Under the head of "plain" boilers come all ordinary cylindrical boilers, with or with out internal furnaces, horizontal or vertical. They are the cheapest and simplest which can be made, and, if properly proportioned, possess a considerable evaporative efficiency. When it is necessary, however, to economize fuel, or space, or both, "multitubular" boilers are used. These derive their name from the fact that in them the flame and gases of combustion are made to pass through a great number of small tubes (surrounded by the water) on their way to the chimney. The steam-generating power of a boiler depends greatly on the extent of surface which it presents to the flame, and it is obvious that a great number of small tubes present a much larger surface than one large tube occupying the space of them all. Thus, with the same heating surface, a multitubular boiler will occupy much less space than a plain one, and at the same time the efficiency of its surface is found to be greater. It is, however, necessarily more expensive and more liable to get out of order. Tubulous boilers differ from multitubular boilers in not only containing tubes, but consisting of them, and having no large cylinders whatever. Their chief advantages are (a) their great strength, for it is easy to make a wrought-iron tube strong enough to withstand pressures far higher than any at present in use; and (b) the peculiarity, that if any accident happens, or explosion occurs. it will only be to one tube at a time, and not to an immense boiler shell (as with the common boiler), and its evil consequences will thus be greatly reduced. For this reason tubulous boilers are often called "safety" boilers. It will be readily understood that there is no distinct line of demarkation between the three classes of which we have been speaking, but that on account of the immense variety of boilers which have been designed and constructed, those of one class pass through gentle gradations into those of the next. The commonest form of boiler for factories, etc., is that known as the Cornish, and shown in fig. 1. It consists simply of a cylindrical shell, a, a, inclosing a much smaller

cylinder, f, f, called a flue. The ends of the flue are open, but the space between it and the shell, which contains the water, is of course closed up and made steam-tight. The fire-grate, d, is in the interior of the flue, and at the end of it is a brick bridge, e, made so as to cause the flame to impinge on the upper side of the flue. The boiler is set in brick-work; and the flame, passing out at the back end of the flue, is made to traverse the whole length of the boiler twice through brick flues before passing away to the chimney.

Fig. 1.

The Cornish boiler has often two internal flues or tubes, which is a much more advantageous construction than that shown in fig. 1. The Galloway boiler (called after its inventor) is a very excellent modification of the Cornish, which in outward appearance it exactly resembles. It has two furnaces, but these join together in one chamber just behind the bridges, and the gases are made to pass through a space considerably narrowed by side pockets projecting inwards in order that they may be well mixed. From this point to the back of the boiler there is just one flue, made oval in section, and crossed by a considerable number of vertical taper tubes, which form a direct communication between the water beneath and that above the flue. These tubes (called Galloway tubes") both promote circulation and strengthen the flue. Multitubular boilers of many kinds are used, both for stationary engines and other purposes, but the largest number of those constructed are certainly for steamers, and a common type of marine boiler is shown in fig. 2. The shell, a, a, is cylindrical, and contains one or more cylindrical furnaces; c is the fire-grate; d, a brick bridge; e, a combustion chamber or flame-box; f. the tubes through which the flame passes back to the front of the boiler; and g, the

b

h

smoke-box at the base of the funnel, . The line k, k, shows the ordinary level of the water in the boiler. Un board ship it is of course an object to take up as little space as possible with boiler and machinery, and at the same time to have boilers which shall use as little coal as possible, both because of the saving in cost, and because of the sav ing in the room taken up by coal. For all these reasons, marine boilers are invariably multitubular.

The varieties of vertical boilers are as numerous as those of horizontal. When dirty water, or water containing much insoluble sediment, has to be used (as e.g. in steam-cranes frequently), they are of the simplest constraction, with nothing but an inner fire-box and an outer shell (both cylindrical), the space between them being filled with water all round and over the top of the fire-box.

Fig. 2.

Fig. 3.

If clean water can be had, however, and it is desired to be at all economical of fuel, some kind of multitubular vertical boiler must be used, and of these probably the best is that known as the Field boiler, and shown in fig. 3. The peculiarity of it consists in the tubes, which are closed at the bottom, and hang down from the top of the fire-box over the grate bars, and contain inner tubes of much smaller diameter. The latter are intended to aid the circulation of the water, which passes down the inner tube and up again through the annular space around it, where, being most exposed to the action of the flame, it is hottest. Of the different varieties of tubulous boilers, those manufactured by Messrs. Howard of Bedford have found most favor; but so far as can be said in the absence of any extended experience as to their working, Sinclair's patent boilers seem to be even more satisfactory. They consist of horizontal wrought-iron welded tubes placed in vertical rows, each row being connected at each end with a vertical tube, also of wrought-iron and of larger diameter. In order that the horizontal tubes may be properly fixed in the vertical ones, a hole must be provided in the side of the latter, opposite the mouth of each of the former. That these holes may be kept tight at any pressure of steam, the ingenious device is adopted of closing them with taper plugs put in from the inside, so that the pressure of steam keeps them shut, and the higher the pressure the less possibility of leakage there is. Locomotive boilers are always multitubular, for much the same reasons as marine boilers. The boiler of a single locomotive often contains 1500 or 1800 sq. ft. of heating surface, and occasionally as much as 2000 sq. feet.

The principal test of the efficiency of a boiler is the quantity of water (generally expressed either in pounds or gallons), which it will evaporate with the consumption of one of coal. Of course this varies very much with the quality of the fuel, but with good pit coal (not dross), a Cornish boiler should evaporate 6 to 8 lbs. of water per lb. coal, and a multitubular boiler, such as fig. 2, about 10 or 11 lb. per lb. coal. The best rate of combustion on the grate varies with the construction of the boiler, from 10 to 18 or 20 lb. per sq. ft. of grate surface per hour.

Boilers are almost invariably made of wrought-iron plates riveted together. The parts most exposed to the action of the flame are made of the best quality of iron, and the other parts of inferior qualities, according to their position in reference to the flame. Occasionally boilers are made of steel, where lightness is the chief requisite, but makers have not yet sufficient confidence in steel plates to use them very largely. Copper is often used in the fire-boxes of locomotives, but seldom in any other description of boiler. Brass boiler tubes are often seen, and on account of its better conducting qualities, brass is to be preferred to iron, but its dearness prevents its superseding iron in the great majority of cases.

Every boiler has, to render it complete and workable, a number of fittings or mountings, of which the following are the principal: A glass gauge to show the level of the water inside the boiler, and gauge-cocks for the same purpose; a gauge to show the pressure of the steam; a cock for admitting water; a cock at the bottom for emptying or "blowing off;" a valve for the discharge of the steam; one or two safety valves, weighted so that, when the pressure of steam in the boiler reaches a certain height, they open and allow the steam to rush into the air; a door by which a man can get in to clean the boiler, etc.

In order to induce a current of air through the furnace so that a proper combustion may be maintained, stationary boilers are generally provided with a chimney of consid erable height, and made of brick or sheet-iron, to which the products of combustion are conducted after they have left the boiler. In locomotive boilers, and in some other cases where a sufficiently tall chimney cannot be made use of, a very powerful current is made by the ejection of the waste steam through a blast-pipe with a contracted nozzle at the base of the chimney. To prevent loss of heat by radiation, and the consequent waste of fuel, boilers should always be covered, in all parts exposed to the atmosphere, with felt or some non-conducting composition.

For further details see also BOILING, MANOMETER, SAFETY-VALVE, STEAM, STEAMENGINE, STEAM-CRANE.

BOILERS (ante). Those most distinctively American are the sectional or water-tube boilers. The Babcock and Wilcox boiler consists of a series of tubes inclined from the front to the rear, and connected at each end by a manifold chamber. The forward ends are connected to the steam-drum, which lies lengthwise of the boiler. The tubes and manifolds are in the fire chamber, and there are two sets of diaphragm plates, by which the hot gas, after rising, is deflected, first downward and then upward, being made to cross the stack of tubes three times before making its exit into the chimney. The water fills the tubes and occupies the lower part of the steam-drum. The tubes of the Root boiler are likewise inclined from front to rear; they are joined at the ends by triangular caps and crow-feet, and the joints are perfected by rubber gaskets. The joints are outside the fire chamber, and the steam-drum lies crosswise of the boiler. The water does not fill all the tube-space within the fire-box, nor enter the steam-drum; by this means dry steam is secured, while the danger of superheating is but slight, as the space not reached by the water lies in the upper and forward part of the fire-box. The Whittingham boiler has its tubes, connections, and steam-drum, all inclosed in the fire-box; the tubes are traversed by interior tubes, or flues, through which the hot gases are conveyed, and thus a large fire surface is secured. The Harrison boiler is made of cast-iron spher

Boiling.

ical shells, 8 in. in external diameter, and of an in. thick; they are cast in sections, 2 or 4 spheres together, are connected by curved necks of 3 in. diameter, and are held together by wrought-iron bolts and caps. The joints are accurately fitted, without 'packing.

The water surface of a boiler is that area of metal which has water within and flame or hot gases without; at this surface the steam is generated. The area which has hot gases without and steam within is superheating surface, at which the steam by the recep tion of heat acquires greater expansive force. The draught-area, or calorimeter, is the cross section of the area traversed by the hot gases from the fire, and may be taken at any point between the furnace and the chimney. Ordinarily, however, it is restricted to the space around the tubes in the water-tube boilers, and to the section of the flues in flue-boilers. That boiler is most efficient which shows the greatest difference between the furnace temperature and that found at the chimney, since that difference indicates the quantity of heat which has been transferred to the water in the generation of steam. If the combustion is complete, the heat of the furnace will depend on the quantity of air furnished, that is, upon the area of the calorimeter, whence it appears that the calorimeter should be large. But if this space be an unbroken volume, much of the hot gas may pass through without impinging against the boiler surface, and delivering its heat, whence it is desirable that the space should be divided thoroughly; and it is evident that a reduced calorimeter may often give better results than a larger one, not properly arranged. A designer of boilers will find important tables on this point in Appleton's Cyclopædia of Mechanics.

Priming is the tendency of the water in the boiler to form spray by the bursting of the steam bubbles when they come to the surface of the water, the spray going forward with the steam into the cylinder. Here it is cooled and accumulates, especially if the exhaust port is not, either by position or capacity, adequate to its discharge. Water is practically incompressible, and if a quantity of it, greater than the volume of the clearance, is found before the piston, near the end of the stroke, it lies between the piston and the cylinder head as mischievous as a mass of metal would be in the same position. Something must yield. The crank pin may be broken, or the cylinder split, or the head burst out, and all rods and gearing will be ruined. Priming is caused by want of steam room, or of area at the surface of the water in the bodies, or by the use of dirty water. The latter cause may be cured by collecting the water in tanks, and giving it time to settle. The others may be avoided by proper construction of the boiler, by checking the steam at the throttle, or by working the engine more expansively. Any sudden removal of pressure, as the opening of the safety-valve, or of the throttle in starting, tends to produce priming, because while the water had, at the instant of the opening, a capacity for steam corresponding to the higher pressure, the diminished pressure sets free a gush of steam that is entirely disproportioned to the ordinary conditions. Some authorities advise the insertion of a perforated plate through which the steam must pass on its way to the cylinder; the water beating against this plate is arrested, and the steam passes on more freely. In some locomotives the steam is taken by a longitudinal perforated pipe, which serves the purpose of the steam dome of usual designs. Boilers in which the steam does not circulate freely because of the disposition of the tubes, are liable to the annoyance of priming.

The term horse power, when applied to the boiler, has a meaning scarcely more definite than when used to indicate the capacity of the engine. In either case, the horsepower realized depends as much upon the method of using the mechanism, as upon its original construction. The best authorities agree that the horse-power of the boiler should indicate the actual evaporation of water, instead of the size of the boiler or the efficiency which may be secured through the engine. The ability to evaporate a cubic foot of water per hour, making steam at 212 F., has been suggested as a suitable unit to be called a horse-power. To ascertain the evaporative power of a boiler by experiment, it is necessary to obtain the weights of fuel and water, and to know the quality of the steam produced. A trial should last 24 hours; steam may be raised, and then fire withdrawn, and the ash-pit cleared, the steam meanwhile being maintained with wood. Coal is then added, and as soon as it is fired, the test begins. Note is taken of the height of water in the gauge, and the water is left at the same height at the end of the test. Coal is carefully weighed in regular amounts and at regular intervals to avoid errors. At the end of the trial the fire is withdrawn, and the remaining coal weighed as soon as possible; this weight, plus that of the ashes made during the experiment, taken from the weight of the coal, gives the weight of fuel consumed. To find the quality of the steam, a tank is provided, which is traversed by a pipe leading to the boiler, the whole apparatus being so arranged as to waste as little heat as possible. The tank is filled with water, and steam is admitted through the pipe in such quantity as may be condensed by the water. We have to note the pressure of the steam, the weight and temperature of the water before steam is admitted, the weight and temperature at the close of the test, the weight and temperature of the water formed from the condensed steam, and the time. Experiment must also be made to test the loss of heat by radiation and evaporation, which is done by heating a given quantity of water to a given temperature in the same tank, and noting the loss in weight and temperature during a given time. To illustrate by an example. Suppose a test made, from which these data have been sccured:

Boiling.

Coal used, 5980 lbs. ; feed water used, 42,820 lbs.; coal withdrawn at end of test, with ashes, 1830 lbs.; hence, coal burned in the test, 5980 minus 1830-4150 lbs. The apparent evaporation per pound of coal is, 42,320+4150-10.2 lbs., if the steam were dry. To test the quality of the steam the described apparatus has been used, and these data noted: Pressure of steam at gauge, 80 lbs.; weight of steam condensed at 95, 204 lbs.; initial temperature of water for condensing, 60°; final temperature, 92°; head of water in tank, 27 in.; time of trial, 24 hours; and by former tests it appears that 4 cu. ft. of water, weighing 62.2 lbs. per ft., pass from the tank per hour, and that the loss of heat by evaporation and radiation is 1480 thermal units per hour. The heat given to the water by the condensing steam in one hour was 4×62.2× (95 minus 60)+1480-8708 thermal units. The steam condensed per hour was 204÷24=8.5 lbs., hence each pound of steam communicated to the water 8708÷8.5=1024.5 thermal units of heat. But this condensed steam was discharged at 95°; to bring it down to the standard of 32° there must have been a farther reduction of 95 minus 30-65 thermal units, showing that the quantity of heat above freezing standard held by a pound of steam as it issued from the boiler was 1024.5+65=1089.5 thermal units. The total heat, above freezing standard, of a pound of dry steam at 80 lbs. pressure (see Rankine, Steam Engine, or Appleton's Cyc. of Mechanics), is 1177.1; it is therefore evident that the steam used in the test contained some moisture. As the temperature of the feed-water was 60°, it had already 28 thermal units of heat per pound above water at 32°, and would require 1177.1 minus 28-1149.1 thermal units to change it to dry steam; but it required 1089.5 minus 28=1061.5 thermal units to change it to steam of the quality observed, hence the actual evaporation was 1061.5÷1149.1-0.91506 of the apparent evaporation. But the apparent evaporation was 8.5 lbs. per pound of coal, and the actual was therefore 7.778 lbs. If the feed-water were at 212, 998.5 thermal units would be required to convert a pound of water into steam. Hence, 1061.5+998.5=10.6= nearly the evaporation per pound from and at 212°.

BOILING (of liquids)—BOILING-POINT. When heat is applied to a vessel containing water, the temperature gradually rises, and vapor comes silently off the surface; but at a certain degree of heat, steam (q.v.) begins to be formed in small explosive bursts at the bottom, and rising through the liquid in considerable bubbles, throws it into commotion. If, after this, the steam is allowed freely to escape, the temperature of the water rises no higher, however great the heat of the fire. The water is then said to boil, and the temperature at which it remains permanent is its boiling-point. The boilingpoint of water is ordinarily 212°; but every liquid has a point of its own. Thus, sulphuric ether boils at 96°; alcohol, at 176; oil of turpentine, at 316'; sulphuric acid, at 620°; and mercury, at 662°. The boiling-point of liquids is constant, under the same conditions, but is liable to be altered by various circumstances. Water with common salt in it, e.g., requires greater heat to make it boil than pure water. The nature of the vessel, too, exerts an influence; in a glass vessel, the boiling-point of water is a degree or two higher than in one of metal, owing to the greater attraction that there is between water and glass than between water and a metal. But what most affects the boilingpoint is variation of pressure. It is only when the barometer stands at 30 in., showing an atmospheric pressure of 15 lbs. on the sq. in., that the boiling-point of water is 212. When the barometer falls, or when part of the pressure is in any other way removed, it boils before coming to 212°, and when the pressure is increased, the boiling-point rises. -Thus, in elevated positions, where there is less air above the liquid to press on its surface, the boiling point is lower than at the level of the sea. An elevation of 510 ft. above the sea-level, makes a diminution of a degree; at higher levels, the difference of elevation corresponding to a degree of temperature in the boiling-point increases; but the rate of variation once ascertained, a method is thus furnished of measuring the heights of mountains. See HEIGHTS, MEASUREMENT OF. At the city of Mexico, 7000 ft. above the sea, water boils at 200°; at Quito, 9000 ft., at 194; and on Donkia mountain, in the Himalaya, at the height of 18,000 ft., Dr. Hooker found it to boil at 180'. Boiling water is thus not always equally hot, and in elevated places, many substances cannot be cooked by boiling, Under the receiver of an air-pump, the same effect is still more strikingly seen; water may be made to boil at the temperature of summer, and ether when colder than ice. In complete vacuo, liquids, in general, boil at a temperature 140° lower than in the open air. The knowledge of this effect of diminished pressure is now largely turned to account in sugar-boiling, in distilling vegetable essences, and in other processes where the substances are apt to be injured by a high temperature. -By increasing the pressure, again, water may be heated to any degree without boiling. Papin's Digester (q. v.) is formed on this principle. Under a pressure of two atmos. pheres, the boiling-point rises to 234°; of four atmospheres, it is 294; of ten atmospheres, 359; of fifty atmospheres, 510°.

In a deep vessel, the water at the bottom has to sustain the pressure not only of the atmosphere, but also of the water above it. At a depth of 34 ft., the pressure of the water above is equal to an atmosphre, or 15 lbs. on the sq. in.; and thus, at the bottom of a vessel of that depth, the water must be heated to 234° before it is at its boilingpoint. This principle has been successfully applied to explain the phenomena of the Geysers (q.v.).

If a small quantity of water be poured into a silver basin, heated above the boiling

Boiling.

point, but below redness, it will begin to boil violently, or perhaps burst into steam at once. But if the basin is heated to redness, the water will gather itself into a globule, and roll about on the hot surface, without becoming heated to the boiling-point. For the explanation of this and other interesting phenomena connected with it, see SPHE ROIDAL CONDITION OF Liquids.

BOILING OF LIQUIDS (ante). As will be understood from the above, the terms liquid and boiling-point are entirely relative, depending upon external agents and upon each other. The statement that water is a liquid is only true under certain conditions. In the arctic regions it is a solid, and in a vessel heated to 212 under ordinary atmos pheric pressure it is a vapor or gas. Nitrous oxide is a liquid under ordinary atmos pheric pressure when reduced below 126 below zero, and the same is true of carbonic acid when reduced to 108.76. Pressure, however, is capable of reducing both these gases to liquids, and modern experiments with various substances are now common in which carbonic acid is liquefied by pressure. Ammonia, commonly a gas, is a liquid when reduced to -28.66. This substance is capable of being absorbed by a very small volume of water under heavy pressure, or, at least, of occupying a very small volume; for we cannot say that the gas is really absorbed; the water assists the pressure in holding the gas in a liquid form. Advantage is taken of this in the working of a certain class of ice-making machines, called ammonia machines (q.v.). Some of the machines, however, depend upon the vaporization of ammonia, anhydrous, or nearly so, for the absorption of sensible heat. The following is a table of the boiling points of various substances:

The investigations of prof. Kopp indicate certain remarkable laws connecting the boiling points of classes of liquids with their chemical constitution. The following tables, calculated from the observations of prof. Kopp and others, show that in the group of alcohols, and the acids derived from them by oxidation-both of which differ in constitution by one molecule of CH2-there is a difference of very nearly 34.2° F. between successive members of the series; and that, moreover, the difference in the boiling points between the alcohols and their respectively derived acids is about 72°.

Other analogous correspondences in the boiling-points of liquids and their chemical constitution were observed; thus in the series of hydrocarbons, homologous with benzole, CH, a difference in the series of CH, was attended with a difference of boiling point of about 43°.

The molecular constitution, or, more strictly speaking, the mutual relations between the molecules of liquids, particularly as regards water, whose affinities are so numerous. exerts a great influence not only upon the boiling-point, but upon the nature or manner of ebullition. Thus, if a clean glass flask is partially filled with ordinary, and, of course, more or less aerated spring water, and heated rapidly with a spirit-lamp, nearly all the air will be expelled first, but before all the air is thus expelled ebullition will commence, and at a point very slightly below 212°. After a little time, more of the air having disappeared, but not entirely, the boiling-point (at 30 in. mercurial pressure) will be 212. By continuing the boiling, however, the mode of ebullition will be found to have