuated. The display also includes modern applications of kinematic mechanisms in manufacturing, automotive design and aerospace technology.

There is also a wonderful modern kinematic model collection of about 300 devices at the Technische Hochschule Aachen under Professor Dittrich. However, they do not have any Reuleaux models. (Their web site also describes all the institutes of kinematic research in Germany.)

Professor Braune was kind enough to give me copies of articles on Reuleaux in German (see Appendix). These articles, dating from around 1899-1940, suggested there were other North American Reuleaux collections at McGill University in Montreal, and Stevens Institute in Hoboken, N.J. Inquiries at McGill indicate that their collection, if it existed in any great number, is gone. One colleague remembers seeing one model several years ago. A call to the curator at the Stevens Institute Library revealed that they have, at most, a few and perhaps only one on display in their library. Also from a phone conversation, the Smithsonian does not have any Reuleaux models either. Professor Dittrich in Aachen thought there was also a collection in St. Petersberg in Russia. However, through his conversation with Professor Artobolesvsky, he believes they were also lost during the War.

In addition to the Reuleaux models, Cornell has about ten old kinematic models by Professor Schröder of the Polytechnisches Arbeits Institute, Darmstadt. These models were cited for an award at the Philadelphia Exposition of 1876 in the section judged by A.D. White of Cornell and Reuleaux. We are currently trying to obtain a biography on Schröder.

In 1878 Sweet decided to leave Cornell which prompted a crisis over the teaching philosophy. A Trustee Committee was set up to reform the mechanical engineering program helped by a generous gift from a successful businessman, Hiram Sibley of Rochester. It was during this time that Andrew White acquired the Reuleaux models although on whose recommendation we don't know. One possibility is that he met both Reuleaux and Thurston at the Worlds Fair in 1876 in Philadelphia, as discussed above. Reuleaux' letter of 1882 to White suggests that the two men had been in contact between 1876 and 1882. (Reuleaux thanked White for help in contacting the Smithsonian.) Then in 1886, White proposed to the Trustees to appoint Robert Thurston as Director of the newly organized Sibley College of Mechanic Arts and Mechanical Engineering. Shortly after his arrival, Thurston wrote to Andrew D. White in a letter dated July 8, 1885 (typed). Thurston had recently been appointed as Director of the newly revised College of Mechanic Arts in a Board of Trustee meeting of June 17, 1885. Thurston notes the need to "equip the department of mechanical engineering." "With the exception of the Reuleaux collection" he asks for funds to buy more models, plans, charts and a "complete collection needs to be obtained." It is not clear whether this refers to the Reuleaux models or other teaching models. In Thurston's letter to White in 1886 asking for financial resources to rebuild the College, he comments that the Reuleaux collection of models, paid for with a Sibley gift of $8,000, was one of the few strong assets of the College.

In the period of 1886 to 1900, Thurston built the mechanical engineering program on a new educational model based on a balance of scientific and mathematical fundamentals, engineering science and practical shop experience. The use of extensive laboratories and collections of teaching models of mechanisms and machines as well as technical artifacts of the day were essential to the students all round education. Using this new philosophy, Thurston catapulted Cornell's Sibley College of Mechanical Engineering into the top rated program in the United States (Calvert, 1967).

The Reuleaux Collection at Cornell is a reminder of how the vision of A.D. White, H. Sibley and Robert Thurston helped change the nature of engineering education in the United States.

The collection of mechanisms is housed in the Sibley School of Mechanical and Aerospace Engineering. When it was originally acquired in 1882 it was housed in the old Engineering campus at the north end of the University. According to an article in the October 17, 1885 issue of Scientific American there was originally a museum of mechanisms and a museum of machines in the then Sibley College of Mechanical Engineering. After the second world war Cornell moved its entire engineering college to the south end of the campus and the kinematic collection was housed in special cabinets in one of the new buildings. By the 1970's, however, space assignments in the College shifted and the Collection space was occupied by Computer Science, so that Mechanical Engineering faculty and students rarely saw or used the models. One of the last faculty to use the collection for teaching was retired Professor Richard Phelan whose design text was well known in earlier decades. After a hiatus of over a decade, however, several younger faculty in design discovered the use of these wonderful machines for teaching. Today the collection is back on the turf of Mechanical and Aerospace Engineering and are publicly on display for students and visitors to view. A few faculty such as myself make use of a dozen or more of the models each year for design and dynamics courses. Inspired by the wonderful kinematic collection in Aachen and Hannover we have motorized one model and hope to motorize reproductions of many other models so that visitors can actuate the models from outside the display cases.

The Engineering Library has embarked on a conservancy program to catalog and make images of the models available on the web. To date we have photographed all 219 Reuleaux models on color slides and hope to make their images available on the web in the form of a virtual museum. Plans are also proposed for quicktime computer movies of selected models, again inspired by the modern German collections.

The state of the models are remarkable. Except for one model, the mechanisms show no rust, Reuleaux having designed the cast iron alloy himself to resist rust. All but a half dozen models actually work and can be actuated by hand. The most marvelous models are the clock and power escapement mechanisms which were designed by Reuleaux to exhibit self excited oscillations driven by a weight on each model. While most of the models represent devices that were common to various technical machines, such as four bar linkages, pumps, or intermittent mechanisms, a small subset actually simulate mathematical curves attendant to kinematic motions, such as cycloid and trochoid curves related to rolling mechanisms.

As noted above, these models were reproduced by the Voigt workshop from Reuleaux's original models under his supervision. Cornell has copies of the Voigt catalog which consists of pictures and a separate list of short descriptions in German. We are using the Reuleaux-Voigt numbering system to catalog the collection (see Table 3).

Researchers from outside Cornell can have access to these models for study by contacting the author of this article.

• He who best understands the machine, who is best acquainted with its essential nature, will be able to accomplish the most by its means.

• ...if the new theory is to lay claim to general interest, it must be capable of producing something new; it must make problems solvable which before could not be solved in any systematic way. This may certainly be said to be the case if it succeeds in making Machine-Kinematics, down to its simplest problems, truly scientific.

• What is left unexamined is however the other, immensely deeper part of the problem, the question: How did the mechanism, or the elements of which it is composed, originate? What laws govern its building up?

• In other words, the invention of a mechanism will be to the scientific kinematist a synthetic problem,–which he can solve by the use of systematic, if also difficult, methods.

[All quotes from the review book Modern Kinematics,

Edited by Arthur G. Erdman, John Wiley & Sons, Inc., New York, 1992.]

The author acknowledges the help of the following:

Cora Jackson, for cataloging and typing

Professor Dietmar Rempfer, Cornell University, for German translation

Professor Emeritus, Richard Phelan, Mechanical and Aerospace Engineering

Professor Reinhard Braune, Institut für Getriebetechnik, University of Hannover, Germany

John Reis for photography