UK 05The Lineage of British Lathes

Lathes of clockmakers and instrument makers (1740–1796)

The ornamental lathe and the slide rest

In 18th-century Britain, the people who owned lathes that could cut metal accurately were not factories but clockmakers, instrument makers and wealthy people who took up turning as a hobby. The typical hobby lathe was the ornamental lathe (rose engine). A wavy plate (rosette) was fixed to the spindle and pressed against a stop while the spindle was turned slowly, so that the spindle rocked and cut flower-like patterns into the work. The Science Museum has one from about 1740, said to have belonged to the 2nd Earl of Macclesfield, who was also President of the Royal Society^[1]. What matters is that on these lathes the tool was not held by hand but supported mechanically by a slide rest. This slide rest was later carried over to the factory lathe^[2].

Tools that "copy" a screw

Clockmakers used small screw-cutting machines to cut a spiral groove in the fusee, the cone-shaped part that evens out the force of the mainspring. A nut fitted on a thread cut on the spindle, and a lever carried the nut's motion to the toolholder, so the lever ratio changed the pitch. The same type appears in a catalog of the British tool dealers Ford, Whitmore & Brunton from about 1775^[2]. On ornamental lathes, too, several "guide screws" fitted to the end of the spindle moved the work along with the spindle and copied short threads. But both were small, and they could copy a screw only as accurately as the original.

Ramsden's precision screws

The man who pursued accuracy itself was the instrument maker Jesse Ramsden. To build dividing engines, machines that engrave fine graduations on circles and straight lines, he built precision screw-cutting lathes in 1777 and 1778. As a guide he used a bar of triangular section, which becomes accurate once two faces are finished, and he fed the tool with a lead screw and change gears. Comparing the finished screw with a scale, he found it out by about one thread in 100, so he used 198:200 change gears to correct it and cut the screw again^[3]. In 1793 Samuel Bentham wrote in a patent that if the motion that advances the tool were linked by gears to the rotation of the work, the feed would be constant; but there was no drawing, and it never became a machine^[4].

What changed on the shop floor. Factory lathes of this period had wooden frames and hand-held tools. Precision lathes belonged to the world of clocks, instruments and hobbies, and were too small to machine factory machinery. Woodbury writes that the idea of a precision lathe built of metal entered the factory lathe through Maudslay, who knew many instrument makers^[3].

Maudslay's screw-cutting lathe and his pupils (1797–1849)

A lathe built entirely of metal

Henry Maudslay was foreman at the workshop of Joseph Bramah, famous for his locks, where he built by hand special machines for making locks. Around 1795 he made a slide rest integral with the tailstock, but it was still light and weak^[4]. After setting up on his own in 1797^[2], Maudslay began building machine tools entirely of metal, frames included. Until then factory machine tools had wooden frames. Metal made them not only stiffer but also made it possible to produce accurate flat guideways^[2].

Lead screw and change gears

Maudslay's first screw-cutting lathe at the Science Museum (c. 1800) has a bed of two parallel triangular bars. The headstock sits on one bar, the slide rest runs straddling both, and a lead screw between them drives the slide rest. The toolholder on the slide rest has an infeed screw with micrometer graduations, and the lead screw was connected to the spindle by change gears (now lost)^[5][3]. According to Roe, at first the lead screw itself was changed for each pitch to be cut^[4].

The second machine, built soon after, had a cast-iron bed, a lead screw of 30 threads per inch and 28 change gears of 15 to 50 teeth. An intermediate gear was carried on a swinging arm so that any combination would mesh, and sample screws from 16 to 100 threads per inch survive^[2][4]. Cutting many pitches with a single lead screw and change gears. Making this combination usable as a factory tool was Maudslay's greatest achievement. Roe cites the fact that contemporaries called the slide rest "Maudslay's go-cart" as evidence that Maudslay was regarded as its originator^[4].

Master screws, flat surfaces and measuring tools

To cut an accurate screw, you first need an accurate lead screw. Maudslay made a tool that produced an "original screw" by turning a cylinder against an inclined knife, so that the knife advanced along the groove it had cut itself^[6]. He used the best of these original screws in wood or soft metal as temporary lead screws to cut steel lead screws, and reduced the error further by reversing the lead screw end for end and by feeding with the average of two lead screws^[3]. He checked the flatness of guideways by scraping three surface plates against one another at the same time, and put the plates on the workbenches for his men to use^[4][2]. As a standard of length he used a bench micrometer of 1805, the "Lord Chancellor"^[7]. Only with all three—guideways, lead screw and measuring tools—in place did the lathe become a machine that could "create" accuracy.

Maudslay's firm also sold treadle lathes with triangular-bar beds to amateurs and small workshops. They remade the slide rest, until then an accessory of the ornamental lathe, as an engineer's tool^[8][2]. On the ornamental lathe side, too, Holtzapffel of London built a lathe in 1815 that, instead of rocking the spindle, used rosettes on a rear shaft to move the slide rest and cut patterns^[2].

Roberts' back gear and automatic feed

Richard Roberts, who worked in Maudslay's works for two years, set up on his own in Manchester in 1816^[11], and in 1817 built a lathe much like today's engine lathe. The machine at the Science Museum has a center height of 9.5 inches and 6 feet between centers, with a stiff bed of three vertical plates joined by cross braces. The spindle has back gear, which reduced the speed of the four-step cone pulley through gears for heavy cuts. The carriage was fed automatically by the lead screw from one end of the bed to the other, and the feed disengaged when it hit a stop^[2][3]. This lathe was used in Roberts' works until 1854, then at the works of the locomotive builder Beyer, Peacock, and entered the museum in 1909^[2]. Woodbury writes that by 1817 factory lathes in the true sense were in use^[3].

Fox, Clement and protecting the lead screw

James Fox of Derby, who built lace machinery and also sold machine tools, built a large lathe 22 feet long in 1830. When not cutting threads, it fed the carriage with a rack, and moved the carriage with a separate feed shaft and worm, using the lead screw only for thread cutting. If the lead screw does not wear, thread-cutting accuracy lasts longer^[3]. Fox's lathes went to France, Germany and Poland, and a Polish works for which Fox supplied a complete set of machines in 1826 still survives as a museum^[2][3]. Joseph Clement, a skilled draftsman, built a screw-cutting lathe with two treadles on a 12-foot bed so that two men could do different jobs on the same lathe, and also built a three-tier compound slide rest for Charles Babbage, of calculating-engine fame^[2].

Whitworth's box bed and half nut

Joseph Whitworth, who had worked under Maudslay, Holtzapffel and Clement, set up on his own in Manchester in 1833^[4][3]. His 1835 patent included a half nut to engage the carriage with the lead screw at will, and an automatic cross feed driven from the lead screw. With an automatic cross feed, faced surfaces come out even^[3]. In 1837 he even patented a mechanism that varied the feed as the tool approached the center when facing, in an attempt to keep the cutting speed constant. It is the same idea as "constant surface speed" on today's CNC lathes^[3].

Whitworth then dropped the triangular bars for a hollow box-section bed. For the same weight it is much stiffer than triangular bars, and it also protects the lead screw from dirt and damage. His 5-inch lathe of 1843 combined the box bed, the half nut, cross feed from the lead screw and a conical spindle bearing that takes thrust^[2]. Woodbury counts Whitworth's change of the framework from slender "architectural" forms to box sections, gaining stiffness by adding metal, as a major contribution on a par with his standards of accuracy^[3].

What changed on the shop floor. Screws went from things made one at a time with file and chisel to things "cut" on a lathe by calculating the change gears. Screws cut from the same lead screw fitted one another, so bolts and nuts could be standardized within a works. Back gear made it possible to machine heavy iron parts, and automatic carriage feed made it possible to turn long shafts to the same diameter without relying on the worker's hand. Parts for steam engines, textile machinery and locomotives were made on these lathes^[3].

From Whitworth's era to the capstan lathe (1850–1913)

The peak of 1851, and a change of leadership

At the 1851 Great Exhibition in London, Whitworth's machine tools were in a class of their own. While 20 other firms showed one to three machines each, Whitworth displayed 23 items, including machines for turning, planing, shaping, drilling, gear cutting and screw cutting and a measuring machine; at the 1862 exhibition, too, he led 60 firms and took up a quarter of the space for machine tools^[2]. Woodbury calls Whitworth in 1850 the most important machine tool builder in the world^[3].

From about the same time, however, new forms of lathe began to be born in the United States: the turret lathe, which arranges tools on a revolving head, and the automatic screw machine, which does everything by cams (see the chapter on American lathes). Gilbert and Woodbury hold that around 1850 the lead passed to the United States and later to Germany^[2][3], and Woodbury writes that the very high regard for Whitworth's designs may have been one reason^[3]. Whitworth himself patented a copying lathe for gun barrels in 1855 and a taper-turning lathe with automatic tool feed in 1857, so he did have the idea of special-purpose automatic lathes^[3].

The feed shaft and the apron

The details of the engine lathe also advanced in Britain. At the 1862 exhibition, Shepherd, Hill & Co. showed a mechanism that took longitudinal and cross feed from a feed shaft at the back of the bed through cone clutches. If the cut was too heavy the clutch slipped and protected the machine, and the lead screw could be reserved for thread cutting^[2]. Whitworth's 6-inch lathe of about 1870 had eight speeds from a four-step cone pulley and back gear, gathered the half-nut lever and the feed selection on the front plate of the carriage (the apron), and used a rack under the bed for hand traverse^[14][2]. The layout of the controls on today's engine lathe was largely settled by this time.

Capstan lathes: many identical parts

At the end of the 19th century, to mass-produce brass parts such as taps, and parts for sewing machines and bicycles, lathes that changed tools by turning a toolhead spread in Britain too. The "brass finisher's lathe" of about 1890 by Smith & Coventry of Manchester had a revolving toolhead (capstan) holding several tools, so the next tool could be brought in without stopping the machine^[20]. In 1892 A. Muir patented a four-sided capstan toolholder to fit an engine lathe^[2].

At the center of this movement was Alfred Herbert of Coventry. Herbert and William Hubbard bought a small engineering works in Coventry in 1889 and began making polishing lathes and hand lathes for bicycle factories^[18]. The Science Museum has a Herbert & Hubbard power lathe of the same year (four-step cone pulley and back gear)^[19]. The firm became Alfred Herbert Ltd. in 1894, and in 1895 it built the No.4 capstan lathe. It had a chasing saddle that could cut both right- and left-hand threads, with a single lever withdrawing the tool and the nut together. More than 24,000 of this type had been built by 1961^[18]. In 1899 came a fully automatic screw-cutting machine fed with stock from a magazine, and in 1900 a single-pulley headstock driven by one belt in place of a cone pulley, with speeds changed by gears inside the headstock^[18]. It was a time when tools were changing from carbon steel to high-speed steel and cutting speeds roughly tripled, so higher speeds and easier speed changes were in demand^[2].

Flat turrets and exchange with America

At the 1901 Glasgow Exhibition, Smith & Coventry showed a flat turret lathe, which The Engineer described as "a challenge to the American makers." The spindle was hollow so that bar stock could be drawn through by hand, and the rack was fitted on the carriage side to lighten the carriage^[22]. The flat turret lathe was a type introduced by Hartness in the US in 1891 (see the chapter on American lathes), and around 1909 Dean, Smith & Grace of Keighley were building a 21-inch flat turret lathe of the Hartness type^[24]. At the same Glasgow exhibition, John Lang & Sons of Johnstone, Scotland, showed a lathe whose bed had square guideways instead of dovetails. Their shell lathe cut with two carriages, left and right, at the same time, and turned the curve of the shell nose with a radius bar^[22].

What changed on the shop floor. With the capstan lathe, small-part work, where a skilled turner had changed tools one at a time, became "setup work": once the stops were set, the same dimensions came out every time. In Britain, the type with the revolving toolhead on a ram (a reciprocating slide) for small work is generally called a capstan lathe, and the type with the toolhead on a saddle that moves with it, for larger work, a turret lathe.

War and mass-production lathes (1914–1945)

Lathes for turning shells

In the First World War Britain needed huge numbers of shells, and lathe makers built shell lathes. In January 1916 Engineering described the Colchester Lathe Company's 9.5-inch lathe built only for rough-turning shells. As it turned only one size of shell, the spindle had a single speed and no overhead countershaft was needed, and it was a sturdy machine that even unskilled workers could operate^[25]. By narrowing the general-purpose lathe down to "a machine that does only one job," makers made up for the shortage of skilled men (for war and machine tools as a whole, see Chapter 3).

Electric motors, cemented carbide and the preselector headstock

From the end of the 19th century electricity became available, and individual drive, with a motor on each machine instead of belts from overhead line shafts, spread. In the 1920s tools tipped with cemented carbide (tungsten carbide) also appeared^[2]. As cutting speeds rose, the time spent changing the spindle speed every time the tool or the diameter changed became costly.

Herbert's answer, in 1934, was the Preoptive (preselector) headstock. While one cut is in progress, the operator selects the next speed on a dial, then changes over with a single lever^[18]. A 1944 textbook explains that the Preoptive headstock of the No.4 capstan lathe had eight speeds from 42 to 1,000 rpm and could be changed to any speed while cutting^[26]. The setup of a capstan lathe now included a tooling chart saying "this tool at this speed," and the combination of tool and speed could be kept by operating the machine rather than by the operator's feel.

Wickman's multi-spindle automatics

To make the same part even faster, you use a multi-spindle automatic with several spindles. The machine was born in the US in the 1890s^[2], but Wickman of Coventry developed its own multi-spindle automatic from the early 1930s and shipped the first machine in 1937. Its distinctive feature was an "automatic setting" system that allowed setup without making special cams for each job. In 1939 the firm built a new factory at Tile Hill, and during the Second World War it built more than 300 five-spindle automatics and had about 300 more built by outside firms, sending its own engineers to help^[30].

What changed on the shop floor. On cam-type automatics, making the cams was itself a major part of the setup. If setup could be done without special cams, multi-spindle automatics could be used even for orders of a few thousand pieces. Together with the preselector headstock, the British lathe innovations of this period aimed at "reducing changeover time."

Postwar: from model engineering lathes to NC (1946–1979)

Myford ML7: a lathe for home and school

Myford Engineering Co. of Beeston (Nottingham) was a company started by Cecil Moore, and by 1935 at the latest it was advertising lathes and accessories in a model engineering magazine^[28][29]. In February 1946 the model engineering magazine The Model Engineer introduced Myford's new lathe, the ML7. It was a small lathe with a 3.5-inch center height and 20 inches between centers, but it incorporated ideas gained from wartime production: a gap bed, ground guideways, a built-in motor and a back-gear shaft placed below the headstock. The article named the gap bed, a large boring table and a hollow tailstock barrel as good features of British lathes^[31]. The article was written for model engineers, and the Science Museum also has an ML7^[32].

German automatics and American lathes

Foreign lathe technology also came into postwar Britain. J. E. Maclaren of BSA Tools and others moved the machine shop of Index-Werke of Esslingen, Germany, to Redditch as war reparations, and built Index automatics at the Index Automatic Machine Co.^[33] (for Index, see the chapter on German lathes). In 1959 John Lang & Sons formed Lang Gisholt Machine Co. with Gisholt of Wisconsin, USA, and began building Gisholt machines at the Johnstone works as well^[36]. At the 1952 International Machine Tool Exhibition at Olympia, London, Dean, Smith & Grace showed a 26-inch lathe suited to both cemented carbide and high-speed steel (20 hp, 24 speeds with three levers, 48 feeds)^[34]. Headstocks had become all-geared, and speeds were selected with levers.

Colchester and Harrison

Among general-purpose engine lathes, it was Colchester that grew. It joined George Cohen's 600 Group in 1954^[35][17], and in 1967 it was recognized for its export record and for "technological innovation in the methods of manufacturing lathes" with the Queen's Award to Industry^[37]. The Triumph 2000, whose first production unit was completed in February 1968, was a general-purpose lathe with angular styling, controls designed around the operator's hands and bright paint, affordable even for small workshops and skilled amateurs^[38]. In 1971 the 600 Group also bought T. S. Harrison & Sons, and so came to own Colchester and Harrison, the two leading British brands of engine lathe^[35].

Plugboards and NC

Automation moved from cams and levers to electrics. Herbert-BSA's NC single-spindle lathe, the Batchmatic 50, was a machine for turning small quantities of parts quickly and cheaply, and was first shown to the American market at a US trade show in 1971^[39]. Herbert's Auto Robot capstan lathe of around 1977 controlled the machine's motions with an electrical sequence, set up on a plugboard. The three cross slides and the speed changes were hydraulic and the turret pneumatic, and its eight tools could be reused across small-lot jobs. H. W. Ward capstan lathes also got a plugboard control called Gearaid^[41][42]. It was a time when NC, programmed with numbers, and plugboards, sequenced by wiring, existed side by side (for NC as a whole, see Chapter 4).

Herbert in trouble

In 1961 Herbert was Britain's largest machine tool maker, with 7,336 employees^[18]. But in the 1970s it ran into difficulties, and in July 1975 the government announced up to £25 million of long-term funding under the Industry Act 1972. Most of the funds were to take the form of shares, to be held by the National Enterprise Board (NEB), which was about to be set up^[40].

★From 1980: contraction and reorganization

In July 1980 the final closure of Alfred Herbert was announced, and in October it went into voluntary liquidation^[43][44]. It was the end of Britain's largest lathe maker, which had lasted more than 90 years together with the capstan lathe.

Instead, it was a Japanese maker that started up a large new NC machine tool factory in Britain. Yamazaki Mazak set up a British company in July 1980 and in June 1987 started its European plant (a CIM plant) in Worcester. In March 1988 NC machine tools built in the British plant were shipped to Japan for the first time^[46]. The Worcester plant still builds machines for Europe today^[48].

British lathe brands also went CNC. Colchester's 2000L is a CNC lathe with a FANUC 6-TB NC unit, and the Science Museum holds it as an example of CNC, which turns the dimensions on a drawing directly into tool motion without relying on cams or the operator's skill^[47]. The company kept changing shape: Colchester and Harrison merged into 600 Lathes in 1992, joined the chuck maker Pratt Burnerd International to form 600 UK in 2009, became Colchester Machine Tool Solutions in 2018, and left the 600 Group in 2022^[17][35]. How to choose a lathe today is covered on the Lathes, NC programs on the NC programming, and cutting tools on the Tooling pages.

Links with Japan

Lathe building in Japan began in the British style. In May 1889 Ikegai Shotaro started work in Shiba-Kanasugi, Tokyo, in a rented shop with two machines, each an English-type lathe, and at the end of that year he and his younger brother Kishiro built a British-style 9-foot lathe for their own use. It is the oldest surviving powered lathe made in Japan and is in the National Museum of Nature and Science as Mechanical Engineering Heritage No. 53^[52][53][55]. It was not until 1905 that Ikegai moved to mass-producing American-type lathes based on specifications from an American engineer; until then it built mainly British-type lathes^[54]. During the First World War, conversely, Ikegai exported lathes to Britain (five in 1915)^[54]. For events on the Japanese side, see the timeline of Chapter 1 of the Japanese archive.

The postwar link is NC. Colchester's 2000L CNC lathe carried a FANUC NC unit^[47]. Then in 1987 Yamazaki Mazak's Worcester plant started up, and the following year British-built NC machine tools were shipped to Japan^[46]. Part of the work of building machine tools in Britain had come to be done in a Japanese maker's factory.

FAQ

Did Maudslay invent the screw-cutting lathe?

The parts themselves (lead screw, change gears and slide rest) already existed on the lathes of clockmakers and instrument makers before Maudslay (Ramsden's precision screw-cutting lathe of 1777, for example). Maudslay's achievement was to bring them together around 1800 in an all-metal lathe of a size usable in a factory, able to cut threads of many pitches with a single lead screw and 28 change gears. His first lathe is at the Science Museum in London.

What did Roberts and Whitworth change in the lathe?

In his 1817 lathe Richard Roberts used back gear (gearing that reduces the spindle speed for heavy cuts) and automatic feed of the carriage from one end of the bed to the other. Joseph Whitworth made the lathe stiffer and easier to operate with the half nut and automatic cross feed of his 1835 patent, and with a hollow box bed in place of triangular bars.

What is Herbert's Preoptive headstock?

A headstock introduced by Alfred Herbert in 1934 that lets the operator preselect the spindle speed. The next speed is selected while cutting and engaged with a single lever. On the No.4 capstan lathe, eight speeds from 42 to 1,000 rpm could be changed while cutting, reducing the setup time each time the tool was changed.

What kind of lathe is the Myford ML7?

A small lathe announced in 1946 by Myford of Beeston, near Nottingham. With a 3.5-inch center height and 20 inches between centers, a gap bed, ground guideways and a built-in motor, it was introduced in a model engineering magazine as a lathe within reach of amateurs. The Science Museum also has one.