Grace's Guide is the leading source of historical information on industry and manufacturing in Britain. This web publication contains 142,061 pages of information and 227,774 images on early companies, their products and the people who designed and built them.
Note: This is a sub-section of Richard Roberts
Richard Roberts' 1817 lathe is on display at London Science Museum.
With its rigid cast iron bed, back-geared headstock, and self-acting traverse, it represents a remarkable leap forward in lathe design. The heavy duty tailstock could be set over to allow taper turning.
The lathe was used at Sharp, Roberts' works in Manchester until 1852. In 1854 it was bought by Beyer, Peacock and Co for use at their Gorton works, where it remained until 1909, when it was presented to the London Science Museum. In 1885 Richard Peacock wrote to the Patent Museum stating that the company had also bought a Roberts screwcutting machine which had been made after the lathe. This was also preserved, but was unfortunately scrapped in the 1950s.
The lathe has been studied in depth by the Rev. Dr. Richard L. Hills and described in his comprehensive biography of Roberts
The lathe has the date 1817 stamped on the cross-slide (photo 3). A question arises over the date, because the bed shows machining marks made by a planing machine. Roberts' first planing machine, made in 1817, was too small to have undertaken this work, and it seems that some years elapsed before he made a larger one. However, there may be a simple explanation for this anomaly, and this will be discussed below. Dr. Hills points out that the lathe is consistent with the description of the large lathe given by Joshua Field after his visit to Roberts on 30 August 1821.
Photo 4 clearly shows planing marks on the bed. Prior to the availability of a suitable planing machine, the guides would have been finished by a combination of chiselling, filing and scraping. Cutting metal using chisels was a hard-won skill, and is now very much a lost art. Now, referring to photo 5, at the transition between the top of the V-guide and the vertical face of the bed casting, we see clear evidence of chisel work, probably superimposed by the work of a planing machine. It therefore possible, and perhaps likely, that at a later stage in its life, the ways would have been refurbished using a planing machine, probably followed by scraping, in order to improve their accuracy or to remove the effects of wear or damage. It would be surprising if severe wear had occurred, because only the tailstock slides along the ways, not the carriage.
The headstock was located, and the tailstock guided, by one V and one flat way. This was kinematically sound, and the arrangement was also favoured by Fox of Derby and came to be very widely used. A potential drawback arises if differential wear occurs between the two guideways, as the centres of the tailstock and headstock would go out of line. Roberts reduced the scope for wear by supporting and guiding the carriage (saddle) by separate dovetail ways on the front face of the bed. Photo 12 shows the bottom dovetail. Roberts's main motivation was to avoid having the thickness of the carriage reducing the maximum diameter of workpiece that could be swung. This arrangement was never widely copied, although it was later applied by James Spencer and Co of Hollinwood and by G. Birch and Co, a small maker of high class machines in Salford, and by at least five makers in the USA for some of their lathes (including Rivett, Wade, and Porter-Cable). Joseph Whitworth and Co also used the arrangement on one of their lathes - a 'double stud turning lathe'.
Photos 6 & 7 show the drive from the lathe spindle to leadscrew. The studded plate provides a simple and compact way of obtaining 7 speed ratios, but it hardly seems conducive to smooth power transmission. It is surprising that none of the studs have broken off, especially those nearer the centre, which are very thin.
Beneath the studded plate are two pairs of spur gears which reduce the speed of the bevel pinion. There is a dog clutch which engages one or other of the bevel gears with the pinion to drive the leadscrew clockwise or anti-clockwise, with neutral in between. Photo 8 shows the mechanism which operates the dog clutch, triggered manually, or automatically when the carriage reaches a predetermined point.
As Dr Hills points out, the ratio of leadscrew to spindle speed was too low for screwcutting, but satisfactory for surfacing. Whether the transmission would be sufficiently regular to givie a fine finish is debatable, but the arrangement would at least allow pre-finishing cuts to be carried out without attention from the operator.
The leadscrew could also be driven directly by a pulley fitted inboard of the bevel gears. Holls suggests that this could have allowed the carriage to be traversed to allow keyways to be cut in components held in the lathe.
Referring to photo 9, the leadscrew engages with the bronze nut. This is geared to the hand crank spindle (top of photo) via and idler gear. The hand crank is shown in photo 10, and it will be seen that it cannot rotate because its outboard fine toothed gear wheel is fixed. This fixes the leadscrew nut, so that the rotating leadscrew will traverse the carriage. By unlatching this gear, the hand crank takes over from the leadscrew.