While it is not the intention of the author to assume to present in this work an exhaustive treatise on lathe design, for the reason that the scope of the plan is not extensive enough to permit it, and for the further reason that it does not seem necessary in view of the objects for which it is written, it does seem thoroughly in keeping with what is proposed, to give such facts as to some of the more important points of design as will serve as cautions to the designer, as interesting lines of thought to the machinist, and as information to the buyer of lathes.

These have been the considerations governing what has been presented in relation to back gearing and triple gearing as well as dimensions of cone steps.

There are several matters intimately connected with this subject which seem to merit still further consideration.

Any one who will take the trouble to examine, as the author frequently has, a lot of lathes in almost any machine shop and to make the most superficial calculations of their speeds, will be surprised at the amount of apparent "guesswork" that has entered into their design. The speed curve shown in Fig. 92 is a case in point. Narrow-faced back gears without any calculations, so far as one may see, of the strain which they must bear; small pinions, whose teeth are soon ground away on account of the sharp angles of their action; a lack of proper proportion between cone dimensions and back gear dimensions; and many other similar faults.

We have all seen lathes with a 3-inch vertical belt on a 5-inch cone, while the overhead horizontal belt driving the countershaft was 4 inches wide on a 15-inch pulley; when every mechanic knows that a horizontal belt will drive more than a vertical one, aside from the difference in the diameters of the pulleys being all in favor of the larger pulley.

When we consider these questions we cannot avoid the conclusion that there have been many good opportunities wasted and that the money spent for these machines should have produced much more really practical and useful machines than it has, and that they should have been capable of turning out a much larger output than we find them doing.

In designing a lathe head-stock the triangle formed by the distance from center to center of the inside V's, as a base, and the lathe center the apex, should be an equilateral triangle. Sufficient material must be provided under the largest step of the cone and the face gear to give the requisite strength and rigidity. The large step on the cone should, fill the remaining space with the exception of a sufficient clearance for the belt. The large diameter of the cone having been thus fixed the diameters of the other steps are a question of proportion, according to how many steps there are and what is to be the smallest diameter. It is common practice to make the steps of the spindle cone and the countershaft cone identical. This, on a five-step cone, will give a spindle speed (without back gears) equal to the countershaft speed when the belt is on the middle step. The spindle speed will be correspondingly faster or slower according as the belt is on the smaller or larger steps of the spindle cone, and in the same proportion as the cone steps are to each other.

The cone diameters having been fixed the back gear ratio must be made to correspond. We cannot have an arbitrary cone proportion and an arbitrary back gear ratio. Only one can be fixed, and the other must be arranged to correspond with it.

A homely proportion, but one that will come out very nearly right in practice, in determining the proper width of face for the cone steps, will be one seventh of the swing of the lathe when no triple gears are used. If triple gears are used it is common practice to make the belt a trifle narrower. The present tendency is towards wider belts and it seems to be a very proper development made necessary by modern shop, conditions and the use of high-speed steel tools. It is altogether probable that belts will be made wider rather than narrower in the future.

There is also a tendency to make the differences between cone step diameters less, which gives a larger diameter to the smaller steps, and consequently more power in the driving mechanism of the lathe and avoids the necessity for very tight belts.

Ordinarily, the width of the face gear should not be less than eight tenths of the width of the cone steps, and the width of the back gear not less than six tenths. Of course, the pitch teeth should be in proper proportion to the width of the fact, d in the larger lathes the pitch of the teeth of the face gear should be one number coarser than that of the back gear.

Usually the face gear has an outside diameter about equal to that of the largest step of the cone, but should not be larger. The outside diameter of the cone pinion should not be much smaller than the smallest cone step. These dimensions thus coming within rather narrow limits, the diameter of the back gears and that of the quill pinion will be governed by the ascertained or the arbitrary back gear ratio.

In order to avoid the interference of a large back gear with the desired form of the head-stock at this point, and to secure a symmetrical contour, the back gear shaft can advantageously be raised above the level of the main spindle. This is permissible to the extent of 1 1/2 inches on a 36-inch swing lathe.

The method of transmitting the power from the spindle to the feeding mechanism, formerly done almost exclusively with a belt on cone pulleys, one of which was upon the head shaft (located below the spindle and driven by it through the medium of gears), and the other on the feed rod. The lead screw was, of course, driven by gears so as to obtain a positive motion. In modern lathes nearly all have gear-driven mechanism for both the rod feed and the screw feed.

The former practice of reversing the feed in the head-stock is now to a large extent abandoned, and this function performed at the apron, and is much more convenient to the operator.

The mechanism for driving the lead screw in thread cutting and for a large range of feeds is now very popular and is accomplished wholly by gears with appropriate levers, shafts, and clutches. Those for driving the feed rod are usually known as "variable-feed devices" and those for thready cutting are known as "rapid change gear devices." It is not unusual to find these combined so that one set of variable-speed gears is adaptable to both functions.

These devices will be considered further on in this work, and representative inventions for these purposes will be given.