Question 82. How does the quantity of steam generated in locomotive boilers in a given time compare with that generated in the boilers of stationary and marine engines?

Answer. Locomotive engine boilers must produce much more steam in a given time, in proportion to their size, than is required of the boilers of any other class of engines, (excepting perhaps those of steam fire-engines,) because the space which locomotive boilers can occupy and also their weight is limited.

Question 83. How is their steam-generating capacity increased above that of marine and stationary boilers?

Answer. By creating a very strong draft of air through the fire and then passing the smoke and heated air through a great many small tubes, which are surrounded by water. By this means the smoke and hot air are divided into many small streams or currents which are exposed to the inside surface of the tubes to which and to the surrounding water their heat is imparted.

Question 84. How is the action of the exhaust steam in producing a draft in the chimney explained?

Answer. The exhaust steam escapes from the cylinders through one or two contracted openings or exhaust-nozzles (f, Plate II, also shown in fig. 40[21]), which point directly up the centre of the chimney or smoke-stack. The exhaust steam escapes from this orifice with great velocity, and expands as it rises, so that it fills the pipe p and the smoke-stack R R. It thus acts somewhat like a plunger or piston forced violently up the chimney, and pushes up the air above it, and, owing to the friction of the particles of air, carries that which surrounds it along up the stack, from which it all escapes finally into the open air, thus leaving a partial vacuum behind in the smoke-box. The external pressure of the atmosphere then forces in air through any and every opening in the smoke-box, to take the place of that already drawn out or exhausted from it. As the only inlet is through the tubes, to which the gases of combustion have free access from the fire-box, and as the external air can only pass through the fire-grate, and through the burning fuel, to reach the fire-box, there is a constant draft of air through the grate as long as the waste steam escapes from the blast-pipe and up the chimney. It is thus that, within certain limits, the more the steam that is required, the more the steam that is produced; for all the steam used in the engine draws in the air in its final escape, to excite the fire to generate more steam.[22] Sometimes one blast-orifice is used for each cylinder, as shown in plates II and III and fig. 40; in other cases the exhaust steam from each cylinder escapes through the same orifice.

Question 85. How much water is it necessary to evaporate in order to furnish the steam required to run an ordinary train at its usual speed?

Answer. For an ordinary “American” locomotive,[23] weighing 60,000 lbs. and with cylinders of 16 inches diameter and 24 inches stroke, from 6,000 to 12,000 lbs. of water must be evaporated per hour.

Question 86. How much water will a pound of coal evaporate in ordinary practice?

Answer. The quantity of water which is converted into steam by a pound of coal varies very materially with the quality of the coal, and the construction and condition of the boiler; but from 6 to 8 lbs. of water per pound of coal is about the average performance of ordinary locomotives. It is, therefore, necessary to burn from 500 to 2,000 lbs. of coal per hour in order to generate the quantity of steam required by ordinary engines.

Question 87. How large a grate is needed to burn this quantity of coal?

Answer. The maximum rate of combustion may be taken at about 125 lbs. of coal on each square foot of grate surface per hour, so that to burn 2,000 lbs. we need a grate with about 16 square feet of surface.

Question 88. How much heating surface is needed for a given size of grate?

Answer. In common practice about 50 square feet of heating surface are given for each square foot of grate. There are, however, no reasons for the proportions of either grate or heating surface which are given, excepting that it has been found that they work well in practice. It is, however, quite certain that the larger a boiler is, and the greater its heating surface in proportion to the steam it must generate, other things being equal, the more economical will it be in its consumption of fuel, or, in other words, the more water will it evaporate per pound of coal.

Question 89. Why is it necessary to use small tubes or flues in order to have the required amount of heating surface?

Answer. Because there is a great deal more surface in a small tube of a given length, in proportion to the space it occupies, than in a large one. Thus a tube two inches in diameter and eleven feet long has 829 square inches of surface, and one four inches in diameter has 1,658 square inches, or just double the quantity. But the four-inch tube occupies four times as much space as the other, as it is twice as high and twice as wide. Therefore, in proportion to the space it occupies, the tube which is two inches in diameter has twice as much surface as the larger one. If we compare a two-inch with an eight-inch tube, we will find that the former has four times as much surface, in proportion to its size, as the eight-inch tube. As the size and weight of locomotive boilers are limited, it is therefore necessary, in order to get the requisite heating surface in the space to which we are confined, to use tubes of small diameter.

Small tubes also have the advantage that they may be made of thinner material, and yet have the same strength to resist a bursting pressure from within, or a collapsing pressure from without, as larger tubes made of thicker metal. The advantage of thin tubes is, that the heat inside of them is conducted to the water outside more rapidly than it would be through thicker metal, which is important when combustion is as rapid as it is in locomotive boilers.

The reason tubes of smaller diameter than two inches are not ordinarily used is because they are then liable to become stopped up with cinders and pieces of unconsumed fuel.

Question 90. How is the fire-box of a locomotive constructed?

Answer. It usually consists of a rectangular box (G, figs. 41 and 42) about three feet wide[24] and, for the size of engine we have selected as an example, about five or five and a half feet long inside. This box is composed of metal plates, either iron, steel or copper, which, excepting on the front side, are from ⁵⁄₁₆ to ³⁄₈ of an inch thick. This box is called the inside shell of the fire-box, and is surrounded by another shell, A B C D E F, fig. 42, of either iron or steel plates, of about the same thickness as those composing the inside. This is called the outside shell of the fire-box and, as already explained, is so much larger than the inside that there is a space, called the water-space, from 2¹⁄₂ to 4¹⁄₂ inches wide, on all the sides of the fire-box between the inner and outer plates.

The top g, g, of the inside shell, which is called the crown-sheet or crown-plate, is flat, whereas the outside shell is arched, as shown in fig. 42. To the front plate of the inside shell the tubes a a′, a a′ are attached. For this reason its thickness is usually made greater than that of the other plates, and is usually from ³⁄₈ to ³⁄₄ of an inch. The edges of one of the plates at each corner of the fire-box, where they are united together, as shown in figs. 41 and 42, are bent at right angles, and the other is fastened to it with rivets from ⁵⁄₈ to ³⁄₄ of an inch in diameter.

The inside and the outside shells of the fire-box are united to each other by a wrought-iron bar or ring (A F, figs. 41 and 42) which completely surrounds the inner shell and closes the water-space between the two shells. This bar is bent and welded to the proper form to extend around the bottom of the inside fire-box, and it is riveted to both shells. The water in the water-space is in free communication with the rest of the water in the boiler; and thus the flat sides of the respective shells of the fire-box are exposed to the full pressure of the steam, which tends to burst the outside shell and collapse the inside one. These flat sides, by themselves, would be unable to resist the strain upon them, but as the strain upon the respective fire-boxes is in opposite directions, and necessarily equal for equal areas of surface, tie-bolts, n, n, n, n, (figs. 41 and 42,) or, as they are called stay-bolts, which are from ³⁄₄ to 1 inch in diameter, are screwed through the plates at frequent intervals, usually from 3¹⁄₂ to 4¹⁄₂ in. apart, so as to connect the two fire-boxes securely together, the ends of the stay-bolts being also riveted or spread out by hammering so as still further to increase their holding power. These bolts, owing to the expansion and contraction of the boiler and other strains to which they are subjected, very frequently break, and if they are made of solid bars of metal there is no way of discovering with certainty whether they are in good condition or not without taking the boiler to pieces. They should therefore be made of the best quality of wrought iron, brass or copper and should also be made tubular, that is they should have a hole through the centre, so that when they break the water will escape at the fracture into the hole and the leak will thus indicate the defect and danger. The latter is much greater from this cause than is usually supposed, and it is not unusual to find on taking a boiler to pieces that a large number of the stay-bolts are broken.

Question 91. How can the strain on the flat surface of a boiler between the stay-bolts be calculated?

Answer. By MULTIPLYING THE AREA IN INCHES BETWEEN ADJACENT STAY-BOLTS BY THE PRESSURE. The reason for this is, that each stay-bolt must sustain the pressure on a part of the plate to which it is attached. Thus in fig. 43 it is plain that the bolt s must sustain the pressure on one-half of that part of the plate between it and the bolts v, t, w, u, around it, or the pressure on the square a b d c, whose sides are equal to the distance (4 inches) between the centres of the bolts. With a pressure of 100 pounds per square inch, the calculation would therefore be: 4 × 4 × 100 = 1,600 lbs. on each bolt.

Stay-bolts should never be subjected to a strain of more than one-eighth or one-tenth of their breaking strength.

Question 92. How do stay-bolts often fail without breaking?

Answer. By tearing or stripping the thread of the bolt, or that in the plate, but oftener perhaps by the stretching of the plates between the holes. With a heavy pressure, the tendency of the plates between the holes, especially if they are heated very hot, is to “bulge” outward and thus stretch the hole in every direction until it is so large that the bolt is drawn out without much injury to the screw-thread.

Question 93. How is the flat-top or crown-sheet strengthened?

Answer. It is sometimes strengthened with stay-bolts similar to those used for the sides, which pass through the inner and outer shells;[25] but usually the crown-sheet is strengthened by a series of iron bars, (f, f, fig. 41 and 42) called crown-bars, placed on edge, and of considerable depth, which are firmly fastened to it by T-head rivets or bolts. The crown-sheet can therefore only be crushed downwards by bending these bars, which are of great strength. They usually extend crosswise of the length of the fire-box, but are sometimes placed lengthwise. These bars bear on the fire-box only at each end, as shown in fig. 42, and are usually made with a projection, k, k, which rests on the edge of the side plates. Iron rings or washers from ³⁄₄ to 1¹⁄₂ inches thick are interposed between the plate and the bars at the points where the bolts or rivets which secure the rivets pass through. This permits the water to circulate under the bars, and prevents the crown-sheet from being burnt or overheated, as it would be if the water were excluded from the whole under surface of the crown-bars. [26]The crown-bars are also attached to the outer shell and the dome by braces, e, e, l, l.

The opening c, fig. 41, at the back end is for the door through which fuel is supplied to the grate.

Question 94. How are the grates constructed?

Answer. They are generally made of cast-iron bars, and for burning coal are usually arranged so that the fire can be shaken by moving the bars. For burning anthracite coal, the grates are sometimes made of wrought-iron tubes, through which a current of water circulates to prevent them from being overheated.

Question 95. How are cinders and burning coals prevented from falling through the grate upon the road?

Answer. By attaching a sheet-iron receptacle or ash-pan (d′ d′, fig. 41) as it is called, under the grate, which it completely encloses from the outside air. This then serves two purposes, as it is often important when the engine is standing still to prevent any access of air to the fire-box, and therefore the ash-pan is made to fit tightly to the fire-box. Suitable doors, or dampers as they are called, are placed in front and behind, and sometimes on the sides, which can be opened or closed to admit or exclude air as may be needed.

Question 96. How are the tubes or flues of a locomotive arranged?

Answer. They are fastened into accurately drilled holes in the tube sheet (a, a, figs. 41 and 42) which forms the front of the fire-box and in similar holes in a plate (a′, a′, fig. 41,) which forms the front end of the cylindrical part of the boiler. They thus connect the fire-box with the smoke-box. The tubes are arranged so that each tube will have a space of from ⁵⁄₈ to ⁷⁄₈ of an inch between it and those adjoining. The position of the holes for the tubes in relation to each other is determined by describing from the centre of one tube (o, fig. 44) a circle with a radius, o k, equal to the sum of the diameter of a tube and the distance which they are intended to be apart, and then subdividing this circle with the radius into six parts, k, r, s, l, g and p. Each point of subdivision and also the centre, o, of the circle will be the centre of a tube. By drawing them from these centres it will be found that the distances a b, c d between adjoining tubes will be the same between all of them. By describing circles from the centres of the outside tubes and subdividing the circles as before the position of other tubes will be determined around those first laid down. This can, of course, be carried out indefinitely. A difference in the arrangement of the tubes will be observed if, when we subdivide the first circle shown in fig. 44, instead of commencing from the intersection of a vertical line we begin from a horizontal line, h i, as shown in fig. 45. In the former case the tubes are said to be in vertical rows, and in the latter in horizontal rows. It is apparent from the figures and as shown by the arrows that the water can circulate in ascending currents more freely when tubes are arranged in vertical rows than when they are arranged horizontally.

Question 97. How are the tubes fastened and made water-tight in the tube-sheets?

Answer. They are inserted into the holes drilled to receive them, and the ends are allowed to project about a quarter of an inch beyond the tube-sheets. A tapered plug, fig. 46, is then driven into the tube, to expand it so that it will fit the hole. A tool is used called a tube-expander, fig. 47, which is what might be called an expanding plug, consisting of a number of sections, a, b, c, d, e, f, g, h, held together by a spring clasp s, which embraces them, as shown in the engraving. This plug when the sections are drawn together is inserted into the mouth of the tube, and the tapered plug p p, is then driven into the opening left in the center of the cluster of sections, which are thus expanded. By this means, the ridge a b c d expands the tube at the inner edge of the tube-sheet, forming a ridge or corrugation, as shown at c c, fig. 48. At the same time the shoulder j k l on the tool expands the outer edge of the tube somewhat as is shown at f f, fig. 48. By repeating this process, and slightly turning the expander each time, the tubes can be made perfectly water-tight. There are other forms of tube-expanders, but the one described, known as Prosser’s expander, is more generally used than any other. In many cases, after the tubes are expanded with the tool described, the outer edge is turned over still more with what is called a thumb-tool, fig. 49, probably from its resemblance in form to a man’s thumb. By placing the curved shoulder a on the end f, fig. 48, of the tube it is turned over, somewhat in the form shown in the engraving, by repeated blows of a hammer on the end of the tool. Copper ferrules, represented by the black shading, a a, are also much used now on the outside of locomotive tubes, and it is said that with them the joints can be kept tight much easier than without. By turning over the outside edge of the tube as shown in fig. 48, it not only protects the copper ferrule, but, as the tubes must act as braces to sustain the pressure of steam in the flat tube-sheets, it gives the joints the requisite strength for resisting such strains.

Question 98. How can the strain on the cylindrical part of a boiler be calculated?

Answer. By multiplying the diameter in inches by the length in inches and the product by the steam-pressure per square inch. Thus for a boiler 48 inches in diameter and 10 feet long with 100 pounds pressure the calculation would be 48 × 120 × 100 = 576,000 lbs.

Question 99. Why do we multiply the diameter, instead of the circumference, by the length, to get the strain on the cylindrical part?

Answer. The reason for multiplying by the diameter instead of by the circumference is because only a portion of the pressure on the inside surface of the boiler exerts a force to burst the shell at any one point. Thus, supposing the following diagram, fig. 50, to represent a section of a boiler, if we have a force acting on the shell in the direction of the line a b, at the point b, where it is exerted against the shell of the boiler, it would be composed of two forces, one acting in the direction b e, and tending to tear the boiler apart on the line c d, and the other acting in the direction f b, to tear it apart on the line h g. It is so with all pressure inside the boiler, excepting that, say a h, which acts exactly at right angles to the line of rupture c d, it is all composed of two forces, only one of which tends to tear the boiler apart at one point. It is therefore only a part of the pressure on the circumference which tends to burst the boiler at a given place, and that part is equivalent to the pressure on a surface whose width is equal to the diameter and not the circumference.

This we know is a little difficult for those to understand who are not familiar with the principles of what is called the “resolution of forces,” and we will therefore try to make it clear in another way.

To do this we will suppose that we have a boiler, a b, fig. 51, made in two halves and bolted together at a and b by flanges. It is evident that if we brought a pressure against the inside of the flanges in the direction of the darts c and d, such a pressure would not have a tendency to tear apart the bolts a and b. Some distortion of the boiler might in fact take place, if, for example, we put a jack-screw inside and forced out the flanges as indicated, without subjecting the bolts to a tensile strain. We see therefore that the forces acting in the direction c and d have no tendency to tear apart the bolts at a and b, but it is only the forces such as e, f and g, which act at right angles to a b, that exert a strain on the flanges.

That this force is equivalent to a pressure on a surface with a width equal to the diameter of the boiler is apparent if we suppose that we have a boiler, a b, fig. 52, and that each half, c and d, is nearly filled with some substance, say wood or cement, which is fitted so tight that no steam can get between it and the shell of the boiler. It is apparent now that if we admit steam into the space f, the force exerted on the bolts a and b is that due to the pressure on the surface of the wood or cement exposed to the steam whose width is equal to the diameter of the boiler. It might be said though that if this substance were elastic, like india-rubber, the effect of the steam would be different. If it were elastic, and a pressure on the surfaces f caused it to spread in the direction g and h so as to produce a pressure in those directions, it would, as has already been shown, not exert a force on the bolts a and b to tear them apart, but have a tendency to rupture the boiler at right angles to a b. The sides of the boiler must therefore have a strength sufficient to resist this force which tends to tear them asunder. If the boiler is made of iron ³⁄₈ inch thick there would be a sectional area of 45 square inches on each side, or a total of 90 square inches, to resist this strain, so that each square inch must bear 6,400 lbs. of strain. The correctness of this rule can be demonstrated by the use of mathematics, which would be out of place here. Its practical truth has however been proved by experiment.

Question 100. How much strain per square inch is good boiler plate capable of resisting, and how much is it safe to subject it to?

Answer. There is great variation in the tensile strength[27] of rolled iron boiler plate, but that of good plate will average about 50,000 pounds per square inch, if the strain is applied in the direction of the “grain” or the fibres of the iron[28], and about ten per cent. less if the strain is applied crosswise of the grain. It has, however, been found by experiment that when a tensile strain is applied to a bar of iron or other material, it is stretched a certain amount in proportion to the length of the bar and to the degree of strain to which it is subjected. It is found that if this strain does not exceed about one-fifth of that which would break the bar, it will recover its original length, or will contract after being stretched, when the strain is removed. The greatest strain which any material will bear without being permanently stretched is called its limit of elasticity, and so long as this is not exceeded no appreciable permanent elongation or “set” will be given to iron by any number of applications of such strains or loads. If, however, the limit of elasticity is exceeded, the metal will be permanently elongated, and this elongation will be increased by repeated applications of the strain until finally the bar will break. At the same time the character of the metal will be altered by the repeated application of strains greater than its elastic limit, and it will become brittle and less able to resist a sudden strain, and will ultimately break short off. It is therefore unsafe to subject iron, or in fact any other material, to strains greater than its elastic limit. This limit for iron boiler plates may be taken at about one-fifth its breaking, or, as it is called, ultimate strength. It should be remembered, however, in this connection, that it often happens that the steam pressure is not the greatest force the boiler must withstand, as sudden or unequal expansion and contraction are probably more destructive, to locomotive boilers especially, than the pressure of the steam.

Question 101. How are the plates of boilers fastened together?

Answer. With rivets, which are made with a head at one end, and are inserted while they are red-hot into holes drilled or punched in the edges of the plates. After they are in the holes a head is formed on the other end, either with blows from hand hammers, or by a machine constructed for the purpose. In these machines the rivet after it is in the holes is brought between a fixed and a movable die, the head which is made with the rivet being placed against the fixed die, and the movable die is then pressed, either by steam or hydraulic pressure with great force against the other end of the rivet, thus forcing the end of the rivet into the form of the die, which is made of the proper shape and size for the rivet head. The powerful pressure which is thus brought on the rivet causes it to be pressed into all parts of the two holes, thus completely filling them both; whereas with hand riveting, the holes are not nearly so completely filled, as it is impossible with blows of a hammer to subject the rivets to so powerful or uniform a pressure as the machine brings upon them.

Question 102. What is the strength of riveted seams compared with that of the solid plate?

Answer. The strength of a riveted seam depends very much upon the arrangement and proportion of the rivets, but with the best design and construction, the seams are always weaker than the solid plates, as it is always necessary to cut away a part of the plate for the rivet holes, which weakens the plate in three ways: 1. By lessening the amount of material to resist the strains. 2. By weakening that left between the holes. 3. By disturbing the uniformity of the distribution of the strains. The first cause of weakness is obvious from an inspection of an ordinary seam, riveted with a single row of rivets, fig. 53. In this we have two plates 7¹⁄₂ inches wide and ³⁄₈ thick fastened with four rivets ¹¹⁄₁₆ inches in diameter and 1⁷⁄₈ inches from centre to centre. The section of the plate calculated with decimals[29] would therefore be .375 × 7.5 = 2.81 square inches. A piece ¹¹⁄₁₆ inch wide and ³⁄₈ inch thick would be removed to form each hole, or a sectional area for the whole plate of .375 × .6875 × 4 = 1.03 square inches, so that the section of the plate would be reduced through the holes 2.81 - 1.03 = 1.78 square inches. In other words, on the dotted line a b it will have only about 63 per cent. of the sectional area of the solid plate.

The second cause of the reduction of strength is owing to the injury sustained by the plates during the process of drilling and punching. The knowledge existing regarding this subject is not very accurate, although numerous experiments have been made to determine the exact amount of weakening caused by punching plates. It is, however, certain that in many cases the strength of the metal left between the holes of boiler plates is reduced from 10 to 30 per cent. by the process of punching. It is probable, however, that soft ductile metal is injured less than that which is harder and more brittle. Some kinds of steel plates are especially liable to injury from punching. It is also probable that the condition of the punch, and the proportions of the die used with it, have much to do with its effect upon the metal

The third cause of weakness is owing to the fact that if one or more holes are made in a plate of any material, and it is then subjected to a tensile strain, the strain, instead of being equally distributed through the section left between the holes, will be greatest in that part of the metal nearest them. This can be illustrated by taking a band of india-rubber, fig. 54, and cutting a round hole in it to represent a rivet hole. If we draw two parallel lines, A B, across the band and then stretch it, the lines, instead of remaining parallel when the band is stretched, will separate most next to the hole, as shown in fig. 55, indicating that the fibres of the rubber nearest the hole are strained most. A similar effect takes place when a plate of iron is stretched, so that a fracture is liable to begin next to the hole, after which the plate will be broken as it were in detail.

Question 103. How may a boiler seam like that shown in fig. 53 break?

Answer. It may break in three different ways:

1. By the plate tearing between the rivet holes on the line a b.

2. By the rivets shearing off.

3. By the plate in front of the rivets crushing, as shown in fig. 56.

Question 104. How can the strength of a boiler seam be calculated at each of these three points?

Answer. The strength through the rivet holes is calculated by TAKING THE AREA IN SQUARE INCHES OF THE METAL WHICH IS LEFT BETWEEN THE RIVET HOLES, AND MULTIPLYING IT BY THE ULTIMATE STRENGTH OF THE METAL AFTER THE HOLES ARE MADE. Thus, in fig. 53, the area of each of the plates between the rivet holes is 1.78 square inches. As already stated, good boiler plate will break at a strain of about 50,000 pounds in the direction of its fibres,[30] but the strength of the metal left between punched holes is probably 20 per cent. and that between drilled holes 10 per cent. less than that of the solid plate. We must, therefore, in calculating the strength of a punched seam, take the ultimate strength of the metal between the holes at only 40,000 pounds per square inch. The calculation for the strength through the holes would therefore be: 1.78 × 40,000 = 71,200 pounds.

It has also been found by experiment that the strength of rivets to resist shearing is about the same as that of good boiler plate to resist tearing apart, or 50,000 lbs. per square inch. The strength of the rivets, therefore, is calculated by MULTIPLYING THE AREA IN SQUARE INCHES OF ONE RIVET BY THE NUMBER OF RIVETS, AND THE PRODUCT BY THE STRENGTH OF THE METAL TO RESIST SHEARING. The calculation for fig. 53 would therefore be:

or a little more than the strength of the plates through the holes.

The resistance offered by a plate to the crushing strain of a rivet has been found also by experiment to be about 90,000 pounds per square inch. It can be proved that the area which resists the crushing strain of a rivet in a plate, fig. 53, IS MEASURED BY MULTIPLYING THE DIAMETER OF THE RIVET BY THE THICKNESS OF THE PLATE. The calculation for the strength of this part of the seam will therefore be: diameter of hole = .6875 × .375 × 4 × 90,000 = 92,812.

The strength of the solid plate would be EQUAL TO ITS SECTIONAL AREA MULTIPLIED BY 50,000 POUNDS, or 7.5 × .375 × 50,000 = 140,625 pounds. The ultimate strength of our seam would then be as follows:

It will thus be seen that the strength of the weakest part of the above seam, fastened with a single row of rivets in punched holes, is very little more than half (50.6 per cent.) of that of the plates. It will be noticed that the weakest part of the seam is the plates between the holes.