THE HISTORICAL DEVELOPMENT OF TRADITIONAL WINDMILL TECHNOLOGY IN ENGLAND
Windmills first appeared in Western Europe in the twelfth century, the earliest references in England concerning a mill at Weedley, Yorkshire, in 1185 plus another at Amberley in Sussex which had been built by that year. They became gradually more common as the Middle Ages progressed. There would have been a setback in the fourteenth century due to the Black Death, which depopulated vast areas and shattered the rural economy, but numbers afterwards recovered.
The mediaeval windmill was often a reserve source of power to be used if the watermill, the more tried and tested method of grinding corn on which it was preferred to rely, was out of action for any reason such as streams or millponds drying up. Consequently it was not used for long periods and fell into decay, requiring expensive repairs. In the long run, however, the windmill proved to be essential as the population grew, expanding into areas where there was no water source close at hand and becoming larger than the capacity of the local watermill to meet its needs, when the distance you had to travel to collect your flour was taken into account. Windmills were built where there was no already established watermill and the terrain was suitable – either flat ground unbroken by many trees, or hilly regions where it was easy for the sails to catch the wind. By the mid-nineteenth century there were roughly ten thousand of them in the United Kingdom although by then, as we shall see, the wind as a means of generating power for the milling industry was gradually becoming obsolete.
The evidence of illustrations in illuminated manuscripts and church carvings and in manorial accounts, which give a breakdown of the repairs needed to a mill and thus a certain amount of technical information, shows that early windmills in England were post mills, with the boxlike main structure resting on an upright post about which it could be turned by means of bearings, so that the arms or sails were kept facing into the wind. The miller would push on a beam, later called a tailpole, projecting from the superstructure at the rear. The process of orienting a mill, whatever its type, so that the sails are in the correct position for the wind to rotate them is called “winding.”
These early post mills were small affairs, containing only one pair of millstones, and very fragile, easily blown down by a strong wind; an expensive liability, as indeed to some extent are windmills today, of both the traditional and modern kind. Most of the structure and machinery would have been of wood, with iron used mainly for bearings. The process of separating or “dressing” the meal, the substance resulting from the grinding of the corn, into its principal components was probably done with a hand sieve.
Gradually, as the population increased and, partly as a result of that, the country grew wealthier, mills could be built larger − indeed had to be, depending on the size of the community they served − with two or more pairs of millstones and specialised equipment for dressing and cleaning the meal. New types of mill began to appear, that is the tower and smock mills in which the machinery was contained within a stationary tower, of wood in the case of a smock mill (so called because of a certain resemblance to the smocks worn by rural labourers). Only the wooden topmost section or cap, the timbers of whose base frame supported the windshaft, the axle which carried the sails, turned to keep the latter into the wind.
The first tower mills are recorded in Europe in the thirteenth century; they were not much of a success, apparently because, being made of brick or stone, they aroused the opposition of carpenters’ guilds who feared their members would be put out of business. However they began to appear again in the late seventeenth and early eighteenth centuries, and from then on gradually became more common. Smock mills first appeared in Europe in the sixteenth century, probably originating in Holland. Meanwhile post mills continued to be built until at least the mid-nineteenth century, modernised with patent sails and fantails and much bigger than their mediaeval and early modern predecessors with room for the three pairs of stones made possible by spur gearing.
At first the effect of technical advance such as occurred in the early modern period and in the Industrial Revolution is to benefit the existing technology. Mills were improved by the use of more iron components, and by the invention of the fantail and of spring and patent sails. However these modifications were initially very expensive, and as a cost-cutting measure a wheel with iron cogs would mesh with one that had wood cogs, an iron poll end be fitted to a wooden windshaft, and one pair of common sails be replaced with springs but not both. Many mills continued to be built in a quite primitive style, the wooden windshafts with distinctive large poll ends being evident in illustrations, and manual winding, until the early nineteenth century and were sometimes not modernised until their last years, if at all.
The post of a post mill was mounted on a pair of horizontal timbers called crosstrees, at the point where they crossed over one another, hence their name, and quartered so it could fit over them at the intersection. It was braced to them by four diagonal timbers called quarterbars going from it to their ends, the whole forming a trestle structure which distributed evenly the load being placed on the post. The crosstrees rest on brick piers to which they need not be fixed, the sheer weight of the mill being enough to keep it resting firmly upon them without being blown over in a strong wind. From the mid-eighteenth century the trestle was frequently enclosed within a circular building called a roundhouse. This served no structural purpose, since it did not support the body of the mill, but was really there to provide extra storage space for sacks of corn, etc. Some mills never had one. Illustrations of post mills in paintings and engravings and on estate maps in the seventeenth and eighteenth centuries generally show them with open trestles and steeply pitched roofs on which there is often mounted a pennant in what seems to have been the fashion of the period.
The post extended up into the body of the mill where it fitted into a socket on the underside of a fixed horizontal timber called a crowntree, with a bearing or pintle located in a bearing which allowed the mill to turn, the timber being reinforced at this point by an iron plate sometimes called a Samson head. Attached to each end of the crowntree at a right angle to it was another horizontal timber called a side girt. On the ends of the side girts were hung the four corner posts of the body, or buck, frame. Around these basic timbers the frame was built up with intermediary members, into which door and window frames were incorporated, to strengthen and stiffen it. A vertical “prick post” supported the “breast beam” which carried the neck of the windshaft. The mill was steadied on its post by a collar between two longitudinal timbers called sheers which passed on either side of it.
To allow the mill to turn the access ladder to the buck first had to be raised off the ground, and this was accomplished by working a hinged lever, called a talthur, on the tailpole, which was connected to them by a rod or chain.
Smock mills were usually battered, that is with sloping sides; tower mills battered or cylindrical. In a smock mill eight (sometimes six, ten or twelve) main vertical timbers called cant posts, which supported the curb (see below) on which the cap turned and between which transoms carried the ends of the floorbeams, rested on wooden sills on top of a brick or stone base which could range in height from being just the foundations through a few courses to two or three storeys. There might be a stage around the mill at some point giving access to common or spring sails in order to reef them; patent sails didn’t need one as much since they were self-reefing, but one might still be provided to assist with maintenance.
The cap base frame consisted of two longitudinal sheers, from which the diagonal members of the fantail (see below) support structure sprang. The sheers were connected by a number of lateral timbers. Those at the front and rear, the weather or breast beam and the tail beam, supported the neck and tail bearings of the windshaft. The central timber, called the sprattle beam, received the top bearing of the vertical main drive shaft, which was incorporated into a casting bolted to its rear side (usually) face. The cap turned on rollers or skids on a track on top of the tower called a curb, essentially a flange with an upper and side face. It was known as a “live” curb if rollers were used and a “dead” one where skids were employed. Beneath and attached to it, either on the inside or the outside of the tower, was a toothed rack with which the winding gear engaged. The cap was steadied and kept in a central position by truck wheels bearing on the inner face of the curb, with sometimes a keep flange to prevent it being lifted off in a strong wind.
Originally smock and tower mills were winded by tailpole, and then by hand using a wheel and chain, the wheel being mounted on an axle at whose other end a toothed nut meshed with the rack. Some mills retained one or the other of these methods of winding to the end of their working lives. In the 1740s Edmund Lee invented the fantail. This was an automatic device resembling in some ways a small windmill and usually mounted on a frame at the rear of the cap. It consisted of a number of blades on spokes radiating from a central hub. Whenever the wind changed direction it would strike the fantail at an angle and cause it to rotate, setting in motion a series of gears the last of which meshed with the teeth of the rack. A hatch in the rear of the cap gave access onto the platform on which the fan supports were mounted for maintenance. Fantails could also be fitted to post mills, on the rear roof gable, the tailpole or the external access ladder to the buck.
Post and smock mills are clad with weatherboards, which overlap each other so that rain bounces off instead of penetrating between them.
Occasionally a post mill body might be erected, or re-erected, on top of a short tower, turning by means of either a curb or a tailpole. This type of mill was known as a composite mill.
Sails were usually made of softwood. They consisted of a frame attached to a timber called a stock which in turn is bolted to a shorter timber called a whip. There were a pair of whips, an inner and an outer, at right angles to one another. Originally they were morticed through the nose of the wooden windshaft, called the poll end, but were later wedged into square castings which made up an iron poll end or, as it was also known, canister. Clamps were sometimes fitted to help strengthen the stock at its weakest point.
Each whip carried two sail frames. The bars of these frames were morticed through the whip at increasing angles to give a twist, called the angle of weather, like that of an aircraft wing or propeller blade, which made for greater aerodynamic efficiency as well as being aesthetically very attractive. Controlling the speed of the mill was essential both for safety and because it affected the quality of the flour.
The oldest type of sail, in use from mediaeval times, is the common or cloth sail. In this a length of canvas, which catches the wind in the same manner as the sail of a sailing ship, is spread over a wooden lattice framework consisting of a pair of longitudinal timbers called hemlaths, which form the leading (inner) and trailing (outer) edge of the sail, and are connected by a number of lateral bars which are stiffened by one or more longitudinal battens called uplongs. Common sails were usually single-sided, with the greater part of the sail, where the canvas was fitted, on the trailing side; the sail bars extended past the stock to the leading edge but no canvas was fitted there. The canvas was sometimes attached to an iron rod running along the inner hemlath and was unfurled when required, to whatever extent was necessary to adjust the speed of the mill, depending on the strength of the wind. The mill needed to be stopped for this to be done, which was a major disadvantage.
A partial solution was devised by Andrew Meikle in 1772; this was the spring sail which, like the patent sail discussed next, had no uplongs. The sail bars divided each sail into bays which contained several hinged shutters. If the shutters were closed this presented to the wind more of a solid surface, thus creating greater resistance to it and stopping the sails from going round too fast. Opening the shutters, which was called “spilling the wind”, had the opposite effect. Each shutter was mounted on hinges bar and all were connected by a crank to a rod travelling down one side of the stock. This rod was in turn connected to a spring on the heel of the sail, near the canister. The tension of the spring needed to be adjusted periodically, using a lever at the tip of the sail. The sails had to be adjusted individually and this again required stopping the mill.
Patent sails were invented by William Cubitt in 1807. In these there were two rows of shutters, one each side of the stock, and the rod to which the shutter bars was linked was connected to a system of levers mounted on the poll end and called a spider because of its resemblance to one; this in turn was connected to the “striking rod”, which passed through the hollow centre of the usually iron windshaft. Moving the rod backwards or forwards by means of a rack and pinion mechanism or a “rocking lever” opened and closed the shutters in a manner similar to an umbrella. The rack and pinion was connected to a chainwheel, or “striking wheel”, from which a chain was hung (as it would be on the lever) with weights attached to it. When the wind rose it would lift the chain, pull back the rod and open the shutters; when it dropped, the rod would travel forward and close them. The weights would counter the action of the wind and so control sail speed. None of the above methods really achieved complete automation as in a patent sail it was necessary to add or remove weights, sometimes replacing one with another of a different size, depending on wind strength.
The drive was taken from the sails to the stones via shafting and a series of cogged gearwheels, the gears being wedged onto the shafts. Internally, going downwards, post mills had a bin floor which contained the bins feeding the millstones, a stone floor and a spout floor where the ground corn was discharged from spouts into first the dressing machinery and then the sacks in which it would be transported to the bakery. The windshaft and the brake- and tailwheels (see below) were so positioned as to be half on the bin floor and half on the stone floor. Smock and tower mills had an additional floor under the cap, called the dust floor, though this could sometimes be combined with the bin floor. Its purpose was to allow dust and grease from the workings in the cap to accumulate at this level rather than fall to the lower floors and contaminate the flour production process. In a tall mill there might also be one or more floors under the spout floor, giving additional storage space and perhaps with an office for the miller.
The windshaft was usually positioned at an angle of rather less than forty-five degrees, to allow the sails to clear the body of the mill. On it was mounted a large gearwheel called the brakewheel, which in early post mills drove directly a single pair of stones by means of a pinion called a stone nut on a spindle or quant cemented into the upper millstone. Post mills generally have the brakewheel mounted further forward on the windshaft than in a tower-type mill. Whether or not this was always the case it allowed the fitting at some point of a second large gearwheel, called a tailwheel, further back on the shaft so that two pairs of stones, one in the head or breast of the mill and one in the tail, could be driven. The brakewheel is so called because a brake is applied to it by means of a lever operated by a rope or chain to stop the mill.
Smock and tower mills generally did not have tailwheels and the brakewheel was at roughly mid-length on the windshaft. In primitive tower mills a single pair of stones was driven straight from the brakewheel, as in post mills, but later when more stones were added the drive became more complex. The brakewheel engaged with a horizontal gear called a wallower mounted on the upright shaft. The lower end of the latter carried another, larger, horizontal gear known as the great spur wheel. Via stone nuts and quants as before, the spur wheel could drive the stones either from above (overdrift), or below (underdrift) as was the practice in watermills. Beneath the great spur wheel the upright shaft turned in a timber let into the floor or, more usually, set between two longer timbers like the cross-bar of a letter H, the whole assembly forming a kind of dummy floor just below the ceiling. Stone nuts could be taken out of mesh with the great spur wheel, if desired, by hand through adjusting the lower or upper (depending on whether the stones were overdrift or underdrift) bearing of the quant with wedges or, if underdrift was the case, by raising a “jack ring” positioned underneath them using a cranking mechanism. In underdrift mills the lower bearing of the quant (and that of the upright shaft in all mills) was called the “footstep bearing”.
Stones were arranged in pairs, of either Peaks (quarried in the Peak District of Derbyshire and used for grinding barley or animal feed) or French Burrs (of finer texture, mainly from La Ferte-ses-Jouarre in France, and for grinding wheat). The upper millstone, the “runner” stone, turned while the lower, the “bedstone”, remained stationary and the corn was ground between the two. The bedstone was sunk into the floor, secured in place with wedging and packing, and supported by a framework of timbers off the main floorbeams. The lower surface of the runner and the upper surface of the bedstone were marked with a pattern of furrows whose sharp edges acted like a pair of scissors to cut open the husks of grain and allow the valuable substance inside to escape and be ground. The runner and bedstone were not actually in contact with one another but the distance between them was very fine and could be adjusted, as we shall see. The edges of the furrows, which could become blunted with time, needed to be kept trimmed and this was done using special tools called bills and thrifts. The process was known as “dressing”, not to be confused with dressing the meal, and could be carried out either by the miller or grinder, if sufficiently skilled at the task, or a specialist stone dresser.
Some later or modified post mills had a wallower, upright shaft and great spur wheel which drove two pairs of stones side by side in the breast, usually from below. The tailstones were retained and driven from the tailwheel as before.
The grain was emptied into the bins at the top of the mill and travelled down chutes into hoppers mounted on frames, called horses, on the wooden tuns or vats which encased the runner stones. These casings could be circular, octagonal or hexagonal. From the hopper the grain would trickle down an inclined hinged trough called a shoe into the eye, the circular hole in the centre of the runner stone where the quant was fixed in place. To speed up the flow of meal to the stones the shoe was agitated by a cam, called a damsel (because of the amount of noise women were believed to make!). These could be extensions of the quant or mounted separately on the stone casing. After the grain fell through the eye it passed between the stones, centrifugal force spreading it out over the surface of the bedstone until, fully ground, it fell between the stone and its casing and down the spout into a sack from which it was then emptied into the bin which fed the dresser, the latter also having spouts for each of the three components of the meal. The stone tended to rise as it turned faster, and the distance between the upper and lower millstone could be adjusted to vary the texture of the meal. This was done automatically by a pair of governors such as those used by James Watt, who in fact is said to have got the idea from windmills, on his steam engines to regulate their speed. On the spout floor the stone spindles, or extensions to them in an overdriven mill, turned in bearings on timbers called bridge trees, which were pivoted in wood or iron hangers. Connected by links to the bridge tree or to the brayer, a beam morticed into it at a right angle, was an iron lever called a steelyard, one end of same being forked and fitted round a collar on the spindle of the governor which was linked to its two arms, on whose ends were weights for balancing. As the runner stone’s speed increased centrifugal force caused the arms of the governor to fly outwards and the collar to rise. This lifted the steelyard which in turn lowered the bridgetree, decreasing the distance between the runner and bedstones. There was provision for fine adjustment to be made by hand, using screws, if the miller, feeling the flour as it emerged from the spouts, felt the quality was not good enough. A bell alarm warned him if the corn was running out (should that happen, friction between the runner and bedstone risked sparks and thus fire). These were operated by a variety of methods, one of the most common being to swivel the bell into a horizontal position and keep it there by means of a string which passed through the hopper and was held down by the weight of the corn in it. When the corn started to run out, the string slackened and the bell fell against the quant or the damsel, which rung it. The miller would then feed in more corn or stop the mill.
A hoist was needed to lift the sacks of corn up through the mill to the bins. In post mills it was located within the apex of the roof on the bin floor and driven from the cogs of the brake- or tailwheel, or by belt from a pulley on the windshaft. In smock or tower mills it was usually on the dust floor and driven by friction from the wallower, which bore on a pulley on a layshaft carrying an iron bollard to which the sack chain was attached. As the spindle turned it wound up the chain and hauled up the sacks hung on the latter. They passed through a hinged flap, called a sack trap, in each floor on their way to the top; the flap would be pushed open by the ascending sack and then fall back into place again. Sack hoists were thrown in and out of gear with their driving wheel by a lever, operated by a rope or chain, or by (dis)engaging a roller with a slack belt which tensioned it.
Ancillary machinery was needed to “dress” the meal, the substance resulting from grinding the corn, that is separate it into its three basic components – fine flour, coarse flour and bran. Originally this was done in a bolter, which consisted of an inclined cylindrical wooden frame mounted on a revolving spindle and covered by a woollen sleeve. The mesh of this sleeve was fine at the top becoming coarser in stages towards the bottom. The meal was fed in at the top end and as the cylinder turned the fine flour passed through the first mesh; the coarser flour was collected in the next while the bran tailed out at the end. To assist the passage of the flour through the cloth a number of wooden bars were positioned so that the cylinder frame would rap against it and be vibrated. The bolter was superseded to some extent by the wire machine, also known as the dresser, flour machine or flour dresser (though these terms could also be applied to bolters), which although operating on the same principle differed in that it had a fixed cylinder covered with various meshes of wire. Mounted on the spindle were four or more brushes which forced the flour through as the mesh as it turned. The jog scry was a third form of flour dresser, which consisted of an inclined chute covered with several grades of mesh. One end was hinged while the other was agitated by cams so that it acted as a sieve. Other machines found were oat crushers (usually of metal and installed during nineteenth- and twentieth-century modifications), smutters which removed the black fungus that sometimes attacks wheat, and grain cleaners, essentially the same type of machine as a smutter with a fan for blowing away dust.
Some mills had several dressers and a jog scry as well although not all of this equipment may have been in use at the time they were examined.
Generally these machines were driven from the brake- or tailwheel in a post mill and the great spur wheel, or a separate “crown wheel” on the upright shaft, in a smock or tower mill via a pinion, layshaft and one or more (as required) pulleys.
Early gearwheels were solid with crude pegs for cogs; these were known as “cow-pop”, from the resemblance to a cow’s udder, gears or trundle wheels. Later the wallower and stone nuts were of “lantern pinion” type, with a number of staves set between two flat horizontal discs. The next development was the compass arm wheel, with four or more arms radiating from the hub like the points of the compass and passing right through the shaft. Then came clasp-arm wheels where two sets of parallel arms at right angles to each other gripped the boss on the windshaft where the wheel was located, being secured in place with packing and wedges. Clasp-arm wheels could be converted from compass arm.
All-iron wheels were usually of compass-arm type in hat the arms were radial. There were some wheels of composite type, with a wooden rim, lands and cants as in a clasp-arm wheel and eight iron compass arms. Wooden clasp-arm wheels often had a ring of iron cogs bolted on, the original wood cogs having been sawn off.
Drives from a steam engine, where installed, fell into two main types. In one the drive was via an external pulley on a layshaft which entered the mill and ended in a nut meshing with the great spur wheel. In the other the engine could be within the mill or some structure built onto it and a secondary upright shaft took the drive to the spur wheel. With engine drives there was normally provision for detaching a few cogs from the wallower so that the sails did not turn.
Windmills were not only for grinding corn; the wind could in fact be harnessed for just about any purpose if it was technically possible. The most common alternative use was the pumping of water.