Vaucanson's Lathe
In 1751 the brilliant French inventor and engineer Jacques de Vaucanson built a pioneering industrial lathe (tour à charioter de Vaucanson), which is on display at the Musée des Arts et Métiers in Paris. Museum photograph here.
It was an early example of an all-metal industrial lathe whose cutting tool was precisely guided by slideways (i.e. a slide lathe), as distinct from the typical wooden lathes of the day, using a hand held cutting tools. It was not a versatile machine, being built for a specific type of workpiece.
The brass carriage holding the tool is traversed by a leadscrew, guided and supported by two polished wrought iron bars of square section. They are set at 45 degrees to serve as V-ways. These bars are in turn supported by two square polished wrought iron bars. The ends of the four polished bars are bolted to a square brass plate, which is attached to a wrought iron space frame. Bolted to each end of this frame is a bracket holding a dead centre, the workpiece being supported between the pair of dead centres.
No drive mechanism is evident, so presumably a cord or belt was taken round the workpiece, or the workpiece may have been mounted on a mandrel together with a pulley. Belt-driven lathes were often worked by a large hand-cranked wheel fixed alongside. If this were the case here, the lathe would have to fixed down to resist the pull from belt tension. No fixing holes are provided, but it would not be difficult to apply clamps.
The fine-pitch leadscrew which traversed the carriage was rotated by hand using a removable crank, which could be fitted at either end.
It is a very early example of a 'slide lathe' with the cutting tool constrained by guides to move in the 'X' and 'Y' planes, and able to traverse the whole length of the workpiece. It was also an early example of a precision all-metal machine tool made purely for heavy duty commercial use, as distinct from those precision machine tools made for the amusement of wealthy owners. This difference is underlined by the rational approach to the finish of the components: a high standard of finish was applied where it was needed, but no unnecessary effort has been expended on the parts of the frame which are not used for location purposes, these being left as-forged.
However, it was by no means a versatile machine, being solely constructed to turn rolls, possibly calendar rolls, up to about 30cm diameter and about 1 metre long. Apparently the rolls were copper-covered, and in service silk cloth was pressed between them to give a moiré appearance. Presumably the rolls would be turned parallel, because it was not well-suited to turning workpieces with stepped diameters - the toolholder lacked rigidity to allow much outwards extension.
One of the adjustable centres would be partially unscrewed to install the workpiece, and then screwed in to support it with the required degree of contact pressure. This is not ideal, because it demands a very high standard of concentricity between the threaded portion of the adjustable centre and the cone machined on the end, by no means easy to achieve in practice. Any lack of concentricity would upset the alignment of the axis of the workpiece in relation to the axis of movement of the tools. This would result in the workpiece being turned tapered, necessitating adjustment of the centre on a trial and error basis.
Removal of the machined workpiece was the reverse of the installation process, except that even more care was required to prevent damage to the surfaces, noting that the workpiece had to be manipulated by two people to remove it from the cage-like frame. It was presumably removed to a simple type of lathe for polishing to remove tool marks.
The rolls probably comprised a copper sleeve tightly fitted onto an iron armature, having small diameter bearing journals on the end. The periphery of the copper sleeve would need to revolve with a high degree of concentricity relative to the bearing journals. Ideally this would be achieved by finish-turning the journals and the periphery at the same setting in the lathe. This was not possible in this lathe. Therefore it is likely that the spindle's journals were machined 'between centres' in another lathe (conical recesses having first been drilled in both ends of the spindles). These same recesses would then be used to locate the roll, on its spindle, in Vaucanson's lathe. Various sources states that the rolls were about 1 metre long. The centres are stated to be about 1.4m apart[1], implying that the spindles would project about 20cm beyond the roll at each end. The same source states that the workpieces were cylinders for Vaucanson's calandre.
If a shorter workpiece was to be accommodated, it appears that the bracket for the fixed centre could be relocated to the 'floating' bracket seen in photo 7. However, this would not be undertaken lightly, because of the work involved in realigning the adjustable centre. Furthermore, the wrought iron frame to which the 'floating' bracket is clamped does not have prepared surfaces to accommodate the bracket. Puzzling.
If a workpiece of smaller diameter was to be machined, the tool on its cross-slide could in principle be extended out. However, this would involve intolerable loss of rigidity due to the increased overhang. Instead, the pair of centres could be repositioned closer to the slideways. An inconvenient arrangement.
The lacks of obvious wear on the slideways and the support centres suggests that it did not hve a very active life.
Felix Rosat was responsible for the production of the lathe, and of Vaucanson's Drilling Machine.[2]