x° <^ , ^ ^ 5. * + o5 '^> *0 c . ? -^ v %• % / ■ '• \ .y % S : J -r, $°- .4- ^ i. -J- tO * v . \ '/ , , ^ ^ \ aV «'■ ^ <«, ' A xV ' P f 4 o o N V *L. V o o ^ ^ % 4 •/' •%■ >* , d> .A . \ I 8 ,-fc* s ^ o x **. V ^ - v \ />. ,** ■*. •\ \ sO o v STEAM-BOILERS. BY CECIL H. PEABODY and EDWARD F. MILLER Professor of Marine Engineering Assistant Professor of Steam and Naval Architecture, Engineering, Massachusetts Institute of Technology. FIRST EDITION, FIRST THOUSAND. NEW YORK : JOHN WILEY & SONS, London: CHAPMAN & HALL, Limited. 1897. Copyright, 1897 C. H. PEABODY and E. F. MILLER /0 b 0^ KORERT PRUMMONLV ELECTROTYPER AND PKINTER, NEW YORK PREFACE. In this book we have attempted to give a clear and con cise statement of facts concerning boilers, and of methods of designing, making, managing, and caring for boilers. Though the book is intended primarily for the use of students in technical schools and colleges, it is hoped that it may be found useful to engineers in general. There is given a description of various types of boilers in common use. Following this is a discussion of combustion, corrosion, and incrustation, with a statement of the most recent investigations and conclusions on these important subjects. We are fortunately able to give a satisfactory table of the compositions of American fuels — the first, so far as we are aware, that has been published. A statement is given of the proper and of the customary sizes and form of furnaces, and of the methods of firing. In the present unsatisfactory condition of the chimney problem we have contented ourselves with giving the ordinary theory and pointing out its defects, together with the common ways of proportioning chimneys. Tables of grate-areas and heating-surfaces, and of other proportions of furnaces and boilers, have been made up from the oest current practice for stationary, locomotive, and marine boilers. In the chapter on strength of boilers we have given briefly the methods and conditions for testing materials and for making boilers, and the properties which such materials IV PREFA CE. should have. Especial attention is given to the properties and proportions of riveted joints, deduced by Professors Lanza and Schwamb from tests at the Watertown Arsenal. Simpler calculations of stresses in the members of boilers are explained, and more complex ones, depending on the theory of elasticity and theories of beams and continuous girders, are illustrated by examples. A description is given of staying and other details affect- ing the design and construction of boilers, and of such acces- sories as safety-valves, gauges, and steam-traps. In order to give a conception of the methods and conditions of boiler- making, we have given a description of a modern bciler-shop and the machinery and processes used in it. In the chapter on boiler-testing we have given the methods used in the laboratories of the Massachusetts Insti- tute of Technology, including gas analysis, measurement of air used, and temperature, determinations in the furnace and chimney. Finally, the principles and methods set forth in tr e earlier chapters are brought together and illustrated by applying them to the design of a boiler of a common type. For our own students this chapter serves as an introduction to a course in machine design given by Professor Schwamb, who has kindly furnished us with methods and materials which he has collected and developed in connection with the design- ing of boilers. In the appendix are given various useful tables, such as logarithms, natural trigonometric functions, areas and circum- ferences of circles, proportions of rods and screws, and proper- ties of saturated steam C. H. P. and E. F. M. Boston, February i, 1897. CONTENTS. CHAPTER I. PAGZ Types of Boilers i CHAPTER II. Fuels and Combustion 37 CHAPTER III. Corrosion and Incrustation 65 CHAPTER IV. Settings, Furnaces, and Chimneys 91 CHAPTER V. Power of Boilers 130 CHAPTER VI. Staying and Other Details 148 CHAPTER VII. Strength of Boilers 170 CHAPTER VIII Boiler Accessories 235 v VI CONTENTS. CHAPTER IX. Shop-practice 272 CHAPTER X. Testing Boilers 300 CHAPTER XI. Boiler Design 323 APPENDIX 357 INDEX 369 STEAM-BOILERS CHAPTER I. TYPES OF BOILERS. Steam-eoilers may be classified according to their form and construction or according to their use. Thus we have horizontal and vertical boilers, internally and externally fired boilers, shell-boilers and sectional boilers, fire-tube and water- tube boilers: the several features mentioned may be combined in various ways so as to give rise to a large number of kinds and forms of boilers. Again, we have stationary, locomotive, and marine boilers, together with a variety of portable and semi-portable boilers. Locomotive boilers are always shell- boilers, internally fired, and with fire-tubes; and the re- strictions of the service have developed a form that has changed little from the beginning, except in the direction of increased size and power. Marine boilers present a much larger variety of form and construction, depending on the steam-pressure used and the size and service of the vessel to which they are supplied. The Scotch or drum boiler is more widely used than any other form at present, but the tendency to use high-pressure steam has led to the introduction of vari- ous forms of water-tube boilers for marine work. The variety of forms and methods of construction of stationary boilers is very wide: each country and section of a country is likely to have its own favorite type. Thus in New England, where 2 STEAM-BOILERS. the water is good, cylindrical tubular boilers are largely used ; in some of the Western States, where water contains mineral impurities, flue-boilers are preferred; and in England, the Lancashire and Galloway boilers are favored; and again, various forms of sectional and water-tube boilers are now widely used. Cylindrical Tubular Boiler. — This type of boiler is shown by Fig. i and by Plate I. It consists essentially of a cylin- drical shell closed at the ends by two flat tube-plates, and of numerous fire- tubes, commonly having a diameter of three or four inches. About two thirds of the volume of the boiler is filled with water, the other third being reserved for steam. The water-line is six or eight inches above the top row of tubes. The tube-plates below the water-line are sufficiently stayed by the tubes ; above the water-line the flat plates are stayed by through rods or stays as in Plate I, by diagonal stays like those shown by Fig. 52, page 154, or otherwise. A pair of cylindrical boilers in brick setting are shown by Figs. 36 and 37, on pages 92 and 93, with the furnaces under the front (right-hand) end. The products of combustion pass back over a bridge-zvall, limiting the furnace, to the back eud, then forward through the tubes and up the uptake to the flue which leads to the chimney. The shell commonly extends beyond the front tube-plate, as shown at the right in Fig. 1, and is cut away to facilitate the arrangement of the uptake. The boiler is usually sup- ported by cast-iron brackets riveted to the shell ; the front brackets may rest on or be fixed to the supporting side walls, but the rear brackets should be given some freedom to avoid unduly straining the boiler by expansion. Thus the rear brackets may rest on rollers, which in turn bear on a horizontal iron plate. The expansion takes place toward the back end of the boiler, and to allow for this expansion a space is left between the back tube-sheet, and the arch of fire-brick back of the boiler. TYPES OF BOILERS. 4 S TEA M-B OIL EKS. The boilers shown by Fig. i and by Plate I each have two steam-nozzles, one near each end. The safety-valve is usually attached to the front nozzle, which is above the fur- nace. The steam-pipe leading steam from the boiler is at- tached to the rear nozzle, which is over the back end of the boiler, where ebullition is less violent, and consequently there is less danger that water will be thrown into the steam-pipe. Boilers of this type commonly have a manhole on top near the middle, and a hand-hole near the bottom of each tube- sheet, as shown on Plate I, to give access to the interior of the boiler and to facilitate washing out. Many boilers are now made with a manhole near the bottom of the front tube- sheet, in addition to the one on top. All parts of the boiler can then be cleaned and inspected whenever desirable. Some of the lower tubes must be left out when there is a manhole in the tube-sheet, but this is of small consequence, as the lower tubes are not efficient, and enough heating-surface can be provided elsewhere. The omission of the lower tubes re- quires also special stays for the portion of the tube-sheet left unsupported. The feed-pipe for the boiler shown by Plate I enters the front head at the left, below the water-line, and runs toward the back end of the boiler, where it may end in a perforated pipe leading across the boiler. The feed-pipe may enter the top of the boiler, near the back end, and terminate in a similar perforated transverse pipe below the water-line. A blow-off pipe leads from the bottom of the shell near the back tube-sheet. On the blow-off pipe there is a plug or valve which may be opened when steam is up, to blow out mud and soft scale that may collect in the boiler. The boiler is com- monly set with a slight inclination toward the rear so that mud may collect near the blow-off pipe. The boiler may be emptied by allowing the water to run out at the blow-off pipe. About half of the shell, two thirds of the back tube-sheet, and all the inside surface of the tubes come in contact with TYPES OF BOILERS. the products of combustion and form the heating-surface ; all the heating-surface is below the water-line. The boiler-setting, shown by Figs. 36 and 37 on pages 92 and 93, is made of brick laid in cement or mortar; all parts that are directly exposed to the fire are lined with fire- brick. The walls have confined air-spaces to reduce transmis- sion of heat. The boiler front is commonly made of cast iron. and has fire-doors leading to the furnace, and ash-pit doors opening from the ash-pit, or space below the grate ; there are also large doors giving access to the tubes through the smoke-box at the front end of the boiler. The furnace is formed by the side walls, the bridge, and the lower part of the boiler front, which latter is lined with fire-brick above the grate. Doors through the rear wall give access to the space back of the bridge. The top of the boiler is covered by a brick arch or by non-conducting material. Two-flue Boiler. — The cylindrical flue-boiler differs from the tubular boiler mainly in replacing the fire-tubes by one or more large flues. Fig. 2 shows such a boiler with two Fig. 2. flues. This type of boiler is usually longer than a tubular boiler, but even so it has less heating-surface and is less efficient in the use of coal. Nevertheless the greater sim- plicity and accessibility for cleaning recommend* it where feed- water is bad. The setting of a flue-boiler resembles that for the cylin- S TEA M-BOIL EKS. drical tubular-boiler. The figure shows two loops at the top of the shell for hanging the boiler; a crude method of sup- porting, suitable only for small and short boilers. Plain Cylindrical Boiler. — In places where fuel is very cheap, especially where it is a waste product, as at sawmills, the plain cylindrical boiler is fre- quently used. Its external ap- pearance is similar to that of the two-flue boiler (Fig. 2), except that there are no flues and the ends are commonly hemispheri- cal or else curved to a radius equal to the diameter of the shell. Such plain cylindrical boilers are also employed to util- ize the waste gases from blast- furnaces. They are commonly 30 to 42 inches in diameter and from 20 to 40 feet long. They have been made 70 feet long. With such extreme lengths spe- cial care must be taken to insure equal distribution of the weight to the supports and to provide for expansion. Lancashire Boiler. — This boiler, shown by Fig. 3, is a two- flue shell-boiler with furnaces in the tubes; it is therefore an internally-fired boiler, in which it differs from the two pre- TYPES OF BOILERS. 7 ceding types, which arc externally-fired. The chief difficulty in the design of these boilers is to provide sufficiently large furnaces without making the external shell too large. As com- pared with the cylindrical tubular boiler, this boiler will be sure to have long, narrow grates, with a shallow ash-pit and a low furnace-crown: the boiler also appears to be deficient in heating-surface. In compensation, radiation and loss of heat from the furnace are almost entirely done away with, and the thick outside shell, with its riveted joints, is not exposed to the fire, as with the tubular boiler. The flues are made in short sections riveted together at the ends, thus forming a series of stiffening rings that add very much to the strength of the flues against collapsing. Conical through-tubes, ver- tical or inclined, give increased heating-surface, break up the currents of the hot gases, improve the circulation of the water, and strengthen the flues. These tubes are small enough at the lower end to pass through the hole cut in the flue for the upper end, and thus are readily put in or taken out for repairs. The flat plates at the ends of the shell are stayed by gusset-stays or triangular flat plates to the shell of the boiler. The boiler is provided with a manhole near the back end and a safety-valve near the front end. Steam is taken through a horizontal dry-pipe, perforated on the top. Galloway Boiler.— This boiler has two furnace-flues at the front end, like the Lancashire boiler. Beyond the furnace the two flues merge into one broad flue, having the upper and lower surfaces stayed by numerous conical through-tubes, like those shown in Fig. 3 for the Lancashire boiler. Cornish Boiler. — This boiler was developed in conjunction with the Cornish engine, and both boiler and engine long had a reputation for high efficiency. It differed from the Lanca- shire boiler in that it had but one flue; it formerly did not have cross-tubes. The one furnace of the Cornish boiler, with a given diameter of shell, can have better proportions than the two furnaces of the Lancashire boiler, but there is even o S TEA M-BOILERS. greater difficulty to get sufficient grate-area and heating-sur- face. The high economy shown by these boilers when used with the Cornish pumping-engine was due to a slow rate of combustion, and to the skill and care of the attendant, who was usually both engineer and fireman, and who was stimu- lated by a system of competition and awards, maintained by the mine-owners in that district. The Lancashire and the Cornish boilers are set in brickwork which forms flues leading around the outside shell, thus mak- ing the shell act as heating-surface. Fig. 4 gives a cross-sec- Fig. 4. tion of the Lancashire boiler and its setting. After the gases from the fires leave the internal flues they are directed into the flue a and come forward ; then they are transferred to the flue b and pass backward ; finally they come forward in the flue c, and are then allowed to pass to the chimney. This forms what is known as a wheel-draught. In some cases the gases divide at the rear and come forward through both side TYPES OF BOILERS y flues a and b, and uniting pass back through c and thence to the chimney, forming a split-draught. Vertical Boilers.— Boilers of this type rave a cylindrical shell with a fire-box in the lower end, and with fire-tubes run- ning from the furnace to the top of the boiler. Large verti- cal boilers have a masonry foundation and a brick ash-pit; small vertical boilers have a cast-iron ash-pit that serves as foundation. Vertical boilers require little floor-space; if properly designed they give good economy, or they may be made light and powerful for their size, when economy is not important. Fig. 5 shows a large vertical boiler designed by Air. Manning. It is made 20 to 30 feet high, so that there is a large heating-surface in the tubes. The shell is enlarged at the fire-box to provide a larger furnace and more area on the grate. The internal shell which forms the fire-box is joined to the external shell by a welded iron ring called the founda- tion-ring. This internal shell should be made of moderate thickness to avoid burning or wasting away under the action of the fire. Being under external pressure, the shell of the fire-box must be stayed to avoid collapsing. For this pur- pose it is tied to the outside shell at intervals of four or five inches each wry, by bolts that are screwed through both shells and riveted over cold, on both ends. The stays near the bottom have each a hole drilled from the outside nearly through to the inside end. Should any stay break or become cracked, steam will escape and give warning to the fireman. The tubes are arranged in concentric circles, leaving a space about ten inches in diameter at the middle of the crown-sheet; the corresponding space in the upper tube- sheet provides for the attachment of the nozzle for the steam outlet. There are numerous hand-holes in the shell outside of the fire-box, some near the crown-sheet, and some near the foun- dation-ring, and these are the only provision for cleaning the IO S TEA M-BOILEKS. WATER LEVEL wywjsM4s& wBsm Fig. 5. TYPES OF BOILERS. II boiler, which consequently is adapted for the use of good feed-water only. The feed-pipe enters the shell at one side and extends across the boiler; it is perforated to distribute the feed-water. The sides of the fire-box, the remaining surface of the tube-sheet allowing for the holes for the tubes, and the inside Fig. 6. of the tubes up to the water-line form the heating-surface: the inside of the tubes above the water-line form the super- 12 S TEA M-B OILERS. heating-surface, since it transmits heat from the gases to the steam and superheats it. This type of boiler has found favor at factories where floor-space is valuable, since a powerful battery of boilers may be placed in a small fire-room. A small vertical boiler adapted for hoisting, pile-driving, and other light work is shown by Fig. 6. It commonly has a short smoke-pipe, into which the exhaust steam from the engine is turned to form a forced draught and give rapid combustion. Under this treatment the upper ends of the tubes frequently g;ve trouble by leaking. To avoid this diffi- culty the tubes are sometimes ended in a sunken or submerged tube-sheet which is kept below the water-line, as shown by Fig. 7. The space between the edge of the tube-sheet Fig. 7. and the outside shell is likely to be contracted, and not to give proper exit for the steam formed on the tubes and crown-sheet. Furthermore, the cone forming the smoke- chamber above the tube-sheet is subjected to external pres- sure and is likely to be weak. A form of vertical boiler having a sunken tube-plate is shown by Fig. 8. It was at one time much used for steam fire-engines, but to save weight it was so crowded with tubes TYPES OF BOILERS. ! 3 and the water-spaces were so contracted that it gave mucl trouble when forced, as at a fire. Fire-engine Boiler. — A boiler for a steam fire-engine should be light and compact, able to make steam quickly and Fig. S. to steam freely when urged. They have small water-space and large heating-surface for their size, but are not economi- cal in the use of fuel. It is customary to use cannel-coal for fire-engines, as it burns freely without clogging. A forced 14 STEAM-BOILERS. draught is obtained by exhausting steam up the smoke-pipe. When standing in the engine-house ready for duty the boilers are kept hot by connecting them to a heating- boiler in the basement. The connection is so made with snap-valves that it is broken by pulling the fire-engine out of position. Figs. 9 and 10 show a vertical section and two half-hori- zontal sections of the Clapp fire-engine boiler. The boiler has a cylindrical shell and a deep internal fire-box. From the crown-sheet a number of fire-tubes lead through the water and steam space to the upper tube-sheet. In the upper part of the fire-box there are a number of water-tubes that start from the side of the fire-box, make several helical coils, and then open into the water-space above the crown-sheet. There are three concentric sets of these helical coils, leaving a cylindrical space in the centre, which is occupied by a series of castings, shown in perspective and partly in section by Fig. 1 1. The casting is formed of an annular torus with a cross-tube, and an inverted U tube above. Water enters at the middle of the cross-tube, passes into the torus, and then up and out at the top of the U. The left half of Fig. 10 shows the helical tubes from above ; the right half shows the arrangement of the fire-tubes and the openings of the water-tubes. Marine Boilers. — A single-ended three-furnace Scotch marine boiler is shown in perspective by Fig. 12; Fig. 13 gives the working drawings of a similar boiler with two fur- naces. The arrangement of the furnaces in the flues, is simi- lar to that for the Lancashire boiler, shown, by Fig. 3. The furnace-flue leads into a combustion-chamber, from which the products of combustion pass through fire-tubes to the uptake, which is bolted onto the front end of the boiler. The flues are from three and a half to four and a half feet in diameter; the size of the boiler depends on the number and size of the flues. Large boilers have as many TYPES OF BOILERS. 15 Fig 9. Fig. 11 i6 S TEA M-B OILERS. as four flues. A three-furnace boiler commonly has three combustion-chambers, while a four-furnace boiler may have two, into each one of which two furnaces lead. Double- ended boilers have furnaces at each end, and resemble two single-ended boilers placed back to back. A double- ended boiler is lighter, cheaper, and occupies less space than OOOOOOOn °°00000nn 0000 °0 OOOono 00o °oooo oooooo oo°°°oooo 8 SSSSS oo8S!SS »o oooppoopoo \\\\\\ ZZMllZ 000000000 oooooooooo Fig. 12. two single-ended boilers. In the best practice there are two distinct sets of combustion-chambers for the two sets of furnaces. To still further lighten double-ended boilers, common combustion-chambers for corresponding furnaces at the two ends have been used. The results from such boilers have not been satisfactory, more especially when TYPES OF BOILERS. 17 it n S^& : ~-— ^-i'izJy : l soitx 7?E + + + t 93 7 FT' : — a 1 8 S TEA M-B OILERS. used under forced draught in the closed stoke-holes of war- ships; there has been so much trouble from leaky tubes under such conditions that forced draught has been aban- doned in many cases, and ships have consequently failed to make the speed anticipated. The circulation of water is defective in all Scotch boilers, and more especially in double-ended boilers. Considerable time — three or four hours — is always allowed for raising steam. Frequently some arrangement is made for drawing cold water from the bottom of the boiler and returning it near the water- line, while steam is raised. Haste and lack of care are liable to cause leakage from unequal expansion. The flue has the highest temperature of any part of the boiler and consequently expands the most, so that some allowance for expansion must be made or it will strain the tube-sheets and cause leaks. The methods of providing for expansion and at the same time stiffening the flues against collapsing under external pressure are shown on pages 210 to 216, and will be described in de- tail later on. Gunboat Boilers. — Some gunboats and other small naval vessels have not room under the deck for Scotch boilers. The form shown by Fig. 14 has been used on such vessels; it has two furnace-flues, leading to a common combustion-chamber, from which fire-tubes lead to the back end of the boiler. Locomotive-boilers. — The typical American locomotive- boiler is shown by Plate II. Fig. 15 gives a perspective view of a boiler of the locomotive type used for small factories, or where steam is required temporarily ; it has no permanent foundation, but is supported on brackets at the fire-box and by a pedestal-bearing on rollers near the back end. The locomotive-boiler consists essentially of a rectangular fire-box and a cylindrical barrel through which numerous tubes pass from the fire-box to the smoke-box, which forms a con- tinuation of the barrel, and from which the products of com- bustion pass up the smoke-stack. TYPES OF BOILERS. *9 The fire-box is joined to the outer shell at the bottom by a forged rectangular foundation-ring, similar (except in shape) Fig. 15. to the foundation-ring of a vertical boiler. Near this ring are several hand-holes for clearing out the space between the fire- box and the shell, commonly called the water-leg. The boiler 20 S TEA M-B OILERS. also "has a manhole at the top of the barrel. The water-leg is stayed by screwed stay-bolts riveted cold at the ends. The flat crown-sheet is stayed to a system of crozun-bars which rest on the side sheets of the fire-box and are also slung from the shell. Plate III shows a locomotive-boiler with a flattened top over the fire-box, to which the crown-sheet is stayed by through-bolts. Other methods and details of stay- ing crown-sheets will be given later. The tubes for a locomotive-boiler are smaller than for sta- tionary boilers (about two inches in diameter) and are spaced much more closely. This is to obtain a large heating-surface required by the high rate of combustion, which often exceeds one hundred pounds of coal per square foot of grate-surface per hour. The boiler works under a strong forced draught, produced by throwing the exhaust up the smoke-stack. The boiler is fastened rigidly to the frame of the locomo- tive at the smoke-box end ; a small longitudinal motion on the frame at the fire-box end is provided by expansion-pads, shown by Fig. 4, Plate II. Locomotive Type of Boiler — Reference has already been made in connection with Fig. 15 to a boiler of locomotive type used for stationary purposes. Plate IV shows a modi- fication of the locomotive type designed by Mr. E. D. Leavitt to give high evaporative efficiency. The boiler represented lias a barrel 90 inches in diameter, and it is 34 feet 4 inches long over all, It supplies steam at 185 pounds pressure to the square inch to the high-duty pumping-engine at Chestnut Hill Reservoir, Boston. The fire-box of this boiler is spread at the bottom to give increased grate-area, and contains two separate furnaces, shown by the section AA on Plate IV. The products of combus- tion pass through openings, shown by section BB, into a com- bustion-chamber, which has the section shown at CC. From the combustion-chamber, the gases pass through tubes to the smoke-box and uptake. As far as the combustion-chamber TYPES OF BOILERS. 21 the top of the boiler is flattened to facilitate the staying of the crown-sheets of the furnace, passages, and combustion-cham- ber; the barrel of the boiler beyond the combustion-chamber is cylindrical. The boiler is somewhat complicated in construction and staying, and must be handled with care, especially in starting, to avoid straining from unequal expansion. It is adapted for the use of good feed-water only. Boilers of the locomotive type were at one time used for torpedo-boats. The fire-box was made shallower than for locomotive-boilers, and forced draught in a closed stoke-hole was used, the rate of combustion being even higher than on locomotives. Whatever may have been the reasons, it was a fact that this type of boiler, which is very reliable on locomo- tives, gave much trouble in torpedo-boats. Water-Tube Boilers. — The boilers thus far considered have an external shell containing a large body of water. Heat is communicated to the water through the shells or through the sides of internal furnaces, and also by carrying the gases through tubes or flues. The boilers and water contained, are heavy and cumbersome, and the shells under high pressure must be made very thick. If the boiler fails either through some defect or through carelessness of attendants, a disastrous explosion is likely to take place. If properly designed and made and if cared for by competent and careful attendants they are safe, reliable, and durable. The large mass of hot water tends to keep a steady pressure, though at the expense of rapidity of raising steam or of meeting a sudden demand for more steam. A large number of water-tube boilers of all sorts of shapes and methods of construction has been devised to overcome the admitted defects of shell-boilers. They all have the larger part of their heating-surface made up of tubes of moder- ate size filled with water. They all have some form of separa- tors, drum, or reservoir in which the steam is separated from 22 S TEA M-B OILERS. the water; some of these boilers have a shell of consider- able size, thus securing a store of hot water and a good free- water surface for disengagement of steam. Such shell, drum, or reservoir is either kept away from the fire or is reached only by gases that have already passed over the surface of water-tubes. The tubes are of moderate or small diameter, and so can be abundantly strong even when made of thin metal. Even if a tube fails through defect in manufacture or through wast- ing during service, it will not cause a true explosion ; and yet the failure of a tube in a confined boiler or fire-room has fre- quently caused death b\ scalding. Water-tube boilers may be made light, powerful, and compact, and are well adapted for use with forced draught. Steam may be raised rapidly from cold water, but pressure falls as rapidly if the fire loses intensity, and fluctuations in pressure are likely to occur. The two greatest difficulties are to secure a proper circulation of water through the tubes and to properly separate the steam from the water. There are many joints that may give trouble by leaking, and some types have numerous hand-holes for cleaning the tubes, which may further increase the chances of petty leaks. A few water-tube boilers will be described as illustrations ; many others equally good will be passed by, since it will be impossible to describe all. Babcock and Wilcox Boiler. — This boiler, which is shown by Figs. 16 and 17, is a water-tube boiler having a cylindrical shell to furnish steam-space, and in which is the free-water surface for the disengagement of steam. The tubes are expanded into vertical headers at each end ; the front-end headers open into a cross-connection in communica- tion with the cylindrical shell, while the back-end headers are connected with a similar cross-connection by slightly inclined pipes. The tubes in each section are staggered so that the tubes taken as a whole are in horizontal rows, but not in ver- TYPES OF BOILERS. 23 tical rows— an arrangement that gives a better spreading of the products of combustion among the tubes. At each end of each tube are hand-holes that give access to the inside of 6 ii m € p i u, c=L ii«i* _^> ^ mm - 111 *< ■11111111 the tube when it needs cleaning or scaling. By the aid of a brick bridge-wall at the end of the furnace and a continuation of this wall formed of special tiles through the tubes, tog-ther 24 STEAM-BOILERS. TYPES OF BOILERS. 2$ with a hanging bridge-wall similarly continued through the tubes, the products of combustion pass over the tubes three times on the way to the uptake at the back end of the boiler. The lower half of the cylindrical shell serves as heating-sur- face, but it is at such a height above the fire and is so shielded by the water-tubes that it is not liable to be over- heated. The boiler is hung from cross-girders front and back, which in turn are supported on iron columns, and the brick setting is only a screen to retain the heat. The circulation of the water in the boiler is down from the shell at the rear to the water-tubes, forward and upward through the tubes, in which course it is partially vaporized and consequently has a less average density, then up into the shell, at the front where the steam and water separate ; the water in the shell flows continually from the front to the rear to supply the current through the tubes. The Heine Boiler, shown by Fig. 42, page 106, resembles the Babcock and Wilcox boiler in general arrangement, but differs in that the tubes are expanded into one large header at each end^made of plate, properly stayed and provided with hand-holes. Again, the gases from the fire are con- strained to pass along the tubes instead of across them, for which purpose there are floors or nearly horizontal bridges of tiles, laid on two or three layers of tubes, instead of the nearly vertical bridges of tiles used in the Babcock and Wilcox boiler. The Root Boiler. — The general appearance of the Root boiler is shown by Fig. 18, and details of construction are shown by Fig. 19. Pairs of tubes are first expanded into headers at the end, as shown by 1, Fig. 19; then several pairs are assembled, as shown by 2, to form a vertical section, by the aid of bends, of which 3 gives further details. The joints between the bends and headers are made tight by aid of a metallic packing-ring shown by 4. The conical bearing on the bend shown by 5 expands the ring into a recess in the header, shown by 6, thus making a steam-tight joint. Each 26 STEAM-BOILERS. Fig. i 8. Fig. ig. TYPES OF BOILERS. 2J section has a steam-drum at the top, as shown in Fig. 18, and at the back end of the steam-drum are pipes leading up into the transverse steam-drum, and downward into a trans- verse water-pipe at mid-height of the boiler. Near each end of the mid-height water-pipe are vertical pipes communicating with the ends of a transverse mud drum, from which a scries of pipes lead to the sections of the boiler. The water circu- lation is down from the back ends to the mud-drums, then forward through the tubes to the front ends of the steam- drums, in which the steam and water separate, the steam passing into the transverse steam-drum, and the water re- turning through the back connections to the sections. The products of combustion pass over the tubes three times before escaping to the chimney. The Stirling Boiler. — This boiler, shown by Fig. 20, has three cylindrical drums at the top and a larger drum at the bottom, connected by tubes having a slight curvature at the ends. The two forward drums at the top have also a connec- tion below the water-line through pipes not indicated. All three upper drums have their steam-spaces connected by piping. The water-line is indicated by a dotted line. The feed-water is introduced into the rear upper drum, from which it passes down through the rear system of pipes, which act mainly as a feed-water heater, and enter the lower drum, where the water deposits any lime compound that it may contain, from whence it may be blown out at intervals. Fire-brick bridges cause the products of combustion to pass in succession through the three systems of water-tubes as shown by the arrows. The Cahall Boiler is a vertical water-tube boiler, shown by Fig. 21. It has an annular drum at the top and a cylin- drical drum at the bottom, connected by tubes and also by two large circulating pipes outside of the brick setting, one of which is drawn in the figure. The fire is in a brick furnace at one side of the boiler, from which the products of 28 STEAM-BOILERS. combustion pass back and forth across the tubes to and from the central space between the tubes. For this purpose there are two iron baffle-plates in the central space, as indicated in the figure. The water-line is carried ac about one third the height of the upper drum, and steam is drawn from a nozzle at the top. Fig. 20. The circulation is down the large exterior pipes to the lower drum, and then up the water-tubes to the upper drum. Man- holes give access to both drums, and in addition there are eight hand-holes in the top of the upper drum, so that any tube may be cut out and replaced without disturbing the others. TYPES OF B01LKKS. 2 9 Sir 1 I P til ! 30 STEAM-BOILERS. Water-tube Marine Boilers. — With the advent of very high steam-pressures on steamships there has been a tendency to replace the Scotch boiler by some form of water-tube boiler. A large number of French merchant steamers and a few French naval vessels have been fitted with Belleville boilers, a type of water-tube boilers that had already found favor for stationary purposes. This type of boiler has also been used to some extent on the Great Lakes. Recently this boiler has been largely introduced in the English Navy. Other water-tube boilers, either designed specially for marine boilers or modified from land boilers, have been used to some extent. In the United States Navy some vessels have been fitted with both shell-boilers and water-tube boilers ; the former are intended for use in ordinary service, and the latter when running at high speed. The objects that are sought in water-tube boilers for steamships are a larger power for the weight and the ability to carry high pressures. It still remains a question whether the water-tube boiler will or can replace the Scotch boiler for ordinary service on steamships. Indeed, it is a question whether there is any real profit in carrying steam at very high pressure. The Belleville Boiler is represented by Fig. 22 ; it con- sists essentially of a series of coils of pipe made up with bends and elbows around which the products of combustion pass on the way to the chimney. At the top there is a steam-drum A, connected by two circulating-pipes B and C, with a drum D at the bottom. From the mud-drum D a rectangular feed- supply runs across the front of the boiler to all the coifs or elements of the boiler. Each element is continuous from the feed-supply to the steam-drum, and is made up of slightly inclined pieces of pipe with horizontal bends or connections at the end. The effect is much as though a helical coil were flattened into two vertical tiers of pipes. The amount of water in the boiler is so small that it cannot be run without TYPES OF BOJLEKS. 3' ^^^553 K . i - i ) ~ \ Tl M Bl^ 32 STEAM-BOILERS. an automatic feed-water regulator, which in turn requires the attention of an expert feed-water tender. The several ele- ments deliver a mixture of water and steam to the steam- drum, which does not appear to act efficiently as a separator, as an external separator is placed between the boiler and the engine. The feed-water is supplied to the steam-drum and passes through the external circulating-pipes to the mud- drum, where it deposits much of its impurities. Fig. 23. Thornycroft Boiler. — The boiler represented by Figs. 23 and 24 was built for the torpedo-boat destroyer, " Daring," by Mr. Thornycroft ; boilers of slightly different forms have been fitted by him, in torpedo-boats and steam-launches. The boiler consists essentially of a large drum or sepa- TYPES OF BOILERS. 33 rat or at the top and three drums at the bottom, connected by a large number of bent-tubes. There is, inside of the casing, a large tube connecting the top drum to the middle drum at the bottom, and this drum is connected to the side drums by smaller pipes. The circulation is down from the top drum to the middle lower drum, and from that to the side drums, then up through all the bent water-tubes to the upper drum, where mingled water and steam is delivered Fig. 24. against a baffle-plate above the water-line. Steam is drawn from a nozzle at the rear end of the top drum. The arrangement of grates and fire-doors is shown in elevation and section by Fig. 23. The middle drum divides the grate into two parts; over that drum is a space which is 34 STEAM-BOILERS. in communication with the uptake, as shown by Fig. 24. The products of combustion pass among the tubes lead- ing from the middle drum ; the tubes to the outer drums intercept the radiant heat which would otherwise strike on the boiler-casing. The boiler-setting is an iron frame, and the casing is thin plate iron lined with incombustible non-conducting material. There are numerous doors through the casing for cleaning the tubes. This boiler has proved very successful with a forced draught, making steam freely and giving little trouble. The boiler contains so small an amount of water that steam may be raised quickly, and any demand for steam can be quickly met. On the other hand, the feed-supply must be regulated with care and skill, and the pressure is liable to fluctuate. The Yarrow Boiler. — The form of boiler used by Mr. Yarrow for torpedo-boats, is shown by Fig. 25. It resem- bles in general arrangement a form used by Mr. Thorny- croft with one grate. It, however, differs radically in certain particulars, namely, in that the tubes are straight and that they enter the upper drum below the water-line, and in that there are no pipes outside the casing to carry water from the upper drum to the lower drum or reservoirs. Some of the tubes.- deliver water and steam to the upper drum, from which steam is drawn ; other tubes carry water from the upper drum to the lower drums. A given tube may act sometimes in one way and sometimes in the other. Naturally those tubes which receive the most heat and make the most steam deliver to the upper drum, and tubes that receive less heat carry down water. The air for the fire is drawn from an iron box or casing outside the boiler-casing, so that the heat escaping from the boiler-casing is largely carried back to the fire, and the fire- room, and also the rest of the vessel, is heated up less. The Almy Boiler. — This boiler, which is represented by TYPES OF BOILERS. 35 Fig. 26, is made of short lengths of pipe screwed into return- bends and into twin unions. At the bottom is a large tube or pipe forming three sides of a square at the sides and back YARROVy BOILER Fig. 25. of the grate. From this water-space the tubes lead into a similar structure at the top. The steam and water are dis- charged into a separator in front of the boiler, from which 36 S TEA M-B OILERS. steam is drawn; while the water separated therefrom, together with the feed-water, passes down through circulating-pipes to the bottom of the boiler. The boiler is provided with a coil feed-water heater Fig. 26. above the main boiler. It is enclosed by a casing lined with non-conducting material. It is intended for general marine work. CHAPTER II. FUELS AND COMBUSTION. THE fuels used for making steam are coal, coke, wood, charcoal, peat, mineral oil, and natural and artificial gas. Various waste and refuse products, such as straw, sawdust, and bagasse, are burned to make steam. All coals appear to be derived from vegetable origin, and they owe their differences to the varying conditions under which they were formed or to the geological changes which they have undergone. Anthracite Coal consists almost entirely of carbon and inorganic matters; it contains little if any hydrocarbon. Some varieties, for example certain coals found in Rhode Island, appear to approach graphite in their characteristics, and are burned with difficulty unless mixed with other coals. Good anthracite is hard, compact, and lustrous, and gives a vitreous fracture when broken. It burns with very little flame unless it is moist, and gives a very intense fire, free from smoke. Even when carefully used, it is liable to break ud under the influence of the high temperature of the furnace when freshly fired, and the fine pieces may be lost with the ash. Semi-anthracite or Semi-bituminous Coal is intermedi- ate in its properties between anthracite coal and bituminous coal; it contains some' hydrocarbon, is less dense than anthra- cite, it breaks with a lamellar fracture, and it burns readily with a short flame. 37 38 STEAM-BOILERS. Bituminous Coals contain a large and varying per cent of hydrocarbons or bituminous matter. Their physical prop- erties and behavior when burning, vary widely and with all intermediate gradations represented, so that classification is difficult. Three kinds may, however, be distinguished, as follows : Dry bituminous coals, which burn freely and with little smoke and without caking. Caking bituminous coals, which swell up, become pasty, and cake together in burning. They are advantageously used for gas-making, Long -flaming bituminous coals, which have a strong ten- dency to produce smoke ; some do and some do not cake while burning. Coke is made from bituminous and semi-bituminous coal by driving off the hydrocarbons by heat. Coke made as a by-product in gas retorts, is weak and friable, and has little value for making steam. Coke made in coking ovens, by partial combustion of the coal which is coked, is of a dark- gray color, porous, hard, and brittle. It has a metallic lustre, and gives out a slight ringing sound when struck. Sulphur in the coal may be burned out in coking, if the coal is moist or if steam is supplied during coking, so that coke may be comparatively free from this noxious element even when made from a poor coal. Coke burns without flame and makes a fierce fire when forced. Lignite, or brown coal, is of more recent geological formation than coal, and is in a manner intermediate between coal and peat. It frequently contains much moisture and mineral matter. It is used where good coal is difficult to get, and while the better varieties form a useful fuel, the poorer qualities have little value. Peat, or turf, is obtained from bogs. It consists of slightly decayed roots of the swamp vegetation mingled with more or less earthy matter. For domestic use it is cut and FUELS AND COMBUSTION. 39 dried in the air. It is little used for making steam, though when pulverized, dried, and compressed it makes a useful artificial fuel. Wood is used for making steam either in remote places where coal is hard to get and timber is plenty, or where saw- dust or other refuse wood is produced in quantity in manufac- turing operations. Wood is also used for kindling coal-fires. One cord of hard wood is equivalent to one ton of anthracite coal ; one cord of yellow-pine is equal to half a ton of coal ; other soft woods are, as a rule, of less value for fuel. Charcoal is made by charring wood ; it is but little used for making steam. Mineral Oil, in the form of crude petroleum or the refuse heavy oil left from the distillation of petroleum, is used for making steam, especially in the neighborhood of the Black Sea oil-field, and by steamers carrying oil from those fields. It is customary to throw the oil into the furnace in the form of finely divided spray through special spraying apparatus worked either with compressed air or with superheated steam. The use of superheated steam has its convenience only to recommend it, for it adds to the inert material to be uselessly heated. Special precautions must be taken, when petroleum is burned, to avoid flooding the furnace with oil and to pre- vent explosions of the vapor and burning of the oil in tanks or receptacles. Gases. — Natural gas from gas-wells has been used for making steam, usually in a crude and wasteful way. Some attempts have been made to use gas made from poor and smoky coal, in producer-furnaces like those used in metallurgi- cal operations ; but the gain to be expected is only the sup- pression of the smoke nuisance, which is rather a social than an economical problem. Artificial Fuels. — The small waste from coals and char- coals, sawdust, and other fine combustible material which cannot be sold in such shape, is sometimes made into cakes or 40 S TEA M-BOILERS. briquettes by mixing it with some adhesive material and then compressing it. The adhesive materials have been wood-tar, coal-tar, or else clay. Tar is available in limited quantities only, and clay is disadvantageous since it adds to the inert material, of which fine fuel is liable to have an excess. Artificial fuels have some advantages for special purposes, and can be stored compactly ; they are used mostly where good fuel is difficult to get. Composition of Fuels. — The composition of a number of American coals, together with the total heat of combustion by Dulong's formula, is given in the table on page 41, which has been kindly furnished by Mr. Henry J. Williams. These results are a part of a very extended investigation by Mr. Williams, to be published in full in the near future. Most of the results are the averages of several separate analyses, and all may be depended upon to give a fair representation of the coals named. Analyses by Mahler of various European coals and of a few American coals, together with the total heat of combus- tion, are given in the table on page 42. The following table gives the composition of several rep- resentative petroleums : COMPOSITION OF PETROLEUMS. Pennsylvania, crude Caucasian, light. . . . heavy . . Petroleum refuse. . . Carbon. Hydrogen. Oxygen. 84.9 13-7 1.4 S6.3 13 6 O.I 86.6 12. 3 I. I 87.1 11.7 I .2 Specific Gravity. 0.8S6 0.884 0.938 0.938 Heat of Combustion. — The number of thermal units de- veloped by the complete combustion of one unit of weight of a fuel is called the heat of combustion. It can be determined by burning the fuel in a properly constructed calorimeter. The most recent and best results are those obtained by the use of Mahler's bomb-calorimeter. This is a strong recep- FUELS AND COMBUSTION. 41 CO H u ed w ML| CO <3 O u < U '2 w s . "^ w < % < ft a 2 ^ H >. CO e5 D z O >> U m to O H < W •DC Q < O V ir ice co 1- "-" f 1! SuioioXtvy r* C^o 04 I -- ? in O r^O en -1- 1 j O O -t O 000 r^ Q\ in O O C> O •SUIUIH31S OO H "1 CJ m p-i -T - -T - O M r^ q^'anqsiiid m 04 r-» m m O in ^ 00 i^ >-• CO CO U"> r^ too irnO 00 ? -1- c>co 1^ jun m uojajg 3dE3 Tf ~ C4 M 000 00 O M in o» O m 1^ U5 'uoiuioioq 04 O C4 ""> i^I-hod m CN co 3 O c a 3 cnO r^ " <~\ M C7> M co co « intci HO - i 'III t^M Nt^ O <7^ — co co O inco a- • O (5 'aniAJSIiBQ ID O "~>CO m 6 6 H tN MtO O in 04 'iippniv 3ig co u-> r-N M Tt OO M r^ - 1^ M -(-0 OCO M c 1 •SUIJJ03 CfnOO O O m co cn in r^O O | ' 3lItASI[3UU03 d i-» -t r» 1-66 r^ in •-. t>» O a r^ " r^ O -t r^ 04 m r^ O co « co m r^ *«A AY r-» r^o co r-« t^ in to co r^ O -<* J3AI^£ M3{^ d - n -r n-tH -r -j- d x^ CO -1-vO 10 in — 1-1 i-nco C^ >- co 3 O C M M O "TO O O -T •BJ *pi3l|JB30 O 00 m r^ r* O O in hh rj- r^ w r-» co a 3 15 O 04 11 CO M C> 04 in — m r> n c Tt- .i •>{33J3 in in co to cr> "- O O « mo r^- co a s t aSjo3Q O CO TO odd " T« -fO O 1- V) hh r-» CO -t O co co Oco — O CI CO T M CO ir>. u-i r-» ci -r r»o «-i 1-co ci cj ino O •SBiuoquDoj O 1^ in O O C co -r - -r in O CO m r^ CO r^ O >-• co vD 1/1M C O coo O ""> MflO N O fl" M co 10 "1 C 1 ir> -t •-" r- •Bd 'uoijua 1-1 CO O IT) r^ c< i-c cj O -i- co co 1-1 w u .tj O -t- r^ c -t- O s in in m .— a! "a" 3 II b A suajjAq IDCO O IT> IO CO 1-1 04 o< C^ it t~* CO Jjj i-h r- 1- O O O CO CO 04 to co CO C «! r^ in O O "Too O co r^ r^ co O cs •qSupT O co -r <- O in O O r^ r^co co in CO w -to r^ in 04 O 04 r^ CO CO l ~ l "53 t/3 : 3 II *c3 f § « 8 *•§ • 3 -C £ .C en 3 a as • a, SI ^ c co It c rt x: | « c V c U 1 & ■a E X9 c e 1 > '. 3 . m '. Ji V3 O <> £ CO 5 EC 42 STEAM-BOILERS. D - O ^ En c J3MO(j oyuo|B3 O W *- O in in r^ too ►" -j- t-i in co M co co co oo oo r^co O CM O -+ - OOOCC N »- T l-H i^o o co r»o O co co co co co co O O O I— 1^ O CM O O r^ 1- o f>Oco -TOO | incih" C* N O I O >-< O Tf T m I r^co CO qsy o o a o co O T -1- r-* co co co o o co O co M -TO m <-i co in O in co r^ in OOO T r^ cm O O O T CO Suipnpxa M M mt N vO CO inco O trN o o OO N (? OOO CO CO CO CO 'j3MOJDtJ[J01E3 paAJ9sqo JSA^OJ DIJUOIB3 •3- co co o co co O cm m r-. T Too co r^ r~ t^ t^ r^ r» r^ O co T a t T -r m in r^co O CO jo 9Aisnp co CM in coo T in co ct> o in 0> T •qsy 885,88 o o o o o o o o r^ o o O OOO OOO h -r o O O in in co O O co m cm co t^ O OOO OOO CM CM O in T in ino m tuiT t co TO OOO T co in OldODS -ojpAH O O O m O in o O r^O T r»co r~» r» n4fi h h O m O O in w O in coo w cm ooo ooo cm o o l-H co O* O T ►H T O* co O O m -1- CM cm o o' o 883 in in cm o* 6 6 - U9Soj; -IN P u e uaSAxQ O s i- 1 inco OvO co O r->. «- C* r* O cm tO O - N co TO to co 0 • in co XT) * N * ' oo • • o • : : '111 pJoj^potf O oo • r* - • . . O O • o • O • • - o . o o o C4 • m in • * in M • • o o> * • o c5> . . d r*- ■ • cd o co '. cn '. -i- s co ' * W O * ' " r> o o ; . ^ .* ! ! .' i ! ; ; i m . i ; ; d r« t • . iA • ■ d - • • hi >-i CJ T r*o co o co co r* " o T ' * " 6 in d • t ••"••••••• d •• • Oco r-« o^OOOOJ « m or* ' *r> ' * * N * O " coco co " * Q u> H.vO ' I CO ' " ]hi •— ' T TO " ] co '■. l • Hi t • • d ■ • • c4 • d ■ d d ci • • *h n cm r» hi o m co t) t m ' O O * " * * "co * inco' n, | co m • • • • * • • • • p» • • * H oo « • ni co O • • co O T O co co O r*co * Oco •'co** cocot* ° ^ ! ! ! I ! I I I I N 1 I ". 6 *!!* d t d • « • • • • • d • d ■ d '• ' ■ •pug A\i3A 'ipAV "III *3AOJ£) S t J3UA\OQ 'VflJA "III '3P!S43AI>I •3in5103>I IB 49AI^ ld equations which differ fundamentally from the equation for the flow of liquids. Only when the changes of temperature and pressure are small, can the hydraulic equation be used at all, and then the approximation is not good Nevertiieless it may be possible to base a working theory of chimneys on the hy- draulic equation, provided that the proper constants can be derived from experiments, and provided that the application of the theory be restricted within limits that are determined by proper investigation. Peclet's Theory of Chimney-draught— The theory of chimney-draught which is commonly given in text-books was. proposed by Peclet many years ago. This theory assumes, first, that the flow of gases through the furnace-flues and chimney, may be represented by an equation like the hydraulic equation just quoted; and second, that the head h in that equation SETTINGS, FURNACES, AND CHIMNEYS. 117 may be calculated by dividing the draught, expressed in pounds per square foot, by the weight of a cubic foot of hot gas. The weight of a cubic foot of hot gas is given by the T expression w~ t and the draught is given by the expression I c (T T\ wH\ — °— -=£) ; so that Peclet's assumption gives .:h = H&-i) (1) Replacing the coefficients k and k x in the hydraulic equation by C and G, which represent the resistance to the flow of gas through the tubes and flues, and the resistance to the flow through the grate, we have or, solving for V the velocity of the hot gas through the chimney, v= v^iii 7 ! _ x Y ! ... (3) i + G + C+l^) in If the area of cross-section of the chimney is A square feet, the volume of hot gas discharged per second is VA, and the weight discharged per second is W^VA.zv^ T< sj~* / '7™' f T y \ 1 ,'.W=Aw V^fffK '~T 1 1 " 7WV" (4) 2 1 8 S TEA M-B OILERS. From some experiments on chimneys and boilers Peclet gives, in connection with this theory, the following values for the coefficients, C7 = 12, /= O.OI2, under the assumption that from 20 to 24 pounds of coal are burned per square foot of grate per hour; the coefficient C does not appear in his equation. Equation (4) is in the proper form for calculating the weight of gas discharged by a given chimney, for which the height, area, and perimeter of cross-section are known. If the weight of gas to be discharged and the area and perimeter are known, the equation for a given case leads to a quad- ratic equation for finding the height H, which can readily be solved numerically. If the weight of gases is known, and the height of the chimney is assumed, then the insertion of linear dimensions in place of A and m leads to an equation of the fourth degree; but as — - is small compared with I -f- G, a solution by approximations can be readily made. It is probable that the equation (4) with the given values for G and f represented satisfactorily the performance of the chimneys which were investigated by Peclet. These chimneys provided draught for boilers then in common use in France, which boilers were probably either plain cylindrical boilers or " double-elephant" boilers. For such boilers the resistance is mainly at the grate. On the other hand, the resistance to the passage of gases through the tubes of cylindrical tubular boilers, locomotive-boilers, and marine boilers is about equal to the resistance to the passage of air through the fuel on the grate. Under the conditions of modern practice in America, the equation deduced by Peclet, using his values for /and C7, gives- results that do not accord with observations or with commoa SETTINGS, FURNACES, AND CHIMNEYS. 1 1 9 practice. The theory consequently is not valuable for pro- portioning chimneys. If the weight of gas discharged by a chimney of given height and cross-section be calculated successively for differ- ent values of T c , the temperature in the chimney, the weight will be found to increase with the temperature, until the tem- perature becomes about 1000 absolute or about 6oo° F. ; be- yond this temperature the weight decreases. The temperature for maximum discharge, as calculated by the equation, may be readily found by aid of the differential calculus. The factor which increases with the temperature is T c ' Differentiating with regard to T c and equating the first differential coefficient to zero gives TA(T c -T a )-}-(T e - r a )* = o. .'. T c = 2 T a . Consequently the maximum discharge of gas will occur when the absolute temperature in the chimney is twice the absolute temperature of the air. If the temperature of the air is yo° F., or ;o + 460.7 = 530.7 degrees absolute, then the temperature to produce the maxi- mum discharge of gas will be, by Peclet's theory. 2 x 530°-7 = io6i°.4 absolute, or io6i°.4 — 46o°.7 — 6oo°-f F. This is about the temperature of melting lead, and books on chimneys frequently say that the temperature in a chimney should not exceed that of melting lead. A tem- perature near this is commonly found in chimneys that are doing good work, a fact that seems to give some support to 1 20 S TEA M-BOILERS. the theory. But the support is entirely fictitious, for the oc- currence of a maximum depends on the assumption that the head h in the hydraulic equation may be replaced by IT \ H\ — — I j j an assumption that can be justified only by ob- servation or experiment. Such observations or experiments are lacking. All we know is that calculations by the equation do not accord with common practice. It is true that 6oo° F. should be a sufficiently high temperature in the chimney to give all the draught required. It will be still better if a lower temperature will suffice. Peclet's Second Theory. — It appears that Peclet was not satisfied with his theory as first propounded, for he after- wards advanced another theory, in which the head is calcu- lated by dividing the draught, expressed in pounds per square foot, by the weight of a cubic foot of cold air. It is noteworthy that this later theory does not show a maximum discharge at 6oo° F. Neither the first nor the second theory is strictly logical ; the value of either as a working theory must consequently de- pend on its adaptability for designing chimneys under condi- tions of ordinary practice. Both theories lack connection, through experiment or observation, with practice, and can- not now be used to advantage. Tests and Observations. — The data from the tests made by Peclet to determine the values of constants in his equation are not now accessible. He gives only the results, namely, G = 12 and/ = 0.012. Prof. Gale' 55 ' reports the following results of tests made on a chimney and boiler of ordinary construction : Area of grate 22.5 sq. ft. Area through tubes 2.74 " Coal per square foot of grate per hour 13.5 pounds. Air per pound of fuel 21 " * Trans. Am. Soc. Mech. Engrs., vol. xi. p. 451. SETTINGS, FURNACES AND CHIMNEYS. 12 Temperature boiler-room 60 ° F Temperature external air }0° F. Height of chimney above grate 72 feet. Area of chimney (round iron stack) 4 sq. ft. Length of horizontal iron flue 24 feet. PRESSURES IN POUNDS PER SQUARE FOOT. Required to produce entrance velocity 0.013 Required to overcome resistance of grate 0.91 Required to overcome resistance of combustion-chamber and boiler-tubes 1.23 Required to overcome resistance of horizontal flue 0.06 Required to produce velocity of discharge , 0.085 Total effective draught 2. 298 Required to overcome resistance of friction o. 19 Total draught 2.4S8 On these results Prof. Gale based a set of constants, to be used in an equation like that given in Peclet's first theory. It does not appear to us that observations on one chimney are sufficient for this purpose. We will note only that his value for the coefficient of friction is /= 0.012 for an unlined iron stack. For a brick chimney he gives f = 0.016. The following table gives the results of a test made at the Massachusetts Institute of Technology on an unlined steel chimney 3 feet in diameter and 100 feet high above the grate. Over the grate At the bridge-wall Half-way between bridge and back end of boiler At back end of boiler In uptake near boiler In stack 34 feet above grate 1 n stack 5 1 feet above grate In stack 68 feet above grate In stack 85 feet above grate Draught. Inches of Water. Tempc Centi Max. Min. Max. 0.240 0.2I8 O.382 O.372 O.410 0.374 0-354 0-334 O.572 0.543 206 0.440 O.414 202 0.334 O.216 O.312 O.168 193 188 O.I22 O.OS6 174 Min. 198 190 187 179 157 122 S TEA M-B OILERS. This chimney now serves two boilers similar to that shown on Plate I, each of which is rated at 80 horse-power. It is in- tended to be sufficient for four such boilers. The heating- surface of each boiler is 11 13, and each has 25.9 square feet of grate area. At the time of the test one boiler had the fire banked, and the combustion at the grate of the working boiler was at the rate of 19.8 pounds per square foot of grate-surface per hour. Kent's Table.— Mr. Wm. Kent* has calculated a table of sizes of chimneys on the following assumptions: 1. The draught-power varies as the square root of the height. 2. Allowance for friction against the sides of the chimney may be made by subtracting from the actual area of the sec- tion of the chimney, a strip two inches wide and as long as the perimeter "of the section. 3. The power of the chimney is directly proportional to the area remaining after the strip is deducted from the actual area. The first assumption is equivalent to using the hydraulic equation on page 1 1 5 in the simplified form ^=— or V=V~2jH, in which H is the height of the chimney in feet and V is the velocity of discharge. The second assumption is purely arbitrary, and can be used only with.n limits, or it may lead to absurd results. Thus a flue 4 inches in diameter would give no draught at all by this assumption. The third assumption follows naturally from the second. If the side of a square chimney be represented by D } then the area is A =D\ * Trans. Am. Soc. Mech. Engs., vol. xi. p. 81. SETTINGS, FURNACES, AND CHIMNEYS. \?l and the strip to bo subtracted is (nearly) 4D x tV = i D = 0.6 V7T If the effective area allowing for the strip is E, then for square chimneys E =A -o.6VA~. The same equation may be used for round chimneys with but little error. Combining the first and third assumptions, Mr. Kent writes the equation H.P. = CE 1/77, which he uses for calculating the commercial horse-power of the chimney, or more properly of the boilers to be connected with the chimney. To obtain a value for the arbitrary constant C he chooses a typical chimney, 80 feet high and 42 inches in diameter, for which the effective area calculated by his method is 9.62 square feet. He says that such a chimney should be capable of carrying a combustion of 120 pounds of coal per hour for each square foot of effective area. If the area of the chimney is one eighth of the area of the grates connected with it, then this is equivalent to a combustion of 120 -^-8 = 15 pounds of coal per square foot per hour — a very common performance. This typical chimney should then burn 9.62 X 120 = 1 154.4 pounds of coal per hour. If it be assumed that nve pounds of coal will be burned per horse-power per hour, tne chimney may be considered to correspond to 1154.4-5 =231 H.P. Substituting in the formula for horse-power 231 = C X 9.62^80. ."• ^=3-33 124 S TEA M-B OILERS. And the horse-power equation, substituting for E its value, may be written H.P. = 3-3304 - 0.6 VA) V7T. The following table has been calculated by the equation just written : SIZES OF CHIMNEYS WITH APPROPRIATE HORSE-POWER OF BOILERS. Height of Chimneys. V 4) <*- v C U ,e Q.S <<3 Comi nercial Horse-powei cr in a 18 23 25 27 O.97 i-77 16 21 35 38 4i I 47 2.41 19 24 49 54 58 62 2 08 3-i4 22 27 65 72 78 83 2 78 3-98 24 30 84 92 100 107 "3 3 58 4.91 27 33 US 125 133 141 4 47 5-94 30 3* 141 152 163 173 182 5 47 7.07 32 39 183 196 208 219 6 57 8.30 35 42 216 231 245 258 271 7 76 9.62 38 48 3" 330 348 365 389 10 44 12.57 43 54 363 427 449 472 503 551 13 5i 15.90 48 60 505 539 565 593 632 692 748 16 98 19.64 54 66 658 694 728 776 849 918 981 20 83 23.76 59 72 792 835 876 934 1023 1 105 1181 25 08 28.27 64 78 995 1038 1 107 1212 1310 1400 29 73 33-18 70 84 1163 1214 1294 1418 T 53i 1637 34 76 38.48 75 90 1344 1415 1496 1639 1770 1893 40 19 44.18 80 96 1537 1616 1720 1876 2027 2167 46 01 50.27 86 The number of pounds of coal per hour that can be burned with a given chimney can be found by multiplying the horse- power given in the table by five. Only that part of the table is filled in which corresponds to ordinary proportions, depending on the judgment of the author of the method. In general it will be better to select proportions from the table for a chimney to be used with a given commercial boiler horse-power, rather than to calculate by the formula, as extraordinary proportions will then be avoided. This table hai been used to a considerable extent, and apparently with satisfactory results. Areas of Chimneys and Flues. — In common practice it is found that satisfactory results are obtained if the area of the SETTINGS, FURNACES, AND CHIMNEYS. 12 section of a chimney is made one eighth of the area of all the grates connected to the chimney. The ratio is sometimes as large as 1/7 and sometimes as small as 1/9, or for tall chim- neys 1/10. Height of Chimney. — Professor Trowbridge* gives the following table of heights of chimney required to give certain rates of combustion, obtained by collecting reliable data and drawing a curve to represent mean results: HEIGHT OF CHIMNEY. Heights in Feet. Pounds of Coal per Square Foot of Section of Chimney per Hour. Pounds of Coal per Square Foot of Grate per Hour. 20 60 7-5 25 68 8-5 SO 76 9-5 35 84 10.5 40 93 11.6 50 105 13.1 60 116 14.5 70 126 15.8 80 135 16.9 90 M4 18.0 100 1^2 19.0 no 160 20.0 The table was made several years ago, but it seems to be conservative and to represent good average practice. By its aid the height of a chimney to give a desired rate of combus- tion can be determined. This height is then to be used with the ordinary ratio of chimney-area to grate-area as just given. Forms of Chimneys.— Chimneys are made of brick or of steel plates. Steel chimneys are always round ; large brick chimneys are usually round ; small ones may be round or square. A round chimney gives a larger draught-area for the same weight of material, and it presents less resistance to the wind. Plate V gives the general arrangement and some detail of two chimneys: one of brick, 175 feet high, and the other of * Heat and Heat-engines, p. 153. 126 STEAM-BOILERS. steel, 200 feet high. The brick chimney is built in two parts : the outer shell, which resists the pressure of the wind ; and the lining, which forms the flue proper, and which may expand when the chimney is full of hot gases without bringing any stress on the shell. The shell has a foundation of rough stone and one course of dressed stone at the surface of the ground. The brickwork is splayed out inside to cover the stone foun- dation, and is drawn in at the top to the same diameter as the inside of the lining. The external form of the top is mainly a matter of appearance. The finish of large tiles at the top sheds rain and keeps water from penetrating the brickwork. The outside of the shell has a straight taper from the base nearly up to the head. A system of internal buttresses, as shown in section at Fig. 3 and Fig. 4, gives the requisite stiff- ness to the shell without an excessive amount cf material. The lining carries its own weight only, being protected from the wind by the external shell; it has a uniform diameter of 6 feet inside, and varies in thickness from 12 inches at the bottom to 4 inches at the top. A rectangular flue with an arched top, leads into the chimney at one side of the founda- tion. The shell of the steel chimney is made of vertical half-inch plates at the base, and is splayed out to give additional bearing on the foundation. Above this portion the shell has a straight taper to the top ; the plates, each 4 feet wide, vary in thick- ness from 3/8 of an inch to 1/4 of an inch. At the top an external finish of light plate is given for the sake of appear- ance. The foundation is of red brick, with a course of stone at the surface of the ground, clamped by a wrought-iron strap. The shell is bolted through a foundation-ring made of cast-iron segments 4 inches thick, and a steel plate 2j inches thick, by long bolts which take hold of anchor-plates bedded in the foundation. The lining of fire-brick varies in thickness from 18 inches at the bottom to 4J inches at the top. It lies against and is carried by the steel shell. The internal diameter of the SETTINGS, FURNACES, AND CHIMNEYS. 1 27 chimney is intended to be 10 feet; at places the size is a little larger on account of the arrangement of the lining. The lining is used to check the escape of heat through the steel shell. It adds nothing to the strength of the chimnev ; on the con- ' trary, it must be carried by the shell There is a chance that moisture may be harbored between the lining ana the shell and give rise to corrosion. Large steel chimneys are compar- atively recent, so that experience does not show whether lined or unlined chimneys are the more durable. Stability of Chimneys.— On account of the concentration of weight on a small area, and the disastrous results that would follow from defective work, the founaanons of an important chimney should be carefully laid by an experienced engineer. A natural foundation is to be preferred, but piling and other artificial methods of preparing the earth for the foundation can be used when necessary. Good natural earth should carry from 2000 to 4000 pounds to the square foot. The base of the chimney should be spread out so that this pressure, or whatever the earth can safely bear, may not be exceeded. In calculating the stability of a chimney it is customary to assume the maximum pressure 01 the wind as 55 pounds per square foot on a flat surface. The pressure of the wind on a round chimney is assumed to be half that on a square chimney having the same width. This method has long been in use, and it has been shown to give abundant stability. Experiments on wind-pressure are difficult and uncertain, and* curiously, the pressure determined by small gauges is com- monly in excess of that shown by large gauges. Thus, cer- tain experiments made during the construction of the Forth Bridge, gave a maximum wind-pressure of 35 pounds per square foot on a large gauge 20 feet long and 15 feet wide, while a small gauge showed a pressure of 41 pounds at the same time. The highest recorded pressure during violent gales, at the Forth Bridge, was that just quoted, namely, 35 pounds to the square foot. Small wind-gauges have shown 128 S TEA M-B OILERS. a pressure of 80 to 100 pounds to the square foot; but such results are discredited, both because it is known that small gauges give too large results, and because buildings were not destroyed as they would have been if exposed to such wind- pressures. To determine whether a chimney is stable, treat it as a cantilever uniformly loaded with 55 pounds to the square foot and find the bending-moments and resultant stresses. The stress will be a tension at the windward side and a com- pression at the leeward side. Calculate the direct stress due to the weight of the chimney, which will be a compression at either side of the chimney. For a brick chimney, subtract the tension due to wind-pressure at the windward side from the compression due to weight : if there is a positive remainder showing a resultant compression the chimney will be stable ; otherwise not, because masonry cannot withstand tension. Again, add the compression due to wind-pressure to the com- pression due to weight, to find the total compression at the leeward side : if the result is not greater than the safe load on masonry, the chimney is strong enough. The safe load may be taken at 8 tons per square foot. A steel chimney may be calculated for compression only, since steel is at least as strong in tension as in compression. The compression should be limited to 10,000 or 12,000 pounds per square inch on the net effective section between rivets. The assumption that rivet-holes are completely filled by the rivets, and that the total compressive strength is not reduced by cutting the rivet- holes, is erroneous. The shearing-resistance of the rivets in the rinp-'Seam should be made equal to the compressive strength of the net section between rivets, in a manner anal- ogous to that used for determining the proportions of boiler- joints. A calculation like that just described must be made for the section of the chimney at the base, for each section where there is a change of thickness or of construction, and for any SETTINGS, FURNACES, AND CHIMNEYS. 1 29 other section where there is reason to suspect weakness or instability. The lining of a brick chimney is to be calculated for com- pression due to weight, at the base and at each section where there is a reduction of thickness. The lining of a steel chim- ney must be counted in when the stress due to weight is determined. A separate calculation must be made for the stability of the foundation of a steel chimney. For this purpose find the total wind-pressure on the chimney and its moment about an axis in the plane of the base of the foundation. Find also the total weight of the entire chimney with its lining, and of the foundation : this will be a vertical force acting through the middle of the foundation. Divide the moment of the wind-pressure by the weight of the chimney and foundation : the result will be the distance from the middle of the founda- tion to the resultant force due to the combined action of wind-pressure and weight. If this resultant force is inside the middle third of the width of the foundation, the chimney will be stable. This brief statement is intended to describe the method of calculating the stability of chimneys, and not to give fulli instructions. The design and calculation for an important chimney should be intrusted only to a competent engineer who has had experience in such work. CHAPTER V. POWER OF BOILERS. THE power of a boiler to make steam depends on the amount of heat generated in the furnace, and on the propor- tion of that heat which is transferred to the water in the boiler. The amount of heat generated depends on the size of the grate, the rate of combustion, and the quality of the coal burned. The transfer of heat to the water in the boiler depends on the amount and arrangement of the heating-sur- face. In practice it is found that each type of boiler has certain general proportions which give good results ; any marked variation from these proportions is likely to give poor economy in the use of coal, or to lead to excessive expense in construction. The capacity of a boiler is commonly stated in boiler horse-power; the economy of a boiler is given in the pounds of steam made per pound of coal. Neither method is entirely satisfactory, but definite meaning is attached to the terms by definitions and conventions. Standard Fuel. — A comparison of the composition and of the total heats of the several kinds of coal given in the table on page 41 shows a great difference in the value of a pound of coal, depending on the district and mine from which it comes. In order to introduce some system into the com- parison of the performance of boilers in different localities it has been proposed that some coal or coals be selected as standards, and that all boiler-tests intended for comparison be made with a standard coal. For this purpose it has been 130 POWER OF BOILERS. 131 proposed to select Lehigh Valley anthracite, Pocahontas semi-bituminous, and Pittsburg bituminous coal. More def- inite comparisons would result if only one coal, such as Poca- hontas, were selected. The objections are, first, that some trouble and expense might be incurred in localities where this coal is not regularly on the market ; and second, that a furnace designed for a given coal may not give its best results with a different kind of coal. There is a notable dif- ference between furnaces designed for anthracite coal and those designed for bituminous coal ; for the rest it appears that the use of a standard coal is a question merely of ex- pediency. In making a boiler-test it is not difficult to make an ap- proximate determination of the per cent of ash in the coal used. When that is done, the economy is usually stated in terms of water evaporated per pound of combustible, as well as per pound of coal. This gives somewhat more definite- ness to the statement ; but as no account is taken of the vola- tile matter in the coal, nor of the oxygen, this method also is indefinite. Value of Coal. — The actual value of a coal for making steam can be determined only by accurate tests with a fur- nace and boiler which are adapted to develop and use the heat that the coal can produce. While many boiler-tests have been made, and there is a good deal of material that could be used for the purpose, there has not yet been made a satisfactory statement of the value of the fuel in common use. It appears probable that the real value of a coal for mak- ing steam is proportional to the total heat of combustion. If this can be shown to be true, then coals should be sold on the basis of heat of combustion, just as steel is required to have certain physical properties which are determined by making proper tests. Quality of Steam. — When the economy of a boiler is stated in terms of water evaporated per pound of coal, it is \y 132 S TEA M-B OILERS. assumed that all the water is evaporated into dry saturated steam. But the steam which leaves the boiler may contain some water, or it may be superheated. The moisture carried along by steam is called priming. The steam from a properly designed boiler, working within its capacity, seldom carries more than three per cent of priming. Under favorable circumstances steam from a boiler will be nearly dry. If steam, after it passes away from the water in the boiler, passes over hot surfaces it will be superheated ; that is, raised to a temperature higher than that of saturated steam at the same pressure. Vertical boilers with tubes through the steam- space give superheated steam. If steam is to be superheated to any considerable extent, it must be passed through a superheater, which usually is in the form of a coil of pipe sub- jected to hot gases outside. Now a boiler filled with water will keep the plates and tubes which form the heating-surface somewhere near the temperature of the water; such heating- surface will endure service for a long time. But superheating- surface is likely to be at a temperature about half-way between, that of the steam inside and the gases outside, and is liable to be rapidly destroyed. For this reason superheated steam, though it gives a notable gain in economy when used in a. steam-engine, is not looked upon with favor. Steam-space. — The steam-space and the free surface for the disengagement of steam should be sufficient to provide for the efficient separation of the steam from the water. Cylin- drical tubular boilers frequently have the steam-space equal to one third of the volume of the boiler-shell. Marine return- tube boilers usually have a smaller ratio of steam-space to water-space. The more logical way appears to be to proportion the steam-space to the rate of steam-consumption by the engine. Thus the ratio of the volume of the steam-space of cylindri- cal boilers to that of the high-pressure cylinder of multiple- POWER OF BOILERS. 1 33 expansion engines varies from 50 : I to 140 : 1. The ratio of the steam-space of a simple locomotive-engine to the volume of the two cylinders is about 6£ : 1. The capacity of the steam-space is sometimes equal to the volume of steam consumed by the engine in 20 seconds. It was found in some experiments with marine boilers having a working-pressure less than 50 pounds per square inch, that a considerable quantity of water was carried away by the steam when the steam-space was equal to the volume of steam con- sumed in 12 seconds, but that no water was carried into the cylinders when the steam-space was equal to the volume of steam used in 1 5 seconds and that no trouble from water was ever experienced when the steam-space was proportioned for 20 seconds. All the preceding discussion refers to engines that run at a considerable speed of rotation — not less than 60 revolutions per minute. Engines that make but few revolutions per min- ute and take steam for only a portion of the stroke require a larger proportion of steam-space. As an example we may •cite the walking-beam engines for paddle-steamers. Equivalent Evaporation.— The heat required to evapo- rate a pound'of water depends on the temperature of the feed- water, the pressure of the steam, and the per cent of priming. For example, if water is supplied to a boiler at 140 F., and is evaporated under the pressure of 80 pounds by the gauge, with 2 per cent of priming, the heat required will be calculated as follows : The heat of the liquid at 140 F., or the heat required to raise a pound of water from 32 F. to that temperature, is 108.2 B. T. U. The heat of the liquid at 94.7 pounds abso- lute, corresponding to 80 pounds by the gauge, is 293.8 B. T. U. Consequently the heat required to raise the feed- water up to the temperature in the boiler is 293.8 - 108.2 = 185.6 B.T.U. The heat of vaporization, or the heat required to change 134 STEA M-B OILERS. a pound of water into steam, at 94.7 pounds absolute, is 886.9/ B. T. U. But 2 per cent of water is found in the steam which comes from the boiler, leaving 98 per cent of steam ; conse- quently the heat required is 0.98 X 886.9 = 869.2 B. T. U. The total amount of heat is therefore 185.6 -f 869.2 = 1054.8 B. T. U. Suppose that each pound of coal evaporates 9 pounds of water, then the heat per pound of coal tranferred to the boiler is 9 X 1054.8= 9493-2 B.T. U. Now the heat required to vaporize a pound of water at 2 12° F., under the pressure of the atmosphere, is 965.8 B. T. U. Dividing the thermal units per pound of coal by this quantity gives 9493.2^965.8 = 9.83, which is called the equivalent evaporation from and at 212 F. This method of stating the economy of a boiler is equiva- lent to using a special thermal unit 965.8 as large as the ther- mal unit defined on page 44. In making calculations involving quantities of wet steam it is convenient to consider the amount of steam present, rather than the percent of priming. In the example just con- sidered, there are 0.02 of water or priming, and 0.98 of steam. The part of a pound which is steam is represented by x. If the heat of vaporization at the pressure of the steam in the boiler is represented by r, the heat of the liquid at that pressure by q, and the heat of the liquid at the temperature of the feed-water by q ; and if, further, there are w pounds of POWER OF BOILERS. 135 water evaporated per pound of coal, — then the equivalent evaporation is w(xr -f q — q ) 965.8 The highest equivalent evaporation per pound of coal is about 12 pounds, and to accomplish this result about 80 per cent of the total heat of combustion must be transferred to the water in the boiler. The complete combustion of a pound of carbon develops 14,650 B. T. U. ; if all this heat could be applied to vaporizing water at 212 F., then the amount of water evap- orated would be 14,650 -i- 965.8 = 15-j- pounds. Few, if any, coals have a greater heat of combustion, con- sequently this figure may be considered to be the maximum equivalent evaporative power of coal. Should any test appear to give a larger evaporative power, or even a power approaching this result, it may be concluded either that there is an error in the test, or that there is a large amount of priming in the steam. Some tests of early forms of water-tube boilers without proper provisions for separating water from the steam, appeared to give extraordinary results; which results were due to the presence of a large amount of priming in the steam. At that time the methods used for determining the amount of priming were difficult and uncer- tain, and were frequently omitted in making boiler-tests. Boiler Horse-power — It has always been the habit to rate and sell boilers by the horse- power. The custom appears to be due to Watt, and at that time the horse-power of a boiler agreed very well with the power of the engine with which it was associated. The traditional method of rating boilers, coming down from that time, was to consider a cubic foot, or 62^- pounds, of water evaporated into steam, as equiva- 136 5 TEA M-B OILERS. lent to one boiler horse-power. This rating is now antiquated, and is seldom or never used. It is now customary to consider 30 pounds of water evap- orated per hour from a temperature of ioo° F., under the pressure of 70 pounds by the gauge, as equivalent to one horse-power. This standard was recommended by a com- mittee of the American Society of Mechanical Engineers.* This standard is equivalent to the vaporization of 34.5 pounds of water per hour from and at 2 12° F. ; it is frequently so quoted. It is also equivalent to 33,320 B. T. U. per hour. Since the power from steam is developed in the engine, and since the economy in the use of steam depends on the engine only, and may vary widely with the type of engine, it appears illogical to assign horse-power to a boiler. The method appears to be justified by custom and convenience. Rate of Combustion. — The rate of combustion is stated in pounds oi coai burnea per square foot of grate- surface per hour. It varies with the draught, the kind of coal, and the skill of the fireman. In general a slow or moderate rate of combustion gives the best results, both because the combustion is more likely to be complete and because the heating-surface of the boiler can then take up a larger portion of the heat generated. A very slow rate of combustion may be uneconomical, because there is a large excess of air admitted through the grate, and because there is a larger proportionate loss of heat by radia- tion and conduction. It is claimed that forced draught may be made to give complete combustion with a small amount of air in excess, and that it should give better economy than slower combustion. It will be remembered that a small amount of carbon monoxide due to incomplete combustion will cause more loss than a large amount of air in excess. Heating-surface. — All the area of the shell, flues, or - Trans., voi. vi, 1881. POWER OF BOILERS. I 37 tubes of a boiler which is covered by water, and exposed to hot gases, is considered to be heating-surface. Any surface above the water-line and exposed to hot gases is counted as superheating-surface. The upper ends of tubes of vertical boilers are in this condition. For a cylindrical tubular boiler the heating-surface in- cludes all that part of the cylindrical shell which is below the supports at the side walls, the rear tube-plate up to the brick- arch which guides the gases into the tubes, and all the inside surface of the tubes. The front tube-plate is not counted as heating-surface. For a vertical boiler like the Manning boiler (page 10) the heating-surface includes the sides and crown of the fire- box and all the inside surface of the tubes up to the water- line. Surface in the tubes above the water-line is superheat- ing-surface. A certain 200-H.P. boiler of this type has 1380 square feet of heating-surface and 470 square feet of super- heating-surface. The heating-surface of a locomotive-boiler consists of the sides and crown of the fire-box and the inside surface of the tubes. The heating-surface of a Scotch boiler consists of the surface of the furnace-flues above the grate and beyond the bridge, the inside of the combustion-chamber, and the inside surface of the tubes. The effective surface of any tube-plate is the surface re- maining after the areas of the openings through the tubes is deducted. Relative Value of Heating-surface. — A review of the kinds and conditions of heating-surface in various kinds of boilers, or even in a particular boiler, shows that the value of heating-surface varies widely. It does not appear possible to assign values to different kinds of heating-surface. We will note only that surfaces like the shell of a cylindrical boiler over the fire, like the inside of a fire-box, or like the flues of 138 S TEA M-B OILERS. a marine boiler, which are exposed to direct radiation from the fire, are the most energetic in their action. Surfaces like combustion-chambers and tube-plates, against which the flames play, are nearly if not quite as good. The inside of small flues and tubes is less favorably situated, more especially as the flame is, under ordinary conditions, rapidly extinguished after it enters such a flue or tube. The length of the flame in small tubes depends on the draught, and with very strong forced draught may extend completely through tubes of some length. The value of heating-surface in a tube rapidly decreases with the length. It is doubtful if there is any advantage in making the length of a horizontal tube more than fifty times the diameter. Tubes of vertical boilers should have twice that length. Ordinary Proportions. — The following table gives the ordinary proportions of various types of boilers: Type of Boiler. a o U o| a 3 Pi Square Feet of Heating-sur- face per Foot of Grate. > 6 cr> hi > rt u < in ■j u rtrs IS (A X 8 to 12 8 to 15 10 tO 20 j 50 to 120 I ) average 75 \ 8 to 15 35 to 45 9 to 15 j 15 to 67 ) ( average 20 j 25 to 30 35 to 40 *48+ 16 60 to 70 40 to 45 30 35 to 45 30 to 40 8 to 10 9 to 10.5 9 to 10.5 6.7 to 8.5 9 to 10.5 7 to 9 9 to 10.5 7 to 9 0.36 0.30 0.23 0.07 0.30 0.11 0.28 0.22 7.0 Cylindrical multitubular. Vertical, Manning 11. 5 11 .1 4-5 Locomotive type, sta- 12.6 Scotch marine 3.3 Water-tube with cylin- der or drum Water -tube with sepa- rator. . 11. 7-3 * 48 heating-surface, 16 superheating-surface. The higher rates of evaporative economy are associated with slower rates of combustion and with larger ratios of heating-surface to grate- surface. POWER OF BOILERS. 1 39 No attempt is made to distinguish the kind or location of heating-surface ; it must be understood that the ordinary ar- rangements and proportions for the several types are followed if this table is to be used in designing boilers. For example, it cannot be expected that heating-surface gained by length- ening the tubes of a locomotive-boiler will add materially to the efficiency of the boiler. This table has been compiled from a large number of ex- amples, and may be taken to represent current good practice. The last two columns giving the grate-surface and heating- surface have been computed on the basis of one horse-power for 34.5 pounds of water evaporated per hour from and at 212° F. The tables on pages 140 to 145 give the principal dimen- sions of notable merchant steamships, of ships in the United States Navy, and of ships in the British Navy. The table on page 146 gives the particulars of boilers on the U. S. S. Brooklyn, the most recent and powerful American cruiser. 140 STEAM-BOILERS. uo paads £"" ro. m ■o co oo oo 00 ro >o 00 VO r» CO 00 CO r^ co p~ r~- 5 oo Ov S o CM ro 'd'H o o ro B 1 lO ov 8 3 5 ro ■ ro CM ro 8 J eg 8 CM in 8 2 in pgjBDipu] CO ■o p-» 2" 00 T Ov vo vo ro •SS3JJ o 8 8v On D m in m m o v8 in p^ m E V Suijijo^ •E3J-B *f • CO o\ 3 CO VO vo in o 02 -3}TM£) ° M CM S M ro vo p^ 8 vo CM aoEjjns %■ ; ^ p~ 00 CM Ov 8 -3UIJB3H t~ ov M ro CM CM 00 ro O VO o VO p^ to 1 °° •3310J1S| Jl. iJL N CM VO ts. t^ so CM P~ _£_ JL VO VO VO vo VO VO p~ ro vo -jL. ^KtfO^IO <5» 8.8 h m m oo m m w m m vo vo w vo h moo o o t~-vo oo m o o-> P^vo P^ OV00 c saqouj ui vc -*oo vo i*» vo o p» mvo co ■*■ C^ H T»-VO M ■*vo ro t^ ■tvo ro p^ Ov >. 'J3puqA3 ill o O v CO vO ro 8 •aSnu o vo P~ cm co in t^ VO VO in* co VO t^ VO X -u ox SSOJQ t* vC > 00 t* ts. \C t^ in c~. *"* Ov 00 vo oo S -JU3UI 8 o rO & o o in 8 8 8 8 8 8 in I 8 § -aDBjdsia o Ov tN m 00 00 V> c c vo O O o VO •iqSnEja "ff p c CM IN p 1 CO CM 8 m m CM CM g CM CM CM CM CM CM p? M vO DC o c c vo CM O VO vo o •qidaa CO u g VO CM P T CO (■ ro m * ■* ro m ro Tt " "b c c c vO vo vo ro vo ro a •qipB3Ja T3 "3 o s •> m -i m ~1 u ■) m m "* *° m IT) in m m m VO vo vo vo vo •q;3u9T m s c P « 8 °° r^ vB VO m vo 8 ro VO v8 .35 in 1 in ■> -i- m "*■ in ■* "0- in m * in ts 8. 8v •3^a| s (X a co 00 oo oo 00 00 o :■: CO a; DC 00 *2 oo 00 00 C CO ~A— a C C C C ~ o o o V P3 E=3 to ^ V U) a B a . a CO T3 tyj o O U c o o O o o h O o 2 '5 C3 H O u J3 U ■a H u •a u H o u u •a o U O U CO CO is o u en (/) (O U3 «5 c m u 0? 23 VU c c V CC u 1 m 3 3 3 s El, cc ~ Ec s Cb b s > ^ — V— b > > fx, o 2 c nl o rt u a a n rt 3 <— s R Z CO u a a c c bu c c 1 r u jC 5 rt i J3 B Hi 1 > 7, C t > bj 'i 5 J. C C H u u 4) CC? c c rt E a V a. LO V 13 > B £ 2 c « c S U u c rs "J : m m m m m o»oo m vo -»- 000 -f N if 0> ->J- in >« mvO VO f> •» On t^ "-" •>»■ m m*0 On « (IN •9DBJJnS-31EJO 0>t^w *■ ■«■ m M fOO in m cvi vo r-< rON •* m - Oh ^ •3inSS3jd-tDB31S m in in uimmo ino if. in in in m o Ofno^ininmmininM in m mioinminminio • m o . m •- • M CN| •sjajiog m -i mvo mo ina»Moo O m»oo mij. m on invo m invo in m ioo in vo •sauiSua o o o 1 o m ^- m o in m o n cvi o m s < •sqq 'JHI J3d •suoj, I^ox vo -«i- C m invo vo vo m *ONfnOM\Ovo io m* mvo ■* t^vo fm»» n m m w 01 IN N •spunoj ui iqSnvuQ jad Ajauiipn^ -uiqDKjMiBjox O\mt-»m'*r^in>'00 0* vo m f^vo in ■*■ On « t- w w OMHioocoxiOMin^uindtHMntiN o- m ►. l-~ m 0>iOM< mvO N 2 mo> is vo m M *-N •iqSncjQ paojoj N3>O00« if f-oo 00 o>N» m M TfVO Tf w n r^ M M M - » N - - M N « •iqJgriHJa m ■ mo • m 0} 3nQ 3SB3JDUI 1U3J J3J iqSnEja P3DJOJ r-~vo vo O 00 m N T)- - if - 00 h nvd m moo vo m t-^vo mvo o O n « mmn c n vo Ov m m o> n •jqSnEja •*■ On On "" v»Ol -' n m VO VO lis C >-T3 C « »J __ o rt o o .. _, CD C _X jc , N Vh cut off and carry short vertical yokes, which are connected by two smaller rods, one above and one below the diagonal stays. The rings forming the barrel of the locomotive are made progressively smaller from the fire-box to the smoke-box; the slight taper toward the front end of the locomotive is found convenient in the design of the machine. Fig. 55 shows two ways of making the furnace-mouth of a locomotive-boiler. In one way the end- plate of the boiler-shell and the corre- sponding plate of the fire-box are flanged iff^ in the same direction, and are riveted out- side of the boiler. In the other case the two plates are flanged into the water-space and the overlapping edges are riveted. Marine Boiler. — The parts requiring staying in the Scotch boiler are the flat ends, the furnaces, and the combustion- chambers. The flat ends above the tubes are stayed by through-stays with nuts inside and with washers and nuts outside the plate. The boiler shown by Fig. 13, page 17, has two rows of through-stays — four in the upper and six in the lower row; two of the upper row pass through the fitting which carries the steam- nozzle. It is found in practice that the tube-sheets of a marine boiler are not sufficiently stayed by plain tubes expanded into the sheets. It is customary to make a portion of the tubes thicker than the others, and to provide these thick tubes with thin nuts outside the tube-plates, so that they may act more effectively as stays. The thick tubes in Fig. 13 are indicated by heavy circles. Sometimes every other tube of each second row is made a thick tube; that is, something more than one fourth of the tubes are stay-tubes. Usually the number is fewer than this. Fig. 55. STAYING AND OTHER DEI' AILS. l6l Below the tubes the front plate is supported in part by the furnace-flues, and in part by through-stays running to the combustion-chamber. There are two such stays above the furnaces and three below the furnaces in the middle of Fio- 13, each if of an inch in diameter. There are also two stays 2\ inches in diameter, one at each side and above the fur- naces. These last stays have one point of attachment to the front end-plate, but each has two points of attachment to the combustion-chamber. For this purpose the rear ends of the stays are bolted to V-shaped forgings, similar to that shown by Fig. 56, page 162. The furnace-flues are corrugated to stiffen them, and thus maintain their form under the external pressure to which they are subjected. The corrugations in Fig. 13 are made up of alternate convex and concave semicircles; other forms of corrugations and other methods of stiffening flues, together with a discussion of the strength of flues, will be given in the next chapter. The front ends of the furnace-flues in Fig. 13 are made as large as the outside of the corrugations; the rear ends are as small as the inside of the corrugations. Such an arrangement makes it easy to remove the furnaces without disturbing the other parts of the boiler and without destroy- ing the flues. The combustion-chambers of a Scotch boiler are made up of flat or curved plates subjected to external pressure, and must be stayed at frequent intervals to prevent collapsing. The sides and bottom of the combustion-chamber in Fig. 13 are stayed to the cylindrical shell of the boiler by screwed stay-bolts, spaced seven inches on centres. The back of the combustion-chamber is stayed in like manner to the back end of the boiler, and thus both of these flat surfaces are secured. The plates used for making the combustion-chamber are thicker than those used for a locomotive fire-box, and conse- quently the stays are spaced wider and are larger in diameter. The top of the combustion-chamber is staved by stay- 1 62 S TEA M-B01LERS. bolts and bridges in a manner that suggests the crown-bar staying of a locomotive fire-box. The space is, however, narrower and the staying is less complicated. Complex Stays. — Sometimes the points to be connected by stays are so numerous that too many through-stays will be -» Fig. 56 required if all points are stayed separately. Thus in Fig. 56 there is an angle-iron riveted to a flat plate, and supported at intervals, as indicated by the two bolts passing through it. Instead of using a through-stay for each bolt, the bolts are coupled by two V-shaped forgings, which forgings are bolted to a through-stay at the angle of the V. There is enough freedom of the bolts in their holes to give equal distribution of the pull on the through-stay. By an exten- sion of this method several points may be supported by one stay-rod. Gusset-stays. — The flat ends of the Lancashire boiler, shown by Fig. 3, page 6, are secured to the cylindrical shell by gusset-stays; such a stay is shown more in detail by Fig. 57. A plate is sheared to the proper form, and is riveted S TA YING A ND O THER D E TA IL S, •65 between two angle-irons along the edges that come against the shell and the flat end. The angle-irons in turn are riveted to the shell or to the flat plate. Gusset-stays have the advantages of simplicity and solidity. They interfere less with the accessibility of the boiler than through-stays or diagonal stays. Their chief defect is that they are very rigid and are apt to localize the springing of the flat plates, which Fig. 57. is caused by unequal expansion of the furnace-flues and shell. Consequently, grooving near gusset-stays is very likely to be found in Lancashire and Cornish boilers. Gusset-stays are also used to some extent in marine boilers, and in locomo- tive-boilers. Spherical Ends. — The ends of cylindrical boilers, or of steam-drums, are commonly curved to form a spherical sur- face, in which case they retain their form under internal pressure and do not need staying. If the radius of the spherical surface is equal to the diameter of the cylindrical surface, the same thickness of plates may be used for both. If the spherical surface has a longer radius, the thickness may be increased. Such disJicd heads of boilers and steam-drums are struck up between dies while at a flanging heat, and are then flanged to give a convenient riveting edge. Steam-domes are short, vertical cylinders of boiler-plate fastened on top of the shell of horizontal boilers. Plates II and III show steam-drums on locomotive-boilers. A steam- drum may be used to advantage when the steam-space is so shallow that there is danger that the ebullition may throw 1 64 S TEA M-B OILERS. water into the pipe leading steam from the boiler. "Locomo- tives usually have steam-domes, for not oniy is the steam- space shallow, but there is danger of splashing of the water in the boiler, especially if the track is rough or sharply curved. Stationary boilers ought to have steam-space enough with- out domes; marine boilers sometimes- have domes, but they are less common than formerly. The additional steam- volume in a steam-dome is insignificant, so that a dome should not be added to increase steam-space of a boiler. The main objection to a steam-dome is that it weakens the boiler-shell, which must be cut away to form a junction with it. The shell may be reinforced, to make partial com- pensation, by a ring or flange of boiler-plate. Such a flange is clearly shown on Plate III, where the longitudinal seam of the ring carrying the dome is purposely placed at the top of the boiler. A similar arrangement is made for the dome on Plate II. Dry-pipe. — Any pipe inside of a boiler for the purpose of leading steam from the boiler is known as a dry-pipe; the pressure in such a pipe is frequently less than that of the steam in the boiler, consequently there is a tendency to dry the steam in the pipe. Dry-pipes are found in locomotive and marine boilers and sometimes in stationary boilers. The dry-pipe of a locomotive opens near the top of the dome. It runs vertically down till it is well below the shell of the barrel, then it runs horizontally through the steam- space and out through the smoke-box tube-sheet. The throttle-valve is at the inlet of the dry-pipe. It is controlled through a bell-crank lever by a rod which enters the head of the boiler from the cab. The marine boiler shown by Fig. 13 has a dry-pipe which is joined to a steam-nozzle at the front end of the boiler. This dry-pipe is pierced with numerous longitudinal slits on STAYING AND OTHER DETAILS. l6$ the upper side; the sum of the area of such slits is seven eighths of the area. through the stop-valve in the steam-pipe. Steam-nozzle. — The stationary boiler shown on Plate I has a cast-iron steam-nozzle at each end. The steam-pipe leading steam from the boiler is bolted to the rear nozzle, and the safety-valves are placed above the front nozzle. Nozzles are often made of cast steel. The best are forged without welds from one piece of steel. Manholes. — A manhole should be large enough'to allow a man to pass easily inside the boiler. That on Plate I is fifteen inches long and eleven inches wide, and has its greatest dimension across the boiler. The manhole there shown is placed inside the shell of the boiler. Both the ring and the cover are forged from steel without a weld. Fig. 58 shows a form of manhole that is Fig. 58, placed outside the shell. This form is commonly made of cast iron, but cast steel manholes of similar form are used to some extent. The manhole-ring should be strong enough to give com- pensation for the plate cut away from the ring on which it is placed. The manhole-cover is placed inside the ring so that it is held up to its seat by the steam-pressure. The cover is drawn up to its seat by a bolt and removable yoke. Some- 1 66 STEAM-BOILERS. , times there are two bolts each with its yoke. A cast-iron manhole naturally has a cast-iron yoke, and a forged manhole has a wrought-iron or steel yoke. The manhole-cover is made steam-tight by a rubber gasket; the form of the cover and its seat are such that the gasket cannot be blown out by the pressure of the steam. Hand-holes are provided at various places on boilers to aid in washing out and cleaning. Thus the boiler on Plate I has a hand-hole near the bottom at each end, and there are several hand-holes near the foundation-ring of the vertical boiler, shown by Fig. 5. The hand-hole covers on Plate I are placed directly against the plate which is not reinforced. Each is held up by a bolt and a small yoke, which has a bearing on the plate completely round the hole. If the yoke has insuffi- cient bearing on the plate, the latter is liable to be damaged and leaks will occur. The hand-holes on the marine boiler shown by Fig. 13 are reinforced by small plates outside the boiler-heads. Washout Plugs. — Instead of hand-holes, washout plugs, two inches or two inches and a half in diameter, are provided near the corners of the foundation-ring of a locomotive fire- box. Such plugs are simply screwed into the outside plate of the boiler. Examples are shown by Plates II and III. Methods of Supporting Boilers. — Horizontal cylindrical boilers are commonly supported on the side walls of the brick setting, by brackets which are riveted to the shell of the boiler. Thus the boiler shown on Plate I has two such brackets on each side; this boiler is about sixteen feet long. If a boiler is as much as eighteen feet long, three brackets are used. The front brackets rest directly on the brickwork, but the other brackets rest on iron rollers, to provide for the expansion of the boiler. The brackets are set so that the plane of support is a little above the middle of the boiler. Fig. 59 shows a common form of bracket, made of cast iron, which is riveted to the shell above the flange of the STAYING AND OTHER DETAILS. I6 7 bracket. A better form with rivets both above and below the flange is shown by Fig. 60. 00 00 Fig. 39. Fig. 60. A detachable bracket, like that shown by Figs. 61 and 62, may be used when the boiler must be put into a building © J - © @ n v\ Fig. 61. through a small aperture. Fig. 62. Fig. 61 gives an end and side elevation and plan of the body of the bracket ; Fig. 62 gives Fig. 63. Fig. 64. a side elevation and plan, with section, of the flange. After the boiler is in place the flange is thrust up into the dovetail 1 68 S TEA M-BOIL ERS. groove in the body of the bracket. The pressure of the flange against the dovetail groove, intensified by the wedging action of the inclined sides, is liable to be excessive. To overcome this difficulty the bracket shown by Figs. 63 and 64 is oftea Fig. 65. r 1 A \ ; 1 "6" Fig. 67. Fig. 69. used. Fig. 63 shows the end elevation and a view from below, of a casting which is riveted to the shell. Fig. 64 shows the same views of a casting which catches into the hollow under Fig. 63 and bears at the top against this same casting, the rivets bolting it to the shell being countersunk. STAYING AND OTHER DETAILS. 1 69 Horizontal boilers, and especially plain cylindrical boilers, are sometimes hung from a support above the boiler, as shown by Figs. 65, 66, and 67. Fig. 65 shows a lug, made of boiler-plate, riveted to the shell of the boiler. The lugs are placed in pairs and the boiler is hung from these lugs by bolts that are supported between transverse beams over the boiler. Fig. 66 differs in substituting a loop for the lug. Fig. 67 shows a method of suspension with two short pieces of plate above the lug, to give some flexibility and provide for expansion. Figs. 68 and 69 show methods of suspending a boiler from the top. These methods are proper only for boilers which have a small diameter. CHAPTER VII. STRENGTH OF BOILERS. The determination of the thickness of boiler-plates, the size of stays, and other elements affecting the strength of a boiler, involves a knowledge of the properties of the materials used and a knowledge of the methods of calculating stresses in the several members of the boiler. A brief statement of these subjects, as applied to boilers, will be given here. Materials Used. — The materials used for making boilers are mild steel, wrought iron, cast iron, malleable iron, copper, bronze, and brass. In order to insure that materials used for making a boiler shall have the proper qualities, it is customary to require that specimens shall be tested in a testing-machine, and that they shall have certain definite properties, such as ultimate tensile strength, elastic limit, and contraction of area at fracture. In order that these properties shall be properly developed, it is essential that specimens shall be of right size and shape, and that the testing shall proceed in a correct method. Testing-machines. — The frame of a testing-machine carries two heads, between which the test-piece is placed, and to which it is fastened by wedges or other clamping devices. One head, called the straining-head, is drawn by screws or by a Jiydraulic piston, and pulls on the test-piece. The other head, called the weighing-head, transmits the pull to some weighing device. Boiler materials are commonly tested in a machine which has the pull applied by screws, driven through gearing by hand or by power; the pull is weighed by a system of levers and knife-edges, arranged like those of a platform 170 STRENGTH OF' BOILERS. *7 l scale. Such a machine should be able to exert a pull of fifty or a hundred thousand pounds. Testing-machines that give a direct tension are commonly arranged to give also a direct compression. There are also machines arranged to give transverse loads, like the load applied to a beam. Forms of Test-pieces. — A test-piece of boiler-plate should be at least one inch wide, planed on both edges, and should be about two feet long. A piece which is loss than eighteen inches long is not fit for testing. Test-pieces eighteen inches to two feet long may be cut directly from bars or rods for making stays or bolts. If a rod is so large that the available testing-machine will not break- it, it is of course possible to turn it down to a smaller diameter, but it would be preferable to send such a rod to a machine that is powerful enough to break it at full size. Test-pieces of cast metal may be cast in the form of rectangular bars, which should be at least one inch wide and an inch thick. If the bars are rough or irregular it may be necessary to plane the edges, or perhaps to plane them all over. Test-pieces of boiler-plate should be cut from the edge of at least one plate of each lot of plates. Sometimes speci- fications require pieces from each plate used for a given boiler. Pieces should be cut from both the side and the end of a plate, for there is a grain developed by rolling either iron or steel boiler-plate, and tests should be made both with the grain and across the grain. Very hard material may require shoulders on the test- pieces to enable the testing-machine to get a proper hold. But iron or steel that is so hard as to require shoulders is much too hard for boiler-making; consequently there will be no reason for providing test-pieces of boiler iron or steel with shoulders. If test-pieces have shoulders, they should be at least ten inches apart. 172 STEAM-BOILERS. Methods jrf Testing.— A test-piece of proper length is first measured to determine the breadth and thickness or else the diameter, as the case may be. A length of eight inches is laid off near the middle of the test- piece, and clamps for measuring the stretch of the piece are ap- plied at the ends of this eight-inch length, as shown by Fig. 70. The piece is then secured in the machine and a load is applied. The distance between the clamps is now measured by a micrometer caliper with an extension-piece. The method of doing this is to place the head of the micrometer against a point on the flange of the clamp at one end, and adjust the length of the micrometer so that it shall just touch the cor- responding point on the other clamp. A little practice will en- able the observer to measure to one or two ten -thousandths of an inch. As the load is increased the test-piece stretches, the in- crease of length being proportion- al to the increase of the load. The stretch is measured on both sides of the test-piece for each increase of load applied. If the test-piece is not straight or exactly aligned in the machine there may be some irregularity in the stretching at Fig. 70. STRENGTH OF BOILERS. 1 73 first, but after a considerable load is applied the piece stretches uniformly until about half the maximum load that the piece can carry has been applied. During the progres? of the test a point is reached beyond which the stretch in creases more rapidly than the load; this is known as the elastic limit. After the elastic limit is reached the clamps are removed and the test proceeds without them, but at about the same rate of loading. A load is soon reached which the piece cannot permanently endure, shown by the fact that the scale- beam will fall though the straining-head remains at rest. This is called the stretch limit. The piece may, however, carry a considerably higher load if the straining-head is kept moving to take up the stretch. Finally, the piece begins to draw down rapidly, somewhere near the middle of its length, and when the piece breaks, the fracture shows about half the area of the piece before testing. Hard materials may draw down little, or not at all; the limit of elasticity may approach the strength of the material. The jaws or wedges of the testing-machine interfere with the stretching or flow of the material gripped by them. The influence of the wedges may extend two or three inches beyond their edge in the testing of boiler-plate. If a piece has shoulders they will have a like effect. Consequently the points at which a clamp is secured to a test-piece should be two or three inches from a shoulder or from the wedges of the machine. The wedges of a machine of a capacity of fifty or a hundred thousand pounds are four or five inches Ion"". They will grip on three inches at the end of a test-piece, but not on less. The test-piece must have eight inches for measuring stretch, two or three inches at each end for flow, and three to five inches at each end in the wedges. Conse- quently the piece must be eighteen or twenty-four inches long. The method just described is slow and laborious, and 774 S TEA M-B OILERS. requires two observers — one to measure stretch and one to weigh. For commercial work an automatic device is often used which registers loads and corresponding elongations. Such devices commonly record the stretch limit instead of the elastic limit; these two points should never be confused. Stress. — The number of pounds of force per square inch is called the stress. The stress is uniform on a piece under direct tension, and is equal to the load divided by the area of transverse section. Stress may be expressed in other units, such as tons per square foot or kilograms per square milli- meter. Strain. — The stretch of a piece, under direct tension, per unit of length is called the strain. If the original length is / . . # and the stretch is a, then the strain is - = s. The Limit of Elasticity is the limiting stress beyond which the strain increases more rapidly than the stress. The limit is not perfectly definite, and can be determined approxi- mately only. A load greater than the elastic limit will pro- duce an appreciable permanent elongation after the load is removed. A stress less than the elastic limit will produce only a slight permanent elongation; such elongation may be inappreciable. Stretch Limit. — The stress at which the scale-beam of a testing-machine will fall while the straining-head is at rest is called the stretch limit. Ultimate Strength. — The maximum stress that a piece will endure in a testing-machine is called the ultimate strength of the material. The strength depends somewhat on the rate of testing. The more rapidly the testing proceeds the higher will be the apparent strength. It is desirable that some standard rate of testing may be adopted by engineers so that results may be strictly comparable. The Modulus of Elasticity is the result obtained by dividing the stress by the strain. If the stress is/ pounds STRENGTH OF BOILERS. 175 per square inch and the strain is s per inch, then the modulus of elasticity is E = t S Reduction of Area. — The area of the test-piece of boiler- plate at the rupture is much less than that of the piece before testing. This reduction is important, as it shows the ductility of the metal, and its ability to change shape without too much distress under the influence of unequal expansion of different members of a boiler. Ultimate Elongation. — After the test-piece is broken the two parts are laid down in a straight line with the broken ends in contact, and the length of the distance between the points of attachments of the measuring clamps is measured. The ratio of the elongation to the original length (eight inches) is called the ultimate elongation. The ultimate elon- gation is generally given in per cent. This is important, for the same reason given for the contraction of area. Compression. — The preceding definitions are given for tension only, for sake of simplicity and brevity; they may be applied to pieces in direct compression if the term stretch or elongation is replaced by compression. Shearing. — Stresses have thus far been considered to be at right angles to the sections of the pieces to which they are applied, and produce either tension or compression at that section. A stress that is not at right angles to a section will tend to produce sliding at that section. A stress that is parallel to a section will tend to produce sliding only, and is called a shearing-stress. If a shearing-stress is uniformly dis- tributed, its intensity may be found by dividing the total force or load by the area of the section. The rivets of a riveted seam are subjected to a shearing- stress. 1 76 S TEA M-B OILERS. Steel. — At the present time boiler-plates are made of mild, open-hearth steel; good wrought-iron plates can be obtained with some difficulty and trouble. Such steel is in reality a tongh, ductile, ingot iron, containing about one fourth of one per cent of carbon; it is nearly free from sulphur and phosphorus. The former impurity makes iron hot-short and the latter makes it cold-short, i.e., brittle when hot or cold. Plates of this material can be obtained of all sizes and thicknesses up to eight feet wide and an inch and a quarter thick. There is no limit to length except convenience of handling. Steel plate lor boiler-making should have the following properties: Tensile strength 55,000 to 60,000 Elastic limit 30,000 to 33,000 Elongation in eight inches.... 25 per cent or more Reduction of area at fracture. . 50 per cent or more. The plate should be free from blisters, lamination, or blow-holes. A piece cut from a plate less than three fourths of an inch thick should endure bending double under a heavy hammer, both hot and cold, without showing cracks. Heavy plates should endure bending at a small radius to a large angle with- out cracking. The upper end of the ingot into which the molten steel from the open-hearth furnace is cast, is liable to be affected by bubbles and other imperfections when the ingot is poured from the top. Such imperfections, if they are not removed, give rise to lamination in the plates, and therefore when the ingot is roiled into blooms the crop end should be cut long enough to remove all the bubbles. There is always a ten- dency, on account of the reduction of prices through com- petition, to reduce the length of the crop end, and conse- STRENGTH OF BOILERS. 1/7 quently steel plates, though having the other required physical properties, are liable to show lamination. To guard against this, test-pieces should be cut from the ends of plates and tested in a testing-machine, and also by bending hot and cold. Ingots have been cast from the bottom, in which case bubbles are likely to be distributed throughout the ingot. Steel plates are sometimes classified as shell-plates and fire-box plates; the latter are supposed to be of special quality to endure the flanging required in the forming of the locomo- tive fire-box, and to endure the stresses in service due to the action of the fire, draughts of air entering through the fire- door, and from the unequal expansion of the fire-box and the parts of the shell to which it is stayed or otherwise connected. There does not appear to be any difference in the chemical and physical characteristics of these two grades, except the somewhat greater ductility of the fire-box plates, due to greater care in making. Angle-irons, T irons, bars, and rods used for staying and fastening boilers may be made of steel if welding is not re- quired in forming them. Blue Heat. — Steel plates, and other forms of mild steel, become brittle at a temperature corresponding, roughly, to a blue heat. A plate that will endure bending double, both hot and cold, is liable to show cracks if bent at a blue heat. In bending, flanging, and forging no work should be done on steel at a blue heat; properly, such work should be done at a bright red heat; work should never be continued after the steel becomes black. After the steel is cold it may be bent as readily as iron at the same temperature. Wrought Iron. — All the stays and fastenings of boilers that are made by welding should be made of tough, ductile wrought iron. Welds made by a skilful smith may have as great a strength as the bar from which they are made. A ductile bar may break in the clear bar instead of in the weld, 178 S TEA M-B OILERS. on account of the hardening due to the work done on the bar at the weld. It is customary to assume that 25 to 50 per cent of the strength of the bar may be lost by welding. Wrought-iron plates of a quality suitable for boiler-making are now more expensive than mild-steel plates, which are in every way as well adapted to the purpose, and which have a higher strength. Consequently we find wrought-iron plates used only when specially demanded. Wrought iron does not show cracks when worked at a blue heat, and in general may endure more abuse in working. This caused wrought iron to be preferred by many after reliable steel was produced cheaply, but boiler-makers now understand the working of steel plates and avoid improper handling. Wrought-iron plates should show a limit of elasticity of 23,000 pounds, and a tensile strength of 45,000 pounds to the square inch. Wrought-iron rods and bolts should have a strength of 48,000 pounds per square inch. Rivets. — The rivets used in boiler-making are either iron, or steel similar in quality to steel used for boiler-plates. A rivet should bend cold around a bar of the same diameter, and it should bend double when hot without frac- ture. The tail should admit of being hammered down when hot till it forms a disk 2-J- times the diameter of the shank, without cracking. The shank should admit of being ham- mered flat when cold, and then punched with a hole equal in diameter to that of the shank, without cracking. The rods from which rivets are made should show a tensile strength of about 55,000 pounds for steel and about 48,000 pounds for wrought iron. The other properties, such as ultimate elongation and contraction of area, should be like those for boiler-plate. The shearing strength of steel rivets is about 45,000 pounds, and of iron rivets about 38,000 pounds; that is, the STRENGTH OF BOILERS. 1 79 shearing strength will be about two thirds of the tensile strength. Cast Iron in different forms will show a tensile strength of 12,000 to 20,000 pounds to the square inch. Gun-iron, which is cast iron made with special care and skill from selected stock, has shown a tensile strength of nearly 30,000 pounds to the square inch. In compression the strength of small pieces may be as high as 80,000 pounds to the square inch, but larger pieces, like columns, fail at 30,000 pounds to the square inch. Cast iron is used for some or all of the parts of sectional boilers, and for fittings such as manholes, though wrought iron is preferable for such purposes. Flat plates at the ends of cylindrical boilers are sometimes made of cast iron. In general, cast iron should never be used when it is sub- jected to severe changes of temperature or to stresses from unequal expansion, and should be replaced by wrought iron or mild steel whenever it is practicable. Couplings, elbows, and other pipe-fittings are made of cast iron. The brittleness is a convenience when changes are to be made, as joints that cannot be opened are readily broken. Malleable Iron, which is cast iron toughened by being deprived of part of the carbon, is used for pipe-fittings and for fittings of steam-boilers. It is used in place of cast iron for sectional boilers and for parts of water-tube boilers. Though tougher than cast iron, and though it will endure forging to some extent, its variability in quality and its unreliability prevent much reduction in weight and size when substituted for cast iron. Copper is largely used in Europe for making fire-boxes of locomotive-boilers and torpedo-boat boilers. Its greater cost is in part offset by the value of the scrap copper after the fire-box is worn out. Copper for fire-boxes, rivets, and stays should have a ten- T SO S TEAM-BOILERS. sile strength of 34,000 pounds to the square inch, and should show an elongation of 20 to 25 per cent in 8 inches. It should not contain more than one-half per cent of impurities. The greater ductility of copper, and its greater thermal conduc- tivity, permitting of greater thickness for furnace-plates, recommends it to European engineers. Copper is largely used on steamships for making piping of all sorts, such as steam-pipes and water-pipes. Such pipes are made of sheet copper, rolled up or hammered to shape, scarfed and brazed at the edges. The pipe is also brazed to brass flanges for coupling lengths of pipe, or for joining to steam-chests or other parts of the engine or boiler. If the brazing is not done with care and skill the brazed joint may lose as much as half the strength of the sheet copper. Several disastrous explosions of such piping have occurred. Conse- quently wrought-iron piping is finding favor for high-pressure steam. Bronze and Composition. Brass.— Bronze is properly an alloy of copper and tin; thus gun-metal is 90 parts of copper to 10 of tin. Compositions of various qualities are made of copper and zinc with more or less tin. Brass is an alloy of copper and zinc; for example, brass smoke-tubes are made of 70 parts of copper to 30 parts of zinc. Lead is added to brass and to composition to reduce the cost and to make the metal work easier. It may be considered as an adulteration, as it cheapens the metal at the expense of the quality. There are many special bronzes, such as phosphor- bronze and aluminium-bronze, which are used for 'special purposes. Brass is used to some extent for smoke-tubes of locomo- tive and other boilers, on account of its greater thermal con- ductivity, by European engineers. In America, brass is used for valves, gauges, and other boiler fittings. Composition or bronze is advantageously used for the valves and seats of safety-valves and wherever the service endured is excep- STRENGTH OF BOILERS. ' l8l tionally hard. Brass is more commonly used because it is cheaper. In a general way it may be said that the cost and quality of brass and composition is proportional to the copper it contains; thus red brass is better and costs more than yellow brass. Many small brass fittings on the market are sold at a price which precludes the use of proper alloys, and they are consequently soft and worthless. Stay-bolts are usually arranged in equidistant horizontal and vertical rows; as an example we may take the stay-bolts in the locomotive fire-box on Plate II. These bolts are 7/8 of an inch in diameter outside of the threads, and are spaced 4 inches on centres. The total load on each stay-bolt with a steam-pressure of 170 pounds to the square inch is 4 X 4 X 170 = 2720 pounds. The diameter of the bolt at the bottom of the screw-thread is about 0.7 of an inch, and the area of the section is about 0.4 of a square inch. The stress is consequently 2720 -f- 0.4 = 6800. Sometimes the area is calculated from the external diam- eter of the bolt, a proceeding which may lead to a gross error. In the present instance the corresponding area is about 0.6 of a square inch, which gives an apparent stress of about 4500. Suppose that the thread is turned off from the body of the bolt, and that the diameter is thereby reduced to 5/8 of an inch. The area of the section is then about 0.3 of an inch, and the stress is 2720 -f- 0.3 = 9000 +. The stress on stay-bolts should always be low to allow for wasting from corrosion, and to allow for unknown addi- tional stresses that may be caused by the unequal expansion of the plates that are tied together by the stay-bolts. 182 STEAM-BOILERS. Stay-rods. — Through-stays like those passing through the steam-space of the marine boiler, shown by Fig. 13, page 17, are treated much like stay-bolts. Thus the stays in question are 14 inches apart horizontally and 13 inches apart vertically. If they are each assumed to support a rectangular area 13 inches wide and 14 inches long, the total force from 160 pounds steam-pressure will be 14 X 13 X 160 = 29120. The diameter of these stays in the body is 2 inches, which gives an area of section of 3.14 square inches. The stress is consequently 29120 -\- 3.14 = 9300 These stay-rods have swaged heads on which the screw- thread is cut, so that the diameter at the bottom of the thread is greater than the diameter of the body. Stay-rods which are used in connection with girders, as on Plate I, will have to carry loads which depend on the surface supported, the steam-pressure, and the number and arrange- ment of the stays. The determination of the load may be difficult and uncertain, but the calculation of the stress for a given load is very simple. Diagonal Stays. — If a stay-rod runs diagonally from a flat plate to the shell of a boiler, it will evidently be subjected Fig. 71. to a greater stress than it would be if it were a through-stay. Thus in Fig. 71 we have at the point a the parallelogram of STRENGTH OF BOILERS. 1 83 forces abcd\ ab is the total pressure supported by the stay, ac is the pull on the stay, and ad is a force that must be taken up by the flat plate. But the triangles abc and aef are simi- lar, so that we have ac af Yae + ef ab ef ef Suppose, for example, that ae is two feet and ef is six feet ; then ac Vt -4- 6" 2 ^=-^- = I -° 54 ' or the pull on the stay is more than five per cent in excess of what a through-stay would be required to support. Gusset-stays are open to the defect that the distribu- tion of stress on the plate forming the stay is uneven and uncertain. It is customary to calculate them on the assump- tion that the resultant stress acts along a medial line, and is evenly distributed over a section at right angles to that line. A low apparent working-stress should be used. \ / Thin Hollow Cylinder. — Let Fig. \ 72 .represent a semicircular steam-drum v ^~- — -'^ closed at the bottom by a thick flat plate. FlG * 72 * If the steam-pressure is/ pounds per square inch, the radius is r t and the length is /, then the pressure on the plate is 2prl. If the thickness of the cylinder is /, and the stress per square inch on the metal of the cylinder is s, then the pull of the cylinder at one end of the plate is stl. 1 84 S TEA M-BOILERS. But this must be equal to half the pressure on the plate, so that stl — prl. pr For safety the stress should not exceed the safe working stress for the material of which the cylinder is made; so that we have It is evident that the pull on the side of the cylinder and the stress per square inch will be the same if another half- cylinder is substituted for this plate, making a complete thin hollow cylinder. Example I. — A thin hollow cylinder five feet in diameter and half an inch thick, working at a pressure of 200 pounds, will be subjected to a stress of 5 X 12 200 X '- $ = 12,000 pounds per square inch. If the cylinder is made of one con- tinuous plate of steel without longitudinal joint, this stress will be about one fifth of the ultimate strength. Example 2. — If it is desired that the stress shall be 9000 pounds in a cylinder 9 feet in diameter when exposed to a pressure of 120 pounds to the square inch, then the thickness of the plate should be pr 9 X 12 t — — : = 120 X -5- 9000 = O.72 of an inch. End Tension on a Cylinder. — In the preceding cylinder we have considered the tension on a section at the side of the STRENGTH OF BOILERS. I 85 •cylinder. Let us now consider the tension on a transverse section. If the cylinder is closed by a flat plate at the end, the area of that plate will be 3.1416?- 2 , and the total force due to a pressure of p pounds per square inch will be 3.i4i6r>. This force will be resisted by a ring of metal having a cir- cumference 2 X 3.1416?-, and a thickness t. The resistance of the ring will be 2 X 3.14167-/5, representing the stress by s. Consequently we shall have 2 X 3.1416^ = 3.1416^/. pr It is evident that the stress from the end pull is half the stress on the section at the side of a cylinder, and conse- quently a cylinder made of homogeneous material without joints will always be ruptured longitudinally. It is also evident that the stress from the end pull will be the same if the end of the cylinder is closed by a spherical surface, or by any other figure, instead of a flat plate. Thin Hollow Sphere- — A section taken through the cen- tre of a sphere is in the same condition as a transverse sec- tion of a thin cylinder, and will be subjected to the same stress, if the sphere has the same thickness and is subjected to the same internal pressure. Formerly the ends of plain cylindrical boilers were made hemispherical, but such ends are difficult to make and are needlessly strong if of the same thickness as the cylindrical 1 86 5 TEA M-B OILERS. shell. It is now the practice to curve such ends to a less radius than that of the cylindrical shell. If the radius of the head is equal to the diameter of the shell, then with the same thickness of plate the stress will be the same per square inch, provided there are no seams in head or shell. The heads usually do not have a seam, and the shells always have a seam; the margin of strength in the head, when the same thickness of plate is used, under this condition may be offset against the possible injury done to the head in shaping it. The construction known as a bumpcd-up head has the edge flanged into a cylindrical form to make a joint with the shell, and to avoid the awkward stress that would be thrown onto the cylindrical shell if the true cylindrical and spherical surfaces were allowed to intersect. If it is inconvenient to curve the head to a radius as small as the diameter of the cylinder, then a thicker plate may be used, with a longer radius. Rivets. — The plates of a boiler are joined at the edges by rivets; rivets are also used in stays and other members. LThe usual form of rivets is shown by Fig. 73. If the diameter of the rivet is D, then the Fig. 73. proportions may be ^S A 5=1-4; B D = 0.7; D = 3/4. STRENGTH OF BOILERS. 187 The length of the rivet will depend on the number and thickness of the plates through which it is to pass. The rivet represented by Fig. 73 has a pan head. Of the rivets shown by Fig. 74, A, B, and C have pan heads, and D and E have round or hemispherical heads. The form of the point of a rivet will depend on the way in which the rivet is driven and on the shape of the tools or dies used for forming the point. The rivet A has a straight ABC conical point; this is the only form that can be made when the rivet is driven by hand with flat-faced hammers. The rivet B has the head formed by a die or snap. The rivet is driven by a few heavy blows of a hammer, and the head is roughly formed; then a die or snap is placed on the point and driven to form the point by a sledge-hammer. C shows a rounded conical point commonly used for machine-driven rivets. The heads of such rivets may have a similar form. D represents the usual form of countersunk rivets: the hemispherical head is not a peculiarity of such rivets; it is occasionally used with any form of point. The rivet E has some fulness or projection at the point beyond the counter- sink. After a rivet is driven, both ends are called heads; the distinction of heads and points is made here for convenience in description. The straight conical form A is liable to be too flat and weak. Its height should be three-fourths the diameter of the rivet. 1 88 STEAM-BOILERS. When rivet-holes are punched in plates they are slightly conical, as shown by B, Fig. 74, which shows the two smaller ends of the rivet-holes placed together to facilitate the proper filling of the hole by the rivets. The other rivet-holes are straight, as they would be if drilled. Riveted Joints. — The proportions of riveted joints, such as diameter and pitch of rivets, are determined in part by practice and in part by a method of calculation to be explained later. In practice it is found necessary to limit the pitch of the rivets, and consequently the diameter, to be used with a given thickness of plate, in order that the joint may be made tight by calking. This limitation frequently makes the joint weaker than it otherwise would be. The edges of plates are either lapped over and riveted, or brought edge to edge and then joined by a cover-plate which is riveted to each of the two plates. The first method makes a lap-joint and the second a butt-joint. Fig. 75 shows a single-riveted lap-joint and Figs. y6 and yy show double-riveted lap-joints. The rivets in Fig. 76 are said to be staggered; the form shown by Fig. yy is called chain-riveting. Butt-joints with two cover-plates are shown by Figs. 80 and 81. The outer cover-plate is narrow, with rivets placed close enough together to provide for sound calking. The inner plate is wider, and as its edges are not calked they may have a row of more widely spaced rivets. These joints, and those shown by Figs. 78 and 79, are designed with the view securing more strength than can be had with a plain lap-joint like Fig. y6, or than can be had with a butt-joint with cover- plates of equal width. Efficiency of a Riveted Joint. — The strength of a riveted joint is always less than that of the solid plate, because some of the plate is cut away by the rivets. This is very evident in the case of a single-riveted joint, such as that shown by Fig. 75 ; it will be found to be true for more complicated joints, such as those shown by Figs. 80 and 81. The efficiency STRENGTH OF BOILERS. 1 89 of a riveted joint is the ratio of the strength of the joint to the strength of the solid plate. The strength and efficiency of a given riveted joint can be properly determined only by direct test on full-sized speci- mens, which have considerable width. Tests on narrow specimens are liable to be misleading. Tests on boiler-joints are expensive, and can be made only on large and power- ful testing-machines. Tests have been made on behalf of the United States Navy Department at the Watertown Arsenal on a large number of single-riveted joints, on a con- siderable number of double-riveted joints, and on a few special joints. A few tests have been made elsewhere on full- sized joints. These tests give us important information that can be used in designing joints for boilers, but we cannot in general select a joint directly from the tests. Methods of Failure. — A riveted joint may fail in one of several methods, depending on the proportions, such as thick- ness of plate and the diameter and pitch of the rivets. This can be clearly seen in case of a single-riveted joint like that shown by Fig. 75. Such a joint may fail: (1) By tearing the plate at the reduced section between the rivets. If the rivets have the diameter d and the pitch /, then the ratio of the area of the reduced section to that of the whole plate is - d P ' (2) By shearing the rivets. (3) By crushing the plate or the rivets at the surface where they are in contact. (4) By cracking the plate between the rivet-hole and the edge of the plate, or by some method of failure due to in- sufficient lap. A riveted joint never fails by this method in practice, because the lap can always be made sufficient. The failure of more complicated joints may occur in various methods, which will be considered in connection with the calculation of some special joints. igo STEAM-BOILERS. Drilled or Punched Plates. — In the better class of boiler- shops it is now the practice to drill rivet-holes in plates after the plates are in place, so that the holes are sure to be fair. Sometimes the holes are punched to a smaller diameter and then drilled out to the final size after the plates are in place. The result is the same as though the holes were drilled in the first place, as the metal near the hole, which was injured in punching, is all removed. The metal remaining between drilled holes does not have its properties changed by the drilling. On the contrary, the metal between punched holes is always injured more or less. In general, soft ductile metal is injured less than hard metal, and further, soft-steel plates are injured less than wrought-iron plates. When boiler-plates are punched and then rolled to form cylindrical shells, some of the holes are liable to come unfair, so that a rivet cannot be passed through. In such cases the holes should be drilled to a larger size, and a rivet of corre- sponding diameter should be substituted. Careless or reck- less workmen sometimes drive in a drift-pin, and stretch or distort the unfair holes so that a rivet can be forced through. The plate is liable to be severely injured by such treatment, and the rivet cannot properly fill the rivet-holes. Unfortu- nately it is difficult or impossible to detect bad work of this kind after the rivets are driven. Tearing". — The metal between the rivet-holes in a riveted joint cannot stretch as a proper test-piece does in the testing- machine, and consequently it shows a greater tensile strength than a test-piece from the same plate. Some tests on single or double riveted joints with small pitches show an excess of strength from this cause, amounting to ten per cent or more. The excess appears to be uncertain and irregular, so that if any allowance is made for it, it should be by a skilled designer after a careful, study of all the tests that have been made. Ordinarily it will be safer to use the tensile strength shown by test-pieces in the testing-machine, especially for joints like Fig. 78, which have a large pitch for some of the rivets. STRENGTH OF BOILERS. 1QI Shearing. — In general it is fair to assume the shearing strength of rivets of iron or steel to be two thirds the tensile strength of the metal from which the rivets are made. Crushing. — It is customary to assume that the pull on a riveted joint is evenly distributed among the rivets in the joint, and to divide the total pull by the number of rivets to find the shearing or crushing force acting on one rivet. It is further customary to assume that the intensity of the crushing force on the surface where the rivet bears on the plate, may be found by dividing the total force on one rivet, by the product of the diameter of a rivet and the thickness of the plate. The crushing-stress on rivets in joints that fail by crushing is found by experiment to be high and irregular. In some cases it has amounted to 150,000 pounds per square inch; in a few tests it is less than 85,000 pounds. It is probable that 95,000 pounds may be used with safety in calculating riveted joints for boilers. Now the stress on the bearing-surface will seldom be so much as one third the ultimate strength, even during a hydraulic test of a boiler, and it is not probable that a joint will be injured in this way unless the stress approaches the ultimate strength. Friction of Riveted Joints. — It is evident that there must be considerable friction between plates that are firmly clamped together by rivets driven hot. It has been proposed to take some account of this friction in calculating riveted joints, or even to allow the friction to be the determining element in proportioning riveted joints. Such a method is shown by experiment to be unsafe, for slipping takes place at all loads, beginning at loads that are much smaller than a safe load, and the effect of friction disappears before a breaking load is reached. Lap. — The distance from the centre of the rivet-hole to the edge of the plate is called the lap. The lap is usually once and a half the diameter of the rivet, a proportion that appears to be satisfactory. 192 S TEA M-B OILERS. Diameter of Rivet. — The minimum diameter of punched holes is determined by the consideration that the punch should not be broken. In the ordinary methods of punching boiler-plates the diameter of the punch should at least be as much as the thickness of the plate. It very commonly is once and a half the thickness of the plate. Drilled rivet-holes may have any diameter. They never have a diameter less than the thickness of the plate. The maximum diameter of rivet to be used with any kind of riveted joint will in general be determined by the considera- tion that the tendency to crush the plate in front of th-e rivet should not be greater than the shearing strength of the rivet. The maximum diameter thus found is liable to give too large a pitch. Pitch. — The maximum pitch for a given plate along a calked edge should be determined by the consideration that the plate should be held up rigidly enough to make a tight joint without excessive calking. The pitch of rivets, like those in the outer row of the joint shown by Fig. 78, need not be governed by this rule. There does not appear to be any explicit rule deduced either from practice or experiment for determining the proper pitch of rivets. Single-riveted Lap-joint. — In the joint shown by Fig. 75 let the thickness of the plate be /, the diameter of the rivet d> and the pitch p, all in inches. Let the tearing strength of the plate be f t = 55,000, the shearing strength be f s = 45,000, and the resistance to crushing be f c — 95,ooo, all for mild steel. Assume the proportions © ooc )©cj)©p , p , © c \-a—f ) © 1 l j Fig. 78. 1st. Tearing between outer row of rivets. The resistance is (P~d)tf t = 47,270. STRENGTH OF BOILERS. 1 97 2d. Tearing between inner row of rivets, and shearing outer row of rivets. The resistance is (P-2d)tf+~f s = 51,150. 4 Since the rivets are iron, f = 38,000. 3d. Shearing three rivets. The resistance is 4 4th. Crushing in front of three rivets. The resistance is ltdf e — 66,800. 5 th. Tearing at inner row of rivets and crushing in front of one rivet in outer roiv. The resistance is (P-2d)/ t + td/ c = 56,641. The strength of a strip of plate 3J inches wide is Itf t = 60,160. The least resistance is offered by the first method, giving for the efficiency 47,270 100 X 6o7T6o= 78.6 per cent. If the inside cover-plate is thinner than the shell, addi- tional complication will be introduced into the calculations for resistance. Double-riveted Lap-joint with Inside Cover-plate. — The arrangement of this joint is shown by Fig. 79. Assume the dimensions: Thickness of shell and of cover-plate, / = 7/16. Diameter of rivets (steel), d = 13/16. Pitch of rivets in lap, 2 T 5 f . Pitch of outer rows of rivets, P= 4|. I98 STEAM-BOILERS. The methods of failure are: 1st. Tearing at outer row of rivets. Resistance (P — d)tf t — 91,740. 2d. Shearing four rivets. Resistance f s = 93,310. I { )-k b © © ©A©X©-©_©_ © ®*© © © p ■ > c ^ph© © 1 Fig. 79. 3d. Tearing at inner row and shearing outer row of rivets. A strip having the width of the pitch of the outer row of rivets will be weakened at the rivets in the lap to the extent of one rivet-hole and half another rivet-hole. The resist- ance is Ttd' (P-i i d)tf + --f s = 105,285. 4th. Crushing in front of four rivets. Resistance 4tdf = 135,080. 5 th. Tearing at inner row of rivets and crushing in front of one rivet. Resistance (P - \\d)tf + tdf = 115,730. STRENGTH OF BOILERS. 199 Strength of strip 4$ inches wide, Ptf t = 111,290. 9 1 1 74° Efficiency = 100 X T T T „ „ = 82.4 per cent. J 11 1,290 ^ v Double-riveted Butt-joint. — The joint shown by Fig. 80 has a cover-plate inside and another, narrower, outside. There are two rows of rivets on each side of the joint. The inner rows are nearer together and pass through both cover- plates. Fig. 80. The outer row of rivets are wider apart and pass through the inner cover-plate only. The dimensions assumed are: Thickness of the plate and of both cover-plates, t = 7/16. Diameter of rivets (iron), 15/16 inch. Pitch of inner row of rivets, 2f. Pitch of outer row of rivets, 5|. There are five ways in which the joint may fail : 1st. Tearing at outer row of rivets. The resistance is (P-d)tf t = 103,770. 200 STEAM-BOILERS. 2d. Shearing tzvo rivets in double shear and one in single shear. If the plate pulls out from between the cover-plates shearing off the rivets, then the rivets in the inner row must be sheared through on both sides of the plate, or they are in double shear. The outer row of rivets are sheared at only one place. There are, consequently, five sections of rivets to be sheared for a strip as wide as the larger pitch. The resist- ance is — — f = 131,100. 4 3d. Tearing at inner row of rivets and shearing one of the outer row of rivets. The resistance is {P -2d)tf + n —f = 107,430. 4th. Crushing in front of three rivets. The resistance is ltdf = 116,880. 5 th. Crushing in front of tzvo rivets and shearing one rivet. The resistance is 7fd* 2tdf + -—f= 104,140. 4 The strength of a strip 5J inches wide is 5iX T \ Xf t = 126,560. The efficiency is 103,770 100- — 7 — ^— = 82 per cent. 126,560 r Triple-riveted Butt-joint. — The joint shown by Fig. 81 has three rows of rivets on each side. Two rows pass through both cover-plates, and the third or outer row passes through the inner cover-plate only. STRENGTH OF BOILERS. 201 The dimensions are: Thickness of shell, / = 7/16. Thickness of both cover-plates, t c = 3/8. Diameter of rivets (steel), d = 15/16. Pitch, inner rows, p = 3§. Pitch, outer row, P= 7}. Fig. 81. The joint may fail in one of five ways : 1st. Tearing at outer rozv of rivets. The resistance is (P-d)tf = 151,890. 2d. Sliearing four rivets in double shear and 07ie in single shear. The resistance is gird fs= 279.450. 3d. By tearing at middle row of rivets (ivhere the pitch is 3| inches) and shearing one rivet. The resistance is nd* (P-2d)tf+ --f s = 160,340. 202 STEAM-BOILERS. 4th. By crushing in front of four rivets mid shearing one rivet. The resistance is nd' 1 4dtf c +— /= 186,830. 4 5th. By crusJiing in front of five rivets. Four of these rivets pass through both cover-plates and will crush at the shell-plate. The fifth rivet passes through the inner cover- plate only, and will crush at that plate, since the cover-plates are thinner than the shell-plate. The resistance is 4 dtf + dt c f c = 189,170. The strength of a strip of plate 7 J inches wide is Ptf= 174,370. The efficiency is 151,890 100 X — = 87 per cent. 174,370 ^ Designing Riveted Joints. — One element of the design of a riveted joint is to secure as high an efficiency for the joint as is consistent with other requirements, such as a proper pitch for calking. A consideration of the example of a single-riveted lap- joint will show that the efficiency can be improved by increas- ing the diameter of the rivet and by increasing the pitch. In the first place, since the joint will fail by tearing between the rivets, simply increasing the pitch with the same size of rivet will give a greater efficiency. If the pitch is increased till the rivet fails, the failure will be by shearing. Now the resistance to crushing is represented by dtf ci while the resistance to shearing is represented by TteP 4 /iJ STRENGTH OF BOILERS. 203 that is, the resistance to crushing increases proportionally as the diameter, while the resistance to shearing increases as the square of the diameter. The shearing resistance increases the more rapidly, and can be made equal to the crushing resist- ance by using a larger rivet. Of course this will demand a further increase of pitch. In the case of the single-riveted lap-joint now under dis- cussion, the proper proportions for a joint that shall be equally strong against shearing, tearing, and crushing can be calculated directly. The usual way is to determine the diameter of the rivets by making them equally strong against shearing and crushing. Equating the expressions for crushing and shearing resistance, we have dtfc=—fn or d ~-f-< 4 J,7t For the case in hand with steel plates 7/16 of an inch thick, and steel rivets, the diameter will be 45,000 n " • /• Having the diameter of the rivets, we may now calculate the pitch by equating the shearing and tearing resistances, which gives ltd* , ~ , fs ltd* —fs = {p-d)tf t , or p= J -?- 7+dm 4 Jt ¥ For the case in hand we have 45,000 n 1. 17* '=55.ooo 4 X T V +I - I7 = 3 - 2 ' The efficiency of the joint is the ratio of the resistance to 204 STEAM-BOILERS. tearing between the rivets to the strength of a strip of plate having a width equal to the pitch, so that the efficiency is flp-d)t ^ p-d fspt P ' In the case in hand the efficiency is i 3-2 - 1. 17 100 3.2 63.4 per cent. But the pitch calculated in this method is too great for proper calking with a plate of the given thickness. The double-riveted lap-joint has three possible ways of failure, which lead to two equations for finding the diameter and pitch of rivets. Equating the shearing and crushing resistance for two rivets, we have 2—-f s = 2dtf ei or d = J ~ ^-^ 4 Is n which will give the same size rivet for a plate of a given thickness as would be found for a single-riveted joint. Now this method has been found to lead to too large a rivet for a single-riveted joint, where a strip having a width equal to the pitch carries one rivet. In the double-riveted joint such a strip carries two rivets, and consequently it is the more cer- tain that the method proposed will give too large a rivet, and of course too large a pitch for proper calking. The advan- tage of double riveting is that smaller rivets may be used to provide the requisite shearing resistance, and the plate may be less cut away at the section between rivets. In designing a double-riveted lap-joint it is customary to assume a diameter for the rivets and then determine the pitch by equating the shearing resistance of two rivets to the tear- ing resistance between the rivets. If the resulting pitch is too large for proper calking, the diameter of the rivets must be reduced. If, on the contrary, the resulting pitch is less than STRENGTH OF BOILERS. 205 may be allowed, a slightly larger diameter and pitch may be used. A design of a joint like the single-riveted lap-joint with inside cover-plate, which has a wide and a narrow pitch, involves some difficulty and complexity. The fundamental idea of such a joint is to make the resistance to tearing at the inner row of rivets (when the pitch is small) plus the shearing of the outer row of rivets greater than the resistance to tear- ing at the outer row of rivets (when the pitch is larger). To insure this condition we may proceed as follows: Equate the resistance to tearing at the outer row of rivets to the resist- ance to tearing at the inner row plus the resistance to shearing one rivet at the outer row. This gives (P - d)tf t = (/> - 2d)tf t + — /„ 4 whence The result is the minimum diameter of rivets allowable. We may now choose a slightly larger diameter of rivets, and then determine the pitch in three different ways, namely, by equating the resistance to tearing at the outer row of rivets, in succession, to the resistance to shearing of three rivets, to the resistance to crushing in front of three rivets, and to the resistance to tearing between the inner rows of rivets and compression before one rivet. The smallest pitch obtained will be the correct one to use with the given diam- eter of rivet. Should the efficiency of the joint be unsatis- factory, an attempt may be made to raise the efficiency by increasing the diameter of the rivets. Practical Considerations. — In proportioning a riveted joint, the following considerations, some of which have already been mentioned, must receive attention: The pitch of rivets near a calked edge must not be too great for proper caulking. / 206 S TEA M-B OILERS. Rivets must not be too near together for convenience in driving. Punched holes must have a diameter greater than the thickness of the plate. A riveted seam must contain a whole number of rivets. Again, it is desirable that similar seams, as for example the longitudinal seams for the several rings of a cylindrical boiler, shall have the same pitch. It is evident that the design of a boiler-joint cannot be considered apart from the general design of the boiler. Flues. — The tendency of internal pressure in a thin hol- low cylinder is to give it a true cylindrical shape; conse- quently, with fair workmanship, the formulae for thin hollow cylinders may be applied to the calculation of boiler-shells subjected to internal pressure. But the tendency of external pressure is to exaggerate any imperfection of shape, and cylindrical flues fail by collapsing. The pressure at which a flue will collapse can be found by direct experiment only. The earliest and for a long time the only tests on the collapsing of flues were those made by Fairbairn, and pub- lished in the Transactions of the Royal Society, in 1858. All of the tubes tested were 0.043 °f an mcn thick; they varied in diameter from 4 inches to 12 inches, and in length from 20 inches to 60 inches. From these tests he deduced the em- pirical formula 806,300 X / 2I9 P /X d in which / is the length of the tube in feet and d and / are the diameter and thickness in inches, while/ is the collapsing pressure in pounds per square inch. Sometimes the exponent of / is made 2 instead of 2.19, for sake of simplicity. As /is commonly a proper fraction, the use of a smaller exponent will give a higher calculated collapsing pressure. STRENGTH OF BOILERS. 20/ The tubes in this series were too small, and more especially too thin, to serve as a proper basis for the calculation of boiler- flues. It is quoted because it has been widely used, and is now used by some engineers. It sometimes gives a calculated pressure higher and sometimes lower than that at which flues will collapse, and its use is liable to lead to disappointment if not to disaster. The following table gives the results of some tests on larger boiler-flues, taken from Hutton's " Steam-boiler Con- struction." The table gives the dimensions and the actual collapsing pressure, and also the collapsing pressure by Fair- bairn's rule and by a rule proposed by Hutton. EXPERIMENTS ON THE COLLAPSING PRESSURE OF BOILER- FLUES. Where or by Whom Made. By Fairbairn By Fairbairn By Fairbairn By Fairbairn ... Engineering Dept., U. S. N. At Greenock By Knight By Knight By Kntght By J. Hovvden & Co., Glas- gow Di mensions. Collapsing Pressure in Pounds per Square Inch. Q-5 — c rt— < 5 c v z c Xi u c c <" c C — .* .a «» •° a C V irS re .b culuted by tton's Rule. X V v a js'O O X re 3 W u J £ H fn feW Ufa UX 2 3 4 5 6 7 7.87 276 5 no log 114 33-5 360 11 99 Si 113 42 420 12 97 7S IOO 42 300 12 127 10S ng 54 36 8 128 3ii 120 38 86 16 450 740 43b 36 24- 8 235 700 21S 36 24 12 468 1568 490 36 48 12 390 784 35o 43 23 17 840 275S 842 On the whole the rule proposed by Hutton gives the most concordant results; in most cases Hutton's rule gives a cal- culated collapsing pressure that is smaller than the actual collapsing pressure; in no case is the calculated result very largely in excess. Fairbairn's rule in some cases shows a very 208 STEAM-BOILERS. close agreement with experiment, but in others it shows a dangerous excess. Hutton's rule is Cf in which / is the length in inches, d is the diameter in inches, and t is the thickness in thirty-seconds of an inch. C is a con- stant which Hutton makes 600 for iron and 660 for steel. Mr. Michael Longridge, as a result of an investigation of many boiler-flues, most of which have endured service for years, but some of which failed, gives a rule in the same form but with a constant 540 instead of 600. For oval tubes and flues it is recommended that the above rules be applied, using for the diameter twice the maximum radius of curvature. Strengthened Flues. — It is clear from inspection of the preceding table of tests on boiler-flues that the collapsing pressure decreases as the length of the flue increases. Account is taken of this in Hutton's formula, by introducing the square root of the length into the denominator of the expres- sion for calculating the collapsing pressure of a flue. Stating the proposition in the converse manner, the reason why a short flue is the stronger is that the ends of the flue are kept in a circular form by the plates to which the flue is riveted. It has been customary to strengthen plain flues by the aid of rings placed at regular intervals. The section of a ring made of angle-iron is shown by Fig. %2a. The ring is riveted to the flue at intervals, a thimble being placed over each rivet to give space for circulation of water between the ring and the flue. The rings were sometimes solid, made of one piece of angle-iron bent up and welded. Most frequently the ring was in halves, which were merely belted together at the joint. Such rings could be easily removed when the flue was taken out of the boiler. A better method of strengthening a flue is to make it of STRENGTH OF BOILERS. 209 short pieces so joined at the ends as to make stiffening rings. Fig. 82 shows three ways in which this can be done. At b is shown the Adamson ring, formed by flanging the edges of the short lengths of flue outwardly, and riveting through a welded iron ring. At c is shown a welded ring of T iron, to which the short lengths can be riveted without flanging. This Fig. 82. method provides for calking both inside and outside. It does not require the flue to be flanged; but flanging by machinery is rapid, and does not give trouble when good iron or steel is used. Material that will not stand flanging should not be used for flues. At d is shown the bowling hoop-ring, which has the advantage that it provides for longitudinal expansion of the flue. Flues for Scotch and other marine boilers with furnace- flues, are stiffened by transverse or helical corrugations, which provide at the same time for longitudinal expansion. A number of methods of corrugating furnace-flues will be illus- trated in connection with tests given on the following pages. Tests on Furnace-flues. — The strength of corrugated and other stiffened flues can be determined only by tests on full- sized specimens. The following tests are taken from a paper by Mr. B. D. Morison, read before the Northeast Coast In- stitution of Engineers and Shipbuilders. 2IO S TEA M-B OILERS. Furnaces made with Adamson Joints. Tests made at the Works of Hall, Russell & Co., Aberdeen, in 1882, and of J. Howden & Co. in Glasgow, in 1887. "ii 11 11 11 ir Date of Test. I8S2 1887 Length Furnace. 6 ft. 5! in. total length. Length of rings : 18*". 19", 19", and 20" 7 ft. •§• in. total length. Length of each ring, 23" 1/ l> c >, tft 2S rt u s t! tS c . _ as 5.8* Qu c- 1 a § w 1st ring 1" 2 » 2d ring 1 5" 3a » 3d ring 43 9/64 15'' 32 ' 4th ring i" ¥ 43.09 3d ring at 700 1st ring at 840, 2d ring at 760, 3d ring at 840, 4th ring at 835 SE^' ^2S 8 + S?H u q 3 N taX ^- . ■S\ c!!o« rt s lap ent ee ens O w vc/)H U CJ 64,213 61,918 64,240 6i,945 Note. — No record of tensile strength of steel ; 28 tons per square inch assumed. The collapsing coefficients are calculated on external diameter of furnace over plane part. STRENGTH OF BOILERS. 21 I ON OO *-" T3 C cd o" ON 00 i— < cd r/) vO vO J 89 ,.to O vO CN O — oo j-- c u a; £Iz;u»« M M H O O O 00 CO j«H C 3 214 STEA M-BOILERS, 2 CO o U o 3-4 o O d u d +i Of co > U 0* s en H a o 2 * 2 o. «° = o o •ajisuax suoj Lz jo pais °J paonpaj •JJ303 Suisd^noo C - .Q X3 T3 to J3 « ' Z 5 s - - 4J OJ 8 " w 02 02 tu en mco meinn O OO f-> CN MQO Tf f» NO CO^h OO en r->0 in t>*0 r-^r-^co 'X -*■ a * d 1U313 O Tf-vo ON N e>>en enco hh r^ o O O N ■ainssajj SuisdB[[of) UI 9DU3J3JJIQ 1S3}B3J£) •saqoui ui si^U J3AO jajaraBiQ UB3J^ m m on q\ i-i en •pug spBg co w O en en O r->. ttnn moo a o O in m tn n vO •3IPPJIM cm rtO n ma>fioo co moo O enco en en m in o» en m ino O •pug juojjj m rt o^i^ci O eno O »i- o c^i co moo r>. mind n -t m in in " P"3 ^M jo qiSuai :s3?B3jf) •SUOIJB3 -nJJ03 jo jaquirtN OBuang jo qiSuaq njtemjeo •HH<-lHfH|O>i-l|a>l-l|0iHC>*4|N r-lfil tn co tN r^ r^ ^ ts r> f^ r^co co co co co co oooooo^c o •1S9J, JO 3JBQ co co co co - - co co co co co co n n oo co oo oo oo oo STRENGTH OF BOILERS. 215 Purves's Patent Furnaces. Official Tests made at Sir John Brown & Co. 's Works at Sheffield in 1889. ^^*^^ V Date of Test. 5-3 IT. flS Dec. 23, 1890. 9i 9l 9l Qii l/l u j. ► jC O <-> C za u a H rt cO- 6 rt CD •a c a w -2 5 ~ O rt is rt S E H 'a •d B c *o u to c u 2 V o r g a 4_l > J= w u 03.2 o ' ) c o c o CO T3 72 c: tx O P ition of 4> bJ5 > v- « b ■o 02 W " U9J^ suoj Lz p CO 00 CO O o in in IN paiS °1 paanp ,_j o 3.rpo;ysdB[[03 CO co CO CO CO CO •pais jo qjsuajis 3 1 IS -r -t- co in in o -uax ajBoiuin CI CJ CJ C4 O r^ CO co O r^ "■Z+ffXrf in vO vO in in CO in 7303 •sdB ll o3 co oi O O in O co CO co CO co CO "3inSS3JJ o> o o O n O ir. O o O SuisdBiio3 !> o O CO co JJBjjAuB IB J3}3 co CJ yu to Cl ^ -U1BIQ UI 3DU3 *-•*. CO C7> ;> ■^ co •J9jtf!a ;S31E3JQ H w M m CO r-» O in " in co vO § C n co IH M »* in U-O -t- vO in Tt" vO u P3&3 en co <=*■ Tf in in >*-! M Cl CO vO CO O 13 o o I/) r^ •a CO CO ^1- in in § n _ LO N ir> vO -r •*f u-> CO in CO co -r Tf u-> in spug IB iei^ p eoK* H« rtbi «H. eshc U;3U3iaS3?B9J9 -r T -sT ^ ^r in suo.ueJ8ru.103 o o f> CT« o O jo jaqain^j H|eu H=o H»> «hr ^iM IT) r~» in O r^ r^ -jnj p qi§u3T : o o vO vO o \b O V H I in C4 d 6 d d d o Q Q Q Q Q u a, rt a; Q 1/2 *-. v STRENGTH OF BOILERS. 2\J Discussion of Results of Tests on Flues. — The stress in a thin hollow cylinder subjected to external fluid pressure may be calculated by an equation having the same form as that for a cylinder subjected to internal pressure; the equa- tion may be deduced by a similar method. Thus the stress will be S t' in which / is the pressure per square inch, r is the radius and / is the thickness, both in inches. In the table we have a column giving the coefficient of collapse calculated by the expression PD T' in which P is the pressure, D is the diameter, and T is the thickness. The coefficient appears consequently to be twice the compressive stress in the flue at the time of collapsing. This coefficient is fairly regular for each style of furnace, and is somewhere near the tensile strength of the metal from which the flue is made; in some cases it is less and in some more than the tensile strength. Now soft steel in the form of short cylinders will begin to flow when the compressive stress in a testing-machine is about equal to the strength of pieces used for tensile tests. In other words, the tensile and compressive strengths are about equal. The furnaces tested appear, then, to have collapsed when the compressive stress was half the ultimate compressive strength of the metal. Now the limit of elasticity for both tension and compression, for soft steel, is about half the ultimate strength, so that the collapse occurred somewhere about the elastic limit. We should not, however, attribute too much importance to this considera- tion, but it will be better to follow ordinary practice and consider the equations used for calculating the safe working 2l8 STEAM-BOILERS. pressure on flues to be empirical, and to depend directly on experiment. Rules for Working Pressure on Flues. — There are three sets of rules for working pressure on flues, which we shall consider; namely, those of the British Board of Trade > those of Lloyd 's Marine Insurance Underwriters, and those of the United States Inspectors of Steam-vessels. These rules are changed from time to time, and include certain directions to inspectors that need not be given here; if a boiler is built for inspection under these or any other rules the only safe way is to obtain the current edition of the rules and see that the boiler conforms thereto, and also that the boiler is properly proportioned according to the best information that can be obtained by the designer. Rules for Plain Flues. — Both Lloyd's and the United States Inspector^ rules use for plain flues an equation in the form 89,600 X r a *- LD in which P is the working pressure in pounds per square inch* L is the length in feet, and T and D are the thickness and diameter in inches. This is Fairbairn's equation with 2 instead of 2.19 for the exponent of T, and with a constant 806,300 89 , 600 = , so that the working pressure is made one ninth of the calci lated collapsing pressure by Fairbairn's rule. The use of so large a factor as nine shows that the rule is not considered adequate. Flues designed under this rule will probably be strong enough. The Board of Trade rule differs only in replacing the STRENGTH OF BOILERS. factor 89,600 by the approximate figure 99,000. The rules, however, require that the pressure shall not be greater than _ 8800 X T P ~ D ' which provides that the stress shall not exceed 4400 pounds per square inch. For corrugated, ribbed, or grooved furnaces (such as the several furnaces for which tests are given) both the Board of Trade and the Inspectors rules give for the working pressure 1400 X T P = D in which P is the working pressure in pounds on the square inch, and Zand D are the thickness and diameter in inches. This rule makes the working stress 7000 pounds per square inch. Lloyd' s rule for these furnaces is given by the equation QT-2) D ' in which T is the thickness in sixteenths of an inch, D is the diameter in inches, measured over the corrugations or ribs of corrugated or ribbed furnaces, and over the plain part of Holmes' furnaces. C is an arbitrary constant having the fol- lowing values: C = 1000 for steel corrugated furnaces when the tensile strength of the material is under 26 tons, and corrugations are 6 inches apart and \\ inches deep. C= 1259 f° r steel furnaces corrugated on Fox's or Mori- son's plans, tensile strength to be between 26 and 30 tons. C = 1 160 for ribbed furnaces with ribs 9 inches apart. C= 912 for spirally-corrugated furnaces. , C = 945 for Holmes' furnaces, when corrugations are not over 16 inches apart and not less than two inches high. 220 STEAM-BOILERS. In this rule the use of T — 2 (in sixteenths of an inch) instead of T is practically an allowance for wasting of the plate to the extent of one eighth of an inch. The working- stress calculated on the assumed diameter will be found by multiplying by sixteen and dividing by two; in case of the first constant the stress is iooo X 16 = 8000 2 pounds per square inch. Fire-tubes. — The thickness usually given to fire-tubes to insure sound welding and to provide for expanding into the tube-sheets is in excess of that required to prevent collapsing. There appears, however, to be no experiments to show the actual collapsing pressure for such tubes. The joint made by expanding the tubes into the tube- sheets of locomotive and cylindrical tubular boilers has been found both by experiment and practice to be strong enough to secure the tube-sheet without additional staying. It is, however, the custom to make part of the fire-tubes of marine drum-boilers thick enough to take a shallow nut outside of the tube-plate; without such stay-tubes there is liable to be leakage at the ends of the tubes. Girders — When a flat surface cannot conveniently be stayed directly, it is customary to stay the surface to girders properly supported at the ends or elsewhere. The crown-bars of the locomotive-boiler shown on Plate II, and the girders over the combustion-chamber of the marine boiler shown by Fig. 13, page 17, may be taken as examples. Again, the channel-irons which are riveted to the flat heads of the cylindrical boiler shown by Plate I act as girders. The load which a girder of given material can safely carry depends on the form and dimensions of the girder, and on the manner of supporting and loading the girder. Some girders, like those over the combustion-chamber in Fig. 13, can be STRENGTH OF BOILERS. 221 calculated by the simple theory of beams; others, like crown- bars for locomotives and the channel-bars on Plate I, can be properly calculated only by the theory of continuous girders. A proper understanding of the theories of beams and of continuous girders can be obtained from standard works on applied mechanics. An adequate statement of even the theory of beams is out of place in a work on boilers; an incomplete statement is unadvisable, since it is liable to be misleading. One simple example will be worked out as an illustration of the use of the beam theory in boiler-design. As an example, we will take the girders over the combus- tion-chamber of the marine boiler shown by Fig. 13, page 17. The girders are spaced 7 inches apart, and each carries three stays spaced 6j inches apart. The load on each stay-bolt at 160 pounds steam-pressure is 7 X 6\X 160 = 7000 pounds, and the total load on one girder is 21,000 pounds. The sup- porting force at each end of the girder is 10,500 pounds. The span of the girder is 22^ inches, and the half-span is 11J inches. The bending-moment at the middle of the girder due to the supporting force acting upward, and to the load on one bolt acting downward, is 10,500 X 1 4 — 7000 X 6J = 74,375 = M. Each girder is made of two plates each 5/8 of an inch thick, and 7 inches deep. The moment of inertia of the section of the girder at the middle is T V X 2 X i X 7 3 = /• The distance of the most strained fibre is 7-^2 = 3^=^. 222 STEAM-BOILERS. The working fibre-stress is consequently My _ 74 375 X 3^ _ „ /_ / - T yx2xf x 7 »- 7257 pounds per square inch. Stayed Flat Plates. — The method of calculating the stresses in a flat plate supported at regular intervals by stays or stay-bolts, such as the sides of a locomotive fire-box, is treated in the theory of elasticity, under the heading of " indefinite plates which are firmly held at a system of points dividing them into rectangular panels. ' ' A complete solution of this problem is possible only when the panels are squares, that is, when the rows of stays are equidistant longitudinally and transversely. If the steam-pressure is represented by fi, the thickness of the plate by /, and the pitch of the stays by a, then the maximum direct stress, which is a tension at certain places and a compression at other, is given by the formula 2 a* The maximum shearing-stress is given by the equation in which .E is the modulus of elasticity of the material. If the sheets of a locomotive fire-box, or other stayed pistes, have a direct tension or compression, proper allowance must be made for it. If stays or stay-bolts are in rows that are not equidistant each way, as for example the through-stays in the steam- space in Fig. 13, page 17, then the largest pitch may be used in the above equations. The actual stress will in such case be less than the calculated stress by an unknown amount. If, STRENGTH OF BOILERS. 223 further, stays are arranged irregularly, the greatest distance in any direction may be used in the equations, but the calcu- lated stress may then be very different from the actual stress; it is, however, always the larger. As an example, we may calculate the stress in a side sheet of the locomotive fire-box shown on Plate II. Here the rows of rivets are four inches apart each way, the plate is 5/16 of an inch thick, and the steam-pressure is 170 pounds. The maximum stress is The shearing-stress in this case is very much smaller. Now the crown-bars are bedded on and are partly sup- ported by the side sheets of the fire-box. The crown-sheet is 72 inches long and 45! inches wide, and has an area of 72 X 45! = 328s square inches, and is subjected to a pressure of 3285 X 170 = 558,450 pounds. The distribution of this load between the side sheets and the sling-stays can be determined only by the cal- culation of the crown-bars as continuous girders, and may be disturbed by the expansion of the fire-box and by other causes. If it be assumed that the side sheets carry half the load on the crown-bars, then one side sheet will carry one fourth of 558,050, or 139,512 pounds. The side sheet is ^2 inches long and 5/16 of an inch thick, so that the stress per square inch from the load on the crown-bars is 139,512 -f 72X A = 62 °° pounds, — about as much as the stress calculated above. The 224 STEAM-BOILERS. total compression on the side sheet is therefore about 12,400 pounds per square inch. This calculation, which appears sufficiently simple, illus- trates the danger of making calculations by formulae without knowing how they are derived and how they should be applied. The formula for staying given above is often quoted without any reference to tensile or compressive stress on the stayed sheet, from other causes; the use of such a formula by one who is unfamiliar with the theory of elasticity may lead to serious error in design. Factor of Safety. — The ratio of the working pressure of a boiler to the pressure at which the boiler or any part of a boiler may be expected to fail quickly, is called the factor of safety for the boiler or for that part of the boiler. It is commonly recommended by writers that a factor of safety of six shall be used for boilers; probably such a factor would be economical for a boiler that is expected to work continuously for many years, as it allows a margin for deteri- oration. If the stresses coming on the parts of a boiler can be determined, a general factor of five will give sufficient security. If the boiler is carefully watched, a factor of four may be used ; many boilers are worked with this factor. The use of an excessively large factor of safety, for example of the factor nine for flues calculated by Fairbairn's equation, shows a lack of confidence in the method. It is proper to make allowance for corrosion of parts like stays: this may be done either by using a larger factor of safety, or by a direct allow- ance; thus all stays, whatever their diameters, may have an eighth of an inch added to the diameter to allow for corrosion. It is of course proper in any structure to make small but im- portant members, such as stays in boilers, large enough to place them beyond any suspicion of failure. Hydraulic Tests of Boilers. — It is customary to subject new boilers to a water-pressure considerably in excess of the working pressure, to discover any leaks at riveted joints, at STRENGTH OF BOILERS. 22$ the tube-sheets, or elsewhere; should there be any gross defect of design or workmanship it will be developed by this hydraulic test. Old boilers after repairs are subjected to a hydraulic test for the same purpose, but the pressure is not carried so high as for new boilers. The pressure applied during a hydraulic test is seldom more than once and a half the working pressure, and as most boilers have an actual factor of safet/y of not more than rive, and frequently of four, it is apparent that the recommenda- tion of some authors, that the test pressure should be twice the working pressure, cannot ordinarily be followed without danger of injuring the boiler. With a factor of safety of six there should be no danger of injuring the boiler by applying a hydraulic pressure equal to twice the working pressure. It should be borne in mind that some of the worst stresses that come on the different parts of the boilers are due to unequal expansion and contraction, and that such stresses are not set up during a hydraulic test. Finally, the fact that a boiler has successfully withstood a hydraulic test is not a con- clusive proof that it is safe; too many unfortunate explosions of boilers, more frequently old boilers, after a hydraulic test, have shown this. The safety of a boiler is to be insured by careful and cor- rect design, honest and thorough workmanship, and intelli- gent care in service. Forms and methods of design and construction that do not admit of ready calculation should be avoided ; in no case should the ordinary hydraulic test be relied upon to guarantee the strength of' parts that cannot be calculated with a fair degree of certainty. If such forms are used in any case, they ought to be tested separately to a pressure of two or three times the working pressure, and some examples of each form and size ought to be tested to destruc- tion. The boiler undergoing a hydraulic test should be carefully inspected, and any notable change of shape or leakage should 226 S TEA M-B OILERS. be investigated to discover the cause. Frequently small leaks that are developed during a test are stopped at once by calking or otherwise, but it is preferable to mark the place of the leak and calk after the pressure is removed. This of course requires another test to find out if the calking is suc- cessful. The pressure is usually applied by filling the boiler entirely full of water and then pumping in enough water, by hand or by power, to supply the leaks and develop the pressure required. If the pumping is done by hand, it is desirable to carefully remove all air from the bo'ler to avoid the labor of compressing air up to the test pressure. If the pumping is done by power, the saving of work is of less consequence, and a little air remaining in the boiler will act as a cushion, and lessen the shocks due to the strokes cf the pump. New boilers are tested on the boiler-shop floor; old boilers are commonly tested in their settings, and in such case the inspection during a test is less convenient and efficient. It is sometimes recommended that hot water shall be used for testing a boiler; but there seems to be no advantage in so doing, as it is unequal expansion, and not merely rise of temperature, that sets up the unknown stresses that are so destructive to the boiler. Of course the use of hot water makes an efficient inspection during the test difficult if not impossible. When there is no other way of applying the hydraulic test to a boiler in its setting, the boiler may be quite filled with water, and then a light fire may be started in the furnace. The expansion of the water will develop the required pressure at a much less temperature than that of steam at the same pressure, and with less danger should the boiler fail. This method cannot be recommended for general use; and in case it is followed care must be taken not to exceed the desired pressure. STRENGTH OF BOILERS. J2/ Hydraulic Test to Destruction. — In 1S88 a boiler-shell, made to represent a part of the shell ot a gunboat boiler, was tested by hydraulic pressure at the Greenock Foundry,* with the intention of bursting it. The shell was 1 1 i'eet long and 7 feet 8 T 3 <- inches mean diameter. It was made of three sec- tions of 19/32 plate, triple-riveted, with butt-joints and double cover-plates at the longitudinal joints, and lapped and double riveted at the ring seams. The rivets were staggered for both longitudinal and ring seams. The end-plates were 20/32 thick, and stayed with through-stays and washers, spaced 14 inches on centres. The stays were ij inches in diameter; the screws at the ends of the stays were 2] inches in diameter. Finally, it may be said that the shell was designed to fulfil the Admiralty specifications for a working pressure of 145 pounds per square inch. The workmanship was of the same degree of excellence usual for boiler-work at that establish- ment. First Test. — The shell was first subjected to the working pressure of 145 pounds, and showed a slight alteration of form due to the tendency of internal pressure to give it a true cylin- drical form. The pressure was then raised to the Admiralty test pressure of 235 pounds, and then to 300 pounds without developing leaks. There were some minor changes of form due to the increase of pressure. The pressure was then removed and the shell returned to its original dimensions. Pressure was then raised to 330 pounds, when there was a slight leak at the manhole door. At 450 pounds pressure the leak at the manhole door exceeded the capacity oi the pumps. There was also a slight leak at the corners ot two butts. The manhole was then strengthened — no other repairs were made. Second Test. — Pressure was raised to 350 pounds and developed a small leak at the manhole. There were slight * Trans. Inst. Naval Arch., vol. xxx. p. 285. 228 S TEA M- BOILERS. leaks at the butt-straps, which were calked at the end of the test. The manhole, however, leaked so that the test was stopped. Third Test. — After additional bolts were put into the manhole cover the pressure was raised to 350 pounds with- out leakage. At 360 pounds the manhole began to leak, and at 580 pounds the test was stopped on that account. The butt-straps opened visibly at the calking and leaked more than before. Fourth Test. — The butt-joints were again calked and additional pumps were employed. The shell was again tight at 350 pounds and the pressure was carried to 620 pounds, at which there was a good deal of leakage at the butt-straps. Only one or two rivets showed signs of leakage; there appeared to be no difference between the hand and machine riveting in this respect. At the pressure of 620 pounds the entire capacity of the pumps was required to supply the leakage. . The distortion of the shell was very marked at the higher pressures, and increased with the pressure; thus the ends- bulged an inch at 520 pounds, about \\ inches at 580 pounds, and nearly two inches at 620 pounds. The sides bulged more irregularly, but to the extent of nearly an inch at 620 pounds. The stays drew down uniformly 1/64 of an inch at 520 pounds, 2/64 at 580 pounds, and 4/64 at 620 pounds. They increased in length 2^ inches at 520 pounds, 3J inches at 580 pounds, and 3f inches at 620 pounds; this accounts for the bulging of the end-plates. The mean tensional strength of the plates from which the shell and butt-straps were made may be taken at 61,500 pounds. At 620 pounds the tension on the plates between the rivet-holes was ^7,504 pounds, or 93 \ per cent of the strength oi the ^olid plate, and there was no serious disturb- ance of the structure. The ring seams increased in diameter about f of an inch, and the shell bulged out between them. STRENGTH OF BOILERS. 220, The various portions of the boiler acted in harmony and showed no special weakness at any point. The butt-joints had the rivets spaced 5 j- inches on centres to give a percen- tage of 83.7 per cent of the plate, and this may have caused the leakage found there. The riveting appeared to be reliable at the extreme pressure reached. This test seems to show that a boiler will give signs of weakness long before it will fail. Such signs of weakness should be carefully investi- gated : if there is any local weakness or deterioration, repairs or alterations may be made; if there are evidences of general deterioration, the working pressure must be reduced, or better, the boiler may be replaced by a new one. Boiler-explosions. — The great destruction of life and property that is liable to be caused by a violent boiler-explo- sion makes it imperative that the causes should be carefully investigated, to the end that explosions may be prevented. With this in view the boiler and its parts, and any wreck or evidence of destruction caused by the explosion should be left undisturbed until the scene of the explosion can be examined by a competent engineer. Of course if any persons are injured by the explosion they must be rescued and cared for immediately, and also any building or structure that is so injured as to threaten life or safety must be attended to at once; but it should be borne in mind that the examination by the engineer for the purpose of determining the cause of the explosion is also in the interest of humanity, since its aim is to avoid future explosions. All idle or simply curious per- sons should be excluded from the scene of the explosion, more especially as such persons are apt to disturb or even carry away things that may be of importance ?.n the study o> the cause and history of the explosion. If the explosion is accompanied by loss of life or injury to person or property, it will be followed by a le^al investigation in which the testi- mony of the engineer or engineers who have examined the scene of the explosion will be of prime importance, as it will 23O STEAM-BOILERS. have a large influence in locating responsibility for the disaster. While various causes may lead to boiler-explosion, it is unfortunately true that by far the greater part of violent explosions are due to the fact that the boiler is too weak to endure service at the regular working pressure. A new boiler may be weak through defective design or workmanship; there can be no excuse for the explosion of a new boiler from weakness, and such explosions in good practice are rare. An old boiler is liable to become weak through local or general corrosion or other deterioration; this amounts to saying that a bciler will eventually wear out. The length of time that a boiler will endure service depends (i) on the design, (2) on the thickness of plates and the quality of the metal, (3) on the workmanship, (4) on the care given it, and (5) on the quality of the feed-water. Definite figures cannot be given for the life of a boiler, since it depends on so many things. The following table gives the number of years several kinds of boilers can endure regular service if they are properly built and cared for: Lancashire, low-pressure 1 5 to 20 years. Locomotive type, stationary 12 to 1 5 Locomotive-boilers , 8 to 12 Vertical boilers 10 to 1 5 Vertical boiler with submerged tubes... . 14 to 18 Horizontal cylindrical tubular 15 to 20 Scotch marine boiler 12 to 1 5 Water-tube boiler 12 to 16 Pipe or coil boiler 5 to 8 By water-tube boiler is here meant a boiler with a shell or drum containing a considerable body of water. By pipe or coil boiler is meant a boiler made up of pipe and pipe- fittings, with a separator. STRENGTH OF BOILERS. 23 j. Horizontal boilers will require one, and vertical boilers two extra sets of tubes, before the shell is condemned. A loco- motive-boiler will require two extra sets of tubes, and the entire fire-box will be renewed once in the life of the boiler. If boilers are subjected to careless or ignorant abuse, they may be used up in a fraction of their proper time of service, especially if cheaply built. This will account for the numer- ous explosions of sawmill boilers and agricultural boilers. It has been pointed out that leakage is frequently a sign of weakness; a perversion of this idea leads to the assumption that a boiler is safe as long as it can be kept from leaking. Too many boiler-explosions have this history: The boiler, after long and satisfactory service, began to leak; a cheap man was employed to repair the boiler, the repairs consisting mainly of excessive calking to stop the leaks; soon after the repairs, perhaps the first time the boiler was fired up, it exploded violently. A fit conclusion of the history is to ascribe the explosion to some obscure cause or to carelessness of the attendant, if he was killed by the explosion. Serious injury may be caused by overheating any part of the heating-surface, due to low water, to defective circulation, or to deposits of non-conducting substance on the plates or tubes. The overheated member, or plates, of the boiler may burst or collapse, and such failure may lead to an explosion of the boiler, but frequently the escape of steam and water will check the fire and relieve the pressure on the boiler. Local failures are dangerous to the boiler attendants, especially in a confined fire-room, as on shipboard. Unless there is direct evidence of overheating, either from known circum- stances before the explosion or from signs on the boiler after explosion, the cause of the failure should be sought elsewhere. If a boiler shows siems of low water or of overheating the fire should be checked by any effectual means. The most ready way of checking the fire is to close the ash-pit doors and throw ashes onto the fire. If there are no ashes at hand, then 232 STEAM-BOILERS. fresh fuel may be used instead, since its first effect is to deaden the fire. There will be time for caring for, or drawing the fire before the fresh fuel is fairly in combustion. An attempt to draw the fire without first deadening it is liable to give a fierce combustion for a short time; moreover, more time is required to draw the fire. If the furnace has a dumping-grate, the fire may be immediately thrown into the ash-pit without waiting to deaden it. The damper should be left open so that if a rupture occurs the steam may escape up the chimney. Mean- while the steam made by the boiler should be disposed of by allowing the engine to run or by any other means, for exam- ple by opening the safety-valve, provided that it is merely a case of overheating, not accompanied by excessive pressure. It will probably be well to start the feed-pumps or to increase the supply of feed-water. Should the introduction of feed- water be badly arranged so that a large volume of cold water will be thrown onto a heated plate, it is possible that starting the feed-pump may cause a contraction which will start a rupture. It has been found by experiment that boiler-flues that have been purposely allowed to become bare and overheated have been saved by suddenly directing a stream of cold feed- water upon them, though such treatment may make them leak at the joints. The heat stored in such hot plates is insignificant as compared with the heat in the water and steam in the boiler. Excessive pressure, especially if it is enough to give good reason to fear an explosion, is more difficult to deal with; the chances of success are less and the risks are greater than when the water is low, but the pressure is not excessive. If possi- ble the fire should be checked and the pressure relieved. The first may be done by throwing on ashes or cold fuel, and the second by running the engine at full load. It is at least doubtful whether starting the feed-pump will reduce the pressure fast enough to do much good, and on the other hand STRENGTH OF BOILERS. 233 there may be cases where such action would start an explo- sion. It is not best to open the safety-valve, since the sudden opening of a large safety-valve gives a shock which may determine the explosion. Some explosions have been re- ported that occurred immediately after the safety-valve opened. A large amount of energy is stored in the steam and water in a boiler in the form of heat. An idea of the amount of energy in any given case may be obtained by a simple calcu- lation. Thus the cylindrical boiler shown on Plate I, at 150 pounds pressure by the gauge, will contain 6600 pounds of water and 22 pounds of steam. Taking 165 pounds absolute to correspond to 150 pounds by the gauge, we find from a table of the properties of steam that 338 thermal units are required to raise one pound of water from freezing-point to 366 F., corresponding to 165 pounds absolute. Now one thermal unit is equivalent to 778 foot-pounds of work. Con- sequently the energy stored in the hot water in the boiler, calculated from freezing-point, is 6600 X 778 X 338 = 1,736,000,000 foot-pounds. After the water is heated to 366 F. there will be required 855.6 thermal units to vaporize one pound into steam at 165 pounds absolute. But 83.6 thermal units will be expended in changing the volume of the fluid when it passes from water into steam, leaving 772 thermal units for the internal heat of the steam. Consequently the heat stored in a pound of steam is 772 -\- 338 thermal units. The equivalent energy stored in 22 pounds of steam is 22 X 778 X (772 + 338) = 19,000,000 foot-pounds. The first point to be noticed is, that there is many times as much energy in the water as in the steam; and the second is, that even a small fraction of the stored energy is suffi- 234 STEAM-BOILERS. cient to account for all the destruction caused by a boiler- explosion. The circumstances of the boiler-explosion will determine how much of the energy stored in the steam and hot water will be developed and how it will be applied. Even in a par- ticular case it is seldom possible to make proper estimates, nor does there appear to be any advantage from doing so. It is, however, curious to know that if the steam and water in the boiler under discussion were placed in a large cylinder with non-conducting walls and allowed to expand behind a piston, down to the pressure of the atmosphere there would be developed 138,000,000 foot-pounds. And further, if this work were expended in raising the boiler and its contents against the attraction of gravity, it could lift them a mile high. CHAPTER VIII. BOILER ACCESSORIES. In this chapter will be described various fittings, attach- ments, and accessories for steam-boilers. Valves are used to control and regulate the flow of fluids in pipes. They are variously named after their forms or uses, such as globe valves, angle-valves, straightway valves, and check-valves. C Fig. 83. Globe Valves are named from the globular form of their cases. The case is separated into two parts by a diaphragm with a passage through its horizontal part, as shown in Fig. 83. The fluid enters at the right, passes under the valve, and 235 236 STEAM-BOILERS. out at the left. The valve is shut by screwing down the handle on the valve-spindle. A stuffing-box around the valve-spindle prevents leakage of fluid. In this valve the seat ^^ S. 5 PIPE TAP Fig. 84. is rounded, and the valve face is a ring of a peculiar composi- tion, let into the valve at R. When the valve is shut, this composition is squeezed down onto the seat and makes a tight joint. If the fluid enters the valve from the right-hand side, the BOILER ACCESSORIES. 2tf valve-spindle may readily be packed to prevent leakage while the valve is closed. If the fluid entered the valve at the other end, it would be necessary to shut off the fluid from the entire pipe in order to pack the valve. Angle-valves. — This form of valve, shown by Fig. 84, has an inlet at the bottom and an outlet at one side, it may take the place of an elbow at a bend in piping. The valve is made in two parts. The upper part carries a ring of sore metal which forms the bearing-surface. The lower part has ribs or wings which enter the opening through the valve-seat and guide the valve to its seat. The valve-spindle has a Fig. 85. screw at the upper end which passes through a yoke entirely outside of the body of the valve. The body of the valve is made of cast iron. The valve, 2 3 8 S TEA M-B OILERS. valve-seat, valve- spindle, and stuffing-box follower are made of brass or composition. This form of valve is frequently used for the stop-valve between the boiler and the main steam-pipe. Straightway or Gate Valve. — This form of valve gives a straight passage through the valve, and offers very little resistance to the flow of fluids when it is open. Fig. 85 represents a Chapman valve, in which the valve is wedge- -c ff^ Fig. 86. shaped and is forced against a wedge-shaped seat. The valve- spindle is held at a fixed height by a collar, and draws up or forces down the valve to open or close it. The body of the valve is of cast iron; the valve, valve-spindle, and stuffing-box are ot brass; the valve-seat is a soft composition. Fig. 86 represents a Peet valve, which has the faces of the valve-seats parallel. The valve itself is made in two pieces, BOILER A CCESSORIES. 239 between which is a peculiar casting, U shaped at the bottom and with wedge-shaped lips at the top. When the valve is shut this casting rests on the bottom of the valve body, and the two halves of the valve are thrown against the parallel valve-seats by the wedge-shaped lips of the casting. When the valve is opened this casting hangs between the two halves of the valve by the under side of the wedge-shaped lips. Check-valves allow fluids to pass in one direction, but not in the other. Fig. 87 represents a lift check- valve; it Fig. 87. Fig. 88. resembles a globe valve without a valve-spindle. Fluid entering at the left will lift the valve and pass out at the right. Should the current be reversed the valve will be promptly closed. Fig. 88 represents a swing check-valve. It offers less resistance to the flow of fluid than the valve shown above, and there is less chance that foreign matter will lodge on the valve-seat. The valve has some looseness where it is fastened to the swinging arm, so that it may properly seat itself. A feed-pipe must always have a check-valve to keep the boiler-pressure from acting on the pump, or injector, when it is not at work. It automatically opens to allow water to pass into the boiler. There should also be a stop-valve (a globe or gate valve) near the boiler which can be shut at will; thus when the check-valve shows signs of leaking the stop-valve 240 STEAM-BOILERS. may be shut, and then the check-valve may be opened and examined. Safety-valves are intended to prevent the pressure oi steam from rising to a dangerous point. In order to accom- plish this, the effective opening of the valve should be suffi- cient to discharge all the steam that the boiler can make when urged to its full capacity. The effective opening is equal to the circumference of the valve-seat multiplied by the lift of the valve, if the valve-seat is flat; if the valve-seat is conical, the lift should be measured at right angles to the seat. Then if / is the vertical lift and if a is the angle which the seat makes with the vertical, the effective lift is / sin a, The lift of a safety-valve rarely exceeds i/io of an inch. A two-inch pop safety-valve, made by the Crosby Gauge and Valve Co., and tested at the laboratory of the Massachusetts Institute of Technology, was found to lift from 0.07 to 0.08 of an inch. The valve had a conical seat with an angle of 45 . The actual flow was about 95 per cent of the calculated flow for this valve. The amount of steam that a boiler can make may be estimated from the grate-area, the rate of combustion, and the evaporation per pound ot coal. The first item is fixed, and the other two, though somewhat indefinite, may be estimated from the type of boiler and the conditions under which it works. For example, a factory boiler having a grate 5 feet by 6 feet may be assumed to burn 18 pounds of coal per square foot of grate-surface per hour, and to evaporate 8 pounds of water per pound of coal. It will therefore generate 5X0X18X8 < : -; = 1.2 pounds of steam per second. 60 X 60 v l The amount of steam which will be delivered by a safety- BOILER ACCESSORIES. 24 I valve may be calculated by an empirical equation proposed by Rankine; it may be written W=A— t 70' in which W is the weight of steam in pounds delivered per second, A is the effective area of discharge in square inches, and / is the absolute pressure of the steam in pounds per square inch. If the weight of steam to be discharged per second is known, then this equation may be used to calculate the effective area; and will then read A - 70W In the example given above the weight of steam per second is 1.2 pounds. If the steam-pressure is 100 pounds absolute (85.3 by the gauge), then the effective area must be 70 X 1.2 A = =0.84 100 of a square inch. If the effective lift be assumed to be 0.075 of an inch, the circumference of the valve-seat should be 0.84 -f- 0.075 = JI - 2 inches, and the diameter should be 3.5 inches. A common rule requires that there shall be an area of 1/3 of a square inch through the valve-seat for each square foot of grate-surface. It so happens that this rule gives almost identically the same result as that just calculated for the above example; thus: 5 X 6 10 square inches, v/ — = 3.5 -f- inches, diameter. 242 STEAM-BOILERS. Should the size of the valve determined by the two methods be different, the larger one must be taken ; for the engineer will desire to fulfil the requirements of the first method for the sake of safety, and the requirements of the second method must be fulfilled if the boiler is to pass inspec- tion. Lever Safety-valve. — The general arrangement and some of the details of a well-made safety-valve are shown by Fig. 8 9 . I WEIGHT 116 LBS. CENTER OF GRAVITY OF LEVER WEIGHT OF LEVER 42 LBS. WEIGHT OF VALVE AND SPINDLE 15 LBS. Fig. 89. The body of the valve is of cast iron, and has an opening at one side from which the escaping steam is led out of the boiler-room through an escape-pipe. The valve and valve-seat are of brass or composition; the bearing-surface is at an angle of 45 with the vertical. The load is applied by a steel spindle, to a point beneath the bearing-surface so that the valve is drawn down to its seat. The spindle passes through a brass ring in the cover to the valve-casing. The load is applied by a lever with a fulcrum at A and a weight at D. It is steadied by guides cast on the cover of the casing; in the figure the valve and body are shov/n in section but the spindle, lever, guides and weight are shown in eleva- tion. It is important that the pins ai: A and B shall be loose in their bearings, and that the spindle shall be free where it BOILER ACCESSORIES. 243 passes through the top of the valve-case, so that the valve may not fail to rise even if the working parts are rusted a little. After a safety-valve has blown off it is liable to leak a little, and such leakage is likely to injure the bearing-surface. In this way safety-valves sometimes get leaky and trouble- some. The proper wa)' is to regrind the valve and make it tight, but if the boiler attendant is careless he may try to stop the leak by jamming the valve on its seat. This may be done by hanging on extra weight, or wedging a piece of wood or metal against the lever. To remove temptation, it is well to have the guides for the lever open at the top, and also to cut off the lever to just the proper length so that the weight cannot be slid farther out. A short lever and a heavy weight are better, for this reason, than a lighter weight and a longer lever. In order to make a calculation of the pressure at which a safety-valve will blow off, we must know the diameter of the valve, the weight of the valve and valve-spindle, the length of the lever and the weight hung at its end, and the weight and centre of gravity of the lever. This last may be found by calculation, or more simply by balancing the lever on a knife-edge. In the example shown by Fig. 89 the valve has a diameter of 5 inches and an area of ^ 6 ^= 19.635 square inches, on which the steam presses. The valve and spindle weigh 15 pounds; this is applied directly at the valve. The weight of 115 pounds at the end of the lever, is 56 inches from the fulcrum at A. It is equiva- lent to a weight of ili*-?^ = l6l0 244 STEAM-BOILERS. pounds at the valve. The weight of the lever is 42 pounds* applied at the centre of gravity C, 20 inches from the fulcrum.. It is equivalent to a weight at the valve of 42 X 20 = 210 4 pounds. The total equivalent weight, or the load on the valve, is 15 -\- 1 6 10 + 2I ° = 1835 pounds. Since the area of the valve is 19.635 square inches, the steam-pressure per square inch required to lift the valve will be 1835 -T- 19-635 = 93-46 pounds. Problems concerning the loading of a safety-valve may be conveniently stated and solved by taking moments about the fulcrum ; that is, by multiplying each weight or force by its distance from the fulcrum. Let the weights of the valve, spindle, lever, and weight be represented by V, S, L, and W. Let a be the distance of the weight from the fulcrum and b be the distance from the fulcrum to the valve, while c is the distance of the centre of gravity of the lever from the fulcrum. The moment of the weight is Wa t and the moment of the lever is Lc. The moment of the valve and spindle is (V -\-S)b. All three moments act downward, and their total effect is equaL to their sum, Wa + Lc + (V+S)&. If the diameter of the valve is d y then the area is %nd*. Representing the steam-pressure above the atmosphere by/* the force acting on the valve is nd* BOILER ACCESSORIES. 245 and the moment of that force is TTd* , This moment acts upward and, when the valve lifts, will be equal to the total downward moment. So that the equa- tion for calculating the load on a lever safety-valve is pb— = Wa + Lc + (V+ S)&. This equation gives for the steam-pressure at which the valve shown by Fig. 89 will lift [Wa + Lc+ (V+ S)5] p ~ " nd'b _ 4(1 1 5 X 56 + 42 X 20 + 1 5 X 4) •'• p ~ 3.H16X 5 2 X 4 .-. p = 93.46 pounds, as found by the previous calculation. For a second example let us find the distance at which the weight of the valve shown by Fig. 89 must be placed from the fulcrum in order that the valve will blow off at 50 pounds above the atmosphere. Solving the general equation for a, we have nd* pb—- Lc- (V+S)& a = W 50 X 4 X 3 '\ 41 X 5 2 - 42 X 20 - 1 5 X 4 . . * = A - 115 ,\ a r= 26.32 inches. 246 S TEA M-F OILERS. For a third example find the weight which should be hung- at the end of the lever if the valve is to blow off at 30 pounds above the atmosphere. Here we have nd 2 pb Lc-(V-\-S)b W= * u 30 X 4 X iL - — X5 -42X20-15x4 .*. W 56 .•. W = 26 pounds. These last two problems can of course be stated and solved much after the first manner applied to the first problem, but the work, which will amount in the end to the same' thing, cannot be so well arranged nor so easily done. Pop Safety-valve. — A defect of the common lever safety-valve is that it does not close promptly when the steam-pressure is reduced, and it is apt to leak after it has. returned to its seat. The valve shown by Fig. 90 has a groove turned in the flange which projects beyond the bearing-surface, and there is another groove between the outer edge of the valve-seat and a ring which is screwed onto the valve-seat. When the valve lifts the escaping steam is twice deflected, once by the groove in the valve and again by the groove at the valve-seat. The reaction of the steam assists the pressure of the steam on the under surface of the valve, and suddenly opens the valve to its full extent. The valve stays wide open till the steam- pressure in the boiler has fallen a few pounds below the blow- ing-off pressure, and then the valve shuts as suddenly as it opens. The ring which is screwed onto the valve-seat has a number BOILER ACCESSORIES. 247 of holes drilled through it to allow steam to escape from the groove at its upper surface. It may also be screwed up or Fig. go. down to adjust its position; a screw at the side of the case clamps it when adjusted. The action of the valve is regulated 248 STEAM-BOILERS. by the number of holes in the ring and by its vertical posi- tion. This valve is loaded by a helical spring. The tension of the spring and the load on the valve is regulated by a sleeve which is screwed down through the top of the valve-case. It is of course possible to load a plain safety-valve in a similar way, or to load a pop-valve with a lever and weight. The valve is extended up in the form of a thin shell to guard the spring from the escaping steam. The valve-spindle is ex- tended through the top of the case, and may be pulled up by a lever when it is desired to ease the valve off from its seat. A drip at the lower right-hand side of the case draws off water which may collect in the case. The valve and its seat, the adjusting-ring on the seat, the valve-spindie, and the bearing-pieces on the spring are all brass. There is also a brass ring inside the shell that extends down from trie co^er and incloses the spring. There should be a little clearance between this brass ring and the shell on the valve so that the valve shall not be cramped. The entire valve-casing, which is made in four parts, is of cast iron. The closeness of regulation by a safety-valve depends mainly on the width of the bearing-surface. Thus a valve with a narrow bearing-surface will close after the pressure in the boiler is reduced a few pounds ; a valve with a wide bear- ing-surface will stay open till the pressure has suffered a serious reduction. By making the bearing-surface very narrow the reduction of pressure maybe made as small as two pounds. For example : a certain valve was made to open at 100 pounds and to close at 98 pounds. When the bearing-surface is narrow it must be made of hard, dense metal to endure the pressure concentrated on it. Hard bronzes, compositions and nickel alloys are used for this purpose. A safety-valve should be set by trial, to blow off at the required pressure as shown by a correct steam-gauge. A safety-valve should occasionally be lifted from its seat to BOILER ACCESSORIES. 249 insure that it is in proper condition. An unexpected opening of a safety-valve or continued leakage shows lack of attention to duty on the part of boiler attendants. While the safety- valve for a boiler should be able to deliver all the steam it can make, it may be considered that the proper function of a safety-valve is to give warning of excessive pressure. The safety of the boiler must always depend on the faithfulness and intelligence of the boiler attendants. Inspection laws commonly require that every boiler shall have two safety-valves, and that one of them shall be locked up in such a manner that it cannot be overloaded by accident or design. Water-column. — The position of the water-level in a boiler is indicated either by a water-glass or by gauge-cocks or by both. These may be connected directly to the front end of the boiler, or they- may be placed on a fitting known as a water-column or combination. Fig. 91 shows a good form of water-column. It is a cast-iron cylinder connected to the steam-space at the top and to the water-space near the bottom. The normal position of the water-level is near the middle. There is at the bottom a globular receiver into which deposits from the water may settle and be blown out at will. In one side of the water-column are brass fittings for the water-glass, which is a strong tube of special make. The glass tube passes through a species of stuffing-box in the brass fitting. The joint is made tight by a rubber ring which fits on the tube and is compressed by a follower screwed onto it. Each fitting has a valve by which steam may be shut off when the tube is cleaned or replaced. A cock at the bottom drains water from the tube; for this purpose the lower valve is closed and the cock is opened. Stout wires at the side of the glass tube guard it from injury. If either valve leading to the water-glass is closed, the level of the water will rise in the tube. If the upper valve is 250 S TEA M-B OILERS. closed, the steam in the upper part of the tube is gradually condensed by radiation, and is replaced by water entering from below. If the lower valve is closed, the condensation of steam from radiation will accumulate and gradually fill the tube. Gauge-glasses are very brittle and, though carefully annealed, are under considerable stress from unequal cooling. WATER CONNECTION Fig. 91. Before a tube is put in it may be cleaned by pouring acid through it, or by drawing a bit of waste through on a string. A wire should never be forced through a glass tube, for the slightest scratch may start a break which will end in reducing the tube to small pieces. When a tube is in place it may be cleaned by closing the lower valve and opening the drainage- cock and allowing steam to blow through. When a boiler is left banked overnight the water-glass BOILER ACCESSORIES. 251 should be shut off, since a breakage may result in drawing the water in the boiler down to the level of the lower end of the tube. In addition to the water-glass, which shows at all times the level of the water, the water-column carries three gauge- cocks. One is set at the desired water-level, one a little above and one a little below. Steam from the steam-space, through the upper gauge-cock, becomes superheated as it blows into the atmosphere and looks blue. The lower cock discharges hot water from the water-space, which flashes into steam as it escapes, but it has a white color, which is very distinct from that of the jet from the steam-space. A good fireman occasionally tests the position of the water-level by using the gauges to be sure that the indication by the water- glass is not erroneous. Engineers on locomotives, and boiler attendants where very high-pressure steam is used, often prefer to depend entirely on the gauge-cocks, and dispense with the water-glass, which may be annoying or dangerous when it breaks. The water-column shown by Fig. 91 has an alarm-whistle, which shows above the main casting, at the right. It is con- trolled by two floats inside the cylinder; one float at the top opens the valve leading to the whistle when the water-level is too high, the other near the bottom blows the whistle when the water-level is too low. If the fire is stirred up under a boiler which has had the fire banked, the water-level rises in the water-glass; the reason being that the circulation is from the front of the boiler to the rear, and that this circulation is maintained by a differ- ence of level between the front and rear ends. On the con- trary, the water-level falls when a boiler which has been steaming freely is checked. Steam-gauges.— The pressure of the steam in a boiler is shown by a spring-gauge which has the external appearance shown by Fig. 92. The essential part is a flattened brass 252 STEAM-BOILERS. tube bent into the arc of a circle as shown by Fig. 93. The section of the tube may be an oval, or it may have two longi- tudinal corrugations as shown by Fig. 94. Fig. 92. Pressure inside of such a tube makes it bulge and tends to straighten it. One end is fixed and is in communication Fig. 93. Fig. 94 with the space where the pressure is to be measured. The other end is closed and is free to move. It is connected by a link to a lever which bears a circular rack in gear with a BOILER ACCESSORIES. 253 pinion. The motion of the free end of the tube is multiplied and is shown by the motion of a needle on the pinion. The scale on the dial is marked by trial to agree with the indica- tions of a mercury column or of a standard gauge. A hair- spring on the pinion (not shown in Fig. 93) takes up the back- lash of the multiplying-gear. The long, flexible spring-tube is liable to vibrate to an undue extent when the gauge is exposed to the jarring of a locomotive. To avoid this difficulty, two short stiffer tubes have their ends connected to a more effective multiplying device, shown by Fig. 95. The greater number of joints in this device makes it less sensitive than the other form. Fig. 95- Since the spring-tube changes its shape if the temperature changes, hot steam should not be allowed to enter it. An inverted siphon or U tube filled with water is, therefore, interposed between the gauge and the steam from the boiler. Safety-plugs, or Fusible Plugs, as shown by Fig. 96, are made of brass and provided with a core of fusible metal. If the plate into which they are screwed is in danger of over- heating, the fusible metal will melt and run out, and steam and water will blow into the furnace. If the fire is not put 254 STEAM-BOILERS. out, it will at least be checked and the attention of the fire- man will be attracted. The melting-point of fusible metals is not always certain, and the plugs not infrequently blow out when there is no ap- parent cause. On the other hand, they sometimes fail to act when the plate is overheated. If the plug is covered with incrustation, the fusible metal may run out without giving warning. The following are some of the places where a fusible plug is used: In the back head of a cylindrical tubu- lar boiler, about three inches above the top row of tubes. In the crown-sheet of a locomotive fire-box. In the lower tube-sheet of a vertical boiler; or sometimes in one of the. tubes a Fig. 96. little above that tube-sheet. In the lower side of the upper drum of a water-tube boiler. The fusible composition has a conical form so that it can- not be blown out by the pressure of the steam. Foster Reducing-valve. — When steam is desired at a less pressure than that of the boiler, it is passed through a reducing-valve like that shown by Fig. 97. The valve H is held open by the spring aty, acting through the toggle-levers a, until the steam-pressure in the exit-pipe B, pressing on the diaphragm D, is able to overcome the spring and close the valve. The pressure at which this may occur is determined by the tension of the spring, which may be regulated by the screw at K. It is expected that the valve will be drawn up so as to admit just the proper amount of steam to the exit-pipe B to maintain the de- sired pressure in it. Valves for this purpose are liable to work intermittently, i.e. they close till the pressure falls BOILER ACCESSORIES. 255 below the proper point, then they open and raise the steam- pressure above that point. The valve is a species of throt- tling-valve, and therefore cannot be expected to remain tight. If the machinery supplied by the reducing-valve is liable to ^^j Fig. 97. be injurea by excessive pressure, there must be a stop-valve beyond the reducing-valve. The stop-valve must be closed when no steam is drawn, and must be used to regulate the supply of steam until the amount drawn exceeds the leakage of the reducing-valve. The Damper-regulator shown by Fig. 98 places the damper in the flue leading to the chimney under the control of the steam-pressure, so that if the pressure of the steam falls, the damper is opened wider to quicken the fire. The pressure of the steam in the boiler is communicated through the pipe a to the lower surface of a diaphragm, and lifts the loaded lever b y which stands half-way between the stops at the middle of its length when the steam-pressure is at the proper point. Should the steam-pressure rise above the 256 S TEA M-B OILERS. proper point it raises the lever and opens a small piston-valve at c, and water from a hydrant flows into d and presses on a piston which lifts the weights at e and so shuts the damper. Fig. 98. The weighted head e of the piston is connected by a chain to the lever/, and closes the valve c as it rises, and so shuts off the water from the hydrant. A regulator of the same form attached to a throttle- valve acts as a reducing-valve, and regulates the pres- sure below the valve with a variation of less than one BOIL ER A CCESSORIES. 257 pound. Fig. 99 shows the steam-valve used when the Locke regulator acts as a reducing-valve. The valve is a double valve which is nearly balanced, but with a slight tendency to rise under steam-pressure, as the lower valve is the larger. The cylindrical part of the valve is cut into V notches, so that the supply of steam is regulated to a nicety when the valve is partially open. The cylindri- cal portion of the valve protects the valve-seat and the valve-face so that the valve may remain tight when closed. Steam-traps. — The object of a steam-trap is to drain condensed water from steam-pipes without allowing steam FlG - 99- to escape. As a rule a trap is placed below the pipe to be drained so that the drip from the pipe will run into it. Some traps that return the condensed water to the boiler do not conform to this rule. Some traps, such as the McDaniels, the Baird, and the Walworth, have a valve under the control of a float, which will allow water to pass but not steam. A Fig. 100. The McDaniels trap is shown by Fig. 100. The drip enters at C and escapes through the exit at E when the valve G is open. This valve is raised by the spherical float when the water rises to a sufficient height. When the water is 258 S TEA M-B OILERS. drained from the pipe served by the trap, the water-level in the trap falls and the valve G is closed. D is a counter weight to balance the weight of the spherical float. The valve at G can be opened by screwing down the screw at A Fig. 102. on to the counterweight. The trap can be emptied through the valve at F. The Baird trap, Fig. 101, has a spherical float D which controls a piston-valve at J. The inlet is at C, and the outlet BOILER A CCESSORIES. 259 at /. The screws A and B allow the valve J to be opened or closed by hand. The Walworth trap (Fig. 102) has a floating bucket into which the drip overflows after the outer case is partially filled. When the bucket sinks it opens a passage through the central spindle, and the water in the bucket is driven out through this spindle. The hand-wheel and screw at the top control a valve which is closed when the trap is working. The Flynn trap (Fig. 103) depends for its action on a head of water acting on a flexible diaphragm. Water may enter at the top or the bottom at ori- fices marked A. It fills the pipe B and the globe C as high as the end of the pipe E, and pro- duces a pressure of about a pound per square inch on the under side of the diaphragm at D. The spring at G produces a pressure of about half a pound per square inch on the upper side of the diaphragm. Conse- quently the valve leading from the chamber F to the escape- pipe H is closed so long as the pipe E remains empty. But when the water overflows the top of the pipe E and fills the chamber F, the water-pressure on top of the diaphragm will be the same as that on the bottom, and the spring at G will open the valve and allow water to escape. If the supply of water Fig. 103. at A ceases, the pipe E will be emptied and the valve will be closed under the influence of the pressure on the under side 260 S TEA M-B OILERS. OUTLET Fig. 104. of the diaphragm. In the trap as actually constructed the pipe H is about 24 inches long, in the figure it is made shorter in proportion. The Curtis trap (Fig. 104) has an expansion-chamber at C which M inlet is closed by a diaphragm A at the bottom, and is filled with a very volatile fluid. So long as the ex- pansion-chamber is immersed in water the pressure of the fluid on the diaphragm is balanced by the spring on the valve-spindle B. If the water is drained away and the chamber is exposed to the temper- ature of steam (212 F. or more), the fluid vaporizes and exerts enough pressure on the diaphragm to compress the spring and close the exit-valve. Return Steam-trap. — The traps thus far considered usu- ally discharge against the pressure of the atmosphere. They may discharge into a closed tank against a pressure that is higher than the atmosphere, but in all cases the pressure in the pipes drained by the trap must be higher than the discharge-pressure. Return steam-traps are arranged to discharge directly into the boiler. The Bundy return-trap, shown by Figs. 105 and 106, is set three feet or more above the water- line in the boiler. It is so 105. made that it is first opened to the pipe to be drained, and fills up under the pressure in that pipe. It is then put in commu- BOILER ACCESSORIES. 26 1 nication with the steam-space and with the water-space of the boiler, and the water previously collected drains into the boiler. The trap consists of a pear-shaped receptacle or closed bowl, hung on trunnions, through which the bowl is filled and emptied. When empty the bowl is raised by a weight and lever; when filled with water it overbalances the weight and Fig. 106. falls. The ring around the bowl limits the motion. The condensed water from the pipe or system of pipes to be drained enters the trap through the check-valve B, which pre- vents water from flowing back from the trap into the pipe to be drained. The trap is emptied through the check-valve A, which prevents water from the boiler from flowing into the trap. At C is a valve under the control of the trap, which receives steam by a special pipe from the boiler. When the trap is empty and is lifted by the weight and lever, the valve C is thrown down and is shut; water then flows in through the valve B from the pipe to be drained, and air escapes from an air-valve below C, which is open in this position of the trap. A check-valve on the air-pipe prevents air from en- 262 S TEA M-BOILERS. tering the trap if a vacuum happens to be formed in it. When the bowl is filled it falls and opens the steam-valve C, and steam enters the bowl through a curved pipe shown in Fig. 1 06. The pressure in the bowl is now equal to that in the boiler, and the water collected flows into the boiler by gravity. Separators. — If steam is carried to a distance in pipes, a ^^^mm^H considerable amount of water of conden. 1^" ,<-,=--— ---iszzp. sation accumulates. It is undesirable to have this water delivered to a steam- engine in any case, but if the water ac- cumulates in a pocket or a sag in the piping, it may come along with the steam in a body whenever there is a sudden change of steam-pressure, and then the engine will be in danger of injury. A good way of removing such water is to allow the steam to come to rest in a steam-drum of suitable size, from which the water is drained by a steam- trap ; the steam meanwhile may flow from a pipe at the top of the drum. A small steam-drum used as separator is likely to fail, from the fact that the steam does not come to rest, or because the entering and leaving currents of steam are not properly separated. The Stratton separator, shown by Fig. 107, brings in the steam at one side of a cylinder, with a whirling motion Fig. 107. * that throws the water onto the side of the cylinder; dry steam escapes through a pipe in the middle. A good steam-separator will remove all but one or two per cent of moisture from steam, even though the entering steam is very wet. Attention has already been called to the use of separators BOILER A CCESSORIES. 263 with some forms of water-tube boilers which do not have a sufficient free water-surface for the disengagement of steam. Feed-water Heaters. — The feed-water supplied to a boiler SAFETY VALVE BLOW -OFF * FEED TO BOILER EXHAUST FEED FROM PUMP a^^^^^SSSOTS^iglgSS u MUD BLOW-OFF Fig. 108. may be heated up to the temperature of the exhaust-steam by passing it through a feed-water heater. Feed-water heaters are sometimes made open, i.e., the steam from the engine 264 S TEA M-BOIL ERS. mingles with and heats the feed-water. Such heaters have the disadvantage that the oil from the engine is carried into the boiler. A closed feed-water heater resembles a surface condenser, and as the steam and water do not mingle, there is no danger of carrying oil from the engine into the boiler. The Wain- wright heater, shown by Fig. 108, has the heating-surface of corrugated copper or brass tubes, of peculiar make, to allow for expansion. The steam from the engine passes around the tubes and the feed-water passes through the tubes. The Berry man feed- water heater, shown by Fig. 109, is arranged to have the exhaust-steam pass through a series of inverted U tubes, around which the feed-water circulates. Live-steam feed-water heaters take steam from the boiler to raise the tem- perature of the feed-water up to, or nearly to, the temperature in the boiler. The principal advantage appears to be that unequal contraction, due to the in- troduction of cold water, is avoided. It is claimed that with some forms of boilers a better circulation is obtained by aid of such a heater. The use of a feed-water heater for removing lime-salts from feed-water has been discussed on page 73, and an ex. ample of such a feed-water heater was illustrated in connection therewith. Feed-pipes. — The temperature of the feed-water is usually much below the temperature in the boiler. It thus mud pipe becomes essential to so locate the inlet, FlG - io 9- and to so distribute the water, that un- due local contractions may not occur; this is of special im- BOILER ACCESSORIES. 265 portance when the supply is intermittent. The feed-pipe for the cylindrical tubular boiler, shown by Plate I, enters che shell near the water-line, through the front head. It is carried along one side of the boiler for about three fourths of its length, and then is carried across over the tubes and opens downward. A feed-pipe is often perforated to give a better distribution of the feed-water. The shell is reinforced by a piece of plate riveted on the outside, where the feed-pipe enters the boiler. The end of the pipe has a long thread cut on it, so that it can be secured through the reinforcing-plate and the boiler-shell, and may then receive a pipe-coupling which connects it to the continu- ation of the feed-pipe inside. Sometimes the feed-water is delivered to an open trough inside the boiler, from which it overflows in a thin sheet. Or a perforated pipe may deliver the water in form of spray in the steam-space. Either method has the advantage that the water comes in contact with stenm and is heated before it mingles with the water in the boiler. There is the disadvan- tage that the steam-pressure may fall off when the feed-water is turned on or is increased. It has already been pointed out that the feed-pipe should have a globe valve near the boiler, and a check-valve between the globe valve and the feed-pump. Feed-pumps. — Boilers are commonly fed by a small direct- acting steam-pump placed in the boiler-room. The steam- consumption per horse-power per hour of such pumps is very large, and yet the total steam used is insignificant. They are cheap and effective, and easily reguiated. Power pumps driven from a large engine are more econom- ical, provided their speed can be regulated; they not infre- quently are arranged to pump a larger quantity than required for feeding the boiler, the excess being allowed to flow back to the suction side of the pump through a relief-valve. When one pump supplies several boilers, a series of diffi- 266 STEAM-BOILERS. culties is liable to arise. First, if the boilers are fed singly in rotation, the krge intermittent supply of feed-water is likely to give rise to local contraction and the water-level in the boiler fluctuates; there is liability that the water-level will fall too low, endangering the heating-surface, or there may be excessive priming when the water-level is high. It appears advisable that the feed should be delivered to all the boilers simultaneously, the supply to each boiler being regulated by its stop-valve; each branch pipe to a particular boiler should be provided with its own check-valve, and the water-level and rate of feeding of each boiler must be carefully watched by the fireman, or by a water-tender if there are many boilers. An injector is conveniently used for feeding a boiler if the feed-water is not too hot; it has the incidental advantage that it heats the water as it feeds it into the boiler. An injector should be connected up with unions, so that it may readily be taken down for inspection. At sea an injector is com- monly used when the boilers are fed from the sea or from a supply-tank. Every boiler should have two independent sources of sup- ply of feed-water, so that there may be some resource if the usual supply gives out. There may be two pumps, or a pump and an injector. A locomotive usually has two injectors. Blow-off Pipe. — The blow-off pipe draws from the lowest part of the boiler, or from some place where sediment may be expected to collect. On the blow-off pipe there is a cock or a valve which is opened to blow out water from the boiler. Sometimes there are both a cock and a valve. A cock has the disadvantage that it may give trouble by sticking; a valve may leak and the leak may not be detected. The pipe should be carried beyond the cock, so that the attendant is not liable to be splashed with hot water, but the pipe should end in the boiler-room or where discharge through the pipe on account of a leaky cock or valve may be sure to BOILER ACCESSORIES. 26/ attract attention. Each individual boiler should have its own blow-off pipe. The blow-off pipe where it passes through the back con- nection is covered with magnesia, asbestos, or fire-brick. In spite of this protection the blow-off pipe may burn off. The device shown by Fig. 1 10 is used to overcome this difficulty. WATER LINE ^ Fig. i io. When the blow-off cock is shut and the valve on the vertical branch is open, there is a continuous circulation of water which keeps the pipe from burning. The valve on the verti- cal branch is closed before the blow-off cock is opened. If a blow-off pipe burns off and water begins to escape, the feed-pump should be run at lull capacity to keep water in the boiler and guard the piates from burning, if that is possible. The fire should then be checked by throwing on wet ashes or by other means, unless escape of steam from the break in the blow-off pipe prevents. Piping to carry steam from a boiler to an engine, for heating buildings, and for other purposes is too important to be considered as accessory to the boiler. A lew remarks, how- ever, may not be out of place- ^68 STEAM-BOILERS. The expansion of the pipe due to changes of temperature should be provided for, or else cracks in the pipe or fittings, or leakage at the joints may be expected. A common way of allowing for expansion is illustrated by Fig. m, which shows o n Fig. hi. the connection from a boiler to the main steam-pipe. When the main steam-pipe expands or contracts, the short nipple between it and the angle-valve turns a little at one or at both ends; in like manner the vertical pipe turns a little at the nozzle or at the elbow. The motion is so small and so dis- tributed as not to give any trouble unless the expansion to be provided for is very large. A large and long straight steam- pipe may require an expansion-joint. A slip-joint may be made of a brass pipe inside a shell with packing-box and fol- lower, arranged something like the piston-rod of an engine. It is essential that the slip-joint shall be in line or it will be cramped and give trouble. For this purpose the ioint may be carried and guided by a cast-iron bed-piate. Fig. in is so arranged that there is no space where water can collect when the boiler is shut off from the main steam- pipe. If the stop-valve were in the vertical pipe, as is some- times the case, then the pipe over the valve would fill up with water when the boiler is shut off, and that water would be BOILER ACCESSORIES. 269 £uddenly blown into the steam-main when the stop-vaive is next opened. A pipe so situated should always have a drip- pipe to draw off condensed water before the valve is opened. As a special example we may mention the pipe leading to an engine, which always has a drip-pipe above the throttle-valve. Pipes that are likely to be troubled by condensation should be continuously drained by a steam-trap. Horizontal pipes are sometimes arranged so that water may collect in them, due to a sag in the pipe or to the fact that they do not properly drain through a side branch. Though the water may lie quiet in such a pocket while the draught of steam is steady, a sudden increase in the velocity of the steam, or a rapid opening of the valve supplying steam to the pipe, will sweep the water up and carry it along with the steam. The danger from the inrush of water to an engine is readily seen, but it is not so well known that the water thus violently thrown against elbows and other fittings give rise to leaks, if it does not burst the fittings. It is to be remembered that steam offers little or no resistance to the movement of water in a pipe, as it is readily condensed either from a slight increase of pressure or by mingling with colder water. Again, water at the temper- ature corresponding with the pressure easily separates, forming bubbles of steam, which as easily collapse, and the shock of impact of the water gives rise to pressures that search out all weak places in the pipe, even at some distance. Drawings for piping commonly represent the work as though it were all in one plane. There is little liability of confusion since the actual piping could usually be swung into one plane, turning in tees and elbows and other fittings. Lengths are given from centre to centre of pipes represented, because the fittings may differ in length. Piping up to two inches in diameter can be cut by hand. Larger sizes are cut by machine. Sizes of pipe are named by the inside diameter; but the actual diameter, especially of small sizes, may be larger than the nominal diameter. 2^0 S TEA M- BOILERS. Pipe sizes are 1, \, f, J, |, I, ij, i£, 2, 2J, 3, 3J, 4, 5, 6, 7, 8, 10, 12, etc. Brass piping is nearer the nominal size than iron piping. Boiler-tubes are named from the outside di- ameter. Pipe-hangers.— When a pipe needs support it is commonly hung from an overhead beam by a wrought-iron ring, a little Fig. 112. larger than the pipe, which is held up by a lag-screw in the beam. If the pipe is long, the expansion is likely to cramp the ring on the pipe and then bring an awkward side strain on the lag-hook ; or it the hook is open in the direction of the expansion, the ring may be lifted out of the hook and so the support at that point may be lost. The hanger shown by Fig. 1 12 has the supporting ring carried by a roller. The track for the roller is carried by lag-screws. In some cases the lag-screws can be advantageously replaced by bolts which pass clear through the beam. Various modifications of this device may be used. For example, the pipe may rest on a BOILER ACCESSORIES. 2J I toller with a hollow face , the roller is on a horizontal bolt which is supported by straps co an overhead beam. Area of Steam-pipe. — In order that the loss of pressure in a steam-pipe due to friction may not be excessive, it is customary to limit the velocity to 5000 or 6000 feet per min- ute. If there are many bends or elbows in the pipe, the velocity may be 4800 feet per minute, or less. Example. — Required the diameter of the main steam-pipe leading from a battery of boilers having an aggregate of 3000 boiler horse-power. Assume the pressure to be 100 pounds by the gauge, or about 115 pounds absolute. Assume also that a boiler horse-power is equivalent to 30 pounds of steam per hour. Then the steam drawn from the boiler in one hour is 30 X 3000 = 90,000 pounds. The steam per minute is consequently 1500 pounds. Now one pound of steam at 1 1 5 pounds absolute has a volume of 3.862 cubic feet. Consequently 1500 X 3.862 = 5793 cubic feet of steam per minute must pass through the steam- main. With a velocity of 5000 feet per minute the area of the pipe must be 5793 -5- 5000=: 1. 157 square feet, or 166.6 square inches. The corresponding diameter is 14J inches. The next larger size of pipe is 16 inches, which will be used. CHAPTER IX, SHOP-PRACTICE. THE method of work in a boiler-shop depends on the size and arrangement of the shop and on the class of work. There are, however, certain general principles which can be recognized in all modern shops. The materials, especially the plates, are received at one end of the shop, near which is a storeroom, and a bench for laying out work. The plates, after they are laid out, pass in succession to the several machines, where they are sheared, punched or drilled, planed, rolled, and riveted. The machines for performing these operations are arranged in order with proper spaces for handling and working. Space is provided where boilers may be assembled and receive their tubes and furnaces. Machines which, like the punch, have much work to do, compared with other machines, may be duplicated. There should be an efficient system for handling the material at the machines and for passing it on from one machine to the next. A good arrangement is to have a swing-crane near each machine ; the spaces served by the several cranes overlap, so that one crane takes material from the next, and so on. It is advantageous, especially in large shops, to have a travelling crane that can handle the largest boiler made, and which can serve any part of the shop. Flanging and smithing are usually done in a separate shop or room. A few machine-tools are needed for doing work on steam-nozzles, manhole rings and covers, etc. A boiler-shop will have an office, a drawing-room, and a 272 SHOP-PR A C TICE. 273 pattern-room, also a storeroom for patterns. These may be conveniently located in the second story. A Boiler-shop. — The application of the general princi- ples just stated and the explanation of details can be best given by aid of an example. A medium-size shop for making cylindrical boilers has been chosen for this purpose; the shop is capable of making any shell boiler of moderate size. This shop will employ sixty or seventy men and can turn out two 100-horse-power boilers per day. It will take about three days to finish one boiler, so that there may be six or more boilers in process of construction at one time. The shop which is represented by Fig. 113 has one end on the street and has a driveway or yard at one side. Plates are received at the street-door by a travelling crane and stored near at hand. The same crane takes plates to the laying-out bench and from there to the crane which serves the shearing- machine. Along one side of the shop are arranged in suc- cession a shearing-machine, two punches, a plate-planer, a set of plate-rolls, and a riveting-machine. Between the punches and nearer the wall is a flange-punch ; near the planer is a forge for scarfing. This series of machines is served by four swing-cranes, and there are also two hydraulic cranes near the riveting-machine. These cranes, which are at the top of a tower thirty feet high, are operated from the working platform of the riveter. There are two shipping- doors where the finished boilers are delivered to teams, and at each door there is a jib-crane for handling the boilers. These jib-cranes and the hydraulic cranes at the riveter have a capacity of eight or ten tons ; the swing-cranes may be much lighter. A shop where large marine boilers are made will have more powerful cranes. The machine-shop is near the receiving-door. Here are the lathes, planers, and drills for doing work on manholes, nozzles, and other fittings ; also a bench for fitting up boiler- fronts. Two drills for boring tube-holes in tube-plates, and 274 S TEA M-B OIL ERS. CO 2 £ 133U1S SHOP-PR A C TICE. 2 ? 5 a boring-mill for facing off the flanges of boiler-heads, are placed in the entrance to the machine-shop, where work can be conveniently brought to them from the boiler-shop. At the end partition of the machine-shop are places for storing boiler-front castings and sheet-iron. The corner of the boiler- shop near the machine-shop is known as the cold-iron shop ; here the uptakes, flues, and dampers are made. This shop has a shearing-machine, three punches, and a set of rolls suitable for. sheet-iron work; also a bench with hand-vises. At the rear of the boiler-shop there is in one corner a store- room for tubes, stay-rods, channel-bars, and finished fittings. In the opposite corner are the forge-shop and the engine- room. These are separated from each other and from the boiler-shop by glass partitions which do not cut off the light, and yet keep the smoke and dust from the forge out of the other rooms. The main line of shafting is near the wall over the shear- ing-machine, punches, and rolls. The shafting for the ma- chine-shop and cold-iron shop is driven by a belt from the main shaft, near the front end of the building. A space is left near the riveter where the plates from the rolls can be assembled and bolted together before going to the riveter. In front of the riveter there is a space about 60 feet wide and 120 feet long where boilers are deposited after leaving the riveter. Here the boilers receive their stays and tubes, here they are calked and receive ail fixtures that are perma- nently attached to the shell. At this place the boilers are tested by hydraulic pressure, usually to one and a half times the working pressure. When complete the boilers are painted and oiled, ready for shipment. To illustrate the method of building a boiler more in detail, the different steps in making a horizontal boiler will be followed in order. Flanging Heads. — Regular sizes of boiler-heads flanged at one operation by machinery can now be bought on the 2 j6 S TEA M-B OILERS. market, and all except the largest shops are in the habit of buying them. The flanging-machine has a former and a die between which the plate is formed under hydraulic pressure while at the proper flanging temperature. No strains due to unequal heating or cooling are set up in this process, and the plate, which is allowed to cool gradually, does not need to be annealed. Irregular sizes and shapes are made in the shop on a special cast-iron anvil, which is about six inches deep, flat on top, and curved at one side to about the radius of the head to be flanged. The corner of the anvil or former is rounded so as not to cut the plate. It is placed near a special low forge where the plate is heated. In flanging, the plate is first marked at short distances on the inner circle of the bend with a prick-punch. A portion of the plate is then heated to a good heat, and the plate is taken to the anvil or former. After adjusting so that the depth of flange overhangs the right distance from the edge of the former, the heated portion of the plate is beaten down Fig. 114. — Lifting-dogs. against the side of the former by wooden mauls and then smoothed with a flatter and sledge. The plate is then heated in a new place and another portion bent. To straighten the head and also to remove the strains set up by this way of flanging, it should be heated to a dull red and allowed to cool gradually. The lifting-dogs represented by Fig. 1 14 are used in lift- SHOP-PRACTICE. 2J? ing and placing the head during the flanging, and in handling plates during other operations. Fig. 115 represents crane-lifts which are used when plates are lifted and carried by cranes. Fig. 115. After the head is flanged, holes for rivets, stay-rivets, and tubes are marked, and all the rivet-holes are punched. Flange-punch. — The holes in the flange are punched by a special machine shown by Fig. 116. The punch is carried by a horizontal wrought-iron plunger which is operated by a cam. The die is carried by a hooked extension of the frame. The head is held horizontal with the flange down ; the flange is dropped between the punch and the die and the lever is STEAM-BOILERS. Fig. 117. pulled to throw the cam into play ; the plunger then makes a stroke and punches a hole. The machine is driven by a belt, with a fast-and-loose pulley. On the shaft with these pulleys is a heavy fly-wheel. A pinion and spur-gear give a slow powerful stroke to the gear which moves the cam. Punch and Holder. — The punch (Fig. 117) is made of a solid piece of tool-steel. It has a flat head and a conical shoulder by which it is held onto the plunger, a short straight body, and a slightly coned point. The point is larger at the cutting edge than back toward the straight body, to avoid friction in the hole. A tit in the middle of the face of the punch catches in the centre- punch mark and centres the hole punched. The holder is made of wrought iron. It screws onto the end of the plunger, grips the punch by the conical shoulder on its head, and draws it down firmly against the plunger. Tube-holes. — There are two ways of cutting the holes for the tubes in boiler-heads. Some- times a small hole is punched at the centre of the hole. A tool like that shown by Fig. 118 is then put in the drill-press. The post in the middle is run through the small hole previously punched or drilled, and the two cutters rapidly cut out the tube-hole to the proper size. The other way is to punch the tube-holes at once to the proper size by a helical punch shown by Fig. 119. The die is made in the form of a ring with a flat face, so that the punch begins to cut at the cor- ~\J Fig, 118. SHOP-PR A C TICE. 2 79 ners, and the metal is removed by a shearing cut. Though not always done, the holes ought to be punched a little under size and then reamed out to give a fair surface against which the tubes may be expanded. Finishing the Flange. — The boiler-heads are placed on the platen of a boring-mill like that shown by Fig. 120, and the edge of the flange is turned off. The heads of marine boilers are often turned to a true cylinder at the flange to insure that they shall exactly fit the cylindrical shell into which they are riveted. This also gives a good surface to calk against. Boring-mill. — A simpler machine than the boring-mill shown by Fig. 120 would answer to turn off the flanges of the boiler-heads. But the machine is useful in other ways and may do the work which is commonly done on a large lathe. The platen is driven much in the same manner as the head of a lathe, through gearing and cone pulleys, to provide for various speeds. This gearing is not well shown in the figure, as it is hidden by the frame. The cutting-tool is ad- justed and controlled much like the tool of a planer. The tool-carriage is on a horizontal cross-head which is supported at the side frame and on a round vertical bar at the middle. The tool can be traversed in and out on the cross-head, and the cross-head may be raised or lowered. For doing some classes of work the cross-head may be set vertically on the guides that are shown on the horizontal bars of the frame near the right-hand end. Or, again, a tool may be carried by the central rod, which can be fed down by the screw at the top. Laying on the Plates.— The first and one of the most important steps in the work on the shell is the marking out of the plates. Generally one man in each shop does all the laying out. After squaring the sheet, he marks off the length and locates the rivet-holes by means of gauges. These 28o S TEA M-B OILERS, gauges have to be made by trial, a suitable allowance being made in them on account of the thickness of the plate for the Fig. 120. change in length due to rolling. There is a gauge for each course, or a set of gauges for each size boiler, and also sets SHOP-PRACTICE. 28l for the same size, but with different thickness of shell. The plates are marked either with a piece of soapstone or with a slate-pencil. Rivet-holes are prick-punched at the centre. Shearing. — When the plate is laid out it is taken from Fig. 121. the bench to the shears and any superfluous stock is cut off. A shearing-machine is shown by Fig. 12 1. The lower knife is fixed and the upper knife is moved by an eccentric inside the head. The eccentric-shaft is coupled to the gear-shaft by a clutch that is controlled by a treadle. The weight of the sliding-head is counterbalanced by a weight and lever at the top. Lugs are shown on the casting near the knives ; when the machine is required to do extra-heavy work, wrought-iron "bolts are put through the lugs and screwed up to strengthen the frame. The machine is driven by a belt with a fast-and-loose pul- ley ; the shaft carrying these pulleys has a pinion gearing into a large gear to give the necessary power for shearing. A fly- wheel steadies the motion of the machine; it must be able to supply the power for shearing-plates without a large reduction in speed. 282 s TEA M-B OILERS. Punch. — After the plate is sheared to size it is taken to one of the punches and all the rivet-holes are punched. Larger openings for man-holes and other fittings are cut out by punch- ing overlapping holes, thus leaving a ragged edge which is afterwards chipped smooth. The plate is not entirely cut away at such large openings, but the piece to be removed is left hanging at three or four places until after the plates are rolled into cylindrical form. If the pieces were removed, there would be less resistance to the rolls at such places and the plates would have a conical form instead of a true cylindrical form. The punches resemble the shears shown by Fig. 120, with a punch and die instead of the knives. Machines are often so made that they either punch or shear. Planing. — After the plate is sheared and punched the edges are planed to a slight angle to give a good calking edge. The planer shown by Fig. 122 has a long narrow bed on which the edge of the plate is laid and to which it is clamped by a follower; the follower is forced down by screws which pass through a beam as shown. The tool-carriage is drawn- back and forth by a leading-screw; the tool is made to cut on both strokes, and is fed by hand between the cuts. Scarfing. — When the plates are joined by a lap-joint the proper corners of each plate are heated in a portable forge near the planer, and are drawn down or scarfed so that the overlapping plates may come close together and not leave a space. Plate-rolls. — The plates for forming the cylindrical shell are bent to shape cold by running them through bending-rolls The horizontal roll represented by Fig. 123 has two parallel rolls below that are driven in the same direction by gearing. The upper roll is adjusted at each end separately, and some care is required or the shell will receive a conical shape instead of a true cylindrical shape. The bearing at one end of the SHOP-PRACTICE. 283 284 S TEA M-B OILERS. roll can be swung out, as shown by the figure, to remove the plate after it is rolled. The rolls may be driven in either direction by crossed and open belts. The plate to be rolled has one edge introduced Fig. 124. between the upper and lower rolls, the upper roll is brought down and the rolls are started up. The plate is run through nearly to the other edge then the top roll is screwed down SHOP- PR A C TICK. 285 farther and the rolls are reversed. Thus the plate is run back and forth and the todp roll is gradually rawn down till the plate acquires the proper form. The extreme edges of the plate arc not bent in this process; they are commonly bent afterwards by hammering them with sledges. Some rolls have a special device for bending the edges; it consists of two short overhanging rolls about fifteen inches long, one concave and the other convex. The ends of the plate are fed through these rolls sideways, and are bent before they are introduced into the long rolls. Vertical rolls, shown by Fig. 124, are coming into use in boiler-shops. They take up less floor-space, and the plate after it is rolled up into cylindrical form is easily hoisted off from the front roll. For this purpose the front roll is counterbal- anced and the top end can be swung out clear from the hous- ing. The figure shows the rolls as erected by the builders; in the boiler-shop the plate at the lower end of the rolls is flush with the floor of the boiler-shop. The width of plate that can be rolled by either horizontal or vertical rolls depends on the length of the rolls, The length of the rolls and the reach of the riveter (to be men- tioned later) determine the width of plate that can be handled in the shop. Assembling and Riveting. — When the plates for a boiler have been punched, planed, and rolled they are assembled in courses, and bolted together ready for riveting. Formerly boilers were commonly punched and riveted ; now it is cus- tomary to punch the rivet-holes one eighth of an. inch smaller than the finished size and then drill to the right size after the boiler is assembled. This is more expeditious than drilling directly, and as all the metal affected by punching is removed it gives as good results. It is the custom in most shops to drill the holes out at the riveting-machine immediately before the rivets are driven and thus each rivet-hole is sure to be true. 286 S TEA M-B OILERS. The shells of heavy marine boilers are drilled after the plates are assembled without previous punching. A few holes are drilled before the plates are rolled and serve for bolting the plates in place when the boiler is assembled. There are two forms of machines for drilling marine-boiler shells. In one the boiler is placed horizontal on rollers so that it may be readily turned, There are two or three upright frames each carrying a drill. The frames may be adjusted lengthwise of the boiler, and the drills may be set at any height or turned at an angle. When a longitudinal seam is drilled the boiler is rotated to bring a row of rivets to a drill, and the frame is trav- ersed from hole to hole. When a ring-seam is drilled the drill is brought to the proper place, and the boiler is rotated so as to bring the rivet-holes in succession to the drill. The other machine has the boiler placed on one end and the verti- cal frames carrying the drills can be rotated into place, and the boiler can be turned on a vertical axis. If plates are punched and riveted without drilling, the holes should be punched from the side of the plate which comes in contact with the other plate. The reason for this is that the die is always a little larger than the punch and the hole is slightly conical, larger at the side where the die holds up the plate. If the smaller ends of the holes in two plates are brought together, then the rivet fills the hole better and draws the plates up more perfectly as the rivet cools. It is clear that three or more overlapping plates should always be drilled, as punched holes cannot always be brought together in a proper manner. This is aside from the desirability of drill- ing all rivet-holes. Returning now to the assembling of a cylindrical boiler, the process is as follows: The back head is put in the rear course or ring of the shell, and is bolted with six or eight bolts through the punched holes. The head and ring are hoisted up to the drill near the riveter, and six or eight holes are drilled at about equal distances around the seam holding the SHOP-PRACTICE. 287 head into the ring or course, and rivets are driven by the machine in these holes. The bolts are now taken from the punched holes, and all the remaining holes are drilled and riveted, completing the ring-seam through the flange of the back head. The reason for driving a few rivets first, at equal intervals, is that the errors of spacing, when any exist, are distributed, and are removed during the subsequent drilling; while such errors might accumulate and give trouble if the seam were riveted in succession beginning at one point, without first driving a few rivets at intervals. After the ring-seam through the flange of the head is completed, the longitudinal seam or seams are drilled and riveted. Here again a few rivets are driven at intervals before the seam is riveted up. A few holes at the ends of the seams are left for convenience in joining onto the next course. The head and first course are now lowered onto the next course, which has been assembled in readiness. A few bolts are put through the punched holes, and the two courses are hoisted up, drilled and riveted in the way already described for the rear course. When all the courses are riveted together the front head is put in with the flange out so that the rivets in that flange can be driven on the machine. The closing seams on a boiler which, like the Scotch boiler, has both heads set with the flange in, must be riveted by hand. Rivets are heated in a small forge near the riveter and are passed to a man inside the boiler, who picks them up in tongs, thrusts them through the holes from within and guides the head of a rivet up to the die which is inside the boiler. Sometimes the rivets are thrust through from without, in which case the man inside the boiler guides the point to the die. On the platform of the machine stand the riveter and two or three helpers. They adjust the boiler so that the rivet is brought between the dies, and the riveter pulls the 288 STEAM-BOILERS. lever which controls the ram, and the outer die is driven against the rivet, forming the head and closing up the rivet in the joint. The holes are drilled about one sixteenth of an inch larger than the rivets. The pressure of the dies varies from 20 to 70 tons, depending on the thickness of the plate ; enough to compress the rivet and fill the hole completely. The rivets, as they cool, shrink and draw the plates firmly together. Riveting-machines. — There are four types of riveting- machines used for boiler-work, depending on the method of moving the ram or plunger which carries the movable die. The motion may be derived from — 1. A cam and toggle. 2. A hydraulic cylinder, 3. A combination of a hydraulic cylinder with a cam and toggle. 4. A steam-cylinder. The cam a?id toggle riveter is now seldom used. In it the ram carrying the movable die is driven by a toggle-joint that is closed by a cam, which in turn is driven by a belt and gearing. The adjustment for different thicknesses of plate is made by a wedge behind the ram, which can be set by aid of a screw. The pressure on the rivet is controlled by the elas- ticity of the frame of the machine and the setting of the wedge ; it cannot be regulated satisfactorily. The hydraulic riveter, in one form or another, is most com- monly used at the present time. With it a definite pressure can be applied to each rivet whatever the thickness of plate. Fig. 125 represents a hydraulic riveter with a reach of 96 inches which can apply a pressure of 150 tons. It consists essentially of two heavy cast-iron levers or beams, bolted together near the middle and at the lower end. One beam carries the fixed die at its upper end ; the other carries the ram and hydraulic cylinder. The stroke of the ram can be adjusted and is controlled by a single lever. The ram moves SHOP-PRACTICE. 289 in straight girders, and may apply an eccentric pressure with- rotatin^ or springing. Some hydraulic riveters have a hydraulic closing device Fig. 125 for holding the plates together while the rivets are driven. Even when furnished it is commonly not used. The reach of a riveting-machine is the distance from the dies to the bed-plate at the middle of the machine. It limits the width of plate that can be riveted by the machine. A portable hydraulic riveter is shown by Fig. 126, which has a reach of 12 inches and can apply a pressure of 75 tons. It can be swung into position by a crane and can be turned to any angle by the gear at the trunnion. This type of ma- chine is used largely for bridgework ; it is sometimes used 290 S TEA M-BOILERS. for riveting nozzles, manhole-rings, brackets, and reinforcing- plates onto boilers. The power for working a hydraulic riveter is derived from either a steam-pump or a power-pump. A heavy geared power-pump is shown by Fig. 127; it is run continuously and delivers water to an accumulator from which water is supplied to the hydraulic cylinder which moves the ram. The accumulator consists essentially of a loaded piston or plunger. Water is pumped into the cylinder of the accu- mulator, and is drawn out by the hydraulic cylinder as needed. When the accumulator reaches the end of its stroke it closes SHOP-PRACTICE. ! 9 I a valve on the pipe from the pump so that it receives no more water; at the same time it opens a by-pass from the delivery to the suction of the pump which continues to run, but has at that time very little resistance to overcome. When Fir. 127. some water has been withdrawn from the accumulator the by- pass is closed and the valve on the delivery-pioe is opened. When a steam-pump is used there is a device for shutting off steam from the pump when the accumulator is near the end of its stroke, and letting it on again when more water is required. An accumulator, shown by Fig. 128, is loaded by scrap- iron in a plate-iron cylinder. Inside the plate-iron cylinder is 292 S TEA M-BOILERS. a cast-iron cylinder which is closed at the top and which moves, on a fixed plunger. This plunger passes through a stuffing- box and is carried by a cast-iron bed-plate. When water is. Fig. 128. pumped into the cylinder through a passage in the fixed plunger, the whole weight of the cylinder, plate-iron casing, and scrap-iron load are lifted. The pressure required to da SHOP-PR A CTICE. 293 this depends on the load; it is the pressure which is exerted on the plunger of the hydraulic cylinder moving the ram. The frame of I beams at the sides forms a guide for the accumulator-cylinder and its load. Another form of accumulator, loaded with heavy cast-iron blocks and without any exterior guides, is shown by Fig. 1 29. Fig. 129. The hydraulic riveter with toggle and cam combines the simplicity of the cam-and-toggle machine with the advantage of a definite and determinable pressure on the rivet, which is the best feature of the hydraulic machine. The toggle bears against the ram at the front end, and against the plunger of a hydraulic cylinder at the back end. The cylinder is connected with an accumulator which is loaded to give the desired pres- sure on the rivet. Suppose that pressure to be 30 tons; then 294 S TEA M-B OILERS. when the cam closes the toggle, the rear end, resting against the hydraulic plunger, remains at rest, and the front end drives the ram and compresses the rivet till a pressure of 30 tons is reached. When that pressure is reached the hydraulic plunger yields, forces water into the accumulator and raises the load on it. When the cam releases the toggle, the hy- draulic plunger moves forward and the load on the accumula- tor falls and drives water into the cylinder. The stroke of the hydraulic plunger may be very short, as the principal part of the stroke of the ram is made before the piunger yields. Fig. 130. There is no loss of water except by leakage, which may be made up from time to time by a hand-pump. This machine gives a definite pressure on the rivet whatever the thickness of the plate, like the plain hydraulic riveter. It has no pump and the accumulator is smaller. If the plunger has a large area, the load on the accumulator need not be very great. A steam-riveter, shown by Fig. 130, has the same exter- SHOP-PRACTICE. 295 nal appearance as a hydraulic riveter, except that the power is applied by the direct pressure of steam on a piston, which must have a large area to give sufficient pressure to drive the rivets properly. The steam-valve is balanced so that it can be easily moved by the working- lever. If the valve is opened slowly, the ram is first moved forward against the rivet and then full pressure is applied to close the joint; but if the valve is opened promptly, the ram strikes a blow like that of a hammer. There is no reason why this cannot be guarded against if the valve is small and the machine is operated care- fully. The fact that the machine is commonly so used that it strikes a blow, and the fact that it is wasteful of steam, have brought the steam-riveter into disrepute except for small or for portable machines. The ram is moved back by the steam before escaping, after a rivet is driven. Hand-riveting. — In a modern boiler-shop almost all the riveting is done by machine because it is cheaper and, espe- cially on heavy work, is more likely to be well done. There are, however, a good many rivets on any boiler that must be driven by hand. In such case the rivet, which may be heated entirely or at the point only, is thrust through the hole from within and is held up by a man inside, who has for this pur' pose a hammer or weight which weighs about 20 pounds on a long handle. He has also an iron hook which he hooks into a rivet-hole, and against which he gets a purchase to hold the rivet up while it is driven. Two men with hammers that weigh about 5 pounds drive the rivet, striking in turn. A few heavy blows are struck to close the joint and partially form the head, then the head is finished in the shape of a straight- sided cone with lighter hammers. If the rivet is long enough to form a good head, and if it is driven with care and skill, hand-riveting may be equal to machine-riveting. If the heads are ill-formed, or if they are too low, the work may be very inferior. Snap-riveting. — This method of riveting, which is espe- 296 5 TEA M-B OILERS. cially convenient ior driving rivets in contracted spaces, has some resemblance to machine-riveting. The rivet is thrust through the hole and held up from within the boiler. The joint is closed and the head is roughly formed by a few blows of a heavy hammer, then a snap or die is held on the rivet and driven with sledge-hammers. For large rivets the sec- tion of the snap should be a parabola, and the head should be relatively small in diameter and high, because this form causes the rivet to fill the hole better and makes sounder work. Tube-expanders.— The tubes are expanded into the tube- sheets to make a steam-tight joint, beginning at tne least acces- sible end. They are commonly a little too long and are cut off at the projecting end by a tube-cutter. The tubes extend through the tubes a slight amount, and are beaded over, after they are expanded, by a special tool. The expanders most commonly used are known as the Prosser and the Dudgeon expanders. The Prosser expander, represented by Fig. 131, is made up ig. 131. of a number of steel segments held in place by a spring on a cylindrical extension of the segments. The acting part of the segments have the form to be given to the tube after it is expanded. The inside of the segments forms a straight hol- low cone into which a steel taper pin fits. The expander is forced into the tube and is expanded by driving in the pin with a hammer. This should be done gradually so as not to distress the metal of the tube too much, and the expander should be frequently slacked back and shifted part way round on account of the spaces between tne segments. SHOP-PRACTICE. 297 The Dudgeon expander, Fig. 132, has a set of rolls, three or more, in a frame. The rolls are forced out against the sides Fig. 132. of the tube by driving in a taper pin. The pin and frame are rotated as the pin is driven, and the rolls gradually force the tube against the tube-plate. Although the two expanders accomplish much the same result, the action is different. The Prosser causes an abrupt Fig. 133. Fig. 134. stretching of the tube and leaves the tube as shown by Fig. 133, bearing at the corners of the plate only. The Dudgeon enlarges the end of the tube and makes it bear against the en- tire thickness of the tube-sheet. Alter the tubes are expanded the ends are beaded over bv 298 STEAM-BOILERS. a special tool, as represented in both figures, which adds to their grip on the plate when they act as stays. A vacuum may possibly be found in a boiler, if it is allowed to cool without admitting air. The Prosser method has an advantage in such case, when the tubes act as struts between the heads. The Dudgeon method will then act by friction only. The rollers might be shaped to give an expansion just inside the plate, instead of making them straight; there is, however, no evidence of trouble from this source in practice. Calking. — The riveted seams of a boiler are made steam- tight by calking, which consists in driving the lower part of the planed edge forcibly against the plate beneath. Fig. 135 shows the form of calking-tool used in hand-calking, the posi- Fig. 135. tion in which it is held, and the way the extreme edge of the plate is compressed against the plate beneath. The acting sur- face of the tool, which is about an inch wide, is ground at an angle of somewhat less than 90 , and the edge is rounded slightly so that it will not cut the lower plate. The tool is slid along the under plate against the edge of the upper plate and struck with a hammer. If the tool is ground to a sharp edge and used carelessly, a groove may be cut in the under plate and serious injury may be done. A pneumatic calking-machine or tool is now used for doing most of the calking in boiler-shops. In general prin- ciple it resembles a rock-drill, and consists of a cylinder in SHOP-PR A C TICE. 2 99 which works a piston and rod on the end of which is the calking-tool. Air is supplied for working the piston, at a pressure of 60 or 80 pounds, through a flexible tube. It makes about 1500 working-strokes a minute, 3/16 of an inch long. The calker, which is about 2\ inches in diameter out- side and 15 inches long over all, is held by a workman who presses it slowly along the seam to be calked. The edge of the tool is well rounded so as not to injure the lower plate. Work can be done four times as rapidly with the pneumatic calker as by hand. Cold-water Test. — After the boiler is calked it is tested to about once and a half the working pressure, with cold water. During the test the boiler is carefully watched to detect any notable change of shape or other sign of faulty design or construction, and important leaks are marked; small leaks are of no consequence, as they will fill up with rust. Important leaks must be calked after the pressure is relieved ; if necessary, pressure may be applied again to see if they are stopped. The method of making this test and the precautions to be observed are given on page 224. If the boiler is examined by a boiler-inspector, he makes his inspection before the boiler is painted, and stamps certain letters on the head or over the fire-door to show that the boiler has passed inspection. Finally the boiler is painted and oiled ready for shipping. CHAPTER X. BOILER-TESTING. The main object of a boiler-test is to determine the amount of water evaporated per pound of coal, or, more ex- actly, the amount of heat transferred to the boiler per pound of coal burned. For this purpose it is necessary to deter- mine : i. The number of pounds of water pumped into the boiler during the test. 2. The number of pounds of coal burned, and the weight of ashes left. 3. The temperature of the feed-water when it enters the boiler. 4. The pressure of the steam in the boiler. 5. The per cent of moisture in the steam discharged from the boiler. It is desirable to determine the conditions of combustion, such as the draught, the weight of air supplied per pound of coal, the composition of the products of combustion, and the temperature of the escaping flue-gases. It is also desirable to have determinations made of the composition of the coal and its total heat of combustion, but, as was explained in Chapter II, these determinations should usually be intrusted to a chemist and to a physicist. Water. — The best and most satisfactory way is to weigh the feed-water directly, in proper tanks or barrels on scales. There should be two barrels or tanks large enough so that the filling, weighing, and emptying may proceed without haste. BOILER- TES TING. 3 O I The scales should be adjusted and tested with a standard weight and should be known to be correct and sensitive. Good commercial platform scales are sufficient for this pur- pose. The weighing-barrels should be placed high enough to discharge into a tank or reservoir from which the feed-water is drawn by a pump or injector. This tank should hold more than both weighing-barrels, so that when it is about half empty an entire barrelful of water may be discharged into it without danger of overfilling it and wasting water. The bar- rels are emptied through large quick-opening lever- valves; this point should receive attention, as any delay caused by small valves is vety annoying. The weighing-barrels are filled either from a water system or by a special pump from a well or reservoir. When a direct- acting steam-pump is used, a quarter-inch by-pass should be carried from the delivery-pipe to the suction-pipe; the pump will then run slowly when the valves on the pipes leading to the weighing-barrels are shut ; when one of these valves is opened the pump starts away promptly, and it slows down again when the valve is shut. If a power-pump is used, it may be convenient to arrange so that it shall run all the time at full power, discharging into the well or reservoir when neither barrel is filling. Weighing water, though simple enough, requires care and intelligence, as any blunder will spoil the test. The observer should proceed systematically. He will naturally start with both barrels filled, weighed and recorded before the test begins. When the level in the feed-tank has fallen so that it can receive a barrelful of water he will open the discharge- valve from one barrel, which should be marked and designated as Barrel No. i. When that barrel is emptied, he will close the valve and weigh the barrel ; the weight empty is set down and subtracted from the weight full to get the weight dis- charged. The record of weights is kept in a table con- 302 Sl'EAM-BOILERS. taining columns for the name of the barrel, weights full, weights empty, weights discharged, and time at which dis- charged. The weight of the barrel empty must be taken each time, as the barrel will not drain completely in the time that can be allowed. Water may now be turned on to fill Barrel No. I, and Barrel No. 2 may be emptied, as occasion demands. Then one barrel may be filling when the other is emptying, and the work may proceed rapidly but without confusion. The errors that a novice is liable to are either to forget to record the weight of a barrelful of water, or to empty a barrel that has not been weighed. It is convenient and almost necessary to have some sort of an index or telltale to show the water-weigher where the water-level is in the feed-tank. For this purpose we may use a float, with a string that runs up over a pulley and is kept taut by a small weight moving over a scale, which is placed in front of the weighing-barrels. This float is not used to determine the level of the water in the feed-tank at the begin- ning and end of the test. At the beginning of the test the level of the water in the feed-tank is marked, and at the end of the test the level is brought to the same mark, so that all the water delivered by the weighing-barrels is drawn out of the feed-tank by the feed-pump. A good way of marking the water-level is to fasten to the side of the tank a piece of wire bent into a hook, with its point projecting slightly above the water-level. This hook will commonly be placed in position before the test begins, and the tank will be filled up to the level so marked before water is drawn from the feed-tank. If water cannot be weighed directly, it may be measured in tanks of known capacity which are alternately filled and emptied. Or the water may be measured by a good water- meter, which must be tested under the conditions of the test to determine its error. Care must be taken to keep the meter BOILER- TES TING. 303 free from air or it will record more than the amount of water which actually passes. Boiler-tests on steamships can scarcely be made without using meters. At the time when the test begins, the water-level is noted at the water-glass, and at the end of the test the water-level is brought to the same place. The best way is to fix a wooden scale near the water-glass and record the height of the water above an arbitrary point on the scale. Sometimes a string is tied around the glass at the water-level when the test is started ; in such case the distance of the string from some fixed point on the fittings of the water-glass must be recorded, so that the string can be replaced if it happens to be moved or if the glass tube breaks. If the water is not brought exactly to the same level at the end as at the beginning of the test, the difference is noted and allowance is made. It has already been pointed out that the apparent height of the water depends to a certain extent on the rate of vaporization and on the rapidity of circulation in the boiler; consequently the boiler must be making steam at the same rate at the times when the water-level is observed for beginning and ending the test. All pipes leading water to or from the boiler, except the feed-pipe, must be disconnected. Steam may be taken for any purpose and through any pipe, so far as the boiler-test is concerned. Frequently the steam used by an engine is determined by weighing the feed-water for a boiler which is used exclusively for that engine. If the boiler is fed by an injector, the steam for running the injector should be taken from the boiler, for it will be condensed by the feed-water and returned to the boiler. A very small amount of the heat (less than two per cent) in the steam supplied to an injector is used in pumping the feed-water; the remainder is used in heating the feed- water and is returned to the boiler. The temperature of the feed-water must be taken before it goes to the injector. If the 304 STEAM-BOILERS. boiler is fed by a direct-acting steam-pump, that pump should be run with steam taken from some other source. If that cannot be done, then the steam used by the pump must be determined and allowed for, unless the exhaust from the pump can be turned into and condensed by the water in the feed-tank, in which case the pump is in the same condition as an injector. The best way of determining the amount of steam used by a steam-pump is to condense it in a small sur- face condenser, and to collect and weigh the condensed water. Or the steam may be run into a barrel filled with cold water, which is weighed before and after steam is run in. This method requires that the barrel shall be emptied when the water begins to vaporize, and filled afresh with cold water. Steam used by a calorimeter for determining the amount of water in steam must be ascertained also; the methods will be given in connection with a description of the instruments. Coal and Ash. — The coal required during a boiler-test should be brought in as required in barrows; it may be fired from the barrow or dumped and fired from the floor, 'i ne barrow should be weighed full and empty, and the difference should be recorded together with the time ; the latter to serve as a check on the record and make sure that a barrow-load is not neglected. The weight of the barrow is usually the same throughout the test. Any coal left unburned is weighed back It is essential that the condition of the fire shall be tne same at the beginning and at the end of the test. There are two methods in vogue for trying to attain this result ; if the test is 24 hours long or more, the condition of the fire is esti- mated by its appearance; if the test is 10 or 12 hours long, the test is started and stopped with the grate empty. These are for tests of factory boilers with a combustion of 15 to 20 pounds of coal per square foot of grate per hour. For tests on marine or locomotive boilers, where the rate of combustion may be twice or five times as rapid, the duration of a test may be correspondingly reduced. BOILER- TES TING. 305 Coal in solid mass will weigh 70 or So pounds to the cubic foot; when lying on a grate it will weigh 50 or 60 pounds. It is difficult to estimate the thickness of the bed of coal on a grate nearer than two inches. But a layer of coal two inches thick will weigh 8 or 10 pounds, which is about half the rate of combustion for a factory boiler. If a test is only ten hours long, the error resulting from a wrong estimate of the thick- ness of the fire may readily be five per cent. If the test lasts twenty-four hours, the error will probably not be more than two per cent, provided a proper method is used. If the condition of the fire is estimated at the beginning and end of the test, the fire should be cleaned and freed from ashes and clinker shortly before the test begins, and should then be spread in rather a thin even layer of clean glowing coal. Its height above the grate should be estimated with reference to some mark in the furnace that can be recognized readily. Just as long before the end of the test the fire should be cleaned and levelled in the same manner, and the thickness should be estimated with reference to the mark chosen at the beginning. The fireman is sure to have a clean bright fire at the beginning of the test, but he is apt to have a fire with much the same appearance that is half clinker at the end. The error from estimation may be very serious in such case, even though the test is 24 hours long. If the test is started and stopped with the grate empty, the boiler must be brought into good working condition about an hour before the test is to start, with all the brickwork thoroughly heated. The fire is allowed to burn low, and the steam-pressure is maintained by reducing the draught of steam from the boiler. Twenty or thirty minutes before the test starts, the fire is drawn or dumped and the grate and asii- pit are cleaned out. A new fire is started with wood, and coai is thrown on as soon as the wood is well alight. The time when coal is thrown on is counted as the beginning of the test. If the steam-pressure falls while the fire is drawn, 306 S TEA M-B OILERS. the stop-valve may be nearly or quite closed to keep it from falling much below the working-pressure. Toward the end of the test the fire is allowed to burn low, and at the end of the test it is drawn out on the boiler-room floor and quenched with as little water as may be, not enough to leave it wet. The unburned coal is picked out by hand and weighed back, the clinker and ashes are separated and weighed together with the clinker withdrawn during the test and the ashes in the ash-pit. If any appreciable amount of coal falls through the grate, a sample from the ash-pit may be picked over by hand to es- timate the proportions of unburned coal in the ash. The coal in the ash is allowed for in calculating the per cent of ash in the coal, but is not added to the coal weighed back, for there is no way of burning coal thus lost through the grate. When a test is started with a wood fire, more or less coal is apt to fall through the grate in starting. This is drawn from the pit and fired over again. It is customary to allow the fire to burn low before draw- ing the fire at the end.of the boiler-test, both because it brings the fire more nearly to the condition at the beginning, and because it is a hard and unpleasant job to draw a thick fire. But the fire should be maintained at its normal condition until the end of the test approaches, and should be a good fire when drawn. Extraordinary results may be obtained by allowing the fire to burn nearly out at the end of the test, a very considerable amount of steam being formed by heat given out by the boiler-setting. It is unnecessary to say that such results are entirely misleading. The wood used for starting the fire is weighed and allowed for on the assumption that a pound of wood is equivalent to 0.4 of a pound of coal. The total weight of wood used is not large. Temperature of Feed-water. — The temperature of the feed-water is taken by a thermometer in a cup filled with oil, screwed into the feed-pipe close to the check-valve. If the BOILER- TES TING. Z°7 temperature varies, it may be read every five minutes: if it is found to be steady, less frequent intervals will do. Pressure of Steam. — The steam-pressure must be very nearly the same at the beginning and end of a test, and should remain nearly constant throughout the test. Read- ings are commonly taken every fifteen minutes, but the fire- man should be required to keep the pressure nearly constant at all times. The steam-pressure is taken by a spring-gauge like that shown by Fig. 92 on page 252. The gauge should be compared with a mercury column or a standard gauge both before and after the test, and a correction should be applied if necessary. If the pipe carrying pressure to the gauge fills up with water, allowance for the pressure of that column of water must be made. Each foot of water will give a pressure of about 0.43 of a pound per square inch. The reading of the barometer should be taken two or three times during a test. The reading in inches of mercury can be reduced to pounds per square inch by multiplying by the weight of a cubic inch of mercury, which is about 0.491 of a pound. Very commonly the pressure of the steam is obtained indirectly by aid of a thermometer set in the steam- pipe. The absolute pressure corresponding to the temperature is then obtained from a table of the properties of saturated steam. The thermometer is readily standardized, and is not so likely to become unreliable as a steam-gauge. Most vertical boilers and some water-tube boilers give superheated steam ; in such case there shouid be both a thermometer and a gauge on the steam-pipe, to indicate tem- perature and pressure. The excess of the temperature by the thermometer above that corresponding to the absolute pressure of the steam, as found in a table of properties of steam, is the degree of superheating. Specific Heat of Superheated Steam. — The mean value °8 s TEA M-B OILERS. given by Regnault for the specific heat of superheated steam is 0.4808, or approximately 0.48. This property of steam can be used in calculating the amount of heat in steam due to superheating. For example, let the pressure by the gauge be 65.3 pounds, and let the temperature be 350 F. by the thermom- eter. The absolute pressure corresponding to 65.3 pounds is 80 pounds, at which saturated steam has the temperature of 31 1°. 8 F. The superheating is consequently 350 F. - 31 1°. 8 F. = 38°.2 F. The heat due to the superheating is 0.48 X 38.2 = 18.3 B. T. U. When the steam is superheated, the formula for equiv- alent evaporation is changed from the form given on page 135 to Wl o.48(*,-Q + *-+?-?■ 1)0 — — , 965.8 in which t s represents the actual temperature of the super- heated steam, and t is the temperature corresponding to the absolute pressure of the steam determined from the reading of the gauge. Priming. — A boiler which has sufficient steam-space and free water- area will deliver steam which contains less than two per cent of moisture. Professor Denton* has pointed out that a jet of steam blowing into the air from a petcock will give a characteristic blue color if there is less than two per cent of water in the steam. If there is more than two per cent of moisture, the jet will be white. Since steam seldom contains less than one per cent of moisture under the usual conditions of ordinary practice, it is possible by this method to estimate the condition of steam with a probable error of one per cent. * Trans. Am. Soc. Mech. Engs., vol. x. p. 349. BOILER- TES TING. 309 The most ready way of determining the condition of steam is by the aid of a throttling-calorimeter, devised by Professor Peabody,* which depends on the fact that the total heat of steam increases with the pressure, so that dry steam be- comes superheated when the pressure is reduced by throttling. If the steam is only slightly primed, superheating will still take place, and the amount of priming can be determined from the temperature and pressure of the steam after it is throttled. If there is much moisture in the steam, it fails to superheat. A good form of this apparatus is shown by Fig. 136, consisting of a reservoir A to which the steam to be tested^is admitted through a half-inch pipe b with a throttling-valve near the reservoir. The steam flows away through an inch pipe d. At f is a gauge for measuring the pressure, and at c there is a deep cup for a ther- mometer to measure the temperature. The boiler-pressure may be taken from a gauge on the main steam-pipe near the calorimeter. It should not be taken from a pipe in which there is a rapid flow of steam as in the pipe b, since the velocity of the steam will affect the gauge-reading, making it less than tne real pressure. The reservoir is wrapped with hair-ielt ana lagged with wood to reduce radia- tion of heat When a test is made the valve on the pipe d is opened wide (this valve is frequently omitted), and the valve at b is opened wide enough to give a pressure of five to fifteen pounds in the reservoir. Readings are then taken of the Fig. 130. Trans. Am. Soc. Mech. Engs., vol. x. p. 327. 310 S TEA M-BOILERS. boiler-gauge, of the gauge at/, and of the thermometer at e. It is well to wait about ten minutes after the instrument is started before taking readings, so that it may be well heated. The method of calculation can be readily understood from the following Example. — The following are the data of a test made with a throttling calorimeter: Pressure of the atmosphere 14.8 pounds. Pressure by the boiler-gauge ,.. 69.8 " Pressure by the calorimeter-gauge.... 12.0 " Temperature in the calorimeter 268°.2 F. The absolute pressure in the boiler was 69.8 + 14-8 = 84.6 pounds, at which the heat of vaporization is 892.7 B. T. U. and the heat of the liquid is 285.3 B. T. U. So that with x part of a pound steam (and 1 — x priming) the heat in one pound of moist steam was 892.8^+285.3, in which x was to be determined. The absolute pressure in the calorimeter was 12 -f- 14.8 = 26.8 pounds, at which the temperature was 243°.9 F>, and the total heat was 1 1 56.4 B. T. U. The heat due to superheating was 0.48(268°. 2 - 243°. 9 ) = 11. 7 B. T. U., and the heat in one pound of steam in the calorimeter was 1 156.4+ 1 1.7 = 1 168. 1 B.T.U. But the process of throttling neither adds nor subtracts heat, consequently 892.8^+ 285.3 = 1168.1, or x = o.< BOILER- TES TING. 3 I l and the priming was 100(1 — 0.988) == I, 2 per cent. The calculation can be conveniently expressed by an equa- tion in which r and q are the heat of vaporization at the abso- lute boiler-pressure, and Aj and t x are the total heat and the temperature at the absolute pressure in the calorimeter, all taken from a table of proportions of steam; while t s is the temperature of the superheated steam in the calorimeter. Then xr+#= A, + 0.48(4-/,,; x = ,i, + 0.48(4-0 -g r It has been found by experiment that no allowance need be made for radiation from the calorimeter if made as de- scribed, provided that 200 pounds of steam are run through it per hour. Now this quantity will flow through an orifice one fourth of an inch in diameter under the pressure of 70 pounds by the gauge, so that if the throttle-valve be replaced by such an orifice the question of radiation need not be con- sidered. In such case a stop-valve will be placed on the pipe to shut off the calorimeter when not in use; it is opened wide when a test is made. If an orifice is not provided, the throttle-valve may be opened at first a very small amount and the temperature in the calorimeter noted after a few min- utes; the valve may be opened a trifle more, whereupon the temperature will usually rise, showing too little steam used. If the valve is opened little by little till the temperature stops rising, it will then be certain that enough steam is used to reduce the error from radiation to a very small amount. Various modifications of the throttling-calorimeter have been proposed, mainly with a view of reducing its size and weight. Almost any of them will prove satisfactory in prac- tice, but some will be found to be liable to error from radia- 312 S TEA M-BOILERS. tion or from the fact that there is not sufficient opportunity for the steam to come to rest and properly develop the super- heating due to throttling. One great advantage of this instrument is that ordinary care with ordinary gauges and thermometers gives sufficient accuracy. For example, with ioo pounds absolute boiler-pressure and with atmospheric pressure in the calorimeter, an error of half a degree by the thermometer, or half a pound by the boiler-gauge, or a third of a pound by the calo- rimeter-gauge will each give an error of one-tenth of a per cent in the priming. If steam contains more than three per cent of priming, the amount of moisturecan be determined by a gooa separator, which will remove nearly all the moisture. It remains then to measure the steam and water sep- arately. The water may be best measured in a calibrated vessel or receiver, while the steam may be condensed and weighed, or may be gauged by allowing it to flow through an orifice of known size. A form of this instrument devised by Professor Carpenter* is shown by Fig. 137. Steam enters a space at the top which has sides of wire gauze and a convex cup at the bottom. The water is thrown against the cup and finds its way through the gauze into an inside chamber or receiver, and rises in a water-glass outside. The receiver is calibrated by trial so that the amount of water may be read directly from a graduated scale. Fig. 137. * Trans. Am. Soc. Mech. Eng?,, vol, xvit. p. 608. BOILER- TES TING. 3 I 3 The steam meanwhile passes into the outer chamber which surrounds the inner receiver, and escapes from an orifice at the bottom. The amount of steam may either be calculated, by a method to be explained, from the diameter of the orifice and the pressure of the steam, or it may be condensed and weighed or measured. The latter is the more accurate way, and it has the advantage that then there is no error from radiation, for the inner receptacle is well protected by the outer chamber, and condensation in the outer chamber is collected and weighed with the steam. If the instrument is well wrapped and lagged, and if a sufficient quantity of steam is used, then the error from radiation can be neglected, just as was found to be. the case with the throttlincr-calorimeter. This instrument, for want of a better name, is called a separator calorimeter; it is a question whether either it or the throttling- calorimeter are properly calorimeters at all, and whether it would not be better to call both priming-gauges. It is customary to take a sample of steam for the calori- meter or priming-gauge through a small pipe leading from the main steam-pipe. The best method of securing a sample is an open question ; indeed it is a question whether we ever get a fair sample. There is no question but that the com- position of the sample is correctly shown by either of the priming-gauges described. It is probable that the best way is to take steam through a pipe which reaches at least half- way across the main steam-pipe, and which is closed at the end and drilled full of small holes. It is better to have the samping-pipe enter the steam-pipe at the side or at the top of the main, so that any water that may trickle along the bottom of the main shall not enter the calorimeter. Again, it is better to take a sample from a pipe through which steam flows upward. The sampling-pipe should be short and well wrapped to avoid radiation. If the steam from the boiler can be wasted during the test, then the entire steam delivered by the boiler may be passed 3 1 4 S TEA M- BOILERS. through a large priming-gauge, and the difficulty of getting a sample may be avoided. Flow of Steam.— It has been shown by Rankine * that the flow of steam through an orifice into the atmosphere may be represented by an empirical equation, 70 in which Wis the number of pounds of steam per second, A is the area of the orifice in square inches, and p is the absolute pressure of the steam. This equation, which has already been mentioned in connection with safety-valves, can be applied only when the absolute steam-pressure is more than double the pressure of the atmosphere; that is, the pressure of the steam must be 15 pounds by the gauge, or more. Experiments made in the laboratory of the Massachusetts Institute of Technology f show that this equation is liable to an error of about two per cent, but this error may be deter- mined by direct experiment for a given orifice under various pressures, and then a correction can be applied which will reduce the error to a fraction of one per cent. It appears then that the use of an orifice to determine the amount of steam in Professor Carpenter's separator priming- gauge is at least questionable unless direct experiments are made to determine the correction to be applied. On the other hand, the amount of steam used by a throttling prim- ing-gauge may be very properly determined by allowing it to flow through an orifice, since the total amount of steam used by the calorimeter is small. The same equation may be used for calculating flow of steam from one reservoir to another provided that the pres- sure in the second reservoir is less than half that in the first * The Engineer, vol. XXVII. p. 359, 1869. f Trans. Soc. Am. Engs., vol. xi. p. 187. BOILER- TES TING. 3 I $ reservoir. This allows us to gauge small quantities of steam used for any purpose, at a pressure that is less than half the boiler-pressure; for example, for running a steam-pump. A convenient arrangement for gauging the flow of steam in an inch pipe consists of a reservoir three feet long, made up of three-inch piping, and fittings divided at the middle by a brass plate through which there is an orifice of proper size. If the pipe carries steam at ioo pounds absolute, at a velocity of ioo feet a second it will deliver 7id* 3-i4i6 X ( T V) 2 X 100 = — X 100 = 0.5455 4 4 cubic feet per second. The density or weight of one cubic foot of steam at 100 pounds absolute is 0.2271 pounds. So that the pipe will carry 0.5455 X 0.2271 = 0.124 of a pound of steam per second. If this weight is put for W in Rankine's equation, and if A is replaced by \ nd 2 , we shall have 7td 2 X 100 o. 124 = 4 X 70 or , /OI24. V 3-Hi X 4 X 70 _ l 6 X 100 3 of an inch, nearly, for the diameter of the orifice for gauging the flow of steam. With an orifice of approximately the right size, the flow of steam may be regulated by a valve below the gauging device; for example, by the throttle-valve of the pump. Flue-gases. — At frequent intervals samples of flue-gases should be taken from various places, such as back of the bridge, from the uptake, and from the chimney. These sam- 3i6 S TEA M-B OILERS. pies are analyzed as soon as may be by Orsat's apparatus, as described on page 56. Though not commonly done, it would be well if a con- tinuous sample could be taken in a reservoir from which samples for analysis could be taken at intervals. Draught-gauge. — The draught given by a chimney is seldom more than an inch or an inch and a half of water. It can be measured roughly by a simple U tube filled with water. An instrument for accurate determinations of draught should be at once simple and certain in its action. The draught-gauge shown by Fig. 138, devised by Prof. Fig. 138. Miller, has been used with satisfaction for this purpose. It consists of two pieces of three-inch brass pipe connected by a half-inch pipe at the bottom. One of the pipes is closed at he top and can be connected to the chimney by a small pipe with a valve as shown. The other piece of brass pipe is open and has a hook-gauge, reading to 1/1000 of an inch, suspended in it. In preparing for a reading, the closed tube or leg is B OIL ER- TES 7 VNG . 3 ' 7 shut off from the chimney and opened to the atmosphere ; the water then stands at the same height aa, a'a, in both legs. The closed leg is now shut off from the air and connection is made with the chimney, whereupon the level falls to bb in the. open leg and rises to b'b' in the closed leg. As the two legs have exactly the same internal diameter, the fall ab is half the draught, measured in inches of water. The hook-gauge is set to the level aa when the closed leg is open to the air, and to the level bb when it is connected to the chimney. The difference of the readings multiplied by 2 is the draught in inches of water. The reading by the hook-gauge can readily give an acuracy of i/iooo of an inch, which is sufficient for this purpose. Pyrometers. — The determination of high temperatures, as in flues and chimneys, is difficult and uncertain. Most commercial pyrometers, depending on the unequal expansion of metals, are unreliable if not misleading; not only is the scale of such a pyrometer likely to be incorrect, but the zero of the scale is liable to change during use. The Chatelier pyrometer has been used with satisfaction at the Massachusetts Institute of Technology for measuring temperatures in flues and chimneys. It consists essentially of a thermoelectric couple made by joining the ends of two wires, one of platinum and the other of platinum alloyed with ten per cent of rodium. All but about four inches of the wire at the junction is incased in fire-clay inside an iron pipe about four feet long. From the wires of the pyrometer con- nection is made to a sensitive galvanometer in a separate observing-room. The deflection of the galvanometer is indi- cated by a ray of light reflected from a mirror on the needle and moving over a graduated scale. The scale is set to read zero when the junction of the wires is at the temperature of the atmosphere. The junction is then immersed successively in baths of substances which melt at various high tempera- tures, such as sulphur and naphthaline. The readings of the 3 1 8 S TEA M-B OILERS. ray of light when the juncture is in such baths fix known points on the arbitrary scale from which intermediate tem- peratures may be estimated directly. It is convenient to use a curve for this purpose with scale-readings for abscissae and with corresponding temperatures for ordinates. After the scale is deterimned the pyrometer may be introduced into the place or places where temperatures are to be measured, and readings are taken from which the temperatures are deter- mined by interpolation on the curve just described. Air-supply. — The air for a furnace may be made to enter through a temporary mouthpiece fitted to the ash-pit doors. This mouthpiece irnay be of galvanized iron, circular in sec- tion and about three feet long. Its cross-section should have an area equal to that of the door or doors leading to the ash- pit. The velocity of the air passing through the mouthpiece can be measured by an anemometer. The area of the mouth- piece multiplied by the velocity in feet per second gives the volume of air supplied to the ash-pit in cubic feet per second. From this may be calculated the volume and weight of air supplied to the ash-pit per hour or for the entire test; which weight divided by the total coal consumption gives the air per pound of coal burned. It should be noted that the anemometer is liable to an error of from two to five per cent, and further that air enter- ing through the fire-doors and elsewhere than through the ash- pit is not measured. Sample Test. — The test given on page 319, made at the Massachusetts Institute of Technology, may serve as an example of a convenient arrangement for reporting the data and results of a boiler-test. The average pressure of the air and of the steam in the boiler are liable to vary slightly during the test; the average pressures were obtained from readings taken at regular inter- vals during the test. The same may be said of the tempera- ture of the feed-water. BOILER-TESTIXG. 319 BOILER-TEST. j) ATE Dec. 28, ""or—Jan. 2, 'Q2. Duration of Test £^ hours. Average pressure of air £fr Sl lbs. per sq. in. " gauge-pressure /00 -Q " " temperature of feed-water I32 - 02 F. Kind of coal used L ackawanna . Per cent of moisture in coa 1. Description of Boilers: Babcock cV Wilcox, No, 10S tubes 4" dia., if 8" lont; ; outside area 1. , 1 907 3 12 " 4" dia. ' 6' " S6.S 2 drums 3' dia., X H ; one-half 0/ shell, 100.2 Boiler No. J_\ Boiler No. , JNo.^,J^in. by^in.^ Gratc-surfaceX >• Area, feet (No.__, in. by in Water-heating surface, feet Ratio of water-heating surface to grate-surface Lbs. coal fired, including coal equivalent of wood Unburned fuel Coal burned, including coal equivalent of wood Average coal burned for '5 minutes Total refuse from coal Total combustible Average combustible for J 5 minutes Average lbs. of air for 7 ->" minutes Air per lb. of coal Air per lb. of combustible Quality of steam, saturated steam taken as unity Total water pumped into boiler and apparently evaporated.. Water apparently evaporated per lb. of coal burned Water actually evaporated, corrected for quality of steam Equivalent water evaporated into dry steam from and at 212° F Equivalent water evaporated into dry steam, from and at 212° F., per lb. of coal burned Equivalent water evaporated into dry steam, from and at 212° F., per pound of combustible Coal burned per sq. foot of grate surface per hour Water evaporated, from and at 212 F., per sq. foot of heat- ing-surface per hour ... * Fires not drawn. 43-it> 64.6JQ 64,630 126.2 S660 55-Q70 i5j3S o Q83 5 4*. 70+ 8.40 530,463 6/4.300 0.8 320 STEAM-BOILERS. The description of the boiler under item I is brief and yet sufficient to identify it, and gives the data for calculating heating-surface. The grate-surface, heating-surface, and their ratio are calculated from the dimensions of the boiler and furnace, and given in the 2d, 3d, and 4th items. The 5th item gives the total weight of coal fired; as the fires were not drawn, no wood was used and no coal was withdrawn at the end of the test. Consequently the 7th item, coal burned, is the same as the 5th. The 9th item gives the weight of all the clinker and ashes produced during the test. The coal burned, minus the refuse, gives the total combustible for the test, set down at item 10. The air-supply is calculated at intervals of 15 minutes during the test, from the anemometer readings and the condi- tion of the atmosphere as it enters the galvanized-iron tem- porary mouthpiece of the furnace. This is likely to vary considerably, being greatest immediately after fresh coal is fired. Item 12 gives the average from the several calculations during the test. The coal and combustible for 15 minutes given by items 8 and 1 1 are calculated for comparison with the air for the same time. Thus the air per pound of coal is calculated by dividing item 12 by item 8; and in like manner the 14th item is calculated from the nth and 12th. The quality of the steam was obtained from time to time during the test by a throttling-calorimeter, like the one for which a description and calculation are given on page 309. The average from the several determinations is given by item 15. The priming was 100(1.000 — 0.983) = 1.7 per cent. The equivalent evaporation for the total coal (given by item 18) was calculated, by a method like that given on page 133, from the temperature of the feed-water, the pressure in BOILER- TES TING. 3 2 1 the boiler, and the quality of the steam; using the total water apparently evaporated given by item 16. The absolute boiler-pressure was 109.9+ M-85 = 124.8 pounds. The corresponding heat of vaporization and heat of the liquid are 871.8 and 3 1 5 . 1 ; the heat of the liquid at I22°.9 (the temperature of the feed-water) is 91.0. Consequently the total equivalent evaporation from and at 212 F. was 548,794(0.983X^8+ 3.5.. -90 = 6l4>300 pounds . The equivalent evaporation per pound of fuel (item 20) is obtained by dividing the quantity just found by the total coal burned (item 7). In like manner the equivalent evapora- tion per pound of combustible is obtained from item 10. The coal burned per square foot of grate-surface per hour is obtained by dividing the total coal burned by the area of the grate and by the duration of the test. Thus 64,639 = 9.8 pounds. 51.3 X 128 The equivalent evaporation per square toot of heating- surface per hour (item 23) is obtained by dividing the total equivalent evaporation (item 19) by the heating-surface and by the duration of the test. Thus 6l 4.3QO - = 2.17 pounds. 2214 X Remark. — In this chapter are given the observations that are required and the precautions to be taken in making an ordinary boiler-test. It is, however, intended rather as a 322 STEAM-BOILERS. description for the student than as a guide for the engineer, who must learn how to make tests by experience. Many of the processes and observations are so simple that they may be intrusted to any careful and intelligent person; the conduct of the test must receive the attention of a competent engineer, for there is no expert work that an engineer may be called upon to do in which there is more chance for error and deception than in making a boiler-test. CHAPTER XI. BOILER DESIGN. In order to bring together the principles and methods which have been given in the preceding chapters, they will be applied to the design of a boiler. Designing of any sort is an art that is guided and controlled by practical considerations and theoretical principles, and which can be acquired by prac- tice only. The design of a boiler, like many other designs, is further modified to meet the requirements of government boards of inspection, or to conform to the inspection-rules of insurance companies. These rules and requirements vary from place to place and from time to time ; they must be known to the designer, but they have no place in a text-book. A simple and common type of boiler has been chosen for design ; the methods, with proper modification, can be applied to other types, and the general principles illustrated are much the same for all types. Type of Boiler. — The kind of boiler used in a given locality depends on custom, on the kind of water used, and on the cost and quality of fuel. Deviation from common prac- tice should be made only for sufficient reason. Where water is bad or where fuel is cheap, the plain cylindrical boiler or a flue-boiler will be chosen. With clean, soft water the cylin- drical tubular boiler, like that shown by Plate I, has been found to be convenient, economical, and cheap. All these boilers have external furnaces, so that the shell is in part exposed to the fire. Now plates exposed directly to the fire should not be more than half an inch thick ; 3/8 of an inch is preferable. Though thicker plates are sometimes used, this 323 324 S TEA M-B OILERS. consideration limits the size of boilers of this type when high pressures are used. The importance of high efficiency for the longitudinal riveted joint becomes apparent in this connec- tion. Internally-fired boilers, like the Lancashire or the Scotch marine boiler, are not limited in diameter by this reason. The marine boiler sometimes has plates an inch and a quarter thick ; the fact that so great a thickness is undesirable some- times serves as a check on the size of such boilers. General Proportions. — Whatever may be the type of boiler chosen, there must be provided — 1. Sufficient grate-area to burn the fuel required under the available draught. 2. Suitable combustion-space to properly burn the fuel. 3. Sufficient area of flues or tubes to carry off the products of combustion. 4. Sufficient heating-surface to absorb the heat generated. 5. Proper water-space to prevent too great a fluctuation of the water-level when there is an irregular demand for steam. 6. Suitable steam-space to prevent too great a fluctuation of pressure when steam is taken at intervals, as for the cyl- inder of a steam-engine. 7. Sufficient free-water area for disengagement of steam. The last three conditions are not fulfilled by most water- tube boilers; some such boilers depend on a separator for disengaging steam from water. Problem for Design. — Let it be required to determine the main dimensions and some of the details of a hori- zontal cylindrical tubular boiler to develop 80-horse power A. S. M. E. standard (page 135). Let the working-pressure be 150 pounds per square inch by the gauge, and the test- pressure 225 pounds, or once and a half the working-pressure. Assume that anthracite coal will be used, and that it will give an equivalent evaporation of 9 pounds of water per pound of coal from and at 212 F. Assume further that 12 BOILER DESIGN. 325 pounds of coal will be burned per square foot of grate-surface per hour. The heating-surface may be about thirty-seven times the grate-surface. Tubes 16 feet long will be used, which length should not much exceed sixty times the diameter. The area through the tubes will be made about 1/7.5 °f the grate-area. Grate - area. — The A. S. M. E. standard requires that 34.5 pounds of water per hour shall be evaporated from and at 212 F. for each horse - power. The total equivalent evaporation will consequently be 80 X 34.5 = 2760 pounds per hour. With an equivalent evaporation of 9 pounds of water per pound of coal the coal burned will be 2760 -T- 9 = 307 pounds per hour. With a rate of combustion of 12 pounds of coal per square foot of grate surface per hour, the grate-area must be 307 -5- 12 =25.6 square feet. Tubes. — A common rule for finding the diameter of tubes is to allow one inch for each four feet of length when soft coal is used, and five feet when hard coal is used. A tube three inches in diameter will very nearly fulfil this condition. The table of proportions of flue-tubes in the Appendix, gives the area of the internal transverse section of such a tube as 6.08 square inches; the external area is 7.07 square inches. The internal circumference is 8.74 inches, and the external circumference is 9.42 inches. 326 S TEA M-B OILERS. The aiea through the tubes has been chosen as 1/7.5 °f the grate-area, equal to 25.6 X 144 . , = 402 square inches. 7-5 Since the area through one tube is 6.08 square inches, there will be required 492 -=-6.08 = 80.8, or, more properly, 81 tubes. It may be found convenient in laying out the tube-sheet to use more than this number of tubes; a less number is of course improper. Steam-space. — A good rule for this type of boiler is ta allow from 0.8 to 1 cubic foot of steam-space per horse- power, which gives from 64 to 80 cubic feet for this boiler. We will assume 80 cubic feet. For sake of comparison, calculations will be made also by rules given on page 132. Thus for certain boilers working at moderate pressures it is found that the steam-space may be made equal to the volume of steam used by the engine in 20 seconds. Suppose that this boiler, though designed for 150 pounds pressure, may run at 70 pounds pressure, and may supply an 80 horse-power engine which uses 30 pounds of steam per horse-power per hour. Now the volume of one pound of steam at 70 pounds by the gauge, or 85 pounds absolute, is 5.125 cubic feet. So that the engine will use 80 X 30 X 5-125 = 12,300 cubic feet of steam in an hour, or 20 ra X 12300 = 68 3600 cubic feet in 20 seconds. This is about the lower limit by the rule used above. It is clear that the steam-space would BOILER DESIGN. $2 J be very small if determined by this rule for an engine using steam at 150 pounds pressure. Another rule makes the steam-space from 50 to 140 times the volume of the high-pressure cylinder of the engine; 50 for very high pressure and high speed, 140 for slow speed and low pressure. For medium speeds and pressures 60 to 90 may be used. The boiler under consideration may supply steam to a triple-expansion engine which has a high-pressure cylinder 9 inches in diameter by 30 inches stroke, so that the volume is 1. 105 cubic feet. According to this the steam-space needed is 66 to 99 cubic feet. Diameter of Boiler. — For this type of boiler the steam- space is commonly made one third and the water-space two thirds of the contents of the boiler. To the contents of the boiler there must be added the space occupied by the tubes to find the volume of the cylindrical shell. Now we have de- cided to use 81 tubes 3 inches in diameter and 16 feet long. The area of the external transverse section has been found to be 7.07 square inches. The space occupied by the tubes is consequently 81 x 7.07 X 16 44 = 64 cubic feet. To this add steam-space, 80 and water- space, 160 Making in all, 304 " iC The cylinder is 16 feet long, so that its transverse area is 304 -r- 16 = 19 square feet; which corresponds to a diameter of 59.02 inches, or nearly 60 inches. This will be taken as the trial diameter; it may re- quire change in proportioning other parts of the boiler. The method of determining the main dimensions of a 328 S TEA M-B OILERS. boiler from the steam-space will require modification if it is applied to any other type of boiler. Even when applied to a given type it leaves much to the judgment of the designer, who may find difficulty in using it unless he is accustomed to working on that particular type. If the designer has at hand the dimension of several boilers of a given type, he may pre- fer to select the main dimensions for a new design directly, with the reservation that such dimensions may be modified as the design proceeds. This is commonly done by the designers of marine and locomotive boilers. Heating-surface. — The heating-surface of a cylindrical tubular boiler consists of all the shell below the supports at the side wall, all the inside of the tubes, and part of the rear tube-plate. Usually half of the cylindrical part of the shell is heating-surface. In the case in hand the heating-surface, exclusive of the tube-plate, will amount to au u r w 3-H 16 X 60 X 16 Shell - )< ^— = 125.7 sq. ft. ^ T „ 8 -74 X 16 Tubes.... 81 X —^ = 943.9 " " Total 1069.6 " " The grate-surface is to be 25.6 square feet, so that the ratio of grate-surface to heating-surface will be at least as good 25.6 : 1069.6 :: 1 : 41^. The actual ratio will be more favorable as it will appear advisable to use more than 81 tubes, and the back tube-sheet remains to be allowed for. Water-level. — It is now necessary to determine the posi- tion of the water-level to see if there will be sufficient free- water surface and sufficient distance from the water-level to the shell above it. BOILER DESIGN S 2 9 Since the whole boiler is cylindrical, the area of the head of the boiler exposed to steam and to water will have the same ratio as that of the steam-space to the water-space. Consequently the area of the head above the water-level must be one third of the total area of the head less the combined areas of the tubes. The area of a circle having a diameter of 60 inches is 2827.4 square inches. The area of 81 tubes each having an external cross-section of 7.07 square inches will be 81 x 7-07 = 572.7 square inches. The area of the head exposed to steam is consequently 2827.4- 572.7 = square inches. We need now to know the height of a seg- ment of a 60-inch circle, which has the area of 751.6 square inches. The second problem in the explanation of the use of a table of segments (see Appendix; gives for the tubular number corresponding to the area 75K6 0.2088; 60 X 60 for which the ratio of the height to the diameter is 0.312. The height of the segment is therefore 0.312 X 60 = 18.7 inches. This gives sufficient height above the water, and sufficient free-water surface. The water-level will be 30 - 18.7 = 11. 3 inches above the centre of the boiler. Factor of Safety. — It has been pointed out that the actual factor of safety of boiler-shells is usually four or five when the boiler is built. The apparent factor of safety for some parts 330 STEAM-BOILERS. like stay-bolts may be greater, but such factors are illusory because the stays may be subjected to considerable irregular stress from unequal expansion. The apparent stress on stay- rods and bolts, from steam-pressure only, is frequently limited by inspection-ruies or by law. The factor of safety of a boiler which has been at work for some years is much affected by corrosion, which acts upon different parts of the boiler very differently, even when the corrosion is uniform. Thus a plate half an inch thick will have 7/8 of its original strength after it has lost 1/16 of an inch by corrosion. The weakest part of the plate, that is, the riveted joint, seldom suffers as much from corrosion as the whole plate at a distance from the joint, because the plate is. protected to some extent by the rivet-heads. Some forms of joint have an internal cover-plate, which protects the plate at the joint and the joint may be nearly as strong after corrosion as before. Very often old weak boilers fail by tearing the corroded plate outside the riveted joint. Stay-rods and bolts suffer much more from corrosion than plates. Thus a rod one inch in diameter has an area of 0.7854 of a square inch. After corrosion to the extent of 1/16 of an inch has taken place the diameter is 7/8 of an inch and the area is 0.6013, which is 0.6013 -*- 0-7854 = 0.766 ot the original area. Compare this with the plate which retains 7/8 or 0.875 of its thickness after the same amount of corrosion. Of course a smaller stay will suffer more, and a larger one less, in proportion. After the sizes of the parts of a boiler are decided upon it is well to make calculation to see that a factor of safety of four will remain after a reasonable amount of corrosion. Or, as in the case of stay-rods, the size may be calculated with a proper factor, and then the diameter may be increased to> allow for corrosion. BOILER DESIGN 33 1 Thickness of Shell. — The final decision of the proper thickness of the shell for the boiler under consideration can- not be made until the efficiency of the joint is known; but the efficiency of any of the complex joints now in vogue can be found only when the thickness of the plate is known. It is therefore convenient to assume a factor of safety of about six and make a preliminary calculation. Thus for the boiler in hand we will get for the thickness (page 183) ,_ I50X30 =049 55,000 h- 6 of an inch. A similar calculation with a factor of five gives t — = 0.q.L 55,000^ 5 of an inch. The shell will be either 7/16 or 1/2 an inch thick. Seven sixteenths will give an apparent factor of safety of 55 ,000 X 7/16 = - 150 X 30 D ' 35 ' After the allowance for the efficiency of the joint has been made this factor will be found to be about 4f . Longitudinal Joint. — The shell-plate is made as thin as possible because it will be exposed to the fire. Consequently the efficiency of the longitudinal riveted joint must be high if the real factor of safety is to be satisfactory. The strength of triple-riveted joints like that shown on page 201 ranges from 85 to 90 per cent. The joint with two cover-plates shown by Fig. 139, will be chosen. Following the method given on page 201, it appears that this joint may fail in one of five ways, for which the resistances are as follows: A. Tearing at outer row of rivets : Resistance = (P •— d)tft. 332 S TEA M-B OILERS. B. Shearing four rivets in double shear and one in single shear : and 2 Resistance = f s . 4 C. Tearing at the middle row of rivets and shearing one rivet: nd* Resistance = (P — 2d)tf t -\ /,. Fig. 139. D. Crushing four rivets and shearing one : Resistance = ^dtf c ~\ / s . E. Crushing five rivets: Resistance = ^dtf c + dt c f c . The diameter of rivet will be found by equating the resistances A and C. .-.(/>- d)tf t = {P- 2d)tf t + ^f s . # 4tf t 4X tVX 55>QQQ „ AQ . . a = — 7- = ■ = 0.05. n fs 7195 1° 00 BOILER DESIGN. 333 The rivet which was used was 13/16 of an inch when driven. There are several methods in which we may find the way in which the joint will fail, and then find therefrom the effi- ciency. One is that shown on page 202 by assuming a pitch and calculating the resistance of the joint to failure in each of the five several ways. Another method is to equate the five several resistances two and two and calculate the pitch ; the least pitch thus found must not be exceeded. Thus Equating B and C, 4 4 ¥ f. X3.H.6x(^ 45,000 \x 7 55,000 ■ 16 4x f 6 Equating A and B, {P -d)tf t =^/,. 9^ 9 X 3.1416 _i6/ 4SJ000. 13 „ v 7 55,000^16 4X T6 Equating A and D, 4 3 3 V S TEA M-B OILERS. ft - 4* ft 3-1416 X PI)' ~ 4X i6 X 55,ooo + 7 X 55.ooo+i6 7A ' 4X T6 Equating A and E, (P-d)tf t = 4dtf c + dt c f e . Jt l It a vi? v 95,OQO a- I3/l6 X 3/8 w 95>QQO | 13 _ 76 i6 X 55,000"^ 7/16 x 55,000+16 /# ' Here t t9 the thickness of the cover-plate, is taken to be 3/8 of an inch. The greatest allowable pitch at the outer row of rivets is evidently 7.4 inches. Instead of going to the labor of solving all four of the above equations, we may find by some other method how the joint is likely to fail, and make up an equation involving those resistances only. Thus a rivet in the outer row may fail by shearing or by crushing at the cover-plate, which is here made thinner than the shell-plate. Equating the re- sistances of the two methods, we have —■/. = *#.. 4 or for a cover-plate 3/8 of an inch thick ^ = 4XJ X 95,000 =iQi 7t 45,000 A rivet 1.01 inch in diameter will consequently be just as BOILER DESIGN. 335 likely to fail by crushing as by shearing. But the resistance to shearing increases as the square of the diameter, while the resistance to crushing increases as the diameter. It is there- fore evident that a rivet larger than i.oi of an inch will fail by crushing, while a smaller rivet will fail by shearing. A similar calculation at the inner row, when the rivet bears against a cover-plate both inside and outside, and will consequently crush against the shell-plate, gives Js = taf e ; 4 , 2 XtV 95,000 , d = l¥ x — = o.6. it 45,000 Here a rivet larger than 0.6 will crush, and one smaller will shear. It is now evident that a 13/16 rivet will shear at the outer row and will crush at the inner row. That is, for this joint the failure will occur by the method D, but not by the methods B or E. Then equating the resistances A and D, and solving for P, we get for the pitch at the outer row 7.4 inches as before. The corresponding pitch at the calking edge of the outer cover-plate is 3.7 inches; we will choose for that pitch 3f inches, making the pitch at the outer row 71 inches. The efficiency of the joint is p _ J 7I J.3 100 = 100 X — ^ = 88.8 per cent. P 7\ In the preceding article the apparent factor of safety based on the whole strength of the shell-plate is 5.35. Al- lowing for the efficiency of the longitudinal joint, the real factor of safety when the boiler is new is 0.888 X 5-35 =475. 336 S TEA M-B OILERS. With this style of joint the shell-plate is protected from corrosion by the inner cover-plate, and the joint will lose little if any efficiency from corrosion. If it be assumed that the plate loses 1/16 of an inch by corrosion during the life of the boiler, then the strength of the plate will be one seventh less after corrosion, and the corresponding factor of safety will be 5.35 X f = 4.6, which may be considered to be sufficient. Ring-seam — The stress on a transverse section of a homogeneous hollow cylinder from internal fluid pressure is one half the stress on a longitudinal section. It will in gen- eral be found that a single- or a double-riveted ring-seam is sufficient for any cylindrical boiler-shell. Marine boilers commonly have double-riveted ring-seams; externally-fired horizontal boilers seldom have the shell more than half an inch thick, and for that thickness, or less, single-riveted ring- seams are used. It is found in practice that ring-seams of horizontal ex- ternally-fired boilers may have a pitch of about 2 T 3 ^ inches for all thicknesses of plate from 1/4 to 1/2 of an inch. The diameters of rivets for such seams may be made about the size given in the following table : Thickness of plate \ T \ f T \ \ Diameter of rivet | \\ f \ \ The ring-seam in question has a circumference of about 3.1416 X 60 = 188.2 inches, which will allow us to use 84 rivets with a pitch of about 2.24 inches. This joint will fail by shearing the rivets. The efficiency of the joint is consequently the ratio of the resistance of a single rivet to shearing, to the resistance of BOILER DESIGN. 33/ a strip of plate as wide as the pitch. Consequently the efficiency is 4 f * _ i X 3-i4i6 X ( -j-iJV X 45, 000 ///« 2.24 X -iV X 55.000 ■433. which is more than half of the efficiency of the longitudinal seam, and will consequently be sufficient. Lap. — The lap, or distance from the centre of the rivet to the edge of the plate, is usually taken as 1.5 times the diam- eter of the rivet used, which makes the distance of the edge of the hole from the edge of the plate equal to the diamccer of the rivet. For the single-riveted ring-seam this makes th^ lap equal to !-5 X If = 1.22. It is customary to calculate the width of lap required on the assumption that the metal between the rivet and the edge of the plate may be treated as a beam of uniform depth, fixed at the ends and loaded uniformly by the force which would be required to shear or crush the rivet, taking, of course, the larger. The width of the' beam is the thickness of the plate, the depth is the distance from the edge of the hole to the edge of the plate, and the length is the diameter of the rivet. Rivets in single-riveted seams fail by shearing. The loau is consequently the shearing resistance The maximum bending moment for a beam of uniform section fixed at the ends and uniformly loaded is equal to the load multiplied by one eighth of the span. The moment of resistance is equal to fl, 33^ STEAM-BOILERS. in which /is the cross-breaking strength (about 55,000), /is the moment of inertia of the section, and y is the distance of the most strained fibre from the neutral axis. Here we have T th 3 Ji I — — , y= -, 12 2 representing the distance from the edge of the hole to the edge of the plate by h. Equating the bending moment to the moment of re- sistance, 4 6 nd % f s 67 x 7 3 x 3-Hi6 x 13 3 w 45*ooo ^ _ X — 0^77 r 7 £% 5 5,000 7/ 16 X -VX16 16 for the case in hand. The lap is consequently 0.77+- x i| = 1.18 2 16 inches for the ring-seam, which is somewhat less than that by the arbitrary rule that it should be once and a half the diam- eter. A similar calculation for the cover-plates with the same diameter of rivet, but with a plate 3/8 of an inch thick, gives for the lap 1.24 or 1^ of an inch, while the arbitrary rule gives 1.03 of an inch. It is probable that the lap may be consider- ably smaller than is given by the calculation by the beam theory, but for lack of direct experimental knowledge on this question it is not wise to make the lap much less than the calculation gives; we will consequently use 1^ of an inch for the lap of the cover-plates. BOILER DESIGN 339 The rivets of the inner rows pass through both cover-plates and are in double shear, and consequently fail by crushing as is shown on page 335. The load to be used tor calculating the lap is therefore the resistance to crushing in front of the rivet ; that is, we here have for the load tdf c . The equation of bending moment and moment of resistance gives 1 t/f ldXtdfc = f - k = = \l 71 x 13 X 13 X 6000 -^ — 4.55 inches. 150 X 16 X 16 This gives for the limiting distance of rivets 4.55 inches. Of course a less distance may be used if convenient. In some cases the pitch of the rivets may be controlled by the system of staying. For example, the rods used with crowfeet are seldom more than \\ of an inch in diameter, because larger rods may bring too large a local stress where they are riveted to the cylindrical shell. Rods one inch or an inch and an eighth are frequently used. A double crow- BOILER DESIGN, 345 foot has four rivets, each of which will carry one fourth of the load on the stay-rod. A stay-rod i \ of an inch in diameter, and limited to a stress of 7500 pounds, may carry a pull in the direction of its length of 7500 X (l}) a = 11,720 pounds. If the rod makes an angle of 20 with the shell-plate, the pull which it will exert perpendicular to the head will be 1 1,720 cos 20 = 1 1,720 X 0.93969 = 1 1,013 pounds, so that each rivet will carry about 2750 pounds. If each rivet supports a square having the side a 2 exposed to the pressure of steam at 150 pounds, then 11,013 = 150 X a,\ or V iqo 8 inches. 5< Laying out Stays. — Having selected the form of staying to be used, the plan must be laid out on the drawing-board, giving proper attention to practical considerations, such as the way in which the stays are to be inserted, and taking care that accessibility is not too much interfered with. Fig. 140 repeats the upper part of the head of the boiler shown by Plate I, with certain additional dotted lines, which will be referred to in the explanation of calculations. The area to be stayed is considered to be limited by the upper row of tubes, and by a dotted line drawn ij of an inch from the inside of the shelh This line is drawn at the right only; it is very nearly the place where the rounded corner of the flange joins the flat surface of the head. The distance of the lowest row of rivets from the top row of tubes, and of the outer row of rivets from the dotted line, may be as great as their maxi mum distance from each other. Rivets should not be placed nearer than 3 inches from the tubes, lest the expansion of the 34-6 s TEA M-B OILERS. tubes should start leaks. Rivets may be placed near the dotted line, if that is convenient. For example, the outer- most row of rivets in crowfoot staying (Fig. 45, page 152) may be at a distance a 9 from the dotted line; for i|- inch stay- rods or - = 0.933. BOILER DESIGN. 35 I Now / and y depend on the form and size of the section of the beam, and, conversely, the size and form of beam required may be determined from them. But as the upper channel-bar is exposed to a greater bending moment and con- sequently must have a larger section than is required for the lower bar, we will defer the discussion of this matter, because it is convenient to make the bars of the same size. Upper Stay-rods. — The flat surface of the boiler-head above the lower channel-bar is supported by the upper channel-bar aided by the angle-iron which is firmly riveted to it, and which will be assumed to act with and form a part of the channel-bar. Following our general convention that the pressure on a portion of the head between two lines of support is divided equally between them, we will assume that the load on the upper channel-bar is due to the steam-pressure on an area bounded at the bottom by a line half-way between the upper and lower channel-bars, and at the top by an arc 3^ inches inside the boiler-shell. On Fig. 140 half of this area is rep- resented by jkl; the arcj/c being about half-way between the root of the flange, shown by the outer dotted boundary line, and the adjacent rivets. In place of the area jkl we will take the rectangular area Imno, bounded at the end by a line at the middle of the end of the channel-bar, and at the top by a line mn so chosen as to make the rectangular area larger than the area it replaces. The width of this area, ////, is g{ inches, so that the load per inch of length is 9£ X 150 = I337-5 pounds. The upper channel-bar may be assimilated to a continuous girder with three unequal spans; the middle span between the stay-rods is 15^ inches, and the end spans between the stay-rods and the roots of the flange of the head are each 11J inches. This makes the end spans nearly 3/4 of the middle span. Now, a continuous girder uniformly loaded 3 5 2 S TEA M-B OILERS. with w pounds per inch of length, which has a middle span / inches long, and two end spans f / inches long, will have for the end-supporting forces ff}«//, and for the middle support- ing forces f|-Jtt>/. The end supporting forces are provided by the shell, which is abundantly able to carry them. The stay-rods, which furnish the middle-supporting forces, must each carry Ui X J 5i X I387-5 = 21,083 pounds. Assuming a working-stress of 9000 pounds per square inch for the stay, the area of the section for a stay is 21,083 -T- 9000 = 2.34 square inches. The corresponding diameter is not quite li-|- of an inch. As rods of this size are not regularly carried in stock, we will take the next larger regular size, namely, if of an inch. This is the size mentioned in connection with the discussion of the lower stay-rods. Upper Channel-bar. — The calculation of the stress in the upper channel-bar will be made by an extension of the same approximate method used with the lower channel-bar. Since the middle span is wider than the end spans, it will be suffi- cient to make a calculation for it only. The calculation is made as for a simple beam supported at the ends, the points of support being at one fourth of the diameter of the thick washer from the middle stay-rod, that is, at the distance of 1 1 of an inch from the stay-rod. The distance between the upper stay-rods is 15^ inches, so that the span of the beam is 1 Si — 2 X if = I2f- inches. The beam is assumed to be loaded with concentrated loads applied at the rivets C, D, E, F> G, and H (Fig. 140); the load on the rivet / is assumed to be carried by the stay-rod directly, and is not included in this calculation. The pair of BOILER DESIGN. 353 rivets D and B, and the several rivets C, G, and If, are assumed to carry the load due to the pressure on the areas marked off by the dotted lines on Fig. 140, each line being drawn half-way between adjacent supporting points, except that the arc at the top is drawn 3] inches from the shell, as already said. The calculation of the loads on these rivets, of the supporting forces, and of the bending moments is simple and direct, but is tedious when stated in detail. We will therefore be contented to say that the bending moment at the middle of the beam is 37,390. Taking, as with the lower channel-bar, a working-stress of 16,000 pounds, we have 16,000/ / 37>39° = -— - — 1 or -= 2 - I 7- y y The makers of steel beams, channel-bars, and angle-irons publish handbooks which give the sizes and properties of the standard forms, including the moment of inertia / and the ratio — , which is called the moment of resistance. From such y a handbook it appears that the moment of resistance of the channel-bar 6" X 2 J" X \" is 1.08, and that the moment of resistance of the 3!" X 3" angle-iron is 1.55; the sum 2.63 is larger than the required moment of resistance given above. These forms are consequently used as shown on Plate I. Brackets. — The boiler shown on Plate I is supported on four cast-iron brackets, each of which is 10 inches wide in the direction of the length of the boiler, and 15J inches long measured circumferentially. Each bracket is riveted to the shell by nine rivets 15/16 of an inch in diameter. Boilers over 16 feet long commonly have six brackets. The brackets are made wide and long in order that the local strains due to carrying the weight of the boiler may not be excessive. The rivets are larger than are used about the boiler, as the pitch is not restricted as in a calked seam. 354 STEAM-BOILERS The brackets are set above the middle line of the boiler so that the flanges may be protected by brickwork. In the case in hand they are 3f inches above the middle; as much as 4J inches is commonly used. The brackets are arranged so that the weight of the boiler and accessories is equally divided among them, and so that there is as little bending-moment as possible on the shell of the boiler. When four brackets are used they may be some- what less than, a fourth of the length of a tube, from the tube- plat^ The load on the brackets may be estimated by calculating the weight of the boiler when entirely full of water, and add- ing the weight of all parts that are supported by the boiler, such as pipes, valves, and brickwork or covering, that may rest on the boiler. One fourth of this load is assigned to each bracket. This load on a bracket should be uniformly dis- tributed over the bearing-surface of the flange, which is com- monly 8 or 9 inches wide. But to guard against the effect of unequal bearing, it is well to assume the bracket to bear near the outer edge — say two inches from the edge. Such an assumption will bring the bearing-force on a bracket on Plate I, 10 inches from the shell. This bearing-force tends to rotate the bracket about its upper edge, and this tendency is resisted by the rivets under the flange, which must be large enough to resist the resulting pull on them. The other rivets are added to give sufficient resistance to shearing all the rivets. There are seldom less than nine rivets in a bracket, all as large as those below the flange, even though fewer would suffice. The bracket is usually made of cast iron, and the dimensions are commonly controlled as much by the condi- tions required for a sound casting as by calculations for strength. The strength may be calculated, treating it as a cantilever, allowing for the web connecting the flange to the body of the casting. Specifications and Contract. — The engineer intrusted BOILER DESIGN. 355 with the design of a boiler prepares a set of working drawings and a set of specifications which give all necessary instructions concerning the material to be used and the methods of con- struction to be followed. The drawings and specifications form a part of the contract with the boiler-maker. Boiler-makers commonly design standard forms of boilers, and in answer to inquiry will furnish a statement or set of specifications for a desired boiler in form of a letter, which letter forms the contract for the boiler. On the next page is given the contract and specifications for the boiler shown on Plate I. 356 STEAM-BOILER. IRON WORKS CO. Boston, Mass. , Feb. /, 1897. Gentlemen : Your letter of received. We will build One (/) Horizontal Tubular Boiler. One Boiler, viz., Sixty (bo) inches diameter by seventeen 2/12 (/7t 2 z) feet long. Containing 84 Tubes 3 inches diameter, by sixteen (ib) feet long. Shell of Boiler of O. H. Fire-box Steel, 7/16" thick, not less than 55,000 nor over 60,000 lbs. Tensile Strength. Not less than 56% reduction 0/ area, and 25% elongation in 8" . Heads of Boiler of O.H. Flange Steel q/ib" thick. Longitudinal Seams Butt Jointed, with double covering-plates, Triple Riveted. Rivet-holes drilled in place, i.e., Rivet-holes punched 1/4" small, courses rolled up, covering-plates bolted on courses. Heads in courses with all holes together perfectly /air. Then rivet-holes drilled to full size. Longitudinal braces without welds, with upset screw ends. Two (2) or three (3) Lugs on each side, and to be provided with wall-plates and expan- sion-rolls. Manhole (internal frame) on top. This frame a steel casting. Two (2) 5" Nozzles on top, A Hand-hole in each head, Fusible Safety Plug in back head. Bottom at back end reinforced and tapped for 2" blowout Internal Feed Pipe placed in Boiler Co.'s style, With Boiler, Castings for setting, viz.; C. I., Overhung Front, Mouth-pieces, Division Plates, Grate Bars, shaking pattern bo" X bo" . Grate Bearers, Ash-pit Door for the brickwork, Back Return Arched T Bars, the Anchor Bolts for Front. One (1) set of six (6) Buckstaves and Tie Rods with the boiler. With the Boiler One (\) 4" Pop Safety Valve, (3)3/4" Gauge Cocks, One (1) 6" Steam Gauge, One (1)3/4" Water Gauge and One (1) Combination Column Boiler tested 225 lbs. per square inch. Inspected and Insured in the sum of $400.00 for one year, by Steam Boiler Inspection & Insurance Co The Boiler Castings and Fixtures as herein specified by name, delivered F. O. B. cars, or at vessel's wharf, or on sidewalk of building, Boston, Mass., for the sum of six hundred and seventy $70.00) dollars net. Very respectfully yours, IRON WORKS CO. P. S. — Specimens will be furnished, one lengthwise and one crossivise, from each plate. To be at least 18" long and planed on edge 1" or i\" wide. These specimens shall show n& blowhole defects and shall bend double cold, at a red heat, and at a flanging heat. APPENDIX. 358 APPENDIX. LOGARITHMS. Nat. Nos. Proportional Parts. 1 2 3 4 5 6 7 8 9 1 7404 12 3 4 5 6 7 8 9 55 7412 74i9 7427 7435 7443 745i 7459 7466 7474 122 3 4 5 5 6 7 56 74S2 7490 7497 7505 7513 75207528 7536 7543 755i 1 2 2 3 4 5 5 6 7 57 7559 7566 7574 7582 7589 759717604 7612 7619 7627 1 2 2 3 4 5 5 6 7 58 7634 7642 7649 7657 7664 7672 7679 7686 7694 7701 112 3 4 4 5 6 7 59 7709 7716 7723 773i 7738 7745 7752 7760 7767 7774 112 3 4 4 5 6 7 60 7782 7789 7796 7803 7810 7818 7825 7832 7839 7S46 1 1 2 3 4 4 5 6 6 61 7853 7860 7868 7875 7882 7889 7896 7903 7910 79 J 7 112 3 4 4 5 6 6 62 79 2 4 793i 7938 7945 7952 7959 7966 7973 7980 7987 112 3 3 4 5 6 6 63 7993 8000 8007 8014 302I 802818035 S041 8048 S055 112 3 3 4 5 5 6 64 S062 S129 3o6g 8136 8075 8082 8089 8096 8102 8109 S176 8116 S122 112 3 3 4 5 5 6 65 8142 8149 8156 8162 8169 8182 S189 1 1 2 3 3 4 5 5 6 66 819=; 8202 8209 S215 8222 8228 8235 8241 8248 8254 112 3 3 4 5 5 6 67 S261 8267 8274 8280 8287 8293 8299 8306 8312 8319 112 3 3 4 5 5 6 68 S325 833i 8338 8344 8351 8357 8363 8370 8376 8382 112 3 3 4 4 5 6 69 3388 8395 8401 8407 S4I4 8420 S426 8432 8439 8445 112 2 3 4 4 5 6 70 8451 8457 8463 8470 8476 8482 848S 8494 8500 S506 112 2 3 4 4 5 6 71 3513 S519 8525 8531 8537 8543 8549 8555 8561 8567 112 2 3 4 4 5 5 72 8573 8579 8585 S591 8597 8603 S609 861s 8621 S627 I I 2 2 3 4 4 5 5 73 8633 S639 8645 S651 8657 S663 8669 8675 S681 S686 112 2 3 4 4 5 5 74 S692 875i S698 S756 8704 8762 S710 876S S7I6 8722 8727 S733 8739 S745 112 2 3 4 4 5 5 75 8774 8779 8785 8791 8797 8802 I I 2 2 3 3 4 5 5 76 8S08 3814 8820 882=; 8831 8837 8S42 8848 SS54 8859 I I 2 2 3 3 4 5 5 77 8865 88 7 1 SS76 S882 88S7 889318899 8904 S910 8915 112 2 3 3 4 4 5 78 8921 3927 8932 893S 8943 89498954 S960 8965 8971 112 2 3 3 4 4 5 79 S976 8982 8987 8993 8998 9004 9009 9015 9020 9025 I I 2 2 3 3 4 4 5 80 9031 9036 9042 9047 9053 9058^063 9069 9074 9079 112 2 3 3 4 4 5 81 90S5 9090 9096 9101 9106 91129117 9122 9128 9133 I I 2 2 3 3 4 4 5 82 9 T 3S 9M3 9149 9154 9159 91659170 9175 9180 9186 112 2 3 3 4 4 5 83 9191 9196 9201 9206 9212 9217:9222 9227 9232 9238 1 1 2 2 3 3 4 4 5 84 9243 9294 9248 9299 9253 9304 925S 9309 9263 9315 92699274 9320 9325 9279 9284 9289 1 1 2 2 3 3 4 4 5 85 9330 9335 9340 1 1 2 ! 2 3 3 4 4 5 86 9345 935o 9355 936o 9365 93709375 93809385 9390 1 1 2 2 3 3 4 4 5 87 9395 9400 94059410 9415 9420.9425 9430 9435 9440 Oil 223 3 4 4 88 9445 945o 9455 946o 9465 9469 ! 9474 94799484 9489 Oil 223 3 4 4 89 9494 9499 9504 9509 9513 95189523 9528 9533 953S Oil 223 3 4 4 90 9542 9547 9552 9557 9562 95669571 9576 958i 9586 1 1 223 3 4 4 91 959° 9595 9600 9605 9609 9614 9619 9624 9628 9633 Oil 2 2 3 3 4 4 92 963S 9643 9647 9652 9657 9661 9666 9671 9675 968c 1 1 2 2 3 3 4 4 93 96S5 9689 96949699 9703 97oS 97i3 9717 97229727 1 1 2 2 3J 2 94 9731 9736 9741 9745 9750 9754 9759 9763 9768 9773 1 1 2 2 3I 3 4 4 1 95 9777 9782 97869791 9795 9800 9805 9809 9ST4 9S18 Oil 223 3 4 4 96 9823 9S27 9832 9836 9S41 9845 9850 985498599863 Oil 223 3 4 4 97 9868 9872 9877 9881 9886 9890 9894 9899 9903 990S 1 1 223 3 4 4 98 9912 9917 9921(9926 9930 9934 9939 9943 9948,9952 Oil 223 3 4 4 99 9956 9961 996519969 9974 9978 9983 9987I9991I9996 Oil 223 3 3 4 APPENDIX. LOGARITHMS. 359 Nat. Nos. Proportional Parts. 1 2 3 4 5 6 7 8 9 oooo 0043 00S6 0128 0170 0212 1 2 3 4 5 6 7 8 9 10 0253 0294 -)334 0374 4 S 12 17 2 i 2 = 29 33 37 11 0414 0453 04920531 0569 0607 0645 0682 0719 0755 4 8 1 1 '5 19 23 26 30 34 12 0792 0S2S 0864 0899 0934 0969 1004 103S 1072 1 106 3 7 10 M '7 21 24 28 31 13 1 1 3Q 1173 1 2( )l - 1239 1271 1303 1335 [367 [399 [430 3 6 10 [3 16 ig 23 26 29 14 1461 1492 1523 1553 1584 1614 1644 [673 1703 1732 3 6 9 12 15 [8 21 24 27 15 1761 1790 I8l8 1S47 1875 1903 193' 1959 19S7 2014 3 6 8 11 14 17 20 22 25 16 2041 206S 2095 2122 214s 2175 2201 2227 2253 2279 3 5 8 1 1 13 16 18 21 24 17 2304 2330 2355 2 3 SO 2405 24302455 24S0 2504 2520 2 5 7 10 12 *5 17 20 22 18 2553 2577 260] 2625 264S 2672 2695 271S 2742 2765 2 5 7 9 12 14 16 19 21 19 27^- 2S10 28332856 2S7S 2900 2923 2945 2967 '29S9 2 4 7 9 II 13 [6 18 20 20 3010 J032 3054 3075 3096 31183*39 3160 3181 3201 2 4 6 8 I I 13 '5 17 19 21 3222 V-M3 3263 3284 3304 3324 3345 33 f >5 33S5 3404 2 4 6 8 K) 12 '4 16 18 22 3424 3444 3464 34S3 3502 3522 354i 356<>35793598 2 4 6 8 K) 12 14 15 17 23 3617 3636 3655 3674 3692 3711 3729 3747 3766 37S4 2 4 6 7 9 1 1 '3 15 i7 24 3S02 3820 3338,3856 3S74 3S92 3909 3927 3945 3962 2 4 5 7 9 1 1 12 14 i() 25 3979 3997 4OI4 403I 4048 4065 40S2 4099 41164133 2 3 c 7 9 10 12 14 15 26 4150 4166 418342OO 4216 4232 4249 4265 4281 4298 2 3 5 7 8 10 11 13 15 27 4314 4330 4346 4362 437S 4393 4409 442544404456 2 3 5 6 S 9 1 1 13 14 28 4472 44S7 45024518 4533 454S 4564 4579 4594 4609 2 3 5 6 8 9 1 1 12 14 29 4624 477i 4639 47S6 4654 4660 4683 469S 4843 4713 4728 4742 4757 1 3 4 6 7 9 10 12 13 30 4S0O 4S14 4829 4S57 4S71 4886 4900 1 3 4 6 7 9 10 11 13 31 49M 492S 4942 405 ^ 4969 49834997 501T 5024 503S r 3 4 6 7 10 11 12 32 5051 5065 5079 5092 5105 5 IK, 5132 5M5 5159 5172 1 3 4 5 7 8 9 n 12 33 5185 519S 5211 5224 5237 5250 5263 5276 5289 5302 1 3 4 5 6 8 9 10 12 34 5315 532S 5340 5353 5366 53785391 5403 54i6 542S 1 3 4 5 6 8 9 10 11 35 544i 5453 5465^478 5490 5 502 5 514 5527 5539 555^ r 2 4 5 6 7 ( i 10 11 36 5563 5575 5587 5599 5611 5623 5635 5647 565SJ5670 r 2 4 5 6 7 8 10 11 37 5682 5694 5 70- 5717 5729 5740 5752 5763 5775 5 7S<» 1 2 3 5 6 7 8 9 10 38 579 s 5S09 5S21 5S32 5S43 ; 5S?= 5866 5877 5888:5899 1 2 3 5 6 7 8 9 10 39 591 1 5922 5933 5944 5955 5966 5977 5988 5999 6010 1 2 3 4 5 7 S 9 10 40 6021 6031 6042 6053 6064 6075 6085 6096 6107 6117 1 2 3 4 5 6 T 9 10 41 612S 613S 6149 6160 6170 6180 6191 6201 6212 6222 1 2 3 4 5 6 7 8 9 42 6232 6243 6253 6263 6274 62S4 6294 6304 6314 6325 1 2 4 5 6 7 8 9 43 6335 6345 63556365 6375 63856395 6405 64i5'6425 1 2 3 4 5 6 7 8 9 44 6435 6444 64546464 6474 64S46493 6503 65136522 1 2 3 4 5 6 7 8 9 45 6532 6542 6551 6561 6571 6580 6590 6599 6609 661 S 1 2 3 4 5 (» 7 8 9 46 6628 6637 66466656 6665 6675 6684 6693 6702 6712 1 2 3 4 5 6 7 7 8 47 6721 6730 67396749 6758 6767 0776 6785 6794 6S03 1 2 3 4 5 5 6 7 S 48 6S12 6S21 6S30 6839 6848 6857 6866 6S7^ 68S46S93 1 2 3 4 4 5 6 7 8 49 6902 691 1 6920 692S 6937 69466955 6964 6972 6981 1 2 3 4 4 ? 6 7 7 S 50 6990 699S 7007 7016 7024 7033 7042 7050 70597067 r 2 3 3 4 5 7 8 51 7076 7084 7093 7101 7110 7118 7126 7135 7M3 7152 1 2 3 3 4 5 6 7 8 52 7160 716S 7177 7185 7193 7202 7210 7218 7226 7235 1 2 2 3 4 ^ 6 7 7 53 7243 7251 7259 7267 727- 72S4 7292 7300 7303 7316 1 2 2 3 4 5 6 6 7 54 7324 7332' 7340 734S 7356!7364 7372 738o 738SI7396 1 2 2 3 4 5 6 6 7 300 APPENDIX. Explanation of the Table for Finding the Area of Segment of a Circle. — The areas given in the table are for a circle I inch in diameter. The diameter is divided into iooo parts, and the area for segments of different heights can be taken directly from the table, since the ratio of the height of the segment to the diameter of the circle is the same as the height of the segment. For a circle whose diameter is other than unity. Given the diameter of the circle and the height of segment. Re- quired area of segment. Divide height of segment by diameter ; find area given in the table opposite this ratio ; multiply this area by the square of the diameter and the result is the re- quired area. Example. — Dia. of circle = 6o n ', height of segment = 18". 1 8 -f- 60 = . 30 ; area in table opposite .30 is .19817. .19817 X 60 X 60 = area of segment = 713.4 sq. in. Given the diameter of the circle and the area of a segment, to find the height. Divide the area of the segment by the square of the diam- eter. Find in the table the area nearest to this, multiply the ratio corresponding to this by the diameter of the circle, and the result is the required height of the segment. Example. — Area of segment = 713.4 sq. in. Diameter of circle = 60" '. Required the height of the segment. 71 3.4 — = .19817. Ratio opposite this is .300. .300 X 60" = 18", the required height. Example — Area of segment == 640 sq. in. Diameter of circle = 50". 640 = .2560; nearest ratio, .362. 50 X 50 .362 X 50 = 18.10", the required height. APPENDIX. 361 TABLE FOR FINDING AREAS OF SEGMENTS OF A CIRCLE. <-> V *i v «-i 4) j ~ O V «o«J •G «->T. c ■C-M — c J3«7, c 1 JZ - — c £*"- bo^ t u bc w « ^ £ a bo u to bo y u "^ u 01 '5 Crj <♦< bo . a bo 4) '3 cf. a bo u C/5 "rr bo . a be u CO KB* a bo c 2 w .i v a 2^1 a! m doQ SoQ iameter Effective Diameter in Inches. Circumfer- ence in Inches. Area in Sq. Inches. of Rod One Foot Long. of Upset Si rew End. of Screw it Root ol Thread. rhreads Area of per Inch Sa-pvyEnd Nu.nl. 1 Inches. Inches. Per Cent. 1/16 .1963 .0031 .OIO i/S .3927 .0123 .041 3/16 •5S9O .0276 .092 1/4 .7854 .0491 .164 5/16 .9817 .0767 .256 3/8 I.1781 . 1 104 .368 7/16 1-3744 .1503 • 501 1/2 1.5708 .1963 .654 3 .620 IO 6f 9/16 I. 7671 .24S5 .828 t .620 IO 2J 5/8 1.9635 . 3068 I.023 7 .731 9 11/16 2.1598 .3712 1.237 I •837 8 4b 3/4 2.3562 .4418 1.473 .837 8 25 13/16 2.5525 .5185 I.72S T 1 A 8 .940 7 34 7/8 2.7489 .6013 2.004 T i I .065 7 48 15/16 2.9452 • 6903 2.301 T 1 I .065 7 29 1 3.I416 .7S54 2.6l8 If I . 160 6 35 1/16 3-3379 .8866 2.955 If I. 160 6 19 1/8 3-5343 .9940 3-313 jl I.2S4 6 30 3/16 3.7306 1. 1075 3.692 J i I.284 6 17 i/4 3.9270 1.2272 4.O9I if I.389 5i 23 5/i6 4.1233 i.353o 4.510 It I.49O 5 29 3/8 4.3197 1.4849 4-950 T 3 I.49O 5 18 7/16 4. 5160 1.6230 5-4IO T 7 1 S I .615 5 26 1/2 4.7124 1.7671 5.890 2 I .712 4i 30 5/8 5-I05I 2.0739 6.913 a* 1.837 4i 28 3/4 5.4978 2.4053 8.018 2i 1 . 962 4i 26 7/8 5.8905 2.7612 9.204 2f 2.0S7 4^ 24 2 6.2832 3.1416 I0.47 2i 2.175 4 18 1/8 6.6759 3.5466 11.82 2f 2.3OO 4 17 28 1/4 7.0686 3.9761 13-25 2.550 4 3/8 7.46T3 4.4301 14.77 3 2.629 3i 23 1/2 7-S540 4.9087 16.36 3* 2-754 3i 21 5/8 8.2467 5.4II9 18.04 3i 2.879 3h 20 3/4 8.6394 5-939° 19.80 3f 3.OO4 3h 19 26 7/8 9.0321 6.4918 21.64 3l 3-225 3i 3 9.4248 7.0686 23.56 31 3.317 3 22 3^4 APPENDIX. LAP-WELDED BOILER-TUBES. u V c/i U w^ e 21 Kg £& +1 u tt, 5J j V a 3 a a 5 > — en rt Si c.n C «T 4; c !5 U _C3 O 4) ,1? OJ3 n a; — Is U T £ go* J=CO rt S) £ go* X u - & f ^_ rt £ c C 8.3 si . C8 — £ c £ u a J3 '53 c/5 w 1-1 H u U H H J hJ co CO I i .86 .072 3 J 4 2.69 .78 ■57 382 4.46 .26 .22 • 71 *H M i. ii .072 3-93 3-47 1.23 .96 3 06 3-45 33 .29 .89 % \Yi 1-33 .083 4.71 4.19 1.77 1.40 2 55 2.86 39 •35 1.24 1% i% 1.56 •095 5-5o 4.90 2.40 1. 91 2 18 2-45 46 .41 1.66 2 2 1. Si • 095 6.28 5-6q 3-14 2-57 I 91 2. 11 52 •47 1. 91 2^ 2 ^ 2*4 2.06 .095 7.07 6.47 3-Q8 3-33 I 70 1.85 59 •54 2.16 2*£ 2.S-8 .109 7-8 5 7-17 4.91 4.09 I 53 1.67 65 .60 2-75 2^ 2% 2-53 .109 8.64 7-95 5-94 5-°3 I 39 1. 51 72 .66 3-04 3,, ^ 2. 7 8 .109 9.42 8.74 7.07 6.08 I 27 *-37 7 ! ^ •73 3-33 3H :^ 301 . 120 10.21 9.46 8.30 7.12 I 17 1.26 S5 •79 3-96 3H 3.26 . 120 11.00 10 24 9.62 8-35 I 09 1. 17 92 .85 4.28 3% 3% 3-5i . 120 11.78 11.03 11.04 9.68 I 02 1.09 oS .C2 4.60 4 f/ 4 3-73 .134 12-57 11.72 12.57 10.94 95 1 .02 1 05 .98 5-47 4*6 4*£ 423 •i34 14.14 13.29 15.90 14.07 85 .90 1 18 I. II 6.17 5 5 4.70 .148 I5-7I 14.78 19.63 17.38 76 .81 1 3i 1-23 7-58 6 6 567 .165 18.85 17.81 28.27 25-25 64 .67 1 57 1.48 10. 16 7 7 6.67 .165 21.99 20.95 38.48 34-94 55 •57 1 83 i-75 11.90 8 8 7.67 .165 25-13 24.10 50.27 46.20 48 •50 2 09 2.01 i3-o5 9 9 8.64 .180 28.27 27.14 63.62 58.63 42 •44 2 35 2 .26 16.76 IO IO 9-59 .203 31-42 30.14 78.54 72. 2Q 38 .40 2 62 251 20.99 ii ii 10.56 .220 34-56 33-17 95-03 87.58 35 •36 2 SS 2.76 25-03 12 12 "•54 .229 37-7o 36.26 113.10 IO4.63 32 •33 3 J 4 3.02 28.46 SCREW-THREADS. Angle of thread 6o°. Flat at top and bottom = § of pitch. Diameter of Diameter at Threads Diameter of Diameter at Threads Screw, Root of Thread. per Inch, Screw, Root of Thread, per Inch, Inches. Inches. No. Inches. Inches. No. -A .185 20 2 1. 712 4Y2 TS .240 18 2*4 1 .962 * 4*6 % • 294 16 2*£ 2. 175 4 TS •344 14 2% 2-425 4 V2 .400 ^3 3 2.629 3Y2 1% •454 12 3*4 2.879 $& ¥s •507 11 3Y2 3.100 3*4 .620 •731 10 9 ■M 3-317 3 4 3-567 3 1 .837 8 4*4 3-798 2% l tt .940 7 \Yi 4.028 2 U *H 1.065 7 4H 4255 2% 1% 1. 160 6 5 4.480 2*6 iU 1.284 6 5*4 4 -730 2*£ *% I-3S9 5% 5Y2 5-053 2% M 1.490 5 sH 5.203 2% iVs 1.615 5 6 5-423 2*4 APPENDIX. 365 WROUGHT-IRON WELDED STEAM-, GAS-, AND WATER-PIPE. Diameter. Transverse Areas. Nominal Number of :kness. Weight per Foot. Threads per Inch of Nominal Actual Thi Actual External. Internal. Internal External. Internal. Inches. Inches. Inches. In ches. Sq. In. Sq. In. Pounds. H •4°5 •27 068 .129 •0573 .241 27 H .543 •364 088 .229 . 104 1 .42 18 % .675 •494 091 .358 .1917 •559 l8 \& .84 • 623 109 •554 .3048 •837 14 H 1.05 .824 113 .866 •5333 1. us 14 1 I -3 I S 1.048 i34 1-358 .8626 1.668 nH «H 1.66 1.38 14 2.164 1.496 2.244 "^ ij| 1.9 1. 611 M5 2.835 2.038 2.678 nH 2 2-375 2.067 i54 4-43 3-356 3.609 »fci 2^ 2.875 2.468 204 0.492 4-784 5-739 8 3 3-5 3.067 217 9.621 7.388 7-536 8 3*6 4- 3.548 226 12.566 9.8S7 9.001 8 4 4-5 4.026 237 15.904 12.73 10.665 8 4^ 5- 4.508 246 I9-635 15-961 12.34 8 5 5-563 5-045 259 24 . 306 19.99 14.502 8 6 6.625 6.065 28 34-472 28.888 18.762 8 7 7.625 7.023 301 45.664 38.738 23.271 8 8 8.625 7.982 322 58.426 50.04 28.177 8 9 9.625 8-937 344 72.76 62.73 33-701 8 10 10 -75 10.019 366 90.763 78.839 40.065 8 11 12 11.25 375 113.098 99 ■ 402 45-95 8 12 "•75 12 375 127.677 113.098 48.985 8 T 3 14 13-25 375 1 53-938 137.887 53-921 8 M 15 1425 375 176.715 159.485 57 • 893 8 15 16 15-25 375 201.062 182 655 61.77 8 18 1725 19.25 21.25 23.25 375 •375 375 375 254-47 314.16 380.134 452 39 233.706 291.04 354-657 424-558 69.66 77-57 85-47 93-37 24 WROUGHT-IRON WELDED EXTRA STRONG PIPE. H .405 .205 .1 .129 •033 .29 27 H •54 •294 .123 .229 .068 •54 18 % •675 .421 .127 .358 •139 •74 18 % .84 •542 .149 •554 •231 1. Og 14 1.05 .736 •157 .866 •452 1.39 14 . 1 1. 315 •951 .182 1-358 •7i 2.17 "^ & 1.66 1.272 .194 2.164 1.271 3 »vl 1.9 1.494 .203 2.835 1-753 3-6 3 nH 2 2-375 *-933 .221 4-43 2-935 5.02 "14 *\b 2.875 2-315 .28 6.492 4.209 7-6 7 8 3 3-5 2.892 •304 9.621 6.569 10.25 8 3^ 4 3-358 .321 12.566 8.856 12.47 8 4 4-5 3-8i8 •34i I5-904 11.449 14.97 8 5 5-563 4.813 •375 24.306 18.193 20.54 8 6 6.625 5-75 •437 34.472 25.967 28.58 8 366 APPENDIX. HEAT OF THE LIQUID— WATER. Temp. Deg. F. Heat of Temp. Deg. F. Heat of Temp. Heat of Temp. Heat of Temp. Heat of Liquid. Liquid. Deg. F. Liquid. Deg. F. Liquid. Deg. F. Liquid. t g t 4 / 1 t <7 t 1 32 o 69 37-12 106 74.O 143 III. 2 180 148.5 33 I.OI 70 38 II 107 75 144 112. 2 181 M9-5 34 2.01 71 39 " 108 76.0 145 113. 3 182 150.6 35 3 02 72 40.11 109 77.0 146 114.3 183 151. 6 36 4-03 73 41. 11 110 78.0 147 115 3 184 152.6 37 5 04 74 42 11 111 79.0 148 116.3 185 153-6 38 6.04 75 43-11 112 80.0 149 H7-3 186 154-6 39 705 76 44 J * 113 81.0 150 118.3 187 155-6 40 8.06 77 45.10 114 82.0 151 119.3 188 156.6 41 9.06 78 46. 10 115 83.0 152 120.3 189 157.6 42 10.07 79 47.09 116 84.0 153 121. 3 190 158.6 43 11.07 80 48.09 117 850 154 122.3 191 159.6 44 12.08 81 49.08 118 86 155 !23-3 192 160.6 45 13-08 82 50 08 119 87.0 156 124-3 193 161. 6 46 14.09 83 51-07 120 8S.1 157 125.4 194 162.6 47 15 09 84 52.07 121 89.1 158 126.4 195 163.7 48 16. 10 85 53.06 122 90.1 159 127.4 196 164.7 49 17.10 86 54- 06 123 91. 1 160 128.4 197 165.7 50 18.10 87 55-05 124 92. 1 161 129.4 198 166.7 51 19. 11 88 56.05 125 93-i 162 130.4 199 167.7 52 20.11 89 57.04 126 94-1 163 I3I-4 200 168.7 53 21. 11 90 58.04 127 95-1 164 132.4 201 169.7 54 22.11 91 59.03 128 96.1 165 133-4 202 170.7 55 23.11 92 60.03 129 97-1 166 134-4 203 171. 7 56 24.1 1 93 61 03 130 98.1 167 135-4 204 172.7 57 25.12 94 62.02 131 99.1 168 136.4 205 173-7 58 26.12 95 63.02 132 100.2 169 137.4 206 174.7 59 27.12 96 64.01 133 101.2 170 138.5 207 175.8 60 28.12 97 65.01 134 102.2 171 139-5 208 176.8 61 29.12 98 66.01 135 103.2 172 140.5 209 177.3 62 30.12 99 67.01 136 104.2 173 141. 5 210 178.8 63 31.12 100 68.01 137 105.2 174 142.5 211 179.8 64 32.12 101 69.01 138 106.2 175 143-5 212 180.8 65 33-12 102 70.00 139 107.2 176 M4-5 66 34- 12 103 71.00 140 108.2 177 145-5 67 35.12 104 72.0 141 109.2 178 146.5 68 36.12 105 73-o 142 110.2 179 147.5 VOLUME AND WEIGHT OF DISTILLED WATER. Temp. Weight of a Temp. Weight of a Temp. Weight of a Degrees Cubic Foot Degrees Cubic Foot Degrees Cubic Foot Fahr. in Pounds. Fahr. in Pounds. Fahr. in Pounds. 32 62.417 90 62.119 160 61.007 39-1 62.425 IOO 62 OOO 170 60.801 40 62.423 IIO 61.867 ISO 60.587 50 62.409 120 61.720 190 60.366 60 62.367 I30 61556 200 60.136 70 62.302 I40 61388 2IO 59.894 80 62.218 I50 61.204 212 59.707 APPENDIX. 367 PROPERTIES OF SATURATED STEAM. English Units. Presure, Temperature, Heat Total Heat. Heat of Volume in Weight in Pressure, Pounds Degrees of the Vaporiza- Cu. Ft. of Pounds of 1 Pounds per Sq.In. Fahr. Liquid. tion. 1 Pound. Cubic Foot per Sq. In. P t S A r s d / 5 162.34 I30.7 II3I-5 IOOO.8 73-22 O.O1366 5 10 193.25 161. 9 II4O.9 979 O 3S.16 O.02621 10 15 213.03 181. 8 II46.9 965.I 26.15 O.03826 15 20 227.95 196.9 II5I.5 954-6 19.91 O.05023 20 25 240.04 209.1 II55-1 946.0 16.13 O.06199 25 30 250.27 219.4 H5S.3 938.9 13-59 O.07360 30 35 259 19 228.4 JI61.O 932.6 H-75 O.08508 35 40 267.13 236.4 II63.4 927.0 10.37 O.09644 40 45 274.29 243.6 1 165.6 922.0 9.287 O.IO77 45 50 280.85 250.2 1167.6 917.4 8.414 O.II88 50 55 2S6.89 256.3 H69.4 913. 1 7.696 O.1299 55 • 60 292.51 261.9 I 171. 2 909- 3 7.096 O. 1409 60 65 297-77 267.2 II72.7 905-5 65S3 O.1519 65 70 302.71 272.2 1 1 74- 3 902. 1 6-144 O.1628 70 75 30708 276.9 II75-7 898.8 5-762 O.1736 75 80 311. So 281.4 1177.0 895.6 5 425 O.1843 80 85 316.02 2S5.8 117S.3 892.5 5-125 O.I95I 85 90 320.04 290.0 1179.6 889.6 4.858 O.2058 90 95 32389 294.0 1180.7 886.7 4.619 O.2165 95 100 327-58 297.9 1181.9 884.0 4 403 O.2271 100 105 33 r -i3 301.6 1182.9 8S1.3 4.206 O.2378 105 110 334-56 3052 1184.0 878.S 4.026 O.2484 110 115 337.86 30S.7 11S5.0 876.3 3.862 O.2589 115 120 341-05 312.0 1186.0 S74.0 3-7H 0.2695 120 125 344-13 315.2 1186.9 871.7 3-572 O.2S0O 125 130 347-12 318.4 1187.8 S69.4 3-444 O.2904 130 135 350.03 321.4 1188.7 867.3 3-323 O.3009 135 140 352.85 324.4 1189-5 865.1 3.212 O.31 13 140 145 355-59 327.2 1 190. 4 863.2 3.107 O.3218 145 150 358.26 330 1191.2 861.2 3. on O.3321 150 155 360.86 332.7 1192.0 859.3 2.919 O.3426 155 160 363-40 335-4 1192.8 857.4 2.833 0.3530 160 165 365.88 338.o II93-6 855.6 2.751 0.3635 165 170 368.29 340. 5 II94.3 853-8 2 676 0.3737 170 175 370.65 343-0 1195.0 852.0 2.603 O.3841 175 180 372.97 345 4 II95.7 850.3 2-535 3945 180 185 375-23 347-8 1 196.4 848.6 2.470 O.4049 185 190 377-44 350.I 1 197. 1 847.0 2.408 0-4I53 190 195 379-61 352.4 1197.7 845-3 2-349 0.4257 195 200 381.73 354-6 1198.4 843.8 2.294 0.4359 200 205 383-82 356.8 1199.0 842.2 2.241 O.4461 205 210 385.87 358.9 1199.6 840.7 2.190 O.4565 210 215 387.8S 361.0 1200.2 839.2 2.142 O.4669 215 220 389.84 363-0 1200.8 837.8 2.096 0.4772 220 225 391-79 365-1 1201.4 836.3 2.051 O.4876 225 230 393.69 367.1 1202.0 834.9 2.009 O.4979 230 235 395-56 369.0 1202.6 833.6 1.968 O.50S2 235 240 397 4i 371.0 1203.2 832.2 1.928 O.51S6 240 245 399.21 372.8 1203.7 830.9 1. 891 O 5289 245 250 400. 59 374 7 1204.2 829.5 1-854 0-5393 250 INDEX PAGE Accumulators 292, 293 Acetic acid 71 Adamson joints 210 Air for combustion 48, 51 dilution 51 loss from excess 61 per pound of coal 53 supply for boiler, measurement of 318 Almy boiler 34 Angle-valves 237 Anthracite coal , 37 Area of circles 362 steam-pipe 271 uptake 341 Areas of segments of circles 360, 361 Ash-pit 5 doors 5 Assembling and riveting boilers. ., 285 Atmosphere, composition of 49 Atomic weight , 45 Babcock & Wilcox 22 Baird's steam-trap 258 Belleville boiler 30 Belpaire fire-box 20, 158. Berryman feed-water heater 264 Blow-off pipes 4, 266 Blowing out brine 88 Blue heat 177 Board of Trade, rules for flues 218 Boilers, Almy 34 Babcock & Wilcox 22 Belleville 30 369 37° INDEX. PAGE Boilers, Cahall 27 Cornish <, 7 cylindrical tubular 2 double-ended 16 fire-engine 13 Galloway 7 gunboat , 18 Heine.... 25 Lancashire 6 Leavitt 20 locomotive 18, 20 Manning 9, 10 marine 14 (water-tube) 30 plain cylindrical 6 Root 25 Stirling 27 Thorny croft 32 two-flue 5 vertical 9, 11, 12, 13 water-tube , 21 (marine) 30 Yarrow 34 Boiler accessories 235 design 523 explosions 227 front , 5 horse-power 135 Boilers, life of 230 methods of supporting , 166 proportions of 138 Boiler-setting 91 shop, plan and description of 273 testing (evaporative) 300 tubes, size and surface 364 Boilers, U. S. S. Brooklyn , 146 Boring-mill 279 Brackets 166, 353 Brass 180 Bridge-wall 2 Bronze 180 Buck-staves < 94 Bumped-up head 186 Bundy steam-trap 260 Butt-joint 199, 201 INDEX. 371 PAGE Cahall boiler 27 Calculation of riveted joints 192 stay-rods 347,351 Calking 298 Calorimeters for steam, Carpenter 312 Peabody 309 Cam and toggle riveting- machine 288 Carbon, heat of combustion of 43, 44 Carbonate of lime 68 Cast iron 1 79 Chapman valves ... 238 Channel-bars, calculation of 348, 352 Charcoal 39 Check-valves 239 Chemistry of combustion .... 44 Chimney draught = 112 forms of 125 height of 125 stability of , 127 Chimneys 112 Kent's table 122, 124 Circles, area of 362 circumference of 362 Circumference of circles „ . , 362 Cleaning fires no Coals, anthracite 37 bituminous 38 caking, bituminous 38 dry bituminous 38 long-flaming bituminous 38 semi-bituminous 38 Coal, air per pound of 53 value of 131 Cold-water test 299 Collapsing pressure 207, 210-216 Coke 38 Combination 249 Combustion, air required for 48, 51 chemistry of 44 heat of 42-44 incomplete 60 rate of 1 36 temperature of 63 Composition 180 of atmosphere 49 2,7 2 INDEX. PAGE Composition of fuels 40 Compression. 175 Conical through-tubes , 7 Copper 179 Cornish boiler 7 Corrosion 84 Corrugated furnace , 211 Crane-lifts 277 Crown-bars. ... 20, 157 Crowfeet 152 Crushing of rivets 191 Curtis steam-trap 260 Cylinder, strength of 183, 184 Cylindrical tubular-boiler setting 91 staying of 148 Damper regulator 255 Density of gases 45 Detachable brackets 167 Diameter of boiler 327 of rivet 192 Diagonal stays 182 Du Long's formula 47 Double-ended boiler 16 Down-draught furnaces 105 Draught of chimneys 112 gauge 316 Howdens system no split 9 wheel 8 Drill for tube-holes 278 Drilled or punched plates 190 Dry pipe 164 Dudgeon tube-expander. . < 297 Economizer, Green's in Efficiency of riveted joints • 188 Elastic limit 174 Elasticity, modulus of 174 Equivalent evaporation 133 Evaporative test of boiler 300 Excess of air, loss from 61 Expanders for tubes 296, 297 Expansion pads 20 Explosions of boilers 227 index. 373 PAGE Factor of safety 224, 329 Farnley furnace 212 Feed-pipes 4, 264 pumps 265 water filter 77 Feed-water heaters (lime-extracting) 72 Berryman 264 Hoppes , 72 Wainwright 263 impurities in (table) 66 organic impurities 81 Filter, feed-water 77 Finishing flanges 279 Fire-doors 5 engine boiler , , 13 tubes 2, 220 Firing 100 Flange-punch 277 Flanging heads 275 Flat plates 222 Flexible tube 252 Flow of steam 314 Flue-gases 316 Flues 206 collapsing, pressure of 207, 210-216 discussion of tests 217 rules for 218 strengthened 2o3 Flynn steam-trap 259 Forms of test-pieces 171 Foundation-ring 9 Fox's corrugated furnace 211 Friction of riveted joints 191 Fuel, artificial 39 standard , , 130 Fuels, composition of 40, 41, 42 Furnace, corrugated s . . . 211-216 Farnley 's 212 Holmes 213 Morison's 216 Purve's 214,215 Furnaces. . . 95 down-draught 105 oil-burning 10S Fusible plugs 253 374 INDEX. PAGE Gas apparatus, Orsat's. 54 natural .... 39 Galloway boiler 7 General proportions of horizontal multitubular boiler 324 Girders , , 220 Globe valves. 235 Grate-area 325 bars gS water 107 Grates, rocking gg Green's economizer in Grooving 87 Gun boat boile* 18 Gun iron I7g Gusset stays 162, 183 Hand-holes 4, 166, 342 Hand-riveting 2g5 Hangers for pipe 270 Heat of combustion ... 44 (carbon) 43, 44 calculation of 46 (fuels) 42 the liquid (water). 366 Heating-surface. 5, 136, 328 of boiler-tubes (table) 364 value of 137 Heine boiler 25 Holmes' furnace 213 Hollow cylinder 183 Hoppe's purifier. 72 Horse-power of boilers 135 Howden's system no Hydraulic accumulators 2g2, 2g3 riveting-machine 288 with cam and toggle 2g3 test . . 224, 227 of boiler 2gg Incomplete combustion, loss from 60 Iron rods, weight of > 363 Kent's table of chimneys 122 Kerosene oil . 84 index. 375 PAGE Lancashire 6 La P J ( )i, 337 joints 192, 194 Lap-joint with welt 196, [98 Laying out plates 279 stays 345 Leavitt boiler 20 Length of sections 341 Lever safety-valve 242 Lewes (marine-boiler scale) 74 Life of boilers 230 Lifting-dogs 276 Lignite 33 Lime-extracting feed- water heater 72 Limit of elasticity 174 Lloyd's rules for flues 21S Locke damper regulator. . . „ 256 Locomotive-boiler iS staying of 155 type 20 Logarithms 358 Longitudinal joint 331 Mahler's composition of fuels 41 formula 4S Malleable iron 179 Manholes 4, 165, 342 Manning boilers , 9, 10 Marine boilers 14 water-tube , 30 proportions of 140 scale 74, 79 staying of 160 Materials 1 70 McDaniels trap 257 Mechanical stokers 102 Methods of failure of riveted joints 1S9 of making boiler-tests 300 of testing plate 172 Mineral impurities 67 matter in water (table) 66 oil 39 Modulus of elasticity 174 Morison's furnace 216 376 INDEX. PAGE Natural sines, cos, and tan 362 trigonometric functions 362 Naval boilers, proportions of. 141, 142 Nozzles 342 Oil-burning furnaces 108 Organic impurities in feed-water 81 Orsat's gas apparatus 54 Peat 38 Peclet's chimney theory. . . 116 Peet valve 238 Petroleums, composition of 40 Pipe, arrangement of steam 267 area and size of (Appendix) blow-off 266 feed 265 hangers for 270 size for given horse-power 271 sketches for ordering 269 Piping 267 Pitch of rivets 192 Pitting 85 Plain cylindrical boiler 6 Plan of boiler-shop 274 Planing-machine , 282 plates 282 Plate planer , 282 rolls , 282 Pop safety-valve 246 Portable riveting-machine 289 Power-pump for riveter 290 Priming 131, 308 Proportions of boilers 138 Properties of saturated steam 367 of steel 176 Prosser tube-expander 296 Pumps 265 Pump for hydraulic riveter 290 Punch > 282 and holder 278 for tube-holes 27S Punched or drilled plates 190 Purve's furnace 214, 215 Pyrometers 317 INDEX. Z/7 PAGE Rate of combustion « 136 Reach of a riveting-machine 289 Reducing-valve 254 Reduction of area . . . 175 Return steam-traps 260 Ring-seam 336 Rivet, diameter of 192 Riveted joints, calculation of 192 designing 202 efficiency of 188 friction of 191 limitations 205 methods of failure , 189 Riveting-machines, cam and toggle 288 hydraulic 288 with cam and toggle 293 portable 289 steam 294 Rivets 178 pitch of , 192 shearing and crushing 191 Rocking-grates 99 Rolls for plate 282 Roney stoker 103 Root boiler , 25 Safety plugs 253 valve 240 Sample boiler-test blank 319 Scale, marine-boiler 74, 79 Scarfing c 282 Screw-threads (table) 364 Sea-water, composition of 74 Segments of circles 360, 361 Selection of type of boiler 323 Semi-bituminous coal 3S Separator 262 Shearing , 175 of rivets 191 plates 2S1 Shears 281 Shop-practice , 272 Size and surface of boiler-tubes 364 of steam-pipe 271 Sizes of steam, gas, and water pipe (table) 365 378 INDEX. PAGE Smoke-box 5 prevention 104 Snap-riveting. 295 Soda 70 Specific heat 45 of superheated steam 307 volumes 45 Specifications and contract for boiler , . . . 354, 356 Sphere, strength of 185 Spherical ends 1 63 Split-draught 9 Stability of chimneys 127 Stay-bolts 156, 181 Stay-rods 182 calculation of 347, 351 Stayed flat plates 222 Staying , 148 (calculation of) 343 cylindrical tubular boiler 148 laying out 345 locomotive-boiler 155 of marine boiler 160 Stays, diagonal 182 Steam-dome 163 flow of 314 gas, and water pipe (table) 365 gauges 251 nozzle 165 piping 267 quality of 131 riveting-machine 294 space 326 tables 367 traps, Baird's 258 Bundy 260 Curtis. 260 Flynn 259 McDaniel's 257 return 260 Walworth 258 Steel 176 Stirling boiler 27 Stokers, mechanical 102 Strain 174 Stratton separator 262 INDEX. 379 PAGE Stress 174 Stretch limit 174 Submerged tube-sheet 12 Sunken tube-sheet 12 Superheating surface 1 1 Surface blow 82 Table of logarithms 353 Tables of properties of saturated steam 367 Tannic acid 7 1 Temperature of combustion 63 Test on furnace flues 207, 210-216 Testing boilers for evaporation 300 Testing-machines , . . 170 Test-pieces • - 1 7 l Testing plate, methods of 172 Thickness of shell ,. . . 331 Thornycroft boiler 32 Throttling calorimeter 309 Through-stays 2 Trowbridge's table of chimneys 125 Tube-expanders 296, 297 Tube-holes, drills for , 278 punch for 278 plates 2 sheet 339 Tubes 325 after expanding 297 Two flue boiler 5 Type of boiler, selection of 323 Ultimate elongation ■ 1 75 strength 174 Uptake 2 area of 341 U. S. Inspectors' rules for flues 218 Valves 235 angle 236 Chapman 238 check 239 gate 23S globe 235 Peet 238 reducing 254 380 INDEX. PAGE Valves safety lever 242 pop 246 Vertical boilers 9, 11, 12, 13 rolls for plate 284 Volumes, specific , . . . 45 Wainvvright feed-water heater 263 Walworth steam-trap 258 Wash-out plugs 166 Water column 249 grate 107 heat of the liquid 366 level 328 tube boilers 21 boiler-setting 94 marine boilers. .. . 30 weight and volume of (table). 366 Wheel-draught 8 William's composition of fuels 41 Wind pressure 127 Wood 39 Wrought iron .- 177 steam, gas, and water pipe 365 Yarrow boiler 34 Zinc in boilers , . . , 77 d PLATE I. PLATE II. -43^— -^---lm^ 1 -^ i."j FIG. 3 Copper Washer Ck— 1 FIG. 5 FIG. 6 PLATE II. FIG. 5 PLATE III. rz% PLUGS E. AND F. |E.78^ F. 110%"fROM BACK END. B^fevf-f-Mr & lis ^,1 \ 3? ,3? '.■2'\V-5 LONGITUDE 4 ^ 2}^ PLUG BACK END T L J2}£ PLUG FRONT END < REAR ELEVATION LOCOMOTIVE BOILER 160 LBS. PRESSURE PLATE III. T ^P-'-7f ■ — 7iSl'^~" w \ Until- -108ft—- FRONT HEAD. SECTION AT X.Y. 210 TUBES-2"OUTSIDE DIAM. LONGITUDINAL SEAM ON OUTSIDE OF FIRE BOX 2K> PLUG BACK END 2J4 PLUG FRONT END SECTION THROUGH FIRE BOX REAR ELEVATION PLATE IV. SECTION C-C LOOKING BACK NG FRONT. PLATE IV. SECTION C-C LOOKING BACK SECTION A-A LOOKING FRONT 10 3 Mo LOOKING FRONT. SHORT-TITLE CATALOGUE OF THE PUBLICATIONS OF JOHN WILEY & SONS, New York. London: CHAPMAN & HALL, Limited. ARRANGED UNDER SUBJECTS. Descriptive circulars sent on application. Books marked with an asterisk are sold at net prices only. All books are bound in cloth unless otherwise stated. AGRICULTURE. Cattle Feeding — Dairy Practice — Diseases of Animals — Gardening, Etc. Armsby's Manual of Cattle Feeding 12mo, $1 75 Downing's Fruit and Fruit Trees 8vo, 5 00 Grotenfelt's The Principles of Modern Dairy Practice. (Woll.) 12mo, 2 00 Kemp's Landscape Gardening 12mo, 2 50 Lloyd's Science of Agriculture 8vo, 4 00 Loudon's Gardening for Ladies. (Downing.) 12mo, 1 50 Steel's Treatise on the Diseases of the Dog 8vo, 3 50 " Treatise on the Diseases of the Ox 8vo, 6 00 Stockbridge's Rocks and Soils 8vo, 2 50 Woll's Handbook for Farmers and Dairymen , . . . .12mo, 1 50 ARCHITECTURE. Building — Carpentry— Stairs— Ventilation, Etc Berg's Buildings and Structures of American Railroads 4to, 7 50 1 Birkmire's American Theatres — Planning and Construction. 8vo, 3 00 " Architectural Iron and Steel 8vo, 3 50 Birkmire's Compound Riveted Girders 8vo, 2 00 " Skeleton Construction in Buildings 8vo, 3 00 1 Carpenter's Heating and Ventilating of Buildings 8vo, $3 00 Downing, Cottages 8vo, 2 50 and Wigktwick's Hints to Architects ,8vo, 2 00 Preitag's Architectural Engineering 8vo, 2 50 Gerhard's Sanitary House Inspection 16mo, 1 00 " Theatre Fires and Panics , 12mo, 1 50 Hatfield's American House Carpenter 8vo, 5 00 Holly's Carpenter and Joiner 18mo, 75 Kidder's Architect and Builder's Pocket-book Morocco flap, 4 00 Merrill's Stones for Building and Decoration 8vo, 5 00 Monckton's Stair Building— Wood, Iron, and Stone 4to, 4 00 Stevens' House Painting 18mo, 75 Worcester's Small Hospitals — Establishment and Maintenance, including Atkinson's Suggestions for Hospital Archi- tecture 12mo, 1 25 World's Columbian Exposition of 1893 4to, 2 50 ARMY, NAVY, Etc. Military Engineering— Ordnance — Port Charges, Etc. Bourne's Screw Propellers 4to, 5 00 Bruff's Ordnance and Gunnery 8vo, 6 00 Buckuill's Submarine Mines and Torpedoes 8vo, 4 00 Chase's Screw Propellers 8vo, 3 00 Cooke's Naval Ordnance 8vo, 12 50 Cronkhite's Gunnery for Non-com. Officers 18mo, morocco, 2 00 De Brack's Cavalry Outpost Duties. (Carr.). . . .18mo, morocco, 2 00 Dietz's Soldier's First Aid 12mo, morocco, 1 25 * Dredge's Modern French Artillery 4to, half morocco, 20 00 " Record of the Transportation Exhibits Building, World's Columbian Exposition of 1893.. 4to, half morocco, 15 00 Dyer's Light Artillery 12mo, 3 00 Hoff's Naval Tactics 8vo, 1 50 Hunter's Port Charges 8vo, half morocco, 13 00 Ingalls's Ballistic Tables 8vo, 1 50 " Handbook of Problems in Direct Fire 8vo, 4 00 Mahau's Advanced Guard 18mo, 1 50 Permanent Fortifications. (Mercur.).8vo, half morocco, 7 50 2 Mercur's Attack of Fortified Places 12mo, $2 00 Elements of the Art of War 8vo, 4 00 Metcalfe's Ordnance and Gunnery 12mo, with Atlas, 5 00 Phelps's Practical Marine Surveying 8vo, 2 50 Powell's Army Officer's Examiner 12mo, 4 00 Reed's Signal Service 50 SharpeV Subsisting Armies 18mo, morocco, 1 50 Strauss and Alger's Naval Ordnance and Gunnery Todd and Whall's Practical Seamanship 8v6, 7 50 Very's Navies of the World 8vo, half morocco, 3 50 Wheeler's Siege Operations 8vo, 2 00 Winthrop's Abridgment of Military Law 12mo, 2 50 Woodhull's Notes on Military Hygiene 12mo, morocco, 2 50 Young's Simple Elements of Navigation.. 12mo, morocco flaps, 2 50 ASSAYING. Smelting — Ore Dressing— Alloys, Etc. Fletcher's Quant. Assaying with the Blowpipe.. 12mo, morocco, 1 50 Furman's Practical Assaying 8vo, 3 00 Kunhardt's Ore Dressing 8vo, 1 50 * Mitchell's Practical Assaying. (Crookes.) 8vo, 10 00 O'Driscoll's Treatment of Gold Ores 8vo, 2 CO Pvicketts and Miller's Notes on Assaying 8vo, 3 00 Thurston's Alloys, Brasses, and Brouzes 8vo, 2 50 Wilson's Cyanide Processes 12rno, 1 50 ASTRONOMY. Practical, Theoretical, and Descriptive. Craig's Azimuth, 4to, 3 50 Doolittle's Practical Astronomy 8vo, 4 00 Gore's Elements of Geodesy 8vo, 2 50 Michie and Harlow's Practical Astronomy 8vo, 3 00 White's Theoretical and Descriptive Astronomy 12mo, 2 00 BOTANY. Gardening for Ladies, Etc. Baldwin's Orchids of New England 8vo, $1 50 Loudon's Gardening for Ladies. (Downing.) 12mo, 1 50 3 2 00 5 00 3 50 5 00 10 00 5 00 2 50 2 50 1 25 Thome's Structural Botany 18mo, $3 25 Westermaier's General Botany. (Schneider.) 8vo, 2 00* BRIDGES, ROOFS, Etc. Cantilever — Draw — Highway — Suspension. (See also Engineering, p. 6.) Boiler's Highway Bridges 8vo, * " The Thames River Bridge 4to, paper, Burr's Stresses in Bridges 8vo, Crehore's Mechanics of the Girder 8vo, Dredge's Thames Bridges 7 parts, Du Bois's Stresses in Framed Structures 4to, Foster's Wooden Trestle Bridges 4to, Greene's Arches in Wood, etc 8vo, •* Bridge Trusses 8vo, " Roof Trusses 8vo, Howe's Treatise on Arches 8vo, Johnson's Modern Framed Structures ... .4to, 10 00 Merriman & Jacoby's Text-book of Roofs and Bridges. Parti., Stresses 8vo, 2 50 Merriman & Jacoby's Text-book of Roofs and Bridges. Part II., Graphic Statics, 8vo. 2 50 Merriman & Jacoby's Text-book of Roofs and Bridges. Part III., Bridge Design Svo, 5 00 Merriman & Jacoby's Text-book of Roofs and Bridges. Part IV., Continuous, Draw, Cantilever, Suspension, and Arched Bridges (In preparation). * Morison's The Memphis Bridge Oblong 4to, 10 00 WaddelPs Iron Highway Bridges 8vo, 4 00 Wood's Construction of Bridges and Roofs 8vo, 2 00 Wright's Designing of Draw Spans 8vo, 2 50 CHEMISTRY. Qualitative — Quantitative — Organic — Inorganic, Etc. Adriance's Laboratory Calculations 12mo, 1 25 Allen's Tables for Iron Analysis 8vo, 3 00 Austen's Notes for Chemical Students 12mo, 1 50 Bolton's Student's Guide in Quantitative Analysis 8vo, 1 50 4 Classen's Analysis by Electrolysis. (Ilerrick.) 8vo, $3 00 Oafts's Qualitative Analysis. (Schaeffer.) 12rno, 1 50 Drecbsel's Chemical Reactions. (Merrill.) 12ino, 1 25 Fresenius's Quantitative Chemical Analysis. (Allen.) 8vo, 6 00 " Qualitative Chemical Analysis. (Johnson.). .. ..8vo, 4 00 Gill's Gas and Fuel Analysis 12mo, 1 25 Hammarsten's Physiological Chemistry (Mandel.) 8vo, 4 00 Ivolbe's Inorganic Chemistry 12mo, 1 50 Handel's Bio-chemical Laboratory 12mo, 1 50 Mason's Water Supply 8vo, 5 00 Miller's Chemical Physics 8vo, 2 00 Mixter's Elementary Text-book of Chemistry 12mo, 1 50 Morgan's Principles of Mathematical Chemistry 12mo, 1 50 " The Theory of Solutions and its Results 12mo, 1 00 Nichols's Water Supply (Chemical and Sanitary) 8vo, 2 50 O'Brine's Laboratory Guide to Chemical Analysis 8vo, 2 00 Perkins's Qualitative Analysis 12mo, 1 00 Pinner's Organic Chemistry. (Austen.) 12mo, 1 50 Ricketts and Russell's Notes on Inorganic Chemistry (Non- metallic) Oblong 8vo, morocco, 75 Schimpf's Volumetric Analysis 12mo, 2 50 Spencer's Sugar Manufacturer's Handbook . 12mo, morocco flaps, 2 00 Stockbridge's Rocks and Soils 8vo, 2 50 Troilius's Chemistry of Iron , 8vo, 2 00 Wiedemann's Chemical Lecture Notes 12mo, 3 00 " Sugar Analysis 8vo, 2 50 Wulling's Inorganic Phar. and Med. Chemistry 12mo, 2 00 DRAWING. Elementary — Geometrical— Topographical. Hill's Shades and Shadows and Perspective 8vo, 2 00 MacCord's Descriptive Geometry 8vo, 3 00 " Kinematics 8vo, 5 00 " Mechanical Drawing 8vo, 4 00 Mahan's Industrial Drawing. (Thompson.) 2 vols., 8vo, 3 50 Reed's Topographical Drawing. (H. A.) 4to, 5 00 Smith's Topographical Drawing. (Macmillan.) 8vo, 2 50 Warren's Descriptive Geometry 2 vols., 8vo, 3 50 5 Warreu's Drafting Instruments 12mo, 1 25 " Free-hand Drawing 12mo, $1 00 " Higher Linear Perspective 8vo, 3 50 " Linear Perspective. .. 12mo, 100 " Machine Construction 2 vols., 8vo, 7 50 " Plane Problems , 12mo, 125 " Primary Geometry 12mo, 75 " Problems and Theorems 8vo, 2 50 " Projection Drawing 12mo, 150 Shades and Shadows 8vo, 3 00 " Stereotomy— Stone Cutting 8vo, 2 50 Whelpley's Letter Engraving ,12mo, 2 00 ELECTRICITY AND MAGNETISM. Illumination— Batteries— Physics. Ant hoi;}' and Brackett's Text-book of Physics (Magie). . . .8vo, 4 00 Barker's Deep-sea Soundings 8vo, 2 00 Benjamin's Voltaic Cell 8vo, 3 00 Cosmic Law of Thermal Repulsion 18mo, 75 Crehorc and Squier's Experiments with a New Polarizing Photo- Chrouograpb 8vo, 3 00 * Dredge's Electric Illuminatious. . . .2 vols., 4to, half morocco, 25 00 Vol. II 4to, 7 50 Gilbert's De magnete. (Mottelay.) 8vo, 2 50 Holman's Precision of Measurements ■ 8vo, 2 00 Michie's Wave Motion Relating to Sound and Light, 8vo, 4 00 Morgan's, The Theory of Solutions and its Results 12mo, Niaudet's Electric Batteries. (Fishback.) 12mo, 2 50 Reagan's Steam and Electrical Locomotives 12rao 2 00 Thurston's Stationary Steam Engines for Electric Lighting Pur- poses 12mo, 1 50 Tillman's Heat 8vo, 1 50 ENGINEERING. Civil — Mechanical— Sanitary, Etc. (See also Bridges, p. 4 ; Hydraulics, p. 8 ; Materials of En- gineering, p. 9 ; Mechanics and Machinery, p. 11 ; Steam Engines and Boilers, p. 14.) Baker's Masonry Construction. „.,„...* .8vo, 5 00 6 Baker's Surveying Instruments 12mo, Black's U. S. Public Works 4to, Butts's Engineer's Field-book 12mo, morocco, Byrne's Highway Construction 8vo, Carpenter's Experimental Engineering 8vo, Church's Mechanics of Engineering — Solids and Fluids 8vo, " Notes and Examples in Mechanics 8vo, CrandaH's Earthwork Tables 8vo, Craudall's The Transition Curve 12mo, morocco, * Dredge's Penn. Railroad Construction, etc. . . Folio, half mor., - Drinker's Tunnelling 4to, half morocco, Eissler's Explosives — Nitroglycerine and Dynamite 8vo, Gerhard's Sanitary House Inspection lo'nio, Godwin's Railroad Engineer'sField-book.l2mo, pocket-bk. form, Gore's Elements of Goodesy 8vo, Howard's Transition Curve Field-book 12mo, morocco flap, Howe's Retaining Walls (New Edition.) 12mo, Hudson's Excavation Tables. Vol. II 8vo, Hut ton's Mechanical Engineering of Power Plants 8vo, Johnson's Materials of Construction 8vo, Johuson's Stadia Reduction Diagram. .Sheet 22A X 28i inches, " Theory and Practice of Surveying. 8vo, Kent's Mechanical Engineer's Pocket-book 12mo, morocco, Kiersted's Sewage Disposal 12mo, Kirk wood's Lead Pipe for Service Pipe 8vo, Mahan's Civil Engineering. (Wood.) 8vo, Merriman and Brook's Handbook for Surveyors. . . .12mo, mor., Merriman's Geodetic Surveying 8vo, " Retaining Walls and Masonry Dams 8vo, Mosely's Mechanical Engineering. (Mahau.) 8vo, Nagle's Manual for Railroad Engineers .12mo, morocco, Pattou's Civil Engineering 8vo, " Foundations 8vo, Rockwell's Roads and Pavements in France 12mo, Ruffuer's Non-tidal Rivers : 8vo, Searles's Field Engineering 12mo, morocco flaps, Searles's Railroad Spiral 12mo, morocco 7 3 00 $5 00 2 50 5 00 G 00 G 00 2 00 1 50 1 50 20 00 25 00 4 00 1 00 2 50 1 50 1 25 1 00 5 00 6 00 50 4 00 5 00 1 25 1 50 5 00 2 00 2 00 2 00 5 00 7 50 5 00 1 25 1 25 3 00 1 50 Siebert and Biggin's Modern Stone Cutting and Masonry. . .8vo, 1 50 Smith's Cable Tramways 4to, $2 50 • * Wire Manufacture and Uses 4to, 3 00 Spalding's Roads and Pavements 12mo, 2 00 " Hydraulic Cement 12mo, Thurston's Materials of Construction 8vo, 5 00 * Trautwine's Civil Engineer's Pocket-book. ..12mo, mor. flaps, 5 00 * ' ' Cross-section Sheet, 25 * Excavations and Embankments 8vo, 2 00 * " Laying Out Curves 12mo, morocco, 2 50 Warren's Stereotomy— Stone Cutting 8vo, 2 50 Webb's Engineering Instruments 12mo, morocco, 1 00 Wegmann's Construction of Masonry Dams 4to, 5 00 Wellington's Location of Railways. . - 8vo, 5 00 Wheeler's Civil Engineering • 8vo, 4 00 Wolff's Windmill as a Prime Mover 8vo, 3 00 HYDRAULICS. Water-wheels— Windmills — Service Pipe — Drainage, Etc. (See also Engineering, p. 6. ) Bazin's Experiments upon the Contraction of the Liquid Vein (Trautwine) 8vo, 2 00 Bovey's Treatise on Hydraulics. 8vo, 4 00 Coffin's Graphical Solution of Hydraulic Problems. 12mo, mor., Ferrers Treatise on the Winds, Cyclones, and Tornadoes. . .8vo, 4 00 Ganguillet & Kutter'sFlow of Water. (Hering& Trautwine ).8vo, 4 00 Hazen's Filtration of Public Water Supply 8vo, 2 00 Kiersted's Sewage Disposal 12mo, 1 25 Kirkwood's Lead Pipe for Service Pipe. 8vo, 1 50 Mason's Water Supply 8vo, 5 00 Merriman's Treatise on Hydraulics. 8vo, 4 00 Nichols's Water Supply (Chemical and Sanitary) 8vo, 2 50 Ruffner's Improvement for Non-tidal Rivers 8vo, 1 25 Wegmaun's Water Supply of the City of New York 4to, 10 00 Weisbach's Hydraulics. (Du Bois.) 8vo, 5 00 Wilson's Irrigation Engineering, ... 8vo. 4 00 Wolff's Windmill as a Prime Mover 8vo, 3 00 Wood's Theory of Turbines 8vo, 2 50 8 MANUFACTURES. Aniline— Boilers— Explosives— Iron— Sioar— Watches — Woollens, Etc. Allen's Tables for Iron Analysis 8vo, $3 00 Beaumont's Woollen and Worsted Manufacture 12mo, 1 50 Bollaud's Encyclopaedia of Founding Terms 12mo, 3 00 The Iron Founder 12mo, 2 50 Supplement 12mo, 2 50 Booth's Clock and Watch Maker's Manual 12mo, 2 00 Bouvier's Handbook on Oil Painting 12mo, 2 00 Eissler's Explosives, Nitroglycerine and Dynamite 8vo, 4 00 Ford's Boiler Making for Boiler Makers 18mo, 1 00 Metcalfe's Cost of Manufactures 8vo, 5 00 Metcalf's Steel— A Manual for Steel Users 12mo, 2 00 Keimann's Aniline Colors. (Crookes.) 8vo, 2 50 * Reisig's Guide to Piece Dyeing 8vo, 25 00 Spencer's Sugar Manufacturer's Handbook. . . .12mo, mor. flap, 2 00 Svedelius's Handbook for Charcoal Burners 12mo, 1 50 The Lathe and Its Uses 8vo, 6 00 Thurston's Manual of Steam Boilers 8vo, 5 00 West's American Foundry Practice 12mo, 2 50 •' Moulder's Text-book 12mo, 2 50 Wiechmann's Sugar Analysis 8vo, 2 50 Woodbury's Fire Protection of Mills 8vo, 2 50 MATERIALS OF ENGINEERING. Stren gth — Elasticity — Resistance, Etc {See also Engineering, p. 6.) Baker's Masonry Construction 8vo, Beardslee and Kent's Strength of Wrought Iron 8vo, Bovey's Strength of Materials 8vo, Burr's Elasticity and Resistance of Materials 8vo, Byrne's Highway Construction 8vo, Carpenter's Testing Machines and Methods of Testing Materials Church's Mechanic's of Engineering — Solids and Fluids 8vo, Du Bois's Stresses in Framed Structures 4to, Hatfield's Trausverse Strains 8vo, Johnson's Materials of Construction 8vo, 9 5 00 1 50 7 50 5 00 5 00 6 00 .0 00 5 00 6 00 Lanza's Applied Mechanics. 8vo, $7 50 " Strength of Wooden Columns 8vo, paper, 50 Merrill's Stones for Building and Decoration 8vo, 5 00 Merriinan's Mechanics of Materials 8vo, 4 00 Patton's Treatise on Foundations 8vo, 5 00 Rockwell's Roads and Pavements in France 12mo, 1 25 Spalding's Roads and Pavements 12mo, 2 00 Hydraulic Cement 12mo, Thurston's Materials of Construction, 8vo, 5 00 Thurston's Materials of Engineering 3 vols., 8vo, 8 00 Vol. I., Non-metallic 8vo, 2 00 Vol. II., Iron and Steel 8vo, 3 50 Vol. III., Alloys, Brasses, and Bronzes 8vo, 2 50 "Weyrauch's Strength of Iron and Steel. (Du Bois.) .8vo, 1 50 Wood's Resistance of Materials 8vo, 2 00 MATHEMATICS. Calculus— Geometry— Trigonometry, Etc. Baker's Elliptic Functions ., . .8vo, 1 50 Ballard's Pyramid Problem 8vo, 1 50 Barnard's Pyramid Problem 8vo, 1 50 Bass's Differential Calculus 12mo, 4 00 Brigg's Plane Analytical Geometiy 12mo, 1 00 Chapman's Theory of Equations 12mo, 1 50 Chessin's Elements of the Theory of Functions Compton's Logarithmic Computations 12mo, 1 50 Craig's Linear Differential Equations 8vo, 5 00 Davis's Introduction to the Logic of Algebra 8vo, 1 50 Halsted's Elements of Geometry c..8vo, 175 " Synthetic Geometry 8vo, 150 Johnson's Curve Tracing 12mo, 1 00 " Differential Equations— Ordinary and Partial 8vo, 3 50 " Integral Calculus 12mo, 1 50 " Least Squares 12mo, 1 50 Ludlow's Logarithmic and Other Tables. (Bass.) 8vo, 2 00 " Trigonometry with Tables. (Bass.) 8vo, 3 00 Mahan's Descriptive Geometry (Stone Cutting) 8vo, 1 50 Merriman and Woodward's Higher Mathematics 8vo, 5 00 Merrirnan's Method of Least Squares 8vo, 2 00 10 Parker's Quadrature of the Circle 8vo, $2 50 Rice and Johnson's Differential and Integral Calculus, 2 vols, in 1, 12mo, 2 50 Differential Calculus 8vo, 3 50 " Abridgment of Differential Calculus 8vo, 1 50 Searles's Elements of Geometry 8vo, 1 50 Touch's Metrology 8vo, 2 50 Warren's Descriptive Geometry 2 vols., 8vo, 3 50 " Drafting Instruments 12mo, 125 Free-hand Drawing 12mo, 100 " Higher Linear Perspective 8vo, 3 50 " Linear Perspective 12mo, 1 00 Primary Geometry 12mo, 75 " Plane Problems. .12mo, 1 25 " Plane Problems 12mo, 125 ' ' Problems and Theorems 8vo, 2 50 " Projection Drawing 12mo, 1 50 Wood's Co-ordinate Geometry 8vo, 2 00 " Trigonometry 12mo, 1 00 Woolf's Descriptive Geometry Royal 8vo, 3 00 MECHANICS-MACHINERY. Text-books and Practical Works. (See also Engineering, p. 6.) Baldwin's Steam Heating for Buildings 12mo, 2 50 Benjamin's Wrinkles and Recipes 12mo, 2 00 Carpenter's Testing Machines and Methods of Testing Materials 8vo, Chordal's Letters to Mechanics 1 2mo, 2 00 Church's Mechanics of Engineering 8vo, G 00 " Notes and Examples in Mechanics 8vo, 2 00 Crehore's Mechanics of the Girder 8vo, 5 00 Cromwell's Belts and Pulleys 12mo, 1 50 Toothed Gearing 12mo, 1 50 Comptou's First Lessons in Metal Working 12mo, 1 50 Dana's Elementary Mechanics 12mo, 1 50 Dingey's Machinery Pattern Making 12mo, 2 00 11 Dredge's Trans. Exhibits Building, World Exposition, 4to, half morocco,$15 00 Du Bois's Mechanics. Vol. I., Kinematics 8vo, 3 50 Vol.11., Statics 8vo, 4 00 Vol. Ill, Kinetics 8vo, 3 50 Fitzgerald's Boston Machinist 18mo, 1 00 Flather's Dynamometers 12mo, 2 00 Rope Driving 12mo, 2 00 Hall's Car Lubrication 12mo, 1 00 Holly's Saw Filing 18mo, 75 Lanza's Applied Mechanics 8vo. 7 50 MacCord's Kinematics 8vo, 5 00 Merriman's Mechanics of Materials , 8vo, 4 00 Metcalfe's Cost of Manufactures 8vo, 5 00 Michie's Analytical Mechanics 8vo, 4 00 Mosely's Mechanical Engineering. (Mahan.) 8vo, 5 00 Richards's Compressed Air * 12mo, 1 50 Robinson's Principles of Mechanism 8vo, 3 00 Smith's Press-working of Metals 8vo, 3 00 The Lathe and Its Uses 8vo, 6 00. Thurston's Friction and Lost Work 8vo, 3 00 " The Animal as a Machine 12mo, 1 00 Warren's Machine Construction 2 vols., 8vo, 7 50 Weisbach's Hydraulics and Hydraulic Motors. (Du Bois.)..8vo, 5 00 " Mechanics of Engineering. Vol. III., Part L, Sec. I. (Klein.) 8vo, 5 00 Weisbach's Mechanics of Engineering. Vol. III., Part I., Sec. II. (Klein.) 8vo, 5 00 Weisbach's Steam Engines. (Du Bois.) ■ 8vo, 5 00 Wood's Analytical Mechanics 8vo, -3 00 " Elementary Mechanics 12mo, 125 Supplement and Key . ... 125 METALLURGY. Iron— Gold— Silver — Alloys, Ets. Allen's Tables for Iron Analysis 8vo, 3 00 Egleston's Gold and Mercury 8vo, 7 50 12 Egleston's Metallurgy of Silver 8vo, $7 50 * Kerl's Metallurgy— Copper and Iron 8vo, 15 00 * " Steel, Fuel, etc 8vo, 15 00 Kunbardt's Ore Dressing in Europe 8vo, 1 50 Metcalf Steel— A Manual for Steel Users 12mo, 2 00 O'Driscoll's Treatment of Gold Ores 8vo, 2 00 Thurston's Iron and Steel 8vo, 3 50 Alloys 8vo, 2 50 "Wilson's C} r anide Processes 12mo, 1 50 MINERALOGY AND MINING. Mine Accidents — Ventilation — Ore Dressing, Etc. Beard's Ventilation of Mines 12mo, Boyd's Resources of South Western Virginia 8vo, Map of South Western Virginia Pocket-book form, Brush aud Penfield's Determinative Mineralogy 8vo, Chester's Catalogue of Minerals 8vo, " Dictionary of the Names of Minerals 8vo, Dana's American Localities of Minerals 8vo, Descriptive Mineralogy. (E. S.) 8vo, half morocco, Mineralogy and Petrography. (J. D. ) 12mo, Minerals and How to Study Them. (E. S.) 12mo, Text-book of Mineralogy. (E. S.) 8vo, inker's Tunnelling, Explosives, Compounds, and Rock Drills. 4to, half morocco, Egleston's Catalogue of Minerals and Synonyms 8vo, Eissler's Explosives— Nitroglycerine and Dynamite 8vo, Goodyear's Coal Mines of the Western Coast 12mo, Hussak's Rock- forming Minerals. (Smith.) 8vo, Ihlseng's Manual of Mining 8vo, Kunbardt's Ore Dressing in Europe , 8vo, O'Driscoll's Treatment of Gold Ores 8vo, Rosenbusch's Microscopical Physiography of Minerals and Rocks. (Iddings.) 8vo, Sawyer's Accidents in Mines 8vo, Stockbridge's Rocks and Soils 8vo, 13 *Dr: 2 50 3 00 2 00 3 50 1 25 3 00 1 00 12 50 2 00 1 50 3 50 25 00 2 50 4 00 2 50 2 00 4 00 1 50 2 00 5 00 7 00 2 50 Williams's Lithology 8vo, $3 00 Wilson's Mine Ventilation 16mo, 1 25 STEAM AND ELECTRICAL ENGINES, BOILERS, Etc. Stationary— Marine— Locomotive — Gas Engines, Etc. (See also Engineering, p. 6.) Baldwin's Steam Heating for Buildings 12mo, Clerk's Gas Engine t 12mo, Ford's Boiler Making for Boiler Makers 18mo, Hemen way 's Indicator Practice 1 2mo, Hoadley's Warm-blast Furuace 8vo, Kneass's Practice and Theory of the Injector 8vo, MacCord's Slide Yalve 8vo, * Maw's Marine Engines Folio, half morocco, Meyer's Modern Locomotive Construction 4to, Peabody and Miller's Steam Boilers 8vo, Peabody's Tables of Saturated Steam 8vo, " Thermodynamics of the Steam Engine 8vo, " Yalve Gears for the Steam Engine 8vo, Pray's Twenty Years with the Indicator Royal 8vo, Pupin and Osterberg's Thermodynamics 12mo, Reagan's Steam and Electrical Locomotives 12mo, Rontgen's Thermodynamics. (Du Bois.) 8vo, Sinclair's Locomotive Running 12mo, Thurston's Boiler Explosion 12mo, Engine and Boiler Trials 8vo, " Manual of the Steam Engine. Part L, Structure and Theory, 8vo, 7 50 Manual of the Steam Engine. Part II., Design, Construction, and Operation 8vo, 2 parts, " Philosophy of the Steam Engine 12mo, " Reflection on the Motive Power of Heat. (Carnot.) 12mo, " Stationary Steam Engines 12mo, " Steam-boiler Construction and Operation 8vo, 14 2 50 4 00 1 00 2 00 1 50 1 50 18 00 10 00 1 00 5 00 2 50 2 50 1 25 2 00 5 00 2 00 1 50 5 00 7 50 12 00 75 2 00 1 50 5 00 Spangler's Valve Gears 8v« >, $2 51 1 Trowbridge's Stationary Steam Engines 4to, boards, 2 50 Weisbach's Steam Engine. (Du Bois.) 8vo, Whitham's Constructive Steam Engineering 8vo, 10 00 Steam-engine Design 8vo, 6 00 Wilson's Steam Boilers. (Flather.) 12mo, 2 50 Wood's Thermodynamics, TIeat Motors, etc 8vo, 4 00 TABLES, WEIGHTS, AND MEASURES. fok actuabies, chemists, engineers, mechanics— metric Tables, Etc. Adrian ee's Laboratory Calculations 12mo, 1 25 Allen's Tables for Iron Analysis 8vo, 3 00 Bixby's Graphical Computing Tables Sheet, 25 Comptou's Logarithms 12mo, 1 50 Crandall's Railway and Earthwork Tables 8vo, 1 50 Egleston's Weights and Measures ISmo, 75 Fisher's Table of Cubic Yards Cardboard, 25 Hudson's Excavation Tables. Vol. II 8vo, 1 00 Johnson's Stadia and Earthwork Tables , 8vo, 1 25 Ludlow's Logarithmic and Other Tables. (Bass.) 12mo, 2 00 Thurston's Conversion Tables 8vo, 1 00 Totten's Metrology 8vo, 2 50 VENTILATION. Steam Heating — House Inspection — Mine Ventilation. Baldwin's Steam Heating 12mo, 2 50 Beard's Ventilation of Mines 12mo, 2 50 Carpenter's Heating and Ventilating of Buildings 8vo, 3 00 Gerhard's Sanitary House Inspection Square 16mo, 1 00 Mott's The Air We Breathe, and Ventilation 16mo, 1 00 Reid's Ventilation of American Dwellings 12mo, 1 50 Wilson's Mine Ventilation 16mo, 1 25 HISCELLANEOUS PUBLICATIONS. Alcott's Gems, Sentiment, Language Gilt edges, 5 00 Bailey's The New Tale of a Tub .8vo, 75 15 $1 50 1 50 1 50 4 00 1 00 1 50 3 00 1 50 2 50 25 3 00 Ballard's Solution of the Pyramid Problem 8vo, Barnard's The Metrological System of the Great Pyramid. .8vo, Emmon's Geological Guide-book of the Rocky Mountains. .8vo, Ferrel's Treatise on the Winds 8vo, Mott's The Fallacy of the Present Theory of Sound. .Sq. 16mo, Perkins's Cornell University Oblong 4to, Rieketts's History of Rensselaer Polytechnic Institute 8vo, Rotherham's The New Testament Critical!}' Emphathized. 12nao, Totteu's An Important Question in Metrology 8vo, Wkitehouse's Lake Mceris Paper, * Wiley's Yosemite, Alaska, and Yellowstone 4to, HEBREW AND CHALDEE TEXT=BOOKS. For Schools and Theological Seminaries. Gesenius's Hebrew and Chaldee Lexicon to Old Testament. (Tregelles.) Small 4to, half morocco, 5 00 Green's Elementary Hebrew Grammar. 12mo, 1 25 " Grammar of the Hebrew Language (New Edition). 8 vo, 3 00 " Hebrew Chrestomathy 8vo, 2 00 Letteris's Hebrew Bible (Massoretic Notes in English). 8vo, arabesque, 2 25 Luzzato's Grammar of the Biblical Cbaldaic Language and the Talmud Babli Idioms 12mo, 1 50 MEDICAL. Bull's Maternal Management in Health and Disease 12mo, 1 00 Hammarsten's Physiological Chemistry. (Mandel.) 8vo, 4 00 Mott's Composition, Digestibility, and Nutritive Value of Food. Large mounted chart, 1 25 Steel's Treatise on the Diseases of the Ox 8vo, 6 00 Treatise on the Diseases of the Dog 8vo, 3 50 Worcester's Small Hospitals — Establishment and Maintenance, including Atkinson's Suggestions for Hospital Archi- tecture 12mo, 1 25 16 r > ^> ^ /* A ^ ^. +*, ..V s V