The best way to true lathe centres is with an emery-wheel. In some lathes there are special fixtures for emery grinding, while in others an attachment to go in the tool post is used. Fig. 1156 shows such an attachment.

In the figure a is a frame to be fastened in the slide rest tool post at the stem a′. It affords journal bearing to the hand wheel b, to the shaft of which is attached the gear-wheel c, which drives a pinion d, on a shaft carrying the emery-wheel e, the operation being obviously to rotate wheel b, and drive the emery-wheel e, through the medium of the multiplying gear-wheels c, d.

The emery-wheel is fed to its depth of cut on the lathe centre p, by the cross feed screw of the lathe, and is traversed by pulling or pushing the knob f, the construction of this part of the device being as follows: g and h are two bushes, a sliding fit in the arms of frame a, but having on top flat places i and j, against which touch the ends of the two set-screws k, l, to prevent them from rotating. The emery-wheel and gear pinion d are fast together, and a pin passes through and holds g and h together. Hence the knob f pushes or pulls, as the case may be, the bushes through the bearings g, h, in the frame a, the pinion and emery-wheel traversing with them. Hence pinion d is traversed to and fro by hand, and it is to admit of this traverse that it requires its great length. The stem a is at such an angle that, if it be placed true with the line of cross feed, the lathe centre will be ground to the proper angle.

Fig. 1157 represents a centre grinding attachment by Trump Brothers, of Wilmington, Delaware. In this device the emery-wheel is driven by belt power as follows. A driving wheel a is bolted to the lathe face plate, and a stand carries at its top the over-head belt pulleys, and at its base the emery-wheel and spindle. This stand at c sets over the tool post, and is secured by a bar passing through c and through the tool post, whose set-screw therefore holds the stand in position. On the end of the emery-wheel spindle is a feed lever, by means of which the emery-wheel may be fed along the lathe centre. Cup piece b is for enabling wheel a to be readily set true on the lathe face plate, one end of b fitting the hub of a, while the other receives the dead centre which is screwed up so that b will hold a in place, while it is bolted to the lathe face plate, and at the same time will hold it true.

In the absence of a centre grinding attachment, lathe centres may be turned true with a cutting tool, and finished with water applied to the tool so as to leave a bright and true surface. They should not, for the finest of work, be finished by filing, even though the file be a dead smooth one, because the file marks cause undue wear both to the lathe centres and the work centres.

The dead centres should be hardened to a straw color, and the live centre to a blue; the former so as to have sufficient strength to resist the strain, and enough hardness to resist abrasion, and the latter to enable it to be trued up without softening it.

When, after turning them up, the centres are put into their places, the tailstock may be moved up the bed so that the dead centre projects but very little from the tailstock, and is yet close to the live centre, and the lathe should be run at its fastest speed to enable the eye to perceive if the live centre runs true, and whether the dead centre is in line with the live one, and the process repeated so that both centres may be tested.

A more correct test, however, may be made with the centre indicator.

Centre Indicators.—On account of the difficulty of ascertaining when a centre runs quite true, or when a very small hole or fine cone as a centre punch mark runs true when chucked in a lathe, the centre indicator is used to make such tests, its object being to magnify any error, and locate its direction. Fig. 1158, from The American Machinist, represents a simple form of this tool, designed by Mr. G. B. Foote, for testing lathe centres. a is a piece of iron about 8 inches long to fit the lathe tool post, b is a leather disk secured to a by a plate c, and serving to act as a holding fulcrum to the indicator needle, which has freedom of movement on account of the elasticity of the leather washer, and on account of the hole shown to pass through a. It is obvious that if the countersunk end of the needle does not run true, the pointed end will magnify the error by as many times as the distance from the needle point to the leather washer is greater than that from the leather washer to the countersunk end of the needle. It is necessary to make several tests with the indicator, rotating the lathe centre a quarter turn in its socket for each test, so as to prove that the centre runs true in any position in the lathe spindle. If it does not run true the error should be corrected, or the centre and the lathe spindle end may be marked by a centre punch done to show in what position the centre must stand to run true.

The tension of the leather washer serves to keep the countersunk against the lathe centre without a very minute end adjustment. Or the same end may be attained by the means shown in Fig. 1159, which is a design communicated by Mr. C. E. Simonds to The American Machinist. The holder is cupped on one side to receive a ball as shown, and has a countersink on the other to permit a free vibration of the needle. The ball is fitted to slide easily upon the needle, and between the ball and a fixed collar is a spiral spring that keeps the ball in contact with its seat in the holder.

One end of the needle is pointed for very small holes or conical recesses, while the other is countersunk for pointed work, as lathe centres. The countersink of the needle may be made less acute than the lathe centre, so that the contact will be at the very point of the lathe centre, the needle not being centre-drilled. The end of the needle that is placed against the work should be as near to the ball or fulcrum as convenient, so as to multiply the errors of work truth as much as possible.

In some forms of centre indicators the ball is pivoted, so that the needle only needs to be removed to reverse it end for end, or for adjusting its distance, it being made a close sliding fit through the ball. Thus, in Fig. 1160 the ball e is held in a bearing cut half in the holder a, and half in cap b, which is screwed to a by screws c d.

Or the ball may be held in a universal joint, and thus work more frictionless. Thus, in Fig. 1161 it is held by the conical points of two screws diametrically opposite in a ring which is held by the conical points of two screws threading through an outer ring, these latter screws being at a right angle to those in the inner ring. The outer ring is held to the holder by the conical points of two screws, all the conical points seating in conical recesses.

It is obvious that the contact of the point of the needle and the work may be more delicately made when there is some elasticity provided, as is the case with the spiral spring in Fig. 1159.

Indicators of this class may be used to test the truth of cylindrical work: thus, in Fig. 1162 is an application to a piece of work between the lathe centres, there being fitted to one end of the needle a fork a that may be removed at pleasure.

One of the difficulties in turning up a lathe centre to run true arises from the difference in cutting speed at the point and at the full diameter of the cone, the speed necessary to produce true smooth work at the point being too fast for the full diameter. This may be remedied on centres for small work, as, say, three inches and less in diameter, by cutting away the back part of the cone, leaving but a short part to be turned up to true the centre.

To permit the cutting off or squaring tool to pass close up to the centre, and thus prevent leaving a burr or projection on the work end, the centre may be thus relieved at the back and have a small parallel relief, as in Fig. 1164 at a, the coned point being left as large as possible, but still small enough to pass within the countersink.

In centres for large and heavy work it is not unusual to provide some kind of an oil way to afford means of lubrication, and an excellent method of accomplishing this object is to drill a hole a, Fig. 1163, to the axis of the centre and let it pass thence to the point as denoted by the dotted line; there may also be a small groove at b in the figure to distribute the oil along the centre, but grooves of this kind make the returning of the centre more difficult and are apt to cause the work centres to enlarge more from wear, especially in turning tapers with the tailstock set over the lathe centre, these being out of line with the work centre.

To enable a broad tool such as a chaser to meet work of smaller diameter than the lathe centre, the latter is cut away on one side as in Fig. 1164. It is obvious also that the flat place being turned uppermost, will facilitate the use of the file on work of smaller diameter than the lathe centre, and that placed in the position shown in the cut, it will permit a squaring tool to pass clear down to the centre and avoid leaving the projecting burr which is left when the tool cannot pass clear down the face to the edge of the countersink of the work centre.

The method to be employed for centring work depends upon its diameter, and upon whether its ends are square or not. When the pieces are cut from a rod or bar in a cutting-off machine, the ends are square, and they may be utilized to set the work by in centring it. Thus, in Fig. 1165 is a top, and in Fig. 1166 is an end view of a simple device, or lathe attachment for centre drilling. s is a stand bolted to the lathe shears and carrying two pins p, which act as guides to the cup chuck or work guide g; between the heads of pins p and the hubs of g are spiral springs, forcing it forward, but permitting it to advance over the drill chuck; the work w is fed forward to the drill. At the dead centre end the work is supported by a female cone centre d in the tail spindle t. The work rests in mouths of g and d, and as the pieces are cut from the rod they are sufficiently straight, and being cut off in a cutting-off machine the ends are presumably square; hence the coned chucks will hold them sufficiently true with the ends, and the alignment of the centre drilled holes will not be impaired by any subsequent straightening processes; for it is to be observed, that if work is centre-drilled and straightened afterwards, the straightening throws the centre holes out of line one with the other, and the work will be more liable to gradually run out of true as its centres wear.

Thus, in Fig. 1167, let w represent a bent piece of work centre-drilled, and the axis of the holes will be in line as denoted by the dotted line, but after the piece is straightened the holes will lie in the planes denoted by the dotted line in Fig. 1168, and there will be a tendency for the work centres to move over towards the sides c d as the wear proceeds.

In Fig. 1169 is shown a centre-drilling machine, which consists of a live spindle carrying the centre-drilling tool, and capable of end motion for the drill feed. The work is held in a universal chuck, and if long is supported by a stay as shown in the figure. The axis of the work being in line with that of the chuck, the work requires no setting.

In this case the centre hole will be drilled true with that part of the work that is held in the chuck, and the alignment of the centre hole will depend upon the length of the rod being supported with its axis in line with the live spindle. If the work is not straightened after drilling, the results produced are sufficiently correct for the requirements; but it follows from what has been said, that work which requires to be straightened and tried for straightness in the lathe should be centred temporarily and not centre-drilled until after the straightening has been done.

In Fig. 1170 is shown a combined centre-drill and countersink not unfrequently used in centring machines. The objection to it is, that the cutting edges of the drill get dull quicker than those of the countersink, and in regrinding them the drill gets shorter. Of course the drill may be made longer than necessary so as to admit of successive grindings, but this entails drilling the centre holes deeper than necessary, until such time as the drill has worn to its proper length. To overcome this difficulty the countersink may be pierced to receive a drill as in Fig. 1171, the drill being secured by a set-screw s.

Among the devices for centring work by hand, or of pricking the centre preparatory for centre-drilling, are the following:—

In Fig. 1172 is a centre-marking square. a b c d represents the back and e the blade of the square. Suppose then that the dotted circle f represents the end of a piece of work, and we apply the square as shown in the cut and mark a line on the end of the work, and then moving the square a quarter turn around the work, draw another line, the point of contact of these two lines (as at g in the cut) will be the centre of the work, or if the work is of large diameter as denoted by the circle h h, by a similar process we obtain the centre e. In this case, however, the ends a b of the square back must be of equal lengths, so that the end faces at a b will form a right angle to the edge of the blade, and this enables the use of the square for ordinary purposes as well as for marking centres.

The point a of the centre punch shown in Fig. 1173 is then placed at the intersection of the two lines thus marked, and a hammer blow produces the required indentation. The centre punch must be held upright or it will move laterally while entering the metal. The part b of the centre punch is tapered so as to obstruct the vision as little as possible, while it is made hexagon or octagon at the upper end to afford a better grip. By increasing the diameter at c, the tool is stiffened and is much less liable to fly out of the fingers when the hammer blow does not fall quite fair.

In Fig. 1174 is shown a device for guiding the centre punch true with the axis of the work, so as to avoid the necessity of finding the same by lines for the centres. It consists of a guide piece b and a parallel cylindrical centre punch a, c representing a piece of work. b is pierced above with a parallel hole fitting and guiding the centre punch, and has a conical hole at the lower end to rest on the work, so that if the device be held upright and pressed down upon the end of the work, and the top of the centre punch is struck with the hammer, the indentation made will be central to the points of contact of the end of the work with the coned hole of b. If then the end of the work has no projecting burrs the centring will be centred true.

In the absence of these devices, lines denoting the location for the conical recess or centre may be made, when either of the following methods may be pursued.

In Fig. 1175 is shown what is known as a pair of hermaphrodite calipers, which consists of two legs pivoted at the upper end; the bent leg is placed against the perimeter of the work, as shown, and held steadily, while with the point a line is marked on the work. This operation is performed from four equidistant (or thereabouts) points on the work, which will appear as shown in Fig. 1176, providing the radius to which the point was set be equal to the radius of the work. The point at which the lines meet is in this case the location for the centre. If, however, the radius to which the points are set is less than the radius of the work, the lines will appear as in Fig. 1177, in which case the location is in the centre of the inscribed square, as denoted by the dot; or if the radius be set too great the lines will appear as in Fig. 1178, and the location for the centre will again be as denoted by the dot.

Another and very old method of marking these lines is to place the work on a pair of parallel pieces and draw the lines across it, as shown in Fig. 1179, in which w represents the work, p, p the parallel pieces of equal thickness, s a stand (termed a scribing block) carrying a needle n, which is held by a thumb screw and bolt at b. The point of the needle is adjusted for the centre of the work, a line is drawn, the work is then rotated, another line drawn, and so on, until the four lines are drawn as in Fig. 1180, when the work may be turned end for end if light, or if heavy the scribing block may be moved to the other end of the work.

The centre locations are here made true with the part of the work that rests on the parallel pieces, and this is in some cases an essential element in the centring.

Thus, in Fig. 1181, it is required to centre a piece true with the journals a b, and it is obvious that those journals may be rested on parallel pieces p, p, and the centres marked by the scribing block on the faces e, f in the manner before described.

If there is a spot in the length of a long piece of work where the metal is scant and out of round, so that it is necessary to centre the work true by that part, the surface gauge and parallel pieces may be used with advantage, but for ordinary centring it is a slow process. When a piece of work is not cylindrical, and it is doubtful if it will clean up, the centring requires care, for it must not always be assumed, that if two diametrically opposite points meet the turning tool at an equal depth of cut, the piece is centred so as to true up to the largest possible diameter.

This is pointed out in Fig. 1182, which is extracted from an article by Professor Sweet. “In a piece of the irregular form a, the points a and b might be even and still be no indication of the best location for the centre, and in the piece b it is evident that if c and d were even, nothing like the largest cylinder could be got from it. In the case of shape a, the two points e and f should be equidistant from the centre, and in the case of shape b, the three points g, h, i should be equidistant from the centre.”

The depth of the centre drill holes should be such as to leave them in the work after it is cut off to its proper length, and will, therefore, be deeper as the amount to be cut off is greater.

The diameter of the centre drill is larger as the size of the work increases, and may be stated as about ³⁄₆₄ for work of about ¹⁄₂ inch, increasing up to ¹⁄₈ inch for work of about an inch, and up to three inches in diameter; for work of a foot or over the centre drill may be ³⁄₁₆ inch in diameter.

The centre drilling and countersinking may, when the work is cut to length, be performed at one operation, but when it requires to be cut to length in the lathe, that should be done before the countersinking. A very simple chuck for centre drilling is shown in Fig. 1183, with a twist drill (which is an excellent tool for centre-drilling). If the work is held in the hand and fed to the drill by the lathe dead centre, the weight of the work will cause the hole to be out of straight with the work axis, unless the grip is occasionally relaxed, and the work made to rotate a half or a quarter turn as the drilling proceeds.

After the work is centre-drilled and cut off to length, it must be finally countersunk, so as to provide ample bearing area for the lathe centres.

The countersinking should be true to the centre hole; and it is sometimes made to exactly fit the lathe centres, and in other cases it is made more acute than the lathe centre, so that the oil may pass up the countersink, while it is bedding itself to the lathe centres.

If the countersinking is done before the end of the work is squared, it will not be true with the centre-drilled hole.

In order that the countersinking may wear true with the centre-drilled hole, it may be made of a more obtuse angle (as, say, one degree) than the lathe centre, as in Fig. 1184, so that the hole may form a guide to cause the lathe centre to wear the countersinking true to the hole, and thus correct any error that may exist.

If the countersink is made more acute than the lathe centre, as shown in Fig. 1185, the wear of its mouth will act as a guide, causing the centre to be true with the countersinking; and when the bearing area extends to the centre-drilled hole, there will be introduced, if that hole does not run true, an element tending to cause the work to run out of true again, because the countersinking will have more bearing area on one side than on the other.

It is to be observed, however, that if the difference between the countersink angle and that of the lathe centre be not more than about one degree, the work centre will bed itself fully to the lathe centre very rapidly, and usually before the first cut is carried over the work, unless the work centres have been made to have unduly large countersinks.

Fig. 1186 represents a half-round countersink, in which the cutting edge is produced by cutting away the coned point slightly below the dotted axial line. This secures two advantages: first, it gives the cutting edge clearance without requiring the grinding or filing such clearance; and, secondly, the cone being the same angle as the lathe centres, filing away more than half of it causes it to give the lathe centre at first a bearing at the small end of the countersink, as in Fig. 1184, and this secures the advantage mentioned with reference to that figure. It is obvious that such a reamer, however, does not produce strictly a cone countersink, as is shown in Fig. 1187, where the cutting away of the cone is carried to excess simply to explain the principle, and the cone becomes an hyperbolic curve.

The small amount, however, that it is necessary to carry the face below the line of centres, practically serves to make the cone somewhat less acute, and is not therefore undesirable.

Another method of forming the half-round countersink is shown in Fig. 1188, in which the cone is of the same angle as the lathe centres; the back a is ground away to avoid its contact with the work and give clearance, while clearance to the cutting edge is obtained by filing or grinding a flat surface b at the necessary angle to the upper face of the cone. In this case it is assumed that the centre-drilling and countersinking are true one with the other. Yet another form of countersink is shown in Fig. 1189, consisting of a cone having three or four teeth. It may be provided with a tit, which will serve as a guide to keep the countersink true with the hole, and this tit may be made a trifle larger in diameter than the hole, and given teeth like a reamer, so as to ream the hole out while the countersinking is proceeding.

Unless one side of a half-round reamer is filed away so as to give the cutting edge alone contact with the bore of the hole, an improper strain is produced both upon the work and the countersink.

In Fig. 1190, for example, is shown, enlarged for clearness of illustration, a hole, and a half-round countersink in section, and it is evident that if the countersink is set central to the hole, it will have contact at a and at b, and a cannot enter the metal to cut without springing towards c.

But when the lathe has made rather more than one-half a revolution, the forcible contact at b will be relieved, and either the work or the countersink will move back towards d. This may be remedied by setting the countersink to one side, as in Fig. 1191, or by cutting it away on one side, as in Fig. 1192, when the half-round reamer will, if the work be rigidly held while being countersunk, act as a cutting tool. But it is more troublesome to hold the work rigidly while countersinking it than it is to simply hold it in the hands, and for these reasons the square centre is an excellent tool to produce true countersinking.

Fig. 1193 represents a square centre, the conical end being provided with four flat sides, two of which appear at a b, or it may have three flat sides which will give it keener cutting edges, and will serve equally well to keep it true with the drilled hole. But it is questionable whether it is not an advantage not to have the cutting edges so keen as is given by the three flat faces, because the less keen the cutting edges are, the more true the countersinking will be with the hole, the extra pressure required to feed the square centre tending to cause it to remain true with the hole notwithstanding any unequal density of the metal on different sides of the hole. An objection to the square centre is that it involves more labor in the grinding to resharpen it, and is not so easy to grind true, but for fine work this is more than compensated for in the better quality of its work.

This labor, however, may be lessened in two ways: first, the faces may be fluted, as in Fig. 1194, at a and at b, or its diameter may be turned down, as in Fig. 1195. In using the square centre it is placed in the position of the live centre and revolved at high speed, all the cutting edges operating simultaneously; the work is fed up by the dead centre and held in the hand.

To prevent the weight of the work from causing the countersinking being out of true with the hole, the work should be occasionally allowed (by relaxing the grip upon it) to make part of a revolution, as explained with reference to centre-drilling without a work guide. Another and simple form of square centre for countersinking is shown in Fig. 1196. It consists of a piece of square steel set into a stock or holder.

Work that is to be hardened and whose centres are, therefore, liable to warp in the hardening, may be countersunk as in Fig. 1197, there being three indentations in the countersink as shown. This insures that there shall be three points of contact, and the work will run steadily and true. Furthermore, the indentations form passages for the oil, facilitating the lubrication and preventing wear both to the work and to the lathe centres.

These indentations are produced after the countersinking by the punch, shown in Fig. 1198. Except when tapers are turned by setting the lathe centres out of line with the lathe shears (as in setting the tailstock over), all the wear falls on the dead centre end of the work, as there is no motion of the work centre on the live centre, hence the work centres will not have worn to a full bearing until the work has been reversed end for end in the lathe.

If it be attempted to countersink a piece of work whose end face is not square, the countersinking will not be true with the centre hole, and furthermore the causes producing this want of truth will continue to operate to throw the work out of true while it is being turned. Thus, in Fig. 1199, a represents a piece of work and b the dead centre; if the side c is higher than side d of the work end, the increased bearing area at c will cause the most wear to occur at d, and the countersink in the work will move over towards d, and it follows that the face of a rough piece of work should be faced before being countersunk. Professor Sweet designed the centre-drilling device shown in Fig. 1200, which consists of a stock fitting the holes for the lathe centres, and carrying what may be called a turret head, in which are the centre drills, facing tools, and countersinks. The turret has 6 holes corresponding to the number of tools it carries, and each tool is held in position by a pin, upon a spring, which projects into the necessary hole, the construction being obvious. The facing tool is placed next to the drill and is followed by a countersink, in whatever direction the turret is rotated to bring the next tool into operation. The work should, on account of the power necessary for the facing, be driven in a chuck.

A similar tool, which may, however, be used for other work besides centring and countersinking, is shown in Fig. 1201. It consists of a stem fitting into the hole of the tail spindle, and carrying a base having a pin d, on which fits a cap having holes b, and set-screws c for fastening drills, countersinks, or cutting tools. The cap is pierced with six taper holes, and a pin projects through the base into these holes to lock the cap in position, this pin being operated by the spring lever shown.

Work that has already been turned, but has had its centres cut off, may be recentred as follows. One end may be held and driven by a chuck, while the other end is held in a steady rest such as was shown in Fig. 802, and the centre may then be formed in the free end by a half-round reamer, such as shown in Fig. 1190, placed in the position of the dead centre, or the square centre may be used in place of the dead centre, being so placed that one of its faces stands vertically, and therefore that two of its edges will operate to cut. The location for the work centre should be centre punched as accurately as possible, and the work is then placed in the lathe with a driver on it, as for turning it up; a crotch, such as shown in Fig. 1202, is then fastened in the lathe tool post, and fed up by the cross-feed screw until it causes the work to run true, and the square centre should then be fed slowly up and into the work, with a liberal supply of oil. If the work runs out of true, the crotch should be fed in again, but care must be taken not to feed it too far. So long as the square centre is altering the position of the centre in the work, it will be found that the feed-wheel of the tailstock will feed by jumps and starts; and after the feeding feels to proceed evenly, the crotch may be withdrawn and the work tried for being true. The crotch, as well as the square centre, should be oiled to prevent its damaging the work surface. It is obvious that in order to prevent the lathe dead centre point from seating at the point or bottom of the work centre, the square centre should be two or three degrees more acute in angle than the lathe dead centre. If the work is tried for truth while running on the square centre, the latter is apt to enlarge the work centre, while the work will not run steadily, hence it is better (and necessary where truth is a requisite) to try the work with the dead centre in place of the square one.

In thus using a square centre to true work, great care should be taken not to cut the work centres too large, and this may be avoided by making the temporary centre-punch centres small, and feeding the crotch rapidly up to the work, until the latter runs true, while the square centre is fed up only sufficiently to just hold the work steady.

To test the truth of a piece of rough work, it may, if sufficiently light, be placed between the lathe centres with a light contact, and rotated by drawing the hand across it, a piece of chalk being held in the right hand sufficiently near to just touch the work, and if the chalk mark extends all round the work, the latter is as true as can be tested by so crude a test, and a more correct test may be made by a tool held in the tool rest. If the test made at various positions in the length of the work shows the work to be bent enough to require straightening, such straightening may be done by a straightening lever.

In shops where large quantities of shafting are produced, there are special straightening tools or devices: thus, Figs. 1203 and 1204 represent two views of a straightening machine. The shaft to be straightened is rotated by the friction caused by its own weight as it lies between rollers, which saves the trouble of placing the shaft upon centres. Furthermore, the belt that is the prime mover of the gears driving these rollers is driven from the line shaft itself without the aid of any belt pulley. The tension of this driving belt is so adjusted that it will just drive the heaviest shaft the machine will straighten; but if the operator grasps the shaft in his hand, the driving mechanism will stop and the belt will slip, the shaft remaining stationary until the operator sets it in motion again with his hand, when the belt ceases to slip and the mechanism again acts to drive the shaft.