Question 105. How can the strength of such a single-riveted boiler seam be increased?

Answer. The most obvious way of increasing the strength of such a seam is to place, or, as it is called, space, the rivets further apart, which would leave more metal between the holes, and thus strengthen the seam at its weakest part. But if this is done, it is said that there is difficulty in keeping the seam water-tight, as the plates are then liable to spring apart between the rivets. Another way of increasing its strength is to drill the rivet holes. As already stated, the difference in the strength of the metal left between drilled and punched holes has been shown to be from 10 to 20 per cent. There is also another advantage in drilling the holes for rivets. In punching them, it is necessary to punch each plate separately, and even with the utmost care and skill it is impossible to get the holes to match perfectly. Some of them will overlap each other, as shown in fig. 57, so that when the rivet is set, it will assume somewhat the form shown in fig. 58. There is then danger that those rivets which fill the holes that match each other will be subjected to an undue strain. If, for example, we have five rivet holes, as shown in fig. 59, and only the centre ones correspond with each other, then the rivets in all the other holes will assume somewhat the form shown in fig. 58, and therefore the centre rivet c, in fig. 59, which fits the holes accurately, must take the strain of the other four until they draw up “to a bearing.” Under such circumstances, which are not unusual, there will be great danger either of shearing off the rivet c, or of starting a fracture in the plates, as indicated by the irregular line a b, between the adjoining rivets. It is also obvious that a rivet like the one in fig. 58 will not hold the plates together so well as one which fits more perfectly, as shown in section in fig. 53, and therefore there is more danger of leakage between the plates from badly fitted rivets than from those which fill the holes more perfectly; consequently rivets which fit imperfectly must be placed nearer together than those which are well fitted. It is true that rivets which are set with a riveting machine fill any inaccuracies of the holes more perfectly than those which are set by hand. But even if they are made to fill the holes as shown in fig. 60 they are still not so strong to resist shearing nor so efficient in holding the plates together as they would be if the holes conformed more perfectly to each other. In drilling the holes, the second plate can be drilled from the holes in the first, so that the holes in each will correspond with each other perfectly. The rivets will therefore fit more accurately, and consequently can be spaced further apart, and still keep the plates tight, and thus have more material between the holes, which is the weakest part of the seam. It has been shown that a rivet ¹¹⁄₁₆ inch in diameter has a resistance to shearing of 18,560 pounds. There is therefore no advantage in spacing such rivets further apart than 1¹³⁄₁₆ from centre to centre, because the metal left between drilled holes that distance apart would be slightly stronger than the rivets. If therefore the rivets are placed further apart, their diameter must be increased. There is, however, a limit beyond which the diameters of rivets cannot be increased with advantage, because if we increase their diameters, their sectional area to resist shearing is increased in proportion to the square of the diameter, whereas the section of metal in the plate to resist crushing is increased only in proportion to the diameter. This will be apparent if we compare a rivet ¹⁄₂ inch with one 1 inch in diameter. The first has a sectional area of .1963 inch, the other .7854 inch, or four times that of the first one. Now the area which resists the crushing strain of the rivets is increased only in proportion to their diameters, or is twice as much for the one as for the other. If, therefore, we increase the diameters of the rivets, we very soon reach a point at which the plate has less strength to resist crushing than the rivet has to resist shearing. The diameter of rivet which will give just the same resistance to both strains varies with the thickness of the plates; with ³⁄₈ inch plates a ⁷⁄₈ rivet will have a resistance to shearing of 30,065 pounds and the plate in front of it a resistance to crushing of 29,530 pounds. A ⁷⁄₈ rivet is, therefore, the largest size which can be used to advantage in ³⁄₈ plates. If now we were to space such rivets so far apart that the metal left between the holes would have a strength just equal to that of the rivets, we would have the strongest possible seam that can be made with a single row of rivets. This distance would be 1³⁄₄ inches between the edges of the rivets, or 2⁵⁄₈ from center to center, as shown in fig. 61. The following table will show the strength of such a seam composed of four rivets, and two plates 10¹⁄₂ inches wide,[31] with drilled holes:

From this it is seen that the strength of the seam with drilled plates is 60 per cent. of that of the solid plates, or it is about 18¹⁄₂ per cent. stronger than that made with plates having punched holes and the rivets nearer together. It should be noted that a great part of the superiority of the seams made with drilled holes is due to the superior accuracy of the work done in that way, which makes it possible to use larger rivets spaced further apart. It is probable that with the use of some recently designed machines, intended to produce greater accuracy in punching rivet holes, part of the above advantage may be realized with that kind of work. The greatest distance that rivets may be spaced apart without incurring danger of leakage between the plates must, however, be determined more by practical than theoretical considerations. It is certain, however, that rivets may be spaced much further apart than they are in ordinary practice, and the seams still be kept tight, if the work is done with sufficient accuracy and care.

Question 106. What other methods are there of making boiler seams which are stronger than those which have been described?

Answer. In this country two rows of rivets are used and also what is called a “welt,” or covering-strip, the latter with both single and double-riveted seams.

Question 107. How are the rivets arranged when two rows are used?

Answer. They are often placed just behind each other as shown in fig. 62, which is called chain-riveting. Such an arrangement of rivets obviously adds nothing to the strength of the seam, because its weakest part, as has already been shown, is the section of the plates through the rivet-holes, and this is not at all strengthened by adding another row of rivets, because the plates are just as liable to break through the rivet-holes with two rows of rivets as they would be with one.

A much better arrangement is to place them alternately in the two rows, as shown in fig. 63. Rivets arranged in that way are said to be staggered, or placed zigzag. The method of laying off the holes and proportioning such a seam in order to get the most strength is, first, to determine the greatest distance which can be allowed between the rivet-holes, and yet keep the seams water-tight. Supposing this is 1⁵⁄₁₆ inches,[32] with plates ³⁄₈ inch thick, the diameter of rivet whose sectional area will give an equal amount of strength must then be calculated. From the preceding data it will be found that for drilled plates the resistance of the portion left between such holes will be 22,148 pounds, and a rivet ³⁄₄ inch in diameter will have a resistance of 22,085 pounds. On the line a b, fig. 63, which is to be the first row of rivets, we will draw one hole, c. From the center of this hole we will describe the arc of a circle, d e, with a radius, c d, equal to the sum of the distance between the holes, 1³⁄₈ inches, and one-half the diameter of the hole, or ³⁄₈ inch, making the radius 1³⁄₄ inches. From d, the intersection of this arc with the line of rivets, a b, with the same radius, we will step off the distance d f will then be the centre of the second rivet-hole on the line a, b. The rivet can therefore be drawn, and from its centre, with the same radius employed before, another arc, d g, should be drawn. If now we draw a rivet-hole (h) between the two arcs and touching each of them, we will have all three of the holes so arranged that the metal between the holes c, f, will be just equal to that between them and the hole h. In other words, the strength of the plates on the line c f is just the same as on the line c h f. Therefore the strength of the plates on the straight line a b is just the same as on the zigzag line c h f, etc. The strength of this seam would therefore be as follows:

The weakest portion of this seam, it will be seen, has a strength of 70 per cent. of the plate, or is 38 per cent. stronger than a single-riveted seam with punched holes. If we were to use ⁷⁄₈ inch rivets and make the spaces between them 1³⁄₄ inches, the strength of a similar seam would be as follows:

This seam would then have a strength of 72 per cent. of the solid plates, or be 42¹⁄₄ per cent. stronger than the ordinary riveted seam with punched holes. It is important to observe that an increase of strength results from the use of larger rivets spaced farther apart than is usual in ordinary practice, and that this is possible only with the best and most accurate workmanship.

Question 108. What is the form of construction of boiler seams made with a welt or covering-strip?

Answer. The plates (a, b, fig. 64) are lapped over each other as for an ordinary seam. Another plate, c, about nine inches wide, is then placed on the inside of the seam and bent so as to conform to the lap of the two plates. The rivets r, whether a double or single row, pass through all three plates, and two more rows of rivets are put next to the edges of the covering plate, c. It is plain that the strength of the seam, r, is increased up to a certain point by an amount just equal to that of the rivets in the edges of the covering plate. If, however, these are placed too close together, the plates a and b will be weaker through the outside rows of rivets than the seam is through either of the outside ones and the middle one taken together. If, for example, we take a single-riveted seam, like that shown in fig. 53, whose strength is only a little more than half that of the solid plate, and should add to it a covering plate, as shown in fig. 64, and then space the rivets in the edges of the covering plate the same distance apart as in the middle seam, then obviously the plates would be just as liable to break through the outer rows of holes as through the center row before the covering plate was added. If, however, the holes in the two outside plates are spaced at say twice the distance apart, or 3³⁄₄ inches, then the only way the seam can break through the outer rows of holes is by shearing the rivets, because the plates between the holes are then stronger than the rivets. But before these rivets can be sheared, the centre seam must give way. Thus the strength of such a seam is equal to THE SUM OF THE STRENGTH AT THE WEAKEST POINTS OF THE MIDDLE AND THE OUTSIDE SEAMS. The strength of the plates between the holes of the outside rows of rivets must, however, be as great as the sum referred to, otherwise the seam will be the weakest at that point, and the failure will occur there. The rivets in the outside rows should be spaced at least twice as far apart as those in the middle seam. The number of rivets to resist shearing will then be 50 per cent. greater, so that the strength of a seam like that shown in fig. 53, with a covering plate added, will be as follows:

An ordinary single-riveted seam with punched holes, with a welt or covering plate added, would thus have a strength equal to 65.3 per cent. of the solid plate, or be 29 per cent. stronger than the seam without the covering plate. It is probable, however, that the injury to the plate from punching the outside rows of holes which are further apart is not so great as it is when they are punched nearer together and to the edge, so that the strength is somewhat greater than our estimate.

The relative strength of the different forms of seams described in percentage of the strength of the solid plate is then as follows:

Question 109. Are there any other forms of boiler seams used?

Answer. In Europe what are called butt-joints are used to some extent. In these the ends of the two plates abut against each other, with a covering strip on one or both sides. This form of joint is, however, not used in this country, and therefore its peculiarities will not be described.

Question 110. How are the seams of boilers made water-tight?

Answer. By what is called caulking. That is, by the use of a blunt instrument somewhat resembling a chisel, the end of which is placed against one or both of the edges of the plates c, d, fig. 53, which are then riveted down by blows of a hammer, somewhat as the joints of a ship are made tight. Before the edges, which are called the caulking edges, of the plates are made tight in this way, they are cut or trimmed off with a chisel. In this process the plate under the edge is often injured seriously by the carelessness of workmen, who sometimes allow the chisel to cut a groove in the plate under the edge, thus weakening it at a point where the greatest strength is needed. There is also danger of forcing the plates apart in the manner shown in fig. 65, if it is done carelessly and with a heavy hammer.

Question 111. How are the flat ends of the boiler strengthened?

Answer. By braces, u, u, r, r, fig. 41, which are fastened to the end plates and to the outer shell of the boiler. They are fastened at one end to ⟂ shaped pieces called crow-feet, which are riveted to the end plates of the boiler. At the other end they are made with a broad foot, which is riveted to the outer shell of the boiler. Especial attention should be given to the form, proportion and arrangement of these braces when a boiler is constructed, and they should be frequently examined while the engine is in use, as they are liable to be neglected or carelessly constructed and to become weakened or broken by corrosion, or the constant strain to which they are subjected. Great ignorance is often displayed in the design and proportions of these braces, especially in their attachments to the shell of the boiler.

Question 112. How should the braces for strengthening the ends of a boiler be proportioned?

Answer. Every part should be made equally strong. If, for example, the brace itself is made of a bar of round iron one inch in diameter, its sectional area would be .7854 square inch. The iron used for these braces should be of such strength that a force of not less than 50,000 lbs. per square inch of transverse section should be required to tear it apart lengthwise. A bar of the size referred to would therefore require not less than 39,270 pounds to pull it apart. All the other parts should be capable of resisting an equal strain. It is, for example, not unusual to find a brace of the size we have described, and even of larger diameter, fastened to a crow-foot which is attached to the boiler plate with two rivets ⁵⁄₈ inches in diameter. A similar fastening for the other end of the brace is also often used. The sectional area of these two rivets is considerably less than that of the brace, and at the same time the strain is brought upon them at such an angle as to have a tendency to “snap” them off. For this reason, and also because a rivet is apt to be deteriorated in strength by the hammering, the rivets should always have a sectional area very nearly or quite double that of the bar which forms the brace. The metal around the eyes through which the pins are inserted should also be carefully proportioned, and the transverse section at any one point should always be at least 1¹⁄₂ times greater than that of the bar. The area of the pins used for attaching the bars to the crow-feet should always be a little more than half that of the bar. That is, an inch bar should have a pin not less than ³⁄₄ inch in diameter. When flat braces are used, it is not unusual to find a bar 3 inches wide with a hole an inch in diameter punched or drilled so near the end that a pin is sure either to pull out the end of the bar or else to break it crosswise at the hole with a much less strain than the brace itself would resist. The ends of braces should always be enlarged enough to give them sufficient strength to resist as much strain as the bar itself. Although these precautions may appear unimportant, and unfortunately are often so regarded, yet it is upon just such details as these that the lives and the safety of every locomotive runner, fireman and others near them are constantly dependent.

Question 113. How much water is usually carried in a locomotive boiler?

Answer. There must always be enough water in the boiler to cover completely all the parts which are exposed to the fire, otherwise they will be heated to so high a temperature as to be very much weakened or permanently injured. In order to be sure that all the heating-surface will at all times be covered with water, it is usually carried so that its surface will be from 4 to 8 inches above the crown-sheet.

Question 114. How much space should there be over the water for steam?

Answer. No exact rule can be given to determine this. It may, however, generally be assumed that the more steam space the better. In order to increase the steam room, locomotive boilers are very generally made in this country with what is called a wagon-top, C, fig. 41, that is, the outside shell of the boiler over the fire-box is elevated from 4 to 12 or even 18 inches above the cylindrical part.

Question 115. What is a steam-dome and for what purpose is it intended?

Answer. A steam-dome, X, fig. 41, is a cylindrical chamber made of boiler-plate and attached to the top of the boiler. Its object is to increase the steam room and to furnish a reservoir which is elevated considerably above the surface of the water, from which the supply of steam to be used in the cylinders can be drawn. The reason for drawing the steam from a point considerably above the water is that during ebullition more or less spray or particles of water are thrown up and mixed with the steam. When this is the case, steam is said to be wet, and when there is little or no unevaporated water mixed with it it is said to be dry. It is found by experience that wet steam is much less efficient than that which is dry. There is also danger that the cylinders, pistons or other parts of the machinery may be injured if much water is carried over from the boiler with the steam, because water will be discharged so slowly from the cylinders that there is not time for it to escape before the piston must complete its stroke, so that the cylinder-heads will be “knocked out,” or the cylinder itself or the piston will be broken. The reason for drawing or “taking” steam from a point considerably above the water is because there is less spray there than there is near the surface, and the hottest steam, which is also the dryest, ascends to the highest part of the steam space.

Question 116. Where is the dome usually placed?

Answer. In this country it is usually placed over the fire-box, but in Europe it is placed further forward, either about the centre of the boiler or near the front end of the tubes. Sometimes two domes are used on engines, in this country, one over the fire-box and another near the front end.

Question 117. How is the steam conducted from the dome to the cylinders?

Answer. By a pipe I m m, fig. 41, called the dry-pipe, which extends from the top of the dome to the front tube-plate. On the front side of the tube-plate and inside the smoke-box two curved pipes, O, fig. 41, (shown also in fig. 40,) called steam-pipes, are attached to the dry-pipe at one end, and to the cylinders at the other. The vertical portion of the dry-pipe in the dome, sometimes called the throttle-pipe, is usually made of cast iron, the horizontal part of wrought iron, and the steam pipes of cast iron.

Question 118. How is the loss of heat from locomotive boilers by radiation and convection prevented?

Answer. By covering the boiler and dome with wood, called lagging, about ⁷⁄₈ inch thick, which is a poor conductor of heat, and then covering the outside of the wood with Russia iron, the smooth, polished surface of which is a poor radiator of heat.

Question 119. What is the smoke-box for?

Answer. The smoke-box Q is simply a convenient receptacle for the smoke before it escapes into the chimney or smoke-stack, which is attached to the top of the smoke-box. It also affords a convenient place for the steam and exhaust pipes, where they are surrounded with hot air and smoke, and not exposed to loss of heat by radiation. The front end of the smoke-box is usually made of cast iron, with a large door in the centre which affords access to the inside.

Question 120. How are the chimneys or smoke-stacks of locomotives constructed?

Answer. The forms of smoke-stacks which have been used are almost numberless. For burning bituminous coal and wood they are generally made with a central pipe, R, fig. 41, and a conical-shaped cast-iron plate, S, called the cone or spark deflector, which, as the latter name implies, is intended to deflect the motion of the sparks and cinders which are carried up with the ascending current of smoke and air in the pipe R, so as to prevent them from escaping into the open air while they are incandescent, or “alive.” A wire netting, t t, is also provided, which is intended as a sort of sieve to enclose the sparks and cinders, and at the same time allow the smoke to escape. The receptacle h h is intended as a chamber in which the burning cinders will be extinguished before they escape. For burning anthracite coal, a simple straight pipe, without a deflector or wire netting, is ordinarily used.

Question 121. What are the proportions and materials usually employed in the construction of smoke-stacks?

Answer. The inside pipe R, fig. 41, is usually made of the same diameter as the cylinders, or an inch or two smaller. For the other dimensions there are no established rules, excepting for the height of the top of the chimney above the rail, which is usually from 14 to 15 feet. The outsides of smoke-stacks are made of sheet iron, but the upper part is now sometimes made of cast iron, so as to withstand the abrasion of the sparks and cinders longer than sheet iron will. For very warm and damp climates, the outsides of smoke-stacks are sometimes made of copper to resist corrosion, which is very destructive to all iron structures in those countries. The wire netting is made of iron or steel wire from ¹⁄₁₀ to ¹⁄₃₀ of an inch in diameter, and with from 3 to 4 meshes to the inch.

Question 122. What is the pipe N, fig. 41, intended for?

Answer. It is intended to conduct the exhaust steam and a portion of the smoke from the bottom of the smoke-box, where the steam escapes, to the base of the smoke-stack. In nearly all European engines, the exhaust steam escapes at the top of the smoke-box just below the aperture of the smoke-stack. There is, however, often difficulty in equalizing the draft in the tubes, that is, to get an equal amount of smoke to pass through all of them. By the use of the pipe P, called the inside pipe or petticoat pipe, it is thought that the draft in the tubes can be equalized much better, as a part of the smoke is drawn out of the smoke-box at the top and part at the bottom of the smoke-box. The pipe is usually made in two parts, which slide into each other like a telescope. The distance of the upper end from the top of the smoke-box and that of the lower end from the bottom can thus be increased or diminished, and if the draft is greater through the upper tubes than through the lower ones, or vice versa, it can be regulated or equalized by simply raising or lowering the top or bottom of the petticoat pipe. Sometimes this pipe is made with openings and a kind of deflectors over them between the two ends. It is then called a flounced petticoat pipe, for obvious reasons.

No exact theory can be stated regarding the proportions of these pipes, or the results effected by them, which can be determined only by practical experience. Some more accurate knowledge concerning them is, however, much needed.