Chapter XII: Examples in Lathe Work (3)

It will be noticed here that the use of the emery stick and oil is entirely dispensed with; but for a less fine polish it may be used, providing it be kept in quick motion radially on the work. The objection to its use is that if there be any speck on the work it is apt to cut a streak or groove following the spot like a comet's tail.

TURNING TAPERS.--There are five methods of turning outside tapers; 1st, by setting over the tailstock of the lathe; 2nd, by the use of a former or taper turning attachment such as was shown in Fig. 508; 3rd, by the use of a compound slide rest; 4th, by means of a lathe in which the head and tailstock are upon a bed that can be set at an angle to the lathe shears on which the lathe carriage slides; and 5th, by causing the cross-feed screw to operate simultaneously with the feed traverse.

Referring to the first method, it is objectionable, inasmuch as that the work axis is thrown at an angle to the axis of the lathe centres, which causes the work centres to wear rapidly, and this often induces them to move their positions and throw the work out of true. Furthermore, the tailstock has to be moved back in line with the live spindle axis for turning parallel again, and this is a troublesome matter, especially when the work is long.

Fig. 1214 shows the manner in which the lathe centres and the work centres have contact, L being the live and B the dead centre; hence C C is the axis of the live spindle which is parallel to the lathe shear slides, which are represented by G; obviously A is the work axis. The wear is greatest at the dead centre end of the work, but there is some wear at the live centre end, because there is at that end also a certain amount of motion of the work centre upon the live centre. Thus, in Fig. 1215, let _c_ represent the live centre axis, _a_ the work axis, D the lathe face plate, and E F the plane of the driver or dog upon the work, and it is obvious that the tail of the driver will when at one part of the lathe revolution stand at E, while when diametrically opposite it will stand at F; hence, during each work revolution the driver moves, first towards and then away from the face plate D, and care must be taken in adjusting the position of the driver to see that it has liberty to move in this direction, for if obstructed in its motion it will spring or bend the work.

To determine how much the tailstock of a lathe must be set over to turn a given taper, the construction shown in Fig. 1216 may be employed. Draw the outline of the work and mark its axis D, draw line C parallel to one side of the taper end, and the distance A between this line and the work axis is the amount the tailstock requires to be set over. This construction is proved in Fig. 1217, in which the piece of work is shown set over, C representing the line of the lathe ways, with which the side F of the taper must be parallel. D is the line of the live spindle, and E that of the work, and the distance B will be found the same as distance A in Fig. 1216.

It may be remarked, however, that in setting the tailstock over it is the point of the dead centre when set adjusted to the work length that must be measured, and not the tailblock itself.

Other methods of setting tailstocks for taper turning are as follows: If a new piece is to be made from an old one, or a duplicate of a piece of work is to be turned, the one already turned, or the old piece as the case may be, may be put in the lathe and we may put a tool in the tool post and set the tailstock over until the tool traversed along the work (the latter remaining stationary) will touch the taper surface from end to end.

If, however, the taper is given as so much per foot, the distance to set the tailstock over can be readily calculated.

Thus, suppose a piece of work has a taper part, having a taper of an inch per foot, the work being three feet long, then there would be three inches of taper in the whole length of the piece and the tailstock requires to be set over one-half of the three inches, or 1-1/2 inches. It will not matter how long the taper part of the work is, nor in what part of the work it is, the rule will be found correct so long as the tailstock is set over one-half the amount obtained by multiplying the full length of the work per foot by the amount of taper per foot.

If we have no pattern we may turn at each end of the part that is to be taper a short parallel place, truing it up and leaving it larger to the same amount at each end than the finished size, and taking care that the parallel part at the small end will all turn out in the finishing. We then fasten a tool in the lathe tool post, place it so that it will clear the metal of the part requiring to be turned taper, and placing it at one extreme end of said part, we take a wedge, or a piece of metal sufficiently thick, and place it to just contact with the turned part of the work and the tool point (adjusting the tool with the cross-feed screw), we then wind the rest to the other end of the required taper part, and inserting same wedge or piece of iron, gauge the distance from the tool point to the work, it being obvious that when the tool point wound along is found to stand at an equal distance from each end of the turned part, the lathe is set to the requisite taper.

Figs. 1218 and 1219 illustrate this method of setting. A represents a piece of work requiring to be turned taper from B to C, and turned down to within 1/32 inch of the required size at E and F. If then we place the tool point H first at one end and then at the other, and insert the piece I and adjust the lathe so that the piece of metal I will just fit between the tool point and the work at each extreme end of the required taper part, the lathe will be set to the requisite taper as near as practicable without trying the work to the taper hole. The parallel part at the small end of the work should be turned as true as possible, or the marks may not be obliterated in finishing the work.

Fig. 1220 (from _The American Machinist_) represents a gauge for setting the tailstock over for a taper. A groove is cut as at E and D, these diameters corresponding to the required taper; a holder A is then put in the tool point, and to this holder is pivoted the gauge B. The tailstock is set over until the point of B will just touch the bottom of the groove at each end of the work.

To try a taper into its place, we either make a chalked stripe along it from end to end, smoothing the chalked surface with the finger, or else apply red marking to it, and then while pressing it firmly into its place, revolve it backwards and forwards, holding it the while firmly to its seat in the hole; we move the longest outwardly projecting end up and down and sideways, carefully noting at which end of the taper there is the most movement. The amount of such movement will denote how far the taper is from fitting the hole, while the end having the least movement will require to have the most taken off it, because the fulcrum off which the movement takes place is the highest part, and hence requires the greatest amount of metal to be taken off.

Having fitted a taper as nearly as possible with the lathe tool, that is to say, so nearly that we cannot find any movement or unequal movement at the ends of the taper (for there is sure to be movement if the tapers do not agree, or if the surfaces do not touch at more than one part of their lengths), we must finish it with a fine smooth file as follows: After marking the inside of the hole with a very light coat of red marking, taking care that there is no dirt or grit in it, we press the taper into the hole firmly, forcing it to its seat while revolving it backwards and forwards.

By advancing it gradually on the forward stroke, the movement will be a reciprocating and yet a revolving one. The work must then be run in the lathe at a high speed, and a smooth file used to ease off the mark visible on the taper, applying the file the most to parts or marks having the darkest appearance, since the darker the marks the harder the bearing has been. Too much care in trying the taper to its hole cannot be taken, because it is apt to mark itself in the hole as though it were a correct fit when at the same time it is not; it is necessary therefore at each insertion to minutely examine the fit by the lateral and vertical movement of projecting part of the taper, as before directed.

A taper or cone should be fitted to great exactitude before it is attempted to grind it, the latter process being merely intended to make the surfaces even.

For wrought-iron, cast-iron, or steel work, oil and emery may be used as the grinding materials (for brass, burnt sand and water are the best). The oil and emery should be spread evenly with the finger over the surfaces of the hole and the taper; the latter should then be placed carefully in its place and pressed firmly to its seat while it is being revolved backwards and forwards, and slowly rotated forward by moving it farther during the forward than during the backward movement of the reciprocating motion.

After about every dozen strokes the taper should be carefully removed from the hole and the emery again spread evenly over the surfaces with the finger, and at and during about every fourth one of the back strokes of the reciprocating movement the taper should be slightly lifted from its bed in the hole, being pressed lightly home again on the return stroke, which procedure acts to spread the grinding material and to make the grinding smooth and even. The emery used should be about number 60 to 70 for large work, about 80 to 100 for small, and flour emery for very fine work.

Any attempt to grind work by revolving it steady in one direction will cause it to cut rings and destroy the surface.

Referring to the second method, all that is necessary in setting a former or taper attachment bar is to set it out of line with the lathe shears to half the amount of taper that is to be turned, the bar being measured along a length equal to that of the work. Turning tapers with a bar or taper-turning attachment possesses the advantage that the tailstock not being set over, the work centres are not thrown out of line with the live centres, and the latter are not subject to the wear explained with reference to Fig. 1214. Furthermore, the tailstock being kept set to turn parallel, the operator may readily change from turning taper to turning parallel, and may, therefore, rough out all parts before finishing any of them, and thus keep the work more true, whereas in turning tapers by setting the tailstock over we are confronted by the following considerations:--

If we turn up and finish the plain part first, the removing of the skin and the wear of the centres during the operation of turning the taper part will cause the work to run out of true, and hence it will not, when finished, be true; or if, on the other hand, we turn up the taper part first, the same effects will be experienced in afterwards turning the plain part. We may, it is true, first rough out the plain part, then rough out the taper part, and finish first the one and then the other; to do this, however, we shall require to set the lathe twice for the taper and once for the parallel part.

It is found in practice that the work will be more true by turning the taper part the last, because the work will alter less upon the lathe centres when changed from parallel to taper turning than when changed from the latter to the former. In cases, however, in which the parts fitting the taper part require turning, it is better to finish the parallel part last, and to then turn up the work fastened upon the taper part while it is fast upon its place: thus, in the case of a piston rod and piston, were we to turn up the parallel part of the rod first and the taper last, and the centres altered during the last operation, when the piston head was placed upon the rod, and the latter was placed in the lathe, the plain part or stem would not run true, and we should require to true the centres to make the rod run true before turning up the piston head. If, however, we first rough out the plain part or stem of the rod, and then rough out and finish the taper part, we may then fasten the head to its place on the rod, and turn the two together; that is to say, rough out the piston head and finish its taper hole; then rough out the parallel part of the rod, but finish its taper end. The rod may then be put together and finished at one operation; thus the head will be true with the rod whether the taper is true with the parallel part of the rod or not. With a taper-turning attachment the rod may be finished separately, which is a great advantage.

If, however, one part of the length of a taper turning attachment is much more used than another, it is apt to wear more, which impairs the use of the bar for longer work, as it affects its straightness and causes the slide to be loose in the part most used, and on account of the wear of the sliding block it is proper to wind the tool out from its cut on the back traverse, or otherwise the tool may cut deeper on the back than on the forward traverse, and thus leave a mark on the work surface.

Referring to the third method, a compound slide rest provides an excellent method of turning tapers whose lengths are within the capacity of the upper slide of the compound rest, because that slide may be used to turn the taper, while the ordinary carriage feed may be used for the parallel parts of the work, and as the tailstock does not require to set over, the work centres are not subject to undue wear.

If the seat for the upper slide of the rest is circular, and the taper is given in degrees of angle, a mark may be made on the seat, and the base of the upper slide may be marked in degrees of a circle, as shown in Fig. 1221, which will facilitate the setting; or the following construction, which is extracted from _Mechanics_, may be employed. Measure the diameter of the slide rest seat, and scribe on a flat surface a circle of corresponding diameter. Mark its centre, as A in Fig. 1222, and mark the line A B. From the centre A mark the point B, whose radius is that of the small end of the hole to be bored. Mark the length of the taper to be turned on the line A G and draw the line G D distant from A B equal to the diameter of the large end of the hole to be bored. Draw the line B D. Then the distance E F is the amount the rest must be swiveled to turn the required taper.

It is obvious that the same method may also be used for setting the rest.

Referring to the fourth method, by having an upper bed or base plate for the head and tailstock, so that the line of lathe centres may be set at the required angle to the [V]s or slides on which the carriage traverses, it affords an excellent means of turning tapers, since it avoids the disadvantages mentioned with regard to other systems, while at the same time it enables the turning of tapers of the full length of the carriage traverse, but it is obvious that the head and tailstock are less rigidly supported than when they are bolted direct to the lathe shears.

In turning tapers it is essential that the tool point be set to the exact height of the work axis, or, in other words, level with the line of centres. If this is not the case the taper will have a curved outline along its length. Furthermore, it may be shown that if a straight taper be turned and the tool be afterwards either raised or lowered, the amount of taper will be diminished as well as the length being turned to a curve.

Figs. 1223 and 1224 demonstrate that the amount of taper will be changed by any alteration in the height of the tool. In Fig. 1223, A B represents the line of centres of the spindle of a lathe, or, in other words, the axis of the work W, when the lathe is set to turn parallel; A C represents the axis of the work or cone when the lathe tailstock is set over to turn the taper or cone; hence the length of the line C B represents the amount the tailstock is set over. Referring now to Fig. 1224, the cone is supposed to stand level, as it will do in the end view, because the lathe centres remain at an equal height from the lathe bed or [V]s, notwithstanding that the tailstock is set over. The tool therefore travels at the same height throughout its whole length of feed; hence, if it is set, as at T, level with the line of centres, its line of feed while travelling from end to end of the cone is shone by the line A B. The length of the line A B is equal to the length of the line B C Fig. 1223. Hence, the line A B, Fig. 1224, represents two things: first, the line of motion of the point of tool T as it feeds along the cone, and second its length represents the amount the work axis is out of parallel with the line of lathe centres. Now, suppose that the tool be lowered to the position shown at I; its line of motion as it feeds will be the line C D, which is equal in length to the line A B. It is obvious, therefore, that though the tool is set to the diameter of the small end, it will turn at the large end a diameter represented by the dotted circle H. The result is precisely the same if the taper is turned by a taper-turning attachment instead of setting the tailstock out of line.

The demonstration is more readily understood when made with reference to such an attachment as the one just mentioned, because the line A B represents the line of tool feed along the work, and its length represents the amount the attachment causes the tool to recede from the work axis. Now as this amount depends upon the set-over of the attachment it will be governed by the degree of that set over, and is, therefore, with any given degree, the same whatever the length of the tool travel may be. All that is required, then, to find the result of placing the tool in any particular position, as at I in the end view, is to draw from the tool point a line parallel to A B and equal in length to it, as C D. The two ends of that line will represent in their distances from the work axis the radius the work will be turned to at each end with the tool in that position. Thus, at one end of the line C D is the circle K, representing the diameter the tool I would turn the cone at the small end, while at the other end the dotted circle H gives the diameter at the large end that the tool would turn to when at the end of its traverse. But if the tool be placed as at T, it will turn the same diameter K at the small end, and the diameter of the circle P at the large end.

We have here taken account of the diameters at the ends only of the work, without reference to the result given at any intermediate point along the cone surface, but this we may now proceed to do, in order to prove that a curved instead of a straight taper is produced if the tool be placed either above or below the line of lathe centres.

In Fig. 1225, D E F C represents the complete outline of a straight taper, whose diameter at the ends is represented in the end view by the outer and inner circles. Now, a line from A to B will represent the axis of the work, and also the line of tool point motion or traverse, if that point is set level with the axis. The line I K in the end view corresponds to the line A B in the side view, in so far that it represents the line of tool traverse when the tool point is set level with the line of centres. Now, suppose the tool point to be raised to stand level with the line G H, instead of at I K, and its line of feed traverse be along the line G H, whose length is equal to that of I K. If we divide the length of G H into six equal divisions, as marked from 1 to 6, and also divide the length of the work in the side view into six equal divisions (_a_ to _f_), we shall have the length of line G H in the first division in the end view (that is, the length from H to G), representing the same amount or length of tool traverse as from the end B of the cone to the line A in the side view. Now, suppose the tool point has arrived at 1; the diameter of work it will turn when in that position is evidently given by the arc or half-circle _h_, which meets the point 1 on G H. To mark that diameter on the side view, we first draw a horizontal line, as _h_ _p_, just touching the top of _h_; a perpendicular dropped from it cutting the line A B, gives the radius of work transferred from the end view to the side view. When the tool point has arrived at 2 on G H in the end view, its position will be shown in the side view at the line _b_, and the diameter of work it will turn is shown in the end view by the half-circle _k_. To transfer this diameter to the side view we draw the line _k_ _g_, and where it cuts the line _b_ in the side view is the radius of the work diameter when the tool has arrived at the point _b_ in the side view. Continuing this process, we mark half-circles, as _l_, _m_, _n_, _o_, and the lines _l_ _r_, _m_ _s_, _n_ _t_, _o_ _u_, by means of which we find in the side view the work radius when the tool has arrived at _c_, _d_, _e_, and _f_ respectively. All that remains to be done is to draw on the side view a line, as _u_ E, that shall pass through the points. This line will represent the outline of the work turned by the tool when its height is that denoted by G H. Now, the line _u_ E is shown to be a curve, hence it is proved that with the tool at the height G H a curved, and not a straight, taper will be turned.

It may now be proved that if the tool point is placed level with the line of centres, a straight taper will be turned. Thus its line of traverse will be denoted by A B in the side view and the line I K in the end view; hence we may divide I K into six equal divisions, and A B into six equal divisions (as _a_, _b_, _c_, &c.). From the points of division I K, we may draw half-circles as before, and from these half-circles horizontal lines, and where the lines meet the lines of division in the side view will be points in the outline of the work, as before. Through these points we draw a line, as before, and this line C F, being straight, it is proven that with the tool point level with the work axis, it will turn a straight taper.

It may now be shown that it is possible to turn a piece of work to a curve of equal curvature on each side of the middle of the work length. Suppose, for example, that the cutting tool stands on top of the work, as in the end view in Fig. 1226, and that while the tool is feeding along the work it also has a certain amount of motion in a direction at right angles to the work axis, so that its line of motion is denoted by the line B B in the top view. The outline of the work turned will be a curve, as is shown in Fig. 1227, in which the line of tool traverse is the line C D. Now the amount of tool motion that occurs during this traverse in a direction at right angles to the work axis is represented by the line F E, because the upper end is opposite to the upper end of C D, while the lower end is opposite the lower end of C D. We may then divide one-half of the length of F E into the divisions marked from 1 to 6. Now, as we have taken half the length of F E, we must also take half the length of the work and divide it into six equal divisions marked from _a_ to _f_. Now, suppose the tool point to stand in the line F S in the end view, its position in the top view will be at C. When it is at 1 on the end view it will have arrived at _p_ in the top view. The radius of work it will then turn is shown in the end view by the length of line running from 1 to the work centre. Take this length, and from _a_ in the work axis set it off on the line _a_ _h_, and make the length equal the height of 1 S. In like manner, when the tool point has arrived at 2, the radius it will cut the work is shown by the length of line _i_; hence from 2 on the work axis we may set off the length of 2 S, making 2 S and _b_ _i_ of equal length. Continuing this process, we make the length of _c_ _k_ equal that of 3 S, the length of _d_ _l_ equal 4 S, and so on. All that remains then is to draw a line, _o_ _g_, that shall meet the tops of these lines. This line will show the curve to which that half of the work length will be turned to. The other half of the work length will obviously be turned to the same curvature.

It is obvious that the curvature of the work outline will be determined by the proportion existing between the length of the work and the amount of tool motion in a direction at right angles to the work axis, or, in other words, between the length of the work and that of the line F E. It is evident, also, that with a given amount of tool motion across the work, the curvature of outline turned will be less in proportion as the work length is greater. Now, suppose that the smaller and the larger diameter of the work, together with its length, are given, and it is required to find how much curvature the tool must have, we may find this and work out the curve it will cut by the construction shown in Fig. 1228, in which the circle K is the smallest and the circle P the largest diameter. The line _m_ C is drawn to just touch the perimeter of K, and this at once gives the amount of cross-motion for the tool. Hence, we may draw the line _m_ B and C B, and from their extremities draw the line B B representing the path of traverse of the tool point. We may then obtain the full curve on one side of the work by dividing one-half the length of _m_ C into six equal divisions and proceeding as before, except that we have here added the lines of division in the second half as from _f_ to _l_. It will be observed that the centre of the curve is at the point where the tool point crosses the axis of the work; hence, by giving to the tool more traverse on one side than on the other of the work axis, the location of the smallest point of work diameter may be made to fall on one side of the middle of the work length.

In either turning or boring tapers that are to drive or force in or together, the amount to be allowed for the fit may be ascertained, so that the work may be made correct without driving each piece to its place to try its fit.

Suppose, for example, that the pieces are turned, and the holes are to be reamed, then the first hole reamed may be made to correct diameter by fit and trial, and a collar may be put on the reamer to permit it to enter the holes so far and no farther.

A taper gauge may then be made as in Fig. 1229, the line a representing the bore of the hole, and line B the diameter of the internal piece, the distance between the two being the amount found by trial to be necessary for the forcing or driving. The same gauge obviously serves for testing the taper of the holes reamed.

CHUCKED OR FACE PLATE WORK.--This class of work requires the most skillful manipulation, because the order in which the work may most advantageously proceed and the method of chucking are often matters for mature consideration.

In a piece of work driven between the lathe centres, the truth of any one part may be perceived at any time while operating upon the others, but in chucked work, such is not always the case, and truth in the work is then only to be obtained by holding it truly. Again, the work is apt to be sprung or deflected by the pressure of the devices holding it, and furthermore the removal of the skin or surface will in light work sometimes throw it out of true as the work proceeds, the reason being already given, when referring to turning plain cylindrical work.

TO TURN A GLAND.--There are three methods of turning a gland: first, the hole and the face on the outside of the flange may be turned first, the subsequent turning being done on a mandrel; second, the hole only may be bored at the first chucking, all the remaining work being done on a mandrel; and, third, the hole, hub, and one radial face may be turned at one chucking, and the remaining face turned at a separate chucking.

If the first plan be adopted, any error in the truth of the mandrel will throw the hole out of true with the hub, which would be a serious defect, causing the gland to jamb against one side of the piston rod, and also of the gland bore. The same evil is liable to result from the second method; it is best, therefore, to chuck the gland by the hub in a universal chuck, and simply face the outer face of the flange, and also its edge. The gland may then be turned end for end, and the hole, the hub, the inside radial flange face, and the hub radial face, may then all be turned at one chucking; there is but one disadvantage in this method, which is that the gland must be unchucked to try its fit in the gland hole, but if standard gauges are used such trial will not be necessary, while if such is not the case and an error of measurement should occur, the gland may still be put on a mandrel and reduced if necessary.

In either method of chucking, the fit of the hole to the rod it is intended for cannot be tested without removing the gland from the chuck.

TO TURN A PLAIN CYLINDRICAL RING ALL OVER IN A UNIVERSAL CHUCK.--Three methods may be pursued in doing this simple job: first, the hole may be bored at one chucking, and the two radial faces and the circumference turned at a second chucking; second, the diameter may be turned, first on the hole and two radial faces turned at a second chucking; and third the hole and one radial face may be turned at one chucking, and the diameter and second radial face at a second chucking. The last method is best for the following reasons. The tool can pass clear over the surfaces at each chucking without danger of coming into contact with the chuck jaws, which would cause damage to both; second, at the last chucking, the chuck jaws being inside the ring, the latter may be tested for truth with a pointer fixed in the tool rest, and therefore set quite true.

It is obvious that at neither chucking should the ring be set so far within the chuck jaws that there will be danger of the tool touching them when turning the radial face.

In the case of a ring too thin to permit this, and of too large a bore to warrant making a mandrel for it, the ring may be held on the outside and bored, and both radial faces turned to within a short distance of the chuck jaws; at the second chucking, the chuck jaws being within the ring bore, the work may be set true with a pointer, as before, and finished.

If, however, a number of such rings were to be turned, it would pay to turn up another and thicker ring, and use it as a mandrel after the bore and one radial face of the ring had been turned.

TO TURN AN ECCENTRIC STRAP AND ECCENTRIC.--The eccentric strap should be turned first, because it can then be taken apart and its fit to the eccentric tried while the latter is in the lathe, which is not the case with the eccentric. The strap should first be held in a universal chuck bolted to the face plate, or held in dogs such as shown in Fig. 893 at C, and one face should be turned. It should then be turned round on the chuck to bore it, and face the other side.

If the shape of the strap will admit it, it is best chucked by plates and bolts holding the face first turned to the face plate, because in this case there will be no pressure tending to spring the straps out of their natural shape; otherwise, however, it may be held in a universal or independent jaw chuck, or if too large for insertion in chucks of this kind (which are rarely made for large lathes) it may be held in dogs such as shown in Fig. 893 at C.

If after an eccentric strap is bored, and the bolts that hold its two halves together have been slackened, its diameter at A and at C, Fig. 1230, be measured, it will be found that A is less than C. The cause of this is partly explained under the head of tension of castings; but it is necessary to add that the diameters at A and at C in the figure are equal while the strap is in the lathe, or until the bolts holding the two halves of the strap together are released, yet so soon as this is done the diameter at A will reduce, the bore becoming an oval.[18]

[18] This occurs in all castings of similar form, as brasses, &c.

Now, it is obvious that the eccentric must be turned to the diameter at C, or otherwise it will have lost motion in the strap. If, however, the eccentric be turned to the diameter of C, the strap cannot be tried on, as it will bind at the corners, as shown in Fig. 1231. To remedy this evil it is usual to put a piece of sheet tin or metal between the joint faces of the two halves of the eccentric straps before they are chucked to turn them, and to bore them too large to the amount of the thickness of sheet metal so employed. After the straps are bored these pieces of metal are removed, and the strap halves bolted together as in Fig. 1230, the diameter at C being that to which the eccentric must be turned.

If the sheet metal so inserted were thick enough, the strap bore will measure the same at A as at C, Fig. 1230. If it were too thick the diameter at A will be greatest, while if too thin the diameter at A will be the least. There is no rule whereby the necessary thickness for a given size of strap may be known, and the workman is usually governed by his experience on castings of similar metal, or from the same moulding shop.

He prefers, however, to be on the safe side by not putting in too great a thickness, because it is easier to scrape away the bore at the joint than it is to file away the joint faces. The following thicknesses for the respective diameters may be considered safe for castings that have not been reheated after casting.

Diameter Thickness of metal to
of place between the
bore. strap valves.
Inches. Inch.
6 1/64
12 1/32
18 3/64
24 1/8

In turning a new strap for an old eccentric, it will be necessary, when taking the diameter of the eccentric, to take a piece of tin of the same thickness as that placed between the eccentric lugs or jaws, and place it between the caliper leg and the eccentric, so that the diameter of the strap across C, Fig. 1230, may be made equal when the tin is removed to the diameter of the eccentric.

In turning up the eccentric, the plain face should be faced first, setting it true, or nearly so, with the circumference of the eccentric, as will be the case if the circumference is held in a universal chuck, but if the hub is so long that this cannot be done because the chuck jaws cannot reach the circumference, the hub itself may be held in an independent jaw chuck.

The face turned may then be turned round, so as to meet the face of the chuck against which it should bed fairly, so as to run true. At this chucking the hole bore, the hub, and the radial faces should be turned, all these surfaces being roughed out before any one surface is finished.

The eccentric must then be again reversed, so that the face of the hub meets, the face plate being held by bolts as shown for a crank in figure, when the work being set to the lines marked (so as to give it the correct amount of throw) may be turned to fit the bore of the strap, the strap being taken apart so as to try it on, which this method of chucking will readily permit.

Now, in an eccentric, the surfaces requiring to be most true one with the other are those of the bore and of the circumference where the strap fits, and since the latter was turned with the hub face to the chuck, and that hub face was turned at the same chucking as the hole was bored (and must, therefore, be true to the bore), the bore and circumference will be as true as it is practicable to get them, because upon the truth of the last chucking alone will the truth of the work depend.

Small eccentrics may be held for all their chuckings in jaw chucks, but not so truly as if chucked on a face plate, because of the difficulty of keeping the radial faces of such jaws true, which occurs by reason of the causes explained with reference to Figs. 848 and 849.

Eccentrics having so much throw upon them as to render it difficult to hold them for the last chucking by the method above given (by bolts through the bore), usually have openings through them on the throw side, and in this case parallel pieces may be placed behind the radial face (on the hub side of the eccentric), such parallel pieces being thick enough to keep the hub face clear of the chuck face, and bolts may be passed through the said opening to hold the eccentric. Another method would be as follows:--

The outside diameter of the eccentric may be gripped in a dog chuck, if the dogs of the chuck project out far enough to reach it (otherwise the dogs may grip the hub of the eccentric), while the hole is bored and the plain face of the eccentric turned. The eccentric must then be reversed in the lathe, and the hub and the radial face on that side must be turned. Then the plain face of the eccentric must be bolted to the face plate by plates placed across the spaces which are made to lighten the eccentric, and by a plate across the face of the hub. The eccentric, being set true to the lines, may then be turned on its outside diameter to fit the strap; to facilitate which fitting, thin parallel strips may be placed between the face plate and the plain face of the eccentric at this last chucking. It will be observed that, in either method of chucking, the outside diameter of the eccentric (that is to say, the part on which the strap fits) is turned with the face which was turned at the same chucking at which the hole was bored, clamped to the face plate. In cases where a number of eccentrics having the same size of bore and the same amount of throw are turned, there may be fitted to the face plate of the lathe a disk (such as shown in Fig. 888), of sufficient diameter to fit the hole of the eccentric, the said disk being fastened to the face plate at the required distance from the centre of the lathe to give the necessary amount of throw to the eccentric. The best method of fastening such a disk to the face plate is to provide it with a plain pin turned true with the disk, and let it fit a hole (bored in the face plate to receive it) sufficiently tightly to be just able to be taken in and out by the hand, the pin being provided with a screw at the end, so that it can be screwed tight by a nut to the face plate. The last chucking of the eccentric is then performed by placing the hole of the eccentric on the disk, which will insure the correctness of the throw without the aid of any lines on the eccentric which may be set as true as the diameter of the casting will permit, and then turned to fit the strap.

TO TURN A CYLINDER COVER.--A cylinder cover affords an example of chucking in which the work done at one chucking requires to be very true with that done at a subsequent chucking, thus the gland hole which is on one side requires to be quite true with the diameter that fits into the cylinder bore, this diameter being on the opposite side.

If the polished or gland side of the cover be turned first, the hole for the packing ring and that for the gland may be bored with the assurance that one will be true with the other, while the polished outside face may be turned at the same chucking.

But when the cover is turned round in the lathe to turn the straight face, though the hole may be set true as far as can be ascertained in its short length, yet that length is too short to be an accurate guide, and the hole for the packing ring may appear true, while that for the gland, being longer, will have any error in the setting, multiplied by reason of its greater length. It is better, therefore, to turn the plain face first, gripping the cover by the gland flange so that the plain radial face, the step that fits the cylinder bore, and the outer edge of the cover flange may be turned at one chucking; then when the cover is turned round in the chuck, the flat face may be set true by resting against the radial surface of the chuck jaws, and the concentric truth may be set by the outer edge of the flange, which, being of the extreme diameter of the cover, will most readily show any want of truth in the setting. If in this case a universal chuck be used, and the work does not run quite true, it may be corrected by slacking the necessary dog or jaw on one side, and tightening up again from the screw of the necessary jaw on the other.

This occurs because from the wear, &c., there is always some small amount of play or lost motion in the jaw screws, and in the mechanism operating them, and by the above means this is taken advantage of to true the work.

If from any cause the work cannot be held for the first chucking by means of the gland hole flange, it must be held by the circumferential edge of the cover, letting the jaws envelop as small a distance over that edge as possible, the protruding part of it may then be turned up as close to the chuck jaws as possible, and this turned part may still be used to set the cover concentrically true at the second chucking.

In a very small cover the gland hole may have a mandrel fitted to it and be turned therefrom on both radial faces, or on one face only, the other being turned at the chucking at which the holes were bored.

In a cover too large to be held in a jaw chuck, the cover may be held in chucking dogs such as shown at C in Fig. 893, the edge protruding as much as possible from the dog screws, and being turned half way across at one chucking, and finished at the second chucking. To set the radial face at the second chucking, the surface gauge, applied as shown in Fig. 894, may be employed. If the bore of the packing ring or piston rod hole is large enough to permit it, that hole and the gland hole may be bored at the same chucking as that at which the plain face and step that fits in the cylinder bore is turned, thus ensuring truth in all the essential parts of the cover.

But in this case these operations should be performed at the last of the two chuckings, so as to eliminate any error that might arise from the casting altering its shape by reason of the removal of the metal on the radial face of the gland hole side of the cover.

TO TURN A PULLEY.--A pulley affords an excellent example of lathe work, because it may be operated upon by several different methods: thus, for boring it may be held, if small, in a dog chuck, with the jaws inside the rim; in a dog chuck with the jaws outside the rim; in a dog chuck by the hub itself (if the hub is long enough). A larger pulley may be chucked for boring by the rim held in a jaw chuck; by the rim held by bolts and plates, or by the rim held by dogs, such as shown in Fig. 893, or by the arms rested on pieces placed between them and the chuck, and then bolts and plates applied to those arms.

The rim may be turned by placing the pulley on a mandrel and driving that mandrel by a dog or carrier; by placing it on a mandrel and driving it by a Clements driver such as shown in Fig. 753, and having two diametrically opposite driving pins, placed to bear against diametrically opposite arms; by holding the arms to the chuck as before described, and performing the boring and facing at one chucking; or by holding the rim on its inside by the chuck jaws, so as to turn and bore the pulley at one chucking, which can be done when the inside of the rim is parallel, or not sufficiently coned to cause it to slip off the jaws, or when the jaws will reach to the centre of the rim width.

The advantages and disadvantages of these various methods are as follows:--

From the weakness of the pulley rim it is apt to distort when held with sufficient chuck-jaw pressure to enable the turning of the rim face and edge. But this would not affect the truth of the hole; hence the rim may be gripped in a chuck to bore the hole and face the hub. If so held it should be held true to the inside face of the rim, so that the bore will be true to the same, and then in turning the outside diameter it will be made as true as possible with the rim, which will preserve the balance of the pulley as much as possible. For these reasons the inside of the rim should be the part set to run true, whatever method of chucking be employed; hence, if the circumstances will permit of holding the hub to bore it, an independent jaw chuck should be employed (that is, of course, a chuck capable of independent jaw movement).

If the pulley be chucked by the arms, it is well-nigh impossible to avoid springing those arms from the pressure of the bolts, &c., holding them, and as a result the pulley face, though turned true, will not be true of itself, nor true with the hole, when the arms are released from such pressure.

If the pulley is of such a large size that its rim must be held by bolts and plates while the boring is progressing, such bolts, &c., must be placed on the outside of the rim, so as not to be in the way when setting the pulley true to the inside of the rim.

A small pulley may be turned on a mandrel driven by a dog, which is the truest method of turning, because the rim is in this case strained by the pressure of the cut only. But a dog will not drive a cut at such a leverage as exists at the rim of a pulley above about 18 inches in diameter; furthermore, in a large wheel there would not be sufficient friction between a mandrel and the pulley bore to drive the roughing cut on the pulley face.

It is necessary, therefore, to drive the pulley from the arms, while holding it on a mandrel, but if it be driven by one arm the whole strain due to driving will fall on that one arm, and on one side of the pulley only, and this will have a tendency to cause the rim at and near its junction with that arm to spring or deflect from its natural position, and, therefore, to be not quite true; all that can be done, therefore, is to drive by two arms with a Clements driver, so as to equalize the pressure on them.