Mr. Ramsden employed a more complex apparatus in originating the screw of his dividing engine for straight lines, which it was essential should contain exactly 20 threads in the inch; a condition uncalled for in the circular engine, in which the equality of the teeth of the wheel required the principal degree of attention. This second screw-cutting apparatus, which may be viewed as an offspring of the circular dividing engine, is represented in plan, in fig. 611, and may be thus briefly explained.
The guide-screw G is turned round by the winch, and in each revolution moves the larger tangent wheel one tooth; the tangent wheel has a small central boss or pulley p, to which is attached the one end of an elastic slip of steel, like a watch-spring; the other end of the slip is connected with the slide s, that carries the tool t, in a right line beside the screw C, which latter is the piece to be cut; and C, is connected with the guide-screw G, by a bevil pinion and wheel, g and c, as 1 to 6.
To proportion the traverse of the tool to the interval or pitch of the screw, two dots were made on the slide s, exactly five inches asunder; and in that space the screw should contain 1"" coils, to be brought about by GOO turns of the handle. The guide-screw was moved that number of revolutions, and the diameter of p, was reduced by trial, until the 600 turns traversed slide exactly from dot to dot; these points were observed at the time through a lens placed in a fixed tube, and having a fine silver wire stretched diametrically across the same as an index.*
The late Mr. Henry Maudslay, devoted an almost incredible amount of labour and expense, to the amelioration of screws and screwing apparatus; which, as regarded the works of the millwright and engineer, were up to that time in a very imperfect state. With the view of producing screws of exact values, he employed numerous modifications of the chain or band of steel, the inclined knife, the inclined plane, and indeed each of the known methods, which however he remodelled as additions to ordinary turning-lathe with a triangular bar; a natural result, as he was then in the frequent habit of constructing that machine, and which received great improvement at his hands.
* See "Description of an Engine for Dividing Straight Lines," pages 13 to 16.
In the construction of his dividing engine for straight lines, Ramsden very closely followed his prior machine for circular lines, if wo conceive the wheel spread out as a rectilinear slide. On the one edge of the main slide which carried the work, was cut a screw-form rack, with twenty teeth per inch, which was moved by a short fixed screw of the same pitch, by means of ratchets of 50, 48, or 32 teeth respectively; the screw could be moved a quantity equal to one single tooth, or to several turns and parts, by means of a treadle. To obtain divisions which were incompatible with the subdivision of the inch into 1000, 960 or 640 parts, the respective values of one tooth, the scale was laid on the slide at an angle to the direction of motion; when the swing frame was placed to traverse the knife at right angles to the path of the slide, the graduations were lengthened; when the knife was traversed at right angles to the oblique position of the scale being divided, they were shortened. This was to a small degree equivalent to having a screw of variable length. In cutting the screw-form teeth of the rectilinear dividing engine, the entire length, namely, 25.6 inches, was first divided very carefully by continual bisection into spaces of eight-tenths of an inch, by hand as usual, and the screw-cutter was placed at zero at each of these divisions, pressed into the edge of the slide, and revolved sixteen times; after three repetitions at each of the principal spaces, the entire length was ratched continuously until the teeth were completed.
It was noticed at page 581, that of all the methods he gave the preference to the inclined knife, applied against a cylinder revolving in the lathe, by means of a slide running upon the bar of the lathe; which besides being very rapid, reduced the mechanism to its utmost simplicity. This made the process to depend almost alone on the homogeneity of the materials, and on the relation between the diameter of the cylinder and the inclination of the knife; whereas in a complex machine, every part concerned in the transmission of motion, such as each axis, wheel and slide, entails its risk of individual error, and may depreciate the accuracy of the result; and to these sources of disturbance, must be added those due to change of temperature, whether arising from the atmosphere or from friction, especially when different metals are concerned.
A rod of wood, generally of alder and about two feet long, was put between the centers, and reduced to a cylinder by a rounder or witchet (fig. 343, p. 487), attached to a slide running on the bar; the slide with the inclined knife was then applied, and the angle of the knife was gradually varied by adjusting screws, until several screws made in succession, were found to agree with some fixed measure. The experiment was then repeated with the same angle, upon cylinders of the same diameter, of tin, brass, and other comparatively soft metals, and hundreds, or it might almost be said, thousands of screws were thus made.
From amongst these screws were selected those which, on trial in the lathe, were found to be most nearly true in their angle, or to have a quiescent gliding motion; and which would also best endure a strict examination as to their pitch or intervals, both with the rule and compasses, and also when two were placed side by side, and their respective threads were compared, as the divisions on two equal scales.
The most favourable screw having been selected, it was employed as a guide-screw, in a simple apparatus which consisted of two triangular bars fixed level, parallel, and about one foot asunder, in appropriate standards with two apertures; the one bar carried the mandrel and popit heads as in the ordinary bar lathe. The slide rest embraced both bars, and was traversed thereupon by the guide-screw placed about midway between the bars; the guide-screw and mandrel were generally connected by three wheels, or else by two or four, when the guide and copy were required to have the reverse direction. The mandrel was not usually driven by a pulley and cord; but on the extremity of the mandrel was fixed a light wheel, with one arm serving as a winch handle for rapid motion in running back; and six or eight radial arms, (after the manner of the steering wheels of large vessels,) by which the mandrel and the screw were slowly handed round during the cut.