Home  

CHAPTER XIX.

SETTING THE VALVES.

THE MEN WHO LEARN VALVE-SETTING.
MOST of intelligent machinists engaged on engine work, make it an object of ambition to learn to set valves; and the operation is mastered as soon as the opportunity offers. It has been a practice in numerous shops for those who have the work of valve-setting to do, to invest the operation with fictitious mystery, to patiently disseminate the belief that valve-setting is an exceedingly difficult matter. Cases sometimes arise where the squaring of an engine's valves is really an arduous task, requiring intimate familiarity with delicate methods of adjustment; but valve-setting, as it is usually practiced in building establishments, in repairing-shops, and in round-houses, is merely a matter of plain measurement.

A man may be a first-class engineer without knowing how to set valves, and familiar acquaintance with the operation will not increase his ability in managing his engine when merely getting a train over the road on time is the consideration; but the method of valve setting is so closely associated with an intelligent appreciation of the valve-motion's philosophy, that most of engineers who take an extended interest in their business, wish to acquire the knowledge of how the valves are set.

BEST WAY TO LEARN VALVE-SETTING.
The best way to learn valve-setting is by taking part in the work. Whatever can be said in books on a subject of this kind, provides but an indifferent substitute for going through the actual operations. But a man's ambition to learn may exceed his opportunities; so, for those who can not get a gang boss to direct them into the art of valve-setting, this description will be made as plain as possible.

When an engine's valve-motion is designed, the sizes of the different parts are arranged; and, if this business is done by a competent engineer, there will only be trifling changes necessary in valve-setting.

PRELIMINARY OPERATIONS.
Let us suppose the engine to be an ordinary eight wheel locomotive, with cylinders 17 X 24 inches. Let us assume that the top and bottom rocker-arms are straight, of equal length, and that the eccentric-rods are connected to the link so as to be opposite the block in full gear. This will make the extreme travel of valve equal the eccentric's throw. We will now look round to see that every thing connected with the motion is ready for valve-setting.

First, it is necessary to see that the wedges are properly set up to hold the driving-boxes in about the same position they will occupy when the engine is at work.

CONNECTING ECCENTRIC-RODS TO LINK.
In looking over the motion, it is well to note that the eccentric-rods are properly connected, — the forward eccentric-rod with the top, the backward eccentric-rod with the bottom, of the link. When the crank-pin is on the forward center, the eccentrics will occupy the position they appear in, in Fig. 15, where the rods are open, and nearly horizontal. The full parts of both eccentrics are advanced towards the crank-pin, so that the centers of the eccentrics are advanced from a perpendicular line drawn through center of axle, a horizontal distance equal to the lap and lead. When the crank-pin is on the back center, the eccentric centers will be behind the axle, and the rods will be crossed as they are seen in Fig. 16. The reason why the rods must be crossed when the crank is in this position, is, that the forward eccentric center is below the axle, and the backward eccentric center is above, As the forward eccentric-rod maintains its connection with the top of the link, and the backward eccentric-rod is at the opposite end, crossing of the rods is inevitable. This fact is worth imprinting on the memory, for I have known of several cases where men got the rods up wrong by putting them open when the engine stood with the crank on the back center.

MARKING THE VALVE-STEM.
In ordinary practice, valves are set with the steam-chest cover down, and the position of the valve on the seat is identified by marks on the valve-stem. Before the cover is put down, the valve is placed as in Fig. 17, just beginning to open the forward steam-port; a thin piece of tin being generally used to gauge the opening When the valve stands in this position, a tram is extended from a center punch-mark c, on the stuffing-box, straight along the valve-stem as far as it will reach; and the point, here located at a, is marked. The valve is then moved forward till it begins to uncover the back port, when another measurement is made with the tram, which locates the point b on the valve-stem. Whatever position the valve may stand on, it may now be identified by the tram, When the tram cuts the space half way between a and b, the valve stands in the middle of the seat.

Some machinists do not believe in tramming from the stuffing-box, as the point is liable to be moved in tightening down the steam-chest cover. These generally measure from a point on the cylinder casting, but that practice has its drawbacks.

LENGTH OF THE VALVE-ROD.
To prove the correct length of the valve-rod, the rocker-arm is set at right angles to the valve-seat, which is its middle position. The valve must now stand on the middle of the seat, which will be indicated by the tram point reaching the dividing point between a and b. Should the valve not be right when the rocker is in its middle position, the rod must be altered to put it right.

ACCURACY ESSENTIAL IN LOCATING THE DEAD CENTER POINTS.
Before proceeding to set the valves, a machinist can not be too careful in locating the exact dead centers. Some men conclude, because there is little motion to the cross-head close to the end of the stroke, that a slight movement of the wheel to one side or the other is of little consequence, and makes no perceptible difference in the relative positions of piston and valve. This is a serious mistake; for, although the piston is moving slowly, the eccentric is proceeding at its ordinary speed, and the valve is moving fast. The loose, quick methods of finding dead centers followed occasionally are not conducive to exactness, and nothing but accuracy is permissible in valve-setting.

FINDING THE DEAD CENTERS.
The best way of finding the true center is by moving the cross-head a measured distance round its extreme travel, recording the extent of movement on the driving-wheel tire, whose motion is uniform; then bisecting the distance between the marks on the tire, when the dividing line will indicate the true center.

Thus: Turn the wheels forward till the cross-head reaches within one-half inch of its extreme travel, as shown in Fig. 18. From a point a on the guide-block, extend a tram on the cross-head, and mark the extreme point reached b. Put a center punch-mark c on the wheel-cover, or other convenient fixed point, and from it extend a tram on the edge of the tire, and scratch an arc d. Now, with tram in hand, watch the cross-head, and have the wheels moved forward slowly. When the cross-head passes the center, and moves back till the tram extending from a will reach the point b, stop the motion. Again tram from the wheel-cover point, and describe a second arc on the tire, which will be at e, now moved to the position which d occupied when the previous measurement was taken. With a pair of dividers bisect the distance between d and e. Mark the dividing point C with a center punch, and put a chalk ring round it. When the wheel stands so that the tram will extend from c to C, the engine will be on the forward dead center.

All the other centers must be found by a similar process.

TURNING WHEELS AND MOVING ECCENTRICS.
When a measurement is going to be made for fore gear, the wheels must be turned forward; and, when it is for the back gear, they must be turned backward. Enough movement of the wheel must be given to take up the lost motion every time the direction of movement is changed. In moving an eccentric, it should also be turned far enough in the opposite direction to take up the lost motion.

SETTING BY THE LEAD OPENING.
Put the reverse-lever in the full forward notch, and place the engine on the forward center. If the lead opening in full gear is to be z inch, advance the forward eccentric till the point a (Fig. 17) on the valve-stem is that distance away from the tram point. Throw the reverse-lever into the full backward notch, turn the wheels forward enough to take up the lost motion, then turn them back to the forward center. Move the backward eccentric (if it needs moving) till the tram, extended on the valve-stem, strikes the same point that it reached for the forward motion. It will be noted here, that the valve occupies the same position for fore and back gear when the engine is on the center. Put the reverse-lever in the forward notch again, and turn the wheels ahead till the back center point is reached. Now tram the valve-stem again, and, if the lead opening be the same for both gears as it was on the forward center, that part of the setting is right. It is a good plan to go over the points a second time to prove their corectness. But it is not likely that the lead opening at the back end will be right on the first trial. Instead of having the correct lead, the valve will probably lap over the port, being what workmen call "blind," or it will have too much lead. Let us assume that our valve is z inch blind. This indicates that the eccentric-rod is too long. We shorten the rod till the valve is at the opening point, and, on turning the engine to the forward center again, we will find that the valve there has lost its lead. But our change has adjusted the valve movement, so that on each center the valve is just beginning to open the steam-port. Advancing the eccentric to give one end z inch lead will now have the same effect upon the other end; and, assuming that the back motion has been subjected to similar treatment with a like result, the lead opening on that side is right. This process must now be repeated with the other side of the engine.

ASCERTAINING THE POINT OF CUT-OFF.
The lead openings being properly arranged, we will proceed to examine how the valves cut off the steam; for it is important that about the same supply of steam should be furnished to each cylinder and to each end of the cylinders. The angularity of the connecting rod tends to give a greater supply of steam to the forward than to the back end of the cylinder; but this inequality is, as has already been explained, usually rectified by locating the hanger-stud a certain distance back of the link arc.

To prove the cut-off, we will try the full gear first. Put the reverse-lever in the full forward notch, starting from the forward center, and turn the wheels ahead. The motion of our engine has been designed so that the cut-off in full gear shall happen at 18 inches of the stroke. With tram in hand, watch the movement of the valve as indicated by the stem marks. As the piston moves away from the forward end of the cylinder, the valve will keep opening till nearly half stroke is reached, when it will begin to return, slowly at first, but with increasing velocity as the point of cutoff is reached. When the point a, Fig. 17, gets so that it will be reached by the tram extended from c, the motion must be stopped; as that indicates the point of cutoff. Now measure on the guide how far the cross-head has traveled from the beginning of the stroke, and mark it down with chalk. Then turn the wheels in the same direction past the back center, and obtain the cut-off for the forward stroke in the same manner. The cutoff for the other cylinder will be found in precisely the method described.

In addition to trying the cut-off in full gear, it is usually tested at half stroke and at 6 inches, or with the reverse-lever in the notches nearest to these points. Some men begin at the first notch, and follow the point of cut-off in every notch till the center is reached, and do the same for back gear.

ADJUSTMENT OF CUT-OFF.
From various causes, it often happens that the cut-off is unequal in the two strokes, or one cylinder may be getting more steam than the other. Suppose, that, on one side of the engine the valve is cutting off at 182 inches in forward gear, while at the other side it is cutting off at 172 inches of the stroke. The most ready way to adjust that inequality is by shortening one link-hanger and lengthening the other till a mean is struck. Where the discrepancy is smaller, it is adjusted by lengthening the hanger at the short side.

A harder inequality to adjust is where the valve cuts off earlier for one end of the cylinder than for the other. In new work this is readily overcome by the saddle-stud, but such a change is seldom admissible in old work. When the points of cut-off have been noted down, it will frequently happen, that, instead of both ends cutting off at 18 inches, one end will show the cut at 17 inches, while the other goes to 19 inches. This indicates something wrong, and demands a search for the origin of the unequal motion. First ascertain if the rocker-arm is not sprung. If that is all right, examine the link, which is probably sprung out of its true radius. To straighten the rocker-arm is an easy matter, but not so with case-hardened links; although some men are very successful in springing them back. Where it is impracticable to remedy an unequal cut-off by correcting the origin of the defect, several plans may be resorted to for obtaining the required adjustment. One of the most common resorts is to equalize the forward motion by throwing out the back motion. Putting the rocker-arm away from its vertical position when the valve is in the middle of the seat, by shortening or lengthening the valve-rod, provides a means of adjustment. Sometimes the equality of lead opening is sacrificed to obtain equality of cut-off. The changes necessary to obtain adjustment of a distorted motion can only be successfully arranged by one who has experience in valve-setting or in valve-motion designing.

In many shops the cut-off is adjusted for the point where the engine does most of the work,—say at 6 inches. Other master mechanics direct the equalization to be made for half stroke, while some take the mean between the half stroke and the ordinary working notch.

The final adjustments in valve-setting ought to be made when the engine is hot.


Table of Contents | Contents Page

Home
Do you have any information you'd like to share on this subject? Please email me!
The Catskill Archive website and all contents, unless otherwise specified,
are 1996-2010 Timothy J. Mallery