Note: Some of the images are thumb-nails: click on them for larger
Last updated: 5 June 2016: David Green
A Coffin planimeter made by Bushnell
The Coffin planimeter is a form of linear planimeter. It is named after its inventor, John Coffin, and originated with Coffin's U.S.Patent 258,993 issued on 6 June 1882 (note 1). Download a pdf copy.
Coffin, of Syracuse, New York, called his design "an averageometer for measuring the average breadth of irregular planes". He "designed the instrument more especially for measuring the mean height of indicator-diagrams taken on steam engines".
Steam engine indicator-diagrams were the main game.
Steam engines played a pivotal role in the Industrial Revolution. Invented in the 17th century, their use was restricted for many years to pumping water out of mines, until James Watt (1736-1819) introduced improvements that made them more widely applicable. In the first decade of the 1800's, the introduction of "strong" steam (ie. super-heated steam) enabled still smaller and more powerful steam engines to be built. The first steam locomotive was demonstrated in 1804 by Richard Trevithick and the first steamboat was launched in 1807. Thereafter, huge numbers of steam engines were made for these and other purposes.
These engines all needed to be maintained and operated at peak efficiency. The tool that helped to achieve this was the steam-indicator (note 2). The output from a steam-indicator is a drawing (the indicator-diagram) that shows the relationship between the pressure in the cylinder and the volume of steam in the cylinder, during one complete cycle of the piston. A typical indicator-diagram is shown opposite.
The power of the engine can be calculated from the indicator diagram.
The power of an engine, or the rate at which it can do work, is expressed as its "horse-power", where one horse power equals 33,000 ft-lbs per minute (note 3). The actual horse-power developed by the steam in the cylinder or cylinders is called the Indicated Horse-Power because it determined by using the Steam Indicator.
It is calculated from the formula:
Indicated Horse-Power = P * L * A * N / 33000, where
P is the mean effective pressure (M.E.P.) on the piston in pounds per square
L is the length of the stroke in feet (for a single-acting engine),
A is the mean nett area of the piston face in square inches, and
N is the number of revolutions per minute.
L, A and N can be fixed for an indicator test, so it remains to determine the M.E.P.
The area of an indicator-diagram represents the work done in one cycle. As the length of the diagram represents the stroke of the engine, the average height of the diagram represents the M.E.P. Thus to calculate the M.E.P. we have to measure the area of the diagram and then divide it by the length of the diagram, scaling the result to account for the strength of the spring.
The area of the diagram can be determined graphically or with a planimeter (note 4). The purpose of Coffin's design was to make this process as simple as possible.
Figure 1 from patent #258,993 shows the design proposed by Coffin. All
later variations utilised the basic form proposed in this patent (although
I have not yet seen a model that uses the arm j in the drawing) but several
inventors patented improvements to it, including:
. L. T Snow (1903),
. F. R. Williams (1903),
. E. F. Chase (1905),
. E. McC. Scoville (1906),
. F. C. Blanchard, E. B. Crocker and P. G. Darling (1910) and
. F. R. Williams (1912).
These improvements related mainly to fixing scales to the instrument to simplify or eliminate subsquent calculations. With Coffin's design you read the area from the planimeter wheel and calculated M.E.P. separately.
Coffin seems to disappear from the scene after registering his patent.
Several manufacturers were associated with Coffin planimeters. None of the
models in my possession has any indication of a date of manufacture so it is
not possible to be sure of their age. Alphabetically, companies manufacturing
or supplying Coffin planimeters (and known to me) are:
. American Schaeffer & Budenberg,
. American Steam Gauge & Valve Mfg. Co.,
. Thompson & Bushnell and
As some of the models incorporate later patents it is possible to make a few guesses at the order in which they were made. Inventors and manufacturers are listed below in my (not necessarily correct) attempt at a time-line. The dates are when the planimeters may have been made, not the life span of the companies.
Ashcroft inscribed on many of their planimeters "Sole Manufacturers - Coffin's Pat. June 6. 1882." If we take them at their word, their's will be the earliest examples. Ashcroft Manufacturing Company, New York City, and Bridgeport, Connecticut, USA are said to have made Tabor indicators from about 1880 to the First World War [Ref 1] so this seems a plausible assumption. Certainly the planimeters closely resemble the patent illustration.
The planimeters are numbered. One of my two is #32058 so, unless the number contains a code, a great many were made. The suggestion that Ashcroft ceased business by around 1918 would exclude the possibility of #32058 meaning planimeter 58 made in 1932.
Where other manufacturers always (?) provided complete units, Ashcroft frequently supplied the planimeter arm in a separate box as a separate item. The wooden cases containing the complete units were robust and very nicely made.
To read the instructions for using the Ashcroft planimeter, click on Part 1 or Part 2.
It seems likely that this model was superceded in 1910.
Thompson sold a Coffin planimeter that appears to be very similar to those made by Ashcroft. The only obvious difference is that the recording drum has a white ivorine surface with the numbers indicated in black, rather than the less legible engraved metal drum of the Ashcroft.
The Thompson & Bushnell Company is reported to have made Coffin planimeters that were used by Ashcroft amongst others. This contradicts Ashcroft's claim to be "Sole Manufacturers". I have not yet seen a Coffin planimeter identified as made by Thompson & Bushnell. Never-the-less the possibility has been raised (note 6).
According to [Ref 1] The Trill Indicator Company, Corry, Pennsylvania, USA. made Triumph and Trill indicators, internal- and external-spring types, from 1901 until the end of the Second World War. They may possibly have made planimeters over the same time span. Although at least 19 years after the patent was registered, they made a Coffin planimeter true to the original patent.
U. S. Patent 718,166 was issued on 13 January 1903 to Levi T. Snow. (Download as pdf).
U. S. Patent 718,166, issued on 13 January 1903 to L. T. Snow, was assigned to John S. Bushnell & Co.
The Bushnell company produced two models of the Coffin planimeter incorporating Snow's improvements - the "Bushnell-Coffin Planimeter" and "Bushnell's Improved Planimeter".
The "Improved" planimeters are usually in a box bearing the legend "Manufactured by John S. Bushell & Co. New York, N. Y.", and "patented January 13, 1903".
The "Improved Planimeter" appears to be the more common, but Bushnell does not seem to have made many of them. Of thirteen for which I have the relevant data, the serial numbers range from #11 to only #408.
My Bushnell-Coffin planimeter (#300) is shown at the top of this page.
U. S. Patent 746,427 was issued on 8 December 1903 to F. R. Williams. It is not known if any of these were made.
U. S. Patent 783,568 was issued on 28 February 1905 to E. F. Chase., (download as pdf).
U. S. Patent 809,019 was issued on 2 January 1906 to E. McC. Scoville. (pdf).
This isn't exactly a Coffin planimeter but it is a linear planimeter, mounted on its own table and used to measure indicator diagrams, so I have included it here.
It is not known if any of these were made.
U. S. Patent 961,836 was issued on 21 June 1910 to F. C. Blanchard, E. B. Crocker and P. G. Darling. (pdf)
They state: Our invention ... consists in improvements which relate to the application and adjustment of diagrams to the planimeter table, the averaging of the ordinates of the diagram curves and the employment of the planimeter either on its special table, or separate therefrom.
U. S. Patent 961,835, issued on 21 June 1910 to F. C. Blanchard, E. B. Crocker and P. G. Darling, was assigned to the Ashcroft Manufacturing Company, and physical copies were made, as shown below.
A Bushnell-Coffin planimeter offered for sale on eBay had a ribbon in the lid and an instruction sheet both claiming: "Manufactured exclusively by the American Steam Gauge & Valve Mfg. Co." (note 5).
A Bushnell-Coffin planimeter offered for sale on eBay (stamped with the number 308) had a ribbon in the lid that appears to be labelled American Schaeffer Budenberg Brooklyn.
My Bushnell-Coffin planimeter (stamped with the number 300) is accompanied by a set of mimeographed instructions with the incomplete identification " ...an Schaeffer and Budenburg".
U. S. Patent 1,049,186 was issued on 31 December 1912 to F. R. Williams. It is not known if any of these were made.
Two companies made a planimeter they called The New Planimeter. I haven't been able to date the following planimeters but they would seem to be contempory with Coffin's patent (1882). They concentrate on the same problem - measuring indicator diagrams.
As The Mechanical Specialities Mfg. Co. was based in Boston and Hine and Robertson were based in New York, they may have made these planimeters at the same time. If not, I don't know which Company was earlier.The Mechanical Specialities Mfg. Co.
This device has all the features of the Coffin design, except that it measures the distance the wheel slips along the bar rather than the distance it rolls.
Conveniently, the instructions are pasted into the lid of the box, as follows: "Place the card under the clips so as to bring the atmospheric line nearly parallel with the grooved guide. Set the wheel against the shoulder, with one foot of the planimeter in the grove, and the tracing point in the highest left-hand position of the card, as at A and B in the engraving. Carry the tracer with a free, easy movement in the direction followed by the hands of a clock, tracing the figure, and returning to the starting point. The number of square inches can then be read on the scale by the distance the wheel has moved from the shoulder. The scale is divided into tenths, and each tenth into halves, which would read .05 of a square inch for each division."
According to the National Museum of American History, Hine & Robertson of New York City made steam-engine indicators and sold planimeters from the 1880s to 1897, when the firm was renamed James L. Robertson & Sons.Codman & Shurtleff
medicalantiques.com lists Codman & Shurtleff (1851-1900) as a minor Boston supplier of surgical instruments during the Civil War. They say the company Codman & Shurtleff & Co. (B. S. Codman, A. M. Shurtleff, and Franklin O. Whitney) was formed in 1857 in 13 Tremont, Boston. When B. S. Codman died 22 February 1894 the partnership dissolved and the firm was placed in receivership. It was bought by A. M. and H. L. Shurtleff in 1896.
Codman & Shurtleff also made (or at least supplied) a portable planimeter set for measuring indicator diagrams. It made use of the early American-style polar planimeters, which perhaps they made.
I would like this page to be historically accurate. Sadly, hard facts are thin on the ground. Some of the information here is based on documentary evidence; some only on Web references that need to be treated as hearsay.Note 1.
All the patents mentioned in this page can be found by visiting the United States Patent and Trademark Office and typing in the patent number. For example, to see a scan of Coffin's patent at the USTPO, type in the patent number 258993. Each page of the patent must be downloaded as a separate tiff file.
The US patents are also mirrored here as pdf files. Click the links.back
T. J. Main and T. Brown, in "The Indicator and Dynamometer" published by Longman and Co. London 1864, observe that:
"The Indicator is one of Watt's inventions, upon which he was accustomed to place great reliance; and it may not, perhaps, be too much to say, that, in his hands, it contributed mainly to his successive improvements in the Steam-engine. After his patent had expired, and the Engine had become public property, the various makers, it seems, did not at first sufficiently value this useful instrument; ... However that may be, such is not the case now; for every engine-maker is careful to apply it, as the best means of testing the working condition of his engine ...
The Indicator appears to fulfil two distinct and very important ends.
1. It enables us to discover whether there are any defects in those parts of the machinery by which the steam is admitted to the piston ... in the hands of a skilful engineer, the Indicator is as the stethoscope of the physician, revealing the secret workings of the inner system, and detecting minute derangements in parts obscurely situated.
2. It discovers, at any instant of time, and under any given circumstances ... what is the actual power of the engine."
In broad outline, the indicator consists of a cylinder with a loose fitting piston (A), which activates a pencil (or metal scribe) B tracing a diagram around a drum (D). Through suitable links the vertical motion of the pencil records (and magnifies) the piston's vertical motion. When the indicator is attached (at E) to a take off point on the engine, the pressure on the underside of the piston due to the steam is balanced by the resistance of a helical spring (S).
The compression of the spring is proportional to the pressure of the steam on the underside of the piston. The strength of the spring is known and is stamped on the spring. For example, a spring marked 400 indicates that a vertical displacement of one inch by the pencil on the diagram corresponds to 400 lbs pressure per square inch of steam on the indicator piston.
The indicator drum carrying the diagram turns on its own axis first in one direction and then the other. On the outward stroke of the engine piston, the cord (C), hooked onto the piston and wrapped round the base of the drum, rotates the drum against the resistance of a coiled spring inside it. On the return stroke this spring rotates the drum in the opposite direction. The indicator cord and an intermediate reducing mechanism are adjusted so that the pencil's horizontal displacement round the drum corresponds with the displacements of the engine piston in the cylinder.
There are some nice pictures of indicators at oldengine.org, for example the Crosbie Steam Gage & Valve Co. Indicator here, the Schaeffer & Budenberg Indicator here and the Dobbie McInnes Steam Indicator here, and other pictures in links at the bottom of those pages. back
The power of an engine, or the rate at which it can do work, can be expressed
in two ways:
* the actual horse-power developed by the steam in the cylinder or cylinders, called the Indicated Horse-Power because it is measured by means of the Steam Indicator.
* the horse-power available for work at the engine shaft and called the Brake Horse-Power.
The brake horse-power is always less than the indicated horse-power because some power is lost within the engine due to the friction of the various working parts. back
The common graphical approach was the "ten ordinates" method. The diagram was divided into ten equal vertical slices and the mean height of each slice was measured using the appropriate scale for the spring used to generate the diagram. The sum of these ten measurements was divided by ten to give the mean, which is an approximation for the mean effective pressure. The choice of ten ordinates was a balance between convenience and accuracy.
To use a planimeter, the area of the diagram was traced and then divided by the width of the diagram, carefully measured to two decimal places. The quotient gives the height in inches, which, multiplied by the scale of the spring, gives the M.E.P. back
American Steam Gauge & Valve Mfg. Co. Boston, Mass.
According to their 1908 catalogue, The American Steam Gauge and Valve Manufacturing Company was established in 1851, incorporated in 1854 and reincorporated in 1903. Their central offices and factory were at 208-220, Camden Street, Boston, Mass.
Is this the same Company as:
. American Steam Gauge Company, Chardon Street, Boston, Massachusetts, USA (with a 'New York Branch' in Dey Street, New York City, and a 'Western Branch' in North Canal Street, Chicago). Made Richards indicators, then Thompson and 'American Thompson' designs, internal- and external-spring, from the early/mid 1870s apparently until taken over by Schaeffer & Budenberg during the First World War. [Ref 1].
. American Steam Gauge Company, 36 Chardon Street, Boston, Mass, incorporated 1854 (Original Steam Gauge Co. Business Established 1851) ... sole manufacturers of ... Amsler's planimeter.back
. "An alternative method of assessing the indicator diagram was provided by the 'Averageometer', patented in the USA in 1882 by John Coffin, made by the Thompson & Bushnell Company of New York City and favoured by (amongst others) Ashcroft, maker of Tabor-type indicators". [Ref 1]
. Thompson & Bushnell Company, New York City, USA. Made the Bachelder indicators from c. 1888 until 1905 or later. [Ref 1]
(1) The Web site http://www.bollee.fsnet.co.uk contained a lot of information about steam indicators but is no longer on the Web. It appeared to be serious about accuracy. Some of the dates here are from that source back.