by Ben Russell
Model of sun and planet gear, c. 1781. This model was used by Boulton & Watt to guide the development of the full-size version for the rotative engine.
The first of Boulton & Watt’s rotative engines set running outside Soho was built for John Wilkinson in 1783, and by 1800, they had built 278 as opposed to 171 pumping engines. Boulton & Watt’s customer list reflects the rotative engine’s utility in a wide range of tasks. Customers included lead works, rope works, malt distilleries, sugar, tobacco and snuff manufactories, druggists, rolling mills, forges and foundries, glazers, bark and cork mills, even a mustard manufacturer.31 So Boulton & Watt’s engine was in considerable demand from customers spread broadly across Britain, and three-quarters of English counties had one at work.32 There were also large concentrations of engines working in relatively small areas, and one of the greatest of these was in Lancashire, with 44 engines erected by Boulton & Watt for cotton mills, along with eleven more in other industries.
But not all was as the raw numbers suggest. Boulton & Watt’s rotative beam engine was a highly effective machine, but there was not necessarily a clear-cut link between steam power and production. Potential customers’ power requirements varied considerably. Overwhelmingly the demand for power – of any type, steam or water – was for small units. For many producers, remaining small offered distinct advantages: less risk of overcapacity in a downturn, the ability to more easily change production to suit markets. Sheer size ‘guaranteed neither efficiency in good times nor viability in bad’.33 Even where a factory was big, it was not necessarily power-hungry: Boulton’s Soho Manufactory and Wedgwood’s Etruria works used steam for some processes, but many of their workshops demanded no more power than could be provided by a foot treadle or a kickwheel.34 It was common, where a large mill building was constructed, only to gradually fill it with machinery, or even sublet space to different companies, offering ‘room and power’ in return for rent.35 The average Boulton & Watt engine built before 1800 developed only 15 horsepower.36 So, potential customers for their rotative engine wanted power, if they wanted it at all, on a small scale.
Watt’s rotative engine, as built by him from 1787 until 1800; illustration from John Farey’s Treatise on the Steam Engine (1827).
Many prospective customers were also very timid in their approach to power. Although they might have had experience of production driven by waterwheels or even horses, they were less likely than miners or colliery owners to have prior experience with any sort of steam engine, meaning they needed more guidance in commissioning and operating new plant. They could be led astray by engine builders like the two Matthew Boulton described as ‘drunken, idle, stupid, careless, conceited rascals [who] say, and their masters seem to believe, that it required the learning and knowledge of a University man to keep an engine in order’.37 There was less enthusiasm for adopting an efficient but relatively unproven Watt engine if there was any chance of it being less than entirely reliable. Best to rely on a tried and tested alternative.
Often that alternative was a variant of Newcomen’s atmospheric engine. This could be connected to a crank and flywheel to make a rotative motion; it might not work very evenly, due to the engine’s single-acting cycle, but could sometimes be made more so by using a heavy weight on the flywheel to act as a counter to the engine’s powerful downstroke.38 Or, more simply, an atmospheric engine could be used alongside a waterwheel, recirculating water from below the wheel to a reservoir above it. If the water supply was reliable, the engine might need to be used only in the event of a summer drought. Both these schemes found favour because manufacturers accepted the atmospheric engine’s higher fuel consumption as a price worth paying for simple and reliable operation. In advance of developing their sun and planet engine, even Boulton & Watt’s ‘Old Bess’ engine was used at their Soho Manufactory to recirculate water over a wheel, and John Wilkinson had one at his ironworks, ‘an engine of great magnitude which brought up the contents of the river as it were at one stroke’, that ‘shook the buildings and ground for a considerable distance’.39
Further, waterwheels remained a serious proposition for many prospective customers. They were by no means a crude alternative to steam power. The same year that Watt patented his separate condenser, 1769, John Smeaton was beginning to apply cast-iron components to a wheel at the Carron Foundry in Scotland, which was ‘found to answer much better than wooden ones’ and was ‘constantly recommended afterwards by Mr Smeaton for other mills which he designed, and by degrees came into very general use’.40 By substitution of cast iron, careful testing and design, Smeaton doubled the waterwheel’s efficiency. And in the first quarter of the nineteenth century, Thomas Hewes did for waterwheels what Watt had done for the steam engine, particularly with the ‘suspension’ waterwheel, where power was transmitted to machinery not via the central axle, requiring heavy construction, but via a gear meshing with teeth around the circumference of the wheel, which only needed support from the lightest of wrought-iron spokes, like a bicycle wheel.41 The capabilities of waterwheels are suggested by one that Jabez Fisher saw at Stockport, 40 feet in diameter but so finely poised that it was ‘turned by about as much water as could go out of a Pint Mug’.42 A waterwheel like that installed at Arkwright’s Cromford mill might generate 12 horsepower but, by 1800, wheels of 80 horsepower were in use.43 In the following decades, they reached 100, even 200 horsepower – far more than the steam engines then available.44 These developments reduced the incentive to adopt steam power.45
If they wished to purchase an engine, customers need not turn just to Boulton & Watt: there were many other competing firms. Some, like Matthew Murray in Leeds, presented a major challenge, being ‘the first to set an example to Boulton and Watt themselves in that superior finish to the steam-engine which has now become general’.46 In Manchester the company of Bateman & Sherratt was acknowledged as having ‘very ingenious and able’ engineers who built engines that were ‘of a small size, very compact, stand in a small space, work smooth and easy, and are scarcely heard in the building where erected’.47 Others were less of a threat: an engine built by Ebenezer Smith and Co. was described by James Lawson, an engine erector for Boulton & Watt, as ‘one of the worse made I ever saw’.48 As in scientific instrument making, there was room alongside the very best for those building machines that were just sufficient for the job in hand. John Farey recorded that in the early 1790s ‘great numbers’ of old atmospheric engines were being constructed to drive mills on account of the cheap coal available locally: ‘These engines answered the purposes for which they were applied, and were used for many years.’49 It is telling, then, that Manchester writer John Aikin recorded that only ‘some few [engines] are also erected in this neighbourhood by Messrs Bolton and Watts’ [sic].50
So, although Matthew Boulton predicted that rotative engines presented ‘a field that is boundless’, this declaration did not necessarily transpose into orders for engines.51 Boulton & Watt’s engine, for all its technical prowess, was one of a number of potential options which could be provided by a range of different suppliers. Out of more than 2,000 engines built in Britain by 1800, two-thirds were of the older atmospheric type, and Boulton & Watt built only one-quarter.52 What mattered more to their customers than the sheer scale and the type of power source used was the modernity and ingenuity of the machinery that it drove. This was reflected in the contrast, noted by visitors to Manchester, between the smoke and confusion outside the mills, which was in large part produced by steam engines, and the precision mechanical movements of the production machinery within.
One of visitors’ abiding memories of Manchester was its ‘dark black smoaky atmosphere’, the product of a growing number of chimneys, each connected to a boiler for raising steam.53 In 1786 there was only a single chimney in the town but by 1801 there were more than 50 and, visiting the town in 1802, Eric Svedenstierna was moved to write how ‘in order to carry away the coal-smoke, the chimneys at most of the mills are taken up high above the roofs.’54
Together they supplied the town with a constant fug of smoke, ‘an inky canopy which seemed to embrace and involve the entire place’.55 Alongside the polluted atmosphere, Manchester at street level must have been an assault on visitors’ senses; in 1835 Alexis de Tocqueville described ‘the crunching wheels of machinery, the shriek of steam from boilers, the regular beat of the looms, the heavy rumble of carts’ as ‘the noises from which you can never escape in the sombre half-light of these streets’.56
If the external effects of engines were the subject of negative comment, the machinery inside, by comparison, left a very favourable impression on those who saw it. In 1782 an anonymous author wrote that ‘those who are lovers of invention, and fond of mechanical improvements, must admire the ingenuity of the cotton mills and the engines lately erected in the neighbourhood of Manchester’.57 In 1786 Joseph Smith & Robert Peel of Manchester summed up the main advantage of the English cotton trade as arising ‘from our machines both for spinning and printing; by means of these we can spin both cheaper and better, and we can print . . . cheaper and better’.58 And from a slightly later vantage point Edward Baines attributed to cotton machines ‘as great a revolution in manufactures as the art of printing effected in literature’.59 Britain became the world’s principal textile manufacturer by solving the mechanical problems of production – and that was most evident among the whirring shafts and spinning flyers inside the mills.
A question arises, then, over who the people were to best deliver this modern, mechanical world. Earlier chapters have explored the links between blacksmithing, instrument making and early machine making. Another major source of engineering skill was among millwrights. This class of men emerged to design and construct windmills and watermills, pumping machines and ‘all the various kinds of rough machinery in use’ into the nineteenth century.60 A vivid description of the millwright has been provided by William Fairbairn. He characterized the millwright as a ‘jack-of-all-trades, who could with equal facility work at the lathe, the anvil, or the carpenter’s bench . . . he thus gained the character of an ingenious, roving, rollicking blade, able to turn his hand to anything’.
Millwrighting was originally a peripatetic job, and millwrights, ‘like other wandering tribes . . . went about the country from mill to mill, with the old song of “kettles to mend” reapplied to the more important fractures of machinery’.61 However, as the eighteenth century progressed, more millwrights were employed at a single site. Fairbairn was apprenticed to the Northumberland millwright John Robinson, who maintained all the machines working at a colliery and who, perhaps because of his wide responsibilities, had a ‘rough, passionate temper’ and ‘indulged in . . . profane swearing, carried to such an excess that an order was scarcely once given unless accompanied by an oath’.62 Elsewhere Fairbairn notes the existence of Millwrights’ Institutes in public houses, where drink and disagreement over points of practical science led to members imparting knowledge with a fist aimed at ‘the sensitive parts of the body, instead of appealing to the higher organs of intellect’.63
Such men, at once foul-mouthed, belligerent and highly skilled, found themselves drawn in two directions. Engines like Watt’s, which required precision manufacture, needed the facilities of a dedicated works, making use of specialized equipment like cylinder-boring machines and lathes. In these places the millwright became subsumed into the machine making profession alongside engineers, turners and fitters, working alongside carpenters, wheelwrights, cabinetmakers, joiners, smiths, clockmakers and others attracted from their original trades by better wages.64 However, a second path allowed millwrights to retain a distinct identity: the engine (or other power source) still had to be assembled on site, and its power transmitted to the production machinery that it drove. This remained the millwright’s specialized domain.65 In fact those writing in the early nineteenth century pointed to Watt’s engine and Arkwright’s cotton-spinning machines as leading to the advance in millwork, making the expert millwright more indispensable than ever.66
The millwright’s particular job was to construct the shafts and wheels known as ‘mill-work’, needed to couple a power source to the production machinery. This entailed planning out the entire scheme, calculating the proportions and strength of the individual parts and determining the arrangement of the machinery. The mill-work was required to be strong, stiff and ‘easy of repair’.67 Fairbairn recalls rising
with the sun in summer, and some hours before it in the winter . . . For the remainder of the day I had either to draw out the work, or to ride fifteen or sixteen miles on a hired hack to consult with proprietors, take dimensions, and arrange the principle and plan on which the work was to be constructed.68
Practical work was matched by skill at negotiation and draughts-manship as well.
All was underpinned by the millwright’s ability to manipulate materials. At first, they worked mainly with timber, forming ‘large square shafts and wooden drums, some of them 4 feet in diameter’ but this, with other associated pieces of mechanism, ‘not only crowded the rooms, but seriously obstructed the light where most required, in the more delicate and refined operations of the different machines’.69 Wood began to be replaced with metal to better withstand the higher speeds at which the machinery was required to operate; bulky millwork was shrunk, the ‘ponderous masses of wood, cast iron, and their enormous bearings and couplings’ giving way to ‘slender rods of wrought iron and light frames or hooks for suspending them, [and] pulleys and straps of moderate diameters and dimensions’.70
So, the means of building the internal systems of a mill were provided by the millwright, using many of the techniques used to construct engines that we have earlier discussed. But alongside them worked a second distinct group of workmen, from the world of clockmaking.
As the eighteenth century progressed, and possession of a timekeeper became as much a mark of social standing as it was a record of timekeeping per se, horology was a growing trade.71 It was also highly skilled, requiring ‘no great strength’ but ‘a mechanic head, a light nice hand; and a strong sight, there being scarce any trade which requires a quicker eye or steadier hand’.72 These factors attracted James Watt, who took an interest in clockmaking alongside his other projects. He and his friend William Small were exchanging notes on a new design of clock in 1771, with Small telling Watt how ‘I have perfected my clock with one wheel of nine inch diameter, which is to tell hours, minutes, and seconds, and strike, and repeat, and be made for thirty shillings.’73 And even before that, in 1758, Watt was in partnership with Joseph Black and Alexander Wilson to make clocks – his workshop contains an incomplete timepiece movement and a number of unfinished clockwheels that may date from this period.74
Thomas Allom, Swainson Birley cotton mill near Preston, Lancashire, 1834. The machinery seen in this view, used to prepare the cotton for spinning, is driven by the belts and shafts suspended from the ceiling above.
‘The Watch Maker’, from The Book of Trades (1824). Here he is using a small watchmaker’s turns, a small lathe, surrounded by a range of clocks and watches.
How clockmaking was organized varied across Britain. At one end of the scale were the makers based in London. The metropolis was home to a complex trade exporting finished clocks all over the world, comprising large numbers of specialized workers – Abraham Rees counted up to eighteen different trades – movement makers, spring makers, enamellers, brass founders and more.75 The movements, springs, chains and cases were all manufactured by different tradesmen, though it was the watchmaker who ‘puts his name upon the plate, and is esteemed the maker, though he ha[d] not made in his shop the smallest wheel belonging to it’.76
In contrast to the London trade stood those making clocks who might have been trained as blacksmiths and in associated trades, building locks or even firearms. Their biggest products would have been large ‘turret’ clocks intended for churches and other places from where the time was told by the wider public.77 These clocks were accurate timepieces but, owing
to their physical size, their construction owed as much to the blacksmith as to the bench-based clockmaker. For this reason, these self-contained and multi-skilled tradesmen have sometimes been called ‘clocksmiths’.78
The nature of the clocksmith’s workshop reflected their combination of skills. It would ideally admit as much natural light as possible, to illuminate intricate work such as polishing, engraving and marking out clock dials.79 The bench would be positioned close to the window and, if very small work was carried out there regularly, might have a groove along its front and ‘wings’ around the sides to catch components rolling off. For very small work, wineglasses broken off at the stem were traditionally used, turned upside down, to protect delicate components placed beneath them while a mechanism was dismantled. But alongside the bench for small work, the workshop would also contain much shared with the blacksmith: ‘forge, bellows, anvil, tongs, hammers, screwplate and taps, bench and hand vices, drills, and pliers’.80 The forge might be placed in a separate, darker room, or as far away from the window as possible, the better to accurately gauge the colour and working temperature of the metalwork being heated upon it.