The depth of the separate cuts necessarily varies with the hardness of the material, the diameter, and the shape of the thread; but it should always be well within the cutting capacity of the edge of the tool, and the strength of the cylinder being cut into the screw, even when that is supported by the backstay. The smallest advance is required for cutting threads in steel; and it may here be said approximately, that one hundredth of an inch is a sufficient advance for the tool at the commencement, in cutting most steel screws of either angular or square threads, the advance being gradually reduced to about one thousandth at the conclusion. In all screw cutting three or four comparatively light, produce a far better result than a less number of heavy cuts, but on the other hand the traverses should not be unnecessarily numerous, to avoid wear on the edge of the tool, which for iron and steel should be constantly supplied with oil to diminish the friction.

Frequently, and especially with long delicate screws, it is found advisable to take a second or even a third cut, without advancing the tool any deeper, the inherent elasticity of both screw and apparatus affording sufficient penetration to still remove a shaving; after which, the tool is again advanced a minute quantity and so on. Sometimes it is precisely the contrary, and it is found that after the tool has received its usual minute advance, its traverse produces no cut. This may arise in part from the elasticity or yielding named, and also in part from the effect of the heat evolved by the friction in removing the metal shaving. The cut should then be tried over again more than once, with the tool still at the same depth, before risking a fresh advance; when most probably in the course of the repetitions, from the subsidence of the interference due to elasticity or heat, it will remove its shaving.

The heat produced in metal screw cutting, as in plain turning, may be lessened by lubrication and the employment of a slow speed; but a small increase in temperature is often sufficient to sensibly elongate the work, which may cause the screw to bend, producing irregularity in the thread. The elongation may render it necessary to ease the pressure from the confinement of the work between the centers, by slightly withdrawing the point of the popit head; but it is a much safer practice to suspend the cutting from time to time, to allow the screw to cool. In some cases, when the screw has been partially cut in the lathe, it is advisable to pass it through a pair of dies, carefully cut on an original or master tap, fig. 551, Vol. II. of the thread and diameter; the screw is then returned to the lathe and its further cutting proceeded with, and this alternation is sometimes repeated more than once. For these, among other reasons, it is apparent that metal screws requiring even tolerable accuracy, will not admit of being hurriedly executed.

The angular tool when placed at the correct vertical inclination required by the thread, cuts a shaving on each of its two sides, and when in perfect condition from grinding a double or angular shaving the shape of its point. As the depth of the cut increases, the two separate shavings or the two sides of the double shaving, gather up and bend over towards each other, impeding the cut and occasionally breaking the point of the tool, by a small portion of the metal removed forcing itself in between that and the work. Should the double shaving prove inconvenient, the cutter bar is slackened in the tool holder of the rest and very slightly shifted to the left, parallel to its former position, and the blade adjusted to the thread with only the left side cutting. Many, invariably adopt this system, cutting only by the one side of the tool so soon as the angular thread has acquired a little depth; either shifting the cutter bar as described, or with the slide lathe, moving it a trifling amount to the left by the traverse of the slide of the rest that is parallel with the bearers. The single shaving then produced freely escapes, until by increasing depth the right side of the tool comes once more into cut, producing the double shaving; when if desirable, the cutter bar is shifted again. As the thread approaches completion, and the advance of the tool is gradually diminished that a thinner shaving may leave a smoother surface, the last fine shavings are taken with both sides of the tool cutting.

Replacing the tool adjusted to the thread in the manner c c 2 described, presents no difficulty but requires care; the cutting edge it is apparent, also requires equally careful replacement both laterally and to the depth of cut, every time after the tool has been removed from the rest to be ground, an operation of rather frequent recurrence in metal screw cutting. The advance of the tool in the tool holder of the rest to the depth of the thread previously cut, is readily determined by the sense of touch, and is given by the hand. The correct lateral adjustment is determined by the eye, observation being generally assisted by a magnifier held in the hand above the point of the tool, and a piece of white paper laid on the slide rest below the screw. The tool is then slightly clamped, withdrawn from the work by the slide of the rest, to be sufficiently fixed, after which it is re-advanced and again examined, to observe whether the fixing has at all altered its relative position. No lateral adjustment of the tool, except that to cause the cut to fall only on one side, should be required with the cutter bar fig. 523; as this remains clamped in the rest, while the blade, which may be replaced in it with accuracy, alone is removed for grinding.

A minute width of the original cylindrical surface of the blank, is usually left to form the top of the thread, which would otherwise be too keen and liable to damage. The thread also forms a sharp knife edge at its termination with the shoulder, a; the sharp weak end is removed with the turning tool, the file or chipping chisel, according to the magnitude of the screw, until the section of the extreme end of the thread is sufficiently strong, and the screws are then polished by the methods described, pages 1072 - 3, Vol. III.