deal in this place with the losses com- efficiency, for the lower limit cannot at mon to them. the best differ much from the temperature of surrounding bodies, but the up

The first cause of loss, viz., by conduction and radiation, is proportionately per is only limited by mechanical diffigreater for small than for large engines, culties; and though it may be shown not only from the fact that the heated that it is impossible to abstract all the agent is generally contained and con- imparted heat in the form of work, the veyed in cylindrical vessels and pipes, in proportion of heat so abstracted being, which the volume varies as the square of, in fact, never greater than one-seventh but the surface only directly as, the di- in the best steam engines of the day, ameter, and thus the radiation is propor- yet the efforts made in this direction tionately greater, but loss by leakage must be attended with success in proporand imperfect lagging is also generally tion as those difficulties are overcome. greater; furthermore, the loss of heat in In the discussion which followed the its transference from the fuel to the reading of Mr. Flannery's paper* on agent is greater, for, as will be shown by high-pressure steam boilers these diffithe results of experiments, the conditions of economical combustion are harder to maintain. Both these sources of loss may, however, be considerably modified by mechanical arrangements.

culties seem to have formed the only ground of objection to their use. Pressures of from 300 to 500 pounds per square inch have been successfully used by Mr. Perkins, and the application of high pressures and speeds is greatly increasing for small steam engines.

greater on a small than on a large scale; but, again, the proportionate waste of fuel is nearly always of less consequence. From the recently issued first reports on "Friction at High Speeds," of the Institution of Mechanical Engineers, it would appear that, at any rate up to the velocities at present used, the friction seems slightly to decrease as the velocity increases.

The second cause of loss, from the necessary rejection of heat, is often overlooked. There is not space to examine The third cause of loss, viz., that from the question as fully as it deserves, but friction and imperfection of machinery, the conclusion to be drawn is that the has a special interest from the high agent should be used between as great speeds now used in some small engines. ranges of temperature as practicable. It may be regarded as certain that the This is an important point in discussing proportionate loss from these causes is their relative efficiency, and since the proportionate range for hot air is greater than for steam or gas, the efficiency of the former in this particular is the higher; but at the same time this very quality renders it less capable of developing so much power, since only small differences of pressure accompany this range, which is limited by practical considerations, such as lubrication, packing, &c. There is, however, another and a more obvious way of regarding the action of these agents so as to compare their efficiencies; the heat rejected in the waste steam or gas has to be entirely restored to the new portion of the agent, by the expenditure of fuel, while in the air engine this is partly effected by compression in the return stroke; and so, though the loss of fuel is not so great in the air engine, the high-back pressure resulting from compression makes the resultant work also not so great, and leads to the conclusion, which is verified in practice, that air engines are not so powerful as steam or gas engines, though more eco

nomical.

The higher the initial temperature of the agent the greater will be its possible

SECTION II.—STEAM.

Steam engines are at present by far the most extensively used of small motors, some on a very small scale being made to develop considerable power.

The generation of steam on a small scale is almost entirely limited to vertical boilers-at any rate, for purposes on land-though the horizontal form, as well as sectional boilers, are also coming into more general use, the portability of the first being a strong recommendation. Figs. 1, 2 and 3 are sections of vertical boilers in common use, the general form being a cylindrical wrought-iron shell containing a furnace, also cylindrical * Vide Minutes of Proceedings Inst. C.E. vol., liv. p. 123.

and concentric with it, and fired through number of different devices for robbing an opening in one side. In other par- them of their heat before this takes ticulars there is considerable difference. place; the oldest and most usual is that In Fig. 1, which, though extensively of the cross tubes c, c, whose section is, used, is deservedly falling into disfavor, however, generally circular; these have the gases pass upward to the funnel the advantage of adding to the strength. through a number of vertical tubes, of the furnace; the furnace, besides beeither a cast-iron top or an enlarged ing subjected to external pressure, is the crown being necessary to allow their vulnerable part in the, by no means unescape to the funnel; this latter device common, event of shortness of water. is shown in the figure, and also an inge- Another plan is to have pendent tubes p, nious arrangement for preventing the p, p, of which the "Field" type is a good earlier destruction of the central tubes, example, as it allows a complete circulaby the tendency of the flames to ascend tion of water by separating the ascend

through them, which is especially the case when the fire is low. This consists of a concave wrought-iron plate (D), by whose under side the gases are reflected and are compelled to ascend round its edge, where combustion of some portion of them takes place, and the side of the furnace is acted on by the flame instead of by the current of cold air, which otherwise would be drawn up and be in contact with it. Fig. 2 is a still more common form, the gases passing up a central flue; but there are a considerable

ing and descending columns, by an inserted tube shown in the diagram. The success of this plan is shown from the fact that no less than 134,700 are now in use. Another good device is applied in the "Davy-Paxman " boiler (Fig. 4), in which the tubes, instead of ending in the furnace, pass down and communicate with the water in the lower part of the boiler, and so establish complete circulation, which is prevented from acting too violently by bafflers placed at their upper ends. These and various other

arrangements, though adding to the at d. Another form of sectional boiler complication of the boiler, are very de- may be briefly described as a square box sirable if effective in action, for the loss of tubes, which run from side to side in of heat in them, which experiments indi- two directions, and through which the cate, is thereby much diminished. Thus flames are made to pass successfully. the boiler of Latham and Bradley, 10 Yet another form, which is used in feet high and 4 feet in diameter, has 132 America, consists of a series of cast-iron square feet of heating surface. Fig. 3 spheres with necks, which are faced, and is an excellent design known as the so connection is made by means of long "Talbot," which presents considerable bolts. The advantages of sectional boilheating surface; the furnace is of casters are summarized in the paper of Mr. iron, which in a recent modification is Flannery,* where there is also a quotalined with firebrick, and the ash box acts tion from the third report of the Admias a water heater. These boilers have ralty on the subject. There is a large

given good results in practice, and the number of inventions of both vertical advantage of a non-tubulous boiler with and sectional boilers, but the foregoing large heating surface is realized by those examples are typical of the forms genwho have experienced the cost and trou- erally used. A small vertical boiler comble which tubes often entail. Fig. 5 is a monly has the engine attached to it, and view of an ordinary sectional boiler, and forms its support; this may be effected it will be seen that the amount of heat- by bolting the engine either to its side, ing surface exposed must cause it to be or by means of a cast-iron bedplate or a a rapid steam generator. The arrange- wrought-iron ring securing it upon its ment a, a, a of firebrick and the iron top. Portability is thus secured, and plates b, b, b divert the flames, and com- the evils arising from alternate expel them to pass well round the tubes, pansion and contraction of the boiler, after which they escape by the flue c. A portion of the steam space is shown

128.

which would be felt in a large engine, are so different in these latter, that no are not in this case appreciable, chiefly comparison can fairly be drawn. This because when used on a small scale the information being required, and also engine is only connected with a fixed data concerning the working of small external object by a belt or bearing at a steam engines of the ordinary type, little distance from it; but it has been through the kindness of Mr. Fry, Manremarked that with small marine en- aging Director of the Bristol Wagon gines, where the case is different, the ob- Works, and of Mr. Arrowsmith, Steam jection is considerable. One excellent Printing Works, Bristol, two series of example of the latter method is afforded experiments were made upon a 2-HP. by the design shown in Fig. 3, in which vertical engine and boiler, manufactured the cylinder is let into the boiler, and, by Dodman, of King's Lynn, belonging besides the benefit of short steam con- to the former, and a 4-HP. horizontal nections, a perfect steam jacket is formed. engine called the Soho, and a vertical The efficiency of small boilers is a boiler made by Tangye, of Birmingham, most important thing to consider, and, belonging to the latter. The data now so largely are they at present in use in given will refer only to the boilers, that

in conjunction with small engines, of the engines being introduced herethat it would be impossible to form a after. During the two days' trial of the satisfactory comparison between small former, which will be called No. 1, and motors without some reliable data on the one day's trial of the latter, No. 2, the subject. Now the paper of Mr. Flan- difficulty was experienced of keeping nery indicates a high efficiency for sec- uniform the quantity of water and the tional boilers, but these are not used to pressure of steam; but during a run of anything like the same extent as the ver- three hours on No. 1, and of four hours tical form; and on turning to different with No. 2, by careful stoking and atsources for information, the author was tention, and by taking observations unable to obtain independent results. every quarter of an hour of revoluFor instance, in the comprehensive man- tions, pressure of steam, and supply of ual of Mr. D. K. Clark, M. Inst. C.E., feeding water-a close approximation to although boilers of marine, locomotive, regularity was obtained. Both the dystationary, and portable engines are fully namometer and the indicator were used, treated, yet the class in question is not and the quantity of feed water and coals mentioned; and the ratio of heating sur- carefully weighed, and, as will be shown face to grate area and other conditions in No. 2, the volumes of steam used agreed fairly well with the volume of the cylinder and ports obtained by actu

A manual of "Rules, Tables, and Data for Mechan

ical Engineers." Blackie and Son.

Boiler.

ally filling these spaces with water afterwards, so that little, if any, priming could have taken place in No. 2, while in No. 1 the difference must not be overlooked, and the boiler efficiency in this case can only be approximate. No. 1 was of the form shown in Fig. 1, and was well lagged all over; No. 2 was of the form represented in Fig. 2, but was deficient in heating surface, having no pendent tubes and but one cross tube, circular in section.

been said it is seen that these boilers are being superseded by more efficient ones, with larger heating surface, as Latham's and Bradley's, which, though little larger than No. 2, has five times as much heating surface, and the field, of which there are 1,380 in use; but the introduction of more heating surface by tubes, &c., offers a larger surface for corrosion and leakage, and therefore involves more repairs; and the general conclusion is that the necessity for small boilers is the

TABLE OF RESULTS OF EXPERIMENTS ON SMALL VERTICAL BOILERS.

There are four ways of using steam on a small scale, which seem to lead to the following classification:

The feed water was warmed to about | weak point in the use of steam on a 100° Fahrenheit. The coal used with small scale. No. 1 was Parkfield; an analysis, with which the author was kindly furnished by Mr. Handal Cossham, shows 82 per cent. of carbon, 5.6 per cent. of hydrogen, and 6.1 per cent. of oxygen, and has a heating power of 14.3 pounds of water, at 212° Fahrenheit, converted into steam; so that, theoretically, the evaporation should have been from 100° Fahrenheit, about 12.5, instead of 3.3...

evaporation should have been about the same, making its efficiency=0.39. Thus, although the ratio of heating surface of No. 2 was considerably less than No. 1, the evaporative power was greater, which seems only to be accounted for by the fact that most of the heating surface of the latter was vertical. If the average evaporative power of large boilers from 100° Farhenheit for this kind of coal be taken at 0.65, it is seen how much inferior these small boilers are to larger ones, although from what has

(1.) Single-acting engines using high pressures and speeds. (2.) Double-acting engines using moderate pressures and speeds.

(3.) Rotary engines.

(4.) Apparatus for causing the pressure or impulse of steam on fluid surfaces to directly raise or force that fluid.

The first three are always non-condensing engines; the last kind can hardly be called engines, being apparatus of the injector or pulsometer type; but as they are used principally as a substitute for other kinds of steam pumps, or manual operations, they must certainly be considered as included by the term Small Motive Power. Thus it may be said that all small steam engines are noncondensing. This is obviously because the gain due to the partial absence of back pressure, which condensation secures, would not be sufficient to compensate for the increased complication and