The authors of this database, Kenneth M Brown and Jan A Verbruggen (whose names will be abbreviated to KMB and JAV respectively) aim to bring together data on all non-rotative pumping engines working on the Cornish cycle, that were ever made, with data on all the sites where they worked. It is also an attempt to establish a common reference system for engines and sites. The documents, notes, and other data collected by KMB constitute the basis for the content of the database. JAV has been mainly concerned with putting the core of this content into an electronic database which allows various searching, viewing and report options. The main use envisaged for these, is to facilitate further study of the history of the Cornish pumping engine. Rotative engines (whims, stamps etc.) would require a different format, and for the present they are excluded from the database. At a (much) later stage possibilities of extension of the range of engine types may be considered. The data for each engine and site are, inevitably, brief and cannot encompass all details. They are heavily referenced, however. There exist other attempts at comprehensive surveys of mining in Cornwall, notably by Dines, Collins, Hamilton Jenkin, Burt, and (unpublished) Brooke. These concentrate mainly on social, economic and company structure aspects; the present database is principally about engines - though for Cornwall the list of mine sites tries to be comprehensive, extending to sites which never had steam pumping.

Cornish Engine Database



The 'Cornish Engine' has been defined by Pole as a single-acting engine with the following somewhat imprecise characteristics.
1st. The greater extent to which the principle of expansion is carried.
2nd. The use of high-pressure steam.
3rd. The form and proportions of the boiler and furnace.
4th. The careful management of the fires and the slow rate of combustion adopted.
5th. The differences in the condensing arrangements, the large size of the passages, the position of the air pump, the pause before the commencement of the stroke, &c.
6th. The difference in the system of the pumps and pit-work. And last, but not least important,
7th. The great care that is taken to economize the heat in every possible way.
Further obstacles to a precise demarcation of the Cornish Engine lie in the tendency of Pole and others to designate engines as 'Boulton & Watt with some modifications'. The following definition may come near the mark.
The Cornish pumping engine is a non-rotative, usually single-acting pumping engine in which steam of c.30-60 psi is used expansively to raise a weight. This weight, descending, operates the pump(s). The engine usually has three double-beat valves, operated by a distinctive valve gear arrangement.
This definition is certainly not perfect, but it covers most of the field (exceptions are e.g. bucket pumping engines, where the intermediary of a weight is not always used).
The start of the Cornish Engine era is, not surprisingly, somewhat diffuse. In 1800 the Watt patent (which had stifled further developments to a considerable extent) expired, and after a transition period, experiments with higher steam pressures, expansive working etc. got under way. Developments were fairly gradual, and a number of engineers were involved. Existing engines (mainly by Boulton & Watt, see e.g. Barton App.III) were often adapted for expansive working at higher steam pressures. Such engines, though not built as 'Cornish engines', then worked on the Cornish cycle.
Barton mentions R.Trevithick's 1812 Wheal Prosper engine E-0022 which, he writes, 'has been described as the first true Cornish engine'. Joel Lean's 'Engine Reporter' started in 1811. The present authors have decided to use this year as a cutoff point without, however, intending to be dogmatic. This means e.g. that earlier engines (usually B&W, occasionally Hornblower or others) have been included if they have found their way into the literature after 1811. A case in point is the 63" Dolcoath Stray Park Engine (E-0029). Also, many such non-Cornish pumping engines have been mentioned in the site notes, to facilitate possible future developments of the database. At the other end of the timeline, some engines have been included which, though not strictly Cornish engines, in some way or other mark the end of the Cornish era, e.g. Marriott's 80" (E-0011).
The mainstream of the database is formed by the mines and engines of the copper-, tin-, and china-clay-producing regions of Cornwall, as served by pumping engines made in the county. This has been supplemented by the use and manufacture of Cornish engines outside the county, including collieries, waterworks, drainage, canal duty and other uses.
As far as sites are concerned, inclusion was initially determined by the (sometimes potential or surmised) presence of Cornish pumping engines. As the database grew, it seemed useful to extend this (for Cornwall only) to all mining sites, irrespective of pumping arrangements. Outside Cornwall, the inclusion criteria remain unchanged. Site data are primarily included to indicate when a site was being worked, to help with determining working periods for engines. Aspects such as ownership, management, financial data, product sales, etc. are largely omitted. Other studies and archives cover these much better than we can hope to do.

Engines On The Move


Mine pumping engines were massive, bulky pieces of machinery. Surprisingly, many were quite mobile. Mining adventures were often rather short-lived, and when they closed down the engines, being the most valuable capital items on the site, would often find new owners. Dismantling, moving, and re- assembling an engine would require specialist crews, often provided by foundries.
This moving around of engines did already occur in the Newcomen and Watt epochs. In 1781 a 36" B&W engine was erected new on Wheal Treasury, then moved to Wheal Gons, and finally (?) to Crenver, all within a few years. Of the surviving engines the 80" one now known as Robinson's (E-10) was quite well-travelled: made in 1855 for S-19 Alfred Consols (Davey's), she was moved to S-8 Crenver & Wheal Abraham in 1864 and rechristened Pelly's. When that mine was wound up in 1875 she remained unsold on the mine, probably because the cylinder was smashed. In 1883 she was bought by S-20 Owen Vean & Tregurtha Downs, working on and off until 1902 as St.Aubyn's. Finally, in 1903 Trestrail re-erected her on Robinson's Shaft at S-25 S.Crofty, where she worked until 1955, the last Cornish engine to work on a Cornish mine.
Sometimes an engine had to be modified to suit a new location, e.g. to adapt her to an existing house. In the 1850s S-516 Wheal Emily had E-766, a 26" Bull engine. About 1860 this engine was acquired by S-354 South Carn Brea and converted to a beam configuration (a major job, one wonders how many of the parts, besides the cylinder, were reused!), and continued life as E-532. Some (unknown) time later, she was re-cylindered to 36", continuing as E-787. In 1877 the bob broke, and the engine had to be rebuilt. Its final fate is as yet unknown.

Universal Reference System


At present, engines are usually identified by size, maker, worksite etc. This is not always satisfactory, particularly where popular sizes or makers are involved (e.g. 70" engines made by Harvey). Also, some engines travelled from one site to another. A unique engine identification may then have its uses. The E-index in this database might develop into such an identification. The database has been set up to assign the index automatically, and to safeguard its uniqueness.
For sites, the situation is somewhat different. The list includes sites outside Cornwall only, if a Cornish pumping engine may have worked there. In Cornwall it attempts to include all mine sites, but this is also largely tied to the steam pumping era, and does not necessarily include all sites back into antiquity. Also, non-mining sites (waterworks, land drainage etc.) have been included. This means, that use of the S-index as a unique reference is largely limited to a steam pumping engine context.

Incomplete Database


How many Cornish pumping engines have ever existed? In the CAU-document 'Engine House Survey, Mineral Tramways Project' (1991) the total number of steam engines erected in Cornwall is mentioned as 'something in the region of 3000'. Some reckon, that there were on the average roughly two rotatives (one stamps and one whim) for each pumping engine, which would put the number of pumping engines at c.1000. In the early period quite a few non-Cornish engines (Watt, Hornblower, Bull, etc.) were built. Many engines were exported, either new or second-hand. If all these factors are taken into account, a very rough estimate of the total number of Cornish pumping engines built, might be 1500 or so. Barton (p.253 footnote) starts from the estimated output of the foundries (1550) and states that for the Cornish period a total (pumping plus rotative) of 2000 may not be excessive. Barton also deems it probable, that the largest number of mining engines (of all types) standing in Cornwall simultaneously, was c.650, and occurred c.1865.
It is not likely that the true number will ever be known, nor that all engines and sites will eventually be tracked down, nor that the data for each will ever be complete.
This incompleteness results in some problems with the orderly reporting (on screen or on paper) of the data. The flexibility and range of the reports and displays help to partially overcome or ameliorate these problems. Two aspects of this deserve an explanation.
* When reporting engines sorted by size, whether fully, or a range of sizes, all engines of unknown size are always included in the report. This may help towards finding clues as to the size of some of those engines.
* When reporting workings for an engine, it would be desirable to order these chronologically. If the starting and/or stopping date of a working are missing, such sorting becomes difficult or impossible. In such cases there is often some indication of working period, but this is in the notes, and cannot be used for sorting. As an aid, effective in all but a few cases, a special time indicator has been defined, which works if it is set at any year within a particular working. If at least one of the dates (i.e. either starting or stopping year) is available, this indicator can be derived automatically. If no starting or stopping date is known, the user is requested to input a value - which may in some cases be an educated guess.
Incomplete data may also result in duplicate engine entries. Example: if an engine is for sale at a certain time and location, it is often not known whether, or to whom, she was sold. If an engine of the same size is started shortly afterwards at another mine, this may well be the same engine - but more evidence is needed to be certain, and meanwhile two separate entries will often be the prudent way to go.